![[ Visit Client Website ]](images/banner.png)
BACKGROUND: Vermont consistently has one of the highest incidence rates of melanoma in the United States. In this investigation, we used cancer registry data to analyze trends of incidence, stage at diagnosis, and tumor location, based on factors including age, sex, geographic location, and health care payer.
METHODS: All analyses used a limited-identifiers dataset of cutaneous melanoma cases from the Vermont Cancer Registry (2001-2010). We used logistic regression to analyze whether there were significant differences in stage at diagnosis and tumor location based on age, sex, and insurance payer type. We conducted standardized incidence ratio (SIR) analyses on sub-county geography levels, ranging in size from census tracts to multiple towns, based on considerations of statistical power and demographic representation.
RESULTS: Cases where Medicaid was the primary insurance payer were more likely than other cases to have an advanced stage (regional or distant SEER Summary Stage) at diagnosis (OR=2.43; 95% CI: 1.19, 4.98). Uninsured cases were more likely than other cases to be diagnosed with invasive disease (spread into surrounding healthy tissues from the layer in which it developed – includes localized, regional, or distant SEER Summary Stages) (OR=2.31; 95% CI = 1.20, 4.47) but were not more likely than other cases to have an advanced stage at diagnosis. Cases sixty years of age or older were less likely to be diagnosed with invasive disease (OR=0.74; 95% CI: 0.64, 0.85) than younger cases were. However, there were no significant associations between age and advanced stage of disease at diagnosis. Eighteen (out of a total of 107) sub-county geographies had SIRs that were statistically different from the state rate and values ranging from 0.34 to 2.23, indicating strong geographic heterogeneity of melanoma rates within the state.
CONCLUSIONS: Our analysis indicates that there are statistically significant trends in payer and geographic area of residence at diagnosis that may be important to communicate to physicians and in targeting melanoma prevention and control efforts. The methods used in this investigation may also be useful in assessing trends in other cancer types and other states.