Background: Central cancer registries provide data to monitor incidence rates of cutaneous melanoma.
Objective: The aim of this study was to assess the completeness of melanoma reporting in the United States.
Methods: Data provided by central cancer registries were used to calculate age-adjusted, average annual incidence rates and were compared by time period (1992-1994, 1995-1997), stage, and program (Surveillance Epidemiology and End Results [SEER] and National Program of Cancer Registries [NPCR]). Completeness was measured with incidence/mortality ratio.
Results: Incidence rates among whites for 1995-1997 from SEER registries ranged from 11.8 to 33.9 per 100,000 population; 18 of 40 NPCR registries were within this range. For 1992-1994, 8 of 30 NPCR registries were within the range of SEER incidence rates. NPCR registry incidence rates were generally higher for 1995-1997 than 1992-1994. The percentage of cases of localized melanoma did not increase substantially in most SEER registries over the study period, but some NPCR registries had substantial increases. Among NPCR registries that had incidence rates comparable with those of SEER in 1995-1997, the incidence/mortality ratios were generally lower among NPCR registries than SEER registries.
Conclusion: Although melanoma incidence rates are generally increasing, part of the increases in incidence rates reported by NPCR registries over the study time period are likely due to increased case ascertainment and reporting.