Prospective study of e-cigarette use and respiratory symptoms in adolescents and young adults.

RATIONALE: Electronic cigarette (e-cigarette) aerosol contains volatile aldehydes, including flavourings and oxidant metals with known pulmonary toxicity. OBJECTIVES: To evaluate the associations of e-cigarette use with symptoms of wheeze, bronchitic symptoms and shortness of breath (SOB) across 4 years of prospective data. METHODS: Participants completed questionnaires on respiratory symptoms and past 30-day e-cigarette, cigarette and cannabis use in 2014 (wave 1; N=2094; mean age 17.3 years, SD=0.6 years). Follow-up information was collected in 2015 (wave 2; n=1609), 2017 (wave 3; n=1502) and 2018 (wave 4; n=1637) using online surveys. Mixed-effects logistic regression models evaluated associations of e-cigarette use with respiratory symptoms. MEASUREMENTS AND MAIN RESULTS: Participants were mostly Hispanic white (51.8%) and evenly representative by sex (49.6% female; 50.4% male). Compared with never e-cigarette users, past 30-day e-cigarette users reported increased odds of wheeze (OR 1.81; 95% CI 1.28, 2.56), bronchitic symptoms (OR 2.06; 95% CI 1.58, 2.69) and SOB (OR 1.78; 95% CI 1.23, 2.57), adjusting for study wave, age, sex, race, lifetime asthma diagnosis and parental education. Effect estimates were attenuated (wheeze (OR 1.41; 95% CI 0.99, 2.01), bronchitic symptoms (OR 1.55; 95% CI 1.18, 2.05), SOB (OR 1.48; 95% CI 1.01, 2.18)), after adjusting additionally for current cigarette use, cannabis use and secondhand exposure to e-cigarettes/cigarettes/cannabis. CONCLUSIONS: E-cigarette use in young adults was associated with respiratory symptoms, independent of combustible cannabis and cigarette exposures.

Treatment patterns for patients initiating novel acute migraine specific medications (nAMSMs) in the context of monoclonal antibodies (mAbs) targeting the calcitonin gene-related peptide (CGRP) pathway.

BACKGROUND: New acute and preventive migraine medications are available, but data on current treatment patterns are limited. This study describes migraine treatment patterns among patients initiating novel acute migraine specific medications (nAMSMs), overall and by prior use of anti-calcitonin gene-related peptide (CGRP) pathway monoclonal antibodies (mAbs). METHODS: In this retrospective cohort study using IQVIA open-source pharmacy and medical claims data, we identified patients with ≥ 1 claim for a nAMSM (ubrogepant, rimegepant, lasmiditan) between 01/01/2020 and 09/30/2020 (index period). Patients were indexed on their first nAMSM claim and stratified into 2 cohorts: patients with prior mAb use (≥ 1 claim for erenumab, fremanezumab, galcanezumab in the 6-month pre-index period) or patients without prior mAb use. Treatment patterns were assessed during the 6-month post-index period. RESULTS: Overall, 78,574 patients were identified (63% indexed on ubrogepant, 34% on rimegepant, and 3% on lasmiditan) with 26,656 patients (34%) having had prior mAb use. In the pre-index period, 79% of patients used non-mAb preventive medications and 75% of patients used acute medications. Following the index nAMSM claim, 65% of patients had ≥ 1 refill and 21% had ≥ 4 refills of their index nAMSM; 10% of patients switched to another nAMSM. Post-index mAb use was observed in 82% of patients with a prior mAb and 15% of patients without. Among patients with pre- and post-index use of acute medications, 38% discontinued ≥ 1 acute medication class in the post-index period. Among patients with concomitant use of traditional preventive medications at index, 30% discontinued ≥ 1 concomitant preventive anti-migraine medication in the post-index period. CONCLUSIONS: Most patients initiating nAMSMs had prior treatment with acute and preventive medications. Approximately one-third of patients had prior treatment with anti-CGRP pathway mAbs. After starting nAMSMs, more than one-third of patients discontinued at least one traditional acute medication and one-third of patients discontinued at least one traditional preventive medication. Despite nAMSM initiation, most patients with prior anti-CGRP pathway mAb use continued mAb use. Around 15% of patients without a prior mAb newly started a mAb. These results provide insight into how nAMSMs and mAbs have been integrated into clinical management of migraine in the real-world.

Inpatient Constipation Among Migraine Patients Prescribed Anti-calcitonin Gene-Related Peptide Monoclonal Antibodies and Standard of Care Antiepileptic Drugs: A Retrospective Cohort Study in a United States Electronic Health Record Database.

INTRODUCTION: Erenumab, an anti-calcitonin gene-related peptide (CGRP) receptor monoclonal antibody (mAb), was approved by the US Food and Drug Administration in May 2018. Constipation with serious complications was added to the Warning and Precautions section in the erenumab Prescribing Information in October 2019 after events were observed during post-marketing surveillance. We aimed to assess and compare the risk of inpatient constipation, and, separately, inpatient constipation with serious complications, among patients with migraine treated with CGRP mAbs and standard of care antiepileptic drugs (AEDs). METHODS: Within Optum's Electronic Health Record Research Database, patients with migraine who initiated erenumab, other CGRP mAbs, and AEDs were identified from May 2018 through March 2020. Erenumab initiators were propensity score-matched separately to initiators of other CGRP mAbs and AEDs. Incident inpatient constipation events, and serious complications, were identified using multiple risk windows for outcome assessment (30-, 60-, 90-day risk windows, and all available follow-up). Odds ratios (ORs) were calculated comparing inpatient constipation risk among matched erenumab initiators relative to comparators. RESULTS: We identified 17,902 erenumab, 13,404 other CGRP mAb, and 49,497 AED initiators who met study criteria. Among matched initiators, the risk of inpatient constipation was 0.46% (95% confidence interval (CI) 0.35-0.60) for erenumab and 0.44% (95% CI 0.33-0.58) for other CGRP mAbs within the 90-day risk window, with a corresponding OR of 1.06 (95% CI 0.72-1.55). Among matched erenumab and AED initiators, inpatient constipation risk was 0.53% (95% CI 0.42-0.66) and 0.76% (95% CI 0.62-0.92), respectively, and the OR was 0.69 (95% CI 0.51-0.94). Few serious complications were observed. CONCLUSION: Patients initiating erenumab had similar risk of inpatient constipation within 90 days of treatment initiation versus patients initiating other CGRP mAbs, and lower risk versus patients initiating AEDs. These findings provide context to events observed during post-marketing surveillance.

Risk for ischemic stroke and coronary heart disease associated with migraine and migraine medication among older adults.

BACKGROUND: Migraine has been associated with cardiovascular disease (CVD) events among middle-aged adults. The objective of this study was to determine the risk for ischemic stroke and coronary heart disease (CHD) events among older adults with versus without migraine. METHODS: This retrospective cohort study was conducted using data from US adults ≥66 years of age with Medicare health insurance between 2008 and 2017. After stratification by history of CVD, patients with a history of migraine were matched 1:4 to those without a history of migraine, based on calendar year, age, and sex. Patients were followed through December 31, 2017 for ischemic stroke and CHD events including myocardial infarction or coronary revascularization. All analyses were done separately for patients with and without a history of CVD. RESULTS: Among patients without a history of CVD (n = 109,950 including n = 21,990 with migraine and n = 87,960 without migraine), 1789 had an ischemic stroke and 3552 had a CHD event. The adjusted hazard ratio (HR) among patients with versus without migraine was 1.20 (95% confidence interval [95%CI], 1.07-1.35) for ischemic stroke and 1.02 (95%CI, 0.93-1.11) for CHD events. Compared to patients without migraine, those with migraine who were taking an opioid medication had a higher risk for ischemic stroke (adjusted HR 1.43 [95%CI, 1.20-1.69]), while those taking a triptan had a lower risk for CHD events (adjusted HR 0.79 [95%CI, 0.67-0.93]). Among patients with a history of CVD (n = 79,515 including n = 15,903 with migraine and n = 63,612 without migraine), 2960 had an ischemic stroke and 7981 had a CHD event. The adjusted HRs (95%CI) for ischemic stroke and CHD events associated with migraine were 1.27 (1.17-1.39) and 0.99 (0.93-1.05), respectively. Patients with migraine taking an opioid medication had a higher risk for ischemic stroke (adjusted HR 1.21 [95%CI, 1.07-1.36]), while those taking a triptan had a lower risk for CHD events (adjusted HR 0.83 [95%CI, 0.72-0.95]), each versus those without migraine. CONCLUSIONS: Older adults with migraine are at increased risk for ischemic stroke. The risk for ischemic stroke among older adults with migraine may differ by migraine medication classes.

Compliance in Controlled E-cigarette Studies.

INTRODUCTION: E-cigarette studies have found that the use of a variety of flavors and customizable devices results in greater use frequency and user satisfaction. However, standardized research e-cigarettes are being developed as closed systems with limited flavor options, potentially limiting user satisfaction. In this study, we explore protocol compliance in an e-cigarette study using a standardized, assigned device with puff time and duration tracking (controlled e-cigarette) and potential limitations that controlled devices and e-liquids can introduce. METHODS: In a crossover study, 49 young adult e-cigarette users were recruited using convenience sampling and assigned a controlled e-cigarette device and flavored or unflavored e-liquids on standardized protocols. E-cigarette use frequency (number of puffs per day, collected from the device) and serum cotinine levels were obtained at each of three study visits over 3 weeks. The correlation of cotinine and e-cigarette use over the preceding week was calculated at each study visit. RESULTS: Correlation of nicotine intake, as measured by serum cotinine, and puff time, as measured by puffs count and duration from the e-cigarette device, as an indicator of study protocol compliance, substantially declined after the first week of the study and were no longer correlated in the remaining study weeks (R2 = 0.53 and p ≤ .01 in week 1, R2 < 0.5 and p > .05 for remaining weeks). CONCLUSIONS: There is an emerging need for controlled e-cigarette exposures studies, but low compliance in the use of assigned devices and e-liquids may be a limitation that needs to be mitigated in future studies. IMPLICATIONS: This study is the first to analyze compliance with instructions to use a standardized e-cigarette device with puff time and duration tracking (controlled e-cigarette) across all subjects and an assigned e-liquid flavor over a 3-week period. We find that protocol compliance, as measured by correlations between e-cigarette use measures and cotinine levels, was only achieved in the first week of the study and declined thereafter. These findings indicate that the assignment of a study device and instruction to only use the study device with assigned e-liquid flavor may not be sufficient to ensure participant compliance with the study protocol. We suggest that additional measures, including behavioral and biological markers, are needed to ensure sole use of the study e-cigarette and e-liquid and to be able to interpret results from controlled e-cigarette studies.

Contribution of tailpipe and non-tailpipe traffic sources to quasi-ultrafine, fine and coarse particulate matter in southern California.

Exposure to traffic-related air pollution (TRAP) in the near-roadway environment is associated with multiple adverse health effects. To characterize the relative contribution of tailpipe and non-tailpipe TRAP sources to particulate matter (PM) in the quasi-ultrafine (PM0.2), fine (PM2.5) and coarse (PM2.5-10) size fractions and identify their spatial determinants in southern California (CA). Month-long integrated PM0.2, PM2.5 and PM2.5-10 samples (n = 461, 265 and 298, respectively) were collected across cool and warm seasons in 8 southern CA communities (2008-9). Concentrations of PM mass, elements, carbons and major ions were obtained. Enrichment ratios (ER) in PM0.2 and PM10 relative to PM2.5 were calculated for each element. The Positive Matrix Factorization model was used to resolve and estimate the relative contribution of TRAP sources to PM in three size fractions. Generalized additive models (GAMs) with bivariate loess smooths were used to understand the geographic variation of TRAP sources and identify their spatial determinants. EC, OC, and B had the highest median ER in PM0.2 relative to PM2.5. Six, seven and five sources (with characteristic species) were resolved in PM0.2, PM2.5 and PM2.5-10, respectively. Combined tailpipe and non-tailpipe traffic sources contributed 66%, 32% and 18% of PM0.2, PM2.5 and PM2.5-10 mass, respectively. Tailpipe traffic emissions (EC, OC, B) were the largest contributor to PM0.2 mass (58%). Distinct gasoline and diesel tailpipe traffic sources were resolved in PM2.5. Others included fuel oil, biomass burning, secondary inorganic aerosol, sea salt, and crustal/soil. CALINE4 dispersion model nitrogen oxides, trucks and intersections were most correlated with TRAP sources. The influence of smaller roadways and intersections became more apparent once Long Beach was excluded. Non-tailpipe emissions constituted ~8%, 11% and 18% of PM0.2, PM2.5 and PM2.5-10, respectively, with important exposure and health implications. Future efforts should consider non-linear relationships amongst predictors when modeling exposures. Implications: Vehicle emissions result in a complex mix of air pollutants with both tailpipe and non-tailpipe components. As mobile source regulations lead to decreased tailpipe emissions, the relative contribution of non-tailpipe traffic emissions to near-roadway exposures is increasing. This study documents the presence of non-tailpipe abrasive vehicular emissions (AVE) from brake and tire wear, catalyst degradation and resuspended road dust in the quasi-ultrafine (PM0.2), fine and coarse particulate matter size fractions, with contributions reaching up to 30% in PM0.2 in some southern California communities. These findings have important exposure and policy implications given the high metal content of AVE and the efficiency of PM0.2 at reaching the alveolar region of the lungs and other organ systems once inhaled. This work also highlights important considerations for building models that can accurately predict tailpipe and non-tailpipe exposures for population health studies.

E-cigarette Product Characteristics and Subsequent Frequency of Cigarette Smoking.

BACKGROUND: There is a dearth of evidence regarding the association of use of electronic cigarettes (e-cigarettes) with certain product characteristics and adolescent and young adult risk of unhealthy tobacco use patterns (eg, frequency of combustible cigarette smoking), which is needed to inform the regulation of e-cigarettes. METHODS: Data were collected via an online survey of participants in the Southern California Children's Health Study from 2015 to 2016 (baseline) and 2016 to 2017 (follow-up) (N = 1312). We evaluated the association of binary categories of 3 nonmutually exclusive characteristics of the e-cigarette used most frequently with the number of cigarettes smoked in the past 30 days at 1-year follow-up. Product characteristics included device (vape pen and/or modifiable electronic cigarette [mod]), use of nicotine in electronic liquid (e-liquid; yes or no), and use for dripping (directly dripping e-liquid onto the device; yes or no). RESULTS: Relative to never e-cigarette users, past-30-day e-cigarette use was associated with greater frequency of past-30-day cigarette smoking at follow-up. Among baseline past-30-day e-cigarette users, participants who used mods (versus vape pens) smoked >6 times as many cigarettes at follow-up (mean: 20.8 vs 1.3 cigarettes; rate ratio = 6.33; 95% confidence interval: 1.64-24.5) after adjustment for sociodemographic characteristics, baseline frequency of cigarette smoking, and number of days of e-cigarette use. After adjustment for device, neither nicotine e-liquid nor dripping were associated with frequency of cigarette smoking. CONCLUSIONS: Baseline mod users (versus vape pen users) smoked more cigarettes in the past 30 days at follow-up. Regulation of e-cigarette device type warrants consideration as a strategy to reduce cigarette smoking among adolescents and young adults who vape.

Risk factors associated with subsequent initiation of cigarettes and e-cigarettes in adolescence: A structural equation modeling approach.

BACKGROUND: Previous youth tobacco research has identified multiple correlated risk factors for initiation of cigarette and e-cigarette use; whether these factors are independently associated with initiation is not known, due to challenges with disentangling the independent effects of these correlated risk factors. METHODS: Students in 11th/12th grade enrolled in the Southern California Children's Health Study were surveyed in 2014 (baseline) and again in 2015 (N = 1553). Structural equation models (SEM) were developed to investigate associations of susceptibility, marketing, and the social environment (as latent factors), and other tobacco use at baseline with cigarette or e-cigarette initiation between baseline and follow-up. Analyses were restricted to baseline never cigarette users (N = 1293) for models evaluating cigarette initiation, and to never e-cigarette users (N = 1197) for models evaluating e-cigarette initiation. RESULTS: In fully-adjusted prospective SEM models, latent factors for cigarette susceptibility, marketing, and the social environment, along with ever e-cigarette use and ever hookah use at baseline were independently associated with cigarette initiation between baseline and follow-up (P < 0.05). Similarly, latent factors for e-cigarette susceptibility, marketing, and the social environment, along with ever hookah use at baseline were associated with e-cigarette initiation between baseline and follow-up (P < 0.05); however, cigarette use at baseline was not associated with e-cigarette initiation in SEM models (P = 0.16). CONCLUSIONS: We identified independent effects of multiple risk factors in SEM models on initiation of cigarettes and e-cigarettes. E-cigarette use was associated with cigarette initiation, but cigarette use was not associated with e-cigarette initiation in fully adjusted models. Research to identify underlying causal mechanisms is warranted.

The Dynamic Relationship Between Asthma and Obesity in Schoolchildren.

Asthma and obesity are among the most prevalent chronic health conditions in children. Although there has been compelling evidence of co-occurrence of asthma and obesity, it is uncertain whether asthma contributes to the development of obesity or obesity contributes to the onset of asthma or both. In this study, we used a joint transition modeling approach with cross-lagged structure to understand how asthma and obesity influence each other dynamically over time. Subjects for this study included 5,193 kindergarten and first-grade students enrolled from 13 communities in 2002-2003 in the Southern California Children's Health Study, with up to 10 years of follow-up. We found that nonobese children with diagnosed asthma at a study visit were at 37% higher odds of becoming obese by the next annual visit compared with children without asthma (odds ratio = 1.38; 95% credible interval: 1.12, 1.71). However, the presence of obesity at the current visit was not statistically significantly associated with asthma onset in the next visit (odds ratio = 1.25; 95% credible interval: 0.94, 1.62). In conclusion, childhood asthma appears to drive an increase in the onset of obesity among schoolchildren, while the onset of obesity does not necessarily imply the future onset of asthma, at least in the short term.

The Potential Effects of Policy-driven Air Pollution Interventions on Childhood Lung Development.

Rationale: Although elevated air pollution exposure impairs lung-function development in childhood, it remains a challenge to use this information to estimate the potential public health benefits of air pollution interventions in exposed populations.Objectives: Apply G-computation to estimate hypothetical effects of several realistic scenarios for future air pollution reductions on lung growth.Methods: Mixed-effects linear regression was used to estimate FEV1 and FVC from age 11 to 15 years in 2,120 adolescents across 3 cohorts (1993-2001, 1997-2004, and 2007-2011). Models included regional pollutants (nitrogen dioxide [NO2] or particulate matter with an aerodynamic diameter ≤2.5 µm [PM2.5]) and other important covariates. Using G-computation, a causal inference-based method, we then estimated changes in mean lung growth in our population for hypothetical interventions on either NO2 or PM2.5. Confidence intervals (CIs) were computed by bootstrapping (N = 1,000).Measurements and Main Results: Compared with the effects of exposure from observed NO2 concentrations during the study period, had communities remained at 1994 to 1997 NO2 levels, FEV1 and FVC growth were estimated to have been reduced by 2.7% (95% CI, -3.6 to -1.8) and 4.2% (95% CI, -5.2 to -3.4), respectively. If NO2 concentrations had been reduced by 30%, we estimated a 4.4% increase in FEV1 growth (95% CI, 2.8-5.9) and a 7.1% increase in FVC growth (95% CI, 5.7-8.6). Comparable results were observed for PM2.5 interventions.Conclusions: We estimated that substantial increases in lung function would occur as a result of interventions that reduce NO2 or PM2.5 concentrations. These findings provide a quantification of potential health benefits of air quality improvement.

Past 30-day co-use of tobacco and marijuana products among adolescents and young adults in California.

BACKGROUND: Limited research among adolescents and young adults (AYA) has assessed tobacco and marijuana co-use in light of specific products. We examined the patterns of past 30-day co-use of tobacco and marijuana products, and the product-specific associations among past 30-day use of these substances. METHODS: Data from three school-based convenience samples of California AYA (aged 15-22) (Sample 1 = 3008; Sample 2 = 1419; Sample 3 = 466) were collected during 2016-2017. Proportions of past 30-day co-use of tobacco (e-cigarettes, cigarettes, hookah, cigars) and marijuana (combustible, vaporized, edible, blunt) were estimated. Multivariable logistic regression analyses examined associations between use of each tobacco and marijuana product for individual samples, then the pooled analysis calculated combined ORs. RESULTS: In the three samples, 7.3-11.3% of participants reported past 30-day co-use. Combinations of e-cigarettes or cigarettes and combustible marijuana were the most common co-use patterns. Past 30-day use of e-cigarettes or cigarettes (vs. non-use) increased the odds of past 30-day use of all marijuana products [e-cigarettes: ORs (95%CI) ranging from 2.5 (1.7, 3.2) for edible marijuana to 4.0 (2.8, 5.2) for combustible marijuana; cigarettes: from 3.2 (2.1, 4.2) for vaporized marijuana to 5.5 (3.8, 7.3) for combustible marijuana]. Past 30-day use of hookah or cigars was positively associated with past 30-day use of three of four marijuana products, except for hookah and vaporized marijuana, and for cigars and combustible marijuana. CONCLUSIONS: Given various co-use patterns and significant associations among tobacco and marijuana products, interventions targeting AYA should address co-use across the full spectrum of specific products for both substances.

Effects of policy-driven hypothetical air pollutant interventions on childhood asthma incidence in southern California.

Childhood asthma is a major public health concern and has significant adverse impacts on the lives of the children and their families, and on society. There is an emerging link between air pollution, which is ubiquitous in our environment, particularly in urban centers, and incident childhood asthma. Here, using data from 3 successive cohorts recruited from the same 9 communities in southern California over a span of 20 y (1993 to 2014), we estimated asthma incidence using G-computation under hypothetical air pollution exposure scenarios targeting nitrogen dioxide (NO2) and particulate matter <2.5 µm (PM2.5) in separate interventions. We reported comparisons of asthma incidence under each hypothetical air pollution intervention with incidence under the observed natural course of exposure; results that may be more tangible for policymakers compared with risk ratios. Model results indicated that childhood asthma incidence rates would have been statistically significantly higher had the observed reduction in ambient NO2 in southern California not occurred in the 1990s and early 2000s, and asthma incidence rates would have been significantly lower had NO2 been lower than what it was observed to be. For example, compliance with a hypothetical standard of 20 ppb NO2 was estimated to result in 20% lower childhood asthma incidence (95% CI, -27% to -11%) compared with the exposure that actually occurred. The findings for hypothetical PM2.5 interventions, although statistically significant, were smaller in magnitude compared with results for the hypothetical NO2 interventions. Our results suggest a large potential public health benefit of air pollutant reduction in reduced incidence of childhood asthma.

Cluster-based bagging of constrained mixed-effects models for high spatiotemporal resolution nitrogen oxides prediction over large regions.

BACKGROUND: Accurate estimation of nitrogen dioxide (NO2) and nitrogen oxide (NOx) concentrations at high spatiotemporal resolutions is crucial for improving evaluation of their health effects, particularly with respect to short-term exposures and acute health outcomes. For estimation over large regions like California, high spatial density field campaign measurements can be combined with more sparse routine monitoring network measurements to capture spatiotemporal variability of NO2 and NOx concentrations. However, monitors in spatially dense field sampling are often highly clustered and their uneven distribution creates a challenge for such combined use. Furthermore, heterogeneities due to seasonal patterns of meteorology and source mixtures between sub-regions (e.g. southern vs. northern California) need to be addressed. OBJECTIVES: In this study, we aim to develop highly accurate and adaptive machine learning models to predict high-resolution NO2 and NOx concentrations over large geographic regions using measurements from different sources that contain samples with heterogeneous spatiotemporal distributions and clustering patterns. METHODS: We used a comprehensive Kruskal-K-means method to cluster the measurement samples from multiple heterogeneous sources. Spatiotemporal cluster-based bootstrap aggregating (bagging) of the base mixed-effects models was then applied, leveraging the clusters to obtain balanced and less correlated training samples for less bias and improvement in generalization. Further, we used the machine learning technique of grid search to find the optimal interaction of temporal basis functions and the scale of spatial effects, which, together with spatiotemporal covariates, adequately captured spatiotemporal variability in NO2 and NOx at the state and local levels. RESULTS: We found an optimal combination of four temporal basis functions and 200 m scale spatial effects for the base mixed-effects models. With the cluster-based bagging of the base models, we obtained robust predictions with an ensemble cross validation R2 of 0.88 for both NO2 and NOx [RMSE (RMSEIQR): 3.62 ppb (0.28) and 9.63 ppb (0.37) respectively]. In independent tests of random sampling, our models achieved similarly strong performance (R2 of 0.87-0.90; RMSE of 3.97-9.69 ppb; RMSEIQR of 0.21-0.27), illustrating minimal over-fitting. CONCLUSIONS: Our approach has important implications for fusing data from highly clustered and heterogeneous measurement samples from multiple data sources to produce highly accurate concentration estimates of air pollutants such as NO2 and NOx at high resolution over a large region.

Association of Changes in Air Quality With Incident Asthma in Children in California, 1993-2014.

Importance: Exposure to air pollutants is a well-established cause of asthma exacerbation in children; whether air pollutants play a role in the development of childhood asthma, however, remains uncertain. Objective: To examine whether decreasing regional air pollutants were associated with reduced incidence of childhood asthma. Design, Setting, and Participants: A multilevel longitudinal cohort drawn from 3 waves of the Southern California Children's Health Study over a period of air pollution decline. Each cohort was followed up from 4th to 12th grade (8 years): 1993-2001, 1996-2004, and 2006-2014. Final follow-up for these data was June 2014. Population-based recruitment was from public elementary schools. A total of 4140 children with no history of asthma and residing in 1 of 9 Children's Health Study communities at baseline were included. Exposures: Annual mean community-level ozone, nitrogen dioxide, and particulate matter less than 10 µm (PM10) and less than 2.5 µm (PM2.5) in the baseline year for each of 3 cohorts. Main Outcomes and Measures: Prospectively identified incident asthma, collected via questionnaires during follow-up. Results: Among the 4140 children included in this study (mean [SD] age at baseline, 9.5 [0.6] years; 52.6% female [n = 2 179]; 58.6% white [n = 2273]; and 42.2% Hispanic [n = 1686]), 525 incident asthma cases were identified. For nitrogen dioxide, the incidence rate ratio (IRR) for asthma was 0.80 (95% CI, 0.71-0.90) for a median reduction of 4.3 parts per billion, with an absolute incidence rate decrease of 0.83 cases per 100 person-years. For PM2.5, the IRR was 0.81 (95% CI, 0.67-0.98) for a median reduction of 8.1 µg/m3, with an absolute incidence rate decrease of 1.53 cases per 100 person-years. For ozone, the IRR for asthma was 0.85 (95% CI, 0.71-1.02) for a median reduction of 8.9 parts per billion, with an absolute incidence rate decrease of 0.78 cases per 100 person-years. For PM10, the IRR was 0.93 (95% CI, 0.82-1.07) for a median reduction of 4.0 µg/m3, with an absolute incidence rate decrease of 0.46 cases per 100 person-years. Conclusions and Relevance: Among children in Southern California, decreases in ambient nitrogen dioxide and PM2.5 between 1993 and 2014 were significantly associated with lower asthma incidence. There were no statistically significant associations for ozone or PM10.

Evaluating the predictive value of measures of susceptibility to tobacco and alternative tobacco products.

BACKGROUND: The "cigarette susceptibility index" has been adapted for other products, yet, the validity of these adapted measures-particularly among youth who have used other tobacco products-has not been evaluated. METHODS: We used prospective data from the Southern California Children's Health Study to evaluate the association of questionnaire measures assessing susceptibility to e-cigarette, cigarette, hookah and cigar/cigarillo/little cigar use at wave 1 (W1; 11th/12th grade) with subsequent initiation between W1 and W2 (16 months later, N = 1453). We additionally examined whether each effect estimate differed by use of other tobacco products at W1. RESULTS: Odds ratios, attributable risk%, and risk differences for product initiation among susceptible vs. non-susceptible youth were consistently higher among never users of any tobacco product than among youth with any tobacco use history. For example, susceptible (vs. non-susceptible) youth with no prior tobacco use had 3.64 times the odds of subsequent initiation of e-cigarettes (95%CI:2.61,5.09), while among users of another product, susceptible (vs. non-susceptible) youth had 1.95 times the odds of e-cigarette initiation (95%CI:0.98,3.89; p-interaction = 0.016). 60.4% of e-cigarette initiation among never users of any product could be attributed to susceptibility, compared to 19.8% among users of another product. The e-cigarette absolute risk difference between susceptible and non-susceptible youth was 21.9%(15.2,28.6) for never users, vs. 15.4%(0.2,30.7) for users of another product. CONCLUSION: Tobacco product-specific susceptibility associations with initiation of use at W2 were markedly attenuated among prior users of other products, demonstrating reduced utility for these measures among subjects using other products.

Understanding the importance of key risk factors in predicting chronic bronchitic symptoms using a machine learning approach.

BACKGROUND: Chronic respiratory symptoms involving bronchitis, cough and phlegm in children are underappreciated but pose a significant public health burden. Efforts for prevention and management could be supported by an understanding of the relative importance of determinants, including environmental exposures. Thus, we aim to develop a prediction model for bronchitic symptoms. METHODS: Schoolchildren from the population-based southern California Children's Health Study were visited annually from 2003 to 2012. Bronchitic symptoms over the prior 12 months were assessed by questionnaire. A gradient boosting model was fit using groups of risk factors (including traffic/air pollution exposures) for all children and by asthma status. Training data consisted of one observation per participant in a random study year (for 50% of participants). Validation data consisted of: (1) a random (later) year in the same participants (within-participant); (2) a random year in participants excluded from the training data (across-participant). RESULTS: At baseline, 13.2% of children had asthma and 18.1% reported bronchitic symptoms. Models performed similarly within- and across-participant. Previous year symptoms/medication use provided much of the predictive ability (across-participant area under the receiver operating characteristic curve (AUC): 0.76 vs 0.78 for all risk factors, in all participants). Traffic/air pollution exposures added modestly to prediction as did body mass index percentile, age and parent stress. CONCLUSIONS: Regardless of asthma status, previous symptoms were the most important predictors of current symptoms. Traffic/air pollution variables contribute modest predictive information, but impact large populations. Methods proposed here could be generalized to personalized exacerbation predictions in future longitudinal studies to support targeted prevention efforts.

Applying Multivariate Segmentation Methods to Human Activity Recognition From Wearable Sensors' Data.

BACKGROUND: Time-resolved quantification of physical activity can contribute to both personalized medicine and epidemiological research studies, for example, managing and identifying triggers of asthma exacerbations. A growing number of reportedly accurate machine learning algorithms for human activity recognition (HAR) have been developed using data from wearable devices (eg, smartwatch and smartphone). However, many HAR algorithms depend on fixed-size sampling windows that may poorly adapt to real-world conditions in which activity bouts are of unequal duration. A small sliding window can produce noisy predictions under stable conditions, whereas a large sliding window may miss brief bursts of intense activity. OBJECTIVE: We aimed to create an HAR framework adapted to variable duration activity bouts by (1) detecting the change points of activity bouts in a multivariate time series and (2) predicting activity for each homogeneous window defined by these change points. METHODS: We applied standard fixed-width sliding windows (4-6 different sizes) or greedy Gaussian segmentation (GGS) to identify break points in filtered triaxial accelerometer and gyroscope data. After standard feature engineering, we applied an Xgboost model to predict physical activity within each window and then converted windowed predictions to instantaneous predictions to facilitate comparison across segmentation methods. We applied these methods in 2 datasets: the human activity recognition using smartphones (HARuS) dataset where a total of 30 adults performed activities of approximately equal duration (approximately 20 seconds each) while wearing a waist-worn smartphone, and the Biomedical REAl-Time Health Evaluation for Pediatric Asthma (BREATHE) dataset where a total of 14 children performed 6 activities for approximately 10 min each while wearing a smartwatch. To mimic a real-world scenario, we generated artificial unequal activity bout durations in the BREATHE data by randomly subdividing each activity bout into 10 segments and randomly concatenating the 60 activity bouts. Each dataset was divided into ~90% training and ~10% holdout testing. RESULTS: In the HARuS data, GGS produced the least noisy predictions of 6 physical activities and had the second highest accuracy rate of 91.06% (the highest accuracy rate was 91.79% for the sliding window of size 0.8 second). In the BREATHE data, GGS again produced the least noisy predictions and had the highest accuracy rate of 79.4% of predictions for 6 physical activities. CONCLUSIONS: In a scenario with variable duration activity bouts, GGS multivariate segmentation produced smart-sized windows with more stable predictions and a higher accuracy rate than traditional fixed-size sliding window approaches. Overall, accuracy was good in both datasets but, as expected, it was slightly lower in the more real-world study using wrist-worn smartwatches in children (BREATHE) than in the more tightly controlled study using waist-worn smartphones in adults (HARuS). We implemented GGS in an offline setting, but it could be adapted for real-time prediction with streaming data.

Exposure measurement error in air pollution studies: A framework for assessing shared, multiplicative measurement error in ensemble learning estimates of nitrogen oxides.

BACKGROUND: Increasingly ensemble learning-based spatiotemporal models are being used to estimate residential air pollution exposures in epidemiological studies. While these machine learning models typically have improved performance, they suffer from exposure measurement error that is inherent in all models. Our objective is to develop a framework to formally assess shared, multiplicative measurement error (SMME) in our previously published three-stage, ensemble learning-based nitrogen oxides (NOx) model to identify its spatial and temporal patterns and predictors. METHODS: By treating the ensembles as an external dosimetry system, we quantified shared and unshared, multiplicative and additive (SUMA) measurement error components in our exposure model. We used generalized additive models (GAMs) with a smooth term for location to identify geographic locations with significantly elevated SMME and explain their spatial and temporal determinants. RESULTS: We found evidence of significant shared and unshared multiplicative error (p < 0.0001) in our ensemble-learning based spatiotemporal NOx model predictions. Unshared multiplicative error was 26 times larger than SMME. We observed significant geographic (p < 0.0001) and temporal variation in SMME with the majority (43%) of predictions with elevated SMME occurring in the earliest time-period (1992-2000). Densely populated urban prediction regions with complex air pollution sources generally exhibited highest odds of elevated SMME. CONCLUSIONS: We developed a novel statistical framework to formally evaluate the magnitude and drivers of SMME in ensemble learning-based exposure models. Our framework can be used to inform building future improved exposure models.

Tobacco Retail Licensing and Youth Product Use.

BACKGROUND: Restricting youth access to tobacco is a central feature of US tobacco regulatory policy, but impact of local tobacco retail licensing (TRL) regulation on cigarette smoking rates remains uncertain. Effects of TRL on other tobacco product use and use as adolescents reach the age to legally purchase tobacco products has not been investigated. METHODS: Prevalences of ever and past 30-day cigarette, electronic cigarette (e-cigarette), cigar, and hookah use were assessed in a survey of a cohort of 1553 11th- and 12th-grade adolescents (mean age: 17.3 years); rates of initiation were evaluated 1.5 years later. An American Lung Association (2014) youth access grade was assigned to each of 14 political jurisdictions in which participants lived on the basis of the strength of the local TRL ordinance. RESULTS: At baseline, participants living in 4 jurisdictions with "A" grades (ie, with most restrictive ordinances) had lower odds of ever cigarette use (odds ratio [OR] 0.61; 95% confidence interval [CI] 0.41-0.90) and of past 30-day use (OR 0.51; 95% CI 0.29-0.89) than participants in 10 D- to F-grade jurisdictions. At follow-up at legal age of purchase, lower odds of cigarette use initiation (OR 0.67; 95% CI 0.45-0.99) occurred in jurisdictions with stronger TRL policy. Lower odds of e-cigarette initiation at follow-up (OR 0.74; 95% CI 0.55-0.99) and of initiation with past 30-day use (OR 0.45; 95% CI 0.23-0.90) were also associated with better regulation. CONCLUSIONS: Strong local TRL ordinance may lower rates of cigarette and e-cigarette use among youth and young adults.

Tobacco Marketing and Subsequent Use of Cigarettes, E-Cigarettes, and Hookah in Adolescents.

INTRODUCTION: Tobacco marketing has expanded from cigarettes to other tobacco products through many promotional channels. Marketing exposure is associated with use of that tobacco product. However, it is unclear if marketing for one product leads to subsequent use of other tobacco products. METHODS: This prospective cohort study assessed self-reported marketing exposure for six tobacco products across five marketing channels in 11th and 12th grade students in 2014. Approximately 16 months later, a follow-up survey was conducted online (N = 1553) to assess initiation of cigarettes, electronic cigarettes (e-cigarettes), and hookah. RESULTS: Adolescent never-smokers with frequent exposure to cigarette marketing on the Internet and in stores are more than two times as likely to begin smoking as young adults (Internet OR = 2.98 [95% CI = 1.56 to 5.66]; stores OR = 2.83 [95% CI = 1.23 to 6.50]). Never users of e-cigarettes were significantly more likely to initiate use, if exposed to Internet, store, and outdoor e-cigarette marketing. Never users of hookah were more likely to use hookah after seeing it marketed in stores. Youth exposed to marketing of e-cigarettes, hookah, cigars, smokeless tobacco, and pipe tobacco in stores were two to three times more likely to begin smoking cigarettes even though the marketed products were not cigarettes. CONCLUSIONS: Adolescent exposure to marketing of tobacco products is associated with initiation of those products as young adults. Exposure to marketing for non-cigarette tobacco products is associated with subsequent cigarette smoking, even when the promoted products are not cigarettes. Future research and interventions should consider the influence of marketing from multiple tobacco products on adolescent tobacco use. IMPLICATIONS: Adolescents grow up in a rich media environment with exposure to tobacco marketing in both their homes (eg, through the Internet and television) and their communities (eg, stores and billboards). This prospective study provides evidence that adolescents exposed to tobacco marketing for multiple tobacco products are more likely to subsequently begin using those products and to begin smoking cigarettes even when the marketing they recall is for different tobacco products. Adolescent exposure to tobacco marketing can increase likelihood of cigarette smoking, e-cigarette, and hookah use with potential lifelong health effects.