Differential respiratory health effects from the 2008 northern California wildfires: A spatiotemporal approach.

We investigated health effects associated with fine particulate matter during a long-lived, large wildfire complex in northern California in the summer of 2008. We estimated exposure to PM2.5 for each day using an exposure prediction model created through data-adaptive machine learning methods from a large set of spatiotemporal data sets. We then used Poisson generalized estimating equations to calculate the effect of exposure to 24-hour average PM2.5 on cardiovascular and respiratory hospitalizations and ED visits. We further assessed effect modification by sex, age, and area-level socioeconomic status (SES). We observed a linear increase in risk for asthma hospitalizations (RR=1.07, 95% CI=(1.05, 1.10) per 5µg/m(3) increase) and asthma ED visits (RR=1.06, 95% CI=(1.05, 1.07) per 5µg/m(3) increase) with increasing PM2.5 during the wildfires. ED visits for chronic obstructive pulmonary disease (COPD) were associated with PM2.5 during the fires (RR=1.02 (95% CI=(1.01, 1.04) per 5µg/m(3) increase) and this effect was significantly different from that found before the fires but not after. We did not find consistent effects of wildfire smoke on other health outcomes. The effect of PM2.5 during the wildfire period was more pronounced in women compared to men and in adults, ages 20-64, compared to children and adults 65 or older. We also found some effect modification by area-level median income for respiratory ED visits during the wildfires, with the highest effects observed in the ZIP codes with the lowest median income. Using a novel spatiotemporal exposure model, we found some evidence of differential susceptibility to exposure to wildfire smoke.

Framing air pollution epidemiology in terms of population interventions, with applications to multipollutant modeling.

Air pollution epidemiology continues moving toward the study of mixtures and multipollutant modeling. Simultaneously, there is a movement in epidemiology to estimate policy-relevant health effects that can be understood in reference to specific interventions. Scaling regression coefficients from a regression model by an interquartile range (IQR) is one common approach to presenting multipollutant health effect estimates. We are unaware of guidance on how to interpret these effect estimates as an intervention. To illustrate the issues of interpretability of IQR-scaled air pollution health effects, we analyzed how daily concentration changes in 2 air pollutants (nitrogen dioxide and particulate matter with aerodynamic diameter ≤ 2.5 µm) related to one another within 2 seasons (summer and winter), within 3 cities with distinct air pollution profiles (Burbank, California; Houston, Texas; and Pittsburgh, Pennsylvania). In each city season, we examined how realistically IQR scaling in multipollutant lag-1 time-series studies reflects a hypothetical intervention that is possible given the observed data. We proposed 2 causal conditions to explicitly link IQR-scaled effects to a clearly defined hypothetical intervention. Condition 1 specified that the index pollutant had to experience a daily concentration change of greater than 1 IQR, reflecting the notion that the IQR is an appropriate measure of variability between consecutive days. Condition 2 specified that the copollutant had to remain relatively constant. We found that in some city seasons, there were very few instances in which these conditions were satisfied (eg, 1 day in Pittsburgh during summer). We discuss the practical implications of IQR scaling and suggest alternative approaches to presenting multipollutant effects that are supported by empirical data.

Air Pollution, Neighbourhood Socioeconomic Factors, and Neural Tube Defects in the San Joaquin Valley of California.

BACKGROUND: Environmental pollutants and neighbourhood socioeconomic factors have been associated with neural tube defects, but the potential impact of interaction between ambient air pollution and neighbourhood socioeconomic factors on the risks of neural tube defects is not well understood. METHODS: We used data from the California Center of the National Birth Defects Study and the Children's Health and Air Pollution Study to investigate whether associations between air pollutant exposure in early gestation and neural tube defects were modified by neighbourhood socioeconomic factors in the San Joaquin Valley of California, 1997-2006. There were 5 pollutant exposures, 3 outcomes, and 9 neighbourhood socioeconomic factors included for a total of 135 investigated associations. Estimates were adjusted for maternal race-ethnicity, education, and multivitamin use. RESULTS: We present below odds ratios (ORs) that exclude 1 and a chi-square test of homogeneity P-value of <0.05. We observed increased odds of spina bifida comparing the highest to lowest quartile of particulate matter <10 µm (PM10 ) among those living in a neighbourhood with: (i) median household income of less than $30 000 per year [OR 5.1, 95% confidence interval (CI) 1.7, 15.3]; (ii) more than 20% living below the federal poverty level (OR 2.6, 95% CI 1.1, 6.0); and (iii) more than 30% with less than or equal to a high school education (OR 3.2, 95% CI 1.4, 7.4). The ORs were not statistically significant among those higher socioeconomic status (SES) neighbourhoods. CONCLUSIONS: Our results demonstrate effect modification by neighbourhood socioeconomic factors in the association of particulate matter and neural tube defects in California.

Spatiotemporal prediction of fine particulate matter during the 2008 northern California wildfires using machine learning.

Estimating population exposure to particulate matter during wildfires can be difficult because of insufficient monitoring data to capture the spatiotemporal variability of smoke plumes. Chemical transport models (CTMs) and satellite retrievals provide spatiotemporal data that may be useful in predicting PM2.5 during wildfires. We estimated PM2.5 concentrations during the 2008 northern California wildfires using 10-fold cross-validation (CV) to select an optimal prediction model from a set of 11 statistical algorithms and 29 predictor variables. The variables included CTM output, three measures of satellite aerosol optical depth, distance to the nearest fires, meteorological data, and land use, traffic, spatial location, and temporal characteristics. The generalized boosting model (GBM) with 29 predictor variables had the lowest CV root mean squared error and a CV-R2 of 0.803. The most important predictor variable was the Geostationary Operational Environmental Satellite Aerosol/Smoke Product (GASP) Aerosol Optical Depth (AOD), followed by the CTM output and distance to the nearest fire cluster. Parsimonious models with various combinations of fewer variables also predicted PM2.5 well. Using machine learning algorithms to combine spatiotemporal data from satellites and CTMs can reliably predict PM2.5 concentrations during a major wildfire event.

Exposure to airborne polycyclic aromatic hydrocarbons during pregnancy and risk of preterm birth.

BACKGROUND: Preterm birth is an important marker of health and has a prevalence of 12-13% in the U.S. Polycyclic aromatic hydrocarbons (PAHs) are a group of organic contaminants that form during the incomplete combustion of hydrocarbons, such as coal, diesel and gasoline. Studies suggest that exposure to PAHs during pregnancy is related to adverse birth outcomes. The aim of this study is to evaluate the association between exposure to PAHs during the pregnancy and preterm birth. METHODS: The study population included births from years 2001 to 2006 of women whose maternal residence was within 20km of the primary monitoring site in Fresno, California. Data in the Fresno area were used to form a spatio-temporal model to assign daily exposure to PAHs with 4, 5, or 6 rings at the maternal residence throughout pregnancy of all of the births in the study area. Gestational age at birth and relevant covariates were extracted from the birth certificate. RESULTS: We found an association between PAHs during the last 6 weeks of pregnancy and birth at 20-27 weeks (OR=2.74; 95% CI: 2.24-3.34) comparing the highest quartile to the lower three. The association was consistent when each quartile was compared to the lowest (OR2nd=1.49, 95% CI: 1.08-2.06; OR3rd=2.63, 95% CI:1.93-3.59; OR4th=3.94, 95% CI:3.03-5.12). Inverse associations were also observed for exposure to PAHs during the entire pregnancy and the first trimester and birth at 28-31 weeks and 20-27 weeks. CONCLUSION: An association between PAH exposure during the 6 weeks before delivery and early preterm birth was observed. However, the inverse association with early preterm birth offers an unclear, and potentially complex, inference of these associations.

The self-reported health of U.S. flight attendants compared to the general population.

BACKGROUND: Few studies have examined the broad health effects of occupational exposures in flight attendants apart from disease-specific morbidity and mortality studies. We describe the health status of flight attendants and compare it to the U.S. population. In addition, we explore whether the prevalence of major health conditions in flight attendants is associated with length of exposure to the aircraft environment using job tenure as a proxy. METHODS: We surveyed flight attendants from two domestic U.S. airlines in 2007 and compared the prevalence of their health conditions to contemporaneous cohorts in the National Health and Nutrition Survey (NHANES), 2005-2006 and 2007-2008. We weighted the prevalence of flight attendant conditions to match the age distribution in the NHANES and compared the two populations stratified by gender using the Standardized Prevalence Ratio (SPR). For leading health conditions in flight attendants, we analyzed the association between job tenure and health outcomes in logistic regression models. RESULTS: Compared to the NHANES population (n =5,713), flight attendants (n = 4,011) had about a 3-fold increase in the age-adjusted prevalence of chronic bronchitis despite considerably lower levels of smoking. In addition, the prevalence of cardiac disease in female flight attendants was 3.5 times greater than the general population while their prevalence of hypertension and being overweight was significantly lower. Flight attendants reported 2 to 5.7 times more sleep disorders, depression, and fatigue, than the general population. Female flight attendants reported 34% more reproductive cancers. Health conditions that increased with longer job tenure as a flight attendant were chronic bronchitis, heart disease in females, skin cancer, hearing loss, depression and anxiety, even after adjusting for age, gender, body mass index (BMI), education, and smoking. CONCLUSIONS: This study found higher rates of specific diseases in flight attendants than the general population. Longer tenure appears to explain some of the higher disease prevalence. Conclusions are limited by the cross-sectional design and recall bias. Further study is needed to determine the source of risk and to elucidate specific exposure-disease relationships over time.

The association of ambient air pollution and traffic exposures with selected congenital anomalies in the San Joaquin Valley of California.

Congenital anomalies are a leading cause of infant mortality and are important contributors to subsequent morbidity. Studies suggest associations between environmental contaminants and some anomalies, although evidence is limited. We aimed to investigate whether ambient air pollutant and traffic exposures in early gestation contribute to the risk of selected congenital anomalies in the San Joaquin Valley of California, 1997-2006. Seven exposures and 5 outcomes were included for a total of 35 investigated associations. We observed increased odds of neural tube defects when comparing the highest with the lowest quartile of exposure for several pollutants after adjusting for maternal race/ethnicity, education, and multivitamin use. The adjusted odds ratio for neural tube defects among those with the highest carbon monoxide exposure was 1.9 (95% confidence interval: 1.1, 3.2) compared with those with the lowest exposure, and there was a monotonic exposure-response across quartiles. The highest quartile of nitrogen oxide exposure was associated with neural tube defects (adjusted odds ratio = 1.8, 95% confidence interval: 1.1, 2.8). The adjusted odds ratio for the highest quartile of nitrogen dioxide exposure was 1.7 (95% confidence interval: 1.1, 2.7). Ozone was associated with decreased odds of neural tube defects. Our results extend the limited body of evidence regarding air pollution exposure and adverse birth outcomes.

Traffic-related air pollution and selected birth defects in the San Joaquin Valley of California.

BACKGROUND: Birth defects are a leading cause of infant morbidity and mortality. Studies suggest associations between environmental contaminants and some structural anomalies, although evidence is limited and several anomalies have not been investigated previously. METHODS: We used data from the California Center of the National Birth Defects Prevention Study and the Children's Health and Air Pollution Study to estimate the odds of 26 congenital birth defect phenotypes with respect to quartiles of seven ambient air pollutant and traffic exposures in California during the first 2 months of pregnancy, 1997 to 2006 (874 cases and 849 controls). We calculated odds ratios (adjusted for maternal race/ethnicity, education, and vitamin use; aOR) for 11 phenotypes that had at least 40 cases. RESULTS: Few odds ratios had confidence intervals that did not include 1.0. Odds of esophageal atresia were increased for the highest versus lowest of traffic density (aOR = 2.8, 95% confidence interval [CI], 1.1-7.4) and PM10 exposure (aOR 4.9; 95% CI, 1.4-17.2). PM10 was associated with a decreased risk of hydrocephaly (aOR= 0.3; 95% CI, 0.1-0.9) and CO with decreased risk of anotia/microtia (aOR = 0.4; 95% CI, 0.2-0.8) and transverse limb deficiency (aOR = 0.4; 95% CI, 0.2-0.9), again reflecting highest versus lowest quartile comparisons. CONCLUSION: Most analyses showed no substantive association between air pollution and the selected birth defects with few exceptions of mixed results.

Ambient air pollution and traffic exposures and congenital heart defects in the San Joaquin Valley of California.

BACKGROUND: Congenital anomalies are a leading cause of infant morbidity and mortality. Studies suggest associations between environmental contaminants and some anomalies, although evidence is limited. METHODS: We used data from the California Center of the National Birth Defects Prevention Study and the Children's Health and Air Pollution Study to estimate the odds of 27 congenital heart defects with respect to quartiles of seven ambient air pollutant and traffic exposures in California during the first 2 months of pregnancy, 1997-2006 (n = 822 cases and n = 849 controls). RESULTS: Particulate matter < 10 microns (PM10 ) was associated with pulmonary valve stenosis [adjusted odds ratio (aOR)Fourth Quartile = 2.6] [95% confidence intervals (CI) 1.2, 5.7] and perimembranous ventricular septal defects (aORThird Quartile = 2.1) [95% CI 1.1, 3.9] after adjusting for maternal race/ethnicity, education and multivitamin use. PM2.5 was associated with transposition of the great arteries (aORThird Quartile = 2.6) [95% CI 1.1, 6.5] and inversely associated with perimembranous ventricular septal defects (aORFourth Quartile = 0.5) [95% CI 0.2, 0.9]. Secundum atrial septal defects were inversely associated with carbon monoxide (aORFourth Quartile = 0.4) [95% CI 0.2, 0.8] and PM2.5 (aORFourth Quartile = 0.5) [95% CI 0.3, 0.8]. Traffic density was associated with muscular ventricular septal defects (aORFourth Quartile = 3.0) [95% CI 1.2, 7.8] and perimembranous ventricular septal defects (aORThird Quartile = 2.4) [95% CI 1.3, 4.6], and inversely associated with transposition of the great arteries (aORFourth Quartile = 0.3) [95% CI 0.1, 0.8]. CONCLUSIONS: PM10 and traffic density may contribute to the occurrence of pulmonary valve stenosis and ventricular septal defects, respectively. The results were mixed for other pollutants and had little consistency with previous studies.

Mapping and modeling airborne urban phenanthrene distribution using vegetation biomonitoring.

To capture the spatial distribution of phenanthrene in an urban setting we used vegetation biomonitoring with Jeffrey pine trees (Pinus jeffreyi). The major challenge in characterizing spatial variation in polycyclic aromatic hydrocarbon (PAH) concentrations within a metropolitan area has been sampling at a fine enough resolution to observe the underlying spatial pattern. However, field and chamber studies show that the primary pathway through which PAHs enter plants is from air into leaves, making vegetation biomonitoring a feasible way to examine the spatial distribution of these compounds. Previous research has shown that phenanthrene has adverse health effects and that it is one of the most abundant PAHs in urban air. We collected 99 pine needle samples from 91 locations in Fresno in the morning on a winter day, and analyzed them for PAHs in the inner needle. All 99 pine needle samples had detectable levels of phenanthrene, with mean concentration of 41.0 ng g-1, median 36.9 ng g-1, and standard deviation of 28.5 ng g-1 fresh weight. The ratio of the 90th:10th percentile concentrations by location was 3.3. The phenanthrene distribution had a statistically significant Moran's I of 0.035, indicating a high degree of spatial clustering. We implemented land use regression to fit a model to our data. Our model was able to explain a moderate amount of the variability in the data (R 2 = 0.56), likely reflecting the major sources of phenanthrene in Fresno. The spatial distribution of modeled airborne phenanthrene shows the influences of highways, railroads, and industrial and commercial zones.

Exposure to traffic-related air pollution during pregnancy and term low birth weight: estimation of causal associations in a semiparametric model.

Traffic-related air pollution is recognized as an important contributor to health problems. Epidemiologic analyses suggest that prenatal exposure to traffic-related air pollutants may be associated with adverse birth outcomes; however, there is insufficient evidence to conclude that the relation is causal. The Study of Air Pollution, Genetics and Early Life Events comprises all births to women living in 4 counties in California's San Joaquin Valley during the years 2000-2006. The probability of low birth weight among full-term infants in the population was estimated using machine learning and targeted maximum likelihood estimation for each quartile of traffic exposure during pregnancy. If everyone lived near high-volume freeways (approximated as the fourth quartile of traffic density), the estimated probability of term low birth weight would be 2.27% (95% confidence interval: 2.16, 2.38) as compared with 2.02% (95% confidence interval: 1.90, 2.12) if everyone lived near smaller local roads (first quartile of traffic density). Assessment of potentially causal associations, in the absence of arbitrary model assumptions applied to the data, should result in relatively unbiased estimates. The current results support findings from previous studies that prenatal exposure to traffic-related air pollution may adversely affect birth weight among full-term infants.

Causal inference in epidemiological studies with strong confounding.

One of the identifiability assumptions of causal effects defined by marginal structural model (MSM) parameters is the experimental treatment assignment (ETA) assumption. Practical violations of this assumption frequently occur in data analysis when certain exposures are rarely observed within some strata of the population. The inverse probability of treatment weighted (IPTW) estimator is particularly sensitive to violations of this assumption; however, we demonstrate that this is a problem for all estimators of causal effects. This is due to the fact that the ETA assumption is about information (or lack thereof) in the data. A new class of causal models, causal models for realistic individualized exposure rules (CMRIER), is based on dynamic interventions. CMRIER generalize MSM, and their parameters remain fully identifiable from the observed data, even when the ETA assumption is violated, if the dynamic interventions are set to be realistic. Examples of such realistic interventions are provided. We argue that causal effects defined by CMRIER may be more appropriate in many situations, particularly those with policy considerations. Through simulation studies, we examine the performance of the IPTW estimator of the CMRIER parameters in contrast to that of the MSM parameters. We also apply the methodology to a real data analysis in air pollution epidemiology to illustrate the interpretation of the causal effects defined by CMRIER.

Usual physical activity and hip fracture in older men: an application of semiparametric methods to observational data.

Few studies have examined the relation between usual physical activity level and rate of hip fracture in older men or applied semiparametric methods from the causal inference literature that estimate associations without assuming a particular parametric model. Using the Physical Activity Scale for the Elderly, the authors measured usual physical activity level at baseline (2000-2002) in 5,682 US men ≥65 years of age who were enrolled in the Osteoporotic Fractures in Men Study. Physical activity levels were classified as low (bottom quartile of Physical Activity Scale for the Elderly score), moderate (middle quartiles), or high (top quartile). Hip fractures were confirmed by central review. Marginal associations between physical activity and hip fracture were estimated with 3 estimation methods: inverse probability-of-treatment weighting, G-computation, and doubly robust targeted maximum likelihood estimation. During 6.5 years of follow-up, 95 men (1.7%) experienced a hip fracture. The unadjusted risk of hip fracture was lower in men with a high physical activity level versus those with a low physical activity level (relative risk = 0.51, 95% confidence interval: 0.28, 0.92). In semiparametric analyses that controlled confounding, hip fracture risk was not lower with moderate (e.g., targeted maximum likelihood estimation relative risk = 0.92, 95% confidence interval: 0.62, 1.44) or high (e.g., targeted maximum likelihood estimation relative risk = 0.88, 95% confidence interval: 0.53, 2.03) physical activity relative to low. This study does not support a protective effect of usual physical activity on hip fracture in older men.

Medication use patterns among urban youth participating in school-based asthma education.

Although pharmaceutical management is an integral part of asthma control, few community-based analyses have focused on this aspect of disease management. The primary goal of this analysis was to assess whether participation in the school-based Kickin' Asthma program improved appropriate asthma medication use among middle school students. A secondary goal was to determine whether improvements in medication use were associated with subsequent improvements in asthma-related symptoms among participating students. Students completed an in-class case-identification questionnaire to determine asthma status. Eligible students were invited to enroll in a school-based asthma curriculum delivered over four sessions by an asthma health educator. Students completed a pre-survey and a 3-month follow-up post-survey that compared symptom frequency and medication use. From 2004 to 2007, 579 participating students completed pre- and post-surveys. Program participation resulted in improvements in appropriate use across all three medication use categories: 20.0% of students initiated appropriate reliever use when "feeling symptoms" (p < 0.001), 41.6% of students reporting inappropriate medication use "before exercise" initiated reliever use (p < 0.001), and 26.5% of students reporting inappropriate medication use when "feeling fine" initiated controller use (p < 0.02). More than half (61.6%) of participants reported fewer symptoms at post-survey. Symptom reduction was not positively associated with improvements in medication use in unadjusted and adjusted analysis, controlling for sex, asthma symptom classification, class attendance, season, and length of follow-up. Participation in a school-based asthma education program significantly improved reliever medication use for symptom relief and prior-to-exercise and controller medication use for maintenance. However, given that symptom reduction was not positively associated with improvement in medication use, pharmaceutical education must be just one part of a comprehensive asthma management agenda that addresses the multifactorial nature of asthma-related morbidity.

Ambient air pollution impairs regulatory T-cell function in asthma.

BACKGROUND: Asthma is the most frequent chronic disease in children, and children are at high risk for adverse health consequences associated with ambient air pollution (AAP) exposure. Regulatory T (Treg) cells are suppressors of immune responses involved in asthma pathogenesis. Treg-cell impairment is associated with increased DNA methylation of Forkhead box transcription factor 3 (Foxp3), a key transcription factor in Treg-cell activity. Because AAP exposure can induce epigenetic changes, we hypothesized that Treg-cell function would be impaired by AAP, allowing amplification of an inflammatory response. OBJECTIVES: To assess whether exposure to AAP led to hypermethylation of the Foxp3 gene, causing impaired Treg-cell suppression and worsened asthma symptom scores. METHODS: Children with and without asthma from Fresno, Calif (high pollution, Fresno Asthma Group [FA], n = 71, and Fresno Non Asthmatic Group, n = 30, respectively), and from Stanford, Calif (low pollution, Stanford Asthma Group, n = 40, and Stanford Non Asthmatic Group, n = 40), were enrolled in a cross-sectional study. Peripheral blood Treg cells were used in functional and epigenetic studies. Asthma outcomes were assessed by Global Initiative in Asthma score. RESULTS: Fresno Asthma Group Treg-cell suppression was impaired and FA Treg-cell chemotaxis were reduced compared with other groups (P ≤ .05). Treg-cell dysfunction was associated with more pronounced decreases in asthma Global Initiative in Asthma score in FA versus the Stanford Asthma Group. Foxp3 was decreased in FA compared with the Fresno Non Asthmatic Group (P ≤ .05). FA also contained significantly higher levels of methylation at the Foxp3 locus (P ≤ .05). CONCLUSION: Increased exposure to AAP is associated with hypermethylation of the Foxp3 locus, impairing Treg-cell function and increasing asthma morbidity. AAP could play a role in mediating epigenetic changes in Treg cells, which may worsen asthma by an immune mechanism.

The association between persistent fetal occiput posterior position and perinatal outcomes: an example of propensity score and covariate distance matching.

In a retrospective cohort study of 18,880 full-term, cephalic singletons born in San Francisco, California, during 1976-2001, the authors used multivariable logistic regression (MVLR) and propensity score analysis (PSA) to examine the association between persistent fetal occiput posterior (OP) position and perinatal outcomes. The principles and applications of these techniques are compared and discussed. Pregnancies with OP positions at delivery were compared with those with occiput anterior positions. Perinatal outcomes were examined as adjusted odds ratios determined by MVLR and PSA and as risk differences determined by propensity score matched bootstrapping based on covariate distance. Persistent OP position was associated with operative delivery and maternal morbidity. The odds ratio estimates based on PSA were somewhat larger than those obtained with standard MVLR, and the confidence intervals were narrower. When statistical inference was evaluated with the permutation test, the results were more consistent with the PSA. These analyses demonstrate that PSA is likely to provide more precise estimates of exposure associations and more reliable statistical inferences than MVLR. The authors show that PSA can be extended with Mahalanobis distance matching to obtain estimates of risk difference between exposed and unexposed subjects that avoid violations of the experimental treatment assignment (positivity) assumption that is required for valid causal inference.

Ambient ozone concentrations and cardiac mortality in Southern California 1983-2000: application of a new marginal structural model approach.

The authors evaluated the association between ambient ozone levels and cardiac mortality in California's South Coast Air Basin during the period 1983-2000 and compared inferences from several types of marginal structural model (MSM) estimators. The authors undertook an ecologic study during the high-ozone seasons among persons over age 55 years. In contrast to conditional regression analysis and MSMs based on G-computation and simple inverse probability-of-treatment weighting (IPTW), an MSM that protected against violation of the experimental treatment assignment (ETA) assumption and considered only those areas that could have experienced both high and low ozone concentrations during 1983-2000 found no consistent evidence that reductions in quarterly 1-hour maximum ozone concentrations from levels above any of the regulatory standards to levels below those standards led to decreases in cardiac mortality; however, it did find evidence of decreases related to a decrease in 8-hour maximum concentrations. The G-computation estimator and simple IPTW estimators were biased because of serious violation of the ETA assumption. These analyses highlight the importance of nonviolation of the ETA assumption for valid inference and the failure of conditional regression to provide marginal estimates in the presence of interactions. Noncausal models also consistently inferred larger associations, which may have been due to bias violation of the ETA assumption on which these models rely.

Secondhand smoke, vascular disease, and dementia incidence: findings from the cardiovascular health cognition study.

Recent studies have found that smoking is associated with an increased risk of dementia, but the effects of secondhand smoke (SHS) on dementia risk are not known to have previously been studied. The authors used Cox proportional hazards marginal structural models to examine the association between self-reported lifetime household SHS exposure and risk of incident dementia over 6 years among 970 US participants in the Cardiovascular Health Cognition Study (performed from 1991 to 1999) who were never smokers and were free of clinical cardiovascular disease (CVD), dementia, and mild cognitive impairment at baseline. In addition, because prior studies have found that SHS is associated with increased risk of CVD and that CVD is associated with increased risk of dementia, the authors tested for interactions between SHS and measures of clinical and subclinical CVD on dementia risk. Moderate (16-25 years) and high (>25 years) SHS exposure levels were not independently associated with dementia risk; however, subjects with >25 years of SHS exposure and >25% carotid artery stenosis had a 3-fold increase (hazard ratio = 3.00, 95% confidence interval: 1.03, 9.72) in dementia risk compared with subjects with no/low (0-15 years) SHS exposure and < or =25% carotid artery stenosis. High lifetime SHS exposure may increase the risk of dementia in elderly with undiagnosed CVD.

Age and cystatin C in healthy adults: a collaborative study.

BACKGROUND: Kidney function declines with age, but a substantial portion of this decline has been attributed to the higher prevalence of risk factors for kidney disease at older ages. The effect of age on kidney function has not been well described in a healthy population across a wide age spectrum. METHODS: The authors pooled individual-level cross-sectional data from 18 253 persons aged 28-100 years in four studies: the Cardiovascular Health Study; the Health, Aging and Body Composition Study; the Multi-Ethnic Study of Atherosclerosis and the Prevention of Renal and Vascular End-Stage Disease cohort. Kidney function was measured by cystatin C. Clinical risk factors for kidney disease included diabetes, hypertension, obesity, smoking, coronary heart disease, cerebrovascular disease, peripheral arterial disease and heart failure. RESULTS: Across the age range, there was a strong, non-linear association of age with cystatin C concentration. This association was substantial, even among participants free of clinical risk factors for kidney disease; mean cystatin C levels were 46% higher in participants 80 and older compared with those <40 years (1.06 versus 0.72 mg/L, P < 0.001). Participants with one or more risk factors had higher cystatin C concentrations for a given age, and the age association was slightly stronger (P < 0.001 for age and risk factor interaction). CONCLUSIONS: There is a strong, non-linear association of age with kidney function, even in healthy individuals. An important area for research will be to investigate the mechanisms that lead to deterioration of kidney function in apparently healthy persons.

A cross-validation deletion-substitution-addition model selection algorithm: Application to marginal structural models.

The cross-validation deletion-substitution-addition (cvDSA) algorithm is based on data-adaptive estimation methodology to select and estimate marginal structural models (MSMs) for point treatment studies as well as models for conditional means where the outcome is continuous or binary. The algorithm builds and selects models based on user-defined criteria for model selection, and utilizes a loss function-based estimation procedure to distinguish between different model fits. In addition, the algorithm selects models based on cross-validation methodology to avoid "over-fitting" data. The cvDSA routine is an R software package available for download. An alternative R-package (DSA) based on the same principles as the cvDSA routine (i.e., cross-validation, loss function), but one that is faster and with additional refinements for selection and estimation of conditional means, is also available for download. Analyses of real and simulated data were conducted to demonstrate the use of these algorithms, and to compare MSMs where the causal effects were assumed (i.e., investigator-defined), with MSMs selected by the cvDSA. The package was used also to select models for the nuisance parameter (treatment) model to estimate the MSM parameters with inverse-probability of treatment weight (IPTW) estimation. Other estimation procedures (i.e., G-computation and double robust IPTW) are available also with the package.