Case reports and case series
Why this level?
A single case series has post-test or pre-test/post-test outcomes, without a comparison group. NHMRC 2009 sourceLevel IV applies when a series reports outcomes after an intervention with nothing to compare against. A report of a new presentation, an unexpected reaction or a disease course answers no intervention question and sits outside this ladder, and a series that sampled patients by exposure may in fact be a cohort.
What question does this help answer?
What did this condition, reaction or treatment course look like in these patients? Reports and series can raise a signal, describe a new or rare presentation, document early experience with a technique, or generate a hypothesis for a comparative study.
When to use it
Use one for a rare or newly recognized disease, an unexpected adverse event, early experience with a new procedure, or a presentation too uncommon for a comparative study to be feasible yet.
When not to use it
Do not use a case series to claim that a treatment works or is safer than an alternative. Without a comparison, the outcomes cannot be separated from the natural course of the condition, regression to the mean, other treatments or the way patients were selected. If patients were identified by an exposure and followed for outcomes, the study may be a cohort that can estimate a risk, and it should be designed and reported as one.
How it works in plain language
Define what makes a patient a case before searching for them: the diagnostic criteria, the source, the time window and whether every eligible patient will be included. Describe exposures, treatments, follow-up and outcomes in the same way for each patient, and account for every eligible patient, including those excluded or lost to follow-up.
Key assumptions and design requirements
- The case definition and inclusion rules are explicit and, where possible, fixed before outcomes are known.
- Consecutive or complete inclusion from a stated source is preferred to choosing memorable cases, and the report says which was done.
- Follow-up duration and completeness are reported for each patient.
- Alternative explanations for the outcomes are considered, including the natural history of the condition.
- Consent and privacy requirements are met. A small series of rare cases can identify patients even without names.
How it is sized
There is no power calculation for a case series, and Mallard does not produce one. Report how many patients met the definition and why the series ended. If a proportion is reported, give its confidence interval, which is often wide.
Clinical example
A clinic sees several patients who develop the same unusual skin reaction after starting a new drug. A series stating the case definition, how patients were found, their exposures, timing and outcomes can raise a signal worth investigating. It cannot estimate how often the reaction occurs among everyone taking the drug, because nobody who took it without the reaction was counted. This example is illustrative.
What to watch out for
A case series has no comparison group, so it cannot show that an exposure or treatment caused the outcomes it describes, however striking they are. Selecting cases because they were memorable, successful or already published exaggerates both benefit and harm, and a study labelled a case series is sometimes a cohort that could have estimated a risk.
How to report it
CARE is the guideline for a single case report and PROCESS for a surgical case series. No reporting guideline dedicated to case series outside surgery in general was found in a PubMed search on 2026-09-16; the ones that turned up are scoped to a single field. For a series outside surgery, state the case definition, how cases were identified, and what follow-up was available. Say whether the patients were consecutive or selected, which is what decides the population the series represents. A reporting guideline is not a design manual: use it to make the study transparent, not as a substitute for careful design. Check the target journal's own requirements as well, which this page cannot know.
Before using this in your workspace
Record the case definition, source and search window, whether inclusion was consecutive, the items collected for every patient, and the claim the series is meant to support. If the aim is really a comparison, reconsider the design before collecting data.
- The CARE guidelines: consensus-based clinical case reporting guideline development (opens in a new tab) · BMJ Case Rep · doi 10.1136/bcr-2013-201554
- Distinguishing case series from cohort studies (opens in a new tab) · Ann Intern Med · doi 10.7326/0003-4819-156-1-201201030-00006
- Appropriate use and reporting of uncontrolled case series in the medical literature (opens in a new tab) · Am J Ophthalmol · doi 10.1016/j.ajo.2010.08.047
- Methodological quality and synthesis of case series and case reports (opens in a new tab) · BMJ Evid Based Med · doi 10.1136/bmjebm-2017-110853
- The CARE guidelines: consensus-based clinical case reporting guideline development (opens in a new tab) · J Med Case Rep · doi 10.1186/1752-1947-7-223
- The PROCESS 2020 Guideline: Updating Consensus Preferred Reporting Of CasE Series in Surgery (PROCESS) Guidelines (opens in a new tab) · Int J Surg · doi 10.1016/j.ijsu.2020.11.005