Evaluating non-randomised intervention studies
Open accessThe DOI matches its Crossref record; 334 of the 334 references it cites were found in a registry or an index.
Abstract
OBJECTIVES: To consider methods and related evidence for evaluating bias in non-randomised intervention studies. DATA SOURCES: Systematic reviews and methodological papers were identified from a search of electronic databases; handsearches of key medical journals and contact with experts working in the field. New empirical studies were conducted using data from two large randomised clinical trials. METHODS: Three systematic reviews and new empirical investigations were conducted. The reviews considered, in regard to non-randomised studies, (1) the existing evidence of bias, (2) the content of quality assessment tools, (3) the ways that study quality has been assessed and addressed. (4) The empirical investigations were conducted generating non-randomised studies from two large, multicentre randomised controlled trials (RCTs) and selectively resampling trial participants according to allocated treatment, centre and period. RESULTS: In the systematic reviews, eight studies compared results of randomised and non-randomised studies across multiple interventions using meta-epidemiological techniques. A total of 194 tools were identified that could be or had been used to assess non-randomised studies. Sixty tools covered at least five of six pre-specified internal validity domains. Fourteen tools covered three of four core items of particular importance for non-randomised studies. Six tools were thought suitable for use in systematic reviews. Of 511 systematic reviews that included non-randomised studies, only 169 (33%) assessed study quality. Sixty-nine reviews investigated the impact of quality on study results in a quantitative manner. The new empirical studies estimated the bias associated with non-random allocation and found that the bias could lead to consistent over- or underestimations of treatment effects, also the bias increased variation in results for both historical and concurrent controls, owing to haphazard differences in case-mix between groups. The biases were large enough to lead studies falsely to conclude significant findings of benefit or harm. Four strategies for case-mix adjustment were evaluated: none adequately adjusted for bias in historically and concurrently controlled studies. Logistic regression on average increased bias. Propensity score methods performed better, but were not satisfactory in most situations. Detailed investigation revealed that adequate adjustment can only be achieved in the unrealistic situation when selection depends on a single factor. CONCLUSIONS: Results of non-randomised studies sometimes, but not always, differ from results of randomised studies of the same intervention. Non-randomised studies may still give seriously misleading results when treated and control groups appear similar in key prognostic factors. Standard methods of case-mix adjustment do not guarantee removal of bias. Residual confounding may be high even when good prognostic data are available, and in some situations adjusted results may appear more biased than unadjusted results. Although many quality assessment tools exist and have been used for appraising non-randomised studies, most omit key quality domains. Healthcare policies based upon non-randomised studies or systematic reviews of non-randomised studies may need re-evaluation if the uncertainty in the true evidence base was not fully appreciated when policies were made. The inability of case-mix adjustment methods to compensate for selection bias and our inability to identify non-randomised studies that are free of selection bias indicate that non-randomised studies should only be undertaken when RCTs are infeasible or unethical. Recommendations for further research include: applying the resampling methodology in other clinical areas to ascertain whether the biases described are typical; developing or refining existing quality assessment tools for non-randomised studies; investigating how quality assessments of non-randomised studies can be incorporated into reviews and the implications of individual quality features for interpretation of a review's results; examination of the reasons for the apparent failure of case-mix adjustment methods; and further evaluation of the role of the propensity score.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source doesn’t support the claim as written.
In the citing paper
“This tool was selected for its comprehensive ability to assess the methodological quality of non‐randomized studies and has shown good reliability and validity.21, 22”
From The frequency of influenza and bacterial coinfection: a systematic review and meta-analysis. (2016) · cites this paper as [22]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe cited source doesn’t support the claim as written.
In the citing paper
“In general, the probability of systematic differences in comparison groups at baseline was low, diminishing the risk of selection bias (58).”
From Mindfulness-Based Program for Infertility: Efficacy study (2013) · cites this paper as [58]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“The EPHPP has also been reported to have better interrater reliability than the Cochrane Collaboration Risk of Bias Tool”
From Improving Mental Health Service Utilization Among Men: A Systematic Review and Synthesis of Behavior Change Techniques Within Interventions Targeting Help-Seeking (2019) · cites this paper as Deeks et al., 2003
Experimental — accuracy not yet measuredIn the citing paper
“In nonrandomized studies, selection bias occurs when participants chosen for the intervention of interest have different characteristics from those allocated to the alternative intervention (or not treated) because the choice of a given intervention might be affected by the discretion of the treating clinician or patient preference, patient characteristics, and clinical history [13].”
From RoBANS 2: A Revised Risk of Bias Assessment Tool for Nonrandomized Studies of Interventions (2023) · cites this paper as [13]
In the citing paper
“The EPHPP is considered suitable to assess the quality of a wide range of quantitative study designs, including interventions (Jackson and Waters, 2005;Deeks et al., 2003).”
From Exploring the role of exposure to green and blue spaces in preventing anxiety and depression among young people aged 14–24 years living in urban settings: A systematic review and conceptual framework (2022) · cites this paper as Deeks et al., 2003
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Part of this paper’s reference list could not be split into single entries.
An Introduction to the Bootstrap
Efron, B., & Tibshirani, R. J. (1994) · doi:10.1201/9780429246593 · Cited by 40,500
(1990) · Choice Reviews Online · doi:10.5860/choice.27-3338 · Cited by 35,305
The central role of the propensity score in observational studies for causal effects
ROSENBAUM, P. R., & RUBIN, D. B. (1983) · Biometrika · doi:10.1093/biomet/70.1.41 · Cited by 31,241
Meta-analysis of Observational Studies in Epidemiology<SUBTITLE>A Proposal for Reporting</SUBTITLE>
Stroup, D. F. (2000) · JAMA · doi:10.1001/jama.283.15.2008 · Cited by 20,623
Assessing the quality of reports of randomized clinical trials: Is blinding necessary?
Jadad, A. R., Moore, R. A., Carroll, D., Jenkinson, C., Reynolds, D. J. M., Gavaghan, D. J., & McQuay, H. J. (1996) · Controlled Clinical Trials · doi:10.1016/0197-2456(95)00134-4 · Cited by 17,582
(1963) · Medical Entomology and Zoology · Cited by 15,085
Statistical Aspects of the Analysis of Data From Retrospective Studies of Disease
(1959) · JNCI: Journal of the National Cancer Institute · doi:10.1093/jnci/22.4.719 · Cited by 14,689
Practical Statistics for Medical Research.
Walter, S. D., & Altman, D. G. (1992) · Biometrics · doi:10.2307/2532320 · Cited by 11,668
Self-citation(1979) · Medical Entomology and Zoology · Cited by 11,252
Health Measurement Scales: A Practical Guide to Their Development and Use
Streiner, D. L., Norman, G. R., & Cairney, J. (1989) · Cited by 9,595
Research works printed but not linked: 19
- Abel U, Koch A. The role of randomization in clinical studies: myths and beliefs. J Clin Epidemiol 1999;52:487-97.
- Feinstein AR, Horwitz RI. Problems in the "evidence" of "evidence-based medicine". Am J Med 1997;103:529-35.
- Horwitz RI, Feinstein AR. The application of therapeutic trial principles to improve the design of epidemiologic research. J Chron Dis 1981;34:575-83.
- Byar DP. Problems with using observational data bases to compare treatments. Stat Med 1991;10:663-6.
- Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995;273:408-12.
- Kunz R, Vist G, Oxman AD. Randomisation to protect against selection bias in healthcare trials (Cochrane Methodology Review). In The Cochrane Library, Issue 4. Oxford: Update Software; 2002.
- Diehl LF, Perry DJ. A comparison of randomized concurrent control groups with matched historical control groups: are historical controls valid? J Clin Oncol 1986;4:1114-20.
- Lipsey MW, Wilson DB. The efficacy of psychological educational, and behavioural treatments: confirmation from meta-analysis. Am Psychol 1993;48:1181-209.
- Miller JN, Colditz GA, Mosteller F. How study design affects outcomes in comparisons of therapy. II: surgical. Stat Med 1989;8:455-66.
- Pocock SJ, Elbourne DR. Randomized trials or observational tribulations? N Engl J Med 2000;342:1907-9.
- Moher D, Jadad AR, Tugwell P. Assessing the quality of randomized controlled trials. Current issues and future directions. Int J Technol Assess Health Care 1996;12:195-208.
- Critical Appraisal Skills Programme. 12 questions to help you make sense of a cohort study. Oxford: Critical Appraisal Skills Programme; 1999.
- Kleijnen J, Knipschild P, Ter Riet G. Clinical trials of homeopathy. BMJ 1991;302:316-32.
- van der Windt DA, van der Heijden GJ, Scholten RJ, Koes BW, Bouter LM. The efficacy of non-steroidal anti-inflammatory drugs for shoulder complaints. J Clin Epidemiol 1995;48:691-704.
- Kleijnen J, Riet ter G, Knipschild P. Vitamine C en verkoudheid. Overzicht van een megadosis literatuur. Ned Tijdschr Geneeskd 1989;133:1535.
- Riet ter G, Kleijnen J, Knipschild P. De meta- analyse als review-methode [De effectiviteit van acupunctuur]. Huisarts en Wetenschap 1989;32: 176-81.
- Nicolucci A, Grilli R, Alexanian A, Apolone G, Torri V, Liberati A. Quality, evolution, and clinical implications of randomized, controlled trials on the treatment of lung cancer. JAMA 1989; 262:2101-7.
- Baker DW, Jones R, Hodges J, Massie BM, Konstam MA, Rose EA. Management of heart failure. III. The role of revascularization in the treatment of patients with moderate or severe left ventricular systolic dysfunction. JAMA 1994;272:1528-34.
- Coleridge Smith P. The management of chronic venous disorders of the leg: an evidence-based report of an International Task Force. Phlebology 1999;14:3-19.
Cited by
Items 1,001–1,024 of 1,024
Jarde, A., Losilla, J.-M., Vives, J., & F. Rodrigo, M. (2013) · International Journal of Clinical and Health Psychology · doi:10.1016/s1697-2600(13)70017-6 · Cited by 50
“In fact, all systematic reviews collecting this type of tools (Deeks et al., 2003; Jarde, Losilla, & Vives, 2012a; Sanderson, Tatt, & Higgins, 2007; Shamliyan, Kane, & Dickinson, 2010; West et al., 2002) agree in criticizing that most of them have not been developed using standard psychometric techniques.”
BackgroundBarbonetti, A., D’Andrea, S., Cavallo, F., Martorella, A., Francavilla, S., & Francavilla, F. (2019) · The Journal of Sexual Medicine · doi:10.1016/j.jsxm.2019.02.014 · Cited by 49
López-Liria, R., Valverde-Martínez, M. Á., López-Villegas, A., Bautista-Mesa, R. J., Vega-Ramírez, F. A., Peiró, S., & Leal-Costa, C. (2022) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph19052534 · Cited by 47
Lloyd, M., Skelton, D. A., Mead, G. E., Williams, B., & van Wijck, F. (2018) · Brain and Behavior · doi:10.1002/brb3.1000 · Cited by 40
Shah, A., Jones, M. P., & Holtmann, G. J. (2020) · Indian Journal of Gastroenterology · doi:10.1007/s12664-020-01107-x · Cited by 39
“& Cochrane risk of bias tool [16] - frequently used for evaluating the quality of randomized controlled trials, and & Newcastle-Ottawa scale [17] - commonly used for evaluating the quality of non-randomized studies (consisting of cohort and case-control studies).”
MethodZhou, M., Guo, B., Wang, Y., Yan, D., Lin, C., & Shi, Z. (2017) · Angiology · doi:10.1177/0003319716665985 · Cited by 39
Wang, S.-B., Wang, Y.-Y., Ungvari, G. S., Ng, C. H., Wu, R.-R., Wang, J., & Xiang, Y.-T. (2017) · Schizophrenia Research · doi:10.1016/j.schres.2016.11.020 · Cited by 38
“org/ either the exposure or outcome of interest: 3 scores (Deeks et al., 2003).”
Robertson, C., Ramsay, C., Gurung, T., Mowatt, G., Pickard, R., Sharma, P., & The UK Robotic Laparoscopic Prostatectomy HTA Study Group (2014) · Research Synthesis Methods · doi:10.1002/jrsm.1102 · Cited by 38
Turicchi, J., O'Driscoll, R., Finlayson, G., Beaulieu, K., Deighton, K., & Stubbs, R. J. (2019) · Obesity Reviews · doi:10.1111/obr.12849 · Cited by 37
“The Downs and Black instrument is an established tool for determination of the quality of a study within a systematic review and meta‐analysis.(44) Two questions related to randomisation were removed as randomisation to groups was not relevant to our outcomes and two questions specific to case‐control and cohort studi…”
MethodOdgers, K., Kershaw, K. A., Li, S. H., & Graham, B. M. (2022) · Behaviour Research and Therapy · doi:10.1016/j.brat.2022.104203 · Cited by 35
Paranoia and negative schema about the self and others: A systematic review and meta-analysis
Humphrey, C., Bucci, S., Varese, F., Degnan, A., & Berry, K. (2021) · Clinical Psychology Review · doi:10.1016/j.cpr.2021.102081 · Cited by 34
Self-Help Interventions for Psychosis: A Meta-Analysis
Scott, A. J., Webb, T. L., & Rowse, G. (2015) · Clinical Psychology Review · doi:10.1016/j.cpr.2015.05.002 · Cited by 33
Jackson, K. F., Hodge, D. R., & Vaughn, M. G. (2010) · Journal of Social Service Research · doi:10.1080/01488371003697780 · Cited by 30
“A substantial number of scales have been used to assess the mythological rigor of studies (Deeks et al., 2003; Olivo et al., 2008; Vaughn & Howard, 2004).”
BackgroundLahiri, S., van Ommeren, M., & Roberts, B. (2017) · Global Public Health · doi:10.1080/17441692.2016.1154585 · Cited by 25
Schiavo, L., Santella, B., Mingo, M., Rossetti, G., Orio, M., Cobellis, L., Maurano, A., Iannelli, A., & Pilone, V. (2025) · Nutrients · doi:10.3390/nu17071216 · Cited by 22
Goal setting with caregivers of adults in the community: a mixed methods systematic review
Tang Yan, H. S., Clemson, L. M., Jarvis, F., & Laver, K. (2014) · Disability and Rehabilitation · doi:10.3109/09638288.2014.884173 · Cited by 21
“the mechanism of allocating subjects) are specific to experimental studies and there are no items related to baseline comparability [31,32].”
BackgroundWu, W., Liu, J., & Zhou, Z. (2020) · Emergency Medicine International · doi:10.1155/2020/1382475 · Cited by 20
Brown, H. E., Atkin, A. J., Panter, J., Corder, K., Wong, G., Chinapaw, M. J. M., & van Sluijs, E. (2014) · BMJ Open · doi:10.1136/bmjopen-2014-005439 · Cited by 20
“The tool has been recommended for use in assessing public health interventions based on acceptable content and construct validity.(25) Quality assessment results will be reported using table 1.”
MethodSystematic reviews and meta‐analyses in coloproctology: interpretation and potential pitfalls
Wille‐Jørgensen, P., & Renehan, A. G. (2008) · Colorectal Disease · doi:10.1111/j.1463-1318.2007.01421.x · Cited by 19
Kok, G. D., Bockting, C. L. H., Burger, H., Hannig, W., Pijnenborg, G. H. M., Cuijpers, P., & Hollon, S. D. (2013) · PLoS ONE · doi:10.1371/journal.pone.0057510 · Cited by 18
“[22] and described as one of the most usable methods for this type of study [22], [23].”
MethodMethodological quality is underrated in systematic reviews and meta-analyses in health psychology
Oliveras, I., Losilla, J.-M., & Vives, J. (2017) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2017.05.002 · Cited by 15
Capogrosso, P., Dimitropolous, K., Russo, G. I., Tharakan, T., Milenkovic, U., Cocci, A., Boeri, L., Gül, M., Bettocchi, C., Carvalho, J., et al. (2024) · International Journal of Impotence Research · doi:10.1038/s41443-022-00592-2 · Cited by 13
(2006) · Systematic Reviews in the Social Sciences · doi:10.1002/9780470754887.ch6 · Cited by 12
Self-citationOctavia, A., Sitthisettapong, T., Kettaratad-Pruksapong, M., & Dewanto, I. (2025) · International Dental Journal · doi:10.1016/j.identj.2024.12.015 · Cited by 6
“41 The recommendation for future studies is to include a control group that accurately reflects the characteristics of the population whenever feasible and to employ randomization for allocating participants within similar groups to mitigate any bias in the research outcomes.”
Metadata
- Authors
- Deeks, J., Dinnes, J., D'Amico, R., Sowden, A., Sakarovitch, C., Song, F., Petticrew, M., & Altman, D.
- Affiliations
- University of Oxford (GB)
- University of Southampton (GB)
- University of York (GB)
- University of Birmingham (GB)
- MRC/CSO Social and Public Health Sciences Unit (GB)
- Venue
- Health Technology Assessment
- Published
- 2003-09
- Volume
- 7
- Issue
- 27
- Pages
- iii-x, 1
- ISSN
- 1366-5278 (ISSN-L)
- 2046-4924
- OpenAlex
- W1981501248
- Semantic Scholar
- 3b3a2281d097cc0c60e93d42d570bd2dfadb6492
- Cited by
- 2,907
- References
- 332
- Citations by year
- 2026: 37 · 2025: 76 · 2024: 107 · 2023: 115 · 2022: 142 · 2021: 188 · 2020: 176 · 2019: 164 · 2018: 148 · 2017: 191 · 2016: 169 · 2015: 185 · 2014: 177 · 2013: 141 · 2012: 123
- FWCI
- 81.72
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Health Technology Assessment (publishedVersion, cc-by)
- PubMed (publishedVersion)
- LSHTM Research Online (London School of Hygiene and Tropical Medicine) (acceptedVersion)
- IRIS UNIMORE (University of Modena and Reggio Emilia) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- ENLIGHTEN (Jurnal Bimbingan dan Konseling Islam) (submittedVersion)
- White Rose Research Online (University of Leeds, The University of Sheffield, University of York) (submittedVersion)
- University of Birmingham Research Portal (University of Birmingham) (submittedVersion)
- UEA Digital Repository (University of East Anglia) (submittedVersion)
- Funders
- Health Technology Assessment programme (96/26/99)
- Topics
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Keywords
- bias in observational studies, internal validity, quality assessment tools, risk adjustment, residual confounding, propensity score methods, meta-epidemiology, treatment effect estimation, prognostic factors, selection bias, confounding, reporting bias, reporting quality, randomized controlled trials
- MeSH terms
- Clinical Trials as Topic, Diagnosis-Related Groups, Humans, Quality Control, Research Design, Reproducibility of Results, Bias, Selection Bias, Patient Selection, Empirical Research
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.