High Melanoma Incidence Associated with Diagnostic Scrutiny Rather Than UV Exposure

Research suggests melanoma incidence patterns in the US are more associated with diagnostic scrutiny than UV radiation exposure.

The frequency of diagnoses is often strongly influenced by diagnostic scrutiny changes, such as new screening tests. This phenomenon was observed in South Korea, during which the advent of ultrasound screens led to a dramatic 15-fold increase in incidence of thyroid cancer.

In the US, cutaneous melanoma is the third most frequently diagnosed cancer despite previously being considered rare. Due to the effect of increased scrutiny on diagnoses, this rise in melanoma incidence was seen as important to examine by the researchers led by Heather Welch, MSc, and Adewole S. Adamson, MD, MPP, of the Division of Dermatology at University of Texas at Austin’s Dell Medical School.

“This rise in incidence is conventionally interpreted as a genuine increase in true cancer occurrence, presumably the result of increased exposure to… UV radiation,” wrote Welch and colleagues. “An alternative interpretation is overdiagnosis, the result of increased diagnostic scrutiny: as more physicians look for and biopsy moles, more melanoma is found.”

In their research, the investigators analyzed records of melanoma incidence around the US and the relative association of proxies for diagnostic scrutiny and exposure to UV radiation.

Research and Methods

The investigators used a cross-sectional ecological analysis, recording information from the 727 continental US counties reporting cancer incidence data through the Surveillance, Epidemiology, and End Results (SEER) program.

They combined SEER data regarding melanoma incidence in the non-Hispanic White population, due to melanoma occurrence being strongly correlated with skin pigmentation. This also allowed the researchers to avoid confounding variables such as race and ethnicity, using one large, distinct group. The data collected from this group came from diagnoses reported from 2012 through 2016.

Information on the specific proportion of the population in a given county screening for melanoma is unavailable in the US. Consequently, the investigators used physician supply level (per 100,000 residents) and median household income to access reliable data. The physicians were either dermatologists or primary care clinicians. The investigators used scatterplots of skin cancer incidence versus proxies for diagnostic scrutiny and UV exposure, assessing the association pairs of each variable using Pearson correlation coefficients.

Study Findings

The investigators found that incidence of melanoma among the sample population varied widely, even in counties with similar UV exposure rates. A notable statistic was that the greater Seattle area, which maintains the lowest daily UV doses in the country, had among the highest incidence of melanoma. Conversely, the lowest melanoma incidences for this same sample population were found in southeastern New Mexico, where UV exposure is some of the highest.

Melanoma incidence in counties with dermatologists was reported as 25% higher than in those without dermatologists. Differences in UV radiation exposure proxies (measured by daily dose of UV rays) do not explain the distinction due to being lower in counties with dermatologists (P = .03). Additionally, melanoma incidence was about 20% lower in counties with shortages of primary care physicians compared to those with more physicians. These differences are not explained by UV radiation proxies.

“Nevertheless, sunburns—particularly early in life—have consistently shown an association with melanoma, and efforts to reduce sunburns are clearly warranted,” they wrote. “At the same time, it is important for clinicians and public health officials not to exaggerate the magnitude of cancer risk factors, particularly for exposures as ubiquitous as the sun. Suggestions that 95% of melanomas are caused by the sun are particularly troubling.”

This study, “Association of UV Radiation Exposure, Diagnostic Scrutiny, and Melanoma Incidence in US Counties,” was published online on JAMA Internal Medicine.