Definition and research guide
This page combines a MeSH definition, publication statistics and selected research abstracts. It is a starting point for reading the literature, not a medical textbook or a systematic evidence review.
Content available: 3 complete abstracts (668 words), 9 references. Full article text is not stored or reproduced here; follow the PMID, DOI or PMC links to read the original sources and their access terms.
Definition
Any deviation of results or inferences from the truth, or processes leading to such deviation. Bias can result from several sources: one-sided or systematic variations in measurement from the true value (systematic error); flaws in study design; deviation of inferences, interpretations, or analyses based on flawed data or data collection; etc. There is no sense of prejudice or subjectivity implied in the assessment of bias under these conditions.
Definition: NLM Medical Subject Headings (MeSH) scope note (Public domain (U.S. National Library of Medicine); attribution requested). View the MeSH record.
Also known as: Bias, Epidemiologic; Epidemiologic Biases; Biase, Epidemiologic; Biases, Epidemiologic; Epidemiologic Biase; Biases; Ecological Bias; Ecological Biases.
Where it fits in the MeSH classification
- Broader topics
- Epidemiologic Factors
- Narrower topics
- Observer Variation; Selection Bias
Research landscape
PubMed indexes 81,407 records tagged with the MeSH heading “Bias” (counts retrieved 2026-10-31). 13,459 of them (16.5%) were published from 2020 onward, and the busiest period so far is 2010-2019 with 29,611 records.
Reviews account for 7,511 records (9.2%) and meta-analyses for 2,146. Randomized controlled trials number 2,472 (3.0%) as a share of all records on the topic.
Show the data
| Period | Records |
|---|---|
| through 1999 | 14,298 |
| 2000-2009 | 24,039 |
| 2010-2019 | 29,611 |
| 2020 onward | 13,459 |
Publication-type chart data
| Type | Records | Share |
|---|---|---|
| Reviews | 7,511 | 9.2% |
| Meta-analyses | 2,146 | 2.6% |
| Randomized controlled trials | 2,472 | 3.0% |
Method: PubMed query "Bias"[MeSH Terms] with publication-type and date filters, retrieved 2026-10-31.
Reading the evidence in public health and nutrition
Population and nutrition studies mostly observe what people eat, breathe or are exposed to and then look for links with health, which shows association and not proof of cause. Self-reported diet is imprecise, people who eat or live differently also differ in other ways, and effects measured across a population may not apply to one person. Check the study design, how exposure was measured, what else was accounted for and the size of the effect, not only whether it was statistically significant.
Evidence from open-access studies
These open-access studies come from PubMed’s best-match results for this heading (papers whose main topic it is), ranked here by citation count, with one recent addition. Each abstract is reproduced in full under its recorded open licence so that you can read the authors’ own words. They illustrate the literature; they are not a systematic summary of it. Citation counts measure attention, not reliability or clinical benefit.
| Study / authors | Year | Source label / design | Europe PMC citations | Licence / source |
|---|---|---|---|---|
| Risk of Bias Evaluation of Cross-Sectional Studies: Adaptation of the Newcastle-Ottawa Scale | 2026 | Research article Design not recorded in structured metadata | 167 | cc by PMID 40293188 · DOI · PMC13665615 |
| Demographic history mediates the effect of stratification on polygenic scores | 2020 | Research article Design not recorded in structured metadata | 123 | cc by PMID 33200985 · DOI · PMC7758063 |
| Confounding and regression adjustment in difference-in-differences studies | 2021 | Research article Design not recorded in structured metadata | 91 | cc by PMID 33978956 · DOI · PMC8522571 |
Risk of Bias Evaluation of Cross-Sectional Studies: Adaptation of the Newcastle-Ottawa Scale [1]
Abstract reading summary — source excerpts
- Opening excerpt
- Cross-sectional studies are widely utilized in medical research to estimate prevalence and examine associations.
- Closing excerpt
- To streamline their application, spreadsheets for both NOS-xs and NOS-x2 are provided.
- Interpretation limits
- Verbatim excerpts from the authors’ abstract; the closing excerpt may describe a finding, limitation or recommendation. Design, population, effect size and uncertainty have not been independently extracted or appraised. Read the complete abstract and linked paper before interpreting the findings.
Read complete abstract (236 words)
Cross-sectional studies are widely utilized in medical research to estimate prevalence and examine associations. As such, they can serve as a significant source of data for systematic reviews. However, specific considerations are necessary when evaluating the risk of bias (RoB) of cross-sectional studies, as several potential biases can undermine the validity, reliability, and robustness of their findings. This article introduces a novel, context-specific tool designed to assess the RoB of cross-sectional studies for use in systematic reviews. The proposed tool represents an adaptation of the Newcastle-Ottawa Scale (NOS), originally developed for cohort and case-control studies. Similar to the original NOS, the new tool (named "NOS-xs") features a nine-star rating system to assess six specific items across three main domains: (i) study sample selection, (ii) assessment of exposure(s) and outcome(s), and (iii) confounding factors. Based on the number of awarded stars, studies are categorized as having high (0-3 stars), moderate (4-6 stars), or low (7-9 stars) RoB. The NOS-xs tool maintains consistency with the original NOS tool, facilitating its integration into systematic reviews that also include cohort and/or case-control studies. While the NOS-xs is suited to analytical cross-sectional studies (i.e., association studies), a simplified version ("NOS-xs2") is also introduced for descriptive cross-sectional studies (i.e., prevalence studies). The NOS-xs2 features a four-star rating system to assess three of the six specific items included in the NOS-xs. To streamline their application, spreadsheets for both NOS-xs and NOS-x2 are provided.
Abstract reproduced from the open-access article under the recorded cc by licence; © the authors. PMID 40293188 · DOI · PMC13665615
Demographic history mediates the effect of stratification on polygenic scores [2]
Abstract reading summary — source excerpts
- Opening excerpt
- Population stratification continues to bias the results of genome-wide association studies (GWAS).
- Closing excerpt
- We show that the effect of population stratification depends not only on allele frequencies and environmental structure but also on demographic history.
- Interpretation limits
- Verbatim excerpts from the authors’ abstract; the closing excerpt may describe a finding, limitation or recommendation. Design, population, effect size and uncertainty have not been independently extracted or appraised. Read the complete abstract and linked paper before interpreting the findings.
Read complete abstract (150 words)
Population stratification continues to bias the results of genome-wide association studies (GWAS). When these results are used to construct polygenic scores, even subtle biases can cumulatively lead to large errors. To study the effect of residual stratification, we simulated GWAS under realistic models of demographic history. We show that when population structure is recent, it cannot be corrected using principal components of common variants because they are uninformative about recent history. Consequently, polygenic scores are biased in that they recapitulate environmental structure. Principal components calculated from rare variants or identity-by-descent segments can correct this stratification for some types of environmental effects. While family-based studies are immune to stratification, the hybrid approach of ascertaining variants in GWAS but reestimating effect sizes in siblings reduces but does not eliminate stratification. We show that the effect of population stratification depends not only on allele frequencies and environmental structure but also on demographic history.
Abstract reproduced from the open-access article under the recorded cc by licence; © the authors. PMID 33200985 · DOI · PMC7758063
Confounding and regression adjustment in difference-in-differences studies [3]
Abstract reading summary — source excerpts
- Opening excerpt
- ObjectiveTo define confounding bias in difference-in-difference studies and compare regression- and matching-based estimators designed to correct bias due to observed confounders.
- Closing excerpt
- We emphasize the importance of thoughtful incorporation of covariates to address confounding bias in difference-in-difference studies.
- Interpretation limits
- Verbatim excerpts from the authors’ abstract; the closing excerpt may describe a finding, limitation or recommendation. Design, population, effect size and uncertainty have not been independently extracted or appraised. Read the complete abstract and linked paper before interpreting the findings.
Read complete abstract (282 words)
ObjectiveTo define confounding bias in difference-in-difference studies and compare regression- and matching-based estimators designed to correct bias due to observed confounders.Data sourcesWe simulated data from linear models that incorporated different confounding relationships: time-invariant covariates with a time-varying effect on the outcome, time-varying covariates with a constant effect on the outcome, and time-varying covariates with a time-varying effect on the outcome. We considered a simple setting that is common in the applied literature: treatment is introduced at a single time point and there is no unobserved treatment effect heterogeneity.Study designWe compared the bias and root mean squared error of treatment effect estimates from six model specifications, including simple linear regression models and matching techniques.Data collectionSimulation code is provided for replication.Principal findingsConfounders in difference-in-differences are covariates that change differently over time in the treated and comparison group or have a time-varying effect on the outcome. When such a confounding variable is measured, appropriately adjusting for this confounder (ie, including the confounder in a regression model that is consistent with the causal model) can provide unbiased estimates with optimal SE. However, when a time-varying confounder is affected by treatment, recovering an unbiased causal effect using difference-in-differences is difficult.ConclusionsConfounding in difference-in-differences is more complicated than in cross-sectional settings, from which techniques and intuition to address observed confounding cannot be imported wholesale. Instead, analysts should begin by postulating a causal model that relates covariates, both time-varying and those with time-varying effects on the outcome, to treatment. This causal model will then guide the specification of an appropriate analytical model (eg, using regression or matching) that can produce unbiased treatment effect estimates. We emphasize the importance of thoughtful incorporation of covariates to address confounding bias in difference-in-difference studies.
Abstract reproduced from the open-access article under the recorded cc by licence; © the authors. PMID 33978956 · DOI · PMC8522571
Key reviews and meta-analyses
The most cited review-type papers among PubMed’s best-match results for this heading, listed with bibliographic details only. Follow the links for the full text and access terms.
- Research article Avoiding bias from weak instruments in Mendelian randomization studies [4]
- Research article Immortal time bias in pharmaco-epidemiology [5]
- Research article Understanding the minimum clinically important difference: a review of concepts and methods [6]
- Research article Methodological quality (risk of bias) assessment tools for primary and secondary medical studies: what are they and which is better? [7]
- Research article Mediation analysis in epidemiology: methods, interpretation and bias [8]
- Research article Case-crossover analyses of air pollution exposure data: referent selection strategies and their implications for bias [9]
References
- Carra MC, Romandini P, Romandini M. Risk of Bias Evaluation of Cross-Sectional Studies: Adaptation of the Newcastle-Ottawa Scale. . 2026. PMID 40293188 · DOI · PMC13665615
- Zaidi AA, Mathieson I. Demographic history mediates the effect of stratification on polygenic scores. . 2020. PMID 33200985 · DOI · PMC7758063
- Zeldow B, Hatfield LA. Confounding and regression adjustment in difference-in-differences studies. . 2021. PMID 33978956 · DOI · PMC8522571
- Burgess S, Thompson SG, CRP CHD Genetics Collaboration. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol. 2011. PMID 21414999 · DOI
- Suissa S. Immortal time bias in pharmaco-epidemiology. Am J Epidemiol. 2008. PMID 18056625 · DOI
- Copay AG, Subach BR, Glassman SD, Polly DW, Schuler TC. Understanding the minimum clinically important difference: a review of concepts and methods. Spine J. 2007. PMID 17448732 · DOI
- Ma LL, Wang YY, Yang ZH, Huang D, Weng H, Zeng XT. Methodological quality (risk of bias) assessment tools for primary and secondary medical studies: what are they and which is better?. Mil Med Res. 2020. PMID 32111253 · DOI · PMC7049186
- Richiardi L, Bellocco R, Zugna D. Mediation analysis in epidemiology: methods, interpretation and bias. Int J Epidemiol. 2013. PMID 24019424 · DOI
- Janes H, Sheppard L, Lumley T. Case-crossover analyses of air pollution exposure data: referent selection strategies and their implications for bias. Epidemiology. 2005. PMID 16222160 · DOI
About this article
- Sources
- NLM Medical Subject Headings (definition, synonyms); NCBI PubMed (counts, identifiers); Europe PMC (open-access abstracts and citation counts).
- Verification
- Every reference was resolved against PubMed on 2026-11-04; title, year and DOI matched.
- Review state
- automated source verified; no human expert review has been recorded yet.
- Version
- 1, last updated 2026-11-04.
See how EvidenceLib is built. Spotted an error? Email support@evidencehelix.com.
Evidence strength
EvidenceHelix's computed view of the study designs behind this topic: how common meta-analyses, randomized trials and reviews are, how recent the literature is, and a plain-language reading. It describes the design mix of the literature. It is not a quality appraisal of any study, and it is not medical advice.