HEALTH CONSULTATION
Health Outcome Data Evaluation
KELLY AIR FORCE BASE
SAN ANTONIO, BEXAR COUNTY, TEXAS
After receiving the petition to perform a public health assessment on neighborhoods north andsoutheast of Kelly Air Force Base, ATSDR requested that the Cancer Registry Division (CRD) ofthe Texas Department of Health evaluate cancer rates in zip code areas 78211 and 78237. Specifically, incidence (cases) and mortality (deaths) data were evaluated for cancers of the colon,pancreas, lung, prostate, breast, and leukemia for the periods 1985-1992 and 1990-1994respectively.
In evaluating the cancer incidence data (Tables A-1 and A-2), the number of cancer cases amongboth males and female residents was either lower than or within the range expected, with theexception of pancreatic cancer which was elevated among males in zip code area 78211. Evaluation of the mortality data (Tables A-3 and A-4) in these two zip codes found the numberof cancer deaths among both male and female residents to be either lower than or within the rangeexpected with the exception of colon cancer and leukemia, which were elevated among males in zip code area 78237.
| MALES | ||||
|---|---|---|---|---|
| Site | Observed | Expected | SIR | 95% CI |
| Colon Pancreas Lung Prostate Leukemia | 24 | 38.9 | 0.6† | 0.4-0.9 |
| FEMALES | ||||
| Site | Observed | Expected | SIR | 95% CI |
| Colon Pancreas Lung Breast Leukemia | 18 | 5.5 | 0.6† | 0.3-0.9 |
Note: The SIR (standardized incidence ratio) is defined as the number of observed cases divided bythe number of expected cases. The latter is based on race-, sex-, and age-specific cancer incidencerates for California for the period 1988-1992. The SIR has been rounded to the first decimal place.
CI: confidence interval
* Significantly higher ( at the 5% level) than expected
† Significantly lower (at the 5% level) than expected
| MALES | ||||
|---|---|---|---|---|
| Site | Observed | Expected | SIR | 95% CI |
| Colon Pancreas Lung Prostate Leukemia | 42 | 47.8 | 0.9 | 0.6-1.2 |
FEMALE | ||||
| Site | Observed | Expected | SIR | 95% CI |
| Colon Pancreas Lung Breast Leukemia | 26 | 47.1 | 0.6† | 0.4-0.8 |
Note: The SIR (standardized incidence ratio) is defined as the number of observed cases divided bythe number of expected cases. The latter is based on race-, sex-, and age-specific cancer incidencerates for California for the period 1988-1992. The SIR has been rounded to the first decimal place.
CI: confidence interval
* Significantly higher ( at the 5% level) than expected
† Significantly lower (at the 5% level) than expected
| MALES | ||||
|---|---|---|---|---|
| Site | Observed | Expected | SMR | 95% CI |
| Colon Pancreas Lung Prostate Leukemia | 11 | 7.5 | 1.5 | 0.7-2.6 |
| FEMALES | ||||
| Site | Observed | Expected | SMR | 95% CI |
| Colon Pancreas Lung Breast Leukemia | 7 | 5.8 | 1.2 | 0.5-2.5 |
Note: The SMR (standardized mortality ratio) is defined as the number of observed deaths dividedby the number of expected deaths. The latter is based on race-, sex-, and age-specific cancer mortalityrates for Texas, 1990-1994. The SMR has been rounded to the first decimal place.
CI: confidence interval
| MALES | ||||
|---|---|---|---|---|
| Site | Observed | Expected | SMR | 95% CI |
| Colon Pancreas Lung Prostate Leukemia | 18 | 9.6 | 1.9* | 1.1-3.0 |
| FEMALES | ||||
| Site | Observed | Expected | SMR | 95% CI |
| Colon Pancreas Lung Breast Leukemia | 11 | 8.8 | 1.3 | 0.6-2.2 |
Note: The SMR (standardized mortality ratio) is defined as the number of observed deaths dividedby the number of expected deaths. The latter is based on race-, sex-, and age-specific cancerincidence rates for Texas, 1990-1994. The SMR has been rounded to the first decimal place.
CI: confidence interval
* Significantly higher ( at the 5% level) than expected
California was initially chosen as the comparison population due to the availability of cancerincidence rates for the Hispanic population in that state. The number of expected cases presentedin Tables B1-B3, therefore, was based on race-, sex-, and age-specific cancer incidence ratespublished by the California Cancer Registry. The initial zip codes of concern were 78211, 78228,and 78237.
During the period 1990-1994, the number of cases observed for cancer of the lung, bladder,kidney, or leukemia was close to the number expected among males and females in zip code78211 (Table B-1). However, a significant excess of liver cancer was observed among the maleresidents and a significant excess of cervical cancer was observed among females. A higher thanexpected number of liver cancer cases was observed among females, although it was of borderlinestatistical significance. In zip code area 78228, the number of cases of lung, bladder, andleukemia cancer observed among males and females was close to the expected number (TableB-2). A significant excess of liver cancer among males was observed as was a significant excessof kidney cancer among males. The number of liver and kidney cancer cases was higher thanexpected among females, although the excesses were of borderline statistical significance. Noexcess of cervical cancer was observed among females. The number of cases observed for lung, bladder, and kidney cancer, as well as leukemia, was close to the number expected among malesin zip code 78237 (Table B-3). The number of cases observed of liver and bladder cancer amongfemales in this zip code was also close to the number expected. There was, however, a significantexcess of liver cancer observed among the male residents, and a significant excess of cervical andkidney cancer in females. A higher than expected number of leukemia cases and a lower than expected number of lung cancer cases were observed among females, although these excesses were of borderline statistical significance.
Additional Liver Cancer Analysis
Liver cancer incidence data was also examined for 15 additional zip codes in the Kelly AFB area. The analysis of incidence data using California as the comparison population during the period1990-1994 indicates a statistically significant excess of liver cancer among males in the 78201,78204, 78205, 78207, and 78227 zip code areas (Table B-4). A higher than expected number ofliver cancer cases was observed among males in zip code 78225, although this excess was ofborderline statistical significance. Among females in the study area during the same time period, astatistically significant excess of liver cancer was observed in zip code areas 78207, 78212, and 78221.
The initial analysis of liver cancer incidence using California rates for comparison found asignificant excess of cases among males in all three of the initial zip code areas of concern with acorresponding elevation (although not statistically significant) among females in two zip codeareas also. The additional liver cancer analysis using California rates for comparison found asignificant excess of cases among males in five of the fifteen zip code areas evaluated and among females in three of the zip code areas.
One of the limitations in using California data for the comparison population is that historicallyliver cancer rates in Texas have been consistently higher than those in California. This differencein background rates in the Texas and California populations is reflected in the standardizedincidence ratios calculated for the zip code areas of concern. Using the California rates as acomparison results in an artificially lower expected number of cases for the San Antoniopopulation and subsequently a higher standardized incidence ratio than if the Texas rates wereused. The magnitude of the standardized incidence ratios for both males and females, however, is substantially reduced when using Texas as the comparison population.
| MALES | ||||
|---|---|---|---|---|
| Site | Observed | Expected | SIR | 95% CI |
| Liver Lung Bladder Kidney Leukemia | 10 | 3.6 | 2.8* | 1.3-5.1 |
| FEMALES | ||||
| Site | Observed | Expected | SIR | 95% CI |
| Liver Lung Cervix Bladder Kidney Leukemia | 5 | 1.6 | 3.1 | 1.0-7.3 |
Note: The SIR (standardized incidence ratio) is defined as the number of observed cases dividedby the number of expected cases. The latter is based on race-, sex-, and age-specific cancerincidence rates for California during the period 1989-1993.
CI: confidence interval
Bold type indicates an excess of borderline statistical significance
* Significantly higher (at the 5% level) than expected
| MALES | ||||
|---|---|---|---|---|
| Site | Observed | Expected | SIR | 95% CI |
| Liver Lung Bladder Kidney Leukemia | 23 | 6.8 | 3.4* | 2.1-5.1 |
| FEMALES | ||||
| Site | Observed | Expected | SIR | 95% CI |
| Liver Lung Cervix Bladder Kidney Leukemia | 8 | 3.3 | 2.4 | 1.0-4.8 |
Note: The SIR (standardized incidence ratio) is defined as the number of observed cases dividedby the number of expected cases. The latter is based on race-, sex-, and age-specific cancerincidence rates for California during the period 1989-1993.
CI: confidence interval
Bold type indicates an excess of borderline statistical significance
* Significantly higher (at the 5% level) than expected
| MALES | ||||
|---|---|---|---|---|
| Site | Observed | Expected | SIR | 95% CI |
| Liver Lung Bladder Kidney Leukemia | 20 | 4.7 | 4.3* | 2.6-6.6 |
| FEMALES | ||||
| Site | Observed | Expected | SIR | 95% CI |
| Liver | 5 | 2.4 | 2.1 | 0.7-4.9 |
Note: The SIR (standardized incidence ratio) is defined as the number of observed cases dividedby the number of expected cases. The latter is based on race-, sex-, and age-specific cancerincidence rates for California during the period 1989-1993.
CI: confidence interval
Bold type indicates an excess of borderline statistical significance
* Significantly higher (at the 5% level) than expected
| MALES | ||||
|---|---|---|---|---|
| Zip Code | Observed | Expected | SIR | 95% CI |
| 78201 | 16 | 4.8 | 3.3* | 1.9-5.4 |
| 78204 | 6 | 1.7 | 3.5* | 1.3-7.7 |
| 78205 | 3 | 0.3 | 10.0* | 2.1-29.2 |
| 78207 | 23 | 7.6 | 3.0* | 1.9-4.5 |
| 78212 | 6 | 3.7 | 1.6 | 0.6-3.5 |
| 78221 | 7 | 3.4 | 2.1 | 0.8-4.2 |
| 78224 | 2 | 1.1 | 1.8 | 0.2-6.6 |
| 78225 | 6 | 2.1 | 2.8 | 1.0-6.2 |
| 78226 | 2 | 0.8 | 2.5 | 0.3-9.0 |
| 78227 | 11 | 3.1 | 3.5* | 1.8-6.3 |
| 78229 | 3 | 1.9 | 1.6 | 0.3-4.6 |
| 78238 | 0 | 1.5 | 0.0 | 0.0-2.5 |
| 78242 | 4 | 1.4 | 2.9 | 0.8-7.3 |
| 78245 | 2 | 1.3 | 1.5 | 0.2-5.6 |
| 78252 | 0 | 0.1 | 0.0 | 0.0-36.9 |
| FEMALES | ||||
| Zip Code | Observed | Expected | SIR | 95% CI |
| 78201 | 7 | 3.0 | 2.3 | 0.9-4.8 |
| 78204 | 3 | 1.0 | 3.0 | 0.6-8.8 |
| 78205 | 0 | 0.2 | 0.0 | 0.0-18.0 |
| 78207 | 15 | 4.3 | 3.5* | 2.0-5.8 |
| 78212 | 8 | 2.2 | 3.6* | 1.6-7.2 |
| 78221 | 7 | 1.6 | 4.4* | 1.8-9.0 |
| 78224 | 1 | 0.6 | 1.7 | 0.0-9.3 |
| 78225 | 3 | 1.0 | 3.0 | 0.6-8.8 |
| 78226 | 2 | 0.3 | 6.7 | 0.8-24.1 |
| 78227 | 4 | 1.4 | 2.9 | 0.8-7.3 |
| 78229 | 0 | 1.0 | 0.0 | 0.0-3.7 |
| 78238 | 0 | 0.7 | 0.0 | 0.0-5.3 |
| 78242 | 2 | 0.6 | 3.3 | 0.4-12.0 |
| 78245 | 1 | 0.5 | 2.0 | 0.1-11.1 |
| 78252 | 0 | 0.0 | 0.0 | --- |
Note: The SIR (standardized incidence ratio) is defined as the number of observed cases dividedby the number of expected cases. The latter is based on race-, sex-, and age-specific cancerincidence rates for California during the period 1989-1993.
CI: confidence interval
Bold type indicates an excess of borderline statistical significance
* Significantly higher (at the 5% level) than expected
Additional Liver Cancer Mortality Analysis
When conducting the additional liver analysis, five zip code areas were evaluated in the mortalityanalysis that were not included in the liver cancer incidence analysis. For the sake of consistency,the results from the same zip code areas were presented in the Health Outcome Data section. The results from the additional zip code areas are presented in Table C-1.
The additional analysis of liver cancer mortality in five zip code areas did not find a significant excess of cases among males or females in any of the zip code areas.
| MALES | ||||
|---|---|---|---|---|
| Zip Code | Observed | Expected | SMR | 95% CI |
| 78212 | 11 | 6.8 | 1.6 | 0.8-2.9 |
| 78229 | 6 | 3.4 | 1.8 | 0.6-3.8 |
| 78238 | 0 | 2.3 | 0.0 | 0.0-1.6 |
| 78245 | 2 | 1.9 | 1.1 | 0.1-3.8 |
| 78252 | 0 | 0.1 | 0.0 | 0.0-36.9 |
| FEMALES | ||||
| Zip Code | Observed | Expected | SMR | 95% CI |
| 78212 | 6 | 5.4 | 1.1 | 0.4-2.4 |
| 78229 | 1 | 2.1 | 0.5 | 0.0-2.7 |
| 78238 | 0 | 1.5 | 0.0 | 0.0-2.5 |
| 78245 | 1 | 0.9 | 1.1 | 0.0-6.2 |
| 78252 | 0 | 0.1 | 0.0 | 0.0-36.9 |
Note: The SMR (standardized mortality ratio) is defined as the number of observed deaths dividedby the number of expected deaths. The latter is based on race-, sex-, and age-specific cancermortality rates for Texas during the period 1990-1995.
CI: confidence interval
Bold type indicates an excess of borderline statistical significance
* Significantly higher (at the 5% level) than expected
