Smartphone photos aid pediatric diagnosis

Photos taken by parents and submitted by smartphone are usually of sufficient quality to allow accurate diagnosis of some common pediatric skin conditions, results of a recent study in JAMA Dermatology suggests.

In the study, which included 40 patient-parent dyads, there was an 83% concordance between photograph-based diagnosis and in-person diagnosis, according to senior author Patrick J. McMahon, M.D., Section of Dermatology, Division of General Pediatrics, Children’s Hospital of Philadelphia.

Providing basic “how-to” instructions seemed to make the quality of parents’ photos even better, but the study was underpowered to determine whether that difference was significant versus untrained parents, investigators said.

Taken together, these findings suggest parent-operated smartphone photography can be an additional method used to provide pediatric dermatologic care, according to Dr. McMahon.

“There's always been this looming question as to whether we can trust the photographs that are coming in from parents (via smartphones),” Dr. McMahon said in an interview with Dermatology Times. “Based on results of this study, I do think in this day and age, we could feel more comfortable using this technology with its current capabilities.”

The study by Dr. McMahon and colleagues is believed to be the first to specifically look at whether parents can take photographs of high enough quality to render a diagnosis, and whether there is a concordance between diagnoses based on those images versus in-person diagnoses.

Previous investigations have mainly focused on the diagnostic value of photographs taken by physicians or other health care providers. In a 2009 study including 135 consecutive children presenting with rash, a concordance rate of 82% was reported for in-person diagnoses versus diagnoses based on photos the investigator took using a 5-megapixel Canon PowerShot digital camera.

In the present study by Dr. McMahon and colleagues, parents used their own personal smartphones to take photographs of their child’s’ skin condition in the examination room. They securely uploaded the photos using MyChart, a smartphone app that interfaces directly with the child’s electronic health record.

The concordance rate was 83% overall, as previously noted, and 89% in the subgroup of 37 cases where the photographs were of high enough quality to make a diagnosis (three cases had image quality that was too poor to provide a diagnosis).

Concordance varied depending on diagnostic category: it was 100% for 3 cases of hemangiomas, 100% for 6 pigmented lesions, and 92% for 13 cases of rash. By contrast, concordance rates were 67% for 3 nodules/tumors and 64% for 14 alopecia-related diagnoses.

As a secondary aim, Dr. McMahon and colleagues sought to determine whether a basic, 3-step instruction sheet on how to take better smartphone photographs would improve image quality. They randomized the parent-patient dyads 1:1 to receive the instruction sheet or not.

Image quality was indeed better for parents who received the instructions, though the finding did not reach statistical significance, Dr. McMahon said. Moreover, there was no statistical difference in diagnostic concordance between parents who received instructions and those who did not.

Parents were also asked to rate their willingness to use a teledermatology app to communicate with a pediatric dermatologist, as opposed to waiting for an in-person appointment. On a scale of 1 to 10, parents most often selected 10 (very willing) and the median score was 8, Dr. McMahon and colleagues reported.

When asked how much they would pay for a such a virtual visit on a scale of $0 to $200, the median price selected was $20, and answers ranged from $0 to $160, according to the investigators.

Overall, Dr. McMahon said he was encouraged by results of the study and the potential for future use of teledermatology based on parent-submitted images.

“It’s nice to know that the doctors are able to look at a picture and have diagnostic accuracy that's good enough, even with some images that are imperfect,” he said.

Summary

• Parent-submitted photos were often sufficient to accurately diagnose pediatric skin conditions.
• Providing parents with instructions on how to take better photos seemed to help improve image quality, but the study was underpowered to show whether the improvement was significant.
• Most parents said they were very willing to use teledermatology apps and indicated they would pay for virtual pediatric dermatology consults.

References

• O’Connor DM, Jew OS, Perman MJ, Castelo-Soccio LA, Winston FK, McMahon PJ. Diagnostic Accuracy of Pediatric Teledermatology Using Parent-Submitted Photographs: A Randomized Clinical Trial. JAMA Dermatol. 2017;153(12):1243-1248. doi:10.1001/jamadermatol.2017.4280.
• Heffner VA, Lyon VB, Brousseau DC, Holland KE, Yen K. Store-and-forward teledermatology versus in-person visits: A comparison in pediatric teledermatology clinic. J Am Acad Dermatol. 2009;60(6):956-961. doi:10.1016/j.jaad.2008.11.026.