Geographic distribution of incident mesothelioma in Connecticut men, 1990-99.

Abstract

Hemminki and Li1 reported that for pleural mesothelioma in Sweden (1970–98) high standardized incidence ratios (SIRs) were found for men in the 3 largest cities and 62.8% of cases had “no apparent occupational explanation suggesting exposure to asbestos” (based on information from linkage with national censuses). Although the Unites States has no national cancer registry or linkage of individual patients with national censuses, a United States study found that of the 38 counties with a statistically significant excess of male deaths in 1968–81 from pleural mesothelioma 15 had an asbestos plant, and 28 had a shipbuilding facility in the county or an adjacent county. One of the latter 28 counties was New London County, Connecticut. The present study examined mesothelioma incidence in men in Connecticut by county and town for the most recent 10 years of diagnosis (1990–99) in the Connecticut Tumor Registry, part of the U.S. National Cancer Institute's (NCI) Surveillance, Epidemiology and End Results (SEER) Program of high-quality population-based cancer registries.3 Using International Classification of Diseases for Oncology (ICD-O-2) morphology codes 9050-9053, 238 malignant mesotheliomas of the pleura, peritoneum or other sites were diagnosed among Connecticut men. Also included were 13 pleural cancers of unspecified histologic type (ICD-O-2 8000-8004) in men. Almost all pleural cancers are mesotheliomas.4 Using the county and town of residence at the time of diagnosis, expected numbers were calculated by multiplying statewide ages 30–34 through 80–84 and 85+ years) and gender-specific incidence rates by population estimates (1990 census for 1990–94 and 2000 census for 1995–99) for each of the 8 counties and 169 towns; accurate population estimates for intercensal years were not available. Observed and expected numbers for 1990–94 and 1995–99 were combined to obtain SIRs for 1990–99; 95% confidence intervals (CI) on each SIR were based on the Poisson distribution. Statistically, significantly elevated SIRs in men were found for 2 adjacent counties (New London and Middlesex; Fig. 1), which are not the most urban counties (total populations 259,088 and 155,071, respectively, in the 2000 census). Fairfield (near New York City), New Haven and Hartford counties, with SIRs <1.00, each had >800,000 population in 2000, and include all of the state's largest cities (>100,000 population). Five towns in the state had significantly elevated SIRs in men, 4 in New London county and 1 in Middlesex county (Fig. 1); all 5 had <40,000 people in 2000. Map of boundaries of Connecticut counties and towns, with standardized incidence ratios (SIRs) for mesothelioma in men by county of residence, 1990–99. Shaded towns had a statistically significantly elevated SIR. A major shipbuilder is located in New London county (in Town 1, Fig. 1). Using “usual occupation” (from hospital reports and death certificates) and information on occupational histories reported (albeit incompletely) by hospitals to the registry, 17 of 54 (31.5%) men in New London county and 2 of 23 (13.0%) in Middlesex county had worked in shipbuilding. Only 1 of 5 towns (Town 2, Fig. 1) had no cases with a history of working in shipbuilding. Although no control group was available, the importance of occupational histories and other data sources was suggested by review of registry records for the 34 cases in these 5 towns. One reportedly had asbestos exposure during World War II, 10 had worked in shipbuilding, 11 in another occupation linked to asbestos exposure5, 6 and another 7 in an occupation linked to mesothelioma risk7 (including a pharmaceutical industry worker with a pathology report mentioning asbestos exposure); the other 5 included “U.S. Navy retired” and “U.S. Coast Guard.” A review of European studies (including one in the area of Lund, Sweden) combining occupational histories with analyses of lung fiber content found that 62–85% of men with mesothelioma had probable occupational exposure to asbestos.8 Perhaps in the Swedish study1 the small number of male cases in agriculture/gardening had less chance of a history (undocumented in the censuses) of occupational asbestos exposure than the small number of professionals and larger numbers of blue collar and non-asbestos manual workers (especially in urban areas). Otherwise, conjectures for the urban excess in Sweden vs. Connecticut would include differences in non-occupational asbestos exposure1 (perhaps also in type and size of asbestos fibers) or in causes of mesothelioma other than asbestos.9 Yours sincerely, Anthony P. Polednak*, * Connecticut Department of Public Health, Hartford CT, USA.