Contents
On this page
- Absence of a single agreed threshold
- Contents of 586 published case series
- Bodies that set patient-number thresholds
- The principal methodological proposal
- Practical significance of the patient-number threshold
- Statistical rationale for a minimum number of patients
- Worked example: nine transplant recipients
- Selecting the article-type label for a manuscript
- Unresolved questions
- Key points
- Application to your study
- Frequently asked questions
Academy · Research Writing · case series patient numbers
Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting
No agreed number of patients distinguishes a case series from a case report, and this disagreement is itself the principal finding rather than a gap awaiting resolution. Published sources set the threshold at two, three, four, five, six or ten patients, or decline to set one, and the largest empirical study of the question concluded that no clear distinction exists in the published literature. For a given manuscript, the number is determined locally by the target journal and the ethics committee, which may not agree with each other.
This question is usually posed as though it had a definitive answer, namely the number of patients required before a report can be called a case series; no such answer exists. An analysis of 586 articles published with “case series” in the title found a median of 7 patients and a range from 1 to 6,432, with nearly a third containing fewer than five. Journals, ethics boards and reference works meanwhile set thresholds at three, four, five, six and ten, and a standard epidemiology text states that the number of patients in a series is not limited. The practical answer is that two numbers are required, those of the journal and of the ethics committee, and they often differ.
Learning objectives
- The thresholds in use, source by source, and the bodies that set each one
- The contents of 586 published “case series” articles and 219 “case report” articles
- The practical significance of the threshold as a trigger for ethics review in several systems
- The one serious methodological proposal in the literature, and the reasoning behind it
- The purpose of a statistical minimum number of patients
- How to select the label for a manuscript, and what the methods section should state
Common misconceptions
- “A case series is five or more patients.” Five is one threshold among several. A hospital institutional review board (IRB) exempts series of five or fewer from review and requires six or more to be submitted; a journal’s author guidelines classify four to nine as a case series; a research ethics board sets the line at four; a standard epidemiology reference says more than one. Five is a defensible local rule, not a definition.
- “There must be a consensus somewhere; I just have not found it.” The largest study of the question analysed 586 articles and concluded that no clear distinction between case reports and case series exists in the published literature. The absence of consensus is documented, not merely unobserved.
- “The distinction is just a labelling convention, so it does not matter.” In several systems, the distinction determines whether a project constitutes research. One research ethics board requires no review for a case report of three or fewer patients but treats four or more as a case series, and therefore as research requiring review. An incorrect label may therefore result in unreviewed research.
- “More patients always means a better article type.” Above a certain size, the article is no longer a series. One journal’s guidelines classify ten or more patients with statistical analysis as an Original Article, and the proposal from the 586-article analysis set an upper limit of ten for a series. A 6,432-patient “case series” exists in the literature, which falls outside the intended use of the label.
- “The patient count is what distinguishes the two designs.” Journals and committees use the patient count because it is verifiable. The designs are distinguished more fundamentally by whether the patients were assembled from a defined population by a stated rule. Two consecutively identified patients from a named cohort are closer in design to a series than nine hand-picked cases are.
Absence of a single agreed threshold
The question of how many patients constitute a case series comprises two distinct questions, which have different answers.
The first is methodological: the point at which a collection of patients begins to function as a study rather than an anecdote. This question has a reasoned answer, which depends on whether a population and a selection rule can be described rather than on a count.
The second is administrative: the category the journal will assign to the manuscript, and whether the ethics committee will require review. This question has a definite but local answer: it depends entirely on the journal and the committee concerned, and the two bodies frequently use different numbers for the same manuscript.
The question appears unanswerable because the literature contains confident, specific, mutually incompatible answers to the second question presented as though they were answers to the first. A journal that classifies three or fewer patients as a case report is not making a claim about epidemiology; it is managing article types in a submission system 37. A research ethics board that treats four patients as research is not disagreeing with the journal about methodology; it is drawing a regulatory line where it needs one 20. Both are legitimate. Neither constitutes the definition, and treating either as the definition is the source of difficulty for authors.
Our companion page on the definitional differences between case series and case reports takes the design question in more detail; this page is about the number.
Contents of 586 published case series
The most useful contribution in this literature is a measurement rather than an opinion. Abu-Zidan and colleagues analysed 586 articles published in 2009 with “case series” in the title and 219 articles typed as case reports, and counted the patients in each 24.
The articles titled “case series” had a median of 7 patients, with a range from 1 to 6,432. Nearly a third — 31.7% — contained fewer than five patients. The articles typed as case reports had a median of 4 patients, with a range from 1 to 178, and 52.1% contained fewer than five.
The finding emerges from a comparison of the two distributions. The medians differ by three patients. Both ranges start at one. Both distributions have a substantial share below five. Each contains articles that, by any of the published thresholds, were labelled as the other type. A single-patient article titled “case series” and a 178-patient article typed as a “case report” both exist in this sample.
The authors’ conclusion was that no clear distinction between case reports and case series exists in the published literature 24. This accurately describes the state of the field, and it is more useful to authors than any individual threshold because it indicates the appropriate course of action: rather than seeking a definition, authors should consult the article-type page of the target journal and the standard operating procedure of the relevant committee.
Patient numbers measured across articles published in 2009, from the analysis of 586 “case series” and 219 “case report” articles that concluded no clear distinction exists between the two types in the published literature.
Bodies that set patient-number thresholds
The table below collects the thresholds that are currently in force in different places. They are not variations on a theme; they are five different rules, and a single manuscript with four patients would be a case report to one body, a case series to three others, and exempt from review under one while constituting research requiring review under another.
Three features of the collection merit attention. First, the lowest threshold and the highest differ by a factor of ten — some sources treat two patients as a series, one treats fewer than ten as one, and one declines to set an upper limit at all 30,36,37. Second, the ethics thresholds and the publishing thresholds were set for different purposes and do not track each other: the research ethics board line at four and the hospital IRB line at six are both about when review is required, while the journal line at four to nine is about which submission category to use 7,20,37. Third, no source in this collection cites another, which is why the disagreement persists rather than resolving.
The practical consequence is that “is this a case series?” is not one question. It is at least two: what should the manuscript be called, and does the project require ethics review. Each should be answered separately, with reference to the relevant local document.
The patient-number thresholds in force, source by source
| Who sets it | Case report | Case series | Above that | What the threshold is for |
|---|---|---|---|---|
| A standard epidemiology reference | One individual patient | More than one patient; the number is not limited | No upper limit set | Defining the terms, with no administrative purpose |
| A journal’s author guidelines (AME Case Reports) | Three or fewer patients | Four to nine patients | Ten or more with statistical analysis becomes an Original Article | Assigning the submission category and word limit |
| A research ethics board | Three or fewer — no review required | Four or more — this is a case series and therefore research requiring review | Not distinguished further | Deciding when a write-up becomes research |
| A hospital IRB | Not defined by count | Five or fewer — exempt from review | Six or more — must be submitted for review | Deciding when a write-up becomes research |
| The proposal from the 586-article analysis | Four or fewer, reported individually | More than four — a floor of five, which is also given as the minimum for statistical pooling | Proposed upper limit of ten | Giving the distinction a methodological basis it currently lacks |
| What the literature actually does | Median 4 patients across 219 articles; range 1 to 178; 52.1% under five | Median 7 patients across 586 articles; range 1 to 6,432; 31.7% under five | Articles titled “case series” run to thousands of patients | Nothing — this is the measured state of practice, not a rule |
Compiled from the published definitions, journal author guidelines, ethics committee rules and the empirical analysis of 586 “case series” and 219 “case report” articles cited in the references for this article. None of these sources cites another, and the thresholds are not reconcilable. Authors should consult the article-type page of their journal and the standard operating procedure of their committee; neither can be predicted from this table.
The principal methodological proposal
Having documented that no distinction existed, the authors of the 586-article analysis proposed one 24. Their proposal has four parts, and the reasoning behind each merits examination because it is the only argued position in this literature, as opposed to an administrative rule.
A case series should include more than four patients — a floor of five 24. Four or fewer patients should be reported individually, as case reports, because with four or fewer the reader needs each patient’s detail and a summary statistic conceals more than it conveys. Five is also proposed as the minimum for statistical pooling, which is the substantive reason for the floor rather than a convention 24. The authors also proposed an upper limit of ten, above which the article is better classified as another type.
The upper limit is the least widely recognised element of the proposal and the one that most often applies in practice. The reasoning is that once there are enough patients to compute meaningful statistics, the relevant design questions are those that apply to observational studies (how patients were identified, what the denominator was, and how complete the follow-up was), and calling the article a case series invites the reader to excuse the absence of answers. This reasoning is consistent with the journal guidelines that classify ten or more patients with statistical analysis as an Original Article rather than a series 37.
This is a proposal, not an adopted standard. No journal is obliged to follow it, and the measured distributions show that the literature does not. Its value lies in serving as a reasoned default when no local rule applies, and as a framework for considering what the number is for.
Practical significance of the patient-number threshold
The most important reason for attending to the threshold is not bibliographic. In several systems the patient count is the trigger for ethics review.
One research ethics board sets the line at three: a case report of three or fewer patients needs no review, while four or more patients makes the project a case series and therefore research requiring approval 20. A hospital IRB sets it elsewhere, exempting case series of five or fewer patients from review and requiring six or more to be submitted 7. Under the first rule, a four-patient write-up is research. Under the second, it is not. The clinician’s actions are identical in both cases.
Two practical implications follow, both concerning sequence. First, the determination must be made before the data are collected, not when the manuscript is drafted. Retrospective ethics approval is difficult or impossible in most systems, and a journal that requests an ethics statement for a five-patient series will not accept “review was not required” without knowing under whose rule that determination was made. Second, the determination should be obtained in writing. “Our committee said it was not research” is not a statement a reviewer can verify; a reference to the committee’s own standard operating procedure, with the threshold it specifies, is verifiable.
Consent is a separate matter from review and is not diminished by larger patient numbers. Each identifiable patient in a series requires consent on the same terms as the single patient in a report, and the more patients a series contains, the more likely it is that one of them is identifiable from the combination of details. Our pages on consent for patient-identifiable material and journal consent requirements for case reports cover the standard, and the worked transplant COVID-19 example takes one case through it.
Statistical rationale for a minimum number of patients
The floor of five in the published proposal is tied to a specific function: it is given as the minimum for statistical pooling. This function merits examination, because it provides the only principled argument for any particular number and indicates what the threshold protects.
Below about five patients, a summary statistic is not merely imprecise — it is misleading in a particular way. A mean of three values hides the individual values, and in a small series the individual values are the evidence. A reader who is given “mean age 41 years” for three patients has lost the information that the ages were 19, 42 and 62, which may be the most clinically important thing in the report 4. Reporting each patient individually, which is what the proposal asks for at four or fewer, preserves it.
Above five, summary reporting becomes justified, and the questions change accordingly. Once a mean is reported, a reader is entitled to ask what population it describes, how the patients were selected, and whether the series is consecutive or assembled. These are observational-study questions, and a series above the floor must answer them. In this sense the number is not arbitrary: it marks the point at which the reporting obligations change.
None of this constitutes a power calculation, and no case series is statistically powered. If the intention is to estimate an effect or test a difference, the relevant number of patients is that given by a sample size calculation, and the article type is not a case series. Our biostatistics support covers that distinction.
Worked example: nine transplant recipients
A concrete example illustrates how the thresholds apply in practice. A report of nine kidney transplant recipients with COVID-19 was published as a “case series” 4. The patients had a mean age of about 41 years and had been transplanted a mean of about five years earlier 4. All nine had fever, five developed acute kidney injury and three were hospitalised.
Nine patients is above every published floor and below the proposed ceiling of ten, and it falls inside the four-to-nine band that one journal’s guidelines define as a case series 37. The label is therefore uncontroversial in this case, which makes the example suitable for examining what the label obliges the authors to do.
The reporting merits attention. Counts are given rather than percentages: nine of nine, five of nine, three of nine. With nine patients a percentage conveys false precision; “55.6% developed acute kidney injury” implies a resolution the sample does not have, whereas five of nine conveys the same information accurately. The means are also supported by a count large enough to make them meaningful, which is the purpose of the floor. However, the reader still requires information that a series label does not supply: how these nine patients were identified, over what period, and out of how many transplant recipients with COVID-19 at the centre. That last number is the denominator, and a series that reports it is substantially more useful than one that does not.
By contrast, a report of two renal transplant recipients with COVID-19 was published and titled as “a case report”, not a case series — even though a standard epidemiology reference would call any collection of more than one patient a series 9,30. Both labels are defensible under some published rule, and this single comparison encapsulates the state of the field.
Selecting the article-type label for a manuscript
The steps should be completed in the following order, because two of them are subject to deadlines and the others are not.
The ethics threshold should be identified first, before any data are collected. The committee’s own standard operating procedure or submission guidance should be consulted to identify the patient number at which a write-up becomes research. If the document does not state one, the committee should be asked and the answer obtained in writing. This step has a deadline because it cannot be completed retrospectively.
The article-type definitions of the target journal should then be consulted. The relevant source is the article-type page, where the numbers are stated, rather than the general instructions to authors. If it specifies a band, that band determines both the label and the word limit. If it does not, the proposed defaults are a reasonable fallback: five or more for a series, four or fewer reported individually, and ten or more reconsidered as an observational study.
The design should then be determined, independently of the count. Authors should establish whether the patients were consecutive or selected, and from what population and over what period. If a source population and a selection rule can be stated, they should be reported, because they confer whatever external validity the series has, which matters more than the number. If they cannot be stated, this should also be acknowledged, and language implying a cohort should be avoided.
Finally, a methods sentence should be written that makes all three elements explicit. The sentence should state the number, the selection, and the ethics determination with the rule under which it was made. That single sentence prevents the most common reviewer query on a small series, which concerns not the label but the source of the patients.
The methods sentence on a small series
Label asserted, origin unstated
We present a case series of patients with COVID-19 following kidney transplantation who were treated at our institution. Ethical approval was not required for this case series. Patient characteristics and outcomes were analysed and are presented as means and percentages.
No patient count in the sentence. The reader cannot tell whether “a case series” here means two patients or two hundred, and the label alone does not narrow it — published “case series” range from 1 to 6,432 patients.
“Ethical approval was not required” with no rule cited. Under one research ethics board’s rule four or more patients makes this research requiring review; under a hospital IRB’s rule five or fewer is exempt. Without naming whose rule applied, the statement cannot be checked.
“Treated at our institution” is not a selection rule. The statement does not indicate whether cases were consecutive or chosen, or over what period. A reader cannot distinguish a complete series from a set of interesting cases, and that distinction is what the series label is intended to convey.
Percentages on an unstated and probably small denominator. If the series is nine patients, “55.6%” asserts a precision the sample cannot support; five of nine conveys the same information accurately.
No denominator for the series itself. The number of transplant recipients under follow-up from whom the nine cases were drawn is not given. Without it, the series provides no information about frequency, and readers are likely to infer a frequency regardless.
Number, selection and determination all visible
We report nine kidney transplant recipients with laboratory-confirmed COVID-19, comprising all such patients identified from the transplant unit’s clinical database between 1 March and 30 June 2020; no patient meeting these criteria was excluded. We describe the series as a case series on the basis of our institutional review board’s standard operating procedure, which exempts series of five or fewer patients from review and requires six or more to be submitted; the project was accordingly submitted and approved as a case series, reference ______. Written consent for publication was obtained from each patient individually. With nine patients we report counts rather than percentages for categorical outcomes, and we do not present comparative statistics; the denominator of transplant recipients under follow-up at the unit during the study period was 214.
The count appears in the first clause, so the label is immediately interpretable and the article type can be checked against the journal’s own band.
The ethics determination names the rule, the threshold and the outcome, which converts an unverifiable assertion into something a reviewer or editor can confirm.
Case identification is described as a complete ascertainment from a named source over a stated period, which is the basis of any claim the series has to be representative.
Counts replace percentages at a sample size where percentages mislead, and the absence of comparative statistics is stated rather than left to be noticed.
The source-population denominator is given, so a reader can see what proportion of the unit’s patients the series represents rather than having to estimate it.
For a small series, the first two questions of a reviewer are typically “how many, and where did they come from”. The second version answers both in advance and names the rule under which the ethics decision was made, which is the one element that an editor cannot verify independently.
Unresolved questions
The scope and limits of the foregoing account should be stated clearly.
There is no consensus threshold, and the largest study of the question found no clear distinction in the published literature. The thresholds in the table are real and in force where they are in force; none is authoritative beyond that.
The proposal of a floor at five and a ceiling at ten is argued rather than adopted. The 586 measured articles do not follow it — 31.7% of articles titled “case series” had fewer than five patients, and the range ran to 6,432 24. There is no evidence that journals or committees have moved towards it.
The claim that five is the minimum for statistical pooling is stated in that proposal as a rationale. This page reports it as such and does not extend it into a general statistical rule, because no such general rule exists for case series.
Whether any of this affects how readers use the literature remains an open question. It has not been shown that mislabelled series are read, cited or trusted differently. What is demonstrable is the administrative consequence: under some rules, a four-patient write-up constitutes unreviewed research. This is sufficient reason to determine the number correctly, irrespective of the epistemological question.
Key points
Key points
No agreed number of patients converts a case report into a case series, and the disagreement is documented rather than merely unresolved. An analysis of 586 articles titled “case series” found a median of 7 patients, a range from 1 to 6,432 and 31.7% containing fewer than five, against a median of 4 and a range of 1 to 178 in 219 articles typed as case reports, and concluded that no clear distinction exists in the published literature. The thresholds in force range from two patients to ten: a standard epidemiology reference sets no limit; one journal’s guidelines use three or fewer, four to nine and ten or more; a research ethics board treats four or more as research requiring review; a hospital IRB exempts five or fewer; and the one argued proposal sets a floor of five, also given as the minimum for statistical pooling, with an upper limit of ten. Two local determinations are therefore required rather than one general answer: the threshold of the ethics committee, obtained in writing before data collection, and the article-type band of the target journal. The methods section should then state the count, the selection rule and the ethics determination, because these three facts matter considerably more than the label.
Application to your study
Application to your study
- The ethics threshold should be obtained in writing first. The patient number at which the committee treats a write-up as research should be established before any data are collected, because retrospective approval is difficult or impossible in most systems.
- The journal’s article-type page, rather than its general author instructions, should be consulted. That page specifies the patient bands and thereby determines both the label and the word limit.
- The patient count should appear in the first clause of the methods. A statement such as “We report nine patients” conveys more than “we present a case series”, given that published series range from 1 to 6,432 patients.
- The selection rule, and not only the setting, should be stated. Whether cases were consecutive or selected, from what source, over what period and with what exclusions is what distinguishes a series from a collection of interesting cases.
- The source-population denominator should be reported where available. Nine cases out of 214 patients under follow-up is considerably more informative than nine cases alone.
- Counts rather than percentages should be reported below about five patients, and counts are preferable up to ten. Five of nine is an accurate representation; 55.6% is not.
- The article type should be reconsidered above ten patients. One journal’s guidelines classify ten patients with statistical analysis as an Original Article, and the published proposal sets ten as the ceiling for a series; at that size, the questions applicable to observational studies apply.
- Consent should be obtained from each patient individually. The requirement for consent does not diminish with larger numbers, and a larger series makes individual identifiability more likely, not less.
Frequently asked questions
There is no agreed number, and sources disagree. A standard epidemiology reference says more than one, with the number not limited. One journal’s author guidelines set three or fewer as a case report, four to nine as a case series, and ten or more with statistical analysis as an Original Article. A research ethics board draws the line at four. A hospital IRB exempts five or fewer from review and requires six or more to be submitted. The one argued methodological proposal sets a floor of five and a ceiling of ten. Moreover, the measured literature follows none of them consistently: across 586 articles titled “case series”, the median was 7 patients with a range from 1 to 6,432, and 31.7% had fewer than five. Authors require the number used by their journal and the number used by their committee, not a general figure.
Consensus is absent, and that absence has been measured rather than assumed. Abu-Zidan and colleagues analysed 586 articles published in 2009 with “case series” in the title and 219 typed as case reports, and concluded that no clear distinction between the two exists in the published literature. The distributions overlap heavily: the medians differ by only three patients, 7 against 4; both ranges start at one; and both have a substantial share below five patients, 31.7% and 52.1% respectively. A single-patient article titled “case series” and a 178-patient article typed as a “case report” both appear in that sample. The sources that do state thresholds set them independently and do not cite one another, which is why the disagreement persists.
In several systems, yes, and this is the most consequential reason to resolve the number early. One research ethics board requires no review for a case report of three or fewer patients but treats four or more as a case series and therefore as research requiring approval. A hospital IRB exempts case series of five or fewer patients from review and requires six or more to be submitted. Under the first rule a four-patient write-up is research; under the second it is not, with no change in what the clinician did. Because retrospective approval is difficult or impossible in most systems, the committee’s threshold should be identified before data are collected, and the determination should be obtained in writing, together with the rule under which it was made, so that the ethics statement can be verified.
Under some rules, yes. A standard epidemiology reference defines a case report as a report on one individual patient and a case series as a collection of information on more than one, which makes two patients a series. A publisher’s article-type definition takes the same line, treating a case series as simply more than one case report with no numeric cutoff. But a journal using the three-or-fewer rule would call two patients a case report, and the argued methodological proposal would require them to be reported individually, since it asks for four or fewer to be written up as separate case reports. Practice is also divided: a report of two renal transplant recipients with COVID-19 was published and titled as “a case report”. The article-type page of the target journal should be followed.
Some sources set one and others explicitly do not. A standard epidemiology reference states that the number of patients in a series is not limited. Against that, one journal’s author guidelines classify ten or more patients with statistical analysis as an Original Article rather than a case series, and the proposal from the 586-article analysis set an upper limit of ten. The reasoning for a ceiling is that once there are enough patients to compute meaningful statistics, the questions that matter are observational-study questions — how patients were identified, what the denominator was, how complete follow-up was — and the case series label invites a reader to excuse the absence of answers. For reference, the measured range of articles titled “case series” ran as high as 6,432 patients.
It is tied to statistical pooling: the proposal gives five as the minimum for pooling, and asks that four or fewer patients be reported individually instead. The reasoning is about what a summary statistic does at small numbers. A mean of three values conceals the individual values, and in a very small series the individual values are the evidence — “mean age 41 years” across three patients loses the fact that the ages were 19, 42 and 62, which may be the clinically important point. Reporting each patient separately preserves it. Above the floor, summary reporting becomes justified but brings obligations with it: once a mean is reported, the reader is entitled to ask what population it describes and how patients were selected.
Counts, not percentages, should be used. A published series of nine kidney transplant recipients with COVID-19 reported that all nine had fever, five developed acute kidney injury and three were hospitalised, which is the appropriate form. “55.6% developed acute kidney injury” asserts a resolution that nine patients cannot support, whereas five of nine conveys the same information without false precision. Means are reasonable at this size if the count behind them is stated: that series gave a mean age of about 41 years and a mean of about five years since transplantation. Comparative statistics should not be presented, and their absence should be stated explicitly. The source-population denominator should also be given where available, since it allows a reader to judge what the series represents.
The facts should be stated, and the label should follow from them. Three elements belong in that sentence. The first is the count, in the first clause, because published series range from 1 to 6,432 patients and the label alone conveys nothing to a reader. The second is the selection rule (consecutive or selected, from what data source, over what period, with what exclusions), because it distinguishes a complete series from a set of interesting cases. The third is the ethics determination with the rule on which it rests: the committee’s threshold should be named, whether review was required under it should be stated, and the reference number should be given if one exists. “Ethical approval was not required for this case series” is unverifiable; “our IRB’s procedure exempts series of five or fewer and we submitted six” is checkable.
References
- Valença MM, Prieto Peres MF, Ramos de Andrade J. Why case reports still matter: reclaiming their epistemic, ethical, and clinical value in contemporary publishing. Headache Medicine. 2026;17(1):72-74. doi:10.48208/HeadacheMed.2026.9 read 9 October 2026. doi.org/10.48208/HeadacheMed.2026.9
- Taheri A, Adibi P, Sabbagh Jafari M, Saeedizadeh M, Rahimi A, Abbasi A. The reporting requirements of case reports and adherence of case report reporting guidelines in medical journals: an analysis of the authors' guide sections. Journal of Medical Case Reports. 2023;17. doi:10.1186/s13256-022-03710-2 read 9 October 2026. doi.org/10.1186/s13256-022-03710-2
- Novoa NM. Editorial: Case-reports, a compendium of useful ideas for our daily activity. Frontiers in Surgery. 2022;9:1026401. doi:10.3389/fsurg.2022.1026401 read 9 October 2026. doi.org/10.3389/fsurg.2022.1026401
- Krishna A, Singh PP, Vardhan H, Kumar O. Better outcome of coronavirus disease 2019 infection in kidney transplant recipients: a case series from Eastern India. Saudi Journal of Kidney Diseases and Transplantation. 2021;32(3):875-879. doi:10.4103/1319-2442.336787 read 9 October 2026. doi.org/10.4103/1319-2442.336787
- Wu J, Xiong L, Li P. Coronavirus disease 2019 (COVID-19) in a Chinese renal transplant recipient: a case report. Translational Andrology and Urology. 2021;10(5):2140-2148. doi:10.21037/tau-21-314 read 9 October 2026. doi.org/10.21037/tau-21-314
- Aquilina J, Neves JB, Tran MGB. An overview of study designs. British Journal of Hospital Medicine. 2020. doi:10.12968/hmed.2020.0127 read 9 October 2026. doi.org/10.12968/hmed.2020.0127
- Case Reports and Case Series (Institutional Review Board guidance). 2020. read 9 October 2026. womans.org
- Arpali E, Akyollu B, Yelken B, Tekin S, Turkmen A, Kocak B. Case report: A kidney transplant patient with mild COVID-19. Transplant Infectious Disease. 2020;22(4):e13296. doi:10.1111/tid.13296 read 9 October 2026. doi.org/10.1111/tid.13296
- Li J, Chen G, Zhang M, Tu S, Chen C. Different clinical presentations of two renal transplant recipients with coronavirus disease 2019: a case report. BMC Infectious Diseases. 2020;20(1). doi:10.1186/s12879-020-05434-4 read 9 October 2026. doi.org/10.1186/s12879-020-05434-4
- Sampayo-Cordero M, et al. Agreement between results of meta-analyses from case reports and clinical studies, regarding efficacy and safety of laronidase therapy in patients with mucopolysaccharidosis type I. Molecular Genetics and Metabolism. 2018;123(2). doi:10.1016/j.ymgme.2018.01.002 read 9 October 2026. doi.org/10.1016/j.ymgme.2018.01.002
- Riley DS, Barber MS, Kienle GS, Aronson JK, von Schoen-Angerer T, Tugwell P, et al. CARE guidelines for case reports: explanation and elaboration document. Journal of Clinical Epidemiology. 2017;89:218-235. doi:10.1016/j.jclinepi.2017.04.026 read 9 October 2026. doi.org/10.1016/j.jclinepi.2017.04.026
- Sayre JW, Toklu HZ, Ye F, Mazza J, Yale S. Case series: from clinical practice to evidence-based medicine in graduate medical education. Cureus. 2017:e1546. doi:10.7759/cureus.1546 read 9 October 2026. doi.org/10.7759/cureus.1546
- Guidelines for writing a clinical case report. Heart Views. 2017;18(3). doi:10.4103/1995-705X.217857 read 9 October 2026. doi.org/10.4103/1995-705X.217857
- Murad MH, Asi N, Alsawas M, Alahdab F. New evidence pyramid. Evidence-Based Medicine. 2016;21:125-127. doi:10.1136/ebmed-2016-110401 read 9 October 2026. doi.org/10.1136/ebmed-2016-110401
- Paz-Pacheco E. The case for case reports: valuable resource in ASEAN. Journal of the ASEAN Federation of Endocrine Societies. 2015;30(1):5-6. doi:10.15605/jafes.030.01.12 read 9 October 2026. doi.org/10.15605/jafes.030.01.12
- Ranganathan P, Bavdekar SB, Pramesh CS. Writing case reports for ecancer. ecancermedicalscience. 2015;9:ed49. doi:10.3332/ecancer.2015.ed49 read 9 October 2026. doi.org/10.3332/ecancer.2015.ed49
- Gagnier JJ, Kienle G, Altman DG, Moher D, Sox H, Riley D. The CARE guidelines: consensus-based clinical case report guideline development. Journal of Clinical Epidemiology. 2014;67(1):46-51. doi:10.1016/j.jclinepi.2013.08.003 read 9 October 2026. doi.org/10.1016/j.jclinepi.2013.08.003
- Nissen T, Wynn R. The clinical case report: a review of its merits and limitations. BMC Research Notes. 2014;7:264. doi:10.1186/1756-0500-7-264 read 9 October 2026. doi.org/10.1186/1756-0500-7-264
- Nissen T, Wynn R. The history of the case report: a selective review. JRSM Open. 2014;5(4):2054270414523410. doi:10.1177/2054270414523410 read 9 October 2026. doi.org/10.1177/2054270414523410
- UHN REB Guidance on Case Reports. 2014. read 9 October 2026. uhnresearch.ca
- Buscemi D, Cevik C. Case Reports: Yes or No. The Southwest Respiratory and Critical Care Chronicles. 2013;1(3):1-3. doi:10.12746/swrccc2013.0103.025 read 9 October 2026. doi.org/10.12746/swrccc2013.0103.025
- Rison RA, Kidd MR, Koch CA. The CARE (CAse REport) guidelines and the standardization of case reports. Journal of Medical Case Reports. 2013;7:261. read 9 October 2026. ecancer.org
- Gagnier JJ, Kienle G, Altman DG, Moher D, Sox H, Riley D; the CARE Group. The CARE guidelines: consensus-based clinical case reporting guideline development. BMJ Case Reports. 2013:bcr-2013-201554. doi:10.1136/bcr-2013-201554 read 9 October 2026. doi.org/10.1136/bcr-2013-201554
- Abu-Zidan FM, Abbas AK, Hefny AF. Clinical "case series": a concept analysis. African Health Sciences. 2012;12(4):557-562. doi:10.4314/ahs.v12i4.25 read 9 October 2026. doi.org/10.4314/ahs.v12i4.25
- Carleton HA, Webb ML. The case report in context. Yale Journal of Biology and Medicine. 2012;85(1):93-96. read 9 October 2026. ecancer.org
- Nissen T, Wynn R. The recent history of the clinical case report: a narrative review. JRSM Short Reports. 2012;3:87. read 9 October 2026. cytojournal.com
- Parente RCM, De Oliveira MA, Celeste RK. Case reports and case series in the era of evidence-based medicine. Brazilian Journal of Videoendoscopic Surgery. 2010;3(2):63-66. read 9 October 2026. asean-endocrinejournal.org
- Carey JC. The importance of case reports in advancing scientific knowledge of rare diseases. Advances in Experimental Medicine and Biology. 2010;686:77-86. doi:10.1007/978-90-481-9485-8_5 read 9 October 2026. doi.org/10.1007/978-90-481-9485-8_5
- Albrecht J, Werth VP, Bigby M. The role of case reports in evidence-based practice, with suggestions for improving their reporting. Journal of the American Academy of Dermatology. 2009;60:412-418. read 9 October 2026. asean-endocrinejournal.org
- Boslaugh S (editor). Case reports and case series (encyclopedia entry). Encyclopedia of Epidemiology (SAGE Publications). 2008:135-136. doi:10.4135/9781412953948.n60 read 9 October 2026. doi.org/10.4135/9781412953948.n60
- Neely JG, Karni RJ, Nussenbaum B, Paniello RC, Fraley PL, Wang EW, et al. Practical guide to understanding the value of case reports. Otolaryngology-Head and Neck Surgery. 2008;138:261-264. read 9 October 2026. pulmonarychronicles.com
- Albrecht J, Meves A, Bigby M. Case reports and case series from Lancet had significant impact on medical literature. Journal of Clinical Epidemiology. 2005;58:1227-1232. doi:10.1016/j.jclinepi.2005.04.003 read 9 October 2026. doi.org/10.1016/j.jclinepi.2005.04.003
- Patsopoulos NA, Analatos AA, Ioannidis JPA. Relative citation impact of various study designs in the health sciences. JAMA. 2005;293(19):2362-2366. read 9 October 2026. ecancer.org
- Bernhard JD. A case of too many case reports. Journal of the American Academy of Dermatology. 2001;45:773-774. read 9 October 2026. asean-endocrinejournal.org
- Vandenbroucke JP. In defense of case reports and case series. Annals of Internal Medicine. 2001;134(4):330-334. doi:10.7326/0003-4819-134-4-200102200-00017 read 9 October 2026. doi.org/10.7326/0003-4819-134-4-200102200-00017
- Article type definitions (Dove Medical Press). read 9 October 2026. files.taylorandfrancis.com
- Guidelines for Authors (article type classification by patient number). AME Case Reports. read 9 October 2026. acr.amegroups.org
- OCEBM Levels of Evidence. read 9 October 2026. asmapp.mssm.edu
Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.
How to cite this article
Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.
Pubrica Academy. (2026, October 9). Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting. Pubrica. https://pubrica.com/academy/case-study-or-series/how-many-patients-does-case-series-should-have-in-comparison-to-case-reports/
Pubrica Academy. “Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting.” Pubrica, 9 Oct. 2026, pubrica.com/academy/case-study-or-series/how-many-patients-does-case-series-should-have-in-comparison-to-case-reports/.
Pubrica Academy (2026) Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting. Pubrica. Available at: https://pubrica.com/academy/case-study-or-series/how-many-patients-does-case-series-should-have-in-comparison-to-case-reports/ (Accessed: 15 October 2026).
Pubrica Academy. “Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting.” Pubrica. October 9, 2026. https://pubrica.com/academy/case-study-or-series/how-many-patients-does-case-series-should-have-in-comparison-to-case-reports/
Pubrica Academy, ‘Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting’ (Pubrica, 9 October 2026) <https://pubrica.com/academy/case-study-or-series/how-many-patients-does-case-series-should-have-in-comparison-to-case-reports/> accessed 15 October 2026.
Pubrica Academy. Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting. [Internet]. Pubrica; 2026 Oct 9 [cited 2026 Oct 15]. Available from: https://pubrica.com/academy/case-study-or-series/how-many-patients-does-case-series-should-have-in-comparison-to-case-reports/
Pubrica Academy, “Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting,” Pubrica, Oct. 9, 2026. [Online]. Available: https://pubrica.com/academy/case-study-or-series/how-many-patients-does-case-series-should-have-in-comparison-to-case-reports/
@misc{how_many_patients_does_case_series_sho_2026,
author = {{Pubrica Academy}},
title = {Patient Numbers in Case Series and Case Reports: Thresholds, Definitions and Reporting},
howpublished = {Pubrica},
year = {2026},
month = {oct},
url = {https://pubrica.com/academy/case-study-or-series/how-many-patients-does-case-series-should-have-in-comparison-to-case-reports/},
note = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.
Patient-count and article-type decision worksheet
The first section should be completed before any data are collected, as it is the only part subject to a deadline; the remainder can be completed during drafting. The worksheet exists because the two determinations, ethics and article type, are made by different bodies using different numbers.
PATIENT COUNT AND ARTICLE TYPE DECISION WORKSHEET
PART A — BEFORE YOU COLLECT ANYTHING (ethics determination)
Number of patients you expect to include: ______
Your ethics committee / IRB / REB name: ________________________
Document consulted (SOP, submission guidance, policy number):
____________________________________________________
The patient-number threshold it states: ______
[ ] At or below the threshold — review not required
[ ] Above the threshold — review required
[ ] Document states no threshold — query sent on (date) ______
Written determination obtained? [ ] yes, reference ____________
[ ] no — DO NOT PROCEED
Note: thresholds in use elsewhere range from 4 (research ethics
board: four or more = research) to 6 (hospital IRB: five or fewer
exempt). Your committee's number is the only one that governs you.
PART B — CONSENT (independent of patient count)
Each patient consented individually for publication?
Patient 1 [ ] 2 [ ] 3 [ ] 4 [ ] 5 [ ] 6 [ ] 7 [ ] 8 [ ] 9 [ ]
Journal's own consent form used where required? [ ] yes [ ] n/a
Identifiability reviewed for the COMBINATION of details, not just
names and dates? [ ] yes
Images, rare diagnoses, exact dates and small-centre details checked
for re-identification risk? [ ] yes
PART C — ARTICLE TYPE (journal determination)
Target journal: ______________________________
Article-type page consulted (URL or section): ___________________
Bands the journal specifies:
Case report: ______ Case series: ______ Other: ______
Your patient count falls in: ______________________
Word limit for that type: ______ Reference limit: ______
If the journal specifies NO bands, fallback defaults from the one
argued proposal in the literature:
[ ] 4 or fewer -> report patients individually as case reports
[ ] 5 to 10 -> case series
[ ] more than 10 -> reconsider as an observational study
PART D — THE DESIGN, INDEPENDENTLY OF THE COUNT
Patients were: [ ] consecutive / complete ascertainment
[ ] selected — state the basis: ______________
Source of cases (database, register, clinic list): ______________
Period covered: from __________ to __________
Exclusions applied, and how many excluded: ______________
Source-population denominator, if obtainable: ______
(e.g. total patients under follow-up at the unit in the period)
[ ] I can state a population and a selection rule
[ ] I cannot — and I will avoid language implying a cohort
PART E — REPORTING FORM
[ ] Patient count appears in the FIRST clause of the methods
[ ] Counts, not percentages, for categorical outcomes below ~10
[ ] Means given only where the count behind them is stated
[ ] No comparative statistics, and that is stated explicitly
[ ] Ethics determination names the committee AND its threshold
[ ] Denominator reported, or its absence acknowledged
PART F — METHODS SENTENCE (draft it here)
"We report ______ patients with ______________, comprising
______________________________ identified from ______________
between __________ and __________. We describe this as a
______________ on the basis of ______________________, which
______________________; the project was ______________ ,
reference ______. Written consent for publication was obtained
from each patient individually." Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.
How Pubrica can help
Pubrica addresses the two determinations separately, because they are made by different bodies using different numbers. Pubrica reviews the article-type definitions of the target journal to identify the band into which the patient count falls and the associated word limit, and checks the ethics statement against the threshold that the committee publishes rather than against a general assumption. In the manuscript, Pubrica revises the methods so that the count, the selection rule and the denominator appear in the first few sentences, and converts percentages back to counts wherever the sample is too small to support them.