Consent and Deidentification of Patient Images in Dermatology Journals: Observational Study

Publication of patient images contributes to research and education in dermatology. However, it is important to protect patients’ privacy and rights. The Committee on Publication Ethics (COPE) and the International Committee of Medical Journal Editors (ICMJE) have provided best practices and recommendations, respectively, for the protection of patients’ rights in scholarly publications [1,2]. Nonetheless, requirements for the deidentification of patient images and for the acquisition of consent to publish such images vary across governing bodies and journals. Our objective was to describe leading dermatology journals’ instructions regarding deidentification and consent to publish patient images as well as the content and readability of consent forms.

This study was exempt from institutional review board review as data were publicly available. Themes regarding the publication and deidentification of patient images, as well as the acquisition of consent, were extracted from COPE and ICMJE [1,2]. On June 9, 2021, the top 20 dermatology journals were determined using Google Scholar, which ranks journals based on the h5-index [3]. Guidelines, instructions for authors, submission checklists, and consent forms on the journals’ websites were reviewed for criteria embodied by the themes extracted from COPE and ICMJE. Legal clauses in consent forms were summarized. Available consent forms were prepared and then assessed for readability using Microsoft Word (Microsoft Corporation), which calculates the Flesch-Kincaid Grade Level (FKGL) [4]. FKGL considers average sentence length and the average number of syllables per word to provide a corresponding US grade level rating [4].

A total of 19 (95%) journals’ online instructions instructed authors to obtain written consent or permission for the publication of patient images (Table 1). The specific instances in which consent was required varied and included recognizable, identifying, identifiable, or possibly identifiable images; images that may, could be used to, could, or potentially identify the person; images in which the person could or can be identified, including by the patient; only if the patient’s face is completely identified; or any or all patient images, regardless of whether a patient is or is not identifiable. Some journals provided specific guidance on identifiable features, such as facial features (n=5), tattoos (n=1), and jewelry (n=1). A total of 11 consent forms were identified from 10 journals (Table 2). All forms emphasized that the individual in a published image may be identified or that anonymity cannot be guaranteed. The average FKGL was 15.3 (range 12.1-22.8).

Instructions regarding the deidentification of patient images and acquisition of consent for publication differed across dermatology journals and incorporated various elements from COPE and ICMJE [1,2]. Most leading dermatology journals instructed authors to obtain written consent or permission to publish patient images. This is in contrast to a study that found that approximately 52% of dental, oral surgery, and otorhinolaryngology journals had a policy regarding clinical images [5]. Although readability scores should be used with caution, consent forms were difficult to read and were written, on average, at a college level based on an FKGL score of 15.3. It has been recommended that materials for patients should be written at the sixth-grade level or lower [6]. Consideration should be given to enhancing consent form readability, which may improve patient understanding. Although we analyzed a small subset of journals from a specific subspecialty, our findings may raise awareness of the need to protect patients’ right to confidentiality by implementing consistent policies for the publication of clinical images.