What makes a case series defensible
A case series describes a group of patients sharing a diagnosis, exposure, procedure or clinical feature. Unlike a single case report, it must define how cases were identified and which cases were included. Without an explicit denominator or comparison group, it usually cannot estimate risk or comparative effectiveness.
Before you start
- A precise case definition.
- A defined time period and setting.
- A reproducible method for finding all eligible cases rather than selecting only memorable cases.
- A fixed variable list and data dictionary.
- A plan for missing data and follow-up.
- Governance/ethics and data-access determination before extracting records.
Approval and governance
Usually needs formal review / authorisation
- Retrospective extraction of identifiable clinical records commonly requires ethics/governance review and data-access authorisation.
- Prospective collection or patient contact generally requires formal review and consent unless an authorised body approves another route.
- Use of identifiable photographs or publication of potentially identifiable details requires specific consent.
May follow a lighter or different route
- Some institutions route small descriptive case series differently from formal research, particularly when they arise from routine care; obtain an authorised classification rather than self-classifying.
- A fully anonymised public dataset may follow a different route if its terms permit the intended analysis.
Do not do this
- Do not select cases because their outcomes look interesting after seeing the data.
- Do not report a “success rate” as if it were comparative effectiveness when there is no appropriate denominator/control group.
- Do not combine data extracted under incompatible definitions without documenting the change.
Mediclinic Middle East publicly states that research projects carried out at MCME are to receive approval from its internal Research and Ethics Committee and applicable local regulatory authorities before initiation. The exact route varies by project, facility and emirate. Dubai projects may involve DSREC depending on applicability. This hub must therefore route users to the Research Office and current local forms rather than declaring a project “ethics exempt.” Institution-specific forms and contacts will be inserted after verification.
Step-by-step workflow
Define the clinical phenomenon and purpose
State whether the series is intended to describe presentation, management, adverse events, procedural experience, a new syndrome or another descriptive objective.
Write a reproducible case definition
Specify diagnostic criteria, procedure, exposure, time window, facility, age range and any exclusions. A second researcher should be able to apply the definition to the same records and identify the same eligible cases.
Define how cases will be found
Examples include an ICD/procedure-code query followed by manual validation, a pathology registry, a departmental procedure log or a prospective consecutive series. Record the search logic and dates.
Determine approvals and data access
Submit the protocol, case-identification method and variable list through the authorised route before extracting identifiable information.
Create the extraction form and data dictionary
Define each field, allowed values, units, source location in the record and coding for unknown/not done. Avoid adding variables after seeing outcomes unless labelled exploratory.
Pilot on a few cases
Test whether the case definition is workable and whether the same variables are available consistently. Correct the form before full extraction and document protocol amendments.
Extract consecutively and track exclusions
Keep a screening log with candidate cases, eligibility decision and reason for exclusion. This reduces cherry-picking and makes the denominator transparent.
Describe the cohort rather than overanalyse it
Use counts, proportions, medians/means with appropriate spread, clinical ranges and follow-up summaries. If comparisons are made, label them exploratory and avoid implying causal effects.
Protect identities
Small series are highly re-identifiable. Aggregate rare characteristics, use relative dates, limit exact locations, and obtain publication consent when individual descriptions or images could identify a patient.
Write transparently
Report setting, case definition, ascertainment method, dates, number screened/included, missing data, individual or aggregate outcomes, limitations and why the observations matter.
Archive and update the tracker
Store approvals, extraction form, dataset version, analysis and final citation; then add the publication to Research Pulse.
Final checklist
- Case definition is explicit.
- Case-finding method is reproducible.
- All eligible/consecutive cases were considered or the sampling approach is disclosed.
- Data access and ethics/governance route documented.
- No unsupported comparative-effectiveness claim.
- Missing data and follow-up reported.
- Publication consent addressed for identifiable details.
Analysis and presentation in more detail
- Create a participant table only if it does not create re-identification risk; otherwise present aggregated characteristics.
- Report the number of candidate cases, number eligible and number included. If the denominator of all treated patients is known, report it separately but do not imply incidence unless the sampling frame supports that inference.
- For continuous variables, report distribution-appropriate summaries and ranges when clinically useful. For binary outcomes, give numerator and denominator.
- If follow-up lengths differ, report the amount of follow-up rather than presenting every patient as equally observed.
- Any inferential comparison inside a small case series should be labelled exploratory and should not be the main justification for clinical claims.
Writing structure
- Title/abstract that identify the series and central clinical lesson.
- Introduction explaining why this group of cases is informative.
- Methods describing case definition, setting, dates, ascertainment, eligibility, variables, approvals and analysis.
- Results describing participant flow, characteristics, management and outcomes without interpretation.
- Discussion comparing observations with literature, explaining plausible interpretations and major selection/measurement limitations.
- Consent/ethics/governance statement and any required patient-publication consent statement.