Author Affiliations: Department of Dermatology, University Hospitals Case Medical Center, Cleveland, Ohio (Ms Aneja and Dr Bordeaux); Case Western Reserve University School of Medicine, Cleveland (Ms Aneja and Dr Bordeaux); Yale University School of Medicine, New Haven, Connecticut (Mr Aneja).
We appreciate Dr Lott's interest in our study demonstrating the association between geographic variation in dermatologist density and melanoma mortality.1 We would like to make the following comments in response.
We are familiar with the Robert Wood Johnson Foundation County Health Rankings database and similar sources. Confounders are an issue in any study that makes use of such databases. In our study we considered potential confounders, including metropolitan county status, health professional shortage area classification, primary care physician density, population older than 65 years, population with high school diploma or equivalent education level, number of hospitals with oncology departments, non-Hispanic white population, health-insured population, median household income, and unemployment rate. While the potential for unrecognized confounders surely exists, we are curious as to what specific previously unrecognized cofounders are concerning, and further, what effect (if any) these unrecognized confounders would have on the outcomes and conclusions of our analysis.
Aneja S, Aneja S, Bordeaux JS. On Dermatologist Density and Melanoma Mortality—Reply. Arch Dermatol. 2012;148(9):1092-1093. doi:10.1001/archdermatol.2012.1727