Volume 13, Number 7—July 2007
Estimating Severe Coccidioidomycosis in California
|Population||Frequency of hospitalization†||RR for discharge diagnosis of coccidioidomycosis||Frequency of hospitalization in highest incidence counties†||RR for discharge diagnosis of coccidioidomycosis in highest incidence counties|
|Hispanic||3.4||0.73 (0.68–0.78)||24.1||0.71 (0.63–0.79)|
|Black||8||2.68 (2.48–2.91)||80.8||2.43 (2.10–2.82)|
|Native American||1.4||0.32 (0.21–0.51)||12.7||0.37 (0.20–0.70)|
|Asian-Pacific Islander||2||0.78 (0.70–0.87)||51||1.62 (1.34–1.97)|
|Male||5||2.14 (2.03–2.27)||21.9||1.67 (1.53–1.85)|
|<14||0.5||0.12 (0.10–0.14)||3.9||0.12 (0.09–0.15)|
|50–69||7.1||2.13 (2.01–2.26)||57.2||1.83 (1.66–2.03)|
|>70||7.3||2.74 (2.54–2.97)||47.0||1.83 (1.58, 2.12)|
|AIDS||127||34.5 (31.0–38.4)||912||31.0 (23.8–40.4)|
|All HIV||51||13.9 (12.5–15.5)||319||10.8 (8.3–14.1)|
*Multivariate results reported for 6,465 cases with no missing data (87%). RR, relative risk; Ref, referent. Values in parentheses are 95% confidence intervals.
†Crude incidence per 100,000 residents in Kern, Tulare, Kings, and San Luis Opispo counties.
‡RR by multivariate Poisson model controlling for year, county, age, race, and sex. For the years 1997–1999, racial-ethnic categories were White, Hispanic, Black, Native American, and Asian-Pacific Islander. For the years 2000–2002, racial-ethnic categories were White, Hispanic, Black, Native American, Asian, Pacific Islander, and multirace. The population of multirace represented ≈1% of populations relevant to the study and was not included in the analysis. California Department of Finance population estimates from 2000–2002 for Asians and Pacific Islanders were combined into a single category, Asian-Pacific Islander, to match coding of the Office of Statewide Health Planning and Development database. All Hispanics in the Department of Finance data were assumed to be of white race.
§Bivariate relative risk.
¶To estimate a denominator for the pregnant population, we estimated the total person-years of pregnancy for each county in California in the following manner. The total number of live births was multiplied by 0.75 (to approximate 9 mo of pregnancy) and added to the total number of fetal deaths, multiplied by 0.56 (to estimate a gestation of 30 weeks). This sum was finally added to the total number of abortions, multiplied by 0.19 (to estimate a gestation of 10 weeks). Estimates of live births and fetal deaths were obtained from the Center for Health Statistics Birth Rate Tables (5). Annual number of abortions was estimated from federal abortion surveillance data from 1997 (6).