Of everything that stops a case report from being published, this is the most common and the most avoidable: consent that was never obtained, obtained incorrectly, or obtained properly and then not reported.
It is also the one problem that cannot be fixed during revision. A missing statistical test can be run. A weak discussion can be rewritten. Consent that was not obtained while the patient was still contactable cannot be retrofitted.
Three different things, routinely confused
| What it is | Who grants it | |
|---|---|---|
| Consent to treatment | Permission for the clinical intervention itself | Patient, before treatment |
| Ethics approval | Institutional authorisation to conduct research | IRB / ethics committee |
| Consent for publication | Permission to publish clinical details and images | Patient, specifically for publication |
These are independent. A signed surgical consent form gives you no right to publish. Ethics approval for a broader study does not cover an individual patient's images. Many institutions do not require ethics review for a single case report at all — and that changes nothing about the consent requirement.
The standard comes from the ICMJE Recommendations and sits on the Declaration of Helsinki: identifying information should not be published without informed consent, and where identifying material is essential to the scientific message, the patient must be told and must agree.
What counts as identifying
Far more than most authors assume. Removing the name is where anonymisation starts, not where it ends.
- Direct identifiers — name, initials, hospital or medical record number, date of birth, address, contact details.
- Dates — admission, surgery, discharge. Combined with a named institution, an exact date narrows the population sharply.
- Images of the face, or any part of it. Also the ear, which is close to unique.
- Distinctive marks — tattoos, birthmarks, scars, jewellery, clothing, nail varnish visible at the edge of a clinical photograph.
- Rare conditions. A diagnosis with a handful of reported cases identifies the patient to anyone who knows them.
- Occupational and social detail — "a 34-year-old professional violinist", "the only paediatric liver transplant at our centre in 2024".
- Imaging with embedded data — see below.
- Pedigree charts in genetics, which identify entire families.
The working test: could anyone who already knows this person — a relative, a colleague, a neighbour — recognise them from what you have published? Not a stranger. Someone who knows them.
The eye bar does not work
The black rectangle across the eyes fails on both counts. It does not reliably prevent recognition, particularly by people who know the patient. And it signals to readers that the patient may not have consented, which is precisely the impression you do not want attached to your work.
Modern practice is straightforward: if the image is identifiable and clinically necessary, obtain explicit consent and publish it. If it is not necessary, crop tightly to the lesion or region of interest so the identifying content is gone entirely. Do not attempt a middle path in which the image is partly obscured and consent is treated as partly required.
Strip the image metadata
This one catches careful authors. DICOM files carry patient name, ID, date of birth, study date and institution in their headers. Photographs from phones and cameras carry EXIF data including timestamps and, frequently, GPS coordinates.
A visually anonymous image can therefore still contain the patient's full name in its file. Export DICOM images as de-identified files or as flattened JPEG/PNG, strip EXIF from photographs, and check the exported file rather than assuming. Also check what is burned into the pixels themselves — many ultrasound and radiograph exports display the patient banner as part of the image.
Obtaining consent properly
Consent must be informed, which means specific. "May I write this up?" is not informed consent. The patient needs to understand what is being published, where, and that it will be permanent and public.
Six elements to cover:
- What will be published — clinical history, investigations, images, and specifically which images.
- Where — an international, open-access, peer-reviewed medical journal, freely readable by anyone including non-clinicians.
- Permanence — publication is effectively irrevocable; an article cannot be reliably removed from the internet once distributed and indexed.
- Anonymity limits — identifying details will be minimised, but complete anonymity cannot be guaranteed, particularly where images or rare conditions are involved. Say this plainly rather than reassuring.
- Voluntariness — refusal will not affect their care in any way. Where you are the treating clinician there is an inherent power asymmetry, and it should be named.
- The right to see the manuscript. ICMJE guidance is that patients should be shown the manuscript before submission where identifying material is involved. Offer this in every case.
Get it in writing, and get it early. Ideally while the patient is still under your care. Tracing a discharged patient eighteen months later to ask about a paper you have already written is a poor position to be in.
Who signs when the patient cannot
- Children — parent or legal guardian signs. Older children and adolescents should also give assent, and it is good practice to record that separately.
- Adults lacking capacity — legal representative, in line with local law.
- Deceased patients — next of kin. Rules vary by jurisdiction; document who consented and their relationship.
- Untraceable patients — raise it with the editor before submitting. Do not omit the statement and hope.
A consent form you can adapt
Use your institution's form if one exists. If not, this covers the required ground — have it checked against local law and your institution's policy before use.
CONSENT FOR PUBLICATION OF CLINICAL DETAILS AND IMAGES
I, [patient name], give permission for information about my medical condition, treatment and outcome, together with the images identified below, to be published in a medical journal.
I understand that:
• The article will be published in an international, peer-reviewed, open-access medical journal and will be freely available online to anyone, including people who are not medical professionals.
• The article will be permanently available and cannot reliably be withdrawn once published.
• My name and direct identifying details will not be published, but complete anonymity cannot be guaranteed — someone who knows me may still recognise me, particularly from images or from details of my condition.
• I may see the manuscript before it is submitted if I wish.
• I may refuse, or withdraw this consent at any time before publication, and my medical care will not be affected in any way.Images covered by this consent: [list or describe each image]
Patient signature ______________________ Date __________
Name in capitals ______________________If signed by a parent, guardian or legal representative:
Name ______________________ Relationship to patient ______________________Clinician obtaining consent ______________________ Date __________
Manuscript shown to patient before submission: ☐ Yes ☐ Declined
Keep the signed form securely, with the patient record. Do not attach it to the submission unless the journal explicitly requests it through a secure route — a form containing identifiers sent by ordinary email is a data breach in its own right.
Reporting it in the manuscript
Consent obtained but not reported reads exactly like consent never obtained. Include an explicit statement — most journals expect it under a Declarations or Ethics heading:
Written informed consent was obtained from the patient for publication of this case report and the accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.
Where a representative consented, say so:
Written informed consent was obtained from the patient's next of kin for publication of this case report and the accompanying images.
The CARE guideline, which most journals apply to case reports, also asks for the patient's perspective on their treatment and outcome where it can be obtained. It is often the most memorable part of a case report, and it is easy to collect at the same time as consent.
Quick checklist before you submit
- ☐ Written consent obtained specifically for publication
- ☐ Signed form stored securely and retrievable
- ☐ Manuscript offered to the patient before submission
- ☐ Direct identifiers removed from text, tables and figures
- ☐ Exact dates generalised where not clinically essential
- ☐ Images cropped to the clinical region; no incidental identifying features
- ☐ DICOM and EXIF metadata stripped, and the exported file checked
- ☐ No identifying banner burned into the image pixels
- ☐ Consent statement present in the Declarations section
- ☐ Consent-giver's relationship stated if not the patient
Our full position is on the Publication Ethics page. For structuring the report itself, see our guide to writing a case report with the CARE checklist, and for what ethics approval does and does not cover, ethics approval and informed consent.