American Journal of Epidemiology update july, 18, 2008
The Incidence of Esophageal Cancer and High-Grade Dysplasia in Barrett’s Esophagus: A Systematic Review and Meta-Analysis
Barrett’s esophagus is a well-recognized precursor of esophageal adenocarcinoma. Surveillance of Barrett’s esophagus patients is recommended to detect high-grade dysplasia (HGD) or early cancer. Because of wide variation in the published cancer incidence in Barrett’s esophagus, the authors undertook a systematic review and meta-analysis of cancer and HGD incidence in Barrett’s esophagus. Ovid Medline (Ovid Technologies, Inc., New York, New York) and EMBASE (Elsevier, Amsterdam, the Netherlands) databases were searched for papers published between 1950 and 2006 that reported the cancer/HGD risk in Barrett’s esophagus. Where possible, early incident cancers/HGD were excluded, as were patients with HGD at baseline. Forty-seven studies were included in the main analysis, and the pooled estimate for cancer incidence in Barrett’s esophagus was 6.1/1,000 person-years, 5.3/1,000 person-years when early incident cancers were excluded, and 4.1/1,000 person-years when both early incident cancer and HGD at baseline were excluded. Corresponding figures for combined HGD/cancer incidence were 10.0 person-years, 9.3 person-years, and 9.1/1,000 person-years. Compared with women, men progressed to cancer at twice the rate. Cancer or HGD/cancer incidences were lower when only high-quality studies were analyzed (3.9/1,000 person-years and 7.7/1,000 person-years, respectively). The pooled estimates of cancer and HGD incidence were low, suggesting that the cost-effectiveness of surveillance is questionable unless it can be targeted to those with the highest cancer risk.
Statin Use and Risk of Prostate Cancer: Results from a Population-based Epidemiologic Study
Epidemiologic studies of statin use in relation to prostate cancer risk have been inconclusive. Recent evidence, however, suggests that longer-term use may reduce risk of more advanced disease. The authors conducted a population-based study of 1,001 incident prostate cancer cases diagnosed in 2002–2005 and 942 age-matched controls from King County, Washington, to evaluate risk associated with statin use. Logistic regression was used to generate odds ratios for ever use, current use, and duration of use. No overall association was found between statin use and prostate cancer risk (odds ratio (OR) = 1.0, 95% confidence interval (CI): 0.8, 1.2 for current use; OR = 1.1, 95% CI: 0.7, 1.8 for >10 years’ use), even for cases with more advanced disease. Risk related to statin use, however, was modified by body mass index (interaction p = 0.04). Obese men (BMI ?30 kg/m2) who used statins had an increased risk (OR = 1.5, 95% CI: 1.0, 2.2) relative to obese nonusers, with a stronger association for longer-term use (OR = 1.8, 95% CI: 1.1, 3.0 for ?5 years’ use). Although statin use was not associated with overall prostate cancer risk, the finding of an increased risk associated with statin use among obese men, particularly use for extended durations, warrants further investigation.
Benign Breast Disease among First-Degree Relatives of Young Breast Cancer Patients
Benign breast disease is associated with increased risk of breast cancer. To further clarify whether there is a genetic link between benign and malignant breast lesions, the authors identified 14,648 first-degree female relatives of 8,807 patients in whom breast cancer was diagnosed at <50 years of age by using Danish nationwide cancer and population registers. Hospital register data were used to follow the relatives for occurrence of benign breast disease from 1977 to 2003 and to calculate rates of benign breast disease in the general population of Danish women for comparison. Risk for relatives was increased for benign breast diseases (observed/expected ratio = 1.54, 95% confidence interval: 1.42, 1.66), particularly for relatives aged <40 years. Higher risks were observed after breast cancer had been diagnosed in the family; however, an increased risk for relatives aged <50 years (observed/expected ratio = 1.24, 95% confidence interval: 1.02, 1.51) was present before breast cancer was diagnosed in the family. Enhanced surveillance of close relatives of breast cancer patients seems to be an important factor to address when investigating the association between benign breast disease and familial breast cancer. A genetic link between benign breast disease and breast cancer was indicated by our data but needs to be confirmed in future studies.
Body Size and Renal Cell Cancer Incidence in a Large US Cohort Study
Renal cell cancer (RCC) incidence has increased in the United States over the past three decades. The authors analyzed the association between body mass index (BMI) and invasive RCC in the National Institutes of Health (NIH)–AARP Diet and Health Study, a large, prospective cohort aged 50–71 years at baseline initiated in 1995–1996, with follow-up through December 2003. Detailed analyses were conducted in a subcohort responding to a second questionnaire, including BMI at younger ages (18, 35, and 50 years); weight change across three consecutive age intervals; waist, hip, and waist-to-hip ratio; and height at age 18 years. Incident RCC was diagnosed in 1,022 men and 344 women. RCC was positively and strongly related to BMI at study baseline. Among subjects analyzed in the subcohort, RCC associations were strongest for baseline BMI and BMI recalled at age 50 years and were successively attenuated for BMI recalled at ages 35 and 18 years. Weight gain in early (18–35 years of age) and mid- (35–50 years of age) adulthood was strongly associated with RCC, whereas weight gain after midlife (age 50 years to baseline) was unrelated. Waist-to hip ratio was positively associated with RCC in women and with height at age 18 years in both men and women.
Non-Hodgkin Lymphoma in Women: Reproductive Factors and Exogenous Hormone Use
Few studies of reproductive hormone exposures and non-Hodgkin lymphoma (NHL) have examined NHL subtypes. Associations between reproductive hormonal factors and risk of all NHL and of two predominant subtypes, diffuse large-cell lymphoma (DLCL) (n = 233) and follicular lymphoma (n = 173), were investigated among women (n = 581) in a large, population-based, case-control study (1,591 cases, 2,515 controls). Controls (n = 836) identified by random digit dialing were frequency matched by age and county to incident NHL cases ascertained in the San Francisco Bay Area of California in 1988–1993. Adjusted unconditional logistic regression was used to obtain odds ratios. More than four pregnancies indicated a possible lower risk of all NHL (odds ratio (OR) = 0.81, 95% confidence interval (CI): 0.55, 1.2; p-trend = 0.06) and of DLCL (OR = 0.53, 95% CI: 0.31, 0.90; p-trend = 0.01). Exclusive use of menopausal hormone therapy for ?5 years was associated with a reduced risk of all NHL (OR = 0.68, 95% CI: 0.48, 0.98) and of DLCL (OR = 0.50, 95% CI: 0.30, 0.85). Oral contraceptive use indicated a lower risk of all NHL (OR = 0.68, 95% CI: 0.49, 0.94), and perhaps DLCL (OR = 0.79, 95% CI: 0.51, 1.2), and of follicular lymphoma (OR = 0.75, 95% CI: 0.46, 1.2). Results suggest that endogenous and exogenous reproductive hormones confer different risks by NHL subtype and are associated with a reduced risk of DLCL in women.
Consumption of trans-Fatty Acid and Its Association with Colorectal Adenomas
trans-Fatty acid consumption is known to have detrimental effects on cardiovascular health, but little is known about its role in digestive tract neoplasia. To investigate the association between colorectal adenomas and trans-fatty acid consumption, the authors utilized data from a cross-sectional study of 622 individuals who underwent complete colonoscopy between 2001 and 2002 at the University of North Carolina Hospitals. Participants were interviewed about demographic, lifestyle, and dietary factors thought to be related to colorectal cancer. trans-Fatty acid consumption, energy adjusted by the residual method, was categorized into quartiles based on its distribution in controls. Compared with participants in the lowest quartile of consumption, those in the highest quartile had an increased prevalence of colorectal adenomas, with an adjusted prevalence odds ratio of 1.86 (95% confidence interval: 1.04, 3.33). The authors further investigated the relation between trans-fatty acid consumption and colorectal neoplasia by examining the adenoma characteristics, with the adjusted prevalence odds ratios showing little or no difference by adenoma location, size, or number. These results suggest that consumption of high amounts of trans-fatty acid may increase the risk of colorectal neoplasia, and they provide additional support to recommendations to limit trans-fatty acid consumption.
Effects of Exposure to Polychlorinated Biphenyls and Organochlorine Pesticides on Thyroid Function during Pregnancy
In this study, the authors’ objective was to determine whether serum concentrations of polychlorinated biphenyls (PCBs), hexachlorobenzene, p,p‘-dichlorodiphenyl trichloroethane (DDT), o,p‘-DDT, and p,p‘-dichlorodiphenyl dichloroethylene (DDE) are associated with thyroid function during pregnancy. These compounds, as well as thyroid-stimulating hormone, total thyroxine, and free thyroxine, were measured in serum samples collected between October 1999 and October 2000 from 334 pregnant women living in the Salinas Valley, California. Data were analyzed by multivariate linear regression. After adjustment for covariates, seven of the 19 PCB congeners detected in more than 75% of participants and the sum of those congeners were negatively associated with free thyroxine concentrations. PCBs 44, 52, and 183 remained significant after the exclusion of two outliers. Hexachlorobenzene concentrations were negatively associated with both free thyroxine and total thyroxine. PCB and hexachlorobenzene concentrations were strongly correlated, which hampered the authors’ ability to identify their independent associations with thyroid function. None of the exposures under study were associated with thyroid-stimulating hormone. Results suggest that exposure to PCBs and/or hexachlorobenzene at background levels may affect thyroid function during pregnancy. These findings are of particular significance, since thyroid hormones of maternal origin may play an essential role in fetal neurodevelopment.
Early Childhood Infection by Human Herpesvirus 8 in Zambia and the Role of Human Immunodeficiency Virus Type 1 Coinfection in a Highly Endemic Area
Kaposi’s sarcoma occurs at high incidence among Zambian adults and children, but there is a paucity of data on human herpesvirus 8 (HHV-8) incidence and routes of infection, especially in children. Between 1998 and 2004, the authors conducted a prospective study of viral transmission in a cohort of 684 children in Lusaka, Zambia, to estimate the annual incidence of HHV-8 from birth through 48 months of age. Maternal and pediatric human immunodeficiency virus type 1 (HIV-1) infection status was also determined. The results, based on 1,532 child-years of follow-up, showed that HHV-8 seroconversion occurs early in life. The incidence rate of HHV-8 seroconversion was 13.8 infections per 100 child-years by 48 months of age. HIV-1-infected children were at substantially higher risk for HHV-8 seroconversion (adjusted hazard ratio = 4.60, 95% confidence interval: 2.93, 7.22). Maternal HIV-1 and HHV-8 infection status were not independently associated with risk of HHV-8 seroconversion in the child. HHV-8 antibody titers in children followed at all consecutive time points revealed seroreversion of HHV-8 antibodies, with undetectable titers in some children at one or more time points after seroconversion. These results demonstrate that cross-sectional serologic screening probably underestimates true HHV-8 seroprevalence in young Zambian children because of fluctuations in detectable antibody titers.
A Simple Screening Approach for Assessing Community Prevalence and Phenotype of Polycystic Ovary Syndrome in a Semiurban Population in Sri Lanka
In most of South Asia, prevalences and phenotypes of polycystic ovary syndrome (PCOS) among women in the community are unknown. The authors aimed to estimate prevalence and phenotype in a community setting in Sri Lanka and to test a valid, feasible screening approach to early diagnosis. A community-based, cross-sectional study was carried out in 2005–2006. A random sample of 3,030 women aged 15–39 years was selected by cluster sampling proportionate to population size. An interviewer-administered questionnaire was utilized to screen for “probable cases” of PCOS based on menstrual history and clinical manifestations of hyperandrogenism. Selected “probable cases” underwent clinical, biochemical, and ovarian ultrasound assessment. The response rate was 96.2% (n = 2,915). A total of 220 (7.5%) “probable cases” were identified: 209 women with oligo/amenorrhea (95%) and 11 women with hirsutism (5%). Further evaluation of the 220 probable cases confirmed 164 newly diagnosed cases of PCOS based on the 2003 Rotterdam diagnostic criteria. With 19 previously diagnosed cases already present, total prevalence was 6.3% (95% confidence interval: 5.9, 6.8). Of the women with “oligo/amenorrhea and/or hirsutism,” 91.1% were confirmed to have PCOS; 99.4% of women with “regular cycles in the absence of clinical hyperandrogenism” were confirmed as normal. The most common phenotypes of PCOS were oligo/amenorrhea and polycystic ovaries (91.4%) and oligo/amenorrhea and hirsutism (48.3%).
Immeasurable Time Bias in Observational Studies of Drug Effects on Mortality
Observational studies suggesting that some drugs are effective at reducing mortality may have been subject to “immeasurable time bias” arising from the unidentified presence of hospitalizations when defining drug exposure with computerized health databases. The author illustrates the bias using a case-control study of 1,313 deaths and 1,313 controls selected from a cohort of 2,049 patients with chronic obstructive pulmonary disease from Saskatchewan, Canada, identified from 1990 and followed up through 1999. Different approaches were used to estimate the rate ratio of death associated with inhaled corticosteroid exposure, defined by a prescription dispensed in the 30-day period prior to the index date. More cases had been hospitalized during the 30-day exposure period (72%) than controls (26%), with lower durations of stay for cases who received an inhaled corticosteroid prescription (9.9 vs.16.2 days), thus introducing variations in measurable exposure times. The raw analysis that did not consider hospitalization found a rate ratio of 0.60 (95% confidence interval (CI): 0.50, 0.73). Alternatively, analyses accounting for variations in measurable times resulted in a rate ratio of 0.93 (95% CI: 0.76, 1.14) when weighted by measurable time, while use of the Kaplan-Meier estimator of the 30-day cumulative incidence of exposure found a rate ratio of 1.35 (95% CI: 1.14, 1.60). In conclusion, immeasurable time bias may be present in several observational database studies suggesting that certain drugs are effective at reducing mortality.
The Public-Use National Health Interview Survey Linked Mortality Files: Methods of Reidentification Risk Avoidance and Comparative Analysis
The National Center for Health Statistics (NCHS) conducts mortality follow-up for its major population-based surveys. In 2004, NCHS updated the mortality follow-up for the 1986–2000 National Health Interview Survey (NHIS) years, which because of confidentiality protections was made available only through the NCHS Research Data Center. In 2007, NCHS released a public-use version of the NHIS Linked Mortality Files that includes a limited amount of perturbed information for decedents. The modification of the public-use version included conducting a reidentification risk scenario to determine records at risk for reidentification and then imputing values for either date or cause of death for a select sample of records. To demonstrate the comparability between the public-use and restricted-use versions of the linked mortality files, the authors estimated relative hazards for all-cause and cause-specific mortality risk using a Cox proportional hazards model. The pooled 1986–2000 NHIS Linked Mortality Files contain 1,576,171 records and 120,765 deaths. The sample for the comparative analyses included 897,232 records and 114,264 deaths. The comparative analyses show that the two data files yield very similar results for both all-cause and cause-specific mortality. Analytical considerations when examining cause-specific analyses of numerically small demographic subgroups are addressed.
Testing Language Effects in Psychiatric Epidemiology Surveys with Randomized Experiments: Results from the National Latino and Asian American Study
To evaluate the prevalence of mental disorders for persons of non-English-language origin, it is essential to use translated diagnostic interviews. The equivalence of translated surveys is rarely tested formally. In the National Latino and Asian American Study (NLAAS), the authors tested whether a carefully translated mental health survey administered in Spanish produced results equivalent to those obtained by the original English version, using a randomized survey experiment. The NLAAS is a nationally representative survey carried out in the United States in 2002–2003. Bilingual respondents from the Latino section of the NLAAS (n = 332) were randomly assigned to receive either a Spanish- or English-language version of the World Mental Health Survey Composite International Diagnostic Interview. In tests of differences in lifetime and 12-month prevalences of 11 diagnoses and four higher-order aggregate disorder categories, in only one case was there an apparent difference between randomized language groups: Lifetime reports of generalized anxiety disorder were more prevalent in the bilingual group assigned to English than in the group interviewed in Spanish. Detailed follow-up analyses did not implicate any specific question in the generalized anxiety disorder protocol. Translation and back-translation of surveys does not guarantee that response probabilities are exactly equivalent. Randomized survey experiments should be incorporated into cross-cultural psychiatric surveys when possible.
Leave a Reply