Volume 11, Issue 1 (3-2022)
JCHR 2022, 11(1): 3-4 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
sefidkar R, Madadizadeh F. A tutorial on Quasi-experimental designs. JCHR 2022; 11 (1) :3-4
URL: http://jhr.ssu.ac.ir/article-1-780-en.html
URL: http://jhr.ssu.ac.ir/article-1-780-en.html
1- Center for Healthcare Data Modeling, Departments of biostatistics and Epidemiology, School of public health, Shahid Sadoughi University of Medical Sciences, Yazd, Iran
2- Center for Healthcare Data Modeling, Departments of biostatistics and Epidemiology, School of public health, Shahid Sadoughi University of Medical Sciences, Yazd, Iran , madadizadehfarzan@gmail.com
2- Center for Healthcare Data Modeling, Departments of biostatistics and Epidemiology, School of public health, Shahid Sadoughi University of Medical Sciences, Yazd, Iran , madadizadehfarzan@gmail.com
Keywords: Quasi-experimental design, Medicine
Full-Text [PDF 350 kb]
(859 Downloads)
| Abstract (HTML) (1650 Views)
Full-Text: (267 Views)
Dear Editor,
A main step in answering a scientific hypothesis in an epidemiological study is deciding which
type of study is suitable to be undertaken, considering methodology, practical considerations and budget and time limitations. Basically, epidemiological studies can be divided into two broad types: observational and interventional studies (experimental studies). In contrast to observational studies which no attempt is made to affect the outcome, to determine the effect of an exposure to the intervention on the natural course of events, a deliberate intervention is made on some or all samples in interventional studies. Depending on whether the units involved in a study are individuals or communities, there are two main types of experimental studies: (i) controlled clinical trials and (ii) community trials (1).
In practice, there are particular circumstances in which the researcher goal is to demonstrate a causal relationship between an independent and dependent variable while the random allocation
of the intervention may not be feasible for
ethical or practical constraints (2). To response this
need, Quasi-experimental designs (QEDs) are increasingly employed.
There are various types of QEDs with different strengths, weaknesses and applications including: interrupted time series designs, designs with control groups, and designs without control
groups (3). Quasi-experiments are included in observational studies class which combine some of the advantages of controlled trials with those of non-experimental.
Comparison of QED and randomized controlled trials
QEDs generate evidence faster with less cost and resources compared with true experimental designs. Similar to randomized controlled trials (RCTs), QEDs is a tool to infer the casual treatment effects, but, contrary to RCTs, patients or clusters of patients self-select into one of several different treatment groups in QEDs (4,5). Although lack of random assignment decreases internal validity and also increases potential for bias, or confounding, QEDs are the best and most valid designs available when random assignment is difficult, unethical or impossible. It can be also applied to validate treatment methods or establish new associations for true experimental designs (6). QEDs have higher external validity compared to RCTs and are pragmatic because they evaluate the real-world efficacy of an intervention in non-laboratory conditions (3). Another difference between RCTs and QEDs is that in QEDs, true control group is not needed and instead, experiment group can be compared with a
group which receives a different experimental treatment (7).
In some clinical research, due to some ethical and practical reasons, the causal effects of an intervention cannot be explored through RCTs. So, in order to answer the research question in such circumstances, investigators should choose the strongest design which is feasible as well. QEDs are suggested to address this shortage.
Author's contribution
R.S and F.M. conceived of the presented idea. R.S. wrote the manuscript with support from F.M., R.S., and F.M. read the manuscript and verified
it.
A main step in answering a scientific hypothesis in an epidemiological study is deciding which
type of study is suitable to be undertaken, considering methodology, practical considerations and budget and time limitations. Basically, epidemiological studies can be divided into two broad types: observational and interventional studies (experimental studies). In contrast to observational studies which no attempt is made to affect the outcome, to determine the effect of an exposure to the intervention on the natural course of events, a deliberate intervention is made on some or all samples in interventional studies. Depending on whether the units involved in a study are individuals or communities, there are two main types of experimental studies: (i) controlled clinical trials and (ii) community trials (1).
In practice, there are particular circumstances in which the researcher goal is to demonstrate a causal relationship between an independent and dependent variable while the random allocation
of the intervention may not be feasible for
ethical or practical constraints (2). To response this
need, Quasi-experimental designs (QEDs) are increasingly employed.
There are various types of QEDs with different strengths, weaknesses and applications including: interrupted time series designs, designs with control groups, and designs without control
groups (3). Quasi-experiments are included in observational studies class which combine some of the advantages of controlled trials with those of non-experimental.
Comparison of QED and randomized controlled trials
QEDs generate evidence faster with less cost and resources compared with true experimental designs. Similar to randomized controlled trials (RCTs), QEDs is a tool to infer the casual treatment effects, but, contrary to RCTs, patients or clusters of patients self-select into one of several different treatment groups in QEDs (4,5). Although lack of random assignment decreases internal validity and also increases potential for bias, or confounding, QEDs are the best and most valid designs available when random assignment is difficult, unethical or impossible. It can be also applied to validate treatment methods or establish new associations for true experimental designs (6). QEDs have higher external validity compared to RCTs and are pragmatic because they evaluate the real-world efficacy of an intervention in non-laboratory conditions (3). Another difference between RCTs and QEDs is that in QEDs, true control group is not needed and instead, experiment group can be compared with a
group which receives a different experimental treatment (7).
In some clinical research, due to some ethical and practical reasons, the causal effects of an intervention cannot be explored through RCTs. So, in order to answer the research question in such circumstances, investigators should choose the strongest design which is feasible as well. QEDs are suggested to address this shortage.
Author's contribution
R.S and F.M. conceived of the presented idea. R.S. wrote the manuscript with support from F.M., R.S., and F.M. read the manuscript and verified
it.
Review: Letter to The Editor |
Subject:
Epidemiology
Received: 2021/06/16 | Accepted: 2022/03/28 | Published: 2022/03/30
Received: 2021/06/16 | Accepted: 2022/03/28 | Published: 2022/03/30
References
1. Aggarwal R, Ranganathan P. Study designs: Part 4 - Interventional studies. Perspect Clin Res. 2019;10(3). [DOI:10.4103/picr.PICR_91_19]
2. Lauren T. An introduction to quasi-experimental designs [Internet]. Available from: https://www.scribbr.com/ methodology/quasi-experimental-design/
3. Schweizer ML, Braun BI, Milstone AM. Research Methods in Healthcare Epidemiology and Antimicrobial Stewardship - Quasi-Experimental Designs. Infect Control Hosp Epidemiol. 2016;37(10). [DOI:10.1017/ice.2016.117]
4. Bärnighausen T, Tugwell P, Røttingen J-A, Shemilt I, Rockers P, Geldsetzer P, et al. Quasi-experimental study designs series- Paper 4: uses and value. J Clin Epidemiol. 2017; [DOI:10.1016/j.jclinepi.2017.03.012]
5. Maciejewski ML. Quasi-experimental design. Biostat Epidemiol. 2020;4(1):38-47. [DOI:10.1080/24709360.2018.1477468]
6. Thompson CB, Panacek EA. Research Study Designs: Experimental and Quasi-Experimental. Air Med J. 2006;25(6):242-6. [DOI:10.1016/j.amj.2006.09.001]
7. Rogers J, Révész A. Experimental and quasi-experimental designs. The Routledge handbook of research methods in applied linguistics. New York: Routledge; 2020. [DOI:10.4324/9780367824471-12]
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution 4.0 International License. |
