statistical research case study

Introduction to Statistical Thinking

Chapter 16 case studies, 16.1 student learning objective.

This chapter concludes this book. We start with a short review of the topics that were discussed in the second part of the book, the part that dealt with statistical inference. The main part of the chapter involves the statistical analysis of 2 case studies. The tools that will be used for the analysis are those that were discussed in the book. We close this chapter and this book with some concluding remarks. By the end of this chapter, the student should be able to:

Review the concepts and methods for statistical inference that were presented in the second part of the book.

Apply these methods to requirements of the analysis of real data.

Develop a resolve to learn more statistics.

16.2 A Review

The second part of the book dealt with statistical inference; the science of making general statement on an entire population on the basis of data from a sample. The basis for the statements are theoretical models that produce the sampling distribution. Procedures for making the inference are evaluated based on their properties in the context of this sampling distribution. Procedures with desirable properties are applied to the data. One may attach to the output of this application summaries that describe these theoretical properties.

In particular, we dealt with two forms of making inference. One form was estimation and the other was hypothesis testing. The goal in estimation is to determine the value of a parameter in the population. Point estimates or confidence intervals may be used in order to fulfill this goal. The properties of point estimators may be assessed using the mean square error (MSE) and the properties of the confidence interval may be assessed using the confidence level.

The target in hypotheses testing is to decide between two competing hypothesis. These hypotheses are formulated in terms of population parameters. The decision rule is called a statistical test and is constructed with the aid of a test statistic and a rejection region. The default hypothesis among the two, is rejected if the test statistic falls in the rejection region. The major property a test must possess is a bound on the probability of a Type I error, the probability of erroneously rejecting the null hypothesis. This restriction is called the significance level of the test. A test may also be assessed in terms of it’s statistical power, the probability of rightfully rejecting the null hypothesis.

Estimation and testing were applied in the context of single measurements and for the investigation of the relations between a pair of measurements. For single measurements we considered both numeric variables and factors. For numeric variables one may attempt to conduct inference on the expectation and/or the variance. For factors we considered the estimation of the probability of obtaining a level, or, more generally, the probability of the occurrence of an event.

We introduced statistical models that may be used to describe the relations between variables. One of the variables was designated as the response. The other variable, the explanatory variable, is identified as a variable which may affect the distribution of the response. Specifically, we considered numeric variables and factors that have two levels. If the explanatory variable is a factor with two levels then the analysis reduces to the comparison of two sub-populations, each one associated with a level. If the explanatory variable is numeric then a regression model may be applied, either linear or logistic regression, depending on the type of the response.

The foundations of statistical inference are the assumption that we make in the form of statistical models. These models attempt to reflect reality. However, one is advised to apply healthy skepticism when using the models. First, one should be aware what the assumptions are. Then one should ask oneself how reasonable are these assumption in the context of the specific analysis. Finally, one should check as much as one can the validity of the assumptions in light of the information at hand. It is useful to plot the data and compare the plot to the assumptions of the model.

16.3 Case Studies

Let us apply the methods that were introduced throughout the book to two examples of data analysis. Both examples are taken from the case studies of the Rice Virtual Lab in Statistics can be found in their Case Studies section. The analysis of these case studies may involve any of the tools that were described in the second part of the book (and some from the first part). It may be useful to read again Chapters  9 – 15 before reading the case studies.

16.3.1 Physicians’ Reactions to the Size of a Patient

Overweight and obesity is common in many of the developed contrives. In some cultures, obese individuals face discrimination in employment, education, and relationship contexts. The current research, conducted by Mikki Hebl and Jingping Xu 87 , examines physicians’ attitude toward overweight and obese patients in comparison to their attitude toward patients who are not overweight.

The experiment included a total of 122 primary care physicians affiliated with one of three major hospitals in the Texas Medical Center of Houston. These physicians were sent a packet containing a medical chart similar to the one they view upon seeing a patient. This chart portrayed a patient who was displaying symptoms of a migraine headache but was otherwise healthy. Two variables (the gender and the weight of the patient) were manipulated across six different versions of the medical charts. The weight of the patient, described in terms of Body Mass Index (BMI), was average (BMI = 23), overweight (BMI = 30), or obese (BMI = 36). Physicians were randomly assigned to receive one of the six charts, and were asked to look over the chart carefully and complete two medical forms. The first form asked physicians which of 42 tests they would recommend giving to the patient. The second form asked physicians to indicate how much time they believed they would spend with the patient, and to describe the reactions that they would have toward this patient.

In this presentation, only the question on how much time the physicians believed they would spend with the patient is analyzed. Although three patient weight conditions were used in the study (average, overweight, and obese) only the average and overweight conditions will be analyzed. Therefore, there are two levels of patient weight (average and overweight) and one dependent variable (time spent).

The data for the given collection of responses from 72 primary care physicians is stored in the file “ discriminate.csv ” 88 . We start by reading the content of the file into a data frame by the name “ patient ” and presenting the summary of the variables:

Observe that of the 72 “patients”, 38 are overweight and 33 have an average weight. The time spend with the patient, as predicted by physicians, is distributed between 5 minutes and 1 hour, with a average of 27.82 minutes and a median of 30 minutes.

It is a good practice to have a look at the data before doing the analysis. In this examination on should see that the numbers make sense and one should identify special features of the data. Even in this very simple example we may want to have a look at the histogram of the variable “ time ”:

A feature in this plot that catches attention is the fact that there is a high concventration of values in the interval between 25 and 30. Together with the fact that the median is equal to 30, one may suspect that, as a matter of fact, a large numeber of the values are actually equal to 30. Indeed, let us produce a table of the response:

Notice that 30 of the 72 physicians marked “ 30 ” as the time they expect to spend with the patient. This is the middle value in the range, and may just be the default value one marks if one just needs to complete a form and do not really place much importance to the question that was asked.

The goal of the analysis is to examine the relation between overweigh and the Doctor’s response. The explanatory variable is a factor with two levels. The response is numeric. A natural tool to use in order to test this hypothesis is the \(t\) -test, which is implemented with the function “ t.test ”.

First we plot the relation between the response and the explanatory variable and then we apply the test:

Nothing seems problematic in the box plot. The two distributions, as they are reflected in the box plots, look fairly symmetric.

When we consider the report that produced by the function “ t.test ” we may observe that the \(p\) -value is equal to 0.005774. This \(p\) -value is computed in testing the null hypothesis that the expectation of the response for both types of patients are equal against the two sided alternative. Since the \(p\) -value is less than 0.05 we do reject the null hypothesis.

The estimated value of the difference between the expectation of the response for a patient with BMI=23 and a patient with BMI=30 is \(31.36364 -24.73684 \approx 6.63\) minutes. The confidence interval is (approximately) equal to \([1.99, 11.27]\) . Hence, it looks as if the physicians expect to spend more time with the average weight patients.

After analyzing the effect of the explanatory variable on the expectation of the response one may want to examine the presence, or lack thereof, of such effect on the variance of the response. Towards that end, one may use the function “ var.test ”:

In this test we do not reject the null hypothesis that the two variances of the response are equal since the \(p\) -value is larger than \(0.05\) . The sample variances are almost equal to each other (their ratio is \(1.044316\) ), with a confidence interval for the ration that essentially ranges between 1/2 and 2.

The production of \(p\) -values and confidence intervals is just one aspect in the analysis of data. Another aspect, which typically is much more time consuming and requires experience and healthy skepticism is the examination of the assumptions that are used in order to produce the \(p\) -values and the confidence intervals. A clear violation of the assumptions may warn the statistician that perhaps the computed nominal quantities do not represent the actual statistical properties of the tools that were applied.

In this case, we have noticed the high concentration of the response at the value “ 30 ”. What is the situation when we split the sample between the two levels of the explanatory variable? Let us apply the function “ table ” once more, this time with the explanatory variable included:

Not surprisingly, there is still high concentration at that level “ 30 ”. But one can see that only 2 of the responses of the “ BMI=30 ” group are above that value in comparison to a much more symmetric distribution of responses for the other group.

The simulations of the significance level of the one-sample \(t\) -test for an Exponential response that were conducted in Question  \[ex:Testing.2\] may cast some doubt on how trustworthy are nominal \(p\) -values of the \(t\) -test when the measurements are skewed. The skewness of the response for the group “ BMI=30 ” is a reason to be worry.

We may consider a different test, which is more robust, in order to validate the significance of our findings. For example, we may turn the response into a factor by setting a level for values larger or equal to “ 30 ” and a different level for values less than “ 30 ”. The relation between the new response and the explanatory variable can be examined with the function “ prop.test ”. We first plot and then test:

The mosaic plot presents the relation between the explanatory variable and the new factor. The level “ TRUE ” is associated with a value of the predicted time spent with the patient being 30 minutes or more. The level “ FALSE ” is associated with a prediction of less than 30 minutes.

The computed \(p\) -value is equal to \(0.05409\) , that almost reaches the significance level of 5% 89 . Notice that the probabilities that are being estimated by the function are the probabilities of the level “ FALSE ”. Overall, one may see the outcome of this test as supporting evidence for the conclusion of the \(t\) -test. However, the \(p\) -value provided by the \(t\) -test may over emphasize the evidence in the data for a significant difference in the physician attitude towards overweight patients.

16.3.2 Physical Strength and Job Performance

The next case study involves an attempt to develop a measure of physical ability that is easy and quick to administer, does not risk injury, and is related to how well a person performs the actual job. The current example is based on study by Blakely et al.  90 , published in the journal Personnel Psychology.

There are a number of very important jobs that require, in addition to cognitive skills, a significant amount of strength to be able to perform at a high level. Construction worker, electrician and auto mechanic, all require strength in order to carry out critical components of their job. An interesting applied problem is how to select the best candidates from amongst a group of applicants for physically demanding jobs in a safe and a cost effective way.

The data presented in this case study, and may be used for the development of a method for selection among candidates, were collected from 147 individuals working in physically demanding jobs. Two measures of strength were gathered from each participant. These included grip and arm strength. A piece of equipment known as the Jackson Evaluation System (JES) was used to collect the strength data. The JES can be configured to measure the strength of a number of muscle groups. In this study, grip strength and arm strength were measured. The outcomes of these measurements were summarized in two scores of physical strength called “ grip ” and “ arm ”.

Two separate measures of job performance are presented in this case study. First, the supervisors for each of the participants were asked to rate how well their employee(s) perform on the physical aspects of their jobs. This measure is summarizes in the variable “ ratings ”. Second, simulations of physically demanding work tasks were developed. The summary score of these simulations are given in the variable “ sims ”. Higher values of either measures of performance indicates better performance.

The data for the 4 variables and 147 observations is stored in “ job.csv ” 91 . We start by reading the content of the file into a data frame by the name “ job ”, presenting a summary of the variables, and their histograms:

All variables are numeric. Examination of the 4 summaries and histograms does not produce interest findings. All variables are, more or less, symmetric with the distribution of the variable “ ratings ” tending perhaps to be more uniform then the other three.

The main analyses of interest are attempts to relate the two measures of physical strength “ grip ” and “ arm ” with the two measures of job performance, “ ratings ” and “ sims ”. A natural tool to consider in this context is a linear regression analysis that relates a measure of physical strength as an explanatory variable to a measure of job performance as a response.

FIGURE 16.1: Scatter Plots and Regression Lines

Let us consider the variable “ sims ” as a response. The first step is to plot a scatter plot of the response and explanatory variable, for both explanatory variables. To the scatter plot we add the line of regression. In order to add the regression line we fit the regression model with the function “ lm ” and then apply the function “ abline ” to the fitted model. The plot for the relation between the response and the variable “ grip ” is produced by the code:

The plot that is produced by this code is presented on the upper-left panel of Figure  16.1 .

The plot for the relation between the response and the variable “ arm ” is produced by this code:

The plot that is produced by the last code is presented on the upper-right panel of Figure  16.1 .

Both plots show similar characteristics. There is an overall linear trend in the relation between the explanatory variable and the response. The value of the response increases with the increase in the value of the explanatory variable (a positive slope). The regression line seems to follow, more or less, the trend that is demonstrated by the scatter plot.

A more detailed analysis of the regression model is possible by the application of the function “ summary ” to the fitted model. First the case where the explanatory variable is “ grip ”:

Examination of the report reviles a clear statistical significance for the effect of the explanatory variable on the distribution of response. The value of R-squared, the ration of the variance of the response explained by the regression is \(0.4094\) . The square root of this quantity, \(\sqrt{0.4094} \approx 0.64\) , is the proportion of the standard deviation of the response that is explained by the explanatory variable. Hence, about 64% of the variability in the response can be attributed to the measure of the strength of the grip.

For the variable “ arm ” we get:

This variable is also statistically significant. The value of R-squared is \(0.4706\) . The proportion of the standard deviation that is explained by the strength of the are is \(\sqrt{0.4706} \approx 0.69\) , which is slightly higher than the proportion explained by the grip.

Overall, the explanatory variables do a fine job in the reduction of the variability of the response “ sims ” and may be used as substitutes of the response in order to select among candidates. A better prediction of the response based on the values of the explanatory variables can be obtained by combining the information in both variables. The production of such combination is not discussed in this book, though it is similar in principle to the methods of linear regression that are presented in Chapter  14 . The produced score 92 takes the form:

\[\mbox{\texttt{score}} = -5.434 + 0.024\cdot \mbox{\texttt{grip}}+ 0.037\cdot \mbox{\texttt{arm}}\;.\] We use this combined score as an explanatory variable. First we form the score and plot the relation between it and the response:

The scatter plot that includes the regression line can be found at the lower-left panel of Figure  16.1 . Indeed, the linear trend is more pronounced for this scatter plot and the regression line a better description of the relation between the response and the explanatory variable. A summary of the regression model produces the report:

Indeed, the score is highly significant. More important, the R-squared coefficient that is associated with the score is \(0.5422\) , which corresponds to a ratio of the standard deviation that is explained by the model of \(\sqrt{0.5422} \approx 0.74\) . Thus, almost 3/4 of the variability is accounted for by the score, so the score is a reasonable mean of guessing what the results of the simulations will be. This guess is based only on the results of the simple tests of strength that is conducted with the JES device.

Before putting the final seal on the results let us examine the assumptions of the statistical model. First, with respect to the two explanatory variables. Does each of them really measure a different property or do they actually measure the same phenomena? In order to examine this question let us look at the scatter plot that describes the relation between the two explanatory variables. This plot is produced using the code:

It is presented in the lower-right panel of Figure  16.1 . Indeed, one may see that the two measurements of strength are not independent of each other but tend to produce an increasing linear trend. Hence, it should not be surprising that the relation of each of them with the response produces essentially the same goodness of fit. The computed score gives a slightly improved fit, but still, it basically reflects either of the original explanatory variables.

In light of this observation, one may want to consider other measures of strength that represents features of the strength not captures by these two variable. Namely, measures that show less joint trend than the two considered.

Another element that should be examined are the probabilistic assumptions that underly the regression model. We described the regression model only in terms of the functional relation between the explanatory variable and the expectation of the response. In the case of linear regression, for example, this relation was given in terms of a linear equation. However, another part of the model corresponds to the distribution of the measurements about the line of regression. The assumption that led to the computation of the reported \(p\) -values is that this distribution is Normal.

A method that can be used in order to investigate the validity of the Normal assumption is to analyze the residuals from the regression line. Recall that these residuals are computed as the difference between the observed value of the response and its estimated expectation, namely the fitted regression line. The residuals can be computed via the application of the function “ residuals ” to the fitted regression model.

Specifically, let us look at the residuals from the regression line that uses the score that is combined from the grip and arm measurements of strength. One may plot a histogram of the residuals:

The produced histogram is represented on the upper panel. The histogram portrays a symmetric distribution that my result from Normally distributed observations. A better method to compare the distribution of the residuals to the Normal distribution is to use the Quantile-Quantile plot . This plot can be found on the lower panel. We do not discuss here the method by which this plot is produced 93 . However, we do say that any deviation of the points from a straight line is indication of violation of the assumption of Normality. In the current case, the points seem to be on a single line, which is consistent with the assumptions of the regression model.

The next task should be an analysis of the relations between the explanatory variables and the other response “ ratings ”. In principle one may use the same steps that were presented for the investigation of the relations between the explanatory variables and the response “ sims ”. But of course, the conclusion may differ. We leave this part of the investigation as an exercise to the students.

16.4 Summary

16.4.1 concluding remarks.

The book included a description of some elements of statistics, element that we thought are simple enough to be explained as part of an introductory course to statistics and are the minimum that is required for any person that is involved in academic activities of any field in which the analysis of data is required. Now, as you finish the book, it is as good time as any to say some words regarding the elements of statistics that are missing from this book.

One element is more of the same. The statistical models that were presented are as simple as a model can get. A typical application will required more complex models. Each of these models may require specific methods for estimation and testing. The characteristics of inference, e.g. significance or confidence levels, rely on assumptions that the models are assumed to possess. The user should be familiar with computational tools that can be used for the analysis of these more complex models. Familiarity with the probabilistic assumptions is required in order to be able to interpret the computer output, to diagnose possible divergence from the assumptions and to assess the severity of the possible effect of such divergence on the validity of the findings.

Statistical tools can be used for tasks other than estimation and hypothesis testing. For example, one may use statistics for prediction. In many applications it is important to assess what the values of future observations may be and in what range of values are they likely to occur. Statistical tools such as regression are natural in this context. However, the required task is not testing or estimation the values of parameters, but the prediction of future values of the response.

A different role of statistics in the design stage. We hinted in that direction when we talked about in Chapter  \[ch:Confidence\] about the selection of a sample size in order to assure a confidence interval with a given accuracy. In most applications, the selection of the sample size emerges in the context of hypothesis testing and the criteria for selection is the minimal power of the test, a minimal probability to detect a true finding. Yet, statistical design is much more than the determination of the sample size. Statistics may have a crucial input in the decision of how to collect the data. With an eye on the requirements for the final analysis, an experienced statistician can make sure that data that is collected is indeed appropriate for that final analysis. Too often is the case where researcher steps into the statistician’s office with data that he or she collected and asks, when it is already too late, for help in the analysis of data that cannot provide a satisfactory answer to the research question the researcher tried to address. It may be said, with some exaggeration, that good statisticians are required for the final analysis only in the case where the initial planning was poor.

Last, but not least, is the theoretical mathematical theory of statistics. We tried to introduce as little as possible of the relevant mathematics in this course. However, if one seriously intends to learn and understand statistics then one must become familiar with the relevant mathematical theory. Clearly, deep knowledge in the mathematical theory of probability is required. But apart from that, there is a rich and rapidly growing body of research that deals with the mathematical aspects of data analysis. One cannot be a good statistician unless one becomes familiar with the important aspects of this theory.

I should have started the book with the famous quotation: “Lies, damned lies, and statistics”. Instead, I am using it to end the book. Statistics can be used and can be misused. Learning statistics can give you the tools to tell the difference between the two. My goal in writing the book is achieved if reading it will mark for you the beginning of the process of learning statistics and not the end of the process.

16.4.2 Discussion in the Forum

In the second part of the book we have learned many subjects. Most of these subjects, especially for those that had no previous exposure to statistics, were unfamiliar. In this forum we would like to ask you to share with us the difficulties that you encountered.

What was the topic that was most difficult for you to grasp? In your opinion, what was the source of the difficulty?

When forming your answer to this question we will appreciate if you could elaborate and give details of what the problem was. Pointing to deficiencies in the learning material and confusing explanations will help us improve the presentation for the future editions of this book.

Hebl, M. and Xu, J. (2001). Weighing the care: Physicians’ reactions to the size of a patient. International Journal of Obesity, 25, 1246-1252. ↩

The file can be found on the internet at http://pluto.huji.ac.il/~msby/StatThink/Datasets/discriminate.csv . ↩

One may propose splinting the response into two groups, with one group being associated with values of “ time ” strictly larger than 30 minutes and the other with values less or equal to 30. The resulting \(p\) -value from the expression “ prop.test(table(patient$time>30,patient$weight)) ” is \(0.01276\) . However, the number of subjects in one of the cells of the table is equal only to 2, which is problematic in the context of the Normal approximation that is used by this test. ↩

Blakley, B.A., Qui?ones, M.A., Crawford, M.S., and Jago, I.A. (1994). The validity of isometric strength tests. Personnel Psychology, 47, 247-274. ↩

The file can be found on the internet at http://pluto.huji.ac.il/~msby/StatThink/Datasets/job.csv . ↩

The score is produced by the application of the function “ lm ” to both variables as explanatory variables. The code expression that can be used is “ lm(sims ~ grip + arm, data=job) ”. ↩

Generally speaking, the plot is composed of the empirical percentiles of the residuals, plotted against the theoretical percentiles of the standard Normal distribution. The current plot is produced by the expression “ qqnorm(residuals(sims.score)) ”. ↩

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base

Methodology

• What Is a Case Study? | Definition, Examples & Methods

What Is a Case Study? | Definition, Examples & Methods

Published on May 8, 2019 by Shona McCombes . Revised on November 20, 2023.

A case study is a detailed study of a specific subject, such as a person, group, place, event, organization, or phenomenon. Case studies are commonly used in social, educational, clinical, and business research.

A case study research design usually involves qualitative methods , but quantitative methods are sometimes also used. Case studies are good for describing , comparing, evaluating and understanding different aspects of a research problem .

Table of contents

When to do a case study, step 1: select a case, step 2: build a theoretical framework, step 3: collect your data, step 4: describe and analyze the case, other interesting articles.

A case study is an appropriate research design when you want to gain concrete, contextual, in-depth knowledge about a specific real-world subject. It allows you to explore the key characteristics, meanings, and implications of the case.

Case studies are often a good choice in a thesis or dissertation . They keep your project focused and manageable when you don’t have the time or resources to do large-scale research.

You might use just one complex case study where you explore a single subject in depth, or conduct multiple case studies to compare and illuminate different aspects of your research problem.

Prevent plagiarism. Run a free check.

Once you have developed your problem statement and research questions , you should be ready to choose the specific case that you want to focus on. A good case study should have the potential to:

• Provide new or unexpected insights into the subject
• Challenge or complicate existing assumptions and theories
• Propose practical courses of action to resolve a problem
• Open up new directions for future research

TipIf your research is more practical in nature and aims to simultaneously investigate an issue as you solve it, consider conducting action research instead.

Unlike quantitative or experimental research , a strong case study does not require a random or representative sample. In fact, case studies often deliberately focus on unusual, neglected, or outlying cases which may shed new light on the research problem.

Example of an outlying case studyIn the 1960s the town of Roseto, Pennsylvania was discovered to have extremely low rates of heart disease compared to the US average. It became an important case study for understanding previously neglected causes of heart disease.

However, you can also choose a more common or representative case to exemplify a particular category, experience or phenomenon.

Example of a representative case studyIn the 1920s, two sociologists used Muncie, Indiana as a case study of a typical American city that supposedly exemplified the changing culture of the US at the time.

While case studies focus more on concrete details than general theories, they should usually have some connection with theory in the field. This way the case study is not just an isolated description, but is integrated into existing knowledge about the topic. It might aim to:

• Exemplify a theory by showing how it explains the case under investigation
• Expand on a theory by uncovering new concepts and ideas that need to be incorporated
• Challenge a theory by exploring an outlier case that doesn’t fit with established assumptions

To ensure that your analysis of the case has a solid academic grounding, you should conduct a literature review of sources related to the topic and develop a theoretical framework . This means identifying key concepts and theories to guide your analysis and interpretation.

There are many different research methods you can use to collect data on your subject. Case studies tend to focus on qualitative data using methods such as interviews , observations , and analysis of primary and secondary sources (e.g., newspaper articles, photographs, official records). Sometimes a case study will also collect quantitative data.

Example of a mixed methods case studyFor a case study of a wind farm development in a rural area, you could collect quantitative data on employment rates and business revenue, collect qualitative data on local people’s perceptions and experiences, and analyze local and national media coverage of the development.

The aim is to gain as thorough an understanding as possible of the case and its context.

Receive feedback on language, structure, and formatting

Professional editors proofread and edit your paper by focusing on:

• Academic style
• Vague sentences
• Style consistency

See an example

In writing up the case study, you need to bring together all the relevant aspects to give as complete a picture as possible of the subject.

How you report your findings depends on the type of research you are doing. Some case studies are structured like a standard scientific paper or thesis , with separate sections or chapters for the methods , results and discussion .

Others are written in a more narrative style, aiming to explore the case from various angles and analyze its meanings and implications (for example, by using textual analysis or discourse analysis ).

In all cases, though, make sure to give contextual details about the case, connect it back to the literature and theory, and discuss how it fits into wider patterns or debates.

If you want to know more about statistics , methodology , or research bias , make sure to check out some of our other articles with explanations and examples.

• Normal distribution
• Degrees of freedom
• Null hypothesis
• Discourse analysis
• Control groups
• Mixed methods research
• Non-probability sampling
• Quantitative research
• Ecological validity

Research bias

• Rosenthal effect
• Implicit bias
• Cognitive bias
• Selection bias
• Negativity bias
• Status quo bias

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

McCombes, S. (2023, November 20). What Is a Case Study? | Definition, Examples & Methods. Scribbr. Retrieved April 9, 2024, from https://www.scribbr.com/methodology/case-study/

Is this article helpful?

Shona McCombes

Other students also liked, primary vs. secondary sources | difference & examples, what is a theoretical framework | guide to organizing, what is action research | definition & examples, what is your plagiarism score.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals

Statistics articles from across Nature Portfolio

Statistics is the application of mathematical concepts to understanding and analysing large collections of data. A central tenet of statistics is to describe the variations in a data set or population using probability distributions. This analysis aids understanding of what underlies these variations and enables predictions of future changes.

Latest Research and Reviews

A bicoherence approach to analyze multi-dimensional cross-frequency coupling in EEG/MEG data

• Alessio Basti
• Guido Nolte
• Laura Marzetti

Response times are affected by mispredictions in a stochastic game

• Paulo Roberto Cabral-Passos
• Antonio Galves
• Claudia D. Vargas

The effect of city reputation on Chinese corporate risk-taking

• Haifeng Jiang

EEG decoders track memory dynamics

Successful memorization could be decoded from brain activity. Here the authors decode human memory success from EEG recordings, suggesting memory is linked to context.

• Jesse K. Pazdera
• Michael J. Kahana

Improvement in variance estimation using transformed auxiliary variable under simple random sampling

• Syed Muhammad Asim
• Soofia Iftikhar

Quantitative analysis of the intensity distribution of optical rogue waves

While a lot of research efforts have been directed towards determining the emergence mechanism of optical rogue waves, less attention has been given to characterizing the level of “rogueness" in optical systems where rogue waves manifest. The authors provide such quantitative description for rogue waves resulting from supercontinuum generation

• Kirill Spasibko
• Radim Filip

News and Comment

Efficient learning of many-body systems

The Hamiltonian describing a quantum many-body system can be learned using measurements in thermal equilibrium. Now, a learning algorithm applicable to many natural systems has been found that requires exponentially fewer measurements than existing methods.

Fudging the volcano-plot without dredging the data

Selecting omic biomarkers using both their effect size and their differential status significance ( i.e. , selecting the “volcano-plot outer spray”) has long been equally biologically relevant and statistically troublesome. However, recent proposals are paving the way to resolving this dilemma.

• Thomas Burger

Disentangling truth from bias in naturally occurring data

A technique that leverages duplicate records in crowdsourcing data could help to mitigate the effects of biases in research and services that are dependent on government records.

• Daniel T. O’Brien

Sciama’s argument on life in a random universe and distinguishing apples from oranges

Dennis Sciama has argued that the existence of life depends on many quantities—the fundamental constants—so in a random universe life should be highly unlikely. However, without full knowledge of these constants, his argument implies a universe that could appear to be ‘intelligently designed’.

• Zhi-Wei Wang
• Samuel L. Braunstein

A method for generating constrained surrogate power laws

A paper in Physical Review X presents a method for numerically generating data sequences that are as likely to be observed under a power law as a given observed dataset.

• Zoe Budrikis

Connected climate tipping elements

Tipping elements are regions that are vulnerable to climate change and capable of sudden drastic changes. Now research establishes long-distance linkages between tipping elements, with the network analysis offering insights into their interactions on a global scale.

• Valerie N. Livina

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Effective Use of Statistics in Research – Methods and Tools for Data Analysis

Remember that impending feeling you get when you are asked to analyze your data! Now that you have all the required raw data, you need to statistically prove your hypothesis. Representing your numerical data as part of statistics in research will also help in breaking the stereotype of being a biology student who can’t do math.

Statistical methods are essential for scientific research. In fact, statistical methods dominate the scientific research as they include planning, designing, collecting data, analyzing, drawing meaningful interpretation and reporting of research findings. Furthermore, the results acquired from research project are meaningless raw data unless analyzed with statistical tools. Therefore, determining statistics in research is of utmost necessity to justify research findings. In this article, we will discuss how using statistical methods for biology could help draw meaningful conclusion to analyze biological studies.

Table of Contents

Role of Statistics in Biological Research

Statistics is a branch of science that deals with collection, organization and analysis of data from the sample to the whole population. Moreover, it aids in designing a study more meticulously and also give a logical reasoning in concluding the hypothesis. Furthermore, biology study focuses on study of living organisms and their complex living pathways, which are very dynamic and cannot be explained with logical reasoning. However, statistics is more complex a field of study that defines and explains study patterns based on the sample sizes used. To be precise, statistics provides a trend in the conducted study.

Biological researchers often disregard the use of statistics in their research planning, and mainly use statistical tools at the end of their experiment. Therefore, giving rise to a complicated set of results which are not easily analyzed from statistical tools in research. Statistics in research can help a researcher approach the study in a stepwise manner, wherein the statistical analysis in research follows –

1. Establishing a Sample Size

Usually, a biological experiment starts with choosing samples and selecting the right number of repetitive experiments. Statistics in research deals with basics in statistics that provides statistical randomness and law of using large samples. Statistics teaches how choosing a sample size from a random large pool of sample helps extrapolate statistical findings and reduce experimental bias and errors.

2. Testing of Hypothesis

When conducting a statistical study with large sample pool, biological researchers must make sure that a conclusion is statistically significant. To achieve this, a researcher must create a hypothesis before examining the distribution of data. Furthermore, statistics in research helps interpret the data clustered near the mean of distributed data or spread across the distribution. These trends help analyze the sample and signify the hypothesis.

3. Data Interpretation Through Analysis

When dealing with large data, statistics in research assist in data analysis. This helps researchers to draw an effective conclusion from their experiment and observations. Concluding the study manually or from visual observation may give erroneous results; therefore, thorough statistical analysis will take into consideration all the other statistical measures and variance in the sample to provide a detailed interpretation of the data. Therefore, researchers produce a detailed and important data to support the conclusion.

Types of Statistical Research Methods That Aid in Data Analysis

Statistical analysis is the process of analyzing samples of data into patterns or trends that help researchers anticipate situations and make appropriate research conclusions. Based on the type of data, statistical analyses are of the following type:

1. Descriptive Analysis

The descriptive statistical analysis allows organizing and summarizing the large data into graphs and tables . Descriptive analysis involves various processes such as tabulation, measure of central tendency, measure of dispersion or variance, skewness measurements etc.

2. Inferential Analysis

The inferential statistical analysis allows to extrapolate the data acquired from a small sample size to the complete population. This analysis helps draw conclusions and make decisions about the whole population on the basis of sample data. It is a highly recommended statistical method for research projects that work with smaller sample size and meaning to extrapolate conclusion for large population.

3. Predictive Analysis

Predictive analysis is used to make a prediction of future events. This analysis is approached by marketing companies, insurance organizations, online service providers, data-driven marketing, and financial corporations.

4. Prescriptive Analysis

Prescriptive analysis examines data to find out what can be done next. It is widely used in business analysis for finding out the best possible outcome for a situation. It is nearly related to descriptive and predictive analysis. However, prescriptive analysis deals with giving appropriate suggestions among the available preferences.

5. Exploratory Data Analysis

EDA is generally the first step of the data analysis process that is conducted before performing any other statistical analysis technique. It completely focuses on analyzing patterns in the data to recognize potential relationships. EDA is used to discover unknown associations within data, inspect missing data from collected data and obtain maximum insights.

6. Causal Analysis

Causal analysis assists in understanding and determining the reasons behind “why” things happen in a certain way, as they appear. This analysis helps identify root cause of failures or simply find the basic reason why something could happen. For example, causal analysis is used to understand what will happen to the provided variable if another variable changes.

7. Mechanistic Analysis

This is a least common type of statistical analysis. The mechanistic analysis is used in the process of big data analytics and biological science. It uses the concept of understanding individual changes in variables that cause changes in other variables correspondingly while excluding external influences.

Important Statistical Tools In Research

Researchers in the biological field find statistical analysis in research as the scariest aspect of completing research. However, statistical tools in research can help researchers understand what to do with data and how to interpret the results, making this process as easy as possible.

1. Statistical Package for Social Science (SPSS)

It is a widely used software package for human behavior research. SPSS can compile descriptive statistics, as well as graphical depictions of result. Moreover, it includes the option to create scripts that automate analysis or carry out more advanced statistical processing.

2. R Foundation for Statistical Computing

This software package is used among human behavior research and other fields. R is a powerful tool and has a steep learning curve. However, it requires a certain level of coding. Furthermore, it comes with an active community that is engaged in building and enhancing the software and the associated plugins.

3. MATLAB (The Mathworks)

It is an analytical platform and a programming language. Researchers and engineers use this software and create their own code and help answer their research question. While MatLab can be a difficult tool to use for novices, it offers flexibility in terms of what the researcher needs.

4. Microsoft Excel

Not the best solution for statistical analysis in research, but MS Excel offers wide variety of tools for data visualization and simple statistics. It is easy to generate summary and customizable graphs and figures. MS Excel is the most accessible option for those wanting to start with statistics.

5. Statistical Analysis Software (SAS)

It is a statistical platform used in business, healthcare, and human behavior research alike. It can carry out advanced analyzes and produce publication-worthy figures, tables and charts .

6. GraphPad Prism

It is a premium software that is primarily used among biology researchers. But, it offers a range of variety to be used in various other fields. Similar to SPSS, GraphPad gives scripting option to automate analyses to carry out complex statistical calculations.

This software offers basic as well as advanced statistical tools for data analysis. However, similar to GraphPad and SPSS, minitab needs command over coding and can offer automated analyses.

Use of Statistical Tools In Research and Data Analysis

Statistical tools manage the large data. Many biological studies use large data to analyze the trends and patterns in studies. Therefore, using statistical tools becomes essential, as they manage the large data sets, making data processing more convenient.

Following these steps will help biological researchers to showcase the statistics in research in detail, and develop accurate hypothesis and use correct tools for it.

There are a range of statistical tools in research which can help researchers manage their research data and improve the outcome of their research by better interpretation of data. You could use statistics in research by understanding the research question, knowledge of statistics and your personal experience in coding.

Have you faced challenges while using statistics in research? How did you manage it? Did you use any of the statistical tools to help you with your research data? Do write to us or comment below!

Frequently Asked Questions

Statistics in research can help a researcher approach the study in a stepwise manner: 1. Establishing a sample size 2. Testing of hypothesis 3. Data interpretation through analysis

Statistical methods are essential for scientific research. In fact, statistical methods dominate the scientific research as they include planning, designing, collecting data, analyzing, drawing meaningful interpretation and reporting of research findings. Furthermore, the results acquired from research project are meaningless raw data unless analyzed with statistical tools. Therefore, determining statistics in research is of utmost necessity to justify research findings.

Statistical tools in research can help researchers understand what to do with data and how to interpret the results, making this process as easy as possible. They can manage large data sets, making data processing more convenient. A great number of tools are available to carry out statistical analysis of data like SPSS, SAS (Statistical Analysis Software), and Minitab.

nice article to read

Holistic but delineating. A very good read.

Rate this article Cancel Reply

Your email address will not be published.

Enago Academy's Most Popular Articles

• Reporting Research

Research Interviews: An effective and insightful way of data collection

Research interviews play a pivotal role in collecting data for various academic, scientific, and professional…

Planning Your Data Collection: Designing methods for effective research

Planning your research is very important to obtain desirable results. In research, the relevance of…

• Language & Grammar

Best Plagiarism Checker Tool for Researchers — Top 4 to choose from!

While common writing issues like language enhancement, punctuation errors, grammatical errors, etc. can be dealt…

• Industry News
• Publishing News

2022 in a Nutshell — Reminiscing the year when opportunities were seized and feats were achieved!

It’s beginning to look a lot like success! Some of the greatest opportunities to research…

• Manuscript Preparation
• Publishing Research

Qualitative Vs. Quantitative Research — A step-wise guide to conduct research

A research study includes the collection and analysis of data. In quantitative research, the data…

2022 in a Nutshell — Reminiscing the year when opportunities were seized and feats…

Sign-up to read more

Subscribe for free to get unrestricted access to all our resources on research writing and academic publishing including:

• 2000+ blog articles
• 50+ Webinars
• 10+ Expert podcasts
• 50+ Infographics
• 10+ Checklists
• Research Guides

We hate spam too. We promise to protect your privacy and never spam you.

I am looking for Editing/ Proofreading services for my manuscript Tentative date of next journal submission:

What should universities' stance be on AI tools in research and academic writing?

13. Study design and choosing a statistical test

Sample size.

Top 99+ Trending Statistics Research Topics for Students

Being a statistics student, finding the best statistics research topics is quite challenging. But not anymore; find the best statistics research topics now!!!

Statistics is one of the tough subjects because it consists of lots of formulas, equations and many more. Therefore the students need to spend their time to understand these concepts. And when it comes to finding the best statistics research project for their topics, statistics students are always looking for someone to help them. 

In this blog, we will share with you the most interesting and trending statistics research topics in 2023. It will not just help you to stand out in your class but also help you to explore more about the world.

If you face any problem regarding statistics, then don’t worry. You can get the best statistics assignment help from one of our experts.

As you know, it is always suggested that you should work on interesting topics. That is why we have mentioned the most interesting research topics for college students and high school students. Here in this blog post, we will share with you the list of 99+ awesome statistics research topics.

Why Do We Need to Have Good Statistics Research Topics?

Table of Contents

Having a good research topic will not just help you score good grades, but it will also allow you to finish your project quickly. Because whenever we work on something interesting, our productivity automatically boosts. Thus, you need not invest lots of time and effort, and you can achieve the best with minimal effort and time. 

What Are Some Interesting Research Topics?

If we talk about the interesting research topics in statistics, it can vary from student to student. But here are the key topics that are quite interesting for almost every student:-

• Literacy rate in a city.
• Abortion and pregnancy rate in the USA.
• Eating disorders in the citizens.
• Parent role in self-esteem and confidence of the student.
• Uses of AI in our daily life to business corporates.

Top 99+ Trending Statistics Research Topics For 2023

Here in this section, we will tell you more than 99 trending statistics research topics:

Sports Statistics Research Topics

• Statistical analysis for legs and head injuries in Football.
• Statistical analysis for shoulder and knee injuries in MotoGP.
• Deep statistical evaluation for the doping test in sports from the past decade.
• Statistical observation on the performance of athletes in the last Olympics.
• Role and effect of sports in the life of the student.

Psychology Research Topics for Statistics

• Deep statistical analysis of the effect of obesity on the student’s mental health in high school and college students.
• Statistical evolution to find out the suicide reason among students and adults.
• Statistics analysis to find out the effect of divorce on children in a country.
• Psychology affects women because of the gender gap in specific country areas.
• Statistics analysis to find out the cause of online bullying in students’ lives. 
• In Psychology, PTSD and descriptive tendencies are discussed.
• The function of researchers in statistical testing and probability.
• Acceptable significance and probability thresholds in clinical Psychology.
• The utilization of hypothesis and the role of P 0.05 for improved comprehension.
• What types of statistical data are typically rejected in psychology?
• The application of basic statistical principles and reasoning in psychological analysis.
• The role of correlation is when several psychological concepts are at risk.
• Actual case study learning and modeling are used to generate statistical reports.
• In psychology, naturalistic observation is used as a research sample.
• How should descriptive statistics be used to represent behavioral data sets?

Applied Statistics Research Topics

• Does education have a deep impact on the financial success of an individual?
• The investment in digital technology is having a meaningful return for corporations?
• The gap of financial wealth between rich and poor in the USA.
• A statistical approach to identify the effects of high-frequency trading in financial markets.
• Statistics analysis to determine the impact of the multi-agent model in financial markets. 

Personalized Medicine Statistics Research Topics

• Statistical analysis on the effect of methamphetamine on substance abusers.
• Deep research on the impact of the Corona vaccine on the Omnicrone variant. 
• Find out the best cancer treatment approach between orthodox therapies and alternative therapies.
• Statistics analysis to identify the role of genes in the child’s overall immunity.
• What factors help the patients to survive from Coronavirus .

Experimental Design Statistics Research Topics

• Generic vs private education is one of the best for the students and has better financial return.
• Psychology vs physiology: which leads the person not to quit their addictions?
• Effect of breastmilk vs packed milk on the infant child overall development
• Which causes more accidents: male alcoholics vs female alcoholics.
• What causes the student not to reveal the cyberbullying in front of their parents in most cases. 

Easy Statistics Research Topics

• Application of statistics in the world of data science
• Statistics for finance: how statistics is helping the company to grow their finance
• Advantages and disadvantages of Radar chart
• Minor marriages in south-east Asia and African countries.
• Discussion of ANOVA and correlation.
• What statistical methods are most effective for active sports?
• When measuring the correctness of college tests, a ranking statistical approach is used.
• Statistics play an important role in Data Mining operations.
• The practical application of heat estimation in engineering fields.
• In the field of speech recognition, statistical analysis is used.
• Estimating probiotics: how much time is necessary for an accurate statistical sample?
• How will the United States population grow in the next twenty years?
• The legislation and statistical reports deal with contentious issues.
• The application of empirical entropy approaches with online grammar checking.
• Transparency in statistical methodology and the reporting system of the United States Census Bureau.

Statistical Research Topics for High School

• Uses of statistics in chemometrics
• Statistics in business analytics and business intelligence
• Importance of statistics in physics.
• Deep discussion about multivariate statistics
• Uses of Statistics in machine learning

Survey Topics for Statistics

• Gather the data of the most qualified professionals in a specific area.
• Survey the time wasted by the students in watching Tvs or Netflix.
• Have a survey the fully vaccinated people in the USA 
• Gather information on the effect of a government survey on the life of citizens
• Survey to identify the English speakers in the world.

Statistics Research Paper Topics for Graduates

• Have a deep decision of Bayes theorems
• Discuss the Bayesian hierarchical models
• Analysis of the process of Japanese restaurants. 
• Deep analysis of Lévy’s continuity theorem
• Analysis of the principle of maximum entropy

AP Statistics Topics

• Discuss about the importance of econometrics
• Analyze the pros and cons of Probit Model
• Types of probability models and their uses
• Deep discussion of ortho stochastic matrix
• Find out the ways to get an adjacency matrix quickly

Good Statistics Research Topics 

• National income and the regulation of cryptocurrency.
• The benefits and drawbacks of regression analysis.
• How can estimate methods be used to correct statistical differences?
• Mathematical prediction models vs observation tactics.
• In sociology research, there is bias in quantitative data analysis.
• Inferential analytical approaches vs. descriptive statistics.
• How reliable are AI-based methods in statistical analysis?
• The internet news reporting and the fluctuations: statistics reports.
• The importance of estimate in modeled statistics and artificial sampling.

Business Statistics Topics

• Role of statistics in business in 2023
• Importance of business statistics and analytics
• What is the role of central tendency and dispersion in statistics
• Best process of sampling business data.
• Importance of statistics in big data.
• The characteristics of business data sampling: benefits and cons of software solutions.
• How may two different business tasks be tackled concurrently using linear regression analysis?
• In economic data relations, index numbers, random probability, and correctness are all important.
• The advantages of a dataset approach to statistics in programming statistics.
• Commercial statistics: how should the data be prepared for maximum accuracy?

Statistical Research Topics for College Students

• Evaluate the role of John Tukey’s contribution to statistics.
• The role of statistics to improve ADHD treatment.
• The uses and timeline of probability in statistics.
• Deep analysis of Gertrude Cox’s experimental design in statistics.
• Discuss about Florence Nightingale in statistics.
• What sorts of music do college students prefer?
• The Main Effect of Different Subjects on Student Performance.
• The Importance of Analytics in Statistics Research.
• The Influence of a Better Student in Class.
• Do extracurricular activities help in the transformation of personalities?
• Backbenchers’ Impact on Class Performance.
• Medication’s Importance in Class Performance.
• Are e-books better than traditional books?
• Choosing aspects of a subject in college

How To Write Good Statistics Research Topics?

So, the main question that arises here is how you can write good statistics research topics. The trick is understanding the methodology that is used to collect and interpret statistical data. However, if you are trying to pick any topic for your statistics project, you must think about it before going any further. 

As a result, it will teach you about the data types that will be researched because the sample will be chosen correctly. On the other hand, your basic outline for choosing the correct topics is as follows:

• Introduction of a problem
• Methodology explanation and choice. 
• Statistical research itself is in the main part (Body Part). 
• Samples deviations and variables. 
• Lastly, statistical interpretation is your last part (conclusion). 

Note:   Always include the sources from which you obtained the statistics data.

Top 3 Tips to Choose Good Statistics Research Topics

It can be quite easy for some students to pick a good statistics research topic without the help of an essay writer. But we know that it is not a common scenario for every student. That is why we will mention some of the best tips that will help you choose good statistics research topics for your next project. Either you are in a hurry or have enough time to explore. These tips will help you in every scenario.

1. Narrow down your research topic

We all start with many topics as we are not sure about our specific interests or niche. The initial step to picking up a good research topic for college or school students is to narrow down the research topic.

For this, you need to categorize the matter first. And then pick a specific category as per your interest. After that, brainstorm about the topic’s content and how you can make the points catchy, focused, directional, clear, and specific. 

2. Choose a topic that gives you curiosity

After categorizing the statistics research topics, it is time to pick one from the category. Don’t pick the most common topic because it will not help your grades and knowledge. Instead of it, please choose the best one, in which you have little information, or you are more likely to explore it.

In a statistics research paper, you always can explore something beyond your studies. By doing this, you will be more energetic to work on this project. And you will also feel glad to get them lots of information you were willing to have but didn’t get because of any reasons.

It will also make your professor happy to see your work. Ultimately it will affect your grades with a positive attitude.

3. Choose a manageable topic

Now you have decided on the topic, but you need to make sure that your research topic should be manageable. You will have limited time and resources to complete your project if you pick one of the deep statistics research topics with massive information.

Then you will struggle at the last moment and most probably not going to finish your project on time. Therefore, spend enough time exploring the topic and have a good idea about the time duration and resources you will use for the project. 

Statistics research topics are massive in numbers. Because statistics operations can be performed on anything from our psychology to our fitness. Therefore there are lots more statistics research topics to explore. But if you are not finding it challenging, then you can take the help of our statistics experts . They will help you to pick the most interesting and trending statistics research topics for your projects. 

With this help, you can also save your precious time to invest it in something else. You can also come up with a plethora of topics of your choice and we will help you to pick the best one among them. Apart from that, if you are working on a project and you are not sure whether that is the topic that excites you to work on it or not. Then we can also help you to clear all your doubts on the statistics research topic. 

Frequently Asked Questions

Q1. what are some good topics for the statistics project.

Have a look at some good topics for statistics projects:- 1. Research the average height and physics of basketball players. 2. Birth and death rate in a specific city or country. 3. Study on the obesity rate of children and adults in the USA. 4. The growth rate of China in the past few years 5. Major causes of injury in Football

Q2. What are the topics in statistics?

Statistics has lots of topics. It is hard to cover all of them in a short answer. But here are the major ones: conditional probability, variance, random variable, probability distributions, common discrete, and many more. 

Q3. What are the top 10 research topics?

Here are the top 10 research topics that you can try in 2023:

1. Plant Science 2. Mental health 3. Nutritional Immunology 4. Mood disorders 5. Aging brains 6. Infectious disease 7. Music therapy 8. Political misinformation 9. Canine Connection 10. Sustainable agriculture

Related Posts

How to Find the Best Online Statistics Homework Help

Why SPSS Homework Help Is An Important aspect for Students?

• Open access
• Published: 06 April 2024

Statistical analyses of ordinal outcomes in randomised controlled trials: a scoping review

• Chris J. Selman   ORCID: orcid.org/0000-0002-1277-5538 1 , 2 ,
• Katherine J. Lee 1 , 2 ,
• Kristin N. Ferguson 4 ,
• Clare L. Whitehead 4 , 5 ,
• Brett J. Manley 4 , 6 , 7 &
• Robert K. Mahar 1 , 3  

Trials volume  25 , Article number:  241 ( 2024 ) Cite this article

294 Accesses

3 Altmetric

Metrics details

Randomised controlled trials (RCTs) aim to estimate the causal effect of one or more interventions relative to a control. One type of outcome that can be of interest in an RCT is an ordinal outcome, which is useful to answer clinical questions regarding complex and evolving patient states. The target parameter of interest for an ordinal outcome depends on the research question and the assumptions the analyst is willing to make. This review aimed to provide an overview of how ordinal outcomes have been used and analysed in RCTs.

The review included RCTs with an ordinal primary or secondary outcome published between 2017 and 2022 in four highly ranked medical journals (the British Medical Journal , New England Journal of Medicine , The Lancet , and the Journal of the American Medical Association ) identified through PubMed. Details regarding the study setting, design, the target parameter, and statistical methods used to analyse the ordinal outcome were extracted.

The search identified 309 studies, of which 144 were eligible for inclusion. The most used target parameter was an odds ratio, reported in 78 (54%) studies. The ordinal outcome was dichotomised for analysis in 47 ( \(33\%\) ) studies, and the most common statistical model used to analyse the ordinal outcome on the full ordinal scale was the proportional odds model (64 [ \(44\%\) ] studies). Notably, 86 (60%) studies did not explicitly check or describe the robustness of the assumptions for the statistical method(s) used.

Conclusions

The results of this review indicate that in RCTs that use an ordinal outcome, there is variation in the target parameter and the analytical approaches used, with many dichotomising the ordinal outcome. Few studies provided assurance regarding the appropriateness of the assumptions and methods used to analyse the ordinal outcome. More guidance is needed to improve the transparent reporting of the analysis of ordinal outcomes in future trials.

Peer Review reports

Randomised controlled trials (RCTs) aim to estimate the causal effect of one or more interventions relative to a control or reference intervention. Ordinal outcomes are useful in RCTs because the categories can represent multiple patient states within a single endpoint. The definition of an ordinal outcome is one that comprises monotonically ranked categories that are ordered hierarchically such that the distance between any two categories is not necessarily equal (or even meaningfully quantifiable) [ 1 ]. Ordinal outcomes should have categories that are mutually exclusive and unambiguously defined and can be used to capture improvement and deterioration relative to a baseline value where relevant [ 2 ]. If an ordinal scale is being used to capture change in patient status, then the ordinal outcome should also be symmetric to avoid favouring a better or worse health outcome [ 2 ]. Commonly used ordinal outcomes in RCTs include the modified-Rankin scale, a 7-category measure of disability following stroke or neurological insult [ 3 , 4 , 5 , 6 ], the Glasgow Outcome Scale-Extended (GOS-E), an 8-category measure of functional impairment post traumatic brain injury [ 7 ], and the World Health Organization (WHO) COVID-19 Clinical Progression Scale [ 8 ], an 11-point measure of disease severity among patients with COVID-19. The WHO Clinical Progression Scale, developed specifically for COVID-19 in 2020 [ 8 ], has been used in many RCTs evaluating COVID-19 disease severity and progression [ 9 , 10 ] and has helped to increase the familiarity of ordinal data and modelling approaches for ordinal outcomes for clinicians and statisticians alike [ 11 ].

Randomised controlled trials that use ordinal outcomes need to be designed and analysed with care. This includes the need to explicitly define the target parameter to compare the intervention groups (i.e. the target of estimation, for example, a proportional odds ratio (OR)), the analysis approach, and whether assumptions used in the analysis are valid. Although this is true for all RCTs, these issues are more complex when using an ordinal outcome compared to a binary or continuous outcome. For example, the choice of target parameter for an ordinal outcome depends on both the research question [ 12 , 13 ] and the assumptions that the analyst is willing to make about the data.

One option is to preserve the ordinal nature of the outcome, which can give rise to a number of different target parameters. Principled analysis of ordinal data often relies on less familiar statistical methods and underlying assumptions. Many statistical methods have been proposed to analyse ordinal outcomes. One approach to estimate the effect of treatment on the distribution of ordinal endpoints is to use a cumulative logistic model [ 14 , 15 ]. This model uses the distribution of the cumulative log-odds of the ordinal outcome to estimate a set of ORs [ 16 ], which, for an increase in the value of a covariate, represents the odds of being in the same or higher category at each level of the ordinal scale [ 15 ]. Modelling is vastly simplified by assuming that each covariate in the model exerts the same effect on the cumulative log odds for each binary split of the ordinal outcome, regardless of the threshold. This is known as the proportional odds (PO) assumption, with the model referred to as ordered logistic regression or the PO model (we refer to the latter term herein). The PO model has desirable properties of palindromic invariance (where the estimates of the parameters are not equivalent when the order of the categories are reversed) and invariance under collapsibility (where the estimated target parameter is changed when categories of the response are combined or removed) [ 17 ]. Studies have shown that an ordinal analysis of the outcome using a PO model increases the statistical power relative to an analysis of the dichotomised scale [ 18 , 19 ]. The target parameter from this model, the proportional or common OR, also has a relatively intuitive interpretation [ 20 , 21 ], representing a shift in the distribution of ordinal scale scores toward a better outcome in an intervention group compared to a reference group.

The PO model approach makes the assumption that the odds are proportional for each binary split of the ordinal outcome. If this assumption is violated then the proportional OR may be misleading in certain circumstances. Specifically, violation to PO can affect type I or II errors and/or may distort the magnitude of the treatment effect. For example, violation of proportional odds can increase the likelihood of making a type I error since the model may incorrectly identify evidence of a relationship between the treatment and outcome. Violation of the proportional odds assumption may also increase the likelihood of a type II error as the model may fail to identify a relationship between the treatment and the ordinal outcome because the model may fail to capture the true complexity of the relationship. In addition, a treatment may exert a harmful effect for some categories of the ordinal outcome but exert a beneficial effect for the remaining categories, which can ‘average’ out to no treatment effect when assuming a constant OR across the levels of the ordinal scale. The violation of PO may be harmful if the interest is also to estimate predicted probabilities for the categories of the ordinal scale, which will be too low or high for some outcomes when PO is assumed. Although the PO assumption will ‘average’ the treatment effect across the categories of the ordinal outcome, this may not be a problem if all of the treatment effects for each cut-point are in the same direction and the research aim is to simply show whether the treatment is effective even in the presence of non-PO. If the PO assumption is meaningfully violated and the interest is either in the treatment effect on a specific range of the outcome or to obtain predicted probabilities for each category of the scale, the PO model can be extended to a partial proportional odds (PPO) model which allows the PO assumption to be relaxed for a specific set or for all covariates in the model [ 22 ]. There are two types of PPO models: the unconstrained PPO model, in which the cumulative log-ORs across each cut-point vary freely across some or all of the cut-points [ 23 ], and the constrained PPO model, which assumes some functional relationship between the cumulative log-ORs [ 21 ]. However, such an approach may be more inefficient than using a PO model [ 24 , 25 ].

Alternative statistical methods that can be used to analyse the ordinal outcome include multinomial regression, which estimates an OR for each category of the ordinal outcome relative to the baseline category. The disadvantage of multinomial regression is that the number of ORs requiring estimation increases with the number of categories in the ordinal outcome. A larger sample size may therefore be required to ensure accurate precision of the many target parameters. Other methods are the continuation ratio model or adjacent-category logistic model, though these models lack two desirable properties: palindromic invariance and invariance under collapsibility [ 15 , 17 , 26 ].

Another option is to use alternative methods, such as the Mann-Whitney U  test or Wilcoxon rank-sum test [ 27 ] (referred to as the Wilcoxon test herein). The Wilcoxon test is equivalent to the PO model with a single binary exposure variable [ 15 , 28 ]. The treatment effect from a Wilcoxon test is the concordance probability that represents the probability that a randomly chosen observation from a treatment group is greater than a randomly chosen observation from a control group [ 29 , 30 ]. This parameter closely mirrors the OR derived from the PO model. Importantly, the direction of the OR from the PO model always agrees with the direction of the concordance probability. The disadvantages of the Wilcoxon test are that the concordance probability may be unfamiliar to clinicians, and the Wilcoxon test cannot be adjusted for covariates.

Another option is to dichotomise the ordinal outcome and use an OR or risk difference as the target parameter, estimated using logistic or binomial regression. This produces an effect estimate with clear clinical interpretations that may be suitable for specific clinical settings. The disadvantage of dichotomising an ordinal outcome is that it means discarding potentially useful information within the levels of the scale. This means that the trial may require a larger sample size to maintain the same level of statistical power to detect a clinically important treatment effect [ 19 ], which may not be feasible in all RCTs depending on cost constraints or the rate of recruitment. The decision to dichotomise may also depend on when the outcome is being measured. This was highlighted in a study that showed that an ordinal analysis of the modified-Rankin scale captured differences in long-term outcomes in survivors of stroke better than an analysis that dichotomised the ordinal outcome [ 3 , 31 ].

An alternative to dichotomisation is to treat the ordinal outcome as continuous and focus on the mean difference as the target parameter. This choice to treat the outcome as continuous may be based on the number of categories, where the more categories, the more the outcome resembles a continuum if proximate categories measure similar states or if the scale reflects a latent continuous variable. This has the advantage that modelling is straightforward and familiar, but it can lead to ill-defined clinical interpretations of the treatment effect since the difference between proximate categories is unequal nor quantifiable. Such an analysis also wrongly assumes that the outcome has an unbounded range.

There has been commentary [ 32 ] and research conducted on the methodology of using ordinal outcomes in certain RCT settings that have mainly focused on the benefit of an ordinal analysis using a PO model [ 19 , 33 , 34 , 35 ], including investigations into the use of a PPO model when the PO assumption is violated [ 36 ]. However, these studies have primarily focused on a limited number of statistical methods and in mostly specific medical areas such as neurology and may not be applicable more generally. Given the growing use of ordinal outcomes in RCTs, it is crucial to gain a deeper understanding of how ordinal outcomes are utilised in practice. This understanding will help identify any issues in the use of ordinal outcomes in RCTs and facilitate discussions on improving the reporting and analysis of such outcomes. To address this, we conducted a scoping review to systematically examine the use and analysis of ordinal outcomes in the current literature. Specifically, we aimed to:

Identify which target parameters are of interest in RCTs that use an ordinal outcome and whether these are explicitly defined.

Describe how ordinal outcomes are analysed in RCTs to estimate a treatment effect.

Describe whether RCTs that use an ordinal outcome adequately report key methodological aspects specific to the analysis of the ordinal outcome.

A pre-specified protocol was developed for this scoping review [ 37 ]. Deviations from the protocol are outlined in Additional file 1 . Here, we provide an overview of the protocol and present the findings from the review which have been reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) checklist [ 38 ].

Eligibility criteria

Studies were included in the review if they were published in one of four highly ranked medical journals ( British Medical Journal (BMJ), New England Journal of Medical (NEJM), Journal of the American Medical Association (JAMA), or The Lancet) between 1 January 2017 and 31 July 2022 and reported the results of at least one RCT (e.g. if reporting results from multiple trials) with either a primary or secondary outcome that was measured on an ordinal scale. These journals were chosen because they are leading medical journals that publish original and peer-reviewed research with primarily clinical aims and have been used in other reviews of trial methodology [ 39 , 40 ]. RCTs were defined using the Cochrane definition of an RCT, which is a study that prospectively assigns individuals to one of two (or more) interventions using some random or quasi-random method of allocation [ 41 ].

Studies were excluded from this review if they were written in a language other than English, since we did not have sufficient resources to translate studies written in another language. We also excluded studies which were purely methodological, where the abstract or full-text was not available, which reported data from non-human subjects, and those that provided a commentary, review opinion, or were description only. Manuscripts that reported only a trial protocol or statistical analysis plan were also excluded, since one of the main objectives of this review was to determine which statistical methods are being used to analyse trial data. Studies that used ordinal outcomes that were measured on a numerical rating or visual analogue scale were also excluded. Although these scales are often considered ordinal, they imply equidistance between contiguous categories, and can conceivably be analysed as continuous data.

Information sources

Studies were identified and included in the review by searching the online bibliographic database, PubMed, executed on 3 August, 2022.

Search strategy

The search strategy for this review was developed by CJS in consultation with KJL and RKM. The search strategy employed terms that have been developed to identify RCTs [ 41 ] and terms that have been used to describe an ordinal outcome in published manuscripts for RCTs. The complete search strategy that was used in this review is described in Table 1 .

Selection of sources of evidence

There was no pre-specified sample size for this review. All eligible studies that were identified via the search strategy were included in the review.

Piloting of the eligibility criteria was conducted by CJS and RKM who independently assessed the titles and abstracts of 20 studies to ensure consistency between reviewers. CJS then performed the search on the PubMed database. All titles and abstracts identified were extracted into Covidence, a web-based tool for managing systematic reviews [ 42 ]. A two-phase screening process was employed, where all abstracts and titles were screened by CJS in the first phase. Those studies that were not excluded were then moved to the second phase of the screening process, where the full text was evaluated against the eligibility criteria by CJS. A random sample of 40 studies were also assessed for eligibility by a second reviewer (one of KJL, RKM, BJM, or CLW). All studies that were deemed eligible were included in the data extraction.

Data extraction

A data extraction questionnaire was developed in Covidence [ 42 ] and piloted by CJS and RKM using a sample of 10 studies, which was further refined. The final version of the questionnaire is shown in Additional file 2 , and a full list of the data extraction items is provided in Table 2 . Data was extracted from both the main manuscript and any supplementary material, including statistical analysis plans. CJS extracted data from all eligible studies in the review. Double data extraction was performed by KJL and RKM on a random sample of 20 studies. Any uncertainties in the screening and data extraction process were discussed and resolved by consensus among all reviewers. Simplifications and assumptions that were made for eligibility and data extraction are outlined in Additional file 1 .

Synthesis of results

The data extracted from Covidence were cleaned and analysed using Stata [ 43 ]. Descriptive statistics were used to summarise the data. Frequencies and percentages and medians and interquartile ranges (IQRs) were reported for categorical and continuous variables respectively. Qualitative data were synthesised in a narrative format.

Results of the search

The initial search identified 309 studies, of which 46 were excluded for not being an RCT. There were 263 studies that underwent full text review. Of these, 119 were excluded: 110 because they did not have an ordinal outcome, and nine because they were not an RCT. In total, 144 studies were eligible for data extraction [ 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 , 90 , 91 , 92 , 93 , 94 , 95 , 96 , 97 , 98 , 99 , 100 , 101 , 102 , 103 , 104 , 105 , 106 , 107 , 108 , 109 , 110 , 111 , 112 , 113 , 114 , 115 , 116 , 117 , 118 , 119 , 120 , 121 , 122 , 123 , 124 , 125 , 126 , 127 , 128 , 129 , 130 , 131 , 132 , 133 , 134 , 135 , 136 , 137 , 138 , 139 , 140 , 141 , 142 , 143 , 144 , 145 , 146 , 147 , 148 , 149 , 150 , 151 , 152 , 153 , 154 , 155 , 156 , 157 , 158 , 159 , 160 , 161 , 162 , 163 , 164 , 165 , 166 , 167 , 168 , 169 , 170 , 171 , 172 , 173 , 174 , 175 , 176 , 177 , 178 , 179 , 180 , 181 , 182 , 183 , 184 , 185 , 186 , 187 ]. A flow diagram of the study selection is shown in Fig. 1 . The questionnaire that was used to extract the data from each study is provided in Additional file 2 .

Flow diagram of the study

Study characteristics

A summary of the study characteristics is presented in Table 3 . The highest proportion of studies were published in the NEJM (61 studies, \(42\%\) ), followed by JAMA (40, 28%) and The Lancet (34, 24%), with only nine studies published in the BMJ ( \(6\%\) ). The number of studies that used an ordinal outcome were higher in 2020 and 2021 ( \(30, 21\%\) in each year) compared to earlier years ( \(21, 15\%\) in 2019, \(24, 17\%\) in 2018 and \(23, 16\%\) in 2017). Nearly all studies were conducted in a clinical setting ( \(141, 98\%\) ). The most common medical condition being studied was stroke ( \(39, 28\%\) ), followed by COVID-19 ( \(22, 16\%\) ) and atopic dermatitis ( \(6, 4\%\) ). The most common medical field was neurology ( \(54, 38\%\) ) followed by infectious diseases ( \(22, 16\%\) , all of which were COVID-19 studies), dermatology ( \(13, 9\%\) ), and psychiatry ( \(12, 9\%\) ). Studies were mostly funded by public sources ( \(104, 72\%\) ). The median number of participants in the primary analysis of the ordinal outcome was 380 (interquartile range (IQR): 202–803).

Of the 144 included studies, 58 (40%) used some form of adaptive design, with 47 ( \(33\%\) ) having explicitly defined early stopping rules for efficacy or futility, 18 ( \(13\%\) ) used sample size re-estimation, three ( \(2\%\) ) used response adaptive randomisation, three ( \(2\%\) ) used covariate adaptive randomisation, three ( \(2\%\) ) were platform trials, and three ( \(2\%\) ) used adaptive enrichment that focused on specific subgroups of patients.

Ordinal outcomes and target parameters

A summary of the properties of the ordinal outcomes used in the studies is shown in Table 4 . An ordinal scale was used as a primary outcome in 59 ( \(41\%\) ) of studies. Most studies used an ordinal scale to describe an outcome at a single point in time ( \(128, 89\%\) ), with 16 studies using an ordinal outcome to capture changes over time ( \(11\%\) ). One study used a Likert scale where the categories were ambiguously defined in the manuscript. Another study used an ordinal outcome to measure change over time, but it was asymmetric and biased towards a favourable outcome. The median number of categories in the ordinal outcome was 7 (IQR: 6–7) and ranged from 3 to 23 categories.

There were 32 studies that determined the sample size in advance based on the ordinal outcome, of which 26 out of 32 studies ( \(81\%\) ) used an analytical approach and 6 out of 32 studies ( \(19\%\) ) used simulation to estimate the sample size. Among those studies that used an analytical approach, five studies reported to have used the Whitehead method and three studies reported to have used a t -test. Among the remaining studies that used an analytical approach, it was unclear which specific method was used to compute the sample size.

The ordinal outcome was dichotomised for analysis in 47 ( \(33\%\) ) studies. Some justifications for the dichotomisation of the ordinal outcome included that it represented a clinically meaningful effect and/or that it was common in the analysis of the outcome in similar studies (reported in 24 studies), that the dichotomised outcome represented an agreeable endpoint based on feedback between clinicians and/or patients and families (two studies), or that the assumptions of the statistical model for the categorical outcome were violated (reported in three studies).

There were a variety of target parameters used for the ordinal outcomes. In 130 studies, the target parameter could be determined; however, 59 of these studies ( \(45\%\) ) did not clearly or explicitly define the target parameter of interest. Of those where the target parameter could be determined based on the information provided in the manuscript (e.g. since it was not reported), an OR was the most common target parameter ( \(78, 54\%\) ), followed by a risk difference ( \(31, 22\%\) ). A difference in mean or median was the target parameter in 11 (8%) and 8 (6%) studies respectively. There were 14 ( \(10\%\) ) studies that did not estimate a target parameter. This was either because the study was descriptive in nature, the analysis used a non-parametric procedure, or the target parameter could not be determined (or some combination thereof).

Statistical methods and assumptions

There was a variety of descriptive measures used to summarise the distribution of the ordinal outcome by intervention groups (Table 5 ). The most common descriptive statistics were frequencies and/or percentages in each category of the ordinal outcome ( \(116, 81\%\) ), followed by the median score across all categories ( \(33, 23\%\) ) and IQRs ( \(31, 22\%\) ). The mean and standard deviation across the categories of the ordinal outcome were only summarised in 16 (11%) and 10 (7%) studies respectively.

Many different statistical methods were used to analyse the ordinal outcome (Table 5 ). The PO model was the most common statistical method used to analyse the ordinal outcome (64, \(44\%\) ) that was used to estimate a proportional OR in 62 studies. In studies that used a PO model for the analysis, the interpretation of the target parameter varied between studies (see Additional file 3 ). The most frequent definition used was that the proportional OR represented an ordinal shift in the distribution of ordinal scale scores toward a better outcome in the intervention relative to the control group ( \(12, 19\%\) ). When the outcome was dichotomised, logistic regression was used in 16 studies ( \(11\%\) of all studies) that usually estimated an OR or a risk difference using g-computation. Seven studies estimated a risk difference or risk ratio using binomial regression. Studies also calculated and reported a risk difference with corresponding \(95\%\) confidence intervals estimated using methods such as the Wald method or bootstrapping ( \(31, 22\%\) ). There were 19 (13%) studies that used a non-parametric method to analyse the ordinal outcome (either dichotomised or not), including the Cochran-Mantel-Haenszel test ( \(15, 10\%\) ) to estimate an odds ratio, the Wilcoxon test ( \(14, 10\%\) ), of which no study reported a concordance probability as the target parameter, or the Fisher’s exact or Chi-Square test (12, \(8\%\) ). Other statistical methods that were used were the Hodges-Lehmann estimator, used to estimate a median difference ( \(3, 2\%\) ) and the Van Elteren test ( \(2, 1\%\) ), an extension of the Wilcoxon test for comparing treatments in a stratified experiment. Linear regression was used in 16 ( \(11\%\) ) studies that tended to estimate a mean or risk difference (despite the model having an unbounded support).

The majority of studies ( \(86, 60\%\) ) did not explicitly check the validity of the assumptions for the statistical method(s) used. For example, no study that analysed the ordinal outcome using linear regression commented on the appropriateness of assigning specific numbers of the outcome categories. Among the 64 studies that used a PO model, 20 (31%) did not report whether the assumption of PO was satisfied. Overall, there were 46 studies that reported checking key modelling assumptions; however, the method that was used to check these assumptions were not reported in 6 ( \(13\%)\) of these studies. The most common way to verify model assumptions was to use statistical methods ( \(31, 67\%\) ), followed by graphical methods ( \(2, 4\%\) ).

Among the 44 studies that assessed the validity of the PO assumption for a PO model, 13 studies ( \(30\%\) ) used a likelihood ratio test, 10 studies ( \(23\%\) ) used the Brant test, and 10 studies ( \(23\%\) ) also used the Score test. Six ( \(14\%\) ) studies assessed the robustness of the PO assumption by fitting a logistic regression model to every level of the ordinal outcome across the scale, in which the OR for each dichotomous break was presented. Two studies assessed the PO assumption using graphical methods, which plotted either the inverse cumulative log odds or the empirical cumulative log odds. It was unclear which method was used to assess the PO assumption in ten studies that reported to have checked the assumption.

There were 12 studies ( \(8\%\) ) that reported using a different statistical method than originally planned. Ten of these studies had originally planned to use a PO model, but the PO assumption was determined to have been violated and an alternative method was chosen. One study removed the covariate that was reported to have violated the PO assumption and still used a PO model to analyse the outcome. Two studies used an unconstrained PPO model that reported an adjusted OR for each binary split of the ordinal outcome. Three studies used a Wilcoxon test, with one study stratifying by a baseline covariate that violated the PO assumption. Another study dichotomised the ordinal outcome for the analysis. One study used a Van Elteren test that estimated a median difference (which inappropriately assumes that there is an equal distance between proximate categories), another used a Poisson model with robust standard errors, and one study retained the analysis despite the violation in PO. Notably, a PPO model was not reported to have been used in studies that reported that a covariate other than the treatment violated the PO assumption. Seven studies also did not report which covariate(s) violated the PO assumption.

Frequentist inference was the most common framework for conducting the analysis (133, 92%), with Bayesian methods being used in eight (6%) studies (where two studies used both), of which all eight studies used an adaptive design. Of those using Bayesian methods, seven studies used a Bayesian PO model for analysis. Of these studies, four used a Dirichlet prior distribution to model the baseline probabilities, and three used a normally distributed prior on the proportional log-OR scale. Two of these studies reported to use the median proportional OR with corresponding \(95\%\) credible interval, while one study reported the mean proportional OR. Three studies reported that the models were fitted with the use of a Markov-chain Monte Carlo algorithm with either 10, 000 (one study) or 100, 000 (two studies) samples from the joint posterior distribution. No study reported how the goodness-of-fit of the model was assessed.

For the 38 studies that collected repeated measurements on the ordinal outcome, 18 adjusted for the baseline measurement ( \(47\%\) ), 14 used mixed effects models ( \(37\%\) ), and four used generalised estimated equations ( \(11\%\) ) to capture the correlation among the repeated measures for an individual.

A range of statistical packages were used for the analysis of the ordinal outcome, with SAS ( \(81, 56\%\) ) and R ( \(35, 24\%\) ) being most common. Twelve ( \(8\%\) ) studies did not report the software used.

This review has provided an overview of how ordinal outcomes are used and analysed in contemporary RCTs. We describe the insight this review has provided on the study design, statistical analyses and reporting of trials using ordinal outcomes.

Target parameter

The target parameter of interest is an important consideration when planning any trial and should be aligned with the research question [ 12 , 13 ]. The most common target parameter in this review was an OR, either for a dichotomised version of the ordinal outcome or in an analysis that used the ordinal scale. When an ordinal analysis was used, it was common that the target parameter was a proportional OR, although there was variation in the interpretation of this parameter between studies. We found that it was most common to interpret the proportional OR as an average shift in the distribution of the ordinal scale scores toward a better outcome in the intervention, relative to the comparator(s) [ 19 , 35 , 188 , 189 ]. In the studies that dichotomised the ordinal outcome, many lacked justification for doing so and, in one case, dichotomisation was used only due to the violation of PO, despite the fact that this changed the target parameter.

Some studies in our review treated the ordinal outcome as if it were continuous, and used a difference in means or medians as the target parameter. These quantities do not represent a clinically meaningful effect when the outcome is ordinal, since proximate categories in the scale are not necessarily separated by a quantifiable or equal distance, which can affect the translation of the trial results into practice. If a study is to use a mean difference then the researchers should justify the appropriateness of assigning specific numbers used to the ordinal outcome categories.

The target parameter and statistical method used to estimate it could not be determined in some studies. Notably, the definition of the target parameter was not explicitly defined in almost half of the studies, despite the current recommendations on the importance of clearly defining the estimand of interest, one component of which is the target parameter [ 12 , 13 ]. Furthermore, there is a lack of clarity in defining the target parameter when a PO model was used, despite the interpretation being analogous to the OR for a binary outcome, but applying to an interval of the ordinal scale instead of a single value. Consistency in the definition of a target parameter in RCTs can allow easy interpretation for clinicians and applied researchers. Explicit definition of the target parameter of interest is essential for readers to understand the interpretation of a clinically meaningful treatment effect, and also reflects the present push within clinical research with regards to estimands [ 12 , 13 ].

Statistical methods

It is important to summarise the distribution of the outcome by intervention group in any RCT. When the outcome is ordinal, frequencies and percentages in each category can provide a useful summary of this distribution. Most studies in this review reported frequencies and percentages in each category, although some studies that dichotomised the outcome only reported these summaries for the dichotomised scale. Some studies reported means and standard deviations across the categories which, as mentioned previously, may not have a valid interpretation.

Although there are a range of statistical methods that can be used to analyse an ordinal outcome, we found that the PO model was the most commonly used. This is likely because the PO model is relatively well-known among statisticians and is quite straightforward to fit in most statistical packages, and it possesses the desirable properties of palindromic invariance and invariance under collapsibility. However, when using this approach to estimate a specific treatment effect across all levels of the outcome, it is important to assess and report whether the PO assumption has been met when the aim is to estimate the treatment effect across the different categories or to estimate predicted probabilities in each category. The validity of the PO assumption is less important when the objective is to understand whether one treatment is ‘better’ on average compared to a comparator. In this review, it was common for studies that used a PO model to define the target parameter that related to a treatment benefiting patients with regard to every level of the outcome scale. However, only 44 out of 64 studies reported to have checked the PO assumption, which highlights the deficiency in this practice. Statistical methods were commonly used to assess the PO assumption, although it may be preferable to avoid hypothesis testing when assessing the PO assumption, particularly with small sample sizes, as these statistical tests can have poor statistical power [ 22 , 190 ]. Also, researchers should keep in mind that when the PO assumption is tested, the type I error of the analysis may change and that p -values and confidence intervals based on the updated model ignore the model-fitting uncertainty [ 191 ].

When the PO assumption was violated, a PPO model was rarely used, and instead baseline covariates were removed from the model to address the departure to PO. The fact that the PPO model is underused could be due to a lack of knowledge that such models exist and can be used to address violations in PO. Such a model could have been particularly useful in these studies that had only covariates other than the treatment variable that violated the PO assumption, as the PPO model could have been used to estimate a single proportional OR for the treatment effect. Of note, however, is that an unconstrained PPO model does not necessarily require ordinality as the categories can be arranged and the model fit would be hardly affected [ 192 ], and that estimated probabilities can be negative [ 193 ].

There are other methods that can be used to assess the validity of the PO assumption, such as plotting the differences in predicted log-odds between different categories of the ordinal outcome that should be parallel [ 16 ]. Another option is to fit a logistic regression model to every level of the ordinal outcome across the scale and compare the estimated ORs and corresponding confidence interval for each binary split of the ordinal outcome or simulating predictive distributions. However, estimating separate ORs in this way can be inefficient, particularly when the ordinal outcome has a high number of categories. Arguably, more important than assessing the validity of the PO assumption is to assess the impact of making compared to not making the assumption. If the treatment effect goes in the same direction across each category of the ordinal scale and the objective is to simply understand whether one treatment is better overall, then departures from PO may not be important. If, however, the interest is in estimating a treatment effect for every level of the ordinal outcome and/or the treatment has a detrimental effect for one end of the ordinal scale but a beneficial effect for the remaining categories, there should be careful consideration as to the validity to the type I and II error and the treatment effect if the PO model is used.

Finally, a handful of studies also used the Wilcoxon, Chi-Square, or Fisher’s exact test (the latter being too conservative [ 194 ] and potentially providing misleading results), where commonly only a p -value, not a target parameter, was reported when these methods were used. The lack of a target parameter for the treatment effect can make it difficult for clinicians to translate the results to practice.

Strengths and limitations

The strengths of this study are that we present a review of a large number of RCTs that used ordinal outcomes published in four highly ranked medical journals to highlight the current state of practice for analysing ordinal outcomes. The screening and data extraction process was conducted systematically, and pilot tests and double data extraction ensured the consistency and reliability of the extracted data. The PRISMA-ScR checklist was used to ensure that reporting has been conducted to the highest standard.

This review does, however, have limitations. The restriction to the PubMed database and four highly ranked medical journals may affect the generalisability of this review. We made this decision given the scoping nature of the review, to ensure reproducibility and to ensure that the total number of studies included in the review was manageable. We also aimed to include studies that are likely to reflect best practice of how research using ordinal outcomes is being conducted and reported upon at present. Given the selected journals represent highly ranked medical journals, these findings are likely to reflect the best-case scenario given these journals' reputation for rigour. In addition, our search strategy may have missed certain phrases or variants (particularly related to an ordinal outcome); however, we attempted to mitigate this through our piloting phase. Finally, we also did not review the protocol papers of the trials that may have included additional information related to the statistical methodology. This includes methods that were planned to be used to assess the PO assumption, and any alternative methods that were to be used instead.

Implications of this research

This review has implications for researchers designing RCTs that use an ordinal outcome. Although the majority of studies included in this review were in the fields of neurology and infectious diseases, the results of this review would apply to RCTs in all medical fields that use an ordinal outcome. We have shown that there is substantial variation in the analysis and reporting of ordinal outcomes in practice. Our results suggest that researchers should carefully consider the target parameter of interest and explicitly report what the target parameter represents; this is particularly important for an ordinal outcome which can be unfamiliar to readers. Defining the target parameter upfront will help to ensure that appropriate analytical methods are used to analyse the ordinal outcome and make transparent the assumptions the researchers are willing to make.

Our review also highlights the need for careful assessment and reporting of the validity of the model assumptions made during the analysis of an ordinal outcome. Doing so will ensure that robust statistical methods that align with the research question and categorical nature of the ordinal outcome are used to estimate a valid, clinically relevant target parameter that can be translated to practice.

Availability of data and materials

The datasets and code generated and/or analysed during the current study are available on GitHub [ 195 ].

Abbreviations

Randomised controlled trial

Proportional odds

Partial proportional odds

Statistical analysis plan

Velleman PF, Wilkinson L. Nominal, ordinal, interval, and ratio typologies are misleading. Am Stat. 1993;47(1):65–72.

Article   Google Scholar  

MacKenzie CR, Charlson ME. Standards for the use of ordinal scales in clinical trials. Br Med J (Clin Res Ed). 1986;292(6512):40–3.

Article   CAS   PubMed   Google Scholar  

Banks JL, Marotta CA. Outcomes validity and reliability of the modified Rankin scale: implications for stroke clinical trials: a literature review and synthesis. Stroke. 2007;38(3):1091–6.

Article   PubMed   Google Scholar  

de la Ossa NP, Abilleira S, Jovin TG, García-Tornel Á, Jimenez X, Urra X, et al. Effect of direct transportation to thrombectomy-capable center vs local stroke center on neurological outcomes in patients with suspected large-vessel occlusion stroke in nonurban areas: the RACECAT randomized clinical Trial. JAMA. 2022;327(18):1782–94.

Hubert GJ, Hubert ND, Maegerlein C, Kraus F, Wiestler H, Müller-Barna P, et al. Association between use of a flying intervention team vs patient interhospital transfer and time to endovascular thrombectomy among patients with acute ischemic stroke in nonurban Germany. JAMA. 2022;327(18):1795–805.

Article   PubMed   PubMed Central   Google Scholar  

Bösel J, Niesen WD, Salih F, Morris NA, Ragland JT, Gough B, et al. Effect of early vs standard approach to tracheostomy on functional outcome at 6 months among patients with severe stroke receiving mechanical ventilation: the SETPOINT2 Randomized Clinical Trial. JAMA. 2022;327(19):1899–909.

Wilson L, Boase K, Nelson LD, Temkin NR, Giacino JT, Markowitz AJ, et al. A manual for the glasgow outcome scale-extended interview. J Neurotrauma. 2021;38(17):2435–46.

Marshall JC, Murthy S, Diaz J, Adhikari N, Angus DC, Arabi YM, et al. A minimal common outcome measure set for COVID-19 clinical research. Lancet Infect Dis. 2020;20(8):e192–7.

Article   CAS   Google Scholar  

Lovre D, Bateman K, Sherman M, Fonseca VA, Lefante J, Mauvais-Jarvis F. Acute estradiol and progesterone therapy in hospitalised adults to reduce COVID-19 severity: a randomised control trial. BMJ Open. 2021;11(11):e053684.

Song AT, Rocha V, Mendrone-Júnior A, Calado RT, De Santis GC, Benites BD, et al. Treatment of severe COVID-19 patients with either low-or high-volume of convalescent plasma versus standard of care: a multicenter Bayesian randomized open-label clinical trial (COOP-COVID-19-MCTI). Lancet Reg Health-Am. 2022;10:100216.

PubMed   PubMed Central   Google Scholar  

Mathioudakis AG, Fally M, Hashad R, Kouta A, Hadi AS, Knight SB, et al. Outcomes evaluated in controlled clinical trials on the management of COVID-19: a methodological systematic review. Life. 2020;10(12):350.

Akacha M, Bretz F, Ohlssen D, Rosenkranz G, Schmidli H. Estimands and their role in clinical trials. Stat Biopharm Res. 2017;9(3):268–71.

Mallinckrodt C, Molenberghs G, Lipkovich I, Ratitch B. Estimands, estimators and sensitivity analysis in clinical trials. CRC Press; 2019.

Walker SH, Duncan DB. Estimation of the probability of an event as a function of several independent variables. Biometrika. 1967;54(1–2):167–79.

McCullagh P. Regression models for ordinal data. J R Stat Soc Ser B Methodol. 1980;42(2):109–27.

Google Scholar  

Harrell FE, et al. Regression modeling strategies: with applications to linear models, logistic and ordinal regression, and survival analysis, vol 3. Springer; 2015.

Ananth CV, Kleinbaum DG. Regression models for ordinal responses: a review of methods and applications. Int J Epidemiol. 1997;26(6):1323–33.

Armstrong BG, Sloan M. Ordinal regression models for epidemiologic data. Am J Epidemiol. 1989;129(1):191–204.

Roozenbeek B, Lingsma HF, Perel P, Edwards P, Roberts I, Murray GD, et al. The added value of ordinal analysis in clinical trials: an example in traumatic brain injury. Crit Care. 2011;15(3):1–7.

Breheny P. Proportional odds models. 2015. MyWeb. https://myweb.uiowa.edu/pbreheny/uk/teaching/760-s13/notes/4-23.pdf .

Abreu MNS, Siqueira AL, Cardoso CS, Caiaffa WT. Ordinal logistic regression models: application in quality of life studies. Cad Saúde Pública. 2008;24:s581–91.

Peterson B, Harrell FE Jr. Partial proportional odds models for ordinal response variables. J R Stat Soc: Ser C: Appl Stat. 1990;39(2):205–17.

Fullerton AS. A conceptual framework for ordered logistic regression models. Sociol Methods Res. 2009;38(2):306–47.

Senn S, Julious S. Measurement in clinical trials: a neglected issue for statisticians? Stat Med. 2009;28(26):3189–209.

Maas AI, Steyerberg EW, Marmarou A, McHugh GS, Lingsma HF, Butcher I, et al. IMPACT recommendations for improving the design and analysis of clinical trials in moderate to severe traumatic brain injury. Neurotherapeutics. 2010;7:127–34.

McFadden D, et al. Conditional logit analysis of qualitative choice behavior.  Front Econ. 1973;105–142.

Wilcoxon F. Individual comparisons by ranking methods. Springer; 1992.

Liu Q, Shepherd BE, Li C, Harrell FE Jr. Modeling continuous response variables using ordinal regression. Stat Med. 2017;36(27):4316–35.

Fay MP, Brittain EH, Shih JH, Follmann DA, Gabriel EE. Causal estimands and confidence intervals associated with Wilcoxon-Mann-Whitney tests in randomized experiments. Stat Med. 2018;37(20):2923–37.

De Neve J, Thas O, Gerds TA. Semiparametric linear transformation models: effect measures, estimators, and applications. Stat Med. 2019;38(8):1484–501.

Ganesh A, Luengo-Fernandez R, Wharton RM, Rothwell PM. Ordinal vs dichotomous analyses of modified Rankin Scale, 5-year outcome, and cost of stroke. Neurology. 2018;91(21):e1951–60.

French B, Shotwell MS. Regression models for ordinal outcomes. JAMA. 2022;328(8):772–3.

Bath PM, Geeganage C, Gray LJ, Collier T, Pocock S. Use of ordinal outcomes in vascular prevention trials: comparison with binary outcomes in published trials. Stroke. 2008;39(10):2817–23.

Scott SC, Goldberg MS, Mayo NE. Statistical assessment of ordinal outcomes in comparative studies. J Clin Epidemiol. 1997;50(1):45–55.

McHugh GS, Butcher I, Steyerberg EW, Marmarou A, Lu J, Lingsma HF, et al. A simulation study evaluating approaches to the analysis of ordinal outcome data in randomized controlled trials in traumatic brain injury: results from the IMPACT Project. Clin Trials. 2010;7(1):44–57.

DeSantis SM, Lazaridis C, Palesch Y, Ramakrishnan V. Regression analysis of ordinal stroke clinical trial outcomes: an application to the NINDS t-PA trial. Int J Stroke. 2014;9(2):226–31.

Selman CJ, Lee KJ, Whitehead CL, Manley BJ, Mahar RK. Statistical analyses of ordinal outcomes in randomised controlled trials: protocol for a scoping review. Trials. 2023;24(1):1–7.

Tricco AC, Lillie E, Zarin W, O’Brien KK, Colquhoun H, Levac D, et al. PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med. 2018;169(7):467–73.

Bell ML, Fiero M, Horton NJ, Hsu CH. Handling missing data in RCTs; a review of the top medical journals. BMC Med Res Methodol. 2014;14(1):1–8.

Berwanger O, Ribeiro RA, Finkelsztejn A, Watanabe M, Suzumura EA, Duncan BB, et al. The quality of reporting of trial abstracts is suboptimal: survey of major general medical journals. J Clin Epidemiol. 2009;62(4):387–92.

Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al.. Cochrane handbook for systematic reviews of interventions. John Wiley & Sons; 2019.

Veritas Health Innovation. Covidence systematic review software. Melbourne; 2022.

StataCorp L. Stata statistical software: Release 17 (2021). College Station: StataCorp LP; 2021.

Hanley DF, Lane K, McBee N, Ziai W, Tuhrim S, Lees KR, et al. Thrombolytic removal of intraventricular haemorrhage in treatment of severe stroke: results of the randomised, multicentre, multiregion, placebo-controlled CLEAR III trial. Lancet. 2017;389(10069):603–11. https://doi.org/10.1016/S0140-6736(16)32410-2 .

Article   CAS   PubMed   PubMed Central   Google Scholar  

Nangia J, Wang T, Osborne C, Niravath P, Otte K, Papish S, et al. Effect of a scalp cooling device on alopecia in women undergoing chemotherapy for breast cancer: the SCALP randomized clinical trial. JAMA. 2017;317(6):596–605. https://doi.org/10.1001/jama.2016.20939 . United States.

Ruzicka T, Hanifin JM, Furue M, Pulka G, Mlynarczyk I, Wollenberg A, et al. Anti-interleukin-31 receptor A antibody for atopic dermatitis. N Engl J Med. 2017;376(9):826–35. https://doi.org/10.1056/NEJMoa1606490 . United States.

Németh G, Laszlovszky I, Czobor P, Szalai E, Szatmári B, Harsányi J, et al. Cariprazine versus risperidone monotherapy for treatment of predominant negative symptoms in patients with schizophrenia: a randomised, double-blind, controlled trial. Lancet. 2017;389(10074):1103–13. https://doi.org/10.1016/S0140-6736(17)30060-0 . England.

Mathieson S, Maher CG, McLachlan AJ, Latimer J, Koes BW, Hancock MJ, et al. Trial of pregabalin for acute and chronic sciatica. N Engl J Med. 2017;376(12):1111–20. https://doi.org/10.1056/NEJMoa1614292 . United States.

Baud O, Trousson C, Biran V, Leroy E, Mohamed D, Alberti C. Association between early low-dose hydrocortisone therapy in extremely preterm neonates and neurodevelopmental outcomes at 2 years of age. JAMA. 2017;317(13):1329–37. https://doi.org/10.1001/jama.2017.2692 . United States.

van den Berg LA, Dijkgraaf MG, Berkhemer OA, Fransen PS, Beumer D, Lingsma HF, et al. Two-year outcome after endovascular treatment for acute ischemic stroke. N Engl J Med. 2017;376(14):1341–9. https://doi.org/10.1056/NEJMoa1612136 . United States.

Kaufman J, Fitzpatrick P, Tosif S, Hopper SM, Donath SM, Bryant PA, et al. Faster clean catch urine collection (Quick-Wee method) from infants: randomised controlled trial. BMJ. 2017;357:j1341. https://doi.org/10.1136/bmj.j1341 .

Costa Leme A, Hajjar LA, Volpe MS, Fukushima JT, De Santis Santiago RR, Osawa EA, et al. Effect of intensive vs moderate alveolar recruitment strategies added to lung-protective ventilation on postoperative pulmonary complications: a randomized clinical trial. JAMA. 2017;317(14):1422–32. https://doi.org/10.1001/jama.2017.2297 . United States.

Breitenstein C, Grewe T, Flöel A, Ziegler W, Springer L, Martus P, et al. Intensive speech and language therapy in patients with chronic aphasia after stroke: a randomised, open-label, blinded-endpoint, controlled trial in a health-care setting. Lancet. 2017;389(10078):1528–38. https://doi.org/10.1016/S0140-6736(17)30067-3 . England.

Wechsler ME, Akuthota P, Jayne D, Khoury P, Klion A, Langford CA, et al. Mepolizumab or placebo for eosinophilic granulomatosis with polyangiitis. N Engl J Med. 2017;376(20):1921–32. https://doi.org/10.1056/NEJMoa1702079 .

Devinsky O, Cross JH, Laux L, Marsh E, Miller I, Nabbout R, et al. Trial of cannabidiol for drug-resistant seizures in the Dravet syndrome. N Engl J Med. 2017;376(21):2011–20. https://doi.org/10.1056/NEJMoa1611618 . United States.

Anderson CS, Arima H, Lavados P, Billot L, Hackett ML, Olavarría VV, et al. Cluster-randomized, crossover trial of head positioning in acute stroke. N Engl J Med. 2017;376(25):2437–47. https://doi.org/10.1056/NEJMoa1615715 . United States.

Juch JNS, Maas ET, Ostelo RWJG, Groeneweg JG, Kallewaard JW, Koes BW, et al. Effect of radiofrequency denervation on pain intensity among patients with chronic low back pain: the Mint randomized clinical trials. JAMA. 2017;318(1):68–81. https://doi.org/10.1001/jama.2017.7918 .

Mohamed S, Johnson GR, Chen P, Hicks PB, Davis LL, Yoon J, et al. Effect of antidepressant switching vs augmentation on remission among patients with major depressive disorder unresponsive to antidepressant treatment: the VAST-D randomized clinical trial. JAMA. 2017;318(2):132–45. https://doi.org/10.1001/jama.2017.8036 .

Kanes S, Colquhoun H, Gunduz-Bruce H, Raines S, Arnold R, Schacterle A, et al. Brexanolone (SAGE-547 injection) in post-partum depression: a randomised controlled trial. Lancet. 2017;390(10093):480–9. https://doi.org/10.1016/S0140-6736(17)31264-3 . England.

Lapergue B, Blanc R, Gory B, Labreuche J, Duhamel A, Marnat G, et al. Effect of endovascular contact aspiration vs stent retriever on revascularization in patients with acute ischemic stroke and large vessel occlusion: the ASTER randomized clinical trial. JAMA. 2017;318(5):443–52. https://doi.org/10.1001/jama.2017.9644 .

Lindley RI, Anderson CS, Billot L, Forster A, Hackett ML, Harvey LA, et al. Family-led rehabilitation after stroke in India (ATTEND): a randomised controlled trial. Lancet. 2017;390(10094):588–99. https://doi.org/10.1016/S0140-6736(17)31447-2 . England.

Berlowitz DR, Foy CG, Kazis LE, Bolin LP, Conroy MB, Fitzpatrick P, et al. Effect of intensive blood-pressure treatment on patient-reported outcomes. N Engl J Med. 2017;377(8):733–44. https://doi.org/10.1056/NEJMoa1611179 .

Hui D, Frisbee-Hume S, Wilson A, Dibaj SS, Nguyen T, De La Cruz M, et al. Effect of lorazepam with haloperidol vs haloperidol alone on agitated delirium in patients with advanced cancer receiving palliative care: a randomized clinical trial. JAMA. 2017;318(11):1047–56. https://doi.org/10.1001/jama.2017.11468 .

Roffe C, Nevatte T, Sim J, Bishop J, Ives N, Ferdinand P, et al. Effect of routine low-dose oxygen supplementation on death and disability in adults with acute stroke: the stroke oxygen study randomized clinical trial. JAMA. 2017;318(12):1125–35. https://doi.org/10.1001/jama.2017.11463 .

Dwivedi R, Ramanujam B, Chandra PS, Sapra S, Gulati S, Kalaivani M, et al. Surgery for drug-resistant epilepsy in children. N Engl J Med. 2017;377(17):1639–47. https://doi.org/10.1056/NEJMoa1615335 . United States.

Nogueira RG, Jadhav AP, Haussen DC, Bonafe A, Budzik RF, Bhuva P, et al. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct. N Engl J Med. 2018;378(1):11–21. https://doi.org/10.1056/NEJMoa1706442 . United States.

Zheng MX, Hua XY, Feng JT, Li T, Lu YC, Shen YD, et al. Trial of Contralateral seventh cervical nerve transfer for spastic arm paralysis. N Engl J Med. 2018;378(1):22–34. https://doi.org/10.1056/NEJMoa1615208 . United States.

Atri A, Frölich L, Ballard C, Tariot PN, Molinuevo JL, Boneva N, et al. Effect of idalopirdine as adjunct to cholinesterase inhibitors on change in cognition in patients with Alzheimer disease: three randomized clinical trials. JAMA. 2018;319(2):130–42. https://doi.org/10.1001/jama.2017.20373 .

Bassler D, Shinwell ES, Hallman M, Jarreau PH, Plavka R, Carnielli V, et al. Long-term effects of inhaled budesonide for bronchopulmonary dysplasia. N Engl J Med. 2018;378(2):148–57. https://doi.org/10.1056/NEJMoa1708831 . United States.

Raskind MA, Peskind ER, Chow B, Harris C, Davis-Karim A, Holmes HA, et al. Trial of prazosin for post-traumatic stress disorder in military veterans. N Engl J Med. 2018;378(6):507–17. https://doi.org/10.1056/NEJMoa1507598 . United States.

Albers GW, Marks MP, Kemp S, Christensen S, Tsai JP, Ortega-Gutierrez S, et al. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging. N Engl J Med. 2018;378(8):708–18. https://doi.org/10.1056/NEJMoa1713973 .

Bath PM, Woodhouse LJ, Appleton JP, Beridze M, Christensen H, Dineen RA, et al. Antiplatelet therapy with aspirin, clopidogrel, and dipyridamole versus clopidogrel alone or aspirin and dipyridamole in patients with acute cerebral ischaemia (TARDIS): a randomised, open-label, phase 3 superiority trial. Lancet. 2018;391(10123):850–9. https://doi.org/10.1016/S0140-6736(17)32849-0 .

Krebs EE, Gravely A, Nugent S, Jensen AC, DeRonne B, Goldsmith ES, et al. Effect of opioid vs nonopioid medications on pain-related function in patients with chronic back pain or hip or knee osteoarthritis pain: the SPACE randomized clinical trial. JAMA. 2018;319(9):872–82. https://doi.org/10.1001/jama.2018.0899 .

Campbell BCV, Mitchell PJ, Churilov L, Yassi N, Kleinig TJ, Dowling RJ, et al. Tenecteplase versus alteplase before thrombectomy for ischemic stroke. N Engl J Med. 2018;378(17):1573–82. https://doi.org/10.1056/NEJMoa1716405 . United States.

Mellor R, Bennell K, Grimaldi A, Nicolson P, Kasza J, Hodges P, et al. Education plus exercise versus corticosteroid injection use versus a wait and see approach on global outcome and pain from gluteal tendinopathy: prospective, single blinded, randomised clinical trial. BMJ. 2018;361. https://doi.org/10.1136/bmj.k1662 .

Sprigg N, Flaherty K, Appleton JP, Al-Shahi Salman R, Bereczki D, Beridze M, et al. Tranexamic acid for hyperacute primary IntraCerebral Haemorrhage (TICH-2): an international randomised, placebo-controlled, phase 3 superiority trial. Lancet. 2018;391(10135):2107–15. https://doi.org/10.1016/S0140-6736(18)31033-X .

Jolly K, Sidhu MS, Hewitt CA, Coventry PA, Daley A, Jordan R, et al. Self management of patients with mild COPD in primary care: randomised controlled trial. BMJ. 2018;361. https://doi.org/10.1136/bmj.k2241 .

Brock PR, Maibach R, Childs M, Rajput K, Roebuck D, Sullivan MJ, et al. Sodium thiosulfate for protection from cisplatin-induced hearing loss. N Engl J Med. 2018;378(25):2376–85. https://doi.org/10.1056/NEJMoa1801109 .

Khatri P, Kleindorfer DO, Devlin T, Sawyer RN Jr, Starr M, Mejilla J, et al. Effect of alteplase vs aspirin on functional outcome for patients with acute ischemic stroke and minor nondisabling neurologic deficits: the PRISMS randomized clinical trial. JAMA. 2018;320(2):156–66. https://doi.org/10.1001/jama.2018.8496 .

Wang Y, Li Z, Zhao X, Wang C, Wang X, Wang D, et al. Effect of a multifaceted quality improvement intervention on hospital personnel adherence to performance measures in patients with acute ischemic stroke in china: a randomized clinical trial. JAMA. 2018;320(3):245–54. https://doi.org/10.1001/jama.2018.8802 . United States.

Fossat G, Baudin F, Courtes L, Bobet S, Dupont A, Bretagnol A, et al. Effect of in-bed leg cycling and electrical stimulation of the quadriceps on global muscle strength in critically ill adults: a randomized clinical trial. JAMA. 2018;320(4):368–78. https://doi.org/10.1001/jama.2018.9592 .

Thomalla G, Simonsen CZ, Boutitie F, Andersen G, Berthezene Y, Cheng B, et al. MRI-guided thrombolysis for stroke with unknown time of onset. N Engl J Med. 2018;379(7):611–22. https://doi.org/10.1056/NEJMoa1804355 . United States.

Perkins GD, Ji C, Deakin CD, Quinn T, Nolan JP, Scomparin C, et al. A randomized trial of epinephrine in out-of-hospital cardiac arrest. N Engl J Med. 2018;379(8):711–21. https://doi.org/10.1056/NEJMoa1806842 . United States.

Wang HE, Schmicker RH, Daya MR, Stephens SW, Idris AH, Carlson JN, et al. Effect of a strategy of initial laryngeal tube insertion vs endotracheal intubation on 72-hour survival in adults with out-of-hospital cardiac arrest: a randomized clinical trial. JAMA. 2018;320(8):769–78. https://doi.org/10.1001/jama.2018.7044 .

Benger JR, Kirby K, Black S, Brett SJ, Clout M, Lazaroo MJ, et al. Effect of a strategy of a supraglottic airway device vs tracheal intubation during out-of-hospital cardiac arrest on functional outcome: the AIRWAYS-2 randomized clinical trial. JAMA. 2018;320(8):779–91. https://doi.org/10.1001/jama.2018.11597 .

Meltzer-Brody S, Colquhoun H, Riesenberg R, Epperson CN, Deligiannidis KM, Rubinow DR, et al. Brexanolone injection in post-partum depression: two multicentre, double-blind, randomised, placebo-controlled, phase 3 trials. Lancet. 2018;392(10152):1058–70. https://doi.org/10.1016/S0140-6736(18)31551-4 . England.

Cooper DJ, Nichol AD, Bailey M, Bernard S, Cameron PA, Pili-Floury S, et al. Effect of early sustained prophylactic hypothermia on neurologic outcomes among patients with severe traumatic brain injury: the POLAR randomized clinical trial. JAMA. 2018;320(21):2211–20. https://doi.org/10.1001/jama.2018.17075 .

Bonell C, Allen E, Warren E, McGowan J, Bevilacqua L, Jamal F, et al. Effects of the Learning Together intervention on bullying and aggression in English secondary schools (INCLUSIVE): a cluster randomised controlled trial. Lancet. 2018;392(10163):2452–64. https://doi.org/10.1016/S0140-6736(18)31782-3 .

Stunnenberg BC, Raaphorst J, Groenewoud HM, Statland JM, Griggs RC, Woertman W, et al. Effect of mexiletine on muscle stiffness in patients with nondystrophic myotonia evaluated using aggregated N-of-1 trials. JAMA. 2018;320(22):2344–53. https://doi.org/10.1001/jama.2018.18020 .

Burt RK, Balabanov R, Burman J, Sharrack B, Snowden JA, Oliveira MC, et al. Effect of nonmyeloablative hematopoietic stem cell transplantation vs continued disease-modifying therapy on disease progression in patients with relapsing-remitting multiple sclerosis: a randomized clinical trial. JAMA. 2019;321(2):165–74. https://doi.org/10.1001/jama.2018.18743 .

Dennis M, Mead G, Forbes J, Graham C, Hackett M, Hankey GJ, et al. Effects of fluoxetine on functional outcomes after acute stroke (FOCUS): a pragmatic, double-blind, randomised, controlled trial. Lancet. 2019;393(10168):265–74. https://doi.org/10.1016/S0140-6736(18)32823-X .

Anderson CS, Huang Y, Lindley RI, Chen X, Arima H, Chen G, et al. Intensive blood pressure reduction with intravenous thrombolysis therapy for acute ischaemic stroke (ENCHANTED): an international, randomised, open-label, blinded-endpoint, phase 3 trial. Lancet. 2019;393(10174):877–88. https://doi.org/10.1016/S0140-6736(19)30038-8 . England.

Basner M, Asch DA, Shea JA, Bellini LM, Carlin M, Ecker AJ, et al. Sleep and alertness in a duty-hour flexibility trial in internal medicine. N Engl J Med. 2019;380(10):915–23. https://doi.org/10.1056/NEJMoa1810641 .

Bath PM, Scutt P, Anderson CS, Appleton JP, Berge E, Cala L, et al. Prehospital transdermal glyceryl trinitrate in patients with ultra-acute presumed stroke (RIGHT-2): an ambulance-based, randomised, sham-controlled, blinded, phase 3 trial. Lancet. 2019;393(10175):1009–20. https://doi.org/10.1016/S0140-6736(19)30194-1 .

Hanley DF, Thompson RE, Rosenblum M, Yenokyan G, Lane K, McBee N, et al. Efficacy and safety of minimally invasive surgery with thrombolysis in intracerebral haemorrhage evacuation (MISTIE III): a randomised, controlled, open-label, blinded endpoint phase 3 trial. Lancet. 2019;393(10175):1021–32. https://doi.org/10.1016/S0140-6736(19)30195-3 .

Turk AS 3rd, Siddiqui A, Fifi JT, De Leacy RA, Fiorella DJ, Gu E, et al. Aspiration thrombectomy versus stent retriever thrombectomy as first-line approach for large vessel occlusion (COMPASS): a multicentre, randomised, open label, blinded outcome, non-inferiority trial. Lancet. 2019;393(10175):998–1008.  https://doi.org/10.1016/S0140-6736(19)30297-1 . England.

Ma H, Campbell BCV, Parsons MW, Churilov L, Levi CR, Hsu C, et al. Thrombolysis guided by perfusion imaging up to 9 hours after onset of stroke. N Engl J Med. 2019;380(19):1795–803. https://doi.org/10.1056/NEJMoa1813046 . United States.

Fischer K, Al-Sawaf O, Bahlo J, Fink AM, Tandon M, Dixon M, et al. Venetoclax and obinutuzumab in patients with CLL and coexisting conditions. N Engl J Med. 2019;380(23):2225–36. https://doi.org/10.1056/NEJMoa1815281 . United States.

Shehabi Y, Howe BD, Bellomo R, Arabi YM, Bailey M, Bass FE, et al. Early sedation with dexmedetomidine in critically ill patients. N Engl J Med. 2019;380(26):2506–17. https://doi.org/10.1056/NEJMoa1904710 . United States.

Johnston KC, Bruno A, Pauls Q, Hall CE, Barrett KM, Barsan W, et al. Intensive vs standard treatment of hyperglycemia and functional outcome in patients with acute ischemic stroke: the SHINE randomized clinical trial. JAMA. 2019;322(4):326–35. https://doi.org/10.1001/jama.2019.9346 .

Widmark A, Gunnlaugsson A, Beckman L, Thellenberg-Karlsson C, Hoyer M, Lagerlund M, et al. Ultra-hypofractionated versus conventionally fractionated radiotherapy for prostate cancer: 5-year outcomes of the HYPO-RT-PC randomised, non-inferiority, phase 3 trial. Lancet. 2019;394(10196):385–95. https://doi.org/10.1016/S0140-6736(19)31131-6 . England.

Pittock SJ, Berthele A, Fujihara K, Kim HJ, Levy M, Palace J, et al. Eculizumab in aquaporin-4-positive neuromyelitis optica spectrum disorder. N Engl J Med. 2019;381(7):614–25.  https://doi.org/10.1056/NEJMoa1900866 . United States.

Gunduz-Bruce H, Silber C, Kaul I, Rothschild AJ, Riesenberg R, Sankoh AJ, et al. Trial of SAGE-217 in patients with major depressive disorder. N Engl J Med. 2019;381(10):903–11.  https://doi.org/10.1056/NEJMoa1815981 . United States.

Nave AH, Rackoll T, Grittner U, Bläsing H, Gorsler A, Nabavi DG, et al. Physical Fitness Training in Patients with Subacute Stroke (PHYS-STROKE): multicentre, randomised controlled, endpoint blinded trial. BMJ. 2019;366:l5101. https://doi.org/10.1136/bmj.l5101 .

Sands BE, Peyrin-Biroulet L, Loftus EV Jr, Danese S, Colombel JF, Törüner M, et al. Vedolizumab versus adalimumab for moderate-to-severe ulcerative colitis. N Engl J Med. 2019;381(13):1215–26.  https://doi.org/10.1056/NEJMoa1905725 . United States.

Cree BAC, Bennett JL, Kim HJ, Weinshenker BG, Pittock SJ, Wingerchuk DM, et al. Inebilizumab for the treatment of neuromyelitis optica spectrum disorder (N-MOmentum): a double-blind, randomised placebo-controlled phase 2/3 trial. Lancet. 2019;394(10206):1352–63.  https://doi.org/10.1016/S0140-6736(19)31817-3 . England.

Cooper K, Breeman S, Scott NW, Scotland G, Clark J, Hawe J, et al. Laparoscopic supracervical hysterectomy versus endometrial ablation for women with heavy menstrual bleeding (HEALTH): a parallel-group, open-label, randomised controlled trial. Lancet. 2019;394(10207):1425–36. https://doi.org/10.1016/S0140-6736(19)31790-8 .

Reddihough DS, Marraffa C, Mouti A, O’Sullivan M, Lee KJ, Orsini F, et al. Effect of fluoxetine on obsessive-compulsive behaviors in children and adolescents with autism spectrum disorders: a randomized clinical trial. JAMA. 2019;322(16):1561–9. https://doi.org/10.1001/jama.2019.14685 .

John LK, Loewenstein G, Marder A, Callaham ML. Effect of revealing authors’ conflicts of interests in peer review: randomized controlled trial. BMJ. 2019;367. https://doi.org/10.1136/bmj.l5896 .

Yamamura T, Kleiter I, Fujihara K, Palace J, Greenberg B, Zakrzewska-Pniewska B, et al. Trial of satralizumab in neuromyelitis optica spectrum disorder. N Engl J Med. 2019;381(22):2114–24.  https://doi.org/10.1056/NEJMoa1901747 . United States.

Hoskin PJ, Hopkins K, Misra V, Holt T, McMenemin R, Dubois D, et al. Effect of single-fraction vs multifraction radiotherapy on ambulatory status among patients with spinal canal compression from metastatic cancer: the SCORAD randomized clinical trial. JAMA. 2019;322(21):2084–94. https://doi.org/10.1001/jama.2019.17913 .

Lascarrou JB, Merdji H, Le Gouge A, Colin G, Grillet G, Girardie P, et al. Targeted temperature management for cardiac arrest with nonshockable rhythm. N Engl J Med. 2019;381(24):2327–37.  https://doi.org/10.1056/NEJMoa1906661 . United States.

Ständer S, Yosipovitch G, Legat FJ, Lacour JP, Paul C, Narbutt J, et al. Trial of nemolizumab in moderate-to-severe prurigo nodularis. N Engl J Med. 2020;382(8):706–16.  https://doi.org/10.1056/NEJMoa1908316 . United States.

Hill MD, Goyal M, Menon BK, Nogueira RG, McTaggart RA, Demchuk AM, et al. Efficacy and safety of nerinetide for the treatment of acute ischaemic stroke (ESCAPE-NA1): a multicentre, double-blind, randomised controlled trial. Lancet. 2020;395(10227):878–87.  https://doi.org/10.1016/S0140-6736(20)30258-0 . England.

Olsen HT, Nedergaard HK, Strøm T, Oxlund J, Wian KA, Ytrebø LM, et al. Nonsedation or light sedation in critically ill, mechanically ventilated patients. N Engl J Med. 2020;382(12):1103–11.  https://doi.org/10.1056/NEJMoa1906759 . United States.

Campbell BCV, Mitchell PJ, Churilov L, Yassi N, Kleinig TJ, Dowling RJ, et al. Effect of intravenous tenecteplase dose on cerebral reperfusion before thrombectomy in patients with large vessel occlusion ischemic stroke: the EXTEND-IA TNK Part 2 randomized clinical trial. JAMA. 2020;323(13):1257–65. https://doi.org/10.1001/jama.2020.1511 .

Deyle GD, Allen CS, Allison SC, Gill NW, Hando BR, Petersen EJ, et al. Physical therapy versus glucocorticoid injection for osteoarthritis of the knee. N Engl J Med. 2020;382(15):1420–29.  https://doi.org/10.1056/NEJMoa1905877 . United States.

Koblan KS, Kent J, Hopkins SC, Krystal JH, Cheng H, Goldman R, et al. A non-D2-receptor-binding drug for the treatment of schizophrenia. N Engl J Med. 2020;382(16):1497–506.  https://doi.org/10.1056/NEJMoa1911772 . United States.

Cao B, Wang Y, Wen D, Liu W, Wang J, Fan G, et al. A trial of lopinavir-ritonavir in adults hospitalized with severe COVID-19. N Engl J Med. 2020;382(19):1787–99. https://doi.org/10.1056/NEJMoa2001282 .

Wang Y, Zhang D, Du G, Du R, Zhao J, Jin Y, et al. Remdesivir in adults with severe COVID-19: a randomised, double-blind, placebo-controlled, multicentre trial. Lancet. 2020;395(10236):1569–78. https://doi.org/10.1016/S0140-6736(20)31022-9 .

Yang P, Zhang Y, Zhang L, Treurniet KM, Chen W, Peng Y, et al. Endovascular thrombectomy with or without intravenous alteplase in acute stroke. N Engl J Med. 2020;382(21):1981–93.  https://doi.org/10.1056/NEJMoa2001123 . United States.

Martins SO, Mont’Alverne F, Rebello LC, Abud DG, Silva GS, Lima FO, et al. Thrombectomy for stroke in the public health care system of Brazil. N Engl J Med. 2020;382(24):2316–26.  https://doi.org/10.1056/NEJMoa2000120 . United States.

Kabashima K, Matsumura T, Komazaki H, Kawashima M. Trial of nemolizumab and topical agents for atopic dermatitis with pruritus. N Engl J Med. 2020;383(2):141–50.  https://doi.org/10.1056/NEJMoa1917006 . United States.

Johnston SC, Amarenco P, Denison H, Evans SR, Himmelmann A, James S, et al. Ticagrelor and aspirin or aspirin alone in acute ischemic stroke or TIA. N Engl J Med. 2020;383(3):207–17.  https://doi.org/10.1056/NEJMoa1916870 . United States.

Lebwohl MG, Papp KA, Stein Gold L, Gooderham MJ, Kircik LH, Draelos ZD, et al. Trial of roflumilast cream for chronic plaque psoriasis. N Engl J Med. 2020;383(3):229–39.  https://doi.org/10.1056/NEJMoa2000073 . United States.

Simpson EL, Sinclair R, Forman S, Wollenberg A, Aschoff R, Cork M, et al. Efficacy and safety of abrocitinib in adults and adolescents with moderate-to-severe atopic dermatitis (JADE MONO-1): a multicentre, double-blind, randomised, placebo-controlled, phase 3 trial. Lancet. 2020;396(10246):255–66.  https://doi.org/10.1016/S0140-6736(20)30732-7 . England.

Rowell SE, Meier EN, McKnight B, Kannas D, May S, Sheehan K, et al. Effect of out-of-hospital tranexamic acid vs placebo on 6-month functional neurologic outcomes in patients with moderate or severe traumatic brain injury. JAMA. 2020;324(10):961–74. https://doi.org/10.1001/jama.2020.8958 .

van der Vlist AC, van Oosterom RF, van Veldhoven PLJ, Bierma-Zeinstra SMA, Waarsing JH, Verhaar JAN, et al. Effectiveness of a high volume injection as treatment for chronic Achilles tendinopathy: randomised controlled trial. BMJ. 2020;370. https://doi.org/10.1136/bmj.m3027 .

Spinner CD, Gottlieb RL, Criner GJ, Arribas López JR, Cattelan AM, Soriano Viladomiu A, et al. Effect of remdesivir vs standard care on clinical status at 11 days in patients with moderate COVID-19: a randomized clinical trial. JAMA. 2020;324(11):1048–57. https://doi.org/10.1001/jama.2020.16349 .

Horne AW, Vincent K, Hewitt CA, Middleton LJ, Koscielniak M, Szubert W, et al. Gabapentin for chronic pelvic pain in women (GaPP2): a multicentre, randomised, double-blind, placebo-controlled trial. Lancet. 2020;396(10255):909–17. https://doi.org/10.1016/S0140-6736(20)31693-7 .

Furtado RHM, Berwanger O, Fonseca HA, Corrêa TD, Ferraz LR, Lapa MG, et al. Azithromycin in addition to standard of care versus standard of care alone in the treatment of patients admitted to the hospital with severe COVID-19 in Brazil (COALITION II): a randomised clinical trial. Lancet. 2020;396(10256):959–67. https://doi.org/10.1016/S0140-6736(20)31862-6 .

Tomazini BM, Maia IS, Cavalcanti AB, Berwanger O, Rosa RG, Veiga VC, et al. Effect of dexamethasone on days alive and ventilator-free in patients with moderate or severe acute respiratory distress syndrome and COVID-19: the CoDEX randomized clinical trial. JAMA. 2020;324(13):1307–16. https://doi.org/10.1001/jama.2020.17021 .

Beigel JH, Tomashek KM, Dodd LE, Mehta AK, Zingman BS, Kalil AC, et al. Remdesivir for the treatment of COVID-19 - final report. N Engl J Med. 2020;383(19):1813–26. https://doi.org/10.1056/NEJMoa2007764 .

Goldman JD, Lye DCB, Hui DS, Marks KM, Bruno R, Montejano R, et al. Remdesivir for 5 or 10 days in patients with severe COVID-19. N Engl J Med. 2020;383(19):1827–37. https://doi.org/10.1056/NEJMoa2015301 .

Cavalcanti AB, Zampieri FG, Rosa RG, Azevedo LCP, Veiga VC, Avezum A, et al. Hydroxychloroquine with or without azithromycin in mild-to-moderate COVID-19. N Engl J Med. 2020;383(21):2041–52. https://doi.org/10.1056/NEJMoa2019014 .

Self WH, Semler MW, Leither LM, Casey JD, Angus DC, Brower RG, et al. Effect of hydroxychloroquine on clinical status at 14 days in hospitalized patients with COVID-19: a randomized clinical trial. JAMA. 2020;324(21):2165–76. https://doi.org/10.1001/jama.2020.22240 .

Martínez-Fernández R, Máñez-Miró JU, Rodríguez-Rojas R, Del Álamo M, Shah BB, Hernández-Fernández F, et al. Randomized trial of focused ultrasound subthalamotomy for Parkinson’s disease. N Engl J Med. 2020;383(26):2501–13.  https://doi.org/10.1056/NEJMoa2016311 . United States.

Hutchinson PJ, Edlmann E, Bulters D, Zolnourian A, Holton P, Suttner N, et al. Trial of dexamethasone for chronic subdural hematoma. N Engl J Med. 2020;383(27):2616–27.  https://doi.org/10.1056/NEJMoa2020473 . United States.

Klein AL, Imazio M, Cremer P, Brucato A, Abbate A, Fang F, et al. Phase 3 trial of interleukin-1 trap rilonacept in recurrent pericarditis. N Engl J Med. 2021;384(1):31–41.  https://doi.org/10.1056/NEJMoa2027892 . United States.

Post R, Germans MR, Tjerkstra MA, Vergouwen MDI, Jellema K, Koot RW, et al. Ultra-early tranexamic acid after subarachnoid haemorrhage (ULTRA): a randomised controlled trial. Lancet. 2021;397(10269):112–8.  https://doi.org/10.1016/S0140-6736(20)32518-6 . England.

Suzuki K, Matsumaru Y, Takeuchi M, Morimoto M, Kanazawa R, Takayama Y, et al. Effect of mechanical thrombectomy without vs with intravenous thrombolysis on functional outcome among patients with acute ischemic stroke: the SKIP randomized clinical trial. JAMA. 2021;325(3):244–53. https://doi.org/10.1001/jama.2020.23522 .

Zi W, Qiu Z, Li F, Sang H, Wu D, Luo W, et al. Effect of endovascular treatment alone vs intravenous alteplase plus endovascular treatment on functional independence in patients with acute ischemic stroke: the DEVT randomized clinical trial. JAMA. 2021;325(3):234–43. https://doi.org/10.1001/jama.2020.23523 .

Veiga VC, Prats JAGG, Farias DLC, Rosa RG, Dourado LK, Zampieri FG, et al. Effect of tocilizumab on clinical outcomes at 15 days in patients with severe or critical coronavirus disease 2019: randomised controlled trial. BMJ. 2021;372. https://doi.org/10.1136/bmj.n84 .

Gordon KB, Foley P, Krueger JG, Pinter A, Reich K, Vender R, et al. Bimekizumab efficacy and safety in moderate to severe plaque psoriasis (BE READY): a multicentre, double-blind, placebo-controlled, randomised withdrawal phase 3 trial. Lancet. 2021;397(10273):475–86.  https://doi.org/10.1016/S0140-6736(21)00126-4 . England.

Reich K, Papp KA, Blauvelt A, Langley RG, Armstrong A, Warren RB, et al. Bimekizumab versus ustekinumab for the treatment of moderate to severe plaque psoriasis (BE VIVID): efficacy and safety from a 52-week, multicentre, double-blind, active comparator and placebo controlled phase 3 trial. Lancet. 2021;397(10273):487–98.  https://doi.org/10.1016/S0140-6736(21)00125-2 . England.

Blauvelt A, Kempers S, Lain E, Schlesinger T, Tyring S, Forman S, et al. Phase 3 trials of tirbanibulin ointment for actinic keratosis. N Engl J Med. 2021;384(6):512–20.  https://doi.org/10.1056/NEJMoa2024040 . United States.

Simonovich VA, Burgos Pratx LD, Scibona P, Beruto MV, Vallone MG, Vázquez C, et al. A randomized trial of convalescent plasma in COVID-19 severe pneumonia. N Engl J Med. 2021;384(7):619–29. https://doi.org/10.1056/NEJMoa2031304 .

Brannan SK, Sawchak S, Miller AC, Lieberman JA, Paul SM, Breier A. Muscarinic cholinergic receptor agonist and peripheral antagonist for schizophrenia. N Engl J Med. 2021;384(8):717–26. https://doi.org/10.1056/NEJMoa2017015 .

Lundgren JD, Grund B, Barkauskas CE, Holland TL, Gottlieb RL, Sandkovsky U, et al. A neutralizing monoclonal antibody for hospitalized patients with COVID-19. N Engl J Med. 2021;384(10):905–14. https://doi.org/10.1056/NEJMoa2033130 .

Bieber T, Simpson EL, Silverberg JI, Thaçi D, Paul C, Pink AE, et al. Abrocitinib versus placebo or dupilumab for atopic dermatitis. N Engl J Med. 2021;384(12):1101–12.  https://doi.org/10.1056/NEJMoa2019380 . United States.

Gordon AC, Mouncey PR, Al-Beidh F, Rowan KM, Nichol AD, Arabi YM, et al. Interleukin-6 receptor antagonists in critically ill patients with COVID-19. N Engl J Med. 2021;384(16):1491–502. https://doi.org/10.1056/NEJMoa2100433 .

Rosas IO, Bräu N, Waters M, Go RC, Hunter BD, Bhagani S, et al. Tocilizumab in hospitalized patients with severe COVID-19 pneumonia. N Engl J Med. 2021;384(16):1503–16. https://doi.org/10.1056/NEJMoa2028700 .

Aspvall K, Andersson E, Melin K, Norlin L, Eriksson V, Vigerland S, et al. Effect of an Internet-delivered stepped-care program vs in-person cognitive behavioral therapy on obsessive-compulsive disorder symptoms in children and adolescents: a randomized clinical trial. JAMA. 2021;325(18):1863–73. https://doi.org/10.1001/jama.2021.3839 .

Langezaal LCM, van der Hoeven EJRJ, Mont’Alverne FJA, de Carvalho JJF, Lima FO, Dippel DWJ, et al. Endovascular therapy for stroke due to basilar-artery occlusion. N Engl J Med. 2021;384(20):1910–20.  https://doi.org/10.1056/NEJMoa2030297 . United States.

Roquilly A, Moyer JD, Huet O, Lasocki S, Cohen B, Dahyot-Fizelier C, et al. Effect of continuous infusion of hypertonic saline vs standard care on 6-month neurological outcomes in patients with traumatic brain injury: the COBI randomized clinical trial. JAMA. 2021;325(20):2056–66. https://doi.org/10.1001/jama.2021.5561 .

Guttman-Yassky E, Teixeira HD, Simpson EL, Papp KA, Pangan AL, Blauvelt A, et al. Once-daily upadacitinib versus placebo in adolescents and adults with moderate-to-severe atopic dermatitis (Measure Up 1 and Measure Up 2): results from two replicate double-blind, randomised controlled phase 3 trials. Lancet. 2021;397(10290):2151–68.  https://doi.org/10.1016/S0140-6736(21)00588-2 . England.

Reich K, Teixeira HD, de Bruin-Weller M, Bieber T, Soong W, Kabashima K, et al. Safety and efficacy of upadacitinib in combination with topical corticosteroids in adolescents and adults with moderate-to-severe atopic dermatitis (AD Up): results from a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2021;397(10290):2169–81.  https://doi.org/10.1016/S0140-6736(21)00589-4 . England.

Dankiewicz J, Cronberg T, Lilja G, Jakobsen JC, Levin H, Ullén S, et al. Hypothermia versus normothermia after out-of-hospital cardiac arrest. N Engl J Med. 2021;384(24):2283–94.  https://doi.org/10.1056/NEJMoa2100591 . United States.

Tariot PN, Cummings JL, Soto-Martin ME, Ballard C, Erten-Lyons D, Sultzer DL, et al. Trial of pimavanserin in dementia-related psychosis. N Engl J Med. 2021;385(4):309–19.  https://doi.org/10.1056/NEJMoa2034634 . United States.

Guimarães PO, Quirk D, Furtado RH, Maia LN, Saraiva JF, Antunes MO, et al. Tofacitinib in patients hospitalized with COVID-19 pneumonia. N Engl J Med. 2021;385(5):406–15. https://doi.org/10.1056/NEJMoa2101643 .

Lawler PR, Goligher EC, Berger JS, Neal MD, McVerry BJ, Nicolau JC, et al. Therapeutic anticoagulation with heparin in noncritically ill patients with COVID-19. N Engl J Med. 2021;385(9):790–802. https://doi.org/10.1056/NEJMoa2105911 .

Goligher EC, Bradbury CA, McVerry BJ, Lawler PR, Berger JS, Gong MN, et al. Therapeutic Anticoagulation with heparin in critically ill patients with COVID-19. N Engl J Med. 2021;385(9):777–89. https://doi.org/10.1056/NEJMoa2103417 .

Schwarzschild MA, Ascherio A, Casaceli C, Curhan GC, Fitzgerald R, Kamp C, et al. Effect of urate-elevating inosine on early Parkinson disease progression: the SURE-PD3 randomized clinical trial. JAMA. 2021;326(10):926–39. https://doi.org/10.1001/jama.2021.10207 .

Halliday A, Bulbulia R, Bonati LH, Chester J, Cradduck-Bamford A, Peto R, et al. Second asymptomatic carotid surgery trial (ACST-2): a randomised comparison of carotid artery stenting versus carotid endarterectomy. Lancet. 2021;398(10305):1065–73. https://doi.org/10.1016/S0140-6736(21)01910-3 .

Paget LDA, Reurink G, de Vos RJ, Weir A, Moen MH, Bierma-Zeinstra SMA, et al. Effect of platelet-rich plasma injections vs placebo on ankle symptoms and function in patients with ankle osteoarthritis: a randomized clinical trial. JAMA. 2021;326(16):1595–605. https://doi.org/10.1001/jama.2021.16602 .

Estcourt LJ, Turgeon AF, McQuilten ZK, McVerry BJ, Al-Beidh F, Annane D, et al. Effect of convalescent plasma on organ support-free days in critically ill patients with COVID-19: a randomized clinical trial. JAMA. 2021;326(17):1690–702. https://doi.org/10.1001/jama.2021.18178 .

LeCouffe NE, Kappelhof M, Treurniet KM, Rinkel LA, Bruggeman AE, Berkhemer OA, et al. A randomized trial of intravenous alteplase before endovascular treatment for stroke. N Engl J Med. 2021;385(20):1833–44.  https://doi.org/10.1056/NEJMoa2107727 . United States.

Korley FK, Durkalski-Mauldin V, Yeatts SD, Schulman K, Davenport RD, Dumont LJ, et al. Early convalescent plasma for high-risk outpatients with COVID-19. N Engl J Med. 2021;385(21):1951–60. https://doi.org/10.1056/NEJMoa2103784 .

Bennell KL, Paterson KL, Metcalf BR, Duong V, Eyles J, Kasza J, et al. Effect of intra-articular platelet-rich plasma vs placebo injection on pain and medial tibial cartilage volume in patients with knee osteoarthritis: the RESTORE randomized clinical trial. JAMA. 2021;326(20):2021–30. https://doi.org/10.1001/jama.2021.19415 .

Ospina-Tascón GA, Calderón-Tapia LE, García AF, Zarama V, Gómez-Álvarez F, Álvarez-Saa T, et al. Effect of high-flow oxygen therapy vs conventional oxygen therapy on invasive mechanical ventilation and clinical recovery in patients with severe COVID-19: a randomized clinical trial. JAMA. 2021;326(21):2161–71. https://doi.org/10.1001/jama.2021.20714 .

Lebwohl MG, Stein Gold L, Strober B, Papp KA, Armstrong AW, Bagel J, et al. Phase 3 Trials of tapinarof cream for plaque psoriasis. N Engl J Med. 2021;385(24):2219–29.  https://doi.org/10.1056/NEJMoa2103629 . United States.

Berger JS, Kornblith LZ, Gong MN, Reynolds HR, Cushman M, Cheng Y, et al. Effect of P2Y12 Inhibitors on survival free of organ support among non-critically ill hospitalized patients with COVID-19: a randomized clinical trial. JAMA. 2022;327(3):227–36. https://doi.org/10.1001/jama.2021.23605 .

Polizzotto MN, Nordwall J, Babiker AG, Phillips A, Vock DM, Eriobu N, et al. Hyperimmune immunoglobulin for hospitalised patients with COVID-19 (ITAC): a double-blind, placebo-controlled, phase 3, randomised trial. Lancet. 2022;399(10324):530–40. https://doi.org/10.1016/S0140-6736(22)00101-5 .

Gadjradj PS, Rubinstein SM, Peul WC, Depauw PR, Vleggeert-Lankamp CL, Seiger A, et al. Full endoscopic versus open discectomy for sciatica: randomised controlled non-inferiority trial. BMJ. 2022;376. https://doi.org/10.1136/bmj-2021-065846 .

Preskorn SH, Zeller S, Citrome L, Finman J, Goldberg JF, Fava M, et al. Effect of sublingual dexmedetomidine vs placebo on acute agitation associated with bipolar disorder: a randomized clinical trial. JAMA. 2022;327(8):727–36. https://doi.org/10.1001/jama.2022.0799 .

Ruijter BJ, Keijzer HM, Tjepkema-Cloostermans MC, Blans MJ, Beishuizen A, Tromp SC, et al. Treating rhythmic and periodic EEG patterns in comatose survivors of cardiac arrest. N Engl J Med. 2022;386(8):724–34.  https://doi.org/10.1056/NEJMoa2115998 . United States.

Renú A, Millán M, San Román L, Blasco J, Martí-Fàbregas J, Terceño M, et al. Effect of intra-arterial alteplase vs placebo following successful thrombectomy on functional outcomes in patients with large vessel occlusion acute ischemic stroke: the CHOICE randomized clinical trial. JAMA. 2022;327(9):826–35. https://doi.org/10.1001/jama.2022.1645 .

van der Steen W, van de Graaf RA, Chalos V, Lingsma HF, van Doormaal PJ, Coutinho JM, et al. Safety and efficacy of aspirin, unfractionated heparin, both, or neither during endovascular stroke treatment (MR CLEAN-MED): an open-label, multicentre, randomised controlled trial. Lancet. 2022;399(10329):1059–69.  https://doi.org/10.1016/S0140-6736(22)00014-9 . England.

Paskins Z, Bromley K, Lewis M, Hughes G, Hughes E, Hennings S, et al. Clinical effectiveness of one ultrasound guided intra-articular corticosteroid and local anaesthetic injection in addition to advice and education for hip osteoarthritis (HIT trial): single blind, parallel group, three arm, randomised controlled trial. BMJ. 2022;377:e068446. https://doi.org/10.1136/bmj-2021-068446 .

Yoshimura S, Sakai N, Yamagami H, Uchida K, Beppu M, Toyoda K, et al. Endovascular therapy for acute stroke with a large ischemic region. N Engl J Med. 2022;386(14):1303–13.  https://doi.org/10.1056/NEJMoa2118191 . United States.

Pérez de la Ossa N, Abilleira S, Jovin TG, García-Tornel Á, Jimenez X, Urra X, et al. Effect of direct transportation to thrombectomy-capable center vs local stroke center on neurological outcomes in patients with suspected large-vessel occlusion stroke in nonurban areas: the RACECAT randomized clinical trial. JAMA. 2022;327(18):1782–94. https://doi.org/10.1001/jama.2022.4404 .

Bösel J, Niesen WD, Salih F, Morris NA, Ragland JT, Gough B, et al. Effect of early vs standard approach to tracheostomy on functional outcome at 6 months among patients with severe stroke receiving mechanical ventilation: the SETPOINT2 randomized clinical trial. JAMA. 2022;327(19):1899–909. https://doi.org/10.1001/jama.2022.4798 .

Perry DC, Achten J, Knight R, Appelbe D, Dutton SJ, Dritsaki M, et al. Immobilisation of torus fractures of the wrist in children (FORCE): a randomised controlled equivalence trial in the UK. Lancet. 2022;400(10345):39–47.  https://doi.org/10.1016/S0140-6736(22)01015-7 . England.

Fischer U, Kaesmacher J, Strbian D, Eker O, Cognard C, Plattner PS, et al. Thrombectomy alone versus intravenous alteplase plus thrombectomy in patients with stroke: an open-label, blinded-outcome, randomised non-inferiority trial. Lancet. 2022;400(10346):104–15.  https://doi.org/10.1016/S0140-6736(22)00537-2 . England.

Mitchell PJ, Yan B, Churilov L, Dowling RJ, Bush SJ, Bivard A, et al. Endovascular thrombectomy versus standard bridging thrombolytic with endovascular thrombectomy within 4 \(\cdot\) 5 h of stroke onset: an open-label, blinded-endpoint, randomised non-inferiority trial. Lancet. 2022;400(10346):116–25.  https://doi.org/10.1016/S0140-6736(22)00564-5 . England.

Wu YW, Comstock BA, Gonzalez FF, Mayock DE, Goodman AM, Maitre NL, et al. Trial of erythropoietin for hypoxic-ischemic encephalopathy in newborns. N Engl J Med. 2022;387(2):148–59.  https://doi.org/10.1056/NEJMoa2119660 . United States.

Menon BK, Buck BH, Singh N, Deschaintre Y, Almekhlafi MA, Coutts SB, et al. Intravenous tenecteplase compared with alteplase for acute ischaemic stroke in Canada (AcT): a pragmatic, multicentre, open-label, registry-linked, randomised, controlled, non-inferiority trial. Lancet. 2022;400(10347):161–9.  https://doi.org/10.1016/S0140-6736(22)01054-6 . England.

Valenta Z, Pitha J, Poledne R. Proportional odds logistic regression—effective means of dealing with limited uncertainty in dichotomizing clinical outcomes. Stat Med. 2006;25(24):4227–34.

Saver JL. Novel end point analytic techniques and interpreting shifts across the entire range of outcome scales in acute stroke trials. Stroke. 2007;38(11):3055–62.

Fullerton AS, Xu J. The proportional odds with partial proportionality constraints model for ordinal response variables. Soc Sci Res. 2012;41(1):182–98.

Faraway JJ. On the cost of data analysis. J Comput Graph Stat. 1992;1(3):213–29.

Clogg CC, Shihadeh ES. Statistical models for ordinal variables. AGE Publications; 1994.

McCullagh P. Generalized linear models. Routledge; 2019.

Crans GG, Shuster JJ. How conservative is Fisher’s exact test? A quantitative evaluation of the two-sample comparative binomial trial. Stat Med. 2008;27(18):3598–611.

Selman CJ. Statistic analyses of ordinal outcomes in RCTs: scoping review data. GitHub. 2023. https://github.com/chrisselman/ordscopingreview . Accessed 31 May 2023.

Download references

Acknowledgements

Not applicable.

This work forms part of Chris Selman’s PhD, which is supported by the Research Training Program Scholarship, administered by the Australian Commonwealth Government and The University of Melbourne, Australia. Chris Selman’s PhD was also supported by a Centre of Research Excellence grant from the National Health and Medical Research Council of Australia ID 1171422, to the Australian Trials Methodology (AusTriM) Research Network. Research at the Murdoch Children’s Research Institute is supported by the Victorian Government’s Operational Infrastructure Support Program. This work was supported by the Australian National Health and Medical Research Council (NHMRC) Centre for Research Excellence grants to the Victorian Centre for Biostatistics (ID1035261) and the Australian Trials Methodology Research Network (ID1171422), including through seed funding awarded to Robert Mahar. Katherine Lee is funded by an NHMRC Career Development Fellowship (ID1127984). Brett Manley is funded by the NHMRC Investigator Grant (Leadership 1). The funding bodies played no role in the study conception, design, data collection, data analysis, data interpretation, or writing of the report.

Author information

Authors and affiliations.

Clinical Epidemiology and Biostatistics Unit, Murdoch Children’s Research Institute, Parkville, VIC, 3052, Australia

Chris J. Selman, Katherine J. Lee & Robert K. Mahar

Department of Paediatrics, University of Melbourne, Parkville, VIC, 3052, Australia

Chris J. Selman & Katherine J. Lee

Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, Parkville, VIC, 3052, Australia

Robert K. Mahar

Department of Obstetrics and Gynaecology, University of Melbourne, Parkville, VIC, 3052, Australia

Kristin N. Ferguson, Clare L. Whitehead & Brett J. Manley

Department of Maternal Fetal Medicine, The Royal Women’s Hospital, Parkville, VIC, 3052, Australia

Clare L. Whitehead

Newborn Research, The Royal Women’s Hospital, Parkville, VIC, 3052, Australia

Brett J. Manley

Clinical Sciences, Murdoch Children’s Research Institute, Parkville, VIC, 3052, Australia

You can also search for this author in PubMed   Google Scholar

Contributions

CJS, RKM, KJL, CLW, and BJM conceived the study and CJS wrote the first draft of the manuscript. All authors contributed to the design of the study, revision of the manuscript, and take responsibility for its content.

Corresponding author

Correspondence to Chris J. Selman .

Ethics declarations

Ethics approval and consent to participate.

As data and information was only extracted from published studies, ethics approval was not required.

Consent for publication

Competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Additional file 1..

Deviations from the protocol. This presents a summary of the deviations from the protocol, with reasons. We also provide an explanation of any simplifications and assumptions that were made for eligibility criteria and data extraction.

Additional file 2.

Data extraction questionnaire. This is a copy of the data extraction questionnaire that will be used for this review in PDF format.

Additional file 3.

Interpretation of the proportional odds ratio in proportional odds models. This presents a summary of the ways that the proportional odds ratio was interpreted across the studies.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Selman, C.J., Lee, K.J., Ferguson, K.N. et al. Statistical analyses of ordinal outcomes in randomised controlled trials: a scoping review. Trials 25 , 241 (2024). https://doi.org/10.1186/s13063-024-08072-2

Download citation

Received : 02 July 2023

Accepted : 22 March 2024

Published : 06 April 2024

DOI : https://doi.org/10.1186/s13063-024-08072-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Ordinal outcomes
• Proportional odds model
• Randomised controlled trials
• Scoping review

ISSN: 1745-6215

• Submission enquiries: Access here and click Contact Us
• General enquiries: [email protected]

ORIGINAL RESEARCH article

This article is part of the research topic.

Energy, Economy, and Climate Interactions: Challenges and Opportunities - Volume II

The Spatial Effect of Integrated Economy on Carbon Emissions in the Era of Big Data: A Case Study of China Provisionally Accepted

• 1 School of Economic and Management, Xi'an University of Technology, China
• 2 School of Business and Circulation, Shaanxi Polytechnic Institute, China

The final, formatted version of the article will be published soon.

The digital economy has the characteristics of resource conservation, which can solve China's high carbon emissions problems. The digital economy can quickly integrate with the real economy, forming an integrated economy. However, it is still unclear whether an integrated economy can effectively reduce carbon emissions and achieve China's 'dual carbon goals'. Therefore, this study takes 30 provinces in China as the research object, constructs the integration economy index system through the statistical data from 2011-2021, and explores the spatial effect of the impact of the integration economy on carbon emissions by using principal component analysis, coupled coordination model and spatial econometric model. The research results are as follows. (1) From 2011 to 2021, the comprehensive economy showed a trend of increasing yearly (from 0.667 to 0.828), and carbon emissions showed a slow decrease (from 0.026 to 0.017). (2) Due to the infiltration of China's economic development from the eastern to the western, the spatial distribution of the integrated economy shows a decreasing trend from east to west. The spatial distribution of carbon emissions may be related to China's industrial layout of heavy industry in the northern, and light industry in the southern, showing a trend of low in the south and high in the north. (3) The integrated economy can significantly reduce carbon emissions (the coefficients of influence, -0.146), and the reduction effect will be more obvious if spatial spillover effects are taken into account (-0.305). ( 4) The eastern coast, the middle reaches of the Yangtze River, and the middle reaches of the Yellow River economic zones all increase carbon emissions at a certain level of significance (0.065, 0.148, and 3.890). The Northeast, South Coastal and Southwest economic zones significantly reduce carbon emissions (-0.220, -0.092, and -0.308). The results of the Northern Coast and Northwest are not significant (-0.022 and 0.095). ( 5) China should tailor regional economic development policies, such as strengthening investment in digital infrastructure in the Northwest Economic Zone and fully leveraging the spatial spillover effects of integrated economy in the Northeast, Southern Coastal, and Southwest Economic Zones to reduce carbon emissions.

Keywords: Integrated economy, carbon emissions, digital economy, Real economy, spatial effect, China

Received: 22 Jan 2024; Accepted: 11 Apr 2024.

Copyright: © 2024 Wang, Ke and Lei. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

* Correspondence: Miss. Qian Ke, Xi'an University of Technology, School of Economic and Management, Xi'an, 710048, Shaanxi, China

People also looked at

Jump to navigation

• Bahasa Malaysia

Are midwife continuity of care models versus other models of care for childbearing women better for women and their babies?

Key messages

Women or their babies who received midwife continuity of care models were less likely to experience a caesarean section or instrumental birth with forceps or a ventouse suction cup, and may be less likely to experience an episiotomy (a cut made by a healthcare professional into the perineum and vaginal wall). They were more likely to experience spontaneous vaginal birth.

Women who experienced midwife continuity of care models reported more positive experiences during pregnancy, labour, and postpartum. Additionally, there were cost savings in the antenatal (care during pregnancy) and intrapartum (care during labour and birth) period.

Further evidence may change our results, and future research should focus on the impact on women with social risk factors, and those with medical complications, and understanding the implementation and scaling up of midwife continuity of care models, with emphasis on low- and middle-income countries.

What are midwife continuity of care models?

Midwife continuity of care models provide care from the same midwife or team of midwives during pregnancy, birth, and the early parenting period in collaboration with obstetric and specialist teams when required.

What did we want to find out?

We wanted to find out how outcomes differed for women or their babies who received a midwife continuity of care model compared to other models of care.

Our main outcomes were: spontaneous vaginal birth, caesarean section, regional anaesthesia (spinal or epidural block to numb the lower part of the body), intact perineum (the area between the anus and the vulva), fetal loss after 24 weeks gestation, preterm birth, and neonatal death.

We also looked at a range of other outcomes, including women’s experience and cost.

What did we do?

We searched for studies that compared midwife continuity of care models with other models of care for pregnant women. We compared and summarised the results of the studies and rated our confidence in the evidence based on factors such as study methods and size.

What did we find?

We found 17 studies involving 18,533 women in Australia, Canada, China, Ireland, and the United Kingdom.

Many of these studies largely focused on women with a lower risk of complications at the start of pregnancy, or those drawn from a specific geographical location. Midwives continued to provide midwifery care in collaboration with specialist and obstetric teams if women developed complications in pregnancy, birth, and postpartum.

Our main results

Women or their babies who received midwife continuity of care models compared to those receiving other models of care were less likely to experience a caesarean section or instrumental vaginal delivery, and may be less likely to experience an episiotomy. They were more likely to experience a spontaneous vaginal birth.

Midwife continuity care models probably make little or no difference to the likelihood of having an intact perineum, and may have little or no impact on the likelihood of preterm birth.

We are uncertain about the effect of midwife continuity of care models on regional anaesthesia, fetal loss after 24 weeks' gestation, and neonatal death.

Women who experienced care from midwife continuity of care models reported more positive experiences during pregnancy, labour, and postpartum. Additionally, there were cost savings in the antenatal and intrapartum period.

What are the limitations of the evidence?

Our confidence in these findings varies and further evidence may change our results. For instance, it is not always clear if the people assessing the outcomes knew which type of care the women received.The evidence for fetal loss after 24 weeks' gestation and neonatal death is based on a very small number of cases and there are not enough studies to be certain about some results. We lack data on important aspects like maternal health status after birth, neonatal readmissions, or infant health status.

Few studies included a specific focus on women at high risk of complications, and none focused on women from disadvantaged backgrounds, indicating a need for future research in these areas. This highlights the need for more comprehensive and diverse studies to strengthen our understanding and confidence in these findings, particularly in varied populations and across different healthcare settings.

Future research should focus on the impact on women with social risk factors, and those with medical complications, and understanding the implementation and scaling up of midwife continuity of care models, with emphasis on low- and middle-income countries.

How up-to-date is this evidence?

This is an update of our previous review. We included evidence up to 17 August 2022.

Women receiving midwife continuity of care models were less likely to experience a caesarean section and instrumental birth, and may be less likely to experience episiotomy. They were more likely to experience spontaneous vaginal birth and report a positive experience. The certainty of some findings varies due to possible risks of bias, inconsistencies, and imprecision of some estimates.

Future research should focus on the impact on women with social risk factors, and those at higher risk of complications, and implementation and scaling up of midwife continuity of care models, with emphasis on low- and middle-income countries.

Midwives are primary providers of care for childbearing women globally and there is a need to establish whether there are differences in effectiveness between midwife continuity of care models and other models of care. This is an update of a review published in 2016.

To compare the effects of midwife continuity of care models with other models of care for childbearing women and their infants.

We searched the Cochrane Pregnancy and Childbirth Trials Register, ClinicalTrials.gov, and the WHO International Clinical Trials Registry Platform (ICTRP) (17 August 2022), as well as the reference lists of retrieved studies.

All published and unpublished trials in which pregnant women are randomly allocated to midwife continuity of care models or other models of care during pregnancy and birth.

Two authors independently assessed studies for inclusion criteria, scientific integrity, and risk of bias, and carried out data extraction and entry. Primary outcomes were spontaneous vaginal birth, caesarean section, regional anaesthesia, intact perineum, fetal loss after 24 weeks gestation, preterm birth, and neonatal death. We used GRADE to rate the certainty of evidence.

We included 17 studies involving 18,533 randomised women. We assessed all studies as being at low risk of scientific integrity/trustworthiness concerns. Studies were conducted in Australia, Canada, China, Ireland, and the United Kingdom. The majority of the included studies did not include women at high risk of complications. There are three ongoing studies targeting disadvantaged women.

Primary outcomes

Based on control group risks observed in the studies, midwife continuity of care models, as compared to other models of care, likely increase spontaneous vaginal birth from 66% to 70% (risk ratio (RR) 1.05, 95% confidence interval (CI) 1.03 to 1.07; 15 studies, 17,864 participants; moderate-certainty evidence), likely reduce caesarean sections from 16% to 15% (RR 0.91, 95% CI 0.84 to 0.99; 16 studies, 18,037 participants; moderate-certainty evidence), and likely result in little to no difference in intact perineum (29% in other care models and 31% in midwife continuity of care models, average RR 1.05, 95% CI 0.98 to 1.12; 12 studies, 14,268 participants; moderate-certainty evidence). There may be little or no difference in preterm birth (< 37 weeks) (6% under both care models, average RR 0.95, 95% CI 0.78 to 1.16; 10 studies, 13,850 participants; low-certainty evidence).

We are very uncertain about the effect of midwife continuity of care models on regional analgesia (average RR 0.85, 95% CI 0.79 to 0.92; 15 studies, 17,754 participants, very low-certainty evidence), fetal loss at or after 24 weeks gestation (average RR 1.24, 95% CI 0.73 to 2.13; 12 studies, 16,122 participants; very low-certainty evidence), and neonatal death (average RR 0.85, 95% CI 0.43 to 1.71; 10 studies, 14,718 participants; very low-certainty evidence).

Secondary outcomes

When compared to other models of care, midwife continuity of care models likely reduce instrumental vaginal birth (forceps/vacuum) from 14% to 13% (average RR 0.89, 95% CI 0.83 to 0.96; 14 studies, 17,769 participants; moderate-certainty evidence), and may reduce episiotomy 23% to 19% (average RR 0.83, 95% CI 0.77 to 0.91; 15 studies, 17,839 participants; low-certainty evidence).

When compared to other models of care, midwife continuity of care models likely result in little to no difference in postpartum haemorrhage (average RR 0.92, 95% CI 0.82 to 1.03; 11 studies, 14,407 participants; moderate-certainty evidence) and admission to special care nursery/neonatal intensive care unit (average RR 0.89, 95% CI 0.77 to 1.03; 13 studies, 16,260 participants; moderate-certainty evidence). There may be little or no difference in induction of labour (average RR 0.92, 95% CI 0.85 to 1.00; 14 studies, 17,666 participants; low-certainty evidence), breastfeeding initiation (average RR 1.06, 95% CI 1.00 to 1.12; 8 studies, 8575 participants; low-certainty evidence), and birth weight less than 2500 g (average RR 0.92, 95% CI 0.79 to 1.08; 9 studies, 12,420 participants; low-certainty evidence).

We are very uncertain about the effect of midwife continuity of care models compared to other models of care on third or fourth-degree tear (average RR 1.10, 95% CI 0.81 to 1.49; 7 studies, 9437 participants; very low-certainty evidence), maternal readmission within 28 days (average RR 1.52, 95% CI 0.78 to 2.96; 1 study, 1195 participants; very low-certainty evidence), attendance at birth by a known midwife (average RR 9.13, 95% CI 5.87 to 14.21; 11 studies, 9273 participants; very low-certainty evidence), Apgar score less than or equal to seven at five minutes (average RR 0.95, 95% CI 0.72 to 1.24; 13 studies, 12,806 participants; very low-certainty evidence) and fetal loss before 24 weeks gestation (average RR 0.82, 95% CI 0.67 to 1.01; 12 studies, 15,913 participants; very low-certainty evidence). No maternal deaths were reported across three studies.

Although the observed risk of adverse events was similar between midwifery continuity of care models and other models, our confidence in the findings was limited. Our confidence in the findings was lowered by possible risks of bias, inconsistency, and imprecision of some estimates.

There were no available data for the outcomes: maternal health status, neonatal readmission within 28 days, infant health status, and birth weight of 4000 g or more.

Maternal experiences and cost implications are described narratively. Women receiving care from midwife continuity of care models, as opposed to other care models, generally reported more positive experiences during pregnancy, labour, and postpartum. Cost savings were noted in the antenatal and intrapartum periods in midwife continuity of care models.

• Borrow, Renew, Request How to borrow materials, request pdf scans, and interlibrary loans .
• Study Spaces Areas for individual and group study and how to reserve them.
• Course Reserves How to access course-related materials reserved by faculty for their students.
• Services for Faculty and Instructors A list of services offered to faculty and instructors at the University of Hawaii at Manoa.
• Library Instruction Request library instruction for your course or register for a workshop.
• Suggest a Purchase Suggest new materials that support teaching, study, or research.
• Other Services Apply for a research carrel or reserve our lactation room.
• Loanable Technology Cables, adaptors, audio and video equipment, and other devices
• Collections An overview of the various library collections.
• Online Databases Search across 100s of library databases.
• Journals Search journals by title or subject.
• Research Guides Guides for subjects, select courses, and general information.
• OneSearch Finds books and other materials in the UH Manoa Library's collection.
• Scholarly Communication Learn about scholarly communication, open access, and our institutional repositories ScholarSpace , eVols , and the UH System Repository .
• Ask a Librarian Get help by email, online form, or phone.
• FAQ Frequently asked questions.
• Accessibility and Disability Information about accessibility and disability.
• Subject Librarians Find a librarian for a specific subject.
• Copyright Help Links to resources about copyright.
• Technology in the Library Wireless access, scanning, printing.
• English 100 Students The starting point for English 100 research.
• Request a Research Appointment Contact us to schedule an in-person appointment.
• Office and Department Contacts View a list of the departments at the library.
• Jobs at the Library Faculty, staff, and student job opportunities.
• Staff Directory Contact information for staff at the library.
• Exhibits Current and past exhibits at the library.
• Support the Library Find out how you can support the library.
• Our Library Annual reports, mission, values, history, and policies.
• Visiting Hours, directions, floor plans
• News, Blogs & Events News, blogs & events from the library.
• Research Guides

Environmental Studies Research Guide

• Data & Statistics
• Reference Sources
• Dissertations & Theses
• Citing Sources
• << Previous: Dissertations & Theses
• Next: Citing Sources >>
• Last Updated: Apr 8, 2024 9:27 AM
• URL: https://guides.library.manoa.hawaii.edu/environmental

An equal opportunity/affirmative action institution . Use of this site implies consent with our Usage Policy .

Terms of Use | UH System | UH Mānoa

• Learning Skills
• Teaching Resources
• Study Spaces
• Ask A Librarian
• Interlibrary Loan
• Library Faculty/Staff (Internal)

2550 McCarthy Mall Honolulu, Hawaii 96822 USA 808-956-7214 (Reference) 808-956-7203 (Circulation)

Library Digital Collections Disclaimer and Copyright information

© University of Hawaiʻi at Mānoa Library

• Open access
• Published: 09 April 2024

Targeting transitioning lung monocytes/macrophages as treatment strategies in lung disease related to environmental exposures

• Aaron D. Schwab 1 ,
• Todd A. Wyatt 2 , 3 , 4 ,
• Grace Moravec 1 ,
• Geoffrey M. Thiele 2 , 5 ,
• Amy J. Nelson 1 ,
• Angela Gleason 1 ,
• Oliver Schanze 1 ,
• Michael J. Duryee 2 , 5 ,
• Debra J. Romberger 2 , 3 ,
• Ted R. Mikuls 2 , 5 &
• Jill A. Poole 1  

Respiratory Research volume  25 , Article number:  157 ( 2024 ) Cite this article

21 Accesses

Metrics details

Environmental/occupational exposures cause significant lung diseases. Agricultural organic dust extracts (ODE) and bacterial component lipopolysaccharide (LPS) induce recruited, transitioning murine lung monocytes/macrophages, yet their cellular role remains unclear.

CCR2 RFP + mice were intratracheally instilled with high concentration ODE (25%), LPS (10 μg), or gram-positive peptidoglycan (PGN, 100 μg) for monocyte/macrophage cell-trafficking studies. CCR2 knockout (KO) mice and administration of intravenous clodronate liposomes strategies were employed to reduce circulating monocytes available for lung recruitment following LPS exposure. Lung tissues and bronchoalveolar lavage fluid (BALF) were collected. Pro-inflammatory and/or pro-fibrotic cytokines, chemokines, and lung extracellular matrix mediators were quantitated by ELISA. Infiltrating lung cells including monocyte/macrophage subpopulations, neutrophils, and lymphocytes were characterized by flow cytometry. Lung histopathology, collagen content, vimentin, and post-translational protein citrullination and malondialdehyde acetaldehyde (MAA) modification were quantitated. Parametric statistical tests (one-way ANOVA, Tukey’smultiple comparison) and nonparametric statistical (Kruskal–Wallis, Dunn’s multiple comparison) tests were used following Shapiro–Wilk testing for normality.

Intratracheal instillation of ODE, LPS, or PGN robustly induced the recruitment of inflammatory CCR2 + CD11c int CD11b hi monocytes/macrophages and both CCR2 + and CCR2 − CD11c − CD11b hi monocytes at 48 h. There were also increases in CCR2 + CD4 + and CD8 + T cells and NK cells. Despite reductions in LPS-induced lung infiltrating CD11c int CD11b hi cells (54% reduction), CCR2 knockout (KO) mice were not protected against LPS-induced inflammatory and pro-fibrotic consequences. Instead, compensatory increases in lung neutrophils and CCL2 and CCL7 release occurred. In contrast, the depletion of circulating monocytes through the administration of intravenous clodronate (vs. vehicle) liposomes 24 h prior to LPS exposure reduced LPS-induced infiltrating CD11c int CD11b hi monocyte-macrophage subpopulation by 59% without compensatory changes in other cell populations. Clodronate liposome pre-treatment significantly reduced LPS-induced IL-6 (66% reduction), matrix metalloproteinases (MMP)-3 (36%), MMP-8 (57%), tissue inhibitor of metalloproteinases (61%), fibronectin (38%), collagen content (22%), and vimentin (40%). LPS-induced lung protein citrullination and MAA modification, post-translational modifications implicated in lung disease, were reduced (39% and 48%) with clodronate vs. vehicle liposome.

Highly concentrated environmental/occupational exposures induced the recruitment of CCR2 + and CCR2 − transitioning monocyte-macrophage and monocyte subpopulations and targeting peripheral monocytes may reduce the adverse lung consequences resulting from exposures to LPS-enriched inhalants.

Introduction

Environmental and occupational lung diseases remain a significant cause of pulmonary impairment worldwide [ 1 ]. Chronic respiratory diseases including chronic obstructive pulmonary disease (COPD), asthma, asthma-like syndrome, byssinosis, hypersensitivity pneumonitis, and pulmonary fibrosis have been associated with exposure to organic dusts. Organic dusts are comprised of particulate matter, components of bacteria, fungi, viruses, pollen, and fragments of animals and plants [ 2 ]. These exposures are common to agriculture and farming, the grain and food processing industry, waste and recycling facilities, the textile and cotton industry, woodworking, concentrated urban areas, flood- and water-damaged buildings, and more [ 3 , 4 , 5 , 6 , 7 ]. Retrospective analysis of the Global Burden of Disease database identified roughly 519,100 deaths and 13.6 million disability-adjusted life years in 2016 from chronic respiratory disease due to occupational airborne exposures [ 8 ]. The incidence and prevalence of interstitial lung disease increased from 1990 to 2019, with occupational exposure implicated as a leading risk factor despite increased awareness and implementation of preventative measures [ 9 ]. However, there remains a paucity of therapeutic options aimed at hastening recovery and/or preventing chronic disease resulting from these exposures.

Lipopolysaccharide (LPS) or endotoxin is found in the outer membrane of gram-negative bacteria [ 10 , 11 , 12 ] and acts as a well-defined component of many organic dusts and disease-causing environmental exposures. A role for gram-positive cell wall components such as peptidoglycans (PGN) in mediating lung disease following organic dust exposures has also been identified. Growing industrialization, intensified agricultural production, climate change, and the rising frequency and severity of extreme weather events have conspired to increase concentrations of aerosolized organic particulate matter and environmental endotoxins [ 13 , 14 , 15 , 16 ]. Organic dusts and its bacterial components engage innate immune signaling pathways (i.e., Toll-like receptors) to initiate airway inflammatory responses marked by the influx of neutrophils, lymphocytes, monocytes, and macrophages with a corresponding release of pro-inflammatory/fibrotic mediators (e.g., tumor necrosis factor (TNF)-α, interleukin (IL)-6, chemoattractants, extracellular matrix proteins). Organic dusts and LPS exposures also induce post-translational modifications in proteins, which may serve to increase inflammation and promote tissue fibrosis [ 17 , 18 , 19 , 20 ]. Although dust and bacterial component-induced lymphocytic lung aggregates were reduced in T and B cell depleted mice, many inflammatory consequences persisted. Thus, there is growing interest in the immunopathogenic role of distinct lung monocyte-macrophage subpopulations recruited and induced following environmental exposures [ 12 , 21 , 22 ]. CC motif chemokine receptor 2 (CCR2) is a critical facilitator of monocyte recruitment and activation via interaction with its high-affinity ligand CCL2 [ 23 , 24 ]. CCR2 + monocytes play a critical role in the onset of inflammation and, upon reaching the inflamed tissue, differentiate into phenotypically and functionally distinct macrophages, capable of modulating inflammatory responses [ 25 ].

The objectives of this study were to first delineate the magnitude and distribution of CCR2 + (and CCR2 − ) monocytes and monocyte-derived macrophages trafficked to the lung following organic dust, LPS, and PGN exposures, then determine whether targeting these lung monocytes-macrophages would attenuate resulting pro-inflammatory and pro-fibrotic responses. We tested the hypotheses that pro-inflammatory and pro-fibrotic responses in lungs from mice exposed to inhaled LPS would be reduced in CCR2 deficient (knockout) mice compared to wild-type mice and that similar protection would be afforded in wild-type mice through the administration of intravenous clodronate liposome to deplete circulating monocytes. Understanding the role of recruited monocytes and monocyte-derived macrophages will provide fundamental knowledge of inflammatory processes following environmental exposures, but also potentially elucidate therapeutic targets to mitigate disease development in at-risk persons.

Environmental exposures

Lipopolysaccharide (LPS) from gram-negative Escherichia coli (O55:B5; Sigma, St. Louis, MO) served as the primary exposure in all experiments. The rationale was that LPS is commercially available and elicits a dose-dependent, reproducible pro-inflammatory lung response in humans and rodents. In studies of monocyte trafficking, comparisons were undertaken using peptidoglycan (PGN) from gram-positive Staphylococcus aureus (Sigma) and an aqueous solution of organic dust extract (ODE) prepared from swine confinement feeding facilities as previously described [ 26 ]. Briefly, settled surface dust (1 g) was incubated in sterile Hank’s Balanced Salt Solution (10 mL; Mediatech, Manassas, VA) for 1 h and centrifuged for 30 min at 2850 × g twice, with the final supernate filter-sterilized (0.22 um) to remove microorganisms and coarse particles. Stock ODE was batch prepared and stored at − 20 °C; aliquots were diluted for each experiment to a final concentration (vol/vol) of 25% in sterile phosphate buffered saline (PBS, pH 7.4; diluent). Endotoxin concentrations were determined using the limulus amebocyte lysate assay (Lonza, Walkersville, MD). Endotoxin levels averaged 1.308–2.616 μg (~ 10–50 EU) for 25% ODE. Prior mass spectrometry studies of ODE have revealed significant amounts of muramic acid (peptidoglycan marker) and 3-hydroxy fatty acids (endotoxin marker), but not ergosterol (fungi marker) as compared to house dust [ 26 ].

Animal exposure model

C57BL/6 and homozygous CCR2 RFP/RFP (B6.129(Cg)-Ccr2 tm2.1lfc /J) mice between 6 and 8 weeks of age were purchased from The Jackson Laboratory (Bar Harbor, ME). In this latter strain (#017586; RRID:IMSR_JAX:017586), a monomeric red fluorescent protein (RFP) sequence replaces the coding sequence of the Ccr2 gene, abolishing gene function and thus referred to as CCR2 knockout (KO) mice. Mouse tail snips were collected and shipped for DNA extraction and targeted CCR2 genotyping (TransnetYX, Cordova, TN) to confirm CCR2 KO. To generate heterozygous CCR2 +/RFP mice in which CCR2 is functional yet marked by RFP expression, CCR2 RFP/RFP mice were bred to C57BL/6 wild-type (WT) mice. For experiments using heterozygous CCR2 +/RFP animals, male and female mice were utilized. For the CCR2 KO (and clodronate liposome studies), male mice were utilized, as male mice had increased inflammatory responses with less experimental variability following LPS exposure, consistent with prior studies [ 12 , 27 ]. Mice were randomized, with AJN, AG, and animal facility staff aware of the randomization, whereas all other authors were blinded. To induce airway inflammation, mice were lightly sedated under isoflurane (VetOne, Boise, ID) and received one treatment with 50 μl of sterile saline (control), ODE (25%), LPS (10 μg), or PGN (100 μg) [ 28 ]. Animals were euthanized 48 h following exposure by isoflurane followed by exsanguination (right axillary blood collection). No respiratory distress, signs of stress, or significant weight loss (defined as > 20%) were observed throughout the exposure period.

Clodronate-induced systemic monocyte/macrophage depletion

In separate studies, C57BL/6 WT mice were administered encapsulated clodronate liposomes intravenously to deplete systemic monocytes and recruited monocyte-derived macrophages [ 29 , 30 , 31 , 32 ]. Clodronate and control liposomes (Liposoma Technology, Amsterdam, Netherlands; 200 μl × 5 mg/ml) were injected into the tail vein one day prior to LPS and saline control exposure.

Lavage fluid cells and lung homogenates

Bronchoalveolar lavage fluid (BALF) was collected using 3 × 1 mL PBS. Total cell numbers from the combined recovered lavage were enumerated using a BioRad TC 20 cell counter with differential cell counts determined from cytospin-prepared slides (cytopro cytocentrifuge, ELITech Group, Logan, UT) stained with Diff-Quick (Siemens, Newark, DE). Lung tissue homogenates were prepared by homogenizing lung samples (1/2 of right lungs) in 500 μl of sterile phosphate buffered saline (PBS) following removal of BALF and blood from the pulmonary vasculature. From cell-free lung tissue homogenates, levels of TNF-α, IL-6, murine neutrophil chemoattractant CXCL1, murine monocyte (and leukocyte) chemoattractants CCL2 and CCL7, and transforming growth factor (TGF)-β were quantitated by ELISA (R&D Systems, Minneapolis, MN) with minimal detectable difference (MDD) of 1.88, 1.6, 2.0, 0.3, 1.5, 31.3 pg/ml, respectively. Additionally, lung tissue homogenates were assessed for regulators of extracellular matrix deposition including matrix metalloproteinase (MMP)-3 and tissue inhibitor of metalloproteinase (TIMP)-1 (ELISA; R&D Systems; MDD of 0.125 and 0.031 ng/ml, respectively) as well as MMP-8 (ELISA; Abcam, Boston, MA; MDD of 0.053 ng/ml).

Lung cell staining and flow cytometry

Following removal of BALF and blood from pulmonary vasculature, harvested lungs (1/2 of right lungs) were subjected to an automated dissociation procedure using a gentleMACS Dissociator instrument (Miltenyi Biotech, Auburn, CA). Viability of total lung cells was assessed by trypan blue exclusion and a LIVE/DEAD Fixable Blue Dead Cell Stain Kit (Invitrogen, Carlsbad, CA). Cell viability was > 99% with no differences by treatment group(s) (data not shown). Lung cells were incubated with CD16/32 (Fc Block, BioLegend, San Diego, CA) to minimize non-specific antibody staining, then stained with monoclonal antibodies against rat anti-mouse; CD45 (clone: 30-F11; BD Biosciences, Franklin Lake, NJ), CD11b (clone: M1/70; BD Biosciences and BioLegend), Ly6G (clone: 1A8; BD Biosciences), CD11c (clone: N418; Invitrogen), CD4 (clone: RM4-5; BD Biosciences), CD8 (clone: 53–6.7; BD Biosciences), CD19 (clone: 1D3; Invitrogen), hamster anti-mouse CD3e (clone: 145-2C11; BD Biosciences and BioLegend), mouse anti-mouse NK1.1 (clone: PK136; BD Biosciences or BioLegend), Ly6C (clone: HK1.4; BioLegend), and F4/80 (clone: QA17A29; BioLegend or clone: T45-2342; BD Biosciences). Cells were acquired on a BD LSRII Yellow/Green cytometer configured with 355-nm, 405-nm, 488-nm, 561-nm, and 633-nm lasers. Post-acquisition, data were exported and stored using the flow cytometry standard (FCS) 3.1 format and analyzed using FlowJo software version 10.8 (FlowJo, RRID:SCR_008520, Ashland, OR).

The gating strategies for Ly6G + neutrophils, CD11c + CD11b lo alveolar (Alv) macrophages (Mɸ), CD11c + CD11b hi activated (Act) alveolar Mɸ, CD11c int CD11b hi recruited/transitioning monocytes-Mɸ, CD11c − CD11b hi monocytes, CD3 + CD4 + T cells, CD3 + CD8 + T cells, CD19 + B cells, and NK cells were performed as previously reported [ 12 , 17 , 18 ] with associated RFP + gating per cell population (Supplemental Fig.  1 and Fig.  1 ). The percentage of all respective cell populations was determined from live CD45 + lung leukocytes after excluding debris and doublets. This percentage was multiplied by the respective total lung cell numbers to determine specific cell population numbers for each animal.

Inhalation of organic dust extract (ODE), lipopolysaccharide (LPS), and peptidoglycan (PGN) induce lung CCR2 + monocyte-macrophages (Mɸ) and monocytes. CCR2 RFP/+ mice were exposed once to ODE (25%), LPS (10 μg), PGN (100 μg), or saline control and euthanized at 48 h. Scatter plots with bars depict mean with SD delineating cells as total (gray), CCR2 + (green), and CCR2 − (red). A Total lung cells enumerated. B Representative contour plot of the four monocyte (mono)-Mɸ subpopulations across groups based upon CD11c and CD11b expression after removal of neutrophils gated from live CD45 + cells after excluding debris and doublets. C RFP ± staining by exposure group and subpopulation. D CD11c + CD11b lo alveolar (Alv) Mɸ, CD11c + CD11b hi activated (Act) Mɸ, CD11c int CD11b hi mono-Mɸ, and CD11c − CD11b. hi monocytes determined by multiplying lung cell % population by total lung cells enumerated from lung sample. Statistical analyses were performed with Kruskal–Wallis with Dunn’s test for multiple comparisons (# p  < 0.05, ## p  < 0.01, ### p  < 0.001, #### p  < 0.0001) vs. respective saline. N  = 19 (saline), 9 (ODE), 8 (LPS), 5 (PGN)

Lung histopathology and post-translational modifications

Following removal of BALF and blood from the pulmonary vasculature, left lungs were excised and inflated to 15 cm H 2 O pressure with 10% formalin (Fisher Scientific, Fair Lawn, NJ) for 24 h to preserve pulmonary architecture [ 18 ]. Fixed left lung lobes were then placed into cassettes, embedded in paraffin, cut (4–5 μm) at midpoint sections to include regions of both large and small airways as well as blood vessels, and stained with hematoxylin and eosin (H&E) or preserved for subsequent IHC. H&E-stained slides of entire lung sections from each animal were reviewed at all scanning magnifications and semi-quantitatively scored for the degree and distribution of lung inflammation. Scores were generated by an expert pathologist blinded to treatment conditions utilizing a previously published scoring system (scored 0 to 4) [ 18 , 33 ] that evaluates the spectrum of inflammatory changes for alveolar and bronchiolar compartments with higher scores indicating greater inflammation.

Lung sections were also stained with modified Masson’s Trichrome and scanned by Aperio scanner (Leica Biosystems, Deer Park, IL). The VENTANA image viewer (version 3.1.4; Roche Diagnostics, Indianapolis, IN) was utilized to capture 10 images per lung section at 20 × from scanned images. Collagen content in trichrome images was quantified as previously described [ 12 , 34 ] using Image J FIJI plugin (version: 2.9.0/1.53t U.S. National Institutes of Health, Bethesda, MD).

To quantify CCR2 + expression of inflammatory monocyte/macrophages, lung sections were stained with anti-CCR2 (1:100, NBP267700, Lot HM0537, Novus, Littleton, CO) and cross absorbed with donkey anti-rabbit (AlexaFluor488, A21206, Lot #2,156,521, Thermo Fisher, Waltham, MA) at 1:100 and processed as previously described [ 12 ]. Slides were mounted with VECTASHIELD® Antifade Mounting Medium with DAPI (4′6-diamindino-2-phenylindole; to identify nuclei)(Cat#H-1200, Lot#ZG1014, Curlingame, CA). Using a Zeiss fluorescent microscope (Zeiss Observer.Z1 Zeiss, White Plains, NY), photographs (10 lung images per mouse) of lung parenchyma were taken, and CCR2 + expression by integrated density was quantified by Image J FIJI plugin.

Citrullinated (CIT) and malondialdehyde acetaldehyde (MAA) modified proteins and vimentin were stained [ 17 ]. Increased in the context of inflammatory lung diseases, vimentin is an extracellular matrix protein that is also targeted by post-translational modifications generated during the process of inflammation and increased oxidative stress. Prior studies by our group have demonstrated robust co-localization of MAA and CIT with vimentin in lung tissues of mice and humans with inflammatory arthritis and lung disease [ 17 , 35 ]. Lung sections were stained with Cy5 rabbit anti-vimentin (Bioss, Woburn, MA, USA, 1:100), Zenon AF 594 label (Invitrogen, Carlsbad, CA, USA) and rabbit polyclonal IgG antibody to MAA [ 17 ], or a mouse monoclonal anti-peptidyl-citrulline antibody (clone F95 IgMκ, Millipore Sigma, Burlington, MA, USA). Detection of the F95 antibody was done using an AF 488 AffiniPure donkey anti-mouse IgM, µ chain specific antibody (Jackson Immunoresearch, West Grove, PA, USA). DAPI (4′,6-diamidino-2-phenylindole; to identify nuclei) was added and samples were sealed with Fluormount-G (Southern Biotech, Birmingham, AL, USA). Fluorochromes detected using a Revolve fluorescent microscope (ECHO, San Diego, CA, USA). Images were quantified using ImageJ, and colocalization was performed using the Image J (RRID:SCR_003070) FIJI plugin Coloc 2 [ 17 , 18 ].

Statistical analysis

Sample-size requirements were extrapolated from previous work assessing post-LPS lung recovery treatments in C57BL/6 [ 12 ]. The mean (± SD) of CD11c int CD11b hi transitioning/recruited monocyte-macrophages was 0.26 × 10 5 (0.09 × 10 5 ) with saline and 6.5 × 10 5 (2.2 × 10 5 ) with LPS treatment 48 h post-exposure; thus, a sample size of N  = 2 in each group would achieve 80% power at the 0.05 level of significance to determine an influx of these cells following inflammatory agent exposure as compared to saline control. Experimental groups for the CCR2 trafficking studies include at least 2 mice for each group. The maximum sample sizes for the CCR2 RFP/+ trafficking studies are N  = 19 (saline/Sal), N  = 9 (ODE), N  = 8 (LPS), and N  = 5 (PGN). A sample size of n  = 5 would achieve 80% power at the 0.05 level of significance to detect a 60% reduction in CD11c int CD11b hi transitioning/recruited monocyte-macrophages with depletion strategies (i.e., CCR2 KO and clodronate liposomes). For the CCR2 WT vs. KO studies, N  = 5 (CCR2 WT-Sal), N  = 5 (CCR2 KO-Sal), N  = 9 (CCR2 WT-LPS) and N  = 9 (CCR2 KO-LPS); and for the clodronate (Clod) versus vehicle (Veh) liposome targeted studies, N  = 8 (Veh + Sal), 8 (Clod + Sal), 8 (Veh + LPS), and 9 (Clod + LPS). Experimental groups for those studies include at least 5 mice for each group. Numbers less than the maximum number reflect limitations in available sample quantity or quality.

Data are presented as the mean ± standard deviation (± SD) with scatter plots depicted for each data point. The Shapiro–Wilk test was utilized to test for normality among treatment groups. If the normality condition was satisfied, parametric statistical tests (one-way ANOVA with subsequent Tukey’s multiple comparison test), and if not satisfied, nonparametric statistical (Kruskal–Wallis with subsequent Dunn’s multiple comparison test) were used to assess differences between any two groups. All statistical analyses were performed using GraphPad Prism (version: 10.1.1) software and statistical significance accepted at a p value < 0.05.

Ethics statement

This study was conducted and reported in accordance with ARRIVE guidelines ( https://arriveguidelines.org ). All animal procedures were also approved by the University of Nebraska Medical Center (UNMC) Institutional Animal Care and Use Committee and were in accordance with the NIH guidelines for the use of rodents.

Inhalant exposures to organic dust extract (ODE), lipopolysaccharide (LPS), and peptidoglycan (PGN) induce lung infiltration of CCR2 + monocyte-macrophage (Mɸ) and monocyte subpopulations

In the first set of experiments, heterozygote CCR2 RFP/+ mice were treated once with ODE (25%), LPS (10 μg), PGN (100 μg), or sterile saline with lung tissue cell infiltrates analyzed at 48 h, as a previous study indicated that this was an optimal time point to detect recruited, infiltrating CD11c int CD11b hi transitioning monocytes-macrophages (Mɸ) following acute LPS treatment [ 12 ]. There were significant increases ( p  < 0.05) in total cells, CCR2 + cells, and CCR2 − cells following ODE, LPS, and PGN as compared to saline control with no difference across the treatment groups (Fig.  1 A). The 4 monocyte-macrophage subpopulations including alveolar (Alv) Mɸ (CD11c + CD11b lo ), activated (Act) Mɸ (CD11c + CD11b hi ), transitioning monocyte-Mɸ (CD11c int CD11b hi ), and monocytes (CD11c − CD11b hi ) were delineated as previously described [ 12 ], with representative contour plots shown in Fig.  1 B. CCR2 RFP + and CCR2 RFP − expression in each of the four monocyte/Mɸ subpopulations by treatment condition are depicted in Fig.  1 C. CCR2 expression was absent on the Alv Mɸ and Act Mɸ subpopulations but was present on transitioning monocyte-Mɸ and monocyte subpopulations with numbers of these cell subpopulations enumerated in Fig.  1 D. The numbers of CCR2 + and CCR2 − transitioning monocyte-Mɸ were significantly increased with ODE, LPS, and PGN treatment as compared to saline control ( p  < 0.05), but the magnitude of the increase was strikingly greater for the CCR2 + (vs. CCR2 − ) transitioning monocyte-Mɸ cells. There were also significant ( p  < 0.05) increases in the numbers of CCR2 + and CCR2 − monocytes with ODE, LPS, and PGN treatment as compared to saline control with similar magnitude of increases for both CCR2 + and CCR2 − monocytes. CCR2 − Act Mɸ were increased with ODE, LPS, and PGN vs. saline, and correspondingly, CCR2 − Alv Mɸ were decreased with ODE, LPS, and PGN vs. saline (Fig.  1 D). Although differences vs. saline were demonstrated, there was no difference in the numbers of the monocytes/Mɸ among ODE, LPS, and PGN. Thus, all environmental exposures examined increased CCR2 + transitioning monocyte-Mɸ and monocyte subpopulations, but there were also increases in CCR2 − monocyte subpopulations and to a lesser degree CCR2 − transitioning monocyte-Mɸ.

Cell surface expression of Ly6C and F4/80 with monocyte/macrophage (Mɸ) subpopulations following inhalant exposures to ODE, LPS, and PGN

Cell surface expression of Ly6C, a predominant marker of monocytes and/or associated with pro-inflammatory and pro-fibrotic properties [ 36 ] by percent expression and mean fluorescence intensity (MFI) were also investigated and summarized (Fig.  2 ). Ly6C expression was low (< 5%) on Sal-Alv Mɸ and ODE-, LPS-, PGN-Act Mɸ (data not shown). In contrast, Ly6C percent and MFI expression were increased on both CCR2 + and CCR2 − ODE-, LPS-, PGN- induced CD11c int CD11b hi monocyte-Mɸ cells vs. saline control (except MFI expression was not increased for these CCR2 − cells following PGN exposure) (Fig.  2 A, B). Moreover, Ly6C MFI expression was significantly ( p  < 0.05) increased on these CCR2 + monocyte-Mɸ cells associated with ODE, LPS, and PGN exposure as compared to the corresponding CCR2 − monocyte-Mɸ cells. Ly6C percent expression was high on all monocyte populations with a significant ( p  < 0.05) increase with LPS-associated CCR2 + and CCR2 − monocytes vs. saline (Fig.  2 A, B). There was an increase in Ly6C MFI expression with ODE and LPS CCR2 − monocytes vs. saline with no difference in intensity of the MFI expression across CCR2 RFP + monocytes. As observed with transitioning monocyte-Mɸ cells, MFI expression was increased in all CCR2 + monocytes as compared to CCR2 − monocytes. These studies demonstrate that Ly6C expression was increased in the recruited CCR2 + cells as well as CCR2 − cells following exposure to environmental agents, and as such, Ly6C alone may not discriminate monocyte-macrophage subpopulations.

Ly6C expression of monocyte-macrophage (Mɸ) and monocyte subpopulations following organic dust extract (ODE), lipopolysaccharide (LPS), and peptidoglycan (PGN) exposure. C57BL/6 mice were exposed once to ODE (25%), LPS (10 μg), PGN (100 μg), or saline control and euthanized at 48 h. Scatter plots with bars depict mean with SD delineating cells as CCR2 + (green) and CCR2- (red). Expression of Ly6C by percent ( A ) and mean fluorescence intensity (MFI) ( B ) across CD11cintCD11bhi monocyte-Mɸ and CD11c-CD11b + monocyte subpopulations as determined by flow cytometry. Statistical analyses were performed with Kruskal–Wallis with Dunn’s test for multiple comparisons (# p  < 0.05 vs. respective saline) and (* p  < 0.05 denoted by line with brackets denoting difference between same inhalant exposure by CCR2 RFP positive vs. negative). N  = 15 (saline), 9 (ODE), 4 (LPS), 5 (PGN)

The monocyte and macrophage marker F4/80 (ADGRE1) [ 37 ] was also investigated across cell subpopulations (Supplemental Fig.  2 ). The percent F4/80 expression was ubiquitous across Sal-Alv Mɸ and ODE-, LPS-, PGN-Act Mɸ subpopulations, but the expression intensity by MFI was increased in the ODE-, LPS-, PGN-Act Mɸ vs. Saline-Alv Mɸ (Supplemental Fig.  2 ). The percent F4/80 expression was high (~ 60–80%) in the transitioning CCR2 + and CCR2 − CD11c int CD11b hi monocyte-macrophage cells (~ 60–80%) and more variable with CD11c − CD11b hi monocytes.

Inhalant exposures to ODE and LPS induce lung infiltration of CCR2 +NK cells and T cells in addition to CCR2 − neutrophils, B cells, and T cells

The number of CCR2 + and CCR2 − neutrophils and lymphocytes was also investigated to capture any non-monocyte/macrophage cell-specific CCR2 expression events at 48 h post environmental agent exposure (Fig.  3 ). CCR2 is recognized to be expressed with NK cells [ 38 ] and activated T cells [ 39 ]. Indeed, there were significant ( p  < 0.05) increases in CCR2 + NK cells, CD3 + CD4 + T cells, and CD3 + CD8 + T cells with ODE and LPS but not PGN treatment vs. saline. There were also significant ( p  < 0.05) increases in CCR2 − CD3 + CD4 + T cells and CCR2 − CD3 + CD8 + T cells. ODE, LPS, and PGN treatment did not increase CCR2 + neutrophils or CD19 + B cells. ODE, LPS, and PGN treatment also increased CCR2 − neutrophils and CD4 + T cells, and ODE and LPS (but not PGN) increased CCR2 − CD8 + T cells.

Organic dust extract (ODE) and lipopolysaccharide (LPS) inhalation induce lung infiltration of CCR2 + NK and T cells. C57BL/6 mice were exposed once to ODE (25%), LPS (10 μg), PGN (100 μg), or saline control and euthanized at 48 h. Scatter plots with bars depict mean with SD delineating cells as total (gray), CCR2 + (green), and CCR2- (red). CD11c-Ly6G + neutrophils, CD19 + B cells, CD3-NK1.1 + NK cells, CD3 + CD4 + T cells, and CD3 + CD8 + T cell infiltrates determined by flow cytometry on live CD45 + cells after exclusion of debris and doublets with lung cell % populations multiplied by total lung cells enumerated from lung sample. Gating strategy depicted in Supplemental Fig.  1 . Statistical analyses were performed with Kruskal–Wallis with Dunn’s test for multiple comparisons (# p  < 0.05, ## p  < 0.01, ### p  < 0.001, #### p  < 0.0001) vs. respective saline. N  = 19 (saline), 9 (ODE), 8 (LPS), 5 (PGN)

Inhalant LPS-induced lung inflammatory responses are not reduced in CCR2 knock-out (KO) mice

Because there were no major differences in ODE-, LPS-, and PGN-induced CCR2 + monocyte-macrophage lung cell infiltrates, LPS was utilized as the prototype environmental inflammatory agent for the remainder of the studies. It was hypothesized that CCR2 knock-out (KO) mice would be protected against LPS-induced lung inflammatory and pro-fibrotic responses due to reduction in the recruitment of transitioning CCR2 + monocyte-Mɸ infiltrates. Although there were significant ( p  = 0.0005) reductions (54% reduction) in LPS-induced CD11c int CD11b hi monocyte-Mɸ cells, there were no significant reductions in LPS-induced pro-inflammatory and pro-fibrotic mediators in lung homogenates including TNF-α, IL-6, CXCL1, MMP-3, MMP-8, and TIMP-1 (Table  1 ) and lung histopathology (data not shown) in CCR2 KO mice vs. WT mice. There were also no differences between CCR2 KO and WT mice for LPS-induced total cells, neutrophils, lymphocytes, and macrophages in BALF as well as LPS-induced lung infiltrates including activated Mɸ, monocytes, T and B lymphocytes, and NK cells. In contrast, LPS-induced lung neutrophils were increased in CCR2 KO (vs. WT) mice, and moreover, there were corresponding, likely compensatory, increases in lung and serum CCL2 and CCL7, chemoattractants that predominately drive monocyte recruitment but also can affect lymphocytes and neutrophils, with the LPS treated CCR2 KO mice.

LPS-induced lung transitioning, infiltrating CD11c int CD11b hi are reduced with systemic delivery of clodronate liposomes

In an alternative approach to deplete recruited lung monocytes-macrophages induced by environmental exposures, intravenous clodronate liposomes (vs. vehicle control liposomes) were dosed one day prior to LPS (and saline) treatment to reduce circulating/systemic reservoir of available monocytes-macrophages with mice euthanized at 48 h following LPS exposure. In these studies, there was a reduction in total lung cells in tissue homogenates associated with a 59% reduction in LPS-induced lung CD11c int CD11b hi monocyte-Mɸ infiltrates in clodronate liposome pre-treated mice compared to control-treated with mice (Fig.  4 ). In contrast, there were no treatment differences in the number of LPS-induced Alv Mɸ, Act Mɸ or monocytes. There were also no treatment differences in the number of inflammatory cells in BALF and no treatment differences in the number of, CD19 + B cells, CD4 + and CD8 + T cells, and NK cells in tissue homogenates following LPS exposure (Supplemental Table  1 ).

LPS-induced transitioning CD11c int CD11b hi monocytes/macrophages are reduced with systemic delivery of clodronate liposomes. Mice were pre-treated with vehicle (Veh) or clodronate (Clod) liposomes 24 h prior to a one-time treatment with LPS (10 μg) or saline (Sal) control and euthanized at 48 h. Scatter plot with bars depicting mean with SD. A Total lung cells enumerated. B Representative contour plot of the four monocyte (mono)-Mɸ subpopulations across groups based upon CD11c and CD11b expression after removal of neutrophils gated from live CD45 + cells after excluding debris and doublets. C Number of CD11c + CD11b lo alveolar (Alv) Mɸ, CD11c + CD11b hi activated (Act) Mɸ, CD11c int CD11b hi mono-Mɸ, and CD11c − CD11b. hi monocytes determined by multiplying lung cell % population by total lung cells enumerated from lung sample. (# p  < 0.05 vs. respective saline) and (* p  < 0.05 denoted by line with brackets denoting difference between groups). N  = 8 (Veh + Sal), 8 (Clod + Sal), 8 (Veh + LPS), 9 (Clod + LPS)

Effects of systemic delivery of clodronate liposomes with LPS-induced lung inflammation, collagen deposition, and infiltrating CCR2 + cells

Lung sections from these same mice pre-treated with vehicle and clodronate liposomes followed by saline and LPS challenge were evaluated for histopathological changes by H&E, collagen deposition by trichrome staining, and CCR2 + cell infiltrates (Fig.  5 A). Although semi-quantitative inflammatory scores following LPS exposure were reduced with clodronate liposome pre-treatment compared to vehicle control (Fig.  5 B), this difference did not reach statistical significance. LPS-induced collagen deposition was reduced by 23% ( p  < 0.05) with clodronate liposome pretreatment (Fig.  5 C). Moreover, LPS-induced CCR2 + cell infiltrates were reduced by 60% ( p  < 0.05) with clodronate liposome pretreatment (Fig.  5 D), consistent with reductions observed in CD11c int CD11b + monocyte- Mɸ demonstrated by flow cytometry.

Effects of systemic clodronate liposome delivery with LPS-induced lung inflammation, collagen, and infiltrating CCR2 + cells. Mice were pre-treated with vehicle (Veh) or clodronate (Clod) liposomes 24 h prior to a one-time treatment with LPS (10 μg) or saline (Sal) control and euthanized at 48 h. A Representative images from treatment groups stained by H&E, trichome, and CCR2 (red) with DAPI nuclei staining (blue) by confocal microscopy. Scatter plots with bars depict mean with SD of semi-quantitative lung inflammatory score ( B ) and integrated density of collagen ( C ), and CCR2 ( D ) quantified per each mouse. Statistical analyses were performed with Kruskal–Wallis with Dunn’s multiple comparison (inflammatory scores) and ANOVA with Tukey’s multiple comparison (collagen content and CCR2) (# p  < 0.05 vs. respective saline) and (* p  < 0.05 denoted by line with brackets denoting difference between groups). N  = 8 (Veh + Sal), 8 (Clod + Sal), 8 (Veh + LPS), 9 (Clod + LPS). Line scale denotes 100 μm

LPS-induced lung pro-fibrotic and inflammatory mediators modulated following systemic delivery of clodronate liposomes

Pre-treatment with intravenous clodronate liposomes (vs. vehicle) also resulted in significant reductions in LPS-induced levels of pro-fibrotic mediators in lung homogenates including MMP-3 (33% reduction), MMP-8 (50% reduction), TIMP-1 (64% reduction), and TGF-β (38% reduction) (Fig.  6 ). Moreover, there were also significant ( p  < 0.05) reductions in LPS-induced pro-inflammatory mediators including IL-6 (72% reduction) and neutrophil chemoattractant CXCL1 (57% reduction) with clodronate (vs. vehicle) liposome pre-treatment (Fig.  6 ). Lung levels of TNF-α induced by LPS exposure were not reduced with clodronate liposome pre-treatment, and there were also no differences in LPS-induced lung CCL2 and CCL7 levels between clodronate and vehicle liposome pre-treatment (Supplemental Table  1 ).

LPS-induced lung pro-fibrotic and inflammatory mediators modulated following systemic delivery of clodronate liposomes. Mice were pre-treated with vehicle (Veh) or clodronate (Clod) liposomes 24 h prior to a one-time treatment with LPS (10 μg) or saline (Sal) control and euthanized at 48 h. Scatter plots with bars depict mean with SD of protein levels of matrix metalloproteinase (MMP)-3, MMP-8, metalloproteinase inhibitor (TIMP-1), transforming growth factor (TGF)-β, IL-6, and neutrophil chemokine CXCL1 of lung homogenate. Statistical analyses were performed with ANOVA and Tukey’s multiple comparison test with significance (# p  < 0.05 vs. respective saline) and (* p  < 0.05 denoted by line with brackets denoting difference between groups) with % reduction noted. N  = 8 (Veh + Sal), 8 (Clod + Sal), 8 (Veh + LPS), 9 (Clod + LPS)

LPS-induced lung CIT and MAA autoantigens and vimentin expression were reduced with systemic delivery of clodronate liposomes

Based upon findings of decreased pro-fibrotic mediators and prior findings demonstrating that repetitive inhalant environmental exposures induce post-translational changes implicated in inflammatory and fibrotic lung disease [ 17 , 18 ], lung tissues were stained for CIT- and MAA-modified antigens as well as vimentin. CIT- and MAA-modified proteins and vimentin were significantly increased following a one-time LPS exposure vs. saline control at 48 h post-LPS exposure (Fig.  7 A, B). Moreover, there were significant ( p  < 0.05) reductions in LPS-induced lung CIT-modified protein expression (39% reduction), MAA-modified protein expression (48% reduction), and vimentin (40% reduction) with clodronate (vs. vehicle) liposome administration.

LPS-induced lung CIT and MAA autoantigens and vimentin expression decrease with systemic clodronate liposome delivery. Mice were pre-treated with vehicle (Veh) or clodronate (Clod) liposomes 24 h prior to a one-time treatment with LPS (10 μg) or saline (Sal) control and euthanized at 48 h. A Representative confocal microscopy images of lung tissue from treatment groups stained for citrulline (CIT, green) and malondialdehyde-acetaldehyde (MAA, red) modified proteins and vimentin (teal). Line scale denotes 70 μm. B Scatter plots with bars depict mean with SD of integrated density of CIT- and MAA-modified proteins and vimentin quantified per each mouse. Statistical analyses were performed with ANOVA with Tukey’s for multiple comparisons (# p  < 0.05 vs. respective saline) and (* p  < 0.05 denoted by line with brackets denoting difference between groups). N  = 5 (Veh + Sal), 7 (Clod + Sal), 8 (Veh + LPS), 9 (Clod + LPS)

Lung disease represents a major cause of occupation-related illness for which therapeutic approaches to alleviate disease burden are lacking. Recruited, infiltrating, and transitioning monocytes-macrophages have been implicated as critical cells in the immunopathogenesis of chronic lung disease. Here, our preclinical animal studies first defined and compared the trafficking of the inflammatory CCR2 + monocytes/macrophages to the lung following organic dust, endotoxin, and peptidoglycan exposure. These studies demonstrated striking increases in CCR2 + recruited/transitioning CD11c int CD11b hi monocyte-Mɸ and CD11c − CD11b hi monocyte subpopulations as well as striking increases in CCR2 − monocytes with these environmental exposures. However, CCR2 KO mice were not protected against inflammatory responses induced following endotoxin exposure despite a reduction (54%) in exposure induced CCR2 + recruited/transitioning CD11c int CD11b hi cells. Instead, systemic depletion of monocytes by intravenous clodronate liposome administration was associated with not only a reduction (59% reduction) in these transitioning monocyte-macrophages, but also a corresponding reduction in endotoxin-induced collagen deposition, extracellular matrix release, vimentin, and lung autoantigen expression. Thus, preventing the influx of circulating/recruited monocytes to the lung, as opposed to specifically targeting CCR2, following environmental exposures may represent a strategic area to further develop to reduce disease burden in occupationally exposed at-risk persons.

Occupational and environmental exposures are inherently complex, with highly concentrated exposures commonplace in a variety of settings, which can initiate and perpetuate the development of lung disease. In our studies aimed at defining the trafficking of monocytes-macrophages to the lung, we observed similar effects following separate exposures to high concentrations of complex organic dust extract (ODE), LPS, and PGN, all of which resulted in a robust increase in monocyte-macrophage recruitment. This supports the current focus of therapeutic approaches for environmental and occupational-associated diseases that have centered on specific agent identification and risk reduction measures and mitigation [ 40 ]. Supporting the relevance of experiments focused on high exposure doses, endotoxin levels encountered in real-world settings are highly variable and often exceed occupational exposure limits in agriculture settings by several orders of magnitude [ 41 , 42 ].

Lung monocytes-macrophages are important in mediating the response to inflammatory bioaerosol exposure, and recruited/transitional monocytes-macrophages specifically are implicated in the transition of acute inflammation to lung fibrosis in animal models and clinical investigations [ 43 ]. CCR2-expressing leukocytes are required for the progression of bleomycin-induced fibrosis, as CCR2 KO mice were protected against collagen deposition, macrophage infiltration, and MMP deposition in this model [ 44 , 45 ]. Likewise, pre-treatment with a CCR2 antagonist reduced lung fibrosis in a mouse model of scleroderma [ 46 ]. Although lung injury promotes the release of CCL2 by several different lung cell types, including airway epithelial cells, and serves as a chemoattractant for CCR2 pro-fibrotic macrophages [ 47 , 48 ], CCL2 neutralizing antibodies did not alter disease progression or mitigate lung function decline in patients with idiopathic pulmonary fibrosis, but instead was associated with an increase in endogenous CCL2 expression and other adverse compensatory changes [ 49 ]. We also found that CCL2 levels were increased in LPS-exposed CCR2 KO mice. The lack of efficacy observed with CCL2 targeting is consistent with our results in which CCL2 KO mice did not appear to be meaningfully protected from LPS-induced lung injury.

In endotoxin exposure models of lung injury, several studies have also demonstrated that there is a decrease in the endotoxin-induced recruitment of peripheral blood monocytes, exudative macrophages, F4/80 + lung cells, lung neutrophils, and lung cell infiltrates in CCR2 KO mice as well as mice depleted of systemic monocytes by clodronate liposomes [ 25 , 50 , 51 , 52 , 53 , 54 ]. Our studies confirm the reduction in infiltrating monocytes/macrophages with both the CCR2 KO and the administration clodronate liposomes, but also simultaneously highlight striking differences in these modeling approaches to reduce profibrotic and inflammatory responses following LPS exposure. In addition to finding no meaningful benefit for globally depleting CCR2-expressing cells in endotoxin-induced lung disease, we demonstrated that CCR2 KO mice had increased neutrophil accumulation with compensatory increases in CCL2 and CCL7 following LPS exposure, suggesting this strategy may actually yield detrimental effects.

Our observations are consistent with a study by Gurczynski and colleagues demonstrating no protection against ɣ-herpesvirus-induced pneumonitis and fibrosis with CCR2 KO mice and that CCR2 + cells played a suppressive role by limiting collagen and IL-17 production [ 55 ]. Another potential compensatory mechanism supported by this study is that the environmental exposures induced the expression of CCR2 + on the CD4 + and CD8 + T-cell infiltrates. Others have demonstrated suppressive roles for CCR2-expressing T cells in lung infections and inflammatory responses [ 56 ]. Moreover, CCR2 is highly expressed on NK cells, and both ODE and LPS induced the recruitment of NK cells. Thus, depletion of CCR2 function with lymphocytes may have negated any potential benefit of reducing CCR2 + monocytes-macrophages. Based on these findings, further investigations are warranted to understand the functional role of CCR2 on infiltrating lymphocytes in the setting of environmental exposure-induced lung disease.

In contrast to CCR2 targeting, the strategy of inhibiting the recruitment of peripheral blood monocytes to the lung through the systemic administration of clodronate liposomes demonstrated benefit with most, but not all, endpoints examined. Beneficial responses are best characterized by a reduction in the pro-fibrotic properties of lung monocytes-macrophages. Namely, endotoxin-induced collagen deposition, MMP3, MMP8, TIMP-1, TGF-β, IL-6, and CXCL1 (but not TNF-⍺) were reduced by depleting peripheral blood monocytes. This corresponded with reductions observed in CIT- and MAA-modified protein generation following endotoxin exposure. This is relevant, as proteins co-modified with CIT and MAA stimulate undifferentiated macrophages towards a mixed M1/M2 phenotype [ 57 ] and secrete soluble factors that drive an aggressive fibroblast phenotype [ 20 ]. Whereas there was no compensatory increase in CCL2 or CCL7 release following the administration of clodronate liposomes as observed with CCR2 KO mice, neither strategy appeared to mitigate the histopathologic changes initiated by acute LPS exposure.

These collective findings underscore the importance of the recruited peripheral blood monocyte transitioning to a lung monocyte/macrophage population in mediating pro-fibrotic processes in the lung as well as generation of post-translationally modified proteins following endotoxin exposure. Therefore, depletion and/or inhibition of “recruitable” peripheral blood monocytes may represent a novel strategy to reduce the burden of lung diseases resulting from select environmental exposures. Whereas CCR2 abolition did not exhibit protection against LPS-induced inflammatory or pro-fibrotic responses, antagonism of other monocyte trafficking receptors (i.e., CCR1, 5, 6, 7), blockade of adhesion molecules (i.e., selectins, integrins, ICAM-1/VCAM-1), or inhibition of chemokines (i.e., CCL2, CCL5, CCL7) may demonstrate therapeutic benefit given depletion of the recruitable monocyte reservoir demonstrated protective effects in our model of acute exposure-induced lung inflammation [ 58 , 59 , 60 ]. Moreover, understanding the mechanisms governing the crosstalk between lung monocytes and airway structural cells including fibroblasts/myofibroblasts is warranted.

In conclusion, high concentration exposure to environmental and occupational exposures including agricultural organic dust extracts, endotoxin, and peptidoglycan induce the recruitment of CCR2 + and CCR2 − peripheral blood monocytes transitioning to lung resident monocytes/macrophages. Depleting peripheral blood monocytes by systemic administration of clodronate liposomes, but not through CCR2 KO animal strategies, resulted in the reduction of endotoxin-induced pro-inflammatory and pro-fibrotic mediators. Developing translational strategies to reduce the recruitment of these cells following exposures may be warranted to reduce risk of lung disease.

Availability of data and materials

Data that support the findings of this study have been deposited in Zenodo. The data are embargoed until manuscript acceptance for publication. The link to the data is below. https://zenodo.org/records/10641513?token=eyJhbGciOiJIUzUxMiJ9.eyJpZCI6IjNjYmNhNDA0LThmMDYtNDBmZS1iMTc2LTZkYTQ1NDZjZjExOCIsImRhdGEiOnt9LCJyYW5kb20iOiIzODkyNWQ0MmUxYzdlNzE4Zjg5YWRkZDVhMzllMTYzNiJ9.KUYVheTKH7IITATY9EiCchsVBH4Db6OoaIi-c2fAakjMpTypBhSTuAHeXMnz-9qoeyVrUzXboA0T7EpsPGMwVw .

Seaman DM, Meyer CA, Kanne JP. Occupational and environmental lung disease. Clin Chest Med. 2015;36(2):249–68 viii-ix.

Article   PubMed   Google Scholar  

Després V, Huffman JA, Burrows SM, Hoose C, Safatov A, Buryak G, et al. Primary biological aerosol particles in the atmosphere: a review. Tellus B Chem Phys Meteorol. 2012;64(1):15598.

Article   Google Scholar  

Park Y, Ahn C, Kim TH. Occupational and environmental risk factors of idiopathic pulmonary fibrosis: a systematic review and meta-analyses. Sci Rep. 2021;11(1):4318.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Schwab AD, Poole JA. Mechanistic and therapeutic approaches to occupational exposure-associated allergic and non-allergic asthmatic disease. Curr Allergy Asthma Rep. 2023;23(6):313–24.

Lee CT, Feary J, Johannson KA. Environmental and occupational exposures in interstitial lung disease. Curr Opin Pulm Med. 2022;28(5):414–20.

Article   CAS   PubMed   Google Scholar  

Blanc PD, Annesi-Maesano I, Balmes JR, Cummings KJ, Fishwick D, Miedinger D, et al. The occupational burden of nonmalignant respiratory diseases. An official American thoracic society and european respiratory society statement. Am J Respir Crit Care Med. 2019;199(11):1312–34.

Article   PubMed   PubMed Central   Google Scholar  

Perlman DM, Maier LA. Occupational lung disease. Med Clin North Am. 2019;103(3):535–48.

Collaborators GOCRRF. Global and regional burden of chronic respiratory disease in 2016 arising from non-infectious airborne occupational exposures: a systematic analysis for the Global Burden of Disease Study 2016. Occup Environ Med. 2020;77(3):142–50.

Collaborators GOCRRF. Global burden of chronic respiratory diseases and risk factors, 1990–2019: an update from the global burden of disease study 2019. EClinicalMedicine. 2023;59:101936.

Basinas I, Sigsgaard T, Kromhout H, Heederik D, Wouters IM, Schlunssen V. A comprehensive review of levels and determinants of personal exposure to dust and endotoxin in livestock farming. J Expo Sci Environ Epidemiol. 2015;25(2):123–37.

Moller W, Heimbeck I, Hofer TP, Khadem Saba G, Neiswirth M, Frankenberger M, et al. Differential inflammatory response to inhaled lipopolysaccharide targeted either to the airways or the alveoli in man. PLoS ONE. 2012;7(4):e33505.

Poole JA, Gaurav R, Schwab A, Nelson AJ, Gleason A, Romberger DJ, et al. Post-endotoxin exposure-induced lung inflammation and resolution consequences beneficially impacted by lung-delivered IL-10 therapy. Sci Rep. 2022;12(1):17338.

Rolph CA, Gwyther CL, Tyrrel SF, Nasir ZA, Drew GH, Jackson SK, et al. Sources of airborne endotoxins in ambient air and exposure of nearby communities—a review. Atmosphere. 2018;9(10):375.

Article   CAS   Google Scholar  

Tager IB, Lurmann FW, Haight T, Alcorn S, Penfold B, Hammond SK. Temporal and spatial patterns of ambient endotoxin concentrations in Fresno, California. Environ Health Perspect. 2010;118(10):1490–6.

Park S, Allen RJ, Lim CH. A likely increase in fine particulate matter and premature mortality under future climate change. Air Qual Atmos Health. 2020;13(2):143–51.

Joshi M, Goraya H, Joshi A, Bartter T. Climate change and respiratory diseases: a 2020 perspective. Curr Opin Pulm Med. 2020;26(2):119–27.

Poole JA, Mikuls TR, Thiele GM, Gaurav R, Nelson AJ, Duryee MJ, et al. Increased susceptibility to organic dust exposure-induced inflammatory lung disease with enhanced rheumatoid arthritis-associated autoantigen expression in HLA-DR4 transgenic mice. Respir Res. 2022;23(1):160.

Mikuls TR, Gaurav R, Thiele GM, England BR, Wolfe MG, Shaw BP, et al. The impact of airborne endotoxin exposure on rheumatoid arthritis-related joint damage, autoantigen expression, autoimmunity, and lung disease. Int Immunopharmacol. 2021;100:108069.

Poole JA, Thiele GM, Ramler E, Nelson AJ, Duryee MJ, Schwab AD, et al. Combined repetitive inhalant endotoxin and collagen-induced arthritis drives inflammatory lung disease and arthritis severity in a testosterone-dependent manner. Am J Physiol Lung Cell Mol Physiol. 2024;326:L239–L251.

Aripova N, Duryee MJ, England BR, Hunter CD, Mordeson JE, Ryan EM, et al. Citrullinated and malondialdehyde-acetaldehyde modified fibrinogen activates macrophages and promotes an aggressive synovial fibroblast phenotype in patients with rheumatoid arthritis. Front Immunol. 2023;14:1203548.

Warren KJ, Wyatt TA, Romberger DJ, Ailts I, West WW, Nelson AJ, et al. Post-injury and resolution response to repetitive inhalation exposure to agricultural organic dust in mice. Safety (Basel). 2017;3(1):10.

Laskin DL, Malaviya R, Laskin JD. Role of macrophages in acute lung injury and chronic fibrosis induced by pulmonary toxicants. Toxicol Sci. 2019;168(2):287–301.

Okabe Y, Medzhitov R. Tissue biology perspective on macrophages. Nat Immunol. 2016;17(1):9–17.

David BA, Kubes P. Exploring the complex role of chemokines and chemoattractants in vivo on leukocyte dynamics. Immunol Rev. 2019;289(1):9–30.

Oliveira VLS, Pollenus E, Berghmans N, Queiroz-Junior CM, Blanter M, Mattos MS, et al. Absence of CCR2 promotes proliferation of alveolar macrophages that control lung inflammation in acute respiratory distress syndrome in mice. Int J Mol Sci. 2022;23(21):12920.

Poole JA, Dooley GP, Saito R, Burrell AM, Bailey KL, Romberger DJ, et al. Muramic acid, endotoxin, 3-hydroxy fatty acids, and ergosterol content explain monocyte and epithelial cell inflammatory responses to agricultural dusts. J Toxicol Environ Health Part A. 2010;73(10):684–700.

Nelson AJ, Roy SK, Warren K, Janike K, Thiele GM, Mikuls TR, et al. Sex differences impact the lung-bone inflammatory response to repetitive inhalant lipopolysaccharide exposures in mice. J Immunotoxicol. 2018;15(1):73–81.

Poole JA, Gleason AM, Bauer C, West WW, Alexis N, Reynolds SJ, et al. alphabeta T cells and a mixed Th1/Th17 response are important in organic dust-induced airway disease. Ann Allergy Asthma Immunol. 2012;109(4):266–73.e2.

Tatham KC, O’Dea KP, Romano R, Donaldson HE, Wakabayashi K, Patel BV, et al. Intravascular donor monocytes play a central role in lung transplant ischaemia-reperfusion injury. Thorax. 2018;73(4):350–60.

Zhang L, Xie W, Zhang J, Shanahan H, Tonello R, Lee SH, et al. Key role of CCR2-expressing macrophages in a mouse model of low back pain and radiculopathy. Brain Behav Immun. 2021;91:556–67.

Zito MA, Koennecke LA, McAuliffe MJ, McNally B, van Rooijen N, Heyes MP. Depletion of systemic macrophages by liposome-encapsulated clodronate attenuates striatal macrophage invasion and neurodegeneration following local endotoxin infusion in gerbils. Brain Res. 2001;892(1):13–26.

Ribeiro-Machado C, Santos SG, Amaral IA, Caldeira J, Pereira P, Barbosa MA, et al. Macrophage-based therapy for intervertebral disc herniation: preclinical proof-of-concept. NPJ Regen Med. 2023;8(1):34.

Tantawy MA, Hatesuer B, Wilk E, Dengler L, Kasnitz N, Weiß S, et al. The interferon-induced gene Ifi27l2a is active in lung macrophages and lymphocytes after influenza A infection but deletion of Ifi27l2a in mice does not increase susceptibility to infection. PLoS ONE. 2014;9(9):e106392.

Chen Y, Yu Q, Xu C-B. A convenient method for quantifying collagen fibers in atherosclerotic lesions by ImageJ software. Int J Clin Exp Med. 2017;10(10):14904–10.

Google Scholar  

England BR, Duryee MJ, Roul P, Mahajan TD, Singh N, Poole JA, et al. Malondialdehyde-acetaldehyde adducts and antibody responses in rheumatoid arthritis-associated interstitial lung disease. Arthritis Rheumatol. 2019;71(9):1483–93.

Gibbons MA, Mackinnon AC, Ramachandran P, Dhaliwal K, Duffin R, Phythian-Adams AT, et al. Ly6Chi monocytes direct alternatively activated pro-fibrotic macrophage regulation of lung fibrosis. Am J Respir Crit Care Med. 2011;184:569–81.

Waddell LA, Lefevre L, Bush SJ, Raper A, Young R, Lisowski ZM, et al. ADGRE1 (EMR1, F4/80) is a rapidly-evolving gene expressed in mammalian monocyte-macrophages. Front Immunol. 2018;9:2246.

Fujimura N, Xu B, Dalman J, Deng H, Aoyama K, Dalman RL. CCR2 inhibition sequesters multiple subsets of leukocytes in the bone marrow. Sci Rep. 2015;5:11664.

Bakos E, Thaiss CA, Kramer MP, Cohen S, Radomir L, Orr I, et al. CCR2 Regulates the immune response by modulating the interconversion and function of effector and regulatory T cells. J Immunol. 2017;198(12):4659–71.

Vlahovich KP, Sood A. A 2019 update on occupational lung diseases: a narrative review. Pulm Ther. 2021;7(1):75–87.

Davidson ME, Schaeffer J, Clark ML, Magzamen S, Brooks EJ, Keefe TJ, et al. Personal exposure of dairy workers to dust, endotoxin, muramic acid, ergosterol, and ammonia on large-scale dairies in the high plains Western United States. J Occup Environ Hyg. 2018;15(3):182–93.

Roque K, Shin KM, Jo JH, Lim GD, Song ES, Shin SJ, et al. Association between endotoxin levels in dust from indoor swine housing environments and the immune responses of pigs. J Vet Sci. 2018;19(3):331–8.

Ogawa T, Shichino S, Ueha S, Matsushima K. Macrophages in lung fibrosis. Int Immunol. 2021;33(12):665–71.

Baran CP, Opalek JM, McMaken S, Newland CA, O’Brien JM Jr, Hunter MG, et al. Important roles for macrophage colony-stimulating factor, CC chemokine ligand 2, and mononuclear phagocytes in the pathogenesis of pulmonary fibrosis. Am J Respir Crit Care Med. 2007;176(1):78–89.

Okuma T, Terasaki Y, Kaikita K, Kobayashi H, Kuziel WA, Kawasuji M, et al. C-C chemokine receptor 2 (CCR2) deficiency improves bleomycin-induced pulmonary fibrosis by attenuation of both macrophage infiltration and production of macrophage-derived matrix metalloproteinases. J Pathol. 2004;204(5):594–604.

Ishikawa M, Yamamoto T. Antifibrogenic effects of C-C chemokine receptor type 2 antagonist in a bleomycin-induced scleroderma model. Exp Dermatol. 2021;30(1):179–84.

Osterholzer JJ, Olszewski MA, Murdock BJ, Chen GH, Erb-Downward JR, Subbotina N, et al. Implicating exudate macrophages and Ly-6C(high) monocytes in CCR2-dependent lung fibrosis following gene-targeted alveolar injury. J Immunol. 2013;190(7):3447–57.

Yang J, Velikoff M, Canalis E, Horowitz JC, Kim KK. Activated alveolar epithelial cells initiate fibrosis through autocrine and paracrine secretion of connective tissue growth factor. Am J Physiol Lung Cell Mol Physiol. 2014;306(8):L786–96.

Raghu G, Martinez FJ, Brown KK, Costabel U, Cottin V, Wells AU, et al. CC-chemokine ligand 2 inhibition in idiopathic pulmonary fibrosis: a phase 2 trial of carlumab. Eur Respir J. 2015;46(6):1740–50.

Dhaliwal K, Scholefield E, Ferenbach D, Gibbons M, Duffin R, Dorward DA, et al. Monocytes control second-phase neutrophil emigration in established lipopolysaccharide-induced murine lung injury. Am J Respir Crit Care Med. 2012;186(6):514–24.

Maus UA, Waelsch K, Kuziel WA, Delbeck T, Mack M, Blackwell TS, et al. Monocytes are potent facilitators of alveolar neutrophil emigration during lung inflammation: role of the CCL2-CCR2 axis. J Immunol. 2003;170(6):3273–8.

Cui TX, Brady AE, Fulton CT, Zhang YJ, Rosenbloom LM, Goldsmith AM, et al. CCR2 mediates chronic LPS-induced pulmonary inflammation and hypoalveolarization in a murine model of bronchopulmonary dysplasia. Front Immunol. 2020;11:579628.

Nakano H, Lyons-Cohen MR, Whitehead GS, Nakano K, Cook DN. Distinct functions of CXCR4, CCR2, and CX3CR1 direct dendritic cell precursors from the bone marrow to the lung. J Leukoc Biol. 2017;101(5):1143–53.

Rosseau S, Hammerl P, Maus U, Walmrath HD, Schutte H, Grimminger F, et al. Phenotypic characterization of alveolar monocyte recruitment in acute respiratory distress syndrome. Am J Physiol Lung Cell Mol Physiol. 2000;279(1):L25–35.

Gurczynski SJ, Procario MC, O’Dwyer DN, Wilke CA, Moore BB. Loss of CCR2 signaling alters leukocyte recruitment and exacerbates gamma-herpesvirus-induced pneumonitis and fibrosis following bone marrow transplantation. Am J Physiol Lung Cell Mol Physiol. 2016;311(3):L611–27.

Milger K, Yu Y, Brudy E, Irmler M, Skapenko A, Mayinger M, et al. Pulmonary CCR2(+)CD4(+) T cells are immune regulatory and attenuate lung fibrosis development. Thorax. 2017;72(11):1007–20.

Aripova N, Duryee MJ, Hunter CD, Ryan EM, Daubach EC, Jones SQ, et al. Peptidyl arginine deiminase expression and macrophage polarization following stimulation with citrullinated and malondialdehyde-acetaldehyde modified fibrinogen. Int Immunopharmacol. 2022;110:109010.

Belchamber KBR, Hughes MJ, Spittle DA, Walker EM, Sapey E. New Pharmacological Tools to Target Leukocyte Trafficking in Lung Disease. Front Immunol. 2021;12:704173.

Gerhardt T, Ley K. Monocyte trafficking across the vessel wall. Cardiovasc Res. 2015;107(3):321–30.

Shi C, Pamer EG. Monocyte recruitment during infection and inflammation. Nat Rev Immunol. 2011;11(11):762–74.

Download references

Acknowledgements

The authors acknowledge Craig Semerad, Victoria B. Smith, and Holly Britton in the Flow Cytometry Research Core Facility at the University of Nebraska Medical Center for aiding with flow cytometry studies. This core facility is administrated through the Office of the Vice Chancellor for Research and supported by state funds from the Nebraska Research Initiative (NRI) and The Fred and Pamela Buffett Cancer Center's National Cancer Institute Cancer Support Grant. Major instrumentation has been provided by the Office of the Vice Chancellor for Research, The University of Nebraska Foundation, the Nebraska Banker's Fund, and by the NIH-NCRR Shared Instrument Program. The authors would also like to thank Marie Nguyen for assistance in manuscript preparation and submission.

Disclosures

JAP has received research reagent from AstraZeneca (no monies) and has been a site investigator for allergy and asthma clinical studies for Takeda, GlaxoSmithKline, Regeneron, Areteia, and AstraZeneca (no monies). TRM received research support from Horizon Therapeutics and has been a consultant for Horizon, Pfizer, UCB, and Sanofi.

National Institute for Occupational Safety and Health grant U54OH010162 (ADS, TAW) and R01OH012045 (JAP), Department of Defense #PR200793 (JAP and TRM). ADS, TAW, and JAP received support from Central States Center of Agricultural Safety and Health (CS-CASH). TAW is supported by grants from the VA (I01 BX005886) and National Institutes of Health (P50 AA030407). TAW is the recipient of a Research Career Scientist Award (IK6 BX005962) from the Department of Veterans Affairs. TRM is also funded by VA Merit I01 BX004660.

Author information

Authors and affiliations.

Division of Allergy & Immunology, University of Nebraska Medical Center, Omaha, NE, USA

Aaron D. Schwab, Grace Moravec, Amy J. Nelson, Angela Gleason, Oliver Schanze & Jill A. Poole

Veterans Affairs Nebraska-Western Iowa Health Care System, Research Service, Omaha, NE, USA

Todd A. Wyatt, Geoffrey M. Thiele, Michael J. Duryee, Debra J. Romberger & Ted R. Mikuls

Division of Pulmonary, Critical Care & Sleep, University of Nebraska Medical Center, Omaha, NE, USA

Todd A. Wyatt & Debra J. Romberger

Department of Environmental, Agricultural and Occupational Health, College of Public Health, University of Nebraska Medical Center, Omaha, NE, USA

Todd A. Wyatt

Division of Rheumatology and Immunology, Department of Internal Medicine, College of Medicine, University of Nebraska Medical Center, Omaha, NE, USA

Geoffrey M. Thiele, Michael J. Duryee & Ted R. Mikuls

You can also search for this author in PubMed   Google Scholar

Contributions

Conceived and designed research: JAP, TRM, GMT; performed experiments: JAP, TRM, GMT, AJN, MJD, ADS, AG, OS, GM, TAW; analyzed data: JAP, TRM, GMT, AJN, MJD, ADS, AG, OS, GM, TAW; interpreted results of experiments: JAP, TRM, GMT, AJN, MJD, ADS, AG, OS, GM, TAW; prepared figures: JAP, TRM, GMT, AJN, MJD, ADS; drafted manuscript: JAP, ADS; edited and revised manuscript: JAP, TRM, GMT, AJN, MJD, ADS, AG, OS, GM, TAW, DJM; approved final version of manuscript: JAP, TRM, GMT, AJN, MJD, ADS, AG, OS, GM, TAW, DJM.

Corresponding author

Correspondence to Jill A. Poole .

Ethics declarations

Ethics approval and consent to participate.

Neither human participants, data, nor tissues were used in these studies. The study was conducted and reported in accordance with ARRIVE guidelines ( https://arriveguidelines.org ). All animal procedures were approved by the University of Nebraska Medical Center (UNMC) Institutional Animal Care and Use Committee (IACUC) and were in accordance with NIH guidelines for the use of rodents.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: supplemental fig. 1..

Gating strategy for identification of non-debris, singlets, live CD45 + myeloid and lymphoid cells. For flow analysis, all panels were first gated as forward scatter-area (FSC-A) x side scatter-area (SSC-A) to omit debris, dead, or apoptotic cells. This was followed by two single cell gates to omit doublets (FSC-A x FSC-heigh (H) and SSC-A x SSC-H), followed by live/dead gate and then CD45 gate to assure removal of any additional dead or apoptotic cells and non-leukocytes. The CD45 + cells were placed on a CD11c x Ly6G gate to select Ly6G + neutrophils. Non-neutrophils were gated for CD19 + B cells (CD19 x SSC gate). This was followed by non-B cells gated on CD11c x CD11b gate to select CD11c + CD11b lo alveolar (Alv) macrophages (Mɸ), CD11c + CD11b hi activated (act) Mɸ, CD11c int CD11b hi transitioning monocytes (Mono)—Mɸ, and CD11c − CD11b hi monocytes (Mono). The negative or non-monocyte/macrophage populations were placed on CD3 x NK1.1 to select CD3 + T cells and CD3 − NK1.1 + NK cells, and then a CD4 x CD8 gate to select CD3 + CD4 + and CD3 + CD8 + T cells. A CCR2 RFP x SSC gate is shown for neutrophils and lymphocytes to demonstrate CCR2 + staining on specific lung cell subpopulations. Lung sample shown is from an LPS-exposed mouse.

Additional file 2: Supplemental Fig. 2.

F4/80 (ADGRE1) expression on monocyte-macrophage (Mɸ) subpopulations following organic dust extract (ODE), lipopolysaccharide (LPS), and peptidoglycan (PGN) inhalation exposure. C57BL/6 mice were exposed once to ODE (25%), LPS (10 μg), PGN (100 μg), or saline control and euthanized at 48 h. Scatter plots with bars depict mean with SD delineating cells as CCR2 + (green) and CCR2 − (red). Expression of F4/80 by percent (A) and mean fluorescence intensity (MFI) (B) across alveolar (Alv) Mɸ, activated (Act) Mɸ, monocyte-Mɸ, and monocyte subpopulations as determined by flow cytometry. Statistical analyses were performed with Kruskal–Wallis with Dunn’s test for multiple comparisons (# p  < 0.05 vs. respective saline) and (* p  < 0.05 denoted by line with brackets denoting difference between same inhalant exposure by CCR2 RFP positive vs. negative). N  = 15 (saline), 9 (ODE), 4 (LPS), 5 (PGN).

Additional file 3: Supplemental Table 1.

LPS-induced airway inflammatory indices not affected with systemic delivery of clodronate liposomes.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Schwab, A.D., Wyatt, T.A., Moravec, G. et al. Targeting transitioning lung monocytes/macrophages as treatment strategies in lung disease related to environmental exposures. Respir Res 25 , 157 (2024). https://doi.org/10.1186/s12931-024-02804-3

Download citation

Received : 09 February 2024

Accepted : 03 April 2024

Published : 09 April 2024

DOI : https://doi.org/10.1186/s12931-024-02804-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Organic dust
• Inflammation
• Lung disease

Respiratory Research

ISSN: 1465-993X

• General enquiries: [email protected]

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Indian J Crit Care Med
• v.25(Suppl 2); 2021 May

An Introduction to Statistics: Choosing the Correct Statistical Test

Priya ranganathan.

1 Department of Anaesthesiology, Critical Care and Pain, Tata Memorial Centre, Homi Bhabha National Institute, Mumbai, Maharashtra, India

The choice of statistical test used for analysis of data from a research study is crucial in interpreting the results of the study. This article gives an overview of the various factors that determine the selection of a statistical test and lists some statistical testsused in common practice.

How to cite this article: Ranganathan P. An Introduction to Statistics: Choosing the Correct Statistical Test. Indian J Crit Care Med 2021;25(Suppl 2):S184–S186.

In a previous article in this series, we looked at different types of data and ways to summarise them. 1 At the end of the research study, statistical analyses are performed to test the hypothesis and either prove or disprove it. The choice of statistical test needs to be carefully performed since the use of incorrect tests could lead to misleading conclusions. Some key questions help us to decide the type of statistical test to be used for analysis of study data. 2

W hat is the R esearch H ypothesis ?

Sometimes, a study may just describe the characteristics of the sample, e.g., a prevalence study. Here, the statistical analysis involves only descriptive statistics . For example, Sridharan et al. aimed to analyze the clinical profile, species distribution, and susceptibility pattern of patients with invasive candidiasis. 3 They used descriptive statistics to express the characteristics of their study sample, including mean (and standard deviation) for normally distributed data, median (with interquartile range) for skewed data, and percentages for categorical data.

Studies may be conducted to test a hypothesis and derive inferences from the sample results to the population. This is known as inferential statistics . The goal of inferential statistics may be to assess differences between groups (comparison), establish an association between two variables (correlation), predict one variable from another (regression), or look for agreement between measurements (agreement). Studies may also look at time to a particular event, analyzed using survival analysis.

A re the C omparisons M atched (P aired ) or U nmatched (U npaired )?

Observations made on the same individual (before–after or comparing two sides of the body) are usually matched or paired . Comparisons made between individuals are usually unpaired or unmatched . Data are considered paired if the values in one set of data are likely to be influenced by the other set (as can happen in before and after readings from the same individual). Examples of paired data include serial measurements of procalcitonin in critically ill patients or comparison of pain relief during sequential administration of different analgesics in a patient with osteoarthritis.

W hat are the T ype of D ata B eing M easured ?

The test chosen to analyze data will depend on whether the data are categorical (and whether nominal or ordinal) or numerical (and whether skewed or normally distributed). Tests used to analyze normally distributed data are known as parametric tests and have a nonparametric counterpart that is used for data, which is distribution-free. 4 Parametric tests assume that the sample data are normally distributed and have the same characteristics as the population; nonparametric tests make no such assumptions. Parametric tests are more powerful and have a greater ability to pick up differences between groups (where they exist); in contrast, nonparametric tests are less efficient at identifying significant differences. Time-to-event data requires a special type of analysis, known as survival analysis.

H ow M any M easurements are B eing C ompared ?

The choice of the test differs depending on whether two or more than two measurements are being compared. This includes more than two groups (unmatched data) or more than two measurements in a group (matched data).

T ests for C omparison

( Table 1 lists the tests commonly used for comparing unpaired data, depending on the number of groups and type of data. As an example, Megahed and colleagues evaluated the role of early bronchoscopy in mechanically ventilated patients with aspiration pneumonitis. 5 Patients were randomized to receive either early bronchoscopy or conventional treatment. Between groups, comparisons were made using the unpaired t test for normally distributed continuous variables, the Mann–Whitney U -test for non-normal continuous variables, and the chi-square test for categorical variables. Chowhan et al. compared the efficacy of left ventricular outflow tract velocity time integral (LVOTVTI) and carotid artery velocity time integral (CAVTI) as predictors of fluid responsiveness in patients with sepsis and septic shock. 6 Patients were divided into three groups— sepsis, septic shock, and controls. Since there were three groups, comparisons of numerical variables were done using analysis of variance (for normally distributed data) or Kruskal–Wallis test (for skewed data).

Tests for comparison of unpaired data

A common error is to use multiple unpaired t -tests for comparing more than two groups; i.e., for a study with three treatment groups A, B, and C, it would be incorrect to run unpaired t -tests for group A vs B, B vs C, and C vs A. The correct technique of analysis is to run ANOVA and use post hoc tests (if ANOVA yields a significant result) to determine which group is different from the others.

( Table 2 lists the tests commonly used for comparing paired data, depending on the number of groups and type of data. As discussed above, it would be incorrect to use multiple paired t -tests to compare more than two measurements within a group. In the study by Chowhan, each parameter (LVOTVTI and CAVTI) was measured in the supine position and following passive leg raise. These represented paired readings from the same individual and comparison of prereading and postreading was performed using the paired t -test. 6 Verma et al. evaluated the role of physiotherapy on oxygen requirements and physiological parameters in patients with COVID-19. 7 Each patient had pretreatment and post-treatment data for heart rate and oxygen supplementation recorded on day 1 and day 14. Since data did not follow a normal distribution, they used Wilcoxon's matched pair test to compare the prevalues and postvalues of heart rate (numerical variable). McNemar's test was used to compare the presupplemental and postsupplemental oxygen status expressed as dichotomous data in terms of yes/no. In the study by Megahed, patients had various parameters such as sepsis-related organ failure assessment score, lung injury score, and clinical pulmonary infection score (CPIS) measured at baseline, on day 3 and day 7. 5 Within groups, comparisons were made using repeated measures ANOVA for normally distributed data and Friedman's test for skewed data.

Tests for comparison of paired data

T ests for A ssociation between V ariables

( Table 3 lists the tests used to determine the association between variables. Correlation determines the strength of the relationship between two variables; regression allows the prediction of one variable from another. Tyagi examined the correlation between ETCO 2 and PaCO 2 in patients with chronic obstructive pulmonary disease with acute exacerbation, who were mechanically ventilated. 8 Since these were normally distributed variables, the linear correlation between ETCO 2 and PaCO 2 was determined by Pearson's correlation coefficient. Parajuli et al. compared the acute physiology and chronic health evaluation II (APACHE II) and acute physiology and chronic health evaluation IV (APACHE IV) scores to predict intensive care unit mortality, both of which were ordinal data. Correlation between APACHE II and APACHE IV score was tested using Spearman's coefficient. 9 A study by Roshan et al. identified risk factors for the development of aspiration pneumonia following rapid sequence intubation. 10 Since the outcome was categorical binary data (aspiration pneumonia— yes/no), they performed a bivariate analysis to derive unadjusted odds ratios, followed by a multivariable logistic regression analysis to calculate adjusted odds ratios for risk factors associated with aspiration pneumonia.

Tests for assessing the association between variables

T ests for A greement between M easurements

( Table 4 outlines the tests used for assessing agreement between measurements. Gunalan evaluated concordance between the National Healthcare Safety Network surveillance criteria and CPIS for the diagnosis of ventilator-associated pneumonia. 11 Since both the scores are examples of ordinal data, Kappa statistics were calculated to assess the concordance between the two methods. In the previously quoted study by Tyagi, the agreement between ETCO 2 and PaCO 2 (both numerical variables) was represented using the Bland–Altman method. 8

Tests for assessing agreement between measurements

T ests for T ime-to -E vent D ata (S urvival A nalysis )

Time-to-event data represent a unique type of data where some participants have not experienced the outcome of interest at the time of analysis. Such participants are considered to be “censored” but are allowed to contribute to the analysis for the period of their follow-up. A detailed discussion on the analysis of time-to-event data is beyond the scope of this article. For analyzing time-to-event data, we use survival analysis (with the Kaplan–Meier method) and compare groups using the log-rank test. The risk of experiencing the event is expressed as a hazard ratio. Cox proportional hazards regression model is used to identify risk factors that are significantly associated with the event.

Hasanzadeh evaluated the impact of zinc supplementation on the development of ventilator-associated pneumonia (VAP) in adult mechanically ventilated trauma patients. 12 Survival analysis (Kaplan–Meier technique) was used to calculate the median time to development of VAP after ICU admission. The Cox proportional hazards regression model was used to calculate hazard ratios to identify factors significantly associated with the development of VAP.

The choice of statistical test used to analyze research data depends on the study hypothesis, the type of data, the number of measurements, and whether the data are paired or unpaired. Reviews of articles published in medical specialties such as family medicine, cytopathology, and pain have found several errors related to the use of descriptive and inferential statistics. 12 – 15 The statistical technique needs to be carefully chosen and specified in the protocol prior to commencement of the study, to ensure that the conclusions of the study are valid. This article has outlined the principles for selecting a statistical test, along with a list of tests used commonly. Researchers should seek help from statisticians while writing the research study protocol, to formulate the plan for statistical analysis.

Priya Ranganathan https://orcid.org/0000-0003-1004-5264

Source of support: Nil

Conflict of interest: None

R eferences