Strata Academy

JBI critical appraisal tools explained – observational study checklists

Choosing the right JBI checklist for case series, cross-sectional studies, cohorts, and qualitative research

1. What is JBI?

The Joanna Briggs Institute (JBI) publishes critical appraisal tools for diverse study designs – analytical observational studies, prevalence studies, case series, case reports, qualitative research, and economic evaluations.

JBI methodology is widely taught in nursing, allied health, public health, and mixed-methods programmes. Each checklist is design-specific; there is no single universal 'JBI score' applicable to every paper type.

JBI tools emphasise whether a study is trustworthy for its stated purpose – describing prevalence, generating hypotheses, or exploring lived experience – rather than forcing causal language inappropriate to the design.

Official checklists are updated periodically. Use the version aligned with your course or the JBI manual current at appraisal time, and cite the checklist name in your appendix.

2. Choosing the correct JBI checklist

Match the checklist to the study design described in methods – not the authors' preferred terminology or journal section headings.

If authors call a study 'prospective observational' but randomly assign treatment, the design is an RCT – use ROB 2, not JBI cohort tools.

Diagnostic accuracy studies should use QUADAS-2 with STARD reporting guidance, not generic observational JBI checklists, because validity and applicability domains differ.

When multiple designs appear (e.g. baseline cross-sectional survey plus follow-up), appraise the component relevant to your research question or acknowledge dual-design complexity explicitly.

• Analytical cross-sectional study → JBI cross-sectional checklist
• Case series (no comparator) → JBI case series checklist
• Case report → JBI case report checklist
• Cohort with exposure–outcome follow-up → JBI cohort checklist
• Qualitative study (interviews, focus groups) → JBI qualitative checklist
• Prevalence survey → JBI prevalence checklist
• Diagnostic test study → QUADAS-2 preferred

3. Typical JBI domains

While items differ by checklist, JBI tools commonly assess whether the study design can answer the question posed, whether sampling was appropriate, and whether measurement of exposures and outcomes was valid and reliable.

Selection bias items ask whether participants represent the target population and whether recruitment could skew findings. Convenience samples from single centres limit generalisability even if analysis is statistically sophisticated.

Measurement bias items examine whether exposure and outcome were defined consistently and measured with instruments that were validated for the population studied.

Confounding and analysis items ask whether statistical methods match the design – e.g. logistic regression assumptions for cross-sectional prevalence, adequate follow-up for cohort studies.

• Appropriateness of design for the question
• Sample and setting – representativeness and recruitment
• Exposure / intervention / phenomenon measured validly
• Outcome / finding measured with acceptable reliability
• Confounding identified and addressed where possible
• Follow-up adequate for cohort designs

4. JBI vs ROB 2 / ROBINS-I / NOS

Intervention studies with randomisation → ROB 2, not JBI. Non-randomised intervention comparisons with causal intent → ROBINS-I is Cochrane's preferred risk-of-bias framework.

JBI excels for descriptive observational designs, prevalence estimates, case series, and qualitative work where the question is not primarily causal effect estimation.

Newcastle–Ottawa Scale (NOS) remains common in meta-analyses of cohort and case–control studies. Some fields expect NOS stars; others expect ROBINS-I – follow your supervisor and journal convention.

CASP checklists overlap pedagogically with JBI for teaching critical appraisal skills but use different wording – pick one framework per paper and apply consistently.

5. Answering JBI items

JBI checklists typically use yes / no / unclear / not applicable. 'Unclear' usually means not reported – document as a reporting limitation rather than assuming the method was adequate.

Summarise overall quality in words (major limitations in selection and measurement) rather than inventing a numeric score unless your course explicitly requires one.

Quote short evidence from the paper for each item – page and section – so examiners can verify your judgement.

For qualitative studies, assess reflexivity, congruity between methodology and analysis, and whether findings are supported by participant quotations.

6. Common mistakes

Using one JBI checklist for every paper in a journal club regardless of design.

Applying JBI cohort tools to RCTs because the setting was a hospital ward.

Ignoring qualitative-specific criteria (reflexivity, theoretical framework, cultural context) on interview studies.

Treating 'unclear' items as 'yes' to produce a favourable summary.

Confusing JBI critical appraisal with JBI systematic review methodology – different toolsets.

7. StrataResearch and JBI

StrataResearch selects JBI-aligned checklists when observational designs such as case series or cross-sectional studies are detected from manuscript structure.

RCTs route to ROB 2; systematic reviews route to AMSTAR 2 and PRISMA; diagnostic accuracy routes to QUADAS-2 – reducing framework mismatch in mixed journal clubs.

Upload via quick analysis to see structured domain output, then compare against your manual JBI worksheet.

Interactive version (quizzes, walkthroughs) loads when JavaScript is enabled.