Risk factors for HIV-1 seroconversion among injection drug users: a case-control study.

OBJECTIVES: A nested case-control study was conducted in Miami, Fla, to determine risk factors associated with human immunodeficiency virus type 1 (HIV-1) seroconversion among injection drug users. METHODS: The study identified 21 incident cases of HIV-1 infection and 76 unmatched controls from two longitudinal cohorts of injection drug users. One cohort consisted of individuals who originally had been recruited from treatment centers; a second cohort was recruited from the "street." Logistic regression analyses that adjusted for age, gender, and race were performed. RESULTS: The final model determined that the primary independent risk factor that best explained the risk for seroconversion was sharing injection equipment in the year prior to conversion; a marginal risk factor was presence of sexually transmitted disease during this same period. CONCLUSIONS: Both an injection component and a sexual component play a role in seroconversion among injection drug users, although the injection component is much stronger.

Epidemiology of cancer among Hispanics in the United States.

BACKGROUND: There are very limited data on cancer among U.S. Hispanics. The data that are available seldom appear outside individual registry reports. Without coalescing and assessing such information, the epidemiology of cancer among Hispanics cannot be understood. PURPOSE: This article presents an overall view of the epidemiology of cancer among Hispanic populations in the United States. METHODS: Major types of cancer were reviewed from geographic areas with 1) large Hispanic/Latino populations and 2) population-based cancer registries that cover 68% of the U.S. Hispanic population. Age-standardized rates, standardized rate ratios, and ranks of the top cancers are presented for Hispanics and non-Hispanics by sex. RESULTS: Hispanic men had consistently lower rates of cancers of the lung, prostate, colon, rectum, and oral cavity as well as of melanoma and non-Hodgkin's lymphoma than non-Hispanic men. Among Hispanic men, higher rates were observed for cancers of the stomach, liver, and gallbladder. Compared with the rates in non-Hispanic women, Hispanic women had consistently lower rates of cancers of the breast, colon, rectum, endometrium, lung, ovary, and oral cavity, non-Hodgkin's lymphoma, and melanoma and higher rates of cancers of the uterine cervix, stomach, liver, and gallbladder. Perhaps more importantly, however, was that, within U.S. Hispanic populations, the top five sites of cancer for women were breast, colon, lung, cervix, and uterine corpus. For men, the top sites were usually prostate, lung, colon, stomach, and rectum. Furthermore, some geographic areas showed differences in the order of importance within these five sites, while in other areas, the top five cancers for Hispanic men included kidney and bladder cancers, liver cancer, and non-Hodgkin's lymphomas. For women, other leading cancers mentioned were ovarian cancer, non-Hodgkin's lymphoma, and gallbladder cancer. CONCLUSIONS: Hispanics had consistently lower rates of breast, lung, prostate, and colon cancers than non-Hispanics. In some geographic regions, however, differences in cancer incidence rates with respect to specific sites existed between the heterogenous Hispanic populations. Furthermore, the rankings demonstrated that, even if Hispanics have lower rates of the most common cancers than non-Hispanics, such sites are still the most important cancers among Hispanic populations. IMPLICATIONS: Cancer is a major problem among U.S. Hispanic populations. The information given in this article can be used to develop etiologic studies and cancer control interventions. Data limitations and recommendations for improving data quality are presented.

Differences in stage at presentation of breast and gynecologic cancers among whites, blacks, and Hispanics.

BACKGROUND: One of the possible assumptions for the higher mortality from cancer of blacks versus whites is that blacks tend to be diagnosed relatively more often with later stage disease. This study examined the stages at diagnosis for female breast cancer and other gynecologic cancers among blacks, non-Hispanic whites, and Hispanic whites. METHODS: Data for Dade County were analyzed in this study, which was based on the tumor registry data from the Florida Cancer Data System from 1981-1989. The stages at diagnosis were classified as in situ, local, regional, and distant according to the International Classification of Diseases for Oncology coding rules. RESULTS: The percentage of early stage diagnoses (in situ and local stage) was significantly lower among blacks and Hispanic whites than among non-Hispanic whites, but only for breast and cervical cancer. There was a significant increase in the percentage of early stage diagnoses of breast cancer among all three groups during 1981-1989, whereas this increase was found only in Hispanic whites for cervical cancer. The percentage of in situ cervical cancers was much lower in women older than age 45, especially among women older than age 65, in all three groups. CONCLUSIONS: These data suggest that screening programs need to be targeted differently for specific age and ethnic groups.

Cancer in south Florida Hispanic women. A 9-year assessment.

BACKGROUND: Cancer incidence varies among race and ethnic subgroups. Relatively little information exists on cancer among Hispanics, and none of it it stratified by race. METHODS: To explore and compare cancer incidence rates of each site of cancer among white Hispanic, white non-Hispanic, black Hispanic, and black non-Hispanic women in Dade County, South Florida, we accessed 9 years of cancer incidence data. Dade County's Hispanic population is the second largest of any US county. For each site, incidence rates, rate ratios, and 95% confidence intervals are presented. RESULTS: Among white Hispanics, there were statistically significant lower rates of cancers of the oral cavity, esophagus, colon, pancreas, lung, breast, ovary, bladder, kidney, melanoma, Kaposi's sarcoma, and non-Hodgkin's lymphoma than among white non-Hispanics. Black Hispanics had significantly lower rates of cancers of the oral cavity, stomach, rectum, lung, cervix, and bladder than black non-Hispanics. However, white Hispanic women had significantly higher rates of cancers of the liver, gallbladder, and uterine cervix than white non-Hispanic women. No in situ breast cancer cases among blacks of either ethnic subgroup was found. CONCLUSION: Both white Hispanic and black Hispanic women generally had lower rates of cancer than their race-specific non-Hispanic counterparts. However, incidence rates among Hispanics differ substantially by race; failure to distinguish them may lead to erroneous conclusions.

Estimates of trace element intakes in Chinese farmers.

Epidemiologic data on the average daily human dietary intake of the trace elements nickel, cadmium, molybdenum and silicon are scarce, worldwide, primarily because foods consumed vary widely over days and seasons. Available full year food allocation records for farmers in rural areas of China provide an unique opportunity to estimate dietary trace elements. In 1988, we measured the concentration of zinc, copper, iron, cadmium, nickel, molybdenum, silicon and selenium in different foods by using food samples collected from 232 villages (distributed among 21 communes). These measurements, combined with the food consumption information from existing food allocation records in each commune for an entire year, allowed us to estimate the average daily consumption of these trace elements by Chinese farmers. With one exception (an association of zinc and esophageal cancer), the variation of dietary trace elements did not show any association with mortality from several common diseases. The estimates for some of the dietary trace elements from some subpopulations were either less or more than current recommendations. No increases in mortality were found, however, from diseases normally associated with either dietary deficiencies or excesses of these elements.

Cancer among Hispanic males in south Florida. Nine years of incidence data.

BACKGROUND: Dade County (Florida) is the largest county in the United States in which a majority of the population is Hispanic. To describe and compare cancer incidence among white and black Hispanic and non-Hispanic males in South Florida and to identify those sites showing important racial-ethnic differences between sub-groups, incidence rates were determined from 9 years of prospective population-based cancer surveillance data. METHODS: All incident cancer cases diagnosed among male residents of Dade County for the period January 1, 1981, through December 31, 1989, were obtained from the statewide tumor registry for Florida, the Florida Cancer Data System. Race- and ethnic-specific age-standardized incidence rates and standardized rate ratios to compare rates among racial-ethnic subgroups for each site are presented. RESULTS: Compared with white non-Hispanics, white Hispanic males had significantly higher rates of cancers of the liver, gallbladder, nasal cavity, penis, and thyroid; acute lymphocytic leukemia; and Kaposi's sarcoma. Significantly lower rates were found for cancers of the esophagus, stomach, colon, rectum, pancreas, lung, testes, bladder, and kidney, and for mesothelioma, melanoma, non-Hodgkin's lymphoma, and chronic lymphocytic leukemia. Black Hispanics had significantly lower rates only of cancers of the esophagus, lung, and prostate than did black non-Hispanics. CONCLUSION: Differences were found in incidence rates by race-ethnicity; while incidence rates for most sites among white Hispanic males were lower than among white non-Hispanics, higher rates for several sites are of particular concern and merit further investigation to determine differences in risk factors and exposures.

Community activism relating to a cluster of breast cancer.

In 1985, residents of a suburban community in South Florida became concerned when several young women were diagnosed with breast cancer. First as individuals, then through a community-based effort, they attempted to gain attention and action on what they believed to be a cluster of breast cancer. Through their efforts to find out whether some agent or toxic exposure existed in the community that might have caused breast cancer, the women formed a community-based organization. This paper describes the activism of the women to resolve the issue through an epidemiologic study of the breast cancer occurrence in their community. Furthermore, it substantiates the need and role of rational community response in resolving community threats and concerns.

The epidemiology of cancer among Hispanic women. The experience in Florida.

To explore cancer incidence among Hispanic women living in Dade County, Florida, data were analyzed from the statewide cancer registry. For all but three sites, Hispanics had lower rates of the 15 most prevalent cancers than non-Hispanics. However, higher rates of cancer among Hispanics were noted for cancers of the gallbladder, liver, and heart and soft tissue. Subgroups of women had significantly higher rates of cervical cancer and thyroid cancer. Lower rates among Hispanics were observed for cancers of the esophagus, vagina, breast, colon, buccal cavity and pharynx, and malignant melanoma. These data suggest that most cancer sites traditionally higher among US Latino women were not higher among Dade Hispanics, and that sites more common among non-Hispanics have not yet shown an increased incidence among Hispanic women in Dade County.

Epidemiology of cancer among Hispanic males. The experience in Florida.

To assess cancer incidence among Hispanic males in Dade County, Florida, data were analyzed from the statewide tumor registry. For all cancer sites (combined), the age standardized rate among Hispanic males was 308.75 cases per 100,000 person-years, compared with 349.55 among non-Hispanics (standardized rate ratio [SRR] = 0.88; 95% confidence interval [CI] = 0.84-0.93). Hispanics experienced significantly greater rates of cancers of the larynx (SRR = 1.58; 1.19-2.09), thyroid (SRR = 3.12; 1.18-8.26), and gallbladder (SRR = 5.45; 1.55-19.15), compared with non-Hispanics, but significantly lower rates of testicular cancer (SRR = 0.17; 0.09-0.37), melanoma (SRR = 0.20; 0.12-0.34), esophagus (SRR = 0.52; 0.49-0.55), stomach (SRR = 0.61; 0.43-0.87), pancreas (SRR = 0.65; 0.45-0.94), kidney and renal pelvis (SRR = 0.68; 0.48-0.97), colon (SRR = 0.79; 0.67-0.94), lung (SRR = 0.82; 0.73-0.94), and Kaposi's sarcoma (SRR = 0.03; 0.00-0.25). These data suggest that Dade Hispanic males have not experienced the cancers traditionally elevated among US Latinos, or those common among non-Hispanics.

Survival in children with perinatally acquired human immunodeficiency virus type 1 infection.

We describe our experience at Jackson Memorial Hospital in Miami, Florida, with 172 children who were given diagnoses of perinatally acquired infection with human immunodeficiency virus type 1 (HIV-1). The 146 mothers of the children acquired HIV-1 through heterosexual contact (69 percent), intravenous drug use (30 percent), or blood transfusion (1 percent). The children presented with symptomatic disease at a median age of eight months; only 21 percent presented after the age of two years. The most common first manifestations of disease were lymphoid interstitial pneumonia (in 17 percent), encephalopathy (in 12 percent), recurrent bacterial infections (in 10 percent), and candida esophagitis (in 8 percent), for which the median survival times from diagnosis were 72, 11, 50, and 12 months, respectively. Nine percent of the children had Pneumocystis carinii pneumonia at a median age of five months and had a median survival of only one month. The median survival for all 172 children was 38 months from the time of diagnosis. Mortality was highest in the first year of life (17 percent), and by proportional-hazard analysis the probability of long-term survival is low. In multivariate analyses, early age at diagnosis and the first identifiable pattern of clinical disease were found to be independently related to survival. We conclude that children with perinatally acquired HIV-1 infection have a very poor prognosis and that most become symptomatic before one year of age. Early diagnosis is important, since there is only a short interval in which to initiate prophylactic or antiviral treatment before progressive disease begins.

Cancer among Haitians in Florida.

Data on cancer rates from West Indian populations are scarce, and to the authors' knowledge there are no published data on cancer rates and distributions among Haitians. Proportional distributions of cancers among three groups of patients living in Florida were compared: Haitian born blacks, United States born blacks, and non-Haitian Caribbean born blacks. The incidence rate of cancer of the cervix among the Haitian and United States born black groups was also compared. Increased rates of certain malignancies associated with viral infection or immunodeficiency were found in the Haitian group. These tumors were hepatocellular carcinoma, nasopharyngeal carcinoma, reticulum cell sarcoma, Kaposi's sarcoma, and carcinoma of the uterine cervix. The age-adjusted incidence rate of carcinoma of the cervix was especially high among Haitian women even with a liberal estimate of the female Haitian population from whom the cases were drawn. Except for cancer of the cervix, the numbers of cancers of interest were small, and age-adjusted incidence rates were not calculated. Continued epidemiological study of larger numbers of patients is needed to evaluate these findings further.

Estrogen replacement therapy and benign breast disease.

The relationship between estrogen replacement therapy (ERT) and the risk of benign breast disease (BBD) was examined among 929 postmenopausal cases and 846 postmenopausal controls identified through a large breast cancer screening program. Prior use of ERT among postmenopausal women was associated with an increased risk of both fibrocystic breast disease [odds ratio (OR) = 1.4; 95% confidence interval (Cl) = 1.1-1.8] and fibroadenoma (OR = 1.6; Cl = 0.8-3.5). The risk of all BBD rose with increasing years of use (OR = 1.9 for greater than or equal to 15 yr of total use) and with years since initial use (OR = 1.6 for greater than or equal to 15 yr since first use). Among users of the conjugated estrogen Premarin, increased risks were found at all but the lowest dose. Finally, risks associated with ERT were higher in women with a bilateral oophorectomy than in other postmenopausal women. These results suggested, overall, an increased risk of BBD associated with ERT.

Age at first birth and breast atypia.

The relation between age at first birth and the occurrence of fibrocystic breast disease of different atypia subtypes was evaluated using data from a cohort study of the health effects of oral-contraceptive use. Multivariate case-control analysis was performed on 218 parous women with biopsied fibrocystic disease and 928 parous controls. Fibrocystic breast disease, as a whole and as atypia subtypes, was found to be unrelated to age at first birth. Furthermore, the estimated effect of age at first birth did not vary from one atypia subtype to another. Women with high parity were found to be at decreased risk of fibrocystic breast disease compared to those with low parity. Age at completion of education, used as an indicator of socioeconomic status, was found to be moderately associated with fibrocystic breast disease. The estimated effect of parity and age at completion of education also did not vary among subtypes. More epidemiologic studies specific to histopathologic classifications of benign breast disease are called for.

Oral contraceptive use and fibrocystic breast disease of different histologic classifications.

The relationship between oral contraceptive (OC) use and occurrence of fibrocystic breast disease (FBD) of different histologic classifications was evaluated with data from a cohort study. Biopsy specimens from 232 women with FBD were classified into different atypia categories. In 96 matched pairs of OC users and nonusers, atypia scores were lower in users than in nonusers. Women without breast diseases (500 OC users and 500 nonusers) were sampled from the original cohort to form a two-stage "anamorphic" study with the 232 cases of FBD. The previously shown inverse association between OC use and FBD occurrence was present and increased with increased length of OC use. However, the "protective effect" of OC use did not vary for different histologic classifications of FBD. The findings from both paired and anamorphic analyses of the data are not consistent with the hypothesis that the use of OC is associated with decreased frequency only of FBD with minimal epithelial atypia.

A prospective cohort study of oral contraceptives and cancer of the endometrium.

A prospective cohort study of approximately 97 300 women was conducted in eastern Massachusetts. Information on use of oral contraceptives, and on gynaecological and reproductive histories was sought through mailed questionnaires, while newly diagnosed cases of cancer of the endometrium were identified from records of 34 area hospitals. Oral contraceptive (OC) use was associated with some increase in endometrial cancer rates (standardized rate ratio = 1.4; 95% confidence interval = 0.9-2.4), although the risk did not vary consistently by length of use, years since first use, or age at initial use. Although some subgroups of users had rates of disease higher than among nonusers, there was little to suggest a consistent association between OC use and incidence of endometrial cancer.

PIP: A prospective cohort study of approximately 97,300 women was conducted in eastern Massachusetts. Information on the use of oral contraceptives (OCs), and on gynecological and reproductive histories was sought through mailed questionnaires, while newly diagnosed cases of endometrial cancer were identified from records of 34 area hospitals. OC use was associated with some increase in endometrial cancer rates (standardized ratio rate-1.4; 95% confidence interval=0.9-2.4), although the risk did not vary consistently by length of use, years since 1st use, or age at initial use. Although some subgroups of users had rates of disease higher than among nonusers, there was little to suggest a consistent association between OC use and incidence of endometrial cancer.

Age at first birth, parity, and breast cancer risk.

The roles of age at first full-term birth and parity as risk factors for breast cancer were examined as part of a large prospective cohort study of oral contraceptive use. Compared to women who first gave birth before age 20 years, women with a first birth between age 20 and 24 years had an SRR of 1.70, and women with first births between age 25 and 29 years and over age 29 had SRRs of 2.19 and 3.18, respectively. Increased incidence rates of breast cancer fell with increasing parity, although standardization for age at first birth diminished the importance of the former as an independent indicator of breast cancer risk.

A prospective cohort study of oral contraceptives and breast cancer.

In 1970 a questionnaire was sent to 97,300 married women aged 25-50 years residing in eastern Massachusetts information was requested on use of oral contraceptives (OC), education, and gynecologic and reproductive histories. Subsequent questionnaires were sent in 1973 and 1979 to a subset of the original cohort (37,000) to obtain recent information. Surveillance of 34 hospitals disclosed 622 cases of breast cancer among respondents diagnosed in a 6 1/2-year period. The rate ratio of breast cancer associated with any use of OC was 0.84 (95% confidence limits: 0.7, 1.1). Although certain subgroups users experienced significantly lower rates of hospitalization for breast cancer than did nonusers (including users aged 25-29, those for whom less than or equal to 5 yr had passed since first use, women with ages at birth of first child either before 25 or after 30, and women who first used OC after the birth of the first child), the risk of breast cancer OC after the birth of the first child), the risk of breast cancer was not consistently lower among users of OC. Among nulliparous women, the rate ratio associated with use of OC was 2.1 (0.9, 5.0).