Showing papers in "Epidemiology in 2010"

Assessing the performance of prediction models: a framework for traditional and novel measures.

Abstract: The performance of prediction models can be assessed using a variety of methods and metrics. Traditional measures for binary and survival outcomes include the Brier score to indicate overall model performance, the concordance (or c) statistic for discriminative ability (or area under the receiver operating characteristic [ROC] curve), and goodness-of-fit statistics for calibration.Several new measures have recently been proposed that can be seen as refinements of discrimination measures, including variants of the c statistic for survival, reclassification tables, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Moreover, decision-analytic measures have been proposed, including decision curves to plot the net benefit achieved by making decisions based on model predictions.We aimed to define the role of these relatively novel approaches in the evaluation of the performance of prediction models. For illustration, we present a case study of predicting the presence of residual tumor versus benign tissue in patients with testicular cancer (n = 544 for model development, n = 273 for external validation).We suggest that reporting discrimination and calibration will always be important for a prediction model. Decision-analytic measures should be reported if the predictive model is to be used for clinical decisions. Other measures of performance may be warranted in specific applications, such as reclassification metrics to gain insight into the value of adding a novel predictor to an established model.

Risk Factors for Falls in Community-dwelling Older People: A Systematic Review and Meta-analysis

Abstract: Background:Falls are the main cause of accidental death in persons aged 65 years or older.Methods:Using MEDLINE and previous reviews, we searched for prospective studies investigating risk factors for falls among community-dwelling older people. For risk factors investigated by at least 5 studies in

Negative controls: a tool for detecting confounding and bias in observational studies.

Abstract: Noncausal associations between exposures and outcomes are a threat to validity of causal inference in observational studies. Many techniques have been developed for study design and analysis to identify and eliminate such errors. Such problems are not expected to compromise experimental studies, where careful standardization of conditions (for laboratory work) and randomization (for population studies) should, if applied properly, eliminate most such noncausal associations. We argue, however, that a routine precaution taken in the design of biologic laboratory experiments--the use of "negative controls"--is designed to detect both suspected and unsuspected sources of spurious causal inference. In epidemiology, analogous negative controls help to identify and resolve confounding as well as other sources of error, including recall bias or analytic flaws. We distinguish 2 types of negative controls (exposure controls and outcome controls), describe examples of each type from the epidemiologic literature, and identify the conditions for the use of such negative controls to detect confounding. We conclude that negative controls should be more commonly employed in observational studies, and that additional work is needed to specify the conditions under which negative controls will be sensitive detectors of other sources of error in observational studies.

To GEE or not to GEE: comparing population average and mixed models for estimating the associations between neighborhood risk factors and health.

Abstract: Two modeling approaches are commonly used to estimate the associations between neighborhood characteristics and individual-level health outcomes in multilevel studies (subjects within neighborhoods). Random effects models (or mixed models) use maximum likelihood estimation. Population average models typically use a generalized estimating equation (GEE) approach. These methods are used in place of basic regression approaches because the health of residents in the same neighborhood may be correlated, thus violating independence assumptions made by traditional regression procedures. This violation is particularly relevant to estimates of the variability of estimates. Though the literature appears to favor the mixed-model approach, little theoretical guidance has been offered to justify this choice. In this paper, we review the assumptions behind the estimates and inference provided by these 2 approaches. We propose a perspective that treats regression models for what they are in most circumstances: reasonable approximations of some true underlying relationship. We argue in general that mixed models involve unverifiable assumptions on the data-generating distribution, which lead to potentially misleading estimates and biased inference. We conclude that the estimation-equation approach of population average models provides a more useful approximation of the truth.

The hazards of hazard ratios.

Abstract: The hazard ratio (HR) is the main, and often the only, effect measure reported in many epidemiologic studies. For dichotomous, non–time-varying exposures, the HR is defined as the hazard in the exposed groups divided by the hazard in the unexposed groups. For all practical purposes, hazards can be thought of as incidence rates and thus the HR can be roughly interpreted as the incidence rate ratio. The HR is commonly and conveniently estimated via a Cox proportional hazards model, which can include potential confounders as covariates. Unfortunately, the use of the HR for causal inference is not straightforward even in the absence of unmeasured confounding, measurement error, and model misspecification. Endowing a HR with a causal interpretation is risky for 2 key reasons: the HR may change over time, and the HR has a built-in selection bias. Here I review these 2 problems and some proposed solutions. As an example, I will use the findings from a Women’s Health Initiative randomized experiment that compared the risk of coronary heart disease of women assigned to combined (estrogen plus progestin) hormone therapy with that of women assigned to placebo. By using a randomized experiment as an example, the discussion can focus on the shortcomings of the HR, setting aside issues of confounding and other serious problems that arise in observational studies. The Women’s Health Initiative followed over 16,000 women for an average of 5.2 years before the study was halted due to safety concerns. The primary result from the trial was a HR. As stated in the abstract and shown in Table 1 of the article, “Combined hormone therapy was associated with a hazard ratio of 1.24.” In addition, Table 2 provided the HRs during each year of follow-up: 1.81, 1.34, 1.27, 1.25, 1.45, and 0.70 for years 1, 2, 3, 4, 5, and 6 , respectively. Thus, the HR reported in the abstract and Table 1 can be viewed as some sort of weighted average of the period-specific HRs reported in Table 2. This bring us to Problem 1: although the HR may change over time, some studies report only a single HR averaged over the duration of the study’s follow-up. As a result, the conclusions from the study may critically depend on the duration of the follow-up. For example, the average HR in the WHI would have been 1.8 if the study had been halted after 1 year of follow-up, 1.7 after 2 years, 1.2 after 5 years, and—who knows—perhaps 1.0 after 10 years. The 24% increase in the rate of coronary heart disease that many researchers and journalists consider as the effect of combined hormone therapy is the result of the arbitrary choice of an average follow-up period of 5.2 years. A trial with a shorter follow-up could have reported an 80% increase, whereas a longer trial might have found little or no increase at all.

Protecting human health from air pollution: shifting from a single-pollutant to a multipollutant approach.

Abstract: To date, the assessment of public health consequences of air pollution has largely focused on a single-pollutant approach aimed at estimating the increased risk of adverse health outcomes associated with the exposure to a single air pollutant, adjusted for the exposure to other air pollutants. However, air masses always contain many pollutants in differing amounts, depending on the types of emission sources and atmospheric conditions. Because humans are simultaneously exposed to a complex mixture of air pollutants, many organizations have encouraged moving towards "a multipollutant approach to air quality." Although there is general agreement that multipollutant approaches are desirable, the challenges of implementing them are vast.

Bias formulas for sensitivity analysis for direct and indirect effects

Abstract: :A key question in many studies is how to divide the total effect of an exposure into a component that acts directly on the outcome and a component that acts indirectly, ie, through some intermediate. For example, one might be interested in the extent to which the effect of diet on blood pre

Ischemic Heart Disease and Stroke in Relation to Blood DNA Methylation

Abstract: Background:Epigenetic features such as DNA hypomethylation have been associated with conditions related to cardiovascular risk. We evaluated whether lower blood DNA methylation in heavily methylated repetitive sequences predicts the risk of ischemic heart disease and stroke.Methods:We quantified blo

Urban ambient particle metrics and health: a time-series analysis.

Abstract: Background:Epidemiologic evidence suggests that exposure to ambient particulate matter is associated with adverse health effects. Little is known, however, about which components of the particulate mixture (size, number, source, toxicity) are most relevant to health. We investigated associations of

Nitrate Intake and the Risk of Thyroid Cancer and Thyroid Disease

Abstract: Thyroid cancer is the most common malignancy of the endocrine system and the eighth most common cancer among women.1 In the United States, the incidence of thyroid cancer has increased substantially since 1980, with an annual percentage increase of 6% during the period 1997–2003.2 This increase may reflect better detection,3 although more recent analyses of US incidence data show that at least some of the increase has occurred for larger tumors and for men and women of all races and ethnicities,4,5 suggesting other factors besides detection. Papillary thyroid cancer accounts for >70% of thyroid tumors in the United States. The only established risk factor is exposure to ionizing radiation, particularly in early childhood.1 Some epidemiologic studies have shown increased risk with goiter6 and number of pregnancies7 and lower risk with intake of fish and cruciferous vegetables.1,6 Ingested nitrate inhibits thyroid uptake of iodide by binding to the sodium-iodide symporter on the surface of thyroid follicles. This reduces the levels of the thyroid hormones triiodothyronine (T3) and thyroxin (T4), which increases thyroid stimulating hormone (TSH). TSH controls thyroid hormone production through a negative feedback loop.8–11 Chronic stimulation of the thyroid gland by TSH can lead to proliferative changes in follicular cells, including hypertrophy and hyperplasia as well as neoplasia.12,13 There is some evidence from human studies that exposure to elevated nitrate levels in drinking water is associated with increased thyroid volume and increased frequency of subclinical thyroid disorders.14,15 Nitrate and nitrite are also precursors in the endogenous formation of N-nitroso compounds, which are potent animal carcinogens that cause thyroid and many other tumors in animal models.16 Ingestion of nitrate and nitrite has also been associated with increased risk of stomach, esophagus, and other cancers in some epidemiologic studies.17 Nitrate is a common contaminant of drinking water, particularly in agricultural areas where application of nitrogen fertilizers since the 1950s has resulted in increasing concentrations of nitrate in drinking water supplies.18–20 Nitrate is also a natural component of the diet, occurring at high levels in green leafy and certain root vegetables. There is some evidence that higher rates of fertilizer application increase nitrate levels in vegetables.21 For example, organically grown lettuce, which does not receive inorganic nitrogen fertilizer, has lower nitrate concentrations than conventionally grown lettuce.22 Thus, intensive agricultural practices may have increased exposure to nitrate from both dietary and drinking water sources. High exposure to nitrate can cause methemoglobinemia in infants. For this reason, nitrate is regulated in public water supplies at a maximum contaminant level (MCL) of 10 mg/L as nitrate-nitrogen (N) (about 45 mg/L as nitrate). Acceptable daily intake values have also been set for dietary intake, with a particular focus on levels in baby foods. However, the regulatory limits for nitrate in food and water have not been extensively studied in relation to other health outcomes.23 To date, epidemiologic studies of thyroid cancer risk have not evaluated nitrate intake in relation to thyroid cancer, and the literature on the relationship with thyroid conditions is limited. We investigated the association between incident thyroid cancer and prevalent hyperthyroidism and hypothyroidism in relation to nitrate intake from drinking water and dietary sources in a prospective cohort of older women in Iowa.

Early postnatal blood manganese levels and children's neurodevelopment.

Abstract: Background:Recent evidence suggests that low-level environmental exposure to manganese adversely affects child growth and neurodevelopment. Previous studies have addressed the effects of prenatal exposure, but little is known about developmental effects of early postnatal exposure.Methods:We studied

Effects of bushfire smoke on daily mortality and hospital admissions in Sydney, Australia.

Abstract: BACKGROUND:: Little research has investigated the health effects of particulate exposure from bushfires (also called wildfires, biomass fires, or vegetation fires), and these exposures are likely to increase, for several reasons. We investigated associations of daily mortality and hospital admissions with bushfire-derived particulates, compared with particulates from urban sources in Sydney, Australia from 1994 through 2002. METHODS:: On days with the highest particular matter (PM)10 concentrations, we assumed PM10 was due primarily to bushfires. We calculated the contribution of bushfire PM10 on these days by subtracting the background PM10 concentration estimated from surrounding days. We assumed PM10 on the remaining days was from usual urban sources. We implemented a Poisson model, with a bootstrap-based methodology, to select optimum smoothed covariate functions, and we estimated the effects of bushfire PM10 and urban PM10, lagged up to 3 days. RESULTS:: We identified 32 days with extreme PM10 concentrations due to bushfires or vegetation-reduction burns. Although bushfire PM10 was consistently associatedwith respiratory hospital admissions, we found no consistent associations with cardiovascular admissions or with mortality. A 10 mug/m increase in bushfire PM10 was associated with a 1.24% (95% confidence interval = 0.22% to 2.27%) increase in all respiratory disease admissions (at lag 0), a 3.80% (1.40% to 6.26%) increase in chronic obstructive pulmonary disease admissions (at lag 2), and a 5.02% (1.77% to 8.37%) increase in adult asthma admissions (at lag 0). Urban PM10 was associated with all-cause and cardiovascular mortality, as well as with cardiovascular and respiratory hospital admission, and these associations were not influenced by days with extreme PM10 concentrations. CONCLUSIONS:: PM10 from bushfires is associated primarily with respiratory morbidity, while PM10 from urban sources is associated with cardiorespiratory mortality and morbidity. Language: en

Prenatal exposure to fine particulate matter and birth weight: variations by particulate constituents and sources.

Abstract: Birthweight is a common indicator of fetal health, and low birth weight (LBW) is associated with risk of infant mortality,1 childhood morbidity,1 and diabetes.2 Links between mother’s exposure to particulate matter (PM) during pregnancy and birth weight have been observed in numerous epidemiologic studies. Several particle-size distributions have been considered. Exposure to particles with aerodynamic diameter ≤2.5 µm (PM2.5) has been associated with lower birth weight in North America (eg, California, Vancouver)3–6 and Europe.7 Exposure to particles with aerodynamic diameter ≤10 µm (PM10) exhibited similar associations in multiple locations,4–6 but not in all areas studied.8 Third-trimester PM10 exposure has been associated with lower birth weight in Southern California, but the effect was not robust to adjustment for ozone.9 Comparison across studies is hindered by differences in study designs,10 and the effects of prenatal exposures to particles on fetal growth are still not fully understood.11 However, collectively the results indicate a relationship between exposure to airborne particles during pregnancy and birth weight. The above-mentioned studies examine particles’ total mass although particles’ chemical composition varies substantially by season and region.12 For example, sulfate levels are higher in the Northeast than other US regions. Scientific evidence on which PM sources or chemical constituents have higher toxicity is one of the largest research gaps with respect to PM.13 Characterizing the health effects of PM components and sources was identified as a priority by a US National Research Council Committee.14 Health effects of various PM sources and constituents have been studied primarily for mortality and hospital admissions.15,16 To date, few studies have considered PM composition in relation to pregnancy outcomes. Some have investigated traffic-related air pollution more generally, using a traffic measure or indicator as a surrogate of exposure to traffic emissions. Lower birth weight has been linked to residence <50 m from a highway4 and to residential distance to major highways.17 PM2.5 absorbance (used to measure black carbon, a marker of traffic-related pollution)7 and distance-weighted traffic density18 have been associated with lower birth weight. We investigated whether exposures to PM2.5 sources and elemental constituents were associated with birth weight at term. Our previous work found higher effect estimates for an association between PM2.5 and lower birth weight among infants of African-American mothers than among infants of white mothers.6 Changes in birth weight may be particularly important for some minorities who are at higher risk of LBW.19 Therefore, we investigated whether any observed effect estimates differ by race.

Traffic-related air pollution and blood pressure in elderly subjects with coronary artery disease.

Abstract: Background:Associations between blood pressure (BP) and ambient air pollution have been inconsistent. No studies have used ambulatory BP monitoring and outdoor home air-pollutant measurements with time-activity-location data. We address these gaps in a study of 64 elderly subjects with coronary arte

Perinatal exposure to hazardous air pollutants and autism spectrum disorders at age 8.

Abstract: Hazardous air pollutants include hundreds of metal, particulate, and volatile organic compounds known to harm human health. Many are plausible candidate exposures in autism etiology because they have neurotoxic or immunotoxic properties,1 and theories of autism etiology include damage to the developing nervous system and immune perturbation.2–4 Airborne chemical exposures may be especially important in contributing to neurodevelopmental disorders because inhaled particles or metals have been found to be delivered directly to the brain through the olfactory bulb.5,6 Although hazardous air pollutants are not routinely monitored in the United States, emissions data have been used to model annual-average census-tract ambient concentrations in the National Air Toxics Assessment (NATA) program. A study in California studied hazardous air pollutants and autism,7 using NATA output to assign exposures. Children with autism spectrum disorders from a state developmental service agency and a large health maintenance organization were compared with children from birth certificate rosters. Hazardous air pollutants were examined individually and combined into composite scores based on chemical structure or mechanism (eg, endocrine disruptors). From a selected list of 25 air pollutants, positive associations were found for the chlorinated solvent group, the metal group, and several individual pollutants. Despite calls for data to fill gaps in our understanding of associations between environmental agents and autism,8,9 this California study remains one of the few studies contributing actual data. Intense investigation of many environmental agents is time- and cost-prohibitive. The NATA model presents an opportunity to screen a large number of pollutants with biologic plausibility for a role in autism. Results of such screenings can be considered in combination with toxicologic literature to identify the candidates that deserve more indepth research. Semi-Bayes hierarchical methods are useful in multiple-comparisons situations in which investigators must choose which comparisons to investigate further. This is true especially when there is substantial cost associated with future investigations,10,11 as there is with autism and environmental exposures. We conducted a screening analysis of hazardous air pollutants and autism spectrum disorders using a prevalent case-control design in North Carolina and West Virginia. As in the California study,7 exposure was assigned using output from the NATA model, but we additionally included some design improvements. We included a more complete group of autism spectrum disorders cases by reviewing records from schools in addition to other sources12 and verified residency of our control group at age 8. To account for the fact that hazardous air pollutants are highly correlated and that more than one of them might have effects on autism, we simultaneously adjusted for measured air pollutants and stabilized the estimates using semi-Bayes models.

Arsenic exposure and risk of spontaneous abortion, stillbirth, and infant mortality.

Abstract: Background: Millions of people worldwide are drinking water with elevated arsenic concentrations. Epidemiologic studies, mainly cross-sectional in design, have suggested that arsenic in drinking water may affect pregnancy outcome and infant health. We assessed the association of arsenic exposure with adverse pregnancy outcomes and infant mortality in a prospective cohort study of pregnant women. Methods: A population-based, prospective cohort study of 2924 pregnant women was carried out during 2002-2004 in Matlab, Bangladesh. Spontaneous abortion was evaluated in relation to urinary arsenic concentrations at gestational week 8. Stillbirth and infant mortality were evaluated in relation to the average of urinary arsenic concentrations measured at gestational weeks 8 and 30. Results: The odds ratio of spontaneous abortion was 1.4 ( 95% confidence interval [CI] = 0.96-2.2) among women with urine arsenic concentrations in the fifth quintile "(249-1253 mu g/L; median = 382 mu g/L), compared with women in the first quintile "(<33 mu g/L). There was no clear evidence of increased rates of stillbirth. The rate of infant mortality increased with increasing arsenic exposure: the hazard ratio was 5.0 (95% CI = 1.4-18) in the fifth quintile of maternal urinary arsenic concentrations (268-2019 mu g/L; median = 390 mu g/L), compared with the first quintile "(<38 mu g/L). Conclusions: We found evidence of increased risk of infant mortality with increasing arsenic exposure during pregnancy, with less evidence of associations with spontaneous abortion or stillbirth risk.

Impact of fine and ultrafine particles on emergency hospital admissions for cardiac and respiratory diseases.

Abstract: Background:Little is known about the short-term effects of ultrafine particles.Methods:We evaluated the effect of particulate matter with an aerodynamic diameter ≤10 μm (PM10), ≤2.5 μm (PM2.5), and ultrafine particles on emergency hospital admissions in Rome 2001–2005. We studied residents aged ≥35

Aircraft Noise, Air Pollution, and Mortality From Myocardial Infarction

Abstract: OBJECTIVE: Myocardial infarction has been associated with both transportation noise and air pollution. We examined residential exposure to aircraft noise and mortality from myocardial infarction, taking air pollution into account. METHODS:: We analyzed the Swiss National Cohort, which includes geocoded information on residence. Exposure to aircraft noise and air pollution was determined based on geospatial noise and airpollution (PM10) models and distance to major roads. We used Cox proportional hazard models, with age as the timescale. We compared the risk of death across categories of A-weighted sound pressure levels (dB(A)) and by duration of living in exposed corridors, adjusting for PM10 levels, distance to major roads, sex, education, and socioeconomic position of the municipality. RESULTS:: We analyzed 4.6 million persons older than 30 years who were followed from near the end of 2000 through December 2005, including 15,532 deaths from myocardial infarction (ICD-10 codes I 21, I 22). Mortality increased with increasing level and duration of aircraft noise. The adjusted hazard ratio comparing 45 dB(A) was 1.3 (95% confidence interval = 0.96-1.7) overall, and 1.5 (1.0-2.2) in persons who had lived at the same place for at least 15 years. None of the other endpoints (mortality from all causes, all circulatory disease, cerebrovascular disease, stroke, and lung cancer) was associated with aircraft noise. CONCLUSION:: Aircraft noise was associated with mortality from myocardial infarction, with a dose-response relationship for level and duration of exposure. The association does not appear to be explained by exposure to particulate matter air pollution, education, or socioeconomic status of the municipality.

Associations of Primary and Secondary Organic Aerosols With Airway and Systemic Inflammation in an Elderly Panel Cohort

Abstract: Background Exposure-response information about particulate air-pollution constituents is needed to protect sensitive populations. Particulate matter <2.5 mm (PM2.5) components may induce oxidative stress through reactive-oxygen-species generation, including primary organics from combustion sources and secondary organics from photochemically oxidized volatile organic compounds. We evaluated differences in airway versus systemic inflammatory responses to primary versus secondary organic particle components, particle size fractions, and the potential of particles to induce cellular production of reactive oxygen species.

Maternal use of acetaminophen, ibuprofen, and acetylsalicylic acid during pregnancy and risk of cryptorchidism

Abstract: Background:Cyclooxygenase (COX) inhibitors—acetaminophen, ibuprofen and acetylsalicylic acid—have endocrine-disruptive properties in the rainbow trout. In humans, aspirin blocks the androgen response to human chorionic gonadotropin (hCG), and, because hCG-stimulated androgen production in utero is c

From midlife to early old age: health trajectories associated with retirement.

Abstract: BACKGROUND: Previous studies report contradictory findings regarding health effects of retirement. This study examines longitudinally the associations of retirement with mental health and physical functioning. METHODS: The participants were 7584 civil servants from the Whitehall II cohort study aged 39-64 years at baseline and 54-76 years at the last follow-up. Self-reported mental health and physical functioning were assessed using the Short Form Medical Outcomes Survey questionnaire, and the scales were scored as T-scores (mean [SD] = 50 [10]). Retirement status and health were assessed with 6 repeated measurements over a 15-year period. RESULTS: The associations between retirement and health were dependent on age at retirement, reason for retirement, and length of time spent in retirement. Compared with continued employment, statutory retirement at age 60 and early voluntary retirement, respectively, were associated with 2.2 (95% confidence interval = 1.7 to 2.8) and 2.2 (1.7 to 2.7) points higher mental health and with 1.0 (0.6 to 1.5) and 1.1 (0.8 to 1.4) points higher physical functioning. Retirement due to ill health was associated with poorer mental health (-0.7 points [-1.62 to 0.2]) and physical functioning (-4.5 points [-5.1 to -3.9]). Within-subject analyses suggested a causal interpretation for statutory and voluntary retirement, but health selection for retirement due to ill health. CONCLUSIONS: Longitudinal analyses of repeat data suggest that health status improves after statutory and voluntarily retirement, although the improvement seems to attenuate over time. By contrast, the association between retirement due to ill health and subsequent poor health seems to reflect selection rather than causation.

Exposure to disinfection by-products, fetal growth, and prematurity: a systematic review and meta-analysis.

Abstract: Background:Exposure to total trihalomethanes in drinking water has been associated with several adverse birth outcomes relating to fetal growth and prematurity.Methods:We carried out a systematic review and meta-analysis of epidemiologic studies featuring original peer-reviewed data on the associati

Impact of pregnancy-induced hypertension on stillbirth and neonatal mortality.

Abstract: Hypertensive disorders of pregnancy complicate 5-8% of pregnancies and are associated with increased risks of perinatal morbidity and mortality,1-3 and maternal morbidity.4 Preeclampsia, part of the spectrum of pregnancy-induced hypertension (PIH), is typically a disease of the first pregnancy, with a reduction in incidence among multiparas.5 The occurrence of PIH in one pregnancy is a strong predictor of recurrence in the next pregnancy,6-9 and recurrent hypertensive disorders is associated with substantially higher risks of adverse perinatal outcomes.10 A study of first births in Norway between 1967-03 showed that the risk of stillbirth in relation to preeclampsia declined substantially between the periods 1967-78 (odds ratio 4.4) and 1991-03 (odds ratio 1.4).11 The rate of births at <32 weeks among preeclamptic mothers tripled during the study period (from 1.6% in 1967-78 to 5.0% in 1991-03), but the decline in stillbirth was not paralleled by a substantial increase in post-natal death.11 Whether similar patterns are evident among multiparous women remains unexplored. Rates of both preeclampsia and gestational hypertension have increased in the United States,12 which underscores the importance of evaluating the burden of perinatal mortality associated with these conditions. Among women who had had preeclampsia in a previous pregnancy, severe gestational hypertension without proteinuria in the next pregnancy was associated with increased risk of preterm birth and small for gestational babies than among women who developed recurrent mild preeclampsia.13 Hauth et al.14 reported higher rates of infant and maternal morbidities in healthy nulliparas who developed severe hypertension or preeclampsia. These studies were, however, relatively small, and mortality could not be properly assessed. Because limited attention has been devoted to pregnancy outcomes in multiparous women with hypertensive disorders of pregnancy, we investigated fetal and neonatal mortality in first and higher order singleton births in the United States. We compared births in 1990-91 to those in 2003-04, among Black and White women. These data, however, do not allow a distinction between preeclampsia and hypertension without proteinuria, and we thus used pregnancy-induced hypertension, comprising hypertension with and without proteinuria, as the entity of interest.

Intelligence in early adulthood and subsequent hospitalization for mental disorders.

Abstract: Background: lower intelligence is a risk factor for several specific mental disorders. It is unclear whether it is a risk factor for all mental disorders, and whether it might be associated with illness severity. We examined the relation of premorbid intelligence with risk of hospital admission and with total admission rates, for the whole range of mental disorders Methods: participants were 1,049,663 Swedish men who took tests of intelligence on conscription into military service and were followed up with regard to hospital admissions for mental disorder, for a mean of 22.6 years. International Classification of Diseases diagnoses were recorded at discharge from the hospital Results: risk of hospital admission for all categories of mental disorder rose with each point decrease in the 9-point IQ score. For a standard deviation decrease in IQ, age-adjusted hazard ratios (95% confidence interval) were 1.60 for schizophrenia (1.55–1.65), 1.49 for other nonaffective psychoses (1.45–1.53), 1.50 for mood disorders (1.47–1.51), 1.51 for neurotic disorders (1.48–1.54), 1.60 for adjustment disorders (1.56–1.64), 1.75 for personality disorders (1.70–1.80), 1.75 for alcohol-related (1.73–1.77), and 1.85 for other substance-use disorders (1.82–1.88). Lower intelligence was also associated with greater comorbidity. Associations changed little on adjustment for potential confounders. Men with lower intelligence had higher total admission rates for mental disorders, a possible marker of clinical severity Conclusions: lower intelligence is a risk factor for the whole range of mental disorders and for illness severity

Parental autoimmune diseases associated with autism spectrum disorders in offspring.

Abstract: Background: Autism spectrum disorders are often idiopathic. Studies have suggested associations between immune response and these disorders. We explored associations between parental autoimmune disorders and children's diagnosis of autism by linking Swedish registries. Methods: Data for each participant were linked across 3 Swedish registries. The study includes 1227 cases and 25 matched controls for each case (30,693 controls with parental linkage). Parental diagnoses comprised 19 autoimmune disorders. We estimated odds ratios (ORs) using multivariable conditional logistic regression. Results: Parental autoimmune disorder was weakly associated with autism spectrum disorders in offspring (maternal OR = 1.6 [95% confidence interval = 1.1–2.2]; paternal OR = 1.4 [1.0–2.0]). Several maternal autoimmune diseases were correlated with autism. For both parents, rheumatic fever was associated with autism spectrum disorders. Conclusions: These data support previously reported associations between parental autoimmune disorders and autism spectrum disorders. Parental autoimmune disorders may represent a critical pathway that warrants more detailed investigation.

The relationship between neighborhood poverty and alcohol use: estimation by marginal structural models.

Abstract: Several studies have reported that alcohol abuse and dependence, as well as other risk behaviors, cluster in contexts of poverty, residential instability, and social isolation.1–5 Most of these studies are cross-sectional and do not account for the fact that neighborhoods change over time, or allow us to assess how such changes might affect alcohol misuse. The question remains whether such multilevel associations are actually due to the influence of neighborhood contextual characteristics on health outcomes such as alcohol abuse, or whether they merely reflect the selection of persons with similar socioeconomic characteristics and health problems into particular types of neighborhoods. Longitudinal studies that follow people and neighborhoods over time are needed to better estimate the nature of the association of neighborhood conditions with alcohol use. Traditionally, longitudinal studies examining the association between neighborhood characteristics and risk behaviors have attempted to address individual selection into neighborhoods by using standard regression models or propensity-score analysis to control tightly for individual-level characteristics (such as socioeconomic position) that are causally related both to the type of neighborhood a person lives in and to the person's use of alcohol. A major concern with such methods, however, is that many of the time-varying potential confounders are also affected by prior neighborhood conditions, and are thus in the causal pathway between the exposure of interest and the outcome, at the same time that they affect the types of neighborhoods that persons move into.6 Individual socioeconomic position, for example, not only contributes to the type of neighborhood a person can afford to live in and the level of alcohol consumed, but it is also a product of the types of income-generating opportunities afforded by the neighborhood socioeconomic environment.5 By controlling for the individual-level composition of neighborhoods, to address individual selection into neighborhoods, traditional regression analytic techniques run the risk of also controlling for individual-level mediators of earlier neighborhood characteristics and thus underestimating the impact that long-term cumulative neighborhood exposure has on health outcomes. Unadjusted estimates are confounded by individual-level characteristics related to selection of persons into neighborhoods.7 Marginal structural models offer a particularly useful tool for research on neighborhoods and health, where there are often time-dependent covariates that act simultaneously as confounders and as intermediate variables in the causal pathway between the neighborhood exposure of interest and the outcome.8 Marginal structural models describe the marginal causal relationship between a time-varying exposure such as neighborhood poverty and alcohol use, and therefore, allow us to control for time-varying confounders without conditioning on these variables. Formally, a marginal structural model for repeated measures is a parametric regression model relating any possible exposure history, up to time t, to the corresponding counterfactual outcome at time t. Marginal structural models are also useful in the case of loss-to-follow-up in longitudinal studies, because they allow us to account for differential loss to follow-up. Assuming ignorable treatment assignment and the absence of differential misclassification,9 the parameters of a marginal structural model can be estimated in an unbiased manner with inverse probability-of-treatment and censoring weighting. This is a product of inverse probability-of-treatment weights and inverse probability-of-censoring weights. Such weighting makes it possible to obtain a comparable “pseudopopulation” in terms of stable and time-varying confounders across levels of the exposure, and thus estimate the unconfounded association between the exposure and outcome without conditioning on the covariate through its inclusion as a predictor in the outcome model.8 A detailed example illustrating how weighting creates an unbiased “pseudopopulation” is provided in the eAppendix (http://links.lww.com/EDE/A397). Using data from a population-based longitudinal study of young adults, we investigated the potentially causal association of neighborhood poverty with 2 important aspects of alcohol consumption: frequency of alcohol consumption and binging. These 2 types of alcohol-related behavior may present contrasting etiologies,7 and neighborhood poverty may have a stronger impact on heavy alcohol consumption than on the consumption gradient. We used marginal structural models to estimate the relationship between cumulative and point-in-time neighborhood poverty and alcohol use behaviors, after appropriately accounting for time-dependent confounders and for loss to follow-up. A directed acyclic graph illustrating the relationship is found in eAppendix (http://links.lww.com/EDE/A397). Because neighborhood poverty is a continuous exposure, pooled logistic regression is not appropriate for estimating inverse probability-of-treatment and censoring weighting treatment weights. Instead, our approach makes use of an estimated log-normal exposure probability density function that correctly accounts for the highly skewed continuous nature of the exposure of interest.

Paraoxonase 1, agricultural organophosphate exposure, and Parkinson disease.

Abstract: Background: Human, animal and cell models support a role for pesticides in the etiology of Parkinson disease. Susceptibility to pesticides may be modified by genetic variants of xenobiotic enzymes, such as paraoxonase, that play a role in metabolizing some organophosphates. Methods: We examined associations between Parkinson disease and the organophosphates diazinon, chlorpyrifos, and parathion, and the influence of a functional polymorphism at position 55 in the coding region of the PON1 gene (PON1-55). From 1 January 2001 through 1 January 2008, we recruited 351 incident cases and 363 controls from 3 rural California counties in a population-based case-control study. Participants provided a DNA sample, and residential exposure to organophosphates was determined from pesticide usage reports and a geographic information system (GIS) approach. We assessed the main effects of both genes and pesticides in unconditional logistic regression analyses, and evaluated the effect of carrying a PON1-55 MM variant on estimates of effects for diazinon, chlorpyrifos, and parathion exposures. Results: Carriers of the variant MM PON1-55 genotype exposed to organophosphates exhibited a greater than 2-fold increase in Parkinson disease risk compared with persons who had the wildtype or heterozygous genotype and no exposure (for diazinon, odds ratio 2.2 95% confidence interval 1.1‐4.5; for chlorpyrifos, 2.6 1.3‐5.4). The effect estimate for chlorpyrifos, was more pronounced in younger-onset cases and controls (60 years) (5.3 1.7‐16). No increase in risk was noted for parathion. Conclusion: The increase in risk we observed among PON1-55 variant carriers for specific organophosphates metabolized by PON1 underscores the importance of considering susceptibility factors when studying environmental exposures in Parkinson disease. (Epidemiology 2010;21: 87‐94)

Witness of intimate partner violence in childhood and perpetration of intimate partner violence in adulthood.

Abstract: BACKGROUND:: At least half a million women are victims of intimate partner violence in the United States annually, resulting in substantial harm. However, the etiology of violence to intimate partners is not well understood. Witnessing such violence in childhood has been proposed as a principal cause of adulthood perpetration, yet it remains unknown whether the association between witnessing intimate partner violence and adulthood perpetration is causal. METHOD:: We conducted a propensity-score analysis of intimate partner violence perpetration to determine whether childhood witnessing is associated with perpetration in adulthood, independent of a wide range of potential confounding variables, and therefore might be a causal factor. We used data from 14,564 U.S. men ages 20 and older from the 2004-2005 wave of the National Epidemiologic Survey on Alcohol and Related Conditions. RESULTS:: Nearly 4% of men reported violent behavior toward an intimate partner in the past year. In unadjusted models, we found a strong association between childhood witnessing of intimate partner violence and adulthood perpetration (for witnessing any intimate partner violence, risk ratio [RR] = 2.6 [95% confidence interval = 2.1-3.2]; for witnessing frequent or serious violence, 3.0 [2.3-3.9]). In propensity-score models, the association was substantially attenuated (for witnessing any intimate partner violence, adjusted RR = 1.6 [1.2-2.0]; for witnessing frequent or serious violence, 1.6 [1.2-2.3]). CONCLUSIONS:: Men who witness intimate partner violence in childhood are more likely to commit such acts in adulthood, compared with men who are otherwise similar with respect to a large range of potential confounders. Etiological models of intimate partner violence perpetration should consider a constellation of childhood factors. Language: en

On the Consistency Rule in Causal Inference: Axiom, Definition, Assumption, or Theorem?

Abstract: Remove the abstract (and move to the Main Text). In two recent communications, Cole and Frangakis and VanderWeele conclude that the consistency rule used in causal inference is an assumption that precludes any side-effects of treatment/exposure on the outcomes of interest. They further develop auxiliary notation to make this assumption formal and explicit. I argue that the consistency rule is a theorem in the logic of counterfactuals and need not be altered, even in cases where different versions of treatment/exposure have side effects on the outcome. Instead, warnings of potential side-effects should be embodied in standard modeling practices, using graphs or structural equation models, in which causal assumptions are given transparent and unambiguous representation. In two recent communications, Cole and Frangakis 1 and VanderWeele 2 interpret the consistency rule of causal inference an assumption that is violated whenever versions of treatment/exposure have unintended side effects on the outcomes of interest. They further develop auxiliary notation to make this assumption formal and explicit. I argue that the consistency rule is in fact a theorem in the logic of counterfactuals and need not be altered. Side-effects can be made explicit in standard modeling practices, whenever transparent models are used. Informally, the consistency rule states that a person’s potential outcome under a hypothetical condition that happened to materialize is precisely the outcome experienced by that person. When expressed formally, this rule reads 3 :