New study finds no significant harm in melanoma screenings by primary care providers

Brown UniversityJul 11 2016

A new study of more than 1,000 primary care melanoma screenings in the western Pennsylvania area in 2014 suggests that overall the practice complies with the old medical maxim to "first do no harm." The numbers show that screenings did not result in a spike in surgical treatments or specialist referrals.

Malignant melanoma kills about 10,000 people in the U.S. every year, but the condition can be cured if caught early. That has led some experts to call for widespread training of primary care providers to conduct screenings at routine visits. Other experts have worried, though, that the practice could lead to rampant overtreatment and unnecessary patient distress if, for example, primary care providers mistake benign lesions for serious cancers. Such competing concerns are among the things under consideration as the U.S. Preventative Services Task Force prepares new skin cancer screening recommendations.

In this context, a team of researchers from multiple centers led by Dr. Martin Weinstock, professor of dermatology in the Warren Alpert Medical School of Brown University and chief of dermatology at the Providence Veterans Affairs Medical Center, decided to study what happened when a cadre of primary care providers in the University of Pittsburgh Medical Center system gained online training in melanoma screening and began looking for the skin cancers in 2014.

The training used was the freely available online course INFORMED, which Weinstock and collaborators developed in 2010.

Skin screening study

To perform the analysis, Weinstock's team reviewed the data from tens of thousands of encounters with patients 35 years and older both in the first eight months of 2013 (before the training occurred) and 2014 (after the training occurred). The researchers also divided care providers into three comparison groups -- one in which about a quarter of providers were trained, one in which only 1 in 11 of the providers were trained, and one in which none were trained. The study design therefore allowed the team to compare several outcomes both over time and across groups with different degrees of training.

The researchers found that neither dermatologist visits nor skin surgeries increased substantially between 2013 and 2014 in any of the groups. Between the groups there was also little difference in how often those outcomes occurred.

The lack of major change was not because the training had no effect, said Weinstock, who also leadas a melanoma program in the Lifespan health system. Between 2013 ad 2014, the group of providers with the most training did produce a 79 percent increase in per-patient melanoma diagnoses. The other groups with less or no training showed no significant increase in diagnoses.

The reason why diagnoses rose significantly in the most trained group but dermatologist visits or surgeries did not is because the number of diagnoses were tiny (24 out of 11,238 patients in 2013 and 48 out of 12,560 patients in 2014). In the same group, in both years, skin surgeries numbered in the hundreds and dermatologist visits numbered in the thousands. There are many reasons why both of those could occur independently of melanoma.

In other words, the newly diagnosed melanoma patients could well have received the proper follow-up care that their diagnoses warranted without radically changing the overall number of dermatologist visits or surgeries. But in the meantime, Weinstock and his colleagues wrote, there was no sign that training primary care providers to screen for melanoma had flooded dermatologists or surgeons with torrents of cases.

"The early evidence from this screening effort indicates that melanoma screening coupled with a modified INFORMED training program neither results in large numbers of skin surgeries nor a dramatic increase in visits to dermatologists, which are two potential adverse downstream consequences that have been of concern as a potential basis of morbidity, distress and costs," the authors wrote in Cancer.

In a follow-up study, Weinstock said, the team plans to analyze metrics of whether screening resulted in significantly greater distress for patients. That will be a test not only of whether primary care screening has that unintended result, but also of whether the INFORMED course is successful in training providers to not only spot cancers, but also reassure patients when lesions are benign.