ScholarGate
Get started
Process / pipelineResearch MethodologyObservational study quality assessmentPipeline

Newcastle-Ottawa Scale for Observational Study Quality Assessment

Also known as: NOS

OriginatorWells et al. (Ottawa Hospital Research Institute)Year2000Sources1Related methods6

The Newcastle-Ottawa Scale (NOS) is a widely used tool for assessing the methodological quality of observational studies (case-control and cohort designs) included in systematic reviews and meta-analyses. Developed by Wells et al. at Ottawa Hospital in 2000, it provides explicit criteria and a star-based scoring system that enables transparent, quantitative comparison of study quality across evidence syntheses.

Key highlights

  • Specifically designed for observational study designs (case-control and cohort), unlike scales developed for RCTs
  • Star-based system (rather than binary judgments) preserves study quality as continuous, allowing nuanced ranking and weighting in meta-analyses
  • Explicitly addresses selection bias, confounding, and measurement, the three critical domains of observational study validity
  • Separate versions for case-control and cohort designs allow design-specific assessment criteria
  • Simple, transparent checklist format facilitates training of reviewers and reproducible assessments

Intuition

This section is available to Pro members. Upgrade to Pro

How it works

This section is available to Pro members. Upgrade to Pro

When to use it

NOS is standard for assessing case-control and cohort study quality in systematic reviews, health technology assessments, and clinical guideline development. Use when observational evidence forms a significant portion of included studies and when subgroup or sensitivity analyses by study quality are planned.

Strengths & limitations

Strengths
  • Specifically designed for observational study designs (case-control and cohort), unlike scales developed for RCTs
  • Star-based system (rather than binary judgments) preserves study quality as continuous, allowing nuanced ranking and weighting in meta-analyses
  • Explicitly addresses selection bias, confounding, and measurement, the three critical domains of observational study validity
  • Separate versions for case-control and cohort designs allow design-specific assessment criteria
  • Simple, transparent checklist format facilitates training of reviewers and reproducible assessments
Limitations
  • No validation of inter-rater reliability or predictive validity; modest evidence that NOS scores correlate with bias magnitude
  • Comparability domain allows only 0–2 stars regardless of number of confounders adjusted; studies adjusting for many confounders do not score higher than those adjusting for one or two key confounders
  • Scoring criteria are somewhat subjective (e.g., 'adequate' follow-up); explicit thresholds (e.g., >80% follow-up) are not provided in all items
  • Does not address publication bias, outcome reporting bias, or qualitative synthesis; quantitative star score may overshadow qualitative assessment
  • User guide is brief; reviewers sometimes disagree on which studies meet criteria without detailed supplementary algorithms

Common pitfalls

This section is available to Pro members. Upgrade to Pro

Applications

This section is available to Pro members. Upgrade to Pro

Frequently asked

Is NOS appropriate for cross-sectional studies?

The original NOS is not designed for cross-sectional studies; it focuses on case-control and cohort designs. For cross-sectional studies, adapt the NOS criteria or use alternative tools such as the CASP Cross-Sectional Checklist or STROBE guidelines.

What is the difference between the case-control and cohort versions?

Case-control studies assess exposure retrospectively (selecting participants by outcome status and looking back); cohort studies assess exposure prospectively (selecting by exposure status and following forward). The NOS versions reflect this: case-control emphasizes control selection and exposure measurement; cohort emphasizes cohort representativeness and follow-up completeness.

Can I use NOS for randomized trials?

No. The NOS is not designed for RCTs. Use Cochrane Risk of Bias 2.0 or CASP RCT Checklist for randomized trials. Although some reviewers have applied NOS to RCTs, it does not capture randomization bias, blinding of outcome assessment, or allocation concealment.

What does a study need to score 7+ stars (high quality)?

A study typically needs: clear, representative case/cohort selection; adjustment for major confounders (earning comparability stars); and validated outcome ascertainment with >80% follow-up (for cohort studies). Exact star distributions vary; consult domain thresholds.

Should I exclude studies scoring <4 stars from my meta-analysis?

Not automatically. Report both analyses: main analysis including all studies, and a sensitivity analysis excluding low-quality studies. Examine whether exclusion changes pooled estimates materially. If estimates are stable, the main analysis is robust to quality; if unstable, describe the impact of bias on your conclusions.

Sources

  1. 1.
    Wells, G. A., Shea, B., O'Connell, D., Peterson, J., Welch, V., Losos, M., & Tugwell, P. (2000). The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Retrieved from Ottawa Hospital Research Institute.
    https://scholar.google.com/scholar?q=Wells%2C%20G.%20A.%2C%20Shea%2C%20B.%2C%20O'Connell%2C%20D.%2C%20Peterson%2C%20J.%2C%20Welch%2C%20V.%2C%20Losos%2C%20M.%2C%20%26%20Tugwell%2C%20P.%20(2000).%20The%20Newcastle-Ottawa%20Sc

You have read it. What now?

Ask about this methodSimilar methods

Cite this page

ScholarGate. (2026, June 3). Newcastle-Ottawa Scale. ScholarGate. https://scholargate.app/research-methodology/newcastle-ottawa-scale