Guidelines to Authors in Writing an Effective Medical Case Report (MCR)
A case report is rarely rejected because the case was not interesting enough. It is rejected because consent was not obtained in the form the journal requires, because the reporting checklist was not followed, or because the single most informative element — a timeline — was never drawn. Those are requirements rather than matters of style, and they can be satisfied before a word of the discussion is written.
- Which reporting checklist item belongs in which section of the manuscript
- What journals require on patient consent, and why it cannot be arranged afterwards
- How to construct the timeline and the patient perspective, the two items most often omitted
- “Follow the target journal’s author instructions and you have covered the guidelines.” Journal instructions deal with formatting, word counts and submission mechanics. They are not a reporting guideline, and a manuscript can satisfy every one of them and still omit the items a reviewer will look for.
What this article covers
A case report is a structured account of one patient, or a small number of patients, written because something about the case is worth other clinicians knowing: an unexpected association, an adverse effect, an atypical presentation of a common condition, or a response to a treatment that was not predicted.
Its value lies in a particular capability. As Nissen and Wynn put it, “the major advantage of case reporting is probably its ability to detect novelties”,3 which is why “case reporting is thus indispensable in drug safety surveillance (pharmacovigilance)”.3 For conditions where a trial is not possible — “for rare disorders randomized controlled trials (RCTs) can be impossible to run” — case reports and series “might be the only evidence available for guiding clinical practice”.3
Its limits are equally specific, and the discussion section of every case report has to acknowledge them. “Causality cannot be inferred from an uncontrolled observation”, and “findings from case reports cannot be generalized”.3 That is why “the case report got a low ranking on the evidence hierarchy”,3 and why the failure mode to guard against is “the tendency or temptation to generalize when there is no justification for it”.3
This article is about the requirements such a report has to meet. It takes the reporting checklist and maps it onto the parts of a manuscript, and it covers the two things that most often stop a case report being published, consent and de-identification, in the detail they need rather than as a line in a list of components.
Three companion articles cover the neighbouring ground, and this one deliberately does not repeat them. How the reporting guideline was developed, and the consensus process behind it, is covered in CARE guidelines for clinical case reporting recommendations. Choosing a case worth reporting and drafting the components is covered in how to write a case report. Turning the finished report into a visual summary is covered in graphical abstracts for case reports.
What none of them covers is consent, which is the subject of two sections below.
The reporting standard, and how to use it
Case reports have a reporting guideline of their own, in the same family as the guidelines that govern trials and systematic reviews. It is called CARE, and it was published in 2013 after a consensus process that “involved 27 participants and resulted in a 13-item checklist”.1 The stated aim was to “develop, disseminate, and implement systematic reporting guidelines for case reports”.1
The problem it was written to solve was not that case reports were uninteresting. It was that “the usefulness of case reports has been limited by inconsistent and incomplete reporting”,2 and that reports “written without guidance from reporting standards are insufficiently rigorous to guide clinical practice or to inform clinical study design”.1 Nissen and Wynn, reviewing the form two years later, described CARE as “an ambitious endeavor to improve completeness and transparency of reports”.3 The claim behind it is modest and testable: “high-quality case reports are more likely when authors follow reporting guidelines”.2
The checklist has thirteen numbered items — title, key words, abstract, introduction, patient information, clinical findings, timeline, diagnostic assessment, therapeutic intervention, follow-up and outcomes, discussion, patient perspective, and informed consent1 — expanded in the published checklist into thirty-three lettered sub-items.4
This distinction decides whether a manuscript is complete, and it is routinely missed. A journal’s author instructions govern format: word limits, file types, reference style, how many figures, where to upload what. A reporting guideline governs content: which pieces of information the report must contain for a reader to evaluate it.
A manuscript can comply perfectly with every instruction a journal publishes and still omit the timeline, the patient perspective and half the diagnostic reasoning. The instructions never asked for them, because that is not what instructions are for.
The practical way to use the checklist is as a drafting tool rather than a final audit. Taken at the end, it produces a list of things to retrofit, and retrofitted items read like retrofitted items. Taken at the start, it tells you what to collect from the record before you begin writing, which is when the dates, the doses and the follow-up intervals are still recoverable.
Figure 1 maps the checklist topics onto the parts of a manuscript. The mapping is the part worth internalising: the items are not in one-to-one correspondence with sections, and six of them live inside the case presentation alone.
One practical note on citing the guideline. The CARE development paper was published simultaneously in more than one journal, so the same work appears in the literature under different volume and page numbers; the EQUATOR Network record cites the BMJ Case Reports version while the full text used throughout this article is the Journal of Medical Case Reports version.1 Cite whichever version you have actually read.
How the guideline was developed, and the consensus process behind it, is covered separately in the companion article on the CARE guidelines. This article is about applying it.
Figures: applying the CARE checklist
Figure 1 maps each CARE checklist item to the section of the case report where it belongs. Figure 2 shows what a case report timeline looks like. Authors can use them to confirm that the case presentation is complete and to prepare the timeline figure that many journals now expect.
The checklist is not a list of sections. Six of its topics sit inside the case presentation alone, which is why that section is where most reports lose marks.
What to look for. The left column is what authors think about; the right column is what the report is assessed against. The case presentation carries six distinct topics, so treating it as one block of narrative is the commonest route to an incomplete report. The three shaded items — the timeline, the patient perspective and informed consent — are the ones with no natural home in the traditional structure, which is exactly why they get left out.
Schematic mapping of checklist topics to manuscript sections; the checklist’s own numbering and sub-items are not reproduced here.
The item most often skipped, because authors have not seen one. It is a figure, not a paragraph, and it carries information the narrative cannot.
What to look for. Two things the narrative hides. The first is the interval between presentation and diagnosis, which a reader can measure here and only infer from prose. The second is the empty stretch: a timeline shows where follow-up was not recorded, which is information in its own right. Drawing one before writing the case presentation also tends to expose dates the notes do not actually support.
Illustrative example constructed for this figure. It describes no real patient and reports no clinical data.
What adherence actually looks like
It is worth knowing which items authors really miss, because the answer is consistent and it is not the items most people worry about. Two studies measured it directly, from opposite ends of the process.
The first examined published work: 1,178 case reports published in PubMed-indexed Indian journals during 2015, scored against the 2013 checklist. Mean adherence was 61.2% (SD 9.2).5 The aggregate is unremarkable. The item-level figures are not.
The second examined manuscripts before any editorial pressure had been applied: 139 case reports submitted to AME medical journals, of which “only 2 case reports (1.4%) reported all 30 subitems”.6
| Checklist item | Published, Indian PubMed-indexed journals, 2015 (n=1,178)5 | Submitted to AME journals (n=139)6 |
|---|---|---|
| Informed consent | 2.8% reported | Reported in all eligible reports |
| Timeline | 4.6% reported | 66.2% unsatisfactory |
| Patient perspective | 0% reported | 98.6% unsatisfactory |
Two things follow from reading the columns together. Consent is the item that responds to being asked for: near-absent where nothing checks it, universal where a journal will not send a manuscript out without it. It is an administrative requirement, and administrative requirements are met when they are enforced.
The timeline and the patient perspective behave differently. They are weak in both columns, which means submission-stage enforcement does not fix them. They are the two items that have no natural home in the structure authors already know, so they get left out at the point of writing rather than at the point of submission. That is why they have sections of their own below.
Consent: the requirement that stops publication
Of everything on this page, this is the item to settle first, because it is the only one that cannot be fixed later. Every other deficiency in a case report can be remedied by more work. If the patient cannot be reached, or declines, the report cannot be published at all, however good the case and however complete the manuscript.
The requirement itself is not in doubt. Nissen and Wynn state the position plainly: “journals today require written informed consent from patients before publishing case reports”.3 The ICMJE Recommendations set out what that means in practice: identifying information “should not be published in written descriptions, photographs, or pedigrees unless the information is essential for scientific purposes and the patient (or parent or guardian) gives written informed consent for publication”.7
Consent for publication is not a signature on a form describing a report in the abstract. The ICMJE is explicit: “informed consent for this purpose requires that an identifiable patient be shown the manuscript to be published”.7
The BMJ consent form puts the same requirement from the patient’s side. The person signing confirms that they “have read the article to be submitted to BMJ”, and that they understand the article “may be published in a journal which is distributed worldwide” and that “complete anonymity cannot be guaranteed”.9
The CARE development paper draws the same line from the author’s side: authors “have an ethical duty to obtain informed consent from the patient”, and “consent becomes informed when the patient or a relative reads the case report and approves its contents”.1
A consent obtained without showing the patient the manuscript does not satisfy this, which is a second reason the conversation cannot sensibly be left until after the writing.
Journals enforce it at the gate rather than at acceptance. BMJ Case Reports states that authors “must have signed informed consent from patients (or relatives/guardians) before submitting”, that “for living patients this is a legal requirement under the UK’s Data Protection legislation”, and that “we will not send your article for review without explicit consent from the patient or guardian”.8 A manuscript without consent is not one that gets reviewed and is then asked for paperwork. It is one that does not enter review.
Three things go wrong with deferring it. The patient may be untraceable by the time the manuscript is finished, which is common where the episode was months earlier or care has transferred. The patient may decline, and Nissen and Wynn note this is a real possibility rather than a formality: “an adequately informed patient might not consent to publication”.3 And the journal may require consent on its own form, which means a consent already obtained in some other wording may not satisfy it.
The cost of asking first is a conversation. The cost of asking last is the whole manuscript.
Consent for publication is also a different thing from consent to treatment, and from the ethics approval an institution may separately require. A patient who agreed to a procedure has not thereby agreed that an account of it will appear in a journal that anyone can read. The conversation has to cover what will be published, where, and that it will be permanent and public.
Consent is revocable, and that has happened
The BMJ form records that the patient “can revoke my consent at any time before publication”.9 The consequences of a withdrawal after publication are not hypothetical. In one documented instance, a case report in the Journal of Medical Case Reports describing an infant with a rare combination of birth defects was removed after the patient’s legal guardian withdrew consent; the retraction notice records that the journal was “informed that they withdrew this consent shortly after publication”, and that “the article is no longer available online in order to protect patient confidentiality”.12
The lesson is not that consent is fragile but that it is a relationship rather than a signature. A patient who understood what was being published, and saw it, is a patient unlikely to withdraw.
Deceased patients, children, and patients who cannot be found
These are the cases where authors most often assume the requirement lapses, and it does not.
- Deceased patients. COPE’s guidance for journals is explicit that the consent requirement “also applies when a report involves deceased persons”.10 Journal policies follow: one states that “in the instance of a deceased patient, consent from the next of kin is required prior to publication”.13
- Where no relative can be traced. Oxford Medical Case Reports states that for a deceased patient whose relatives cannot be traced, “the case report can only be considered for publication if it is sufficiently anonymised”.11 That is a narrow exception resting entirely on anonymisation, not a waiver.
- Children and patients lacking capacity. Consent comes from the parent or guardian,7 and the CARE explanation document puts it as, for children, “obtain consent from a guardian”.2 The retracted report described above is the case where exactly that consent was later withdrawn.
One practical point follows from all of this: obtain consent in the form the journal you are targeting requires, which means identifying the journal before drafting rather than after.
De-identification, and what you must not do
Consent and anonymity are often treated as the same requirement. They are separate, and satisfying one does not discharge the other: a patient may consent and still be identifiable to colleagues, neighbours or family in ways they did not anticipate. The ICMJE puts the relationship between them the other way round, as a trigger: “informed consent should be obtained if there is any doubt that anonymity can be maintained”.7
The honest position on how far anonymity can be guaranteed is the one Nissen and Wynn state: “a guarantee for full confidentiality is not always possible”.3 COPE says the same to editors, that “the journal cannot guarantee anonymity”.10 That is a reason to be careful about what is included, and a reason for the consent conversation to be explicit rather than reassuring.
What makes a patient identifiable is rarely the name, which everyone removes. The ICMJE names “names, initials, or hospital numbers” as the obvious category and then adds the instruction that does the real work: “nonessential identifying details should be omitted”.7 The detail that identifies a patient is more often one of these.
- Exact dates. A date of admission combined with a named institution can identify one person. The BMJ Case Reports template instructs authors directly that “place names and calendar dates are to be avoided”.8 Intervals such as day 0 and day 9 carry the clinical information without the calendar, which is one reason a timeline is drawn in intervals.
- Images. Facial features are obvious; less obvious are tattoos, jewellery, distinctive anatomy, and identifiers visible in radiographs and scan headers. The partial measure most often reached for does not work: the ICMJE states that “masking the eye region in photographs of patients is inadequate protection of anonymity”.7
- An unusual combination. Age, occupation, ethnicity and a rare condition may each be unremarkable alone and jointly unique. The rarer the case, the smaller the population it could be drawn from, which is precisely the tension in reporting rare conditions. COPE notes that “individuals in reports are highly identifiable” and that “examples of identifying information are descriptions of individual case histories”, meaning the history itself and not merely the labels attached to it.10
- Institutional and geographic detail. Naming a small unit narrows the field sharply, and the same template instruction covers it: “remove all details that identify the patient”.8
The tempting solution to an identifiable patient is to change a detail: shift the age, alter the occupation, move the date. Journal policy addresses this directly. Oxford Medical Case Reports states that “patient data should never be altered or falsified to attain anonymity”.11
The ICMJE frames the same boundary as a test rather than a prohibition: where identifying characteristics are de-identified, “authors should provide assurance, and editors should so note, that such changes do not distort scientific meaning”.7 The two are consistent. Generalising a detail that carries no clinical weight is de-identification. Altering one that does is falsifying the record the report exists to convey, and a reader can no longer tell which details are real.
Where a detail is both identifying and clinically necessary, the options are to omit it and say that it has been omitted, to describe it less precisely, or to obtain consent specific to including it. Inventing a replacement is not among them.
Finally, the review has to be systematic rather than impressionistic, because the number of things that can identify a patient is larger than most authors assume. COPE’s guidance points to a published list of twenty-three potential identifiers.10 Reading the manuscript once more for anything that feels identifying is not the same as checking it against a list.
The manuscript, element by element
What follows is what each part of the manuscript has to contain, in the order a reader meets them. The checklist mapping in Figure 1 shows which reporting items each section is carrying.
Title
The checklist is specific here: “the area of focus and ‘case report’ should appear in the title”.4 The second half matters for retrieval as much as for honesty: readers and databases filter on study type, and a title that does not identify the report as a case report will be missed by people looking for exactly that.
Resist the construction that promises more than one patient can deliver. A title asserting that a treatment is effective is making a claim a single case cannot support; a title saying that a treatment was followed by an unexpected outcome is reporting what happened.
Abstract
The abstract carries four things, and the checklist frames two of them as questions to answer rather than headings to fill: “what is unique about this case? what does it add to the medical literature?” at the start, and “what are the main ‘take-away’ lessons from this case?” at the end.4 Between them sit what happened and what the outcome was. Many journals impose a structured format and a word limit, and those are the journal's requirements rather than the guideline's — check them before drafting rather than cutting afterwards.
Introduction
Two or three paragraphs. The checklist asks you to “briefly summarize why this case is unique with medical literature references”,4 and the last three words are the operative ones. This is where a brief literature search earns its place: the claim that a presentation is rare is checkable, and a reviewer will check it.
Case presentation
This is the section that carries six separate reporting topics, and the commonest structural fault in case reports is writing it as a single continuous narrative. The six are patient information, clinical findings, the timeline, diagnostic assessment, therapeutic intervention, and follow-up and outcomes.
| Topic | What belongs here | What is commonly missing |
|---|---|---|
| Patient information | “De-identified demographic information and other patient specific information”,4 the presenting concern, and relevant history | Relevant comorbidities and medication, omitted as background |
| Clinical findings | “The relevant physical examination (PE) and other clinical findings”,4 described rather than interpreted | Negative findings that rule alternatives out |
| Timeline | The sequence and intervals, as a figure | The whole item — see below |
| Diagnostic assessment | Investigations, their results, the reasoning, and “diagnostic challenges (such as access, financial, or cultural)”4 | Tests that were considered and not done, and why |
| Therapeutic intervention | “Administration of intervention (such as dosage, strength, duration)”,4 and any changes | Adherence, and the reason for any change of treatment |
| Follow-up and outcomes | What happened after, including “adverse and unanticipated events”4 | How long follow-up ran, and what was not measured |
Writing these as six labelled passages, even if the journal prints them as continuous prose, is the simplest way to find out which one is thin.
Discussion
The discussion sets the case against the published literature and says what it adds. Three things belong in it that are often absent: the strengths and limitations of this report, the alternative explanations that were considered, and an explicit statement of what the checklist calls “the primary ‘take-away’ lessons of this case report”.4
The limitation paragraph is not a formality. “Causality cannot be inferred from an uncontrolled observation”, and “findings from case reports cannot be generalized”.3 Saying so protects the report from the obvious objection and costs two sentences.
One further asymmetry is worth knowing about when deciding what to report. Journals favour positive findings in this literature as in others: in one review, “only 5% of case reports and 10% of case series reported treatment failures”.3 A case in which a treatment did not work is therefore both underrepresented and, for that reason, more informative than its publication rate suggests.
Conclusion and references
The conclusion should be short and should follow from what was reported, not from what the authors hope is true. References should support the claims about what is already known; a case report with no references has not established that the case is unusual.
The timeline
The timeline is the item authors most often skip, usually because they have not seen one and assume the chronology in the narrative covers it. It does not, and Figure 2 shows why.
The checklist asks for “a timeline of relevant information from the patient’s history and this episode of care”,4 and the development paper is specific about the form: “we recommend a timeline (item 7) in the form of a table or figure”, because “the timeline should show how the key events of the case unfolded”.1 The explanation document describes what that achieves — “a succinct summary of one or more key events in the case, enabling the reader to readily view core elements of the case report”.2
In practice it places events above a time axis and interventions below it, with real intervals rather than sequence words. Two kinds of information appear in it that prose conceals.
The first is duration. “The patient was subsequently diagnosed” tells a reader that a diagnosis followed; a timeline tells them it took nine days, which may be the most clinically interesting fact in the report. The second is absence. A gap on a timeline shows where nothing was recorded, and a reader can see immediately whether a twelve-month outcome rests on continuous follow-up or on a single late contact.
Constructing the timeline first has a diagnostic effect on the manuscript. Dates that the notes do not actually support become visible, because they have to be placed. Intervals that were assumed prove to be unrecorded. Discovering this before the case presentation is drafted is far cheaper than discovering it after a reviewer asks.
The patient perspective
The patient perspective is a distinct checklist item: “when appropriate, the patient can share their perspective on their case”,4 put more strongly in the explanation document as “the patient should share their perspective or experience whenever possible”.2 It is not a substitute for consent, which is covered separately above.
It is frequently omitted, and the reasons given are usually practical: the patient is no longer contactable, or the episode was too long ago. Where that is the case, saying so is better than silence, because an absent item and an unexplained absent item read differently to a reviewer.
Where it can be obtained, it does something the rest of the report cannot. A clinical narrative records what was observed and done; the patient perspective records what the episode was like, which sometimes contains the detail that makes the case instructive, such as a symptom that was not asked about, or a reason for a gap in treatment that the notes record only as non-adherence. The explanation document makes the stronger claim that “the report of a truly novel treatment may rely heavily on a patient’s perspective”.2
The practical instruction is modest: “provide patients with an opportunity to briefly share their perspectives on the episode of care”.2 It is an opportunity offered, not an interview conducted.
Before you submit
Four things are worth settling before a manuscript goes anywhere, because each of them is expensive to fix later.
Consent, in the form the journal requires. This is first because it is the one that cannot be remedied if the patient cannot be reached. See the two sections above.
The CARE checklist, completed. Many journals ask for it at submission. Completing it is also the fastest way to find the items you have missed, since each line requires you to name a page. The evidence above is the argument for doing it: of 139 manuscripts submitted to one journal group, only two reported every sub-item.6
The timeline, drawn. As a figure, with the manuscript referring to it.
Institutional requirements, checked. Whether a case report needs ethics committee review varies by institution and jurisdiction, and the answer is not inferable from the journal's rules. Ask before submission rather than discovering at acceptance that a form was required.
Common mistakes
| Mistake | Stage | What it costs | What to do instead |
|---|---|---|---|
| Writing first, seeking consent afterwards | Before drafting | If the patient cannot be reached or declines, the work cannot be published at all | Obtain consent in the journal’s required form before investing in the manuscript |
| Treating journal author instructions as the reporting guideline | Drafting | Formatting is satisfied while reporting items are missing | Work from the reporting checklist, then apply the journal’s format |
| Writing the case presentation as one narrative | Drafting | Six distinct reporting topics get compressed, and the thin ones are invisible | Draft the six as labelled passages, then join them if the journal prefers prose |
| Omitting the timeline | Drafting | Intervals and gaps in follow-up stay hidden from the reader | Draw it as a figure before the case presentation is written |
| Omitting the patient perspective with no explanation | Drafting | A reviewer cannot tell whether it was impossible or merely skipped | Include it, or state why it could not be obtained |
| Claiming more than one case can support | Discussion | The central claim exceeds the design, which is the easiest objection to make | Report what happened, state the limitations, and leave causation alone |
| Assuming removing the name is enough | Throughout | A patient may remain identifiable from dates, images or an unusual combination of details | Review the whole manuscript, including figures, for identifying material |
Two things decide whether a case report gets published, and neither is the case itself. The first is consent, obtained in the form the journal requires and obtained before the manuscript is written, because it is the one requirement that cannot be remedied if the patient cannot be reached. The second is the CARE checklist, which is not the same thing as the journal's author instructions: six of its thirteen items sit inside the case presentation alone, and three of them, the timeline, the patient perspective and consent, have no natural home in the traditional structure. Those three are measurably the ones that go missing: in one sample of 1,178 published case reports, consent was reported in 2.8%, the timeline in 4.6% and the patient perspective in none at all.5 Draw the timeline before you draft the case presentation. It will show you which dates your notes do not actually support.
How to apply this to your work
- Check the target journal’s consent requirement first, and obtain consent in that form before drafting.
- Confirm whether your institution requires ethics review for a case report, since this does not follow from the journal’s rules.
- Draw the timeline as a figure, and let it expose the dates the record does not support.
- Draft the case presentation as six labelled passages, one per reporting topic, before joining them into prose.
- Obtain the patient perspective, or record why it could not be obtained.
- Complete the CARE checklist, naming a page for each item, and treat any blank as a drafting task.
- Review the whole manuscript for identifying material, including figures, dates and unusual detail combinations.
- State the limitations explicitly, including that a single case cannot establish causation or frequency.
Common questions
What is a medical case report?
A structured account of one patient, or a small number of patients, written because something about the case is worth other clinicians knowing — an unexpected association, an adverse effect, an atypical presentation, or an unanticipated response to treatment. It is a report of what happened, not a study of whether something works.
Do I need the patient’s consent to publish a case report?
Yes, and it should be obtained before the manuscript is written rather than afterwards. It is the single requirement that cannot be remedied later: if the patient cannot be reached or declines, the report cannot be published, whatever has already been invested in it. The ICMJE goes further than a signature and requires that “an identifiable patient be shown the manuscript to be published”.7 Journals differ in the form they require, so check the target journal first.
Is removing the patient’s name enough to anonymise a case report?
No. A patient can remain identifiable from exact dates, from images, from the institution named, or from an unusual combination of clinical details. Masking the eye region in a photograph is specifically named by the ICMJE as “inadequate protection of anonymity”.7 What you must not do is alter or invent details to disguise a patient: “patient data should never be altered or falsified to attain anonymity”.11
Are the journal’s author instructions the same as the CARE guidelines?
No, and conflating them is a common route to an incomplete report. Author instructions cover formatting, word limits, file types and submission mechanics. CARE covers what information the report must contain: thirteen items, expanded into thirty-three sub-items.4 A manuscript can comply fully with the first and still omit items a reviewer will look for.
What is a case report timeline and do I really need one?
It is a figure placing events above a time axis and interventions below it, with real intervals. It is CARE item 7, and it carries information prose does not: how long a diagnosis took, and where follow-up was not recorded. Drawing it before the case presentation also exposes dates that the clinical record does not support. It is also one of the two weakest items in practice, judged unsatisfactory in 66.2% of one sample of submitted manuscripts.6 Figure 2 shows an example.
What is the patient perspective section?
A short account, in the patient’s own words where possible, of their experience of the episode. It is CARE item 12, and it is not a substitute for consent. It is the single most-omitted item: unreported in all 1,178 published reports in one sample,5 and unsatisfactory in 98.6% of submitted manuscripts in another.6 Where it cannot be obtained, because the patient is no longer contactable or the episode was long ago, say so: an unexplained omission and an explained one read differently.
How long should a case report be?
That is a journal question rather than a guideline question, and the limits vary widely, so take the figure from the target journal before drafting rather than cutting afterwards. What the length should not do is squeeze out the reporting items: a short word limit is a reason to write the discussion tightly, not to drop the timeline.
Can a case report prove that a treatment caused an outcome?
No. “Causality cannot be inferred from an uncontrolled observation”, and “findings from case reports cannot be generalized”.3 A case report can show that an association is possible and worth investigating, which is a real contribution, and it remains “indispensable in drug safety surveillance”.3 Stating the limitation explicitly in the discussion removes the most obvious objection to the paper.
Declared interest: Pubrica provides medical writing and publication support commercially. Nothing on this page should be read as suggesting that a service can obtain consent on an author’s behalf or make an identifiable case report publishable. Consent is between the clinician and the patient, and where it cannot be obtained the report cannot be published.
References and sources
- Gagnier JJ, Kienle G, Altman DG, Moher D, Sox H, Riley D. The CARE guidelines: consensus-based clinical case reporting guideline development. Journal of Medical Case Reports. 2013;7:223. doi:10.1186/1752-1947-7-223 read 20 October 2026. The same guideline was published simultaneously in more than one journal; this is the version used here.View the full text
- 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 20 October 2026 as the accepted manuscript deposited in the Oxford University Research Archive.View the repository record
- 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 20 October 2026.View the full text
- CARE Group. The CARE checklist. care-statement.org. Thirteen numbered items, expanded into thirty-three lettered sub-items in the published checklist. read 20 October 2026.View the checklist
- Ravi R, Mulkalwar A, Thatte UM, Gogtay NJ. Medical case reports published in PubMed-indexed Indian journals in 2015: adherence to 2013 CARE guidelines. Indian Journal of Medical Ethics. Published online 23 April 2018. doi:10.20529/IJME.2018.036 read 20 October 2026.View the full text
- Yang F, Cheng Y, Zhang K. Adherence of unpublished case reports to the Case Report (CARE) guidelines: a retrospective cross-sectional analysis of 139 case report manuscripts initially submitted to AME medical journals. AME Medical Journal. 2022;7:36. doi:10.21037/amj-22-94 read 20 October 2026.View the full text
- International Committee of Medical Journal Editors. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals: Protection of Research Participants. read 20 October 2026.View the recommendation
- BMJ Case Reports. Standard template for case reports, April 2022, and How to submit a BMJ Case Report. read 20 October 2026. The template was read from an institutional copy; the submission page from BMJ.View the submission guidance
- BMJ. Patient consent form, July 2019. read 20 October 2026.View the consent form
- Barbour V, on behalf of COPE Council. Journals’ best practices for ensuring consent for publishing medical case reports: guidance from COPE. Committee on Publication Ethics, version 1, December 2016. read 20 October 2026 from a journal-hosted copy of the COPE document.View COPE’s related case guidance
- Oxford Medical Case Reports. Journal policies, including patient consent and anonymity. Oxford University Press. read 20 October 2026.View the policies
- Scudellari M. Chinese medical case study erased after guardian consent withdrawn. Retraction Watch, 28 April 2015. Reporting the retraction notice of a case report in the Journal of Medical Case Reports. read 20 October 2026.Read the report
- European Medical Journal. Obtaining patient consent to publish a case report in EMJ. Journal guidance document, no author or date printed. read 20 October 2026.View the guidance
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.
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.
Work through these before drafting, and again before submission.
[ ] Target journal identified and its consent requirement read [ ] Patient consent obtained in that journal's required form [ ] Consent obtained before the manuscript was drafted [ ] Institutional position on ethics review for case reports confirmed [ ] Timeline drawn as a figure, with real intervals [ ] Dates on the timeline supported by the clinical record [ ] Case presentation drafted as six labelled reporting topics [ ] Negative findings and ruled-out alternatives included [ ] Treatment doses, duration and any changes recorded [ ] Length of follow-up stated, including what was not measured [ ] Patient perspective obtained, or its absence explained [ ] Reporting checklist completed with a page number against each item [ ] Whole manuscript reviewed for identifying material, figures included [ ] No detail altered or invented in order to anonymise [ ] Limitations stated, including that one case cannot establish causation
We help you structure a case report against the reporting checklist, prepare the timeline and patient perspective, and make sure the consent and de-identification requirements are met before submission.