Diagnosing diagnostic error of endometriosis: a secondary analysis of patient experiences from a mixed-methods survey

To analyse endometriosis diagnostic errors made by clinicians as reported by patients with endometriosis.

This study deductively analysed qualitative data as part of a larger mixed-methods research study examining 'invalidating communication' by clinicians concerning patients' symptoms. Data analysed were responses to an open-ended prompt asking participants to describe an interaction with a clinician prior to their diagnosis in which they felt their symptoms were dismissed. We used three validated taxonomies for diagnosing diagnostic error (Diagnosis Error Evaluation and Research (DEER), Reliable Diagnosis Challenges (RDC) and generic diagnostic pitfalls taxonomies).

A total of 476 relevant interactions with clinicians were identified from 444 patients to the open-ended prompt, which identified 692 codable units using the DEER taxonomy, 286 codable units using the RDC taxonomy and 602 codable diagnostic pitfalls. Most prevalent subcategories among these three taxonomies were inaccurate/misinterpreted/overlooked critical piece of history data (from DEER Taxonomy; n=291), no specific diagnosis was ever made (from diagnostic pitfalls taxonomy; n=271), and unfamiliar with endometriosis (from RDC Taxonomy; n=144).

Examining a series of patient-described diagnostic errors reported by patients with surgically confirmed endometriosis using three validated taxonomies demonstrates numerous areas for improvement. These findings can help patients, clinicians and healthcare organisations better anticipate errors in endometriosis diagnosis and design and implement education efforts and safety to prevent or mitigate such errors.