Reporting guidelines: which one, and why they exist
Checklists that make a study reportable, reproducible, and reviewable. Most journals now require one.
Reporting guidelines exist because a large proportion of published research could not be evaluated, replicated or included in a synthesis, for the mundane reason that essential information was missing. They are not a quality standard for the study; they are a completeness standard for the report. A badly designed trial reported to CONSORT is still a badly designed trial — but at least you can tell.
| Design | Guideline | What it is for |
|---|---|---|
| Randomised trial | CONSORT (+ extensions for non-pharmacological and social interventions) | Flow diagram, randomisation, blinding, all outcomes pre-specified |
| Trial protocol | SPIRIT | What you will do, published before you do it |
| Systematic review / meta-analysis | PRISMA 2020 | Search strategy, screening, flow diagram, risk-of-bias assessment |
| Review protocol | PRISMA-P; register on PROSPERO | Pre-specification of a review |
| Observational study | STROBE | Cohort, case-control and cross-sectional reporting |
| Qualitative research | COREQ (interviews/focus groups) or SRQR | Reflexivity, sampling, analysis, coding, trustworthiness |
| Single case report | CARE | Timeline, intervention, outcomes, patient perspective, consent |
| Single-case experimental design | SCRIBE | Phase design, replication, visual and statistical analysis |
| Quality improvement | SQUIRE 2.0 | Context, intervention, iterations, unintended consequences |
| Diagnostic accuracy | STARD | Reference standard, blinding, flow, 2×2 table |
| Economic evaluation | CHEERS | Perspective, horizon, discounting, sensitivity analysis |
| Any psychology manuscript | APA JARS (Quantitative / Qualitative / Mixed) | The APA’s own reporting standards |
Psychotherapy-specific things the generic checklists under-specify
- Therapist characteristics: how many, what training, what supervision, what allegiance. Therapist effects exceed treatment effects, and most trials report nothing about them.
- Treatment fidelity AND competence: whether the protocol was followed, and whether it was delivered well. These are different and both are usually missing.
- The comparator, described in enough detail to know whether it was bona fide.
- Deterioration and adverse events. Psychotherapy trials have historically not reported harms at all, which is a serious and now widely criticised omission.
- Attrition by arm, with reasons, and how missing data were handled.
- Allegiance of the investigators, declared rather than inferred.
Sources
Schulz, K. F., Altman, D. G., & Moher, D. (2010). CONSORT 2010 statement: Updated guidelines for reporting parallel group randomised trials. BMJ, 340, c332
guidelinePage, M. J., et al. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71
guidelineTong, A., Sainsbury, P., & Craig, J. (2007). Consolidated criteria for reporting qualitative research (COREQ). International Journal for Quality in Health Care, 19(6), 349–357
guidelineGagnier, J. J., et al. (2013). The CARE guidelines: Consensus-based clinical case reporting guideline development. BMJ Case Reports
guideline