Showing papers in "Epidemiology in 2000"
TL;DR: In this paper, the authors introduce marginal structural models, a new class of causal models that allow for improved adjustment of confounding in observational studies with exposures or treatments that vary over time, when there exist time-dependent confounders that are also affected by previous treatment.
Abstract: In observational studies with exposures or treatments that vary over time, standard approaches for adjustment of confounding are biased when there exist time-dependent confounders that are also affected by previous treatment. This paper introduces marginal structural models, a new class of causal models that allow for improved adjustment of confounding in those situations. The parameters of a marginal structural model can be consistently estimated using a new class of estimators, the inverse-probability-of-treatment weighted estimators.
4,655 citations
TL;DR: The marginal structural Cox proportional hazards model is described and used to estimate the causal effect of zidovudine on the survival of human immunodeficiency virus-positive men participating in the Multicenter AIDS Cohort Study.
Abstract: Standard methods for survival analysis, such as the time-dependent Cox model, may produce biased effect estimates when there exist time-dependent confounders that are themselves affected by previous treatment or exposure. Marginal structural models are a new class of causal models the parameters of which are estimated through inverse-probability-of-treatment weighting; these models allow for appropriate adjustment for confounding. We describe the marginal structural Cox proportional hazards model and use it to estimate the causal effect of zidovudine on the survival of human immunodeficiency virus-positive men participating in the Multicenter AIDS Cohort Study. In this study, CD4 lymphocyte count is both a time-dependent confounder of the causal effect of zidovudine on survival and is affected by past zidovudine treatment. The crude mortality rate ratio (95% confidence interval) for zidovudine was 3.6 (3.0-4.3), which reflects the presence of confounding. After controlling for baseline CD4 count and other baseline covariates using standard methods, the mortality rate ratio decreased to 2.3 (1.9-2.8). Using a marginal structural Cox model to control further for time-dependent confounding due to CD4 count and other time-dependent covariates, the mortality rate ratio was 0.7 (95% conservative confidence interval = 0.6-1.0). We compare marginal structural models with previously proposed causal methods.
1,586 citations
TL;DR: It is suggested that elevated levels air pollutants are associated with potentially life-threatening arrhythmia leading to therapeutic interventions by an implanted cardioverter defibrillator.
Abstract: Air pollution episodes have been associated with increased cardiovascular hospital admissions and mortality in time-series studies. We tested the hypothesis that patients with implanted cardioverter defibrillators experience potentially life-threatening arrhythmias after such air pollution episodes. We compared defibrillator discharge interventions among 100 patients with such devices in eastern Massachusetts, according to variations in concentrations of particulate matter, black carbon, and gaseous air pollutants that were measured daily for the years 1995 through 1997. A 26-ppb increase in nitrogen dioxide was associated with increased defibrillator interventions 2 days later (odds ratio = 1.8; 95% confidence interval = 1.1-2.9). Patients with ten or more interventions experienced increased arrhythmias in association with nitrogen dioxide, carbon monoxide, black carbon, and fine particle mass. These results suggest that elevated levels air pollutants are associated with potentially life-threatening arrhythmia leading to therapeutic interventions by an implanted cardioverter defibrillator.
633 citations
TL;DR: Evidence was found in each city that the effect of a single day's exposure to PM10 was manifested across several days, and the effect was spread over several days and did not reach zero until 5 days after the exposure.
Abstract: Many studies have reported associations between air pollution and daily deaths. Those studies have not consistently specified the lag between exposure and response, although most have found associations that persisted for more than 1 day. A systematic approach to specifying the lag association would allow better comparison across sites and give insight into the nature of the relation. To examine this question, I fit unconstrained and constrained distributed lag relations to the association between daily deaths of persons 65 years of age and older with PM10 in 10 U.S. cities (New Haven, Birmingham, Pittsburgh, Canton, Detroit, Chicago, Minneapolis, Colorado Springs, Spokane, and Seattle) that had daily monitoring for PM10. After control for temperature, humidity, barometric pressure, day of the week, and seasonal patterns, I found evidence in each city that the effect of a single day's exposure to PM10 was manifested across several days. Averaging over the 10 cities, the overall effect of an increase in exposure of 10 microg/m3 on a single day was a 1.4% increase in deaths (95% confidence intervals (CI) = 1.15-1.68) using a quadratic distributed lag model, and a 1.3% increase (95% CI = 1.04-1.56) using an unconstrained distributed lag model. In contrast, constraining the model to assume the effect all occurs in one day resulted in an estimate of only 0.65% (95% CI = 0.49-0.81), indicating that this constraint leads to a substantial underestimate of effect. Combining the estimated effect at each day's lag across the 10 cities showed that the effect was spread over several days and did not reach zero until 5 days after the exposure. Given the distribution of sensitivities likely in the general population, this result is biologically plausible. I also found a protective effect of barometric pressure in all 10 locations.
584 citations
TL;DR: The results suggest that appreciable magnetic field effects, if any, may be concentrated among relatively high and uncommon exposures, and that studies of highly exposed populations would be needed to clarify the relation of magnetic fields to childhood leukemia.
Abstract: We obtained original individual data from 15 studies of magnetic fields or wire codes and childhood leukemia, and we estimated magnetic field exposure for subjects with sufficient data to do so. Summary estimates from 12 studies that supplied magnetic field measures exhibited little or no association of magnetic fields with leukemia when comparing 0.1-0.2 and 0.2-0.3 microtesla (microT) categories with the 0-0.1 microT category, but the Mantel-Haenszel summary odds ratio comparing >0.3 microT to 0-0.1 microT was 1.7 (95% confidence limits = 1.2, 2.3). Similar results were obtained using covariate adjustment and spline regression. The study-specific relations appeared consistent despite the numerous methodologic differences among the studies. The association of wire codes with leukemia varied considerably across studies, with odds ratio estimates for very high current vs low current configurations ranging from 0.7 to 3.0 (homogeneity P = 0.005). Based on a survey of household magnetic fields, an estimate of the U.S. population attributable fraction of childhood leukemia associated with residential exposure is 3% (95% confidence limits = -2%, 8%). Our results contradict the idea that the magnetic field association with leukemia is less consistent than the wire code association with leukemia, although analysis of the four studies with both measures indicates that the wire code association is not explained by measured fields. The results also suggest that appreciable magnetic field effects, if any, may be concentrated among relatively high and uncommon exposures, and that studies of highly exposed populations would be needed to clarify the relation of magnetic fields to childhood leukemia.
528 citations
TL;DR: In this article, the relative contributions of fine and coarse particles on respiratory symptoms and peak expiratory flow in schoolchildren were examined. But, the authors concluded that fine particles, especially fine sulfate particles, have much stronger acute respiratory effects than coarse particles.
Abstract: Numerous studies have reported associations between airborne particles and a range of respiratory outcomes from symptoms to mortality. Current attention has been focused on the characteristics of these particles responsible for the adverse health effects. We have reanalyzed three recent longitudinal diary studies to examine the relative contributions of fine and coarse particles on respiratory symptoms and peak expiratory flow in schoolchildren. In the Harvard Six Cities Diary Study, lower respiratory symptoms in a two-pollutant model were associated with an interquartile range increment in fine particles [(for 15 microg/m3 particulate matter (PM) <2.5 microm in aerodynamic diameter (PM2.5), odds ratio = 1.29, 95% confidence limits (CL) = 1.06, 1.57] but not coarse particles (for 8 microg/m3 PM2.5-10, odds ratio = 1.05, 95% CL = 0.90, 1.23). In Uniontown, PA, we found that peak flow was associated with fine particles (for 15 microg/m3 PM2.1, peak flow = -0.91 liters/minute, 95% CL = -0.14, -1.68), especially fine sulfate particles, but not with coarse particles (for 15 microg/m3 PM2.1-10, +1.04 liters/minute, 95% CL = -1.32, +3.40). We found similar results for an equivalent children's cohort in State College, PA. We conclude that fine particles, especially fine sulfate particles, have much stronger acute respiratory effects than coarse particles.
430 citations
TL;DR: Strong evidence is provided that ingestion of inorganic arsenic is associated with human lung cancer in northern Chile in a case-control study involving patients diagnosed with lung cancer between 1994 and 1996 and frequency-matched hospital controls.
Abstract: Cities in northern Chile had arsenic concentrations of 860 microg/liter in drinking water in the period 1958-1970 Concentrations have since been reduced to 40 microg/liter We investigated the relation between lung cancer and arsenic in drinking water in northern Chile in a case-control study involving patients diagnosed with lung cancer between 1994 and 1996 and frequency-matched hospital controls The study identified 152 lung cancer cases and 419 controls Participants were interviewed regarding drinking water sources, cigarette smoking, and other variables Logistic regression analysis revealed a clear trend in lung cancer odds ratios and 95% confidence intervals (CIs) with increasing concentration of arsenic in drinking water, as follows: 1, 16 (95% CI = 05-53), 39 (95% CI = 12-123), 52 (95% CI = 23-117), and 89 (95% CI = 40-196), for arsenic concentrations ranging from less than 10 microg/liter to a 65-year average concentration of 200-400 microg/liter There was evidence of synergy between cigarette smoking and ingestion of arsenic in drinking water; the odds ratio for lung cancer was 320 (95% CI = 72-1980) among smokers exposed to more than 200 microg/liter of arsenic in drinking water (lifetime average) compared with nonsmokers exposed to less than 50 microg/liter This study provides strong evidence that ingestion of inorganic arsenic is associated with human lung cancer
420 citations
TL;DR: It is indicated that urban air pollution increases lung cancer risk and that vehicle emissions may be particularly important.
Abstract: We conducted a population-based case-control study among men 40-75 years of age encompassing all cases of lung cancer 1985-1990 among stable residents of Stockholm County 1950-1990. Questionnaires to subjects or next-of-kin (primarily wives or children) elicited information regarding smoking and other risk factors, including occupational and residential histories. A high response rate (>85%) resulted in 1,042 cases and 2,364 controls. We created retrospective emission databases for NOx/NO2 and SO2 as indicators of air pollution from road traffic and heating, respectively. We estimated local annual source-specific air pollution levels using validated dispersion models and we linked these levels to residential addresses using Geographical Information System (GIS) techniques. Average traffic-related NO2 exposure over 30 years was associated with a relative risk (RR) of 1.2 (95% confidence interval 0.8-1.6) for the top decile of exposure, adjusted for tobacco smoking, socioeconomic status, residential radon, and occupational exposures. The data suggested a considerable latency period; the RR for the top decile of average traffic-related NO2 exposure 20 years previously was 1.4 (1.0-2.0). Little association was observed for SO2. Occupational exposure to asbestos, diesel exhaust, and other combustion products also increased the risk of lung cancer. Our results indicate that urban air pollution increases lung cancer risk and that vehicle emissions may be particularly important.
413 citations
TL;DR: Exposure to increased levels of ambient PM10 and possibly CO during pregnancy may contribute to the occurrence of preterm births in Southern California.
Abstract: We evaluated the effect of air pollution exposure during pregnancy on the occurrence of preterm birth in a cohort of 97,518 neonates born in Southern California. We used measurements of carbon monoxide (CO), nitrogen dioxide, ozone, and particulate matter less than 10 microm (PM10) collected at 17 air-quality-monitoring stations to create average exposure estimates for periods of pregnancy. We calculated crude and adjusted risk ratios (RRs) for premature birth by period-specific ambient pollution levels. We observed a 20% increase in preterm birth per 50-microg increase in ambient PM10 levels averaged over 6 weeks before birth [RRcrude = 1.20; 95% confidence interval (CI) = 1.09-1.33] and a 16% increase when averaging over the first month of pregnancy (RRcrude = 1.16; 95% CI = 1.06-1.26). PM10 effects showed no regional pattern. CO exposure 6 weeks before birth consistently exhibited an effect only for the inland regions (RRcrude = 1.13; 95% CI = 1.08-1.18 per 3 parts per million), and during the first month of pregnancy, the effect was weak for all stations (RRcrude = 1.04; 95% CI = 1.01-1.09 per 3 parts per million). Exposure to increased levels of ambient PM10 and possibly CO during pregnancy may contribute to the occurrence of preterm births in Southern California.
345 citations
TL;DR: It was found that a number of reproductive and contraceptive factors that suppress ovulation, including gravidity, breast feeding, and oral contraception, reduced the risk of ovarian cancer and environmental factors and medical conditions that increased risk included talc use, endometriosis, ovarian cysts, and hyperthyroidism.
Abstract: Previous epidemiologic observations consistently suggest that suppression of ovulation, tubal ligation, and hysterectomy reduce the risk of ovarian cancer and that perineal talc use increases the risk. We examined these and other risk factors in the context of a new hypothesis: that inflammation may play a role in ovarian cancer risk. Ovulation entails ovarian epithelial inflammation; talc, endometriosis, cysts, and hyperthyroidism may be associated with inflammatory responses of the ovarian epithelium; gynecologic surgery may preclude irritants from reaching the ovaries via ascension from the lower genital tract. We evaluated these risk factors in a population-based case-control study. Cases 20-69 years of age with a recent diagnosis of epithelial ovarian cancer (767) were compared with community controls (1,367). We found that a number of reproductive and contraceptive factors that suppress ovulation, including gravidity, breast feeding, and oral contraception, reduced the risk of ovarian cancer. Environmental factors and medical conditions that increased risk included talc use, endometriosis, ovarian cysts, and hyperthyroidism. Gynecologic surgery including hysterectomy and tubal ligation were protective. Tubal ligation afforded a risk reduction even 20 or more years after the surgery. The spectrum of associations provides support for the hypothesis that inflammation may mediate ovarian cancer risk.
333 citations
TL;DR: The data support earlier studies suggesting that prenatal environmental tobacco smoke exposure, in addition to maternal smoking, affects infant health.
Abstract: We examined the association of exposure to environmental tobacco smoke with birth weight and gestational age in a large, prospective study We also compared these endpoints between infants of active maternal smokers and those of non-smoking, non-ETS exposed women Pregnant women were interviewed by telephone during the first trimester, and pregnancy outcome was determined for 99% Among the 4,454 singleton live births that could be linked to their birth certificate, we confirmed increased risks of low birth weight and small for gestational age with heavier maternal smoking (> 10 cigarettes/day), as well as noting an increased risk for "very preterm" birth ( or = 30 years) than those of younger mothers, as well as among non-whites High environmental tobacco smoke exposure (> or = 7 hours/day in non-smokers) was moderately associated with low birth weight (adjusted odds ratio (AOR) 18, 95% confidence limits (95% CL) = 082, 41), preterm birth (AOR 16, 95% CL = 087, 29), and most strongly with very preterm birth (AOR 24, 95% CL = 10, 53) These associations were generally greater among non-whites than whites The data support earlier studies suggesting that prenatal environmental tobacco smoke exposure, in addition to maternal smoking, affects infant health
TL;DR: The results indicate that traffic-related air pollution leads to increased prevalence of atopic sensitizations, allergic symptoms, and diseases.
Abstract: Traffic emissions are a major source of air pollution in Western industrialized countries. To investigate the association between traffic-related air pollution and parameters of atopy, we studied 317 children 9 years of age living near major roads in two urban areas and one suburban area of a city in West Germany. Atopic sensitization was analyzed by skin-prick testing and determination of allergen-specific serum immunoglobulin E. Parents recorded allergic symptoms in a symptom diary, and physicians assessed allergic diseases. Personal NO2 exposure and NO2 concentrations in front of each child's home were measured. Outdoor NO2 was a good predictor for traffic exposure but a poor predictor for NO2 exposure at the personal level. Atopy was found to be related to outdoor NO2 (odds ratio for the association between symptoms of allergic rhinitis and outdoor NO2 = 1.81; 95% confidence interval = 1.02-3.21) but not to personal NO2 (odds ratio for the association between symptoms of allergic rhinitis and personal NO2 = 0.99; 95% confidence interval = 0.55-1.79). When the analysis was restricted to urban areas, we found that hay fever, symptoms of allergic rhinitis, wheezing, sensitization against pollen, house dust mites or cats, and milk or eggs were associated with outdoor NO2. The results indicate that traffic-related air pollution leads to increased prevalence of atopic sensitizations, allergic symptoms, and diseases.
TL;DR: The associations observed here lend support to the hypothesis that pubertal events may be affected by pre- and postnatal exposure to organohalogens, and perinatal PBB exposure was associated with earlier pubic hair stage in breastfed girls, but little association was found with breast development.
Abstract: Accidental contamination of the Michigan food chain with polybrominated biphenyls (PBBs) led to the exposure of more than 4,000 individuals in 1973. Because PBB exposure is suspected to disrupt endocrine function, we assessed pubertal development in females 5–24 years of age (N = 327) who were expos
TL;DR: Results of this prospective study support the hypothesis that the estrogen metabolism pathway favoring 2-hydroxylation over 16&agr;-hydroxyylation is associated with a reduced risk of invasive breast cancer risk in premenopausal women.
Abstract: Experimental and clinical evidence suggests that 16alpha-hydroxylated estrogen metabolites, biologically strong estrogens, are associated with breast cancer risk, while 2-hydroxylated metabolites, with lower estrogenic activity, are weakly related to this disease. This study analyzes the association of breast cancer risk with estrogen metabolism, expressed as the ratio of 2-hydroxyestrone to 16alpha-hydroxyestrone, in a prospective nested case-control study. Between 1987 and 1992, 10,786 women (ages 35-69 years) were recruited to a prospective study on breast cancer in Italy, the "Hormones and Diet in the Etiology of Breast Cancer" (ORDET) study. Women with a history of cancer and women on hormone therapy were excluded at baseline. At recruitment, overnight urine was collected from all participants and stored at -80 degrees C. After an average of 5.5 years of follow-up, 144 breast cancer cases and four matched controls for each case were identified among the participants of the cohort. Among premenopausal women, a higher ratio of 2-hydroxyestrone to 16alpha-hydroxyestrone at baseline was associated with a reduced risk of breast cancer: women in the highest quintile of the ratio had an adjusted odds ratio (OR) for breast cancer of 0.58 [95% confidence interval (CI) = 0.25-1.34]. The corresponding adjusted OR in postmenopausal women was 1.29 (95% CI = 0.53-3.10). Results of this prospective study support the hypothesis that the estrogen metabolism pathway favoring 2-hydroxylation over 16alpha-hydroxylation is associated with a reduced risk of invasive breast cancer risk in premenopausal women.
TL;DR: A systematic review of case-control studies comparing EBV serology in MS patients and controls finds a role of EBV in the etiology of MS, and suggests that these results are not readily explained by an aspecific immune activation among MS patients.
Abstract: It has been suggested that the Epstein-Barr virus (EBV) plays a role in the etiology of multiple sclerosis (MS), but individual epidemiologic studies have been inconclusive, in part because of the high prevalence of previous infection among individuals without MS. We conducted a systematic review of case-control studies comparing EBV serology in MS patients and controls. Eight published investigations were identified, including a total of 1,005 cases and 1,060 controls. The summary odds ratio of MS comparing EBV seropositive individuals with EBV seronegative individuals was 13.5 (95% CI = 6.3-31.4). The strength and consistency of this association and the high sensitivity and specificity of EBV serology suggest that these results are not readily explained by an aspecific immune activation among MS patients. These findings support a role of EBV in the etiology of MS.
TL;DR: The pregnancies of women who were both obese and diabetic were 3.1 times as likely to result in an offspring with a defect than were those of nonobese, nondiabetic women, which suggests that obesity and diabetes mellitus may act synergistically in the pathogenesis of congenital anomalies.
Abstract: This study was designed to evaluate the effects of maternal obesity and diabetes mellitus on the risk of nonchromosomal congenital defects. We used data from 22,951 pregnant women enrolled in a prospective cohort study of early prenatal exposures and pregnancy outcome. The relative risks [prevalence ratios (PRs)] of major nonchromosomal congenital defects associated with obesity and diabetes, alone or in combination, were calculated using multiple logistic regression analysis. In this study, in the absence of diabetes, obese women (body mass index > or =28) had no higher risk, overall, of having an offspring with a major defect [PR = 0.95; 95% confidence interval (CI) = 0.62-1.5]. Their offspring, however, did have a higher prevalence of certain types of defects, including orofacial clefts; club foot; cardiac septal defects; and, to a lesser extent, hydrocephaly and abdominal wall defects. Women with pre-existing or gestational diabetes who were not obese also had no excess risk overall of having offspring affected by a major defect (PR = 0.98; 95% CI = 0.43-2.2), although they did have a higher prevalence of musculoskeletal defects. The pregnancies of women who were both obese and diabetic were 3.1 times as likely (95% CI = 1.2-7.6) to result in an offspring with a defect than were those of nonobese, nondiabetic women, which suggests that obesity and diabetes mellitus may act synergistically in the pathogenesis of congenital anomalies. The defects were largely craniofacial or musculoskeletal.
TL;DR: It is suggested that living on busy roads is associated with a higher risk for a sensitization to pollen and could possibly be interpreted as an indication for interactions between pollen and air pollutants.
Abstract: We examined the association between the presence of an allergic sensitization and seasonal allergic diseases or symptoms and the exposure to road traffic in Basel, Switzerland. Traffic counts at the domiciles of subjects ranged from 24 to 32,504 cars per 24 hours, with a median of 1,624. To investigate the relation of road traffic and allergies, we matched the data of the traffic inventory of Basel with those of the 820 participants of the SAPALDIA study (Swiss Study on Air Pollution and Lung Diseases in Adults), ages 18-60 years, who had completed a detailed respiratory health questionnaire and had undergone allergy testing (skin prick tests and serologic examinations). We observed a positive association with a sensitization to pollen that was most pronounced among persons with a duration of residence of at least 10 years. The odds ratios (adjusted for educational level, smoking behavior, number of siblings, age, sex, and family history of atopy) for cars, contrasting four exposure categories with the lowest quartile as referent category, were 1.99 [95% confidence interval (CI) = 0.91-4.38], 2.47 (95% CI = 1.06-5.73), and 2.83 (95% CI = 1.26-6.31). These results suggest that living on busy roads is associated with a higher risk for a sensitization to pollen and could possibly be interpreted as an indication for interactions between pollen and air pollutants. We did not, however, find a similar relation between motor vehicle traffic and hay fever or seasonal allergic symptoms, and we saw no trend that increasing traffic exposure was associated with a rise in sensitization rates to indoor allergens.
TL;DR: It is indicated that incomplete and reversible inhibition of platelet cyclooxygenase by non-aspirin nonsteroidal antiinflammatory drugs is not sufficient to produce clinically detectable cardiovascular protection comparable with that achieved by low-dose aspirin through irreversible inactivation of platelets cyclo oxygengenase.
Abstract: The antiplatelet effect of aspirin reduces the risk of clinical manifestations of atherothrombosis by approximately 25% in secondary prevention settings. Data are limited in primary prevention of coronary heart disease, and even more in women. Here, we estimate the effects of aspirin and non-aspirin nonsteroidal antiinflammatory drugs in the primary prevention of myocardial infarction in postmenopausal women. We followed a cohort of 164,769 women, 50-74 years of age, registered in the General Practice Research Database in the United Kingdom, from January 1991 through December 1995. For aspirin and non-aspirin nonsteroidal antiinflammatory drugs, the risk of myocardial infarction associated with current use was compared with risk in non-users, using a nested case-control analysis. Overall, the relative risk of myocardial infarction associated with current use of aspirin of more than 1 month's duration was 0.56 [95% confidence interval (95% CI) = 0.26-1.21], and that of nonfatal myocardial infarction was 0.28 (95% Cl = 0.08-0.91). Chronic use of nonsteroidal antiinflammatory drugs was not associated with a protective effect (relative risk = 1.32; 95% Cl = 0.97-1.81). These findings indicate that incomplete and reversible inhibition of platelet cyclooxygenase by non-aspirin nonsteroidal antiinflammatory drugs is not sufficient to produce clinically detectable cardiovascular protection comparable with that achieved by low-dose aspirin through irreversible inactivation of platelet cyclooxygenase.
TL;DR: It is concluded that maternal perception of exposure to racial discrimination during pregnancy may be associated with very low birth weight in their infants.
Abstract: We performed a hospital-based case-control study of African-American mothers to explore the relation between a mother's perception of exposure to racial discrimination during pregnancy and very low birth weight. We administered a structured questionnaire to low-income mothers of very low birth weight ( 2500 gm; N = 60) infants. The unadjusted and adjusted odds ratio of very low birth weight for maternal exposure to racial discrimination were 1.9 (0.5-6.6) and 3.2 (0.9-11.3), respectively. We conclude that maternal perception of exposure to racial discrimination during pregnancy may be associated with very low birth weight in their infants.
TL;DR: The food frequency questionnaire performed similarly to food frequency questionnaires used in other cohort studies, indicating similar ability to examine diet-disease relations.
Abstract: We assessed the validity and reproducibility of a self-administered 68-item food frequency questionnaire completed in 1992-1993 by approximately 185,000 adults. Four hundred forty-one participants completed four 24-hour dietary recall interviews over a 1-year period and a repeat administration of the food frequency questionnaire. For 20 nutrients and 10 food groups, measured nutrient intakes, but not food group intakes, were consistently lower by food frequency questionnaire than by recall. Energy-adjusted, attenuation-corrected Pearson validity correlations ranged from 0.12 to 0.80, with a median of 0.58. Reproducibility measures were generally high, with a median of 0.69. The food frequency questionnaire performed similarly to food frequency questionnaires used in other cohort studies, indicating similar ability to examine diet-disease relations.
TL;DR: The association of background-level dioxin exposure with the prevalence of diabetes in these data may well be due to reasons other than causality, although a causal contribution cannot be wholly dismissed.
Abstract: Data from several epidemiologic studies suggest that exposure to unusually high amounts of dioxin (2,3,7,8-tetrachlorodibenzo-p-dioxin) increases the risk of diabetes mellitus, and experimental data suggest that the mechanism for this is decreased cellular glucose uptake. To investigate the dose-response relation more closely, we examined the association of serum dioxin level with prevalence of diabetes mellitus and with levels of serum insulin and glucose among 1,197 veterans in the Air Force Health Study who never had contact with dioxin-contaminated herbicides and whose serum dioxin level was within the range of background exposure typically seen in the United States ( or =5.2 ng/kg lipid) was 1.71 (95% confidence interval = 1.00-2.91). The association was slightly attenuated after adjustment for serum triglycerides. Whether adjustment for serum triglycerides was appropriate, however, cannot be determined with available data. The association of background-level dioxin exposure with the prevalence of diabetes in these data may well be due to reasons other than causality, although a causal contribution cannot be wholly dismissed.
TL;DR: This method can detect threshold and other nonlinear relations in epidemiologic studies, and is applied to analyze the association between PM10 and daily deaths in ten U.S. cities, finding that the association appears linear down to the lowest levels observed in the study.
Abstract: Air pollution has been associated with daily mortality in numerous studies over the last decade. Although considerable attention has focused on issues of potential confounding in these associations, little has been done to address the question of what the shape of the dose-response relation looks like. The question of whether a threshold exists for these relations is of particular concern, with regard to both this application and many other epidemiologic questions. Nonparametric smoothing is widely used to control for the potentially nonlinear relations between covariates and daily deaths but has been little used to model the air pollution associations. Because sampling variability, among other factors, can introduce considerable noise into the estimates of linear dose-response curves, quantitative summaries have been widely used to come up with best linear fits. The same ability of meta-analytic techniques to average out noise applies to nonparametric smooth estimates in individual cities. We have developed a method of applying these techniques to combining nonparametric smooths. Using simulation studies, we show that this method can detect threshold and other nonlinear relations in epidemiologic studies, and we then apply it to analyze the association between PM10 and daily deaths in ten U.S. cities. We find that the association appears linear down to the lowest levels observed in the study. This method is generally applicable in settings where data from multiple studies can be combined.
TL;DR: Analysis of the association between estriol levels in 188 women in the 17th, 25th, 33rd, and 37th weeks of pregnancy and the birth weights of their infants suggests that, on an aggregate level, birth weight can be used as a proxy variable of intrauterine Estriol exposure.
Abstract: In epidemiologic studies of perinatal exposures, birth weight has been proposed as a proxy variable for intrauterine estrogen exposure. To assess the validity of this assumption, we performed analyses of the association between estriol levels in 188 women in the 17th, 25th, 33rd, and 37th weeks of pregnancy and the birth weights of their infants. We found a general increase in mean cumulative estriol dose with increasing birth weight category throughout pregnancy. In late pregnancy, mean pregnancy estriol level of mothers of infants in the highest birth weight category (>4,500 gm) was twice as high as that of mothers of infants in the lowest category (<2,500 gm), 775 nmol/liter and 392 nmol/liter, respectively. Smoking lowered the maternal estriol levels by 20% or more throughout pregnancy. With smoking and birth weight included in a regression analysis, maternal age, placental weight, and infant ponderal index did not add any explanatory power to the model. Our data suggest that, on an aggregate level, birth weight can be used as a proxy variable of intrauterine estriol exposure.
TL;DR: It is indicated that fetal growth retardation rather than preterm delivery of male infants is the main gestational factor underlying the associations of asthma and atopic dermatitis but does not explain the apparent increase over time of asthma or atopic diseases.
Abstract: Gestational factors have been hypothesized to play a role in the susceptibility to asthma and atopic dermatitis. We examined whether fetal growth was associated with asthma and atopic dermatitis separately in a population of 4,795 male conscripts born between 1973 and 1975 in Denmark. The prevalence of asthma was 4.7%. The prevalence odds ratio of asthma in conscripts with a birth weight below 2,501 g was 1.5 (95% confidence interval = 0.7-3.1) compared with conscripts with a birth weight of 3,001-3,500 g, adjusted for gestational age and potential confounders. The adjusted prevalence odds ratio among conscripts born before 34 gestational weeks was 0.8 (95% confidence interval = 0.3-2.0) compared with conscripts born at term. The prevalence of atopic dermatitis was 1.0%. The prevalence odds ratio of atopic dermatitis among those with a birth weight below 2,501 g was 3.0 (95% confidence interval = 0.8-11.9) compared with those whose birth weight was between 3,001 and 3,500 g. Men whose gestational age had been below 34 weeks had an adjusted prevalence odds ratio of 0.3 (95% confidence interval = 0.0-3.1). These findings indicate that fetal growth retardation rather than preterm delivery of male infants is the main gestational factor underlying the associations but does not explain the apparent increase over time of asthma or atopic diseases.
TL;DR: Among the non-work-related factors, only body mass index ≥25 kg/m 2 and immigrant status was associated with a slight increase in relative risk, and training in the use oftransfer devices, and regular use of transfer devices reduced the relative risk from patient transfer.
Abstract: We conducted a case-referent study to identify and quantify work-related and non-work-related risk indicators for reported over-exertion back injuries among nursing personnel. The source population was all nursing personnel employed in the Stockholm County hospitals during a 32-month period. The 240 cases and 614 referents completed questionnaires about occupation, type of clinic, working hours, shift work, patient transfers, perceived exertion, back pain, prior back injury, job strain, body mass index (BMI), smoking, immigrant status, physical training, and self-rated fitness. The highest relative risks (RR) were observed for work-related factors: working at an orthopedic clinic (RR = 5.2; 95% CI = 2.7-10.2), > or =1 patient transfer/shift (RR = 2.7; 95% CI = 1.6-4.5), and working full-time (RR = 2.4; 95% CI = 1.6-3.6). Training in the use of transfer devices, and regular use of transfer devices, reduced the relative risk from patient transfer. Among the non-work-related factors, only body mass index > or =25 kg/m2 and immigrant status was associated with a slight increase in relative risk.
TL;DR: Risks of perinatal death and having a low birth weight infant increased with dose of radiotherapy directed to the abdomen and there was no evidence of an increased risk of having a spontaneous abortion or an infant with a birth defect.
Abstract: This Ontario province-wide cohort study was conducted to compare the risk of adverse pregnancy outcomes in female childhood cancer survivors who received abdominal-pelvic radiation and/or chemotherapy with alkylating agents with the risk among those who were treated by non-sterilizing alkylating agents with the risk among those who were treated by non-sterilizing surgery only. Females in Ontario, Canada, diagnosed in 1964-1988 before age 20 with a histologically confirmed malignancy and who had survived for at least 5 years, attained age 18, and were alive at the time of study, were identified through the Ontario Cancer Registry. We ascertained pregnancy outcomes by a telephone-administered questionnaire. Treatment data were abstracted from medical records for 830 subjects 18-49 years of age, the analysis comprised 340 survivors who had one or more pregnancies after treatment. There was no evidence of an increased risk of having a spontaneous abortion or an infant with a birth defect. Survivors receiving abdominal-pelvic radiation were more likely to have a low birth weight infant (odds ratio estimate [OR] = 3.64; 95% confidence interval [CI] = 1.33-9.96), a premature low birth weight infant (OR = 3.29; 95% CI = 0.97-11.1), or an infant who died in the perinatal period (OR = 2.41; 95% CI = 0.50-11.5), compared with those receiving surgery. Risks of perinatal death and having a low birth weight infant increased with dose of radiotherapy directed to the abdomen.
TL;DR: It is suggested that women who experience stressful life events around the time of conception or early gestation may be at increased risk of delivering infants with certain congenital anomalies.
Abstract: We used data from a population-based case-control study to explore the relation between certain life events during the periconceptional period and several types of congenital anomalies. We ascertained cases from pregnancies ending in 1987-1989 and randomly selected controls from eligible liveborn infants. In telephone interviews, women reported deaths of anyone close to them. They also reported job losses or separations/divorces, for themselves or anyone close to them. Experiencing at least one stressful event during the periconceptional period was associated with a prevalence odds ratio of 1.4-1.5 for the delivery of infants with conotruncal heart defects, neural tube defects, and isolated cleft lip with or without palate. These associations tended to be restricted to women who were not obese and women with less than or equal to a high school education. This study suggests that women who experience stressful life events around the time of conception or early gestation may be at increased risk of delivering infants with certain congenital anomalies.
TL;DR: The results of this population-based cohort study add further support to the value of NSAIDs as a candidate for primary prevention of colorectal tumors.
Abstract: Chronic treatment with nonsteroidal antiinflammatory drugs (NSAIDs) has been associated with a reduced risk of colorectal cancer, but less information is available on the relationship between NSAIDs and colorectal adenoma. We carried out a population-based cohort study with nested case-control analysis to determine the association between the use of aspirin and individual NSAIDs and the risk of colorectal adenoma. The General Practice Research Database in the United Kingdom was the source population. We followed 943,903 persons who were 40-79 years of age and free of colorectal adenoma or other cancer at baseline, which varied between January 1994 and September 1997. There were 1,864 incident cases of colorectal adenoma, for an incidence rate of 6.8 per 10,000 person-years. Compared with non-users, long-term users (1 year and more) of nonaspirin NSAIDs had a 40% decreased risk of colorectal adenoma (relative risk = 0.6; 95% confidence interval = 0.4-0.9). Long-term NSAID use was still associated with a reduced risk 1 year after stopping NSAID treatment. Use of most individual NSAIDs conferred a reduced risk. The risk of developing colorectal adenoma was reduced in long-term users of aspirin at doses of 300 mg daily (relative risk = 0.6; 95% confidence interval = 0.4-1.0), but reduced risk was not evident with daily doses of 75 and 150 mg aspirin. These results add further support to the value of NSAIDs as a candidate for primary prevention of colorectal tumors.
TL;DR: The most important evidence for a benefit of early detection of prostate cancer using prostate-specific antigen (PSA) testing would be a decline in prostate cancer mortality rates to levels below those existing before diagnostic use of PSA testing as mentioned in this paper.
Abstract: It has been noted that the most important evidence for a benefit of early detection of prostate cancer using prostate-specific antigen (PSA) testing would be a decline in prostate cancer mortality rates to levels below those existing before diagnostic use of PSA testing. We document a decrease in U.S. prostate cancer mortality rates in white men less than 85 years of age to levels below those existing in 1986, the year use of PSA testing was approved. In fact, for men 60-79 years of age, prostate cancer mortality rates were lower in 1997 than in any year since 1950. Although it has been argued that the decrease in prostate cancer mortality rates began too soon to be explained by PSA testing, stage-specific survival rates indicate that a rapid decrease in mortality may be explained by the large number of high-grade prostate cancers detected before metastasis. If recent decreases in U.S. prostate cancer mortality rates are due to early detection using PSA testing, randomized clinical trials investigating PSA testing will show early evidence of a mortality benefit.
TL;DR: This work provides additional code to obtain epidemiologic output from GLIMMIX, illustrated with new data on diet and breast cancer from the European Community Multicenter Study on Antioxidants, Myocardial Infarction, and Breast Cancer (EURAMIC).
Abstract: Previous work has shown that multilevel modeling can be a valuable technique for epidemiologic analysis. The complexity of using this approach, however, continues to restrict its general application. A critical factor is the lack of flexible and appropriate software for multilevel modeling. SAS provides a macro, GLIMMIX, that can be used for multilevel modeling, but that is not sufficient for a complete epidemiologic analysis. We here provide additional code to obtain epidemiologic output from GLIMMIX, illustrated with new data on diet and breast cancer from the European Community Multicenter Study on Antioxidants, Myocardial Infarction, and Breast Cancer (EURAMIC). Our results give epidemiologists an easily used tool for fitting multilevel models.