Parent-provided smartphone photos may be used for the accurate diagnosis of skin conditions in children, researchers reported.
Results from a randomized clinical trial of 40 patient-parent dyads found that overall concordance between photograph-based diagnosis and in-person diagnosis was 83% (95% CI 71-94%, κ=0.81), according to Daniel M. O'Connor, MD, of the Children's Hospital of Philadelphia (CHOP), and colleagues.
"Parents can reliably take high-quality photographs of their child's skin condition using smartphone cameras," the authors wrote in JAMA Dermatology. "This finding suggests that direct-to-patient pediatric teledermatology should not be limited by image quality, especially when appropriate photography instructions are provided."
O'Connor's group prospectively collected data on patient-parent dyads from a pediatric dermatology clinic at CHOP from March 1 to Sept. 30, 2016. Mean age of children was 7 years and just more than half (n=22) were female.
They randomized the patient-parent dyads 1:1 to an intervention group that received a simple, three-step instruction sheet on how to take photos with a smartphone or a control group that didn't receive instructions. Clinicians were blinded to whether parents had received photography instructions.
All parents were asked to use their personal smartphones to take photos of their child's skin condition in the clinic examination room and then upload the photos to the child's electronic health record via MyChart, an application linked to lab results, appointment information, current medications, and immunization history. If the child presented for more than one skin condition, the parent was instructed to choose the primary condition.
Cohen κ was used to quantify concordance between the two groups.
O'Connor's group reported that diagnostic concordance between photograph-based diagnosis and in-person diagnosis was 89% (95% CI 75%-97%, κ=0.88) among the 37 dyads with photographs that were of high enough quality to make a diagnosis. Diagnostic accuracy varied by diagnostic category with the highest concordance seen for birthmarks (100%), rashes (92%), and alopecia-related diagnoses (64%).
Parents who received photographic instructions shared more and higher quality images compared with those didn't (mean photograph quality rating scale score 8.9 versus 9.5, P=0.04), but no statistical difference in diagnostic concordance was observed (85% versus 80%).
Overall, parents rated their willingness to use a teledermatology as opposed to waiting for an in-person appointment as an 8 out of 10 (10=very willing), and 80% were willing to pay a median price of $20 (range $0-$160) to use the application.
The researchers concluded that the findings support a recent recommendation that telemedicine can be successfully performed by physicians who are part of a patient's medical home or health system.
"When a teledermatology diagnosis is uncertain or requires further management, there must be a system in place to facilitate appropriate follow-up with the patient's primary clinician, dermatologist, or other specialists. Without the opportunity for in-person follow-up, patients may not receive accurate diagnoses when teledermatology is insufficient, may be sent to emergency departments, or may not receive further care at all," they wrote
Study limitations included the small sample size, use of only one pediatric dermatologist to provide remote diagnoses, and limited generalizability.