Showing papers in "American Journal of Epidemiology in 2003"

Socioeconomic Inequalities in Depression: A Meta-Analysis

Abstract: Low socioeconomic status (SES) is generally associated with high psychiatric morbidity, more disability, and poorer access to health care. Among psychiatric disorders, depression exhibits a more controversial association with SES. The authors carried out a meta-analysis to evaluate the magnitude, shape, and modifiers of such an association. The search found 51 prevalence studies, five incidence studies, and four persistence studies meeting the criteria. A random effects model was applied to the odds ratio of the lowest SES group compared with the highest, and meta-regression was used to assess the dose-response relation and the influence of covariates. Results indicated that low-SES individuals had higher odds of being depressed (odds ratio = 1.81, p < 0.001), but the odds of a new episode (odds ratio = 1.24, p = 0.004) were lower than the odds of persisting depression (odds ratio = 2.06, p < 0.001). A dose-response relation was observed for education and income. Socioeconomic inequality in depression is heterogeneous and varies according to the way psychiatric disorder is measured, to the definition and measurement of SES, and to contextual features such as region and time. Nonetheless, the authors found compelling evidence for socioeconomic inequality in depression. Strategies for tackling inequality in depression are needed, especially in relation to the course of the disorder.

Statistical Analysis of Correlated Data Using Generalized Estimating Equations: An Orientation

Abstract: The method of generalized estimating equations (GEE) is often used to analyze longitudinal and other correlated response data, particularly if responses are binary. However, few descriptions of the method are accessible to epidemiologists. In this paper, the authors use small worked examples and one real data set, involving both binary and quantitative response data, to help end-users appreciate the essence of the method. The examples are simple enough to see the behind-the-scenes calculations and the essential role of weighted observations, and they allow nonstatisticians to imagine the calculations involved when the GEE method is applied to more complex multivariate data.

Estimating the Relative Risk in Cohort Studies and Clinical Trials of Common Outcomes

Abstract: Logistic regression yields an adjusted odds ratio that approximates the adjusted relative risk when disease incidence is rare (<10%), while adjusting for potential confounders. For more common outcomes, the odds ratio always overstates the relative risk, sometimes dramatically. The purpose of this paper is to discuss the incorrect application of a proposed method to estimate an adjusted relative risk from an adjusted odds ratio, which has quickly gained popularity in medical and public health research, and to describe alternative statistical methods for estimating an adjusted relative risk when the outcome is common. Hypothetical data are used to illustrate statistical methods with readily accessible computer software.

Incidence of Parkinson’s Disease: Variation by Age, Gender, and Race/Ethnicity

Abstract: The goal of this study was to estimate the incidence of Parkinson's disease by age, gender, and ethnicity. Newly diagnosed Parkinson's disease cases in 1994-1995 were identified among members of the Kaiser Permanente Medical Care Program of Northern California, a large health maintenance organization. Each case met modified standardized criteria/Hughes diagnostic criteria as applied by a movement disorder specialist. Incidence rates per 100,000 person-years were calculated using the Kaiser Permanente membership information as the denominator and adjusted for age and/or gender using the direct method of standardization. A total of 588 newly diagnosed (incident) cases of Parkinson's disease were identified, which gave an overall annualized age- and gender-adjusted incidence rate of 13.4 per 100,000 (95% confidence interval (CI): 11.4, 15.5). The incidence rapidly increased over the age of 60 years, with only 4% of the cases being under the age of 50 years. The rate for men (19.0 per 100,000, 95% CI: 16.1, 21.8) was 91% higher than that for women (9.9 per 100,000, 95% CI: 7.6, 12.2). The age- and gender-adjusted rate per 100,000 was highest among Hispanics (16.6, 95% CI: 12.0, 21.3), followed by non-Hispanic Whites (13.6, 95% CI: 11.5, 15.7), Asians (11.3, 95% CI: 7.2, 15.3), and Blacks (10.2, 95% CI: 6.4, 14.0). These data suggest that the incidence of Parkinson's disease varies by race/ethnicity.

Evaluating Medication Effects Outside of Clinical Trials: New-User Designs

Abstract: Recent clinical trials demonstrating that hormone replacement therapy (HRT) does not prevent coronary heart disease in women have again raised doubts concerning observational studies. Although much of the explanation probably lies in what might be called the "healthy HRT user" effect, another contributing factor may be that most observational studies included many prevalent users: women taking HRT for some time before study follow-up began. This practice can cause two types of bias, both of which plausibly may have contributed to the discrepancy between observational and randomized studies. First, prevalent users are "survivors" of the early period of pharmacotherapy, which can introduce substantial bias if risk varies with time, just as in studies of operative procedures that enroll patients after they have survived surgery. This article provides several examples of medications for which the hazard function varies with time and thus would be subject to prevalent user bias. Second, covariates for drug users at study entry often are plausibly affected by the drug itself. Investigators often do not adjust for these factors on the causal pathway, which may introduce confounding. A new-user design eliminates these biases by restricting the analysis to persons under observation at the start of the current course of treatment. This article thus argues that such designs should be used more frequently in pharmacoepidemiology.

Genital Human Papillomavirus Infection: Incidence and Risk Factors in a Cohort of Female University Students

Abstract: Incidence data on human papillomavirus (HPV) infection are limited, and risk factors for transmission are largely unknown. The authors followed 603 female university students in Washington State at 4-month intervals between 1990 and 2000. At each visit, a sexual and health questionnaire was completed and cervical and vulvovaginal samples were collected to detect HPV DNA. At 24 months, the cumulative incidence of first-time infection was 32.3% (95% confidence interval: 28.0, 37.1). Incidences calculated from time of new-partner acquisition were comparable for enrolled virgins and nonvirgins. Smoking, oral contraceptive use, and report of a new male sex partner--in particular, one known for less than 8 months before sex occurred or one reporting other partners--were predictive of incident infection. Always using male condoms with a new partner was not protective. Infection in virgins was rare, but any type of nonpenetrative sexual contact was associated with an increased risk. Detection of oral HPV was rare and was not associated with oral-penile contact. The data show that the incidence of HPV associated with acquisition of a new sex partner is high and that nonpenetrative sexual contact is a plausible route of transmission in virgins.

Using Intake Biomarkers to Evaluate the Extent of Dietary Misreporting in a Large Sample of Adults: The OPEN Study

Abstract: This paper describes the Observing Protein and Energy Nutrition (OPEN) Study, conducted from September 1999 to March 2000. The purpose of the study was to assess dietary measurement error using two self-reported dietary instruments-the food frequency questionnaire (FFQ) and the 24-hour dietary recall (24HR)-and unbiased biomarkers of energy and protein intakes: doubly labeled water and urinary nitrogen. Participants were 484 men and women aged 40-69 years from Montgomery County, Maryland. Nine percent of men and 7% of women were defined as underreporters of both energy and protein intake on 24HRs; for FFQs, the comparable values were 35% for men and 23% for women. On average, men underreported energy intake compared with total energy expenditure by 12-14% on 24HRs and 31-36% on FFQs and underreported protein intake compared with a protein biomarker by 11-12% on 24HRs and 30-34% on FFQs. Women underreported energy intake on 24HRs by 16-20% and on FFQs by 34-38% and underreported protein intake by 11-15% on 24HRs and 27-32% on FFQs. There was little underreporting of the percentage of energy from protein for men or women. These findings have important implications for nutritional epidemiology and dietary surveillance.

Is Obesity Associated with Major Depression? Results from the Third National Health and Nutrition Examination Survey

Abstract: Data from the Third National Health and Nutrition Examination Survey (1988-1994) were used to examine the relation between obesity and depression. Past-month depression was defined using criteria from the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, and was measured with the Diagnostic Interview Schedule. Obesity was defined as a body mass index (weight (kg)/height (m) 2 ) of 30 or higher. The authors compared risks of depression in obese and normal-weight (body mass index 18.5-24.9) persons. Obesity was associated with past-month depression in women (odds ratio (OR) = 1.82, 95% confidence interval (Cl): 1.01, 3.3) but was not significantly associated in men (OR = 1.73, 95% Cl: 0.56, 5.37). When obesity was stratified by severity, heterogeneity in the association with depression was observed. Class 3 (severe) obesity (body mass index ≥40) was associated with past-month depression in unadjusted analyses (OR = 4.98, 95% Cl: 2.07, 11.99); the association remained strong after results were controlled for age, education, marital status, physician's health rating, dieting for medical reasons, use of psychiatric medicines, cigarette smoking, and use of alcohol, marijuana, and cocaine. These findings suggest that obesity is associated with depression mainly among persons with severe obesity. Prospective studies will be necessary to clarify the obesity-depression relation but await the identification of potential risk factors for depression in the obese.

Maternal Stress and Preterm Birth

Abstract: This study examined a comprehensive array of psychosocial factors, including life events, social support, depression, pregnancy-related anxiety, perceived discrimination, and neighborhood safety in relation to preterm birth (<37 weeks) in a prospective cohort study of 1,962 pregnant women in central North Carolina between 1996 and 2000, in which 12% delivered preterm. There was an increased risk of preterm birth among women with high counts of pregnancy-related anxiety (risk ratio (RR) = 2.1, 95% confidence interval (CI): 1.5, 3.0), with life events to which the respondent assigned a negative impact weight (RR = 1.8, 95% CI: 1.2, 2.7), and with a perception of racial discrimination (RR = 1.4, 95% CI: 1.0, 2.0). Different levels of social support or depression were not associated with preterm birth. Preterm birth initiated by labor or ruptured membranes was associated with pregnancy-related anxiety among women assigning a high level of negative impact weights (RR = 3.0, 95% CI: 1.7, 5.3). The association between high levels of pregnancy-related anxiety and preterm birth was reduced when restricted to women without medical comorbidities, but the association was not eliminated. The prospective collection of multiple psychosocial measures on a large population of women indicates that a subset of these factors is associated with preterm birth.

Structure of Dietary Measurement Error: Results of the OPEN Biomarker Study

Abstract: Multiple-day food records or 24-hour dietary recalls (24HRs) are commonly used as "reference" instruments to calibrate food frequency questionnaires (FFQs) and to adjust findings from nutritional epidemiologic studies for measurement error. Correct adjustment requires that the errors in the adopted reference instrument be independent of those in the FFQ and of true intake. The authors report data from the Observing Protein and Energy Nutrition (OPEN) Study, conducted from September 1999 to March 2000, in which valid reference biomarkers for energy (doubly labeled water) and protein (urinary nitrogen), together with a FFQ and 24HR, were observed in 484 healthy volunteers from Montgomery County, Maryland. Accounting for the reference biomarkers, the data suggest that the FFQ leads to severe attenuation in estimated disease relative risks for absolute protein or energy intake (a true relative risk of 2 would appear as 1.1 or smaller). For protein adjusted for energy intake by using either nutrient density or nutrient residuals, the attenuation is less severe (a relative risk of 2 would appear as approximately 1.3), lending weight to the use of energy adjustment. Using the 24HR as a reference instrument can seriously underestimate true attenuation (up to 60% for energy-adjusted protein). Results suggest that the interpretation of findings from FFQ-based epidemiologic studies of diet-disease associations needs to be reevaluated.

Comparison of Logistic Regression versus Propensity Score When the Number of Events Is Low and There Are Multiple Confounders

Abstract: The aim of this study was to use Monte Carlo simulations to compare logistic regression with propensity scores in terms of bias, precision, empirical coverage probability, empirical power, and robustness when the number of events is low relative to the number of confounders. The authors simulated a cohort study and performed 252,480 trials. In the logistic regression, the bias decreased as the number of events per confounder increased. In the propensity score, the bias decreased as the strength of the association of the exposure with the outcome increased. Propensity scores produced estimates that were less biased, more robust, and more precise than the logistic regression estimates when there were seven or fewer events per confounder. The logistic regression empirical coverage probability increased as the number of events per confounder increased. The propensity score empirical coverage probability decreased after eight or more events per confounder. Overall, the propensity score exhibited more empirical power than logistic regression. Propensity scores are a good alternative to control for imbalances when there are seven or fewer events per confounder; however, empirical power could range from 35% to 60%. Logistic regression is the technique of choice when there are at least eight events per confounder.

Association between Eating Patterns and Obesity in a Free-living US Adult Population

Abstract: Some studies have suggested that eating patterns, which describe eating frequency, the temporal distribution of eating events across the day, breakfast skipping, and the frequency of eating meals away from home, may be related to obesity. Data from the Seasonal Variation of Blood Cholesterol Study (1994-1998) were used to evaluate the relation between eating patterns and obesity. Three 24-hour dietary recalls and a body weight measurement were collected at five equally spaced time points over a 1-year period from 499 participants. Data were averaged for five time periods, and a cross-sectional analysis was conducted. Odds ratios were adjusted for other obesity risk factors including age, sex, physical activity, and total energy intake. Results indicate that a greater number of eating episodes each day was associated with a lower risk of obesity (odds ratio for four or more eating episodes vs. three or fewer = 0.55, 95% confidence interval: 0.33, 0.91). In contrast, skipping breakfast was associated with increased prevalence of obesity (odds ratio = 4.5, 95% confidence interval: 1.57, 12.90), as was greater frequency of eating breakfast or dinner away from home. Further investigation of these associations in prospective studies is warranted.

Trends of Probable Post-Traumatic Stress Disorder in New York City after the September 11 Terrorist Attacks

Abstract: The authors investigated trends in probable post-traumatic stress disorder (PTSD) prevalence in the general population of New York City in the first 6 months after the September 11 terrorist attacks. Three random digit dialing telephone surveys of adults in progressively larger portions of the New York City metropolitan area were conducted 1 month, 4 months, and 6 months after September 11, 2001. A total of 1,008, 2,001, and 2,752 demographically representative adults were recruited in the three surveys, respectively. The current prevalence of probable PTSD related to the September 11 attacks in Manhattan declined from 7.5% (95% confidence interval: 5.7, 9.3) 1 month after September 11 to 0.6% (95% confidence interval: 0.3, 0.9) 6 months after September 11. Although the prevalence of PTSD symptoms was consistently higher among persons who were more directly affected by the attacks, a substantial number of persons who were not directly affected by the attacks also met criteria for probable PTSD. These data suggest a rapid resolution of most of the probable PTSD symptoms in the general population of New York City in the first 6 months after the attacks. The psychological consequences of a large-scale disaster in a densely populated urban area may extend beyond persons directly affected by the disaster to persons in the general population.

Modifiers of the Temperature and Mortality Association in Seven US Cities

Abstract: This paper examines effect modification of heat- and cold-related mortality in seven US cities in 1986–1993. City-specific Poisson regression analyses of daily noninjury mortality were fit with predictors of mean daily apparent temperature (a construct reflecting physiologic effects of temperature and humidity), time, barometric pressure, day of the week, and particulate matter less than 10 µm in aerodynamic diameter. Percentage change in mortality was calculated at 29°C apparent temperature (lag 0) and at –5°C (mean of lags 1, 2, and 3) relative to 15°C. Separate models were fit to death counts stratified by age, race, gender, education, and place of death. Effect estimates were combined across cities, treating city as a random effect. Deaths among Blacks compared with Whites, deaths among the less educated, and deaths outside a hospital were more strongly associated with hot and cold temperatures, but gender made no difference. Stronger cold associations were found for those less than age 65 years, but heat effects did not vary by age. The strongest effect modifier was place of death for heat, with out-of-hospital effects more than five times greater than in-hospital deaths, supporting the biologic plausibility of the associations. Place of death, race, and educational attainment indicate vulnerability to temperature-related mortality, reflecting inequities in health impacts related to climate change. climate; education; ethnic groups; heat; mortality; poverty; socioeconomic factors; weather

Post-Traumatic Stress Disorder and Chronic Fatigue Syndrome-like Illness among Gulf War Veterans: A Population-based Survey of 30,000 Veterans

Abstract: The authors estimated the prevalence of post-traumatic stress disorder (PTSD) and illness resembling chronic fatigue syndrome (CFS) in the entire population of Gulf War and non-Gulf-War veterans. They also evaluated the relation between the extent of deployment-related stress and the risk of either PTSD or CFS. In 1995-1997, the authors conducted a health survey in which these two symptom-based medical diagnoses in a population-based sample of 15,000 Gulf War veterans representing four military branches and three unit components (active, reserve, and National Guard) were compared with those of 15,000 non-Gulf veteran controls. Gulf War veterans, compared with non-Gulf veteran controls, reported significantly higher rates of PTSD (adjusted odds ratio = 3.1, 95% confidence interval: 2.7, 3.4) and CFS (adjusted odds ratio = 4.8, 95% confidence interval: 3.9, 5.9). The prevalence of PTSD increased monotonically across six levels of deployment-related stress intensity (test for trend: p < 0.01), while the prevalence of CFS rose only at the low end of the stress spectrum. While deployment-related stress could account for the higher risks of both PTSD and CFS, additional factor(s) unique to the Gulf environment may have contributed to the risk of CFS among Gulf War veterans.

Obesity and the Risk of Newly Diagnosed Asthma in School-age Children

Abstract: To determine the relation between obesity and new-onset asthma among school-age children, the authors examined longitudinal data from 3,792 participants in the Children's Health Study (Southern California) who were asthma-free at enrollment. New cases of physician-diagnosed asthma, height, weight, lung function, and risk factors for asthma were assessed annually at five school visits between 1993 and 1998. Incidence rates were calculated, and proportional hazards regression models were fitted to estimate the adjusted relative risks of new-onset asthma associated with percentile of body mass index (weight (kg)/height (m) 2 ) and indicators of overweight (>85th body mass index percentile) and obesity (>95th body mass index percentile). The risk of new-onset asthma was higher among children who were overweight (relative risk (RR) = 1.52, 95% confidence interval (Cl): 1.14, 2.03) or obese (RR = 1.60, 95% Cl: 1.08, 2.36). Boys had an increased risk associated with being overweight (RR = 2.06, 95% 1.33, 3.18) in comparison with girls (RR = 1.25, 95% Cl: 0.83, 1.88). The effect of being overweight was greater in nonallergic children (RR = 1.77, 95% Cl: 1.26, 2.49) than in allergic children (RR = 1.16, 95% CI: 0.63, 2.15). The authors conclude that being overweight is associated with an increased risk of new-onset asthma in boys and in nonallergic children.

Risk Factors for Ectopic Pregnancy: A Comprehensive Analysis Based on a Large Case-Control, Population-based Study in France

Abstract: This case-control study was associated with a regional register of ectopic pregnancy between 1993 and 2000 in France. It included 803 cases of ectopic pregnancy and 1,683 deliveries and was powerful enough to investigate all ectopic pregnancy risk factors. The main risk factors were infectious history (adjusted attributable risk = 0.33; adjusted odds ratio for previous pelvic infectious disease = 3.4, 95% percent confidence interval (CI): 2.4, 5.0) and smoking (adjusted attributable risk = 0.35; adjusted odds ratio = 3.9, 95% CI: 2.6, 5.9 for >20 cigarettes/day vs. women who had never smoked). The other risk factors were age (associated per se with a risk of ectopic pregnancy), prior spontaneous abortions, history of infertility, and previous use of an intrauterine device. Prior medical induced abortion was associated with a risk of ectopic pregnancy (adjusted odds ratio = 2.8, 95% CI: 1.1, 7.2); no such association was observed for surgical abortion (adjusted odds ratio = 1.1, 95% CI: 0.8, 1.6). The total attributable risk of all the factors investigated was 0.76. As close associations were found between ectopic pregnancy and infertility and between ectopic pregnancy and spontaneous abortion, further research into ectopic pregnancy should focus on risk factors common to these conditions. In terms of public health, increasing awareness of the effects of smoking may be useful for ectopic pregnancy prevention.

Social engagement and disability in a community population of older adults: the New Haven EPESE.

Abstract: This paper examines the effect of social engagement on disability among community-dwelling older adults in 1982-1991. Data were collected from the New Haven, Connecticut, site of the Established Populations for Epidemiologic Studies of the Elderly. Baseline social engagement was measured by using 11 items related to social and productive activity. Disability data consisted of a six-item measure of activities of daily living, a three-item measure of gross mobility, and a four-item measure of basic physical functions. Nine waves of yearly disability data were analyzed by using generalized estimating equations models. After adjustment for age, gender, race, and physical activity, significant cross-sectional associations (p's < 0.001) were found between social engagement and all three measures of disability, with more socially engaged older adults reporting less disability. Social engagement also showed small, but negative interaction effects with follow-up-time outcomes (p's < 0.01), indicating that the protective effect of social engagement decreased slightly during follow-up. Results suggest a strong, but not necessarily causal association of social engagement with disability. Promotion of social engagement may still be important for the prevention of disability.

Use of Agricultural Pesticides and Prostate Cancer Risk in the Agricultural Health Study Cohort

Abstract: The authors examined the relation between 45 common agricultural pesticides and prostate cancer incidence in a prospective cohort study of 55,332 male pesticide applicators from Iowa and North Carolina with no prior history of prostate cancer. Data were collected by means of self-administered questionnaires completed at enrollment (1993-1997). Cancer incidence was determined through population-based cancer registries from enrollment through December 31, 1999. A prostate cancer standardized incidence ratio was computed for the cohort. Odds ratios were computed for individual pesticides and for pesticide use patterns identified by means of factor analysis. A prostate cancer standardized incidence ratio of 1.14 (95% confidence interval: 1.05, 1.24) was observed for the Agricultural Health Study cohort. Use of chlorinated pesticides among applicators over 50 years of age and methyl bromide use were significantly associated with prostate cancer risk. Several other pesticides showed a significantly increased risk of prostate cancer among study subjects with a family history of prostate cancer but not among those with no family history. Important family history-pesticide interactions were observed.

Chronic Venous Disease in an Ethnically Diverse Population The San Diego Population Study

Abstract: In a 1994-1998 cross-sectional study of a multiethnic sample of 2,211 men and women in San Diego, California, the authors estimated prevalence of the major manifestations of chronic venous disease: spider veins, varicose veins, trophic changes, and edema by visual inspection; superficial and deep functional disease (reflux or obstruction) by duplex ultrasonography; and venous thrombotic events based on history. Venous disease increased with age, and, compared with Hispanics, African Americans, and Asians, non-Hispanic Whites had more disease. Spider veins, varicose veins, superficial functional disease, and superficial thrombotic events were more common in women than men (odds ratio (OR) = 5.4, OR = 2.2, OR = 1.9, and OR = 1.9, respectively; p < 0.05), but trophic changes and deep functional disease were less common in women (OR = 0.7 for both; p < 0.05). Visible (varicose veins or trophic changes) and functional (superficial or deep) disease were closely linked; 92.0% of legs were concordant and 8.0% discordant. For legs evidencing both trophic changes and deep functional disease, the age-adjusted prevalences of edema, superficial events, and deep events were 48.2%, 11.3%, and 24.6%, respectively, compared with 1.7%, 0.6%, and 1.3% for legs visibly and functionally normal. However, visible disease did not invariably predict functional disease, or vice versa, and venous thrombotic events occurred in the absence of either.

Invited Commentary: Advancing Theory and Methods for Multilevel Models of Residential Neighborhoods and Health

Abstract: In less than 10 years, the field of epidemiology has been transformed. During this time, multilevel modeling has gone from a little-known and perhaps even unwelcome method of analysis to a household name. Multilevel models appeared much earlier in the fields of demography, sociology of education, and criminology (refer, for example, to Raudenbush et al. (1), Mason et al. (2), and Nuttal et al. (3)), but public health and epidemiology have lagged behind in conceptualizing and measuring how contexts affect individual-level health risks and outcomes. The growth in the publication of multilevel studies has paralleled and contributed greatly to the reemergence of social epidemiologic research (4–11). Multilevel models assessing the effects of neighborhood residential environments on health outcomes have been the most common type of contextual study to date. Research has examined associations between neighborhood characteristics, frequently socioeconomic position, and a variety of health outcomes, including perinatal health (12– 15), mortality (16, 17), health behaviors (18, 19), women’s health (20, 21), heart disease (22–24), disability (25), and child health (26) among others. Furthermore, many reviews have been published concerning the statistical methods and rationale for, as well as the quality of, the existing multilevel studies on health (27–34). Yet, it is perhaps still premature to celebrate the exponential growth of multilevel epidemiologic studies on neighborhood residential effects on health. On the one hand, the growth in publication of these studies reflects increased interest in and support for social epidemiologic approaches. This increased support is encouraging in light of recent controversies about the legitimacy of social epidemiology as a subfield of epidemiology (35–40). On the other hand, these studies have emerged in public health and epidemiology without the appropriate theoretical and methodological foundations to guide their implementation. Hence, despite dozens of multilevel studies of neighborhoods and health, we still lack a clear picture of the intervention and policy implications of this body of work. In this issue of the Journal, Buka et al. (41) examine whether levels of neighborhood social support affect the birth weights of African-American and White mothers residing in Chicago, Illinois. Their paper contributes to the small, but growing literature on multilevel analyses of neighborhood residential effects on perinatal outcomes (12–15, 42). It offers several methodological advancements over previous multilevel studies on neighborhood residence and health. Moreover, it highlights additional theoretical and methodological challenges that multilevel analysis, given its recent appearance in public health, must overcome if significant gains are to be made in understanding disease etiology and designing appropriate interventions and policies to prevent adverse health. These theoretical and methodological issues are the topic of this commentary.

Stressful Life Events and Risk of Breast Cancer in 10,808 Women: A Cohort Study

Abstract: The authors prospectively investigated the relation between stressful life events and risk of breast cancer among 10,808 women from the Finnish Twin Cohort. Life events and breast cancer risk factors were assessed by self-administered questionnaire in 1981. A national modification of a standardized life event inventory was used, examining accumulation of life events and individual life events and placing emphasis on the 5 years preceding completion of the questionnaire. Through record linkage with the Finnish Cancer Registry, 180 incident cases of breast cancer were identified in the cohort between 1982 and 1996. The multivariable adjusted hazard ratio for breast cancer per one-event increase in the total number of life events was 1.07 (95% confidence interval (CI): 1.00, 1.15). This risk estimate rose to 1.35 (95% CI: 1.09, 1.67) when only major life events were taken into account. Independently of total life events, divorce/separation (hazard ratio (HR) = 2.26, 95% CI: 1.25, 4.07), death of a husband (HR = 2.00, 95% CI: 1.03, 3.88), and death of a close relative or friend (HR = 1.36, 95% CI: 1.00, 1.86) were all associated with increased risk of breast cancer. The findings suggest a role for life events in breast cancer etiology through hormonal or other mechanisms.

Determinants of Clearance of Human Papillomavirus Infections in Colombian Women with Normal Cytology: A Population-based, 5-Year Follow-up Study

Abstract: Little is known about the factors that influence clearance of human papillomavirus (HPV), the primary cause of cervical carcinoma. A total of 227 women cytologically normal and HPV positive at baseline were identified from a population-based cohort of 1,995 Bogota, Colombia, women aged 13-85 years followed between 1993 and 2000 (mean follow-up, 5.3 years). HPV DNA detection and viral load determination were based on a GP5+/GP6+ polymerase chain reaction enzyme immunoassay. Rate ratio estimates for HPV clearance were calculated by using methods for interval-censored survival time data. Analyses were based on 316 type-specific HPV infections. HPV 16 had a significantly lower clearance rate than infections with low-risk types (rate ratio (RR) = 0.47, 95% confidence interval (CI): 0.32, 0.72), HPV types related to HPV 16 (types 31, 33, 35, 52, 58) had intermediate clearance rates (RR = 0.62, 95% CI: 0.47, 0.94), and other high-risk types did not show evidence of slower clearance compared with low-risk types. Infections with single and multiple HPV types had similar clearance rates. There was no evidence of a dose-response relation between clearance and viral load. Observed was slower clearance in parous women (RR = 0.64, 95% CI: 0.47, 0.89) and faster clearance in ever users of oral contraceptives (RR = 1.38, 95% CI: 1.07, 1.77).

Neighborhood Support and the Birth Weight of Urban Infants

Abstract: Differences in maternal characteristics only partially explain the lower birth weights of infants of African-American women. It is hypothesized that economic and social features of urban neighborhoods may further account for these differences. The authors conducted a household survey of 8,782 adults residing in 343 Chicago, Illinois, neighborhoods to assess mean levels of perceived social support and used US Census data to estimate neighborhood economic disadvantage. Data on birth weight and maternal risk factors were gathered from 95,711 birth certificates (1994-1996). Before statistical adjustment of the data, infants born to African-American mothers were found to be, on average, 297 g lighter than those born to White mothers. After adjustment for individual-level risk factors, this difference was reduced to 154 g. For African-American mothers only, mean birth weight decreased significantly as the neighborhood level of economic disadvantage increased. For White mothers only, a significant positive association was found between perceived levels of neighborhood social support and infant birth weight. Adding these neighborhood-level predictors to the model reduced the adjusted White versus African-American difference in birth weight to 124 g. Results support the hypothesis that neighborhood-level factors are significantly associated with infant birth weight.

Neurodegenerative Diseases and Exposure to Pesticides in the Elderly

Abstract: The authors investigated the hypothesis that exposure to pesticides could be related to central nervous system disorders in a prospective cohort study of 1,507 French elderly (1992-1998). Lower cognitive performance was observed in subjects who had been occupationally exposed to pesticides. In men, the relative risks of developing Parkinson's disease and Alzheimer's disease for occupational exposure assessed by a job exposure matrix were 5.63 (95% confidence interval: 1.47, 21.58) and 2.39 (95% confidence interval: 1.02, 5.63), respectively, after confounding factors were taken into account. No association was found with having a primary job in agriculture or with environmental pesticide exposure, nor was an association found in women. These results suggest the presence of neurologic impairments in elderly persons who were exposed occupationally to pesticides.

5,10-Methylenetetrahydrofolate Reductase Polymorphisms and Leukemia Risk: A HuGE Minireview

Abstract: Leukemias commonly arise as a result of DNA translocations, inversions, or deletions in genes regulating blood cell development or homeostasis. Folate deficiency has been associated with uracil misincorporation into DNA and DNA double strand breaks during uracil excision repair, thus increasing the risk of chromosomal aberrations. Methylenetetrahydrofolate reductase (MTHFR) directs 5,10-methylenetetrahydrofolate toward methionine synthesis at the expense of DNA synthesis. Two MTHFR polymorphisms, C677T and A1298C, have been associated with reduced enzyme activity and C677T with altered distribution of intracellular folate metabolites. Rapidly replicating cell types, such as hematopoietic cells, may be especially sensitive to changes in the availability of intracellular folate. Three case-control studies have evaluated the association between MTHFR polymorphisms and the risk of acute leukemia, and they suggest that both adults and children with the variant forms of MTHFR have a decreased risk of lymphoid leukemias. However, no modification in risk has been observed for myeloid leukemias, suggesting that differences in folate requirements or susceptibility to chromosomal damage may exist between myeloid and lymphoid cells. Further investigation into the association between MTHFR polymorphisms and the risk of leukemia is warranted. It should include larger sample sizes and other polymorphisms in folate metabolism and address interactions with folate status.

Factors Associated with Increased and Decreased Risk of Campylobacter Infection: A Prospective Case-Control Study in Norway

Abstract: In 1999-2000, a prospective case-control study of sporadic, domestically acquired campylobacteriosis was conducted in three counties in Norway to identify preventable risk factors and potentially protective factors. A total of 212 cases and 422 population controls matched by age, sex, and geographic area were enrolled. In conditional logistic regression analysis, the following factors were found to be independently associated with an increased risk of Campylobacter infection: drinking undisinfected water, eating at barbecues, eating poultry bought raw, having occupational exposure to animals, and eating undercooked pork. The following factors were independently related to a decreased risk: eating mutton, eating raw fruits or berries, and swimming. Results indicated that infection is more likely to occur as a result of cross-contamination from raw poultry products than because of poultry consumption per se. Drinking undisinfected water, reported by 53% of cases, was a leading risk factor in this study. Drinking water may constitute the common reservoir linking infection in humans and animals, including poultry and wild birds. Insight into the ecology of Campylobacter in freshwater ecosystems may be required to understand the epidemiology of campylobacteriosis. The possibility that certain foods confer protection against campylobacteriosis deserves exploration.

Effect of Highly Active Antiretroviral Therapy on Time to Acquired Immunodeficiency Syndrome or Death using Marginal Structural Models

Abstract: To estimate the net (i.e., overall) effect of highly active antiretroviral therapy (HAART) on time to acquired immunodeficiency syndrome (AIDS) or death, the authors used inverse probability-of-treatment weighted estimation of a marginal structural model, which can appropriately adjust for time-varying confounders affected by prior treatment or exposure. Human immunodeficiency virus (HIV)-positive men and women (n = 1,498) were followed in two ongoing cohort studies between 1995 and 2002. Sixty-one percent (n = 918) of the participants initiated HAART during 6,763 person-years of follow-up, and 382 developed AIDS or died. Strong confounding by indication for HAART was apparent; the unadjusted hazard ratio for AIDS or death was 0.98. The hazard ratio from a standard time-dependent Cox model that included time-varying CD4 cell count, HIV RNA level, and other time-varying and fixed covariates as regressors was 0.81 (95% confidence interval: 0.61, 1.07). In contrast, the hazard ratio from a marginal structural survival model was 0.54 (robust 95% confidence interval: 0.38, 0.78), suggesting a clinically meaningful net benefit of HAART. Standard Cox analysis failed to detect a clear net benefit, because it does not appropriately adjust for time-dependent covariates, such as HIV RNA level and CD4 cell count, that are simultaneously confounders and intermediate variables.

Effects of Moderate and Vigorous Physical Activity on Heart Rate Variability in a British Study of Civil Servants

Abstract: Original article can be found at: http://ajeoxfordjournalsorg/archive/ Copyright Johns Hopkins Bloomberg School of Public Health DOI: 101093/aje/kwg120 [Full text of this article is not available in the UHRA]

The Interval between Successive Cases of an Infectious Disease

Abstract: The interval between successive cases of an infectious disease is determined by the time from infection to infectiousness, the duration of infectiousness, the time from infection to disease onset (incubation period), the duration of any extra-human phase of the infectious agent, and the proportion clinically affected among infected individuals. The interval is important in the interpretation of infectious disease surveillance and trend data, in the identification of outbreaks, and in the optimization of quarantine and contact tracing. This paper discusses the properties of these intervals, as measured between transmission events or between clinical onsets of successive infected individuals, noting the determinants of their ranges and frequency distributions, the circumstances under which secondary cases may arise before primaries, and under which the infection transmission interval will be different from the interval between clinical onsets of successive cases. It discusses the derivation of interval distribution statistics from descriptive data given in standard textbooks, with illustrations from published data on outbreaks, households, and epidemiologic tracing. Finally, it discusses the implications of such measures for studies of secondary attack rates, for the persistence of infection in human communities, for outbreak response, and for elimination or eradication programs.