Do you see an issue with self-diagnosis? There is potential for a patient to self-diagnose incorrectly and may have some other disorder similar to Dupuytren (possibly an unknown autoimmune disorder). Do you worry that the population of self-diagnosed patients in your database who may have incorrectly self-diagnosed would skew analysis results?
Modified from a question by Holly Stephens (Booz Allen Hamilton)
How will patients be identified? Patients are identified either by prior physician diagnosis or by self-diagnosis. Because of familial involvement, self-diagnosed by a relative is not uncommon. Self-diagnosis is an accurate and validated method when patients refer to images [Gormasz 2011, Pervulesco 2011]. Visual references are included in the International Dupuytren Data Bank patient enrollment form.
Misdiagnosis is possible. Flexor tendinitis ("trigger finger"), late effects of injury, ulnar nerve palsy, rheumatoid, or spasticity may superficially resemble Dupuytren. There is also a likelihood of physician misdiagnosis: on average, a person first diagnosed by a physician (rather than by a family member or self-diagnosis) will have seen at least two physicians before being correctly diagnosed. That's actually a worse track record than the above authors reported with self-diagnosis using visual guides.
What are the odds? I ran a Dupuytren specialty practice in which patients traveled to my office for in-office same-day Dupuytren procedures. I requested that prospective patients send deidentified pictures of their hands for me to review before scheduling an appointment. This was to give them feedback regarding expectations and to visually confirm the diagnosis. Based on my experience screening pictures of thousands of patients who submitted hand selfies in this scenario, I would estimate the wrong self-diagnosis rate as well under 1%, probably closer to 0.1%. Having said that, the great majority of patients scheduling appointments were doing so because of obvious finger contractures. Self-diagnosis accuracy is probably worse for those with early-stage disease: that's also true with physician diagnosis.
Incorrect self-diagnosis is unlikely to be an issue with our pilot study, as participants will be selected based on multiple factors: family history of Dupuytren; early age of diagnosis; multiple sites of involvement. We'll start with the distinctive phenotype which is hard to mistake.
I'm glad that you brought this up. It would make sense to update the protocol to have disease cohort people submit a hand picture for visual confirmation. Thanks!
The questions, answers and comments posted under this Expert Q&A forum reflect the views of the author(s) and should not be construed to represent views or policies held by the FDA. Experts answer questions at their discretion and thus not all questions may be addressed. The Expert Q&A forum is also not intended to address questions pertaining to medical advice, diagnosis or treatment.