Clarifying the causal contrast: An empirical example applying the prevalent new user study design.

PURPOSE: The prevalent new user design extends the active comparator new user design to include patients switching to a treatment of interest from a comparator. We examined the impact of adding "switchers" to incident new users on the estimated hazard ratio (HR) of hospitalized heart failure. METHODS: Using MarketScan claims data (2000-2014), we estimated HRs of hospitalized heart failure between patients initiating GLP-1 receptor agonists (GLP-1 RA) and sulfonylureas (SU). We considered three estimands: (1) the effect of incident new use; (2) the effect of switching; and (3) the effect of incident new use or switching, combining the two population. We used time-conditional propensity scores (TCPS) and time-stratified standardized morbidity ratio (SMR) weighting to adjust for confounding. RESULTS: We identified 76 179 GLP-1 RA new users, of which 12% were direct switchers (within 30 days) from SU. Among incident new users, GLP-1 RA was protective against heart failure (adjHRSMR = 0.74 [0.69, 0.80]). Among switchers, GLP-1 RA was not protective (adjHRSMR = 0.99 [0.83, 1.18]). Results in the combined population were largely driven by the incident new users, with GLP-1 RA having a protective effect (adjHRSMR = 0.77 [0.72, 0.83]). Results using TCPS were consistent with those estimated using SMR weighting. CONCLUSIONS: When analyses were conducted only among incident new users, GLP-1 RA had a protective effect. However, among switchers from SU to GLP-1 RA, the effect estimates substantially shifted toward the null. Combining patients with varying treatment histories can result in poor confounding control and camouflage important heterogeneity.

Validating claims-based definitions for deprescribing: Bridging the gap between clinical and administrative data.

BACKGROUND: Limited research has evaluated the validity of claims-based definitions for deprescribing. OBJECTIVES: Evaluate the validity of claims-based definitions of deprescribing against electronic health records (EHRs) for deprescribing of benzodiazepines (BZDs) after a fall-related hospitalization. METHODS: We used a novel data linkage between Medicare fee-for-service (FFS) and Part D with our health system's EHR. We identified patients aged ≥66 years with a fall-related hospitalization, continuous enrollment in Medicare FFS and Part D for 6 months pre- and post-hospitalization, and ≥2 BZD fills in the 6 months pre-hospitalization. Using a standardized EHR abstraction tool, we adjudicated deprescribing for a sub-sample with a fall-related hospitalization at UNC. We evaluated the validity of claims-based deprescribing definitions (e.g., gaps in supply, dosage reductions) versus chart review using sensitivity and specificity. RESULTS: Among 257 patients in the overall sample, 44% were aged 66-74 years, 35% had Medicare low-income subsidy, 79% were female. Among claims-based definitions using gaps in supply, the prevalence of BZD deprescribing ranged from 8.2% (no refills) to 36.6% (30-day gap). When incorporating dosage, the prevalence ranged from 55.3% to 65.8%. Among the validation sub-sample (n = 47), approximately one-third had BZDs deprescribed in the EHR. Compared to EHR, gaps in supply from claims had good sensitivity, but poor specificity. Incorporating dosage increased sensitivity, but worsened specificity. CONCLUSIONS: The sensitivity of claims-based definitions for deprescribing of BZDs was low; however, the specificity of a 90-day gap was >90%. Replication in other EHRs and for other low-value medications is needed to guide future deprescribing research.

Visualizing External Validity: Graphical Displays to Inform the Extension of Treatment Effects from Trials to Clinical Practice.

BACKGROUND: In the presence of effect measure modification, estimates of treatment effects from randomized controlled trials may not be valid in clinical practice settings. The development and application of quantitative approaches for extending treatment effects from trials to clinical practice settings is an active area of research. METHODS: In this article, we provide researchers with a practical roadmap and four visualizations to assist in variable selection for models to extend treatment effects observed in trials to clinical practice settings and to assess model specification and performance. We apply this roadmap and visualizations to an example extending the effects of adjuvant chemotherapy (5-fluorouracil vs. plus oxaliplatin) for colon cancer from a trial population to a population of individuals treated in community oncology practices in the United States. RESULTS: The first visualization screens for potential effect measure modifiers to include in models extending trial treatment effects to clinical practice populations. The second visualization displays a measure of covariate overlap between the clinical practice populations and the trial population. The third and fourth visualizations highlight considerations for model specification and influential observations. The conceptual roadmap describes how the output from the visualizations helps interrogate the assumptions required to extend treatment effects from trials to target populations. CONCLUSIONS: The roadmap and visualizations can inform practical decisions required for quantitatively extending treatment effects from trials to clinical practice settings.

The association between wealth and sleep medication use in a nationally-representative sample of older Medicare beneficiaries.

BACKGROUND: Sleep disorders are common among older adults, leading to high prevalence of over-the-counter and prescription sleep medication use. Socioeconomically disadvantaged individuals have higher prevalence of sleep disorders. Frequent use of sleep medications can increase the risk of falls. Little is known about the association between wealth and sleep medication use in older adults. METHODS: We conducted a cross-sectional analysis using a nationwide sample of 7603 Medicare beneficiaries (65+ years) from Round 1 (2011) of the National Health and Aging Trends Study. We measured self-reported wealth as the sum of assets (retirement savings, stocks/bonds, checking/savings accounts, business assets, and home value) minus liabilities (mortgage, credit card, and medical debt). Self-reported sleep medication use in the past month was categorized as frequent (5-7 nights/week), sometimes (1-4 nights/week), or never (0 night/week). We estimated differences in the prevalence of sleep medication use by quintiles of wealth using crude and adjusted binomial regression models. Individuals missing sleep medication information were excluded. RESULTS: Median wealth was $152 582 (IQR: $24 023-412 992). Sixteen percent reported frequent sleep medication use, 15% reported some use, and 70% reported no use. Frequent sleep medication use was more common in lower wealth quintiles (lowest: 20%, highest: 12%). Alternatively, some use was more common in higher wealth quintiles (lowest: 11%, highest: 18%). Results were similar after adjustment for demographic factors, anxiety, depression, and sleep disorders. CONCLUSIONS: In this study, less wealthy older adults had higher prevalence of frequent sleep medication use. This may lead to dependency or increased fall risk in this vulnerable population.

Translation of a Claims-Based Frailty Index From the International Classification of Diseases, Ninth Revision, Clinical Modification to the Tenth Revision.

The Faurot frailty index (FFI) is a validated algorithm that uses enrollment and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)-based billing information from Medicare claims data as a proxy for frailty. In October 2015, the US health-care system transitioned from the ICD-9-CM to the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Applying the Centers for Medicare and Medicaid Services General Equivalence Mappings, we translated diagnosis-based frailty indicator codes from the ICD-9-CM to the ICD-10-CM, followed by manual review. We used interrupted time-series analysis of Medicare data to assess the comparability of the pre- and posttransition FFI scores. In cohorts of beneficiaries enrolled in January 2015-2017 with 8-month frailty look-back periods, we estimated associations between the FFI and 1-year risk of aging-related outcomes (mortality, hospitalization, and admission to a skilled nursing facility). Updated indicators had similar prevalences as pretransition definitions. The median FFI scores and interquartile ranges (IQRs) for the predicted probability of frailty were similar before and after the International Classification of Diseases transition (pretransition: median, 0.034 (IQR, 0.02-0.07); posttransition: median, 0.038 (IQR, 0.02-0.09)). The updated FFI was associated with increased risks of mortality, hospitalization, and skilled nursing facility admission, similar to findings from the ICD-9-CM era. Studies of medical interventions in older adults using administrative claims should use validated indices, like the FFI, to mitigate confounding or assess effect-measure modification by frailty.

Missing data approaches in longitudinal studies of aging: A case example using the National Health and Aging Trends Study.

PURPOSE: Missing data is a key methodological consideration in longitudinal studies of aging. We described missing data challenges and potential methodological solutions using a case example describing five-year frailty state transitions in a cohort of older adults. METHODS: We used longitudinal data from the National Health and Aging Trends Study, a nationally-representative cohort of Medicare beneficiaries. We assessed the five components of the Fried frailty phenotype and classified frailty based on their number of components (robust: 0, prefrail: 1-2, frail: 3-5). One-, two-, and five-year frailty state transitions were defined as movements between frailty states or death. Missing frailty components were imputed using hot deck imputation. Inverse probability weights were used to account for potentially informative loss-to-follow-up. We conducted scenario analyses to test a range of assumptions related to missing data. RESULTS: Missing data were common for frailty components measured using physical assessments (walking speed, grip strength). At five years, 36% of individuals were lost-to-follow-up, differentially with respect to baseline frailty status. Assumptions for missing data mechanisms impacted inference regarding individuals improving or worsening in frailty. CONCLUSIONS: Missing data and loss-to-follow-up are common in longitudinal studies of aging. Robust epidemiologic methods can improve the rigor and interpretability of aging-related research.

Breaking Research Silos and Stimulating "Innovation at the Edges" in Epidemiology.

Epidemiological training often requires specialization in a subdiscipline (e.g., pharmacoepidemiology, genetic epidemiology, social epidemiology, or infectious disease epidemiology). While specialization is necessary and beneficial, it comes at the cost of decreased awareness of scientific developments in other subdisciplines of epidemiology. In this commentary, we argue for the importance of promoting an exchange of ideas across seemingly disparate epidemiologic subdisciplines. Such an exchange can lead to invaluable opportunities to learn from and merge knowledge across subdisciplines. It can promote "innovation at the edges," a process of borrowing and transforming methods from one subdiscipline in order to develop something new and advance another subdiscipline. Further, we outline specific actionable steps at the researcher, institution, and professional society level that can promote such innovation.

Looking fear in the face: Adults but not adolescents gaze at social threat during observational learning.

Observational threat learning is an indirect pathway to learn about what is safe versus dangerous. Visual gaze patterns may be an important measure to understand the underlying mechanisms of social threat learning. However, little research has considered this type of learning or attention allocation during observational learning, and even less has examined it across development. The study examined visual gaze patterns during observational threat acquisition amongst adolescents and adults. Ninety-three adolescents (13-17 years) and 78 adults (18-34 years) underwent a differential observational acquisition task by watching a video wherein a learning model was presented with colored bells. One bell (conditioned stimulus, CS+) was associated with an electric stimulation to the learning model's arm (social unconditioned stimulus, social US), while the other bell was not (CS-). Eye tracking was used during each trial. First, all participants, regardless of age, fixated longer on the face of the learning model when the CS+ was presented than when the CS- was presented. Second, adolescents averted their gaze from the learning model's face when the learning model received an electrical stimulation, whereas adults fixated more on the learning model's face when the stimulation was administered. Finally, adolescents understood the CS+-US contingency less than adults, stemming from the different gaze pattern for the age groups to the social US. Results replicate previous findings in adults and extend them to adolescents, emphasizing the importance of the learning model's facial expressions in conveying information on what is safe versus dangerous in the environment. Moreover, results showed developmental differences in gaze patterns during observational threat learning; this translated into poorer understanding of the CS+-US association amongst adolescents.

Observational extinction reduces fear and its retention among adolescents and adults.

Observational fear learning has been notably understudied in fear learning research, especially the potential benefits of observational extinction. Additionally, no studies have investigated observational extinction during adolescence, a key developmental stage for the emergence of anxiety disorders. This study compared observational and direct fear extinction following observational fear acquisition among adolescents and adults. Participants underwent differential observational fear acquisition. Next, participants underwent one of three fear extinctions: (1) observational extinction, (2) direct extinction, (3) no-extinction. Thereafter, participants underwent a US reminder and finally, a direct exposure test. Skin conductance response (SCR), self-reported fear, and risk assessment were measured. Differential observational fear acquisition was demonstrated by both age groups across all measures. During extinction, the observational and direct extinction groups did not differ in differential extinction, however, during the US reminder the observational group exhibited higher SCR compared to the direct group, but not the no-extinction group. Finally, during the direct exposure test, the no-extinction group showed greater differential risk assessment and higher SCR compared to the observational and direct groups. Additionally, adolescents showed poorer discrimination between the CSs compared to adults. Observational extinction successfully reduced fear and its retention amongst youth and adults pointing to the potential benefit of vicarious exposure therapy.

Alternative analytic and matching approaches for the prevalent new-user design: A simulation study.

PURPOSE: To describe the creation of prevalent new user (PNU) cohorts and compare the relative bias and computational efficiency of several alternative analytic and matching approaches in PNU studies. METHODS: In a simulated cohort, we estimated the effect of a treatment of interest vs a comparator among those who switched to the treatment of interest using the originally proposed time-conditional propensity score (TCPS) matching, standardized morbidity ratio weighting (SMRW), disease risk scores (DRS), and several alternative propensity score matching approaches. For each analytic method, we compared the average RR (across 2000 replicates) to the known risk ratio (RR) of 1.00. RESULTS: SMRW and DRS yielded unbiased results (RR = 0.998 and 0.997, respectively). TCPS matching with replacement was also unbiased (RR = 0.999). TCPS matching without replacement was unbiased when matches were identified starting with patients with the shortest treatment history as initially proposed (RR = 0.999), but it resulted in very slight bias (RR = 0.983) when starting with patients with the longest treatment history. Similarly, creating a match pool without replacement starting with patients with the shortest treatment history yielded an unbiased estimate (RR = 0.997), but matching with the longest treatment history first resulted in substantial bias (RR = 0.903). The most biased strategy was matching after selecting one random comparator observation per individual that continued on the comparator (RR = 0.802). CONCLUSIONS: Multiple analytic methods can estimate treatment effects without bias in a PNU cohort. Still, researchers should be wary of introducing bias when selecting controls for complex matching strategies beyond the initially proposed TCPS.

Unique associations between conditioned cognitive and physiological threat responses and facets of anxiety symptomatology in youth.

This study examined associations between anxiety symptomatology and cognitive and physiological threat responses during threat learning in a large sample of children and adolescents. Anxiety symptomatology severity along different dimensions (generalized anxiety, separation anxiety, social anxiety, and panic symptoms) was measured using parental and self-reports. Participants completed differential threat acquisition and extinction using an age-appropriate threat conditioning task. They then returned to the lab after 7-10 days to complete an extinction recall task that also assessed threat generalization. Results indicated that more severe overall anxiety was associated with greater cognitive and physiological threat responses during acquisition, extinction, and extinction recall. During acquisition and extinction, all anxiety dimensions manifested greater cognitive threat responses, while panic, separation anxiety, and social anxiety symptoms, but not generalized anxiety, were related to heightened physiological threat responses. In contrast, when we assessed generalization of cognitive threat responses, we found only generalized anxiety symptoms were associated with greater threat response generalization. The study provides preliminary evidence of specificity in threat responses during threat learning across youth with different anxiety symptoms.

Effects of anticholinergic and sedative medication use on fractures: A self-controlled design study.

BACKGROUND/OBJECTIVES: Unintentional falls are a leading cause of injury for older adults, and evidence is needed to understand modifiable risk factors. We evaluated 1-year fall-related fracture risk and whether dispensing of medications with anticholinergic/sedating properties is temporally associated with an increased odds of these fractures. DESIGN: A retrospective cohort study with nested self-controlled analyses conducted between January 1, 2014, and December 31, 2016. SETTING: Twenty percent nationwide, random sample of US Medicare beneficiaries. PARTICIPANTS: New users of medications with anticholinergic/sedating properties who were 66+ years old and had Medicare Parts A, B, and D coverage but no claims for medications with anticholinergic/sedating properties in the year before initiation were eligible. MEASUREMENTS: We followed new users of medications with anticholinergic/sedating properties until first non-vertebral, fall-related fracture (primary outcome), Medicare disenrollment, death, or end of study data. We estimated the 1-year risk with corresponding 95% confidence intervals (CIs) of first fracture after new use. We applied the self-controlled case-crossover and case-time-control designs to estimate odds ratios (ORs) and 95% CIs by comparing anticholinergic and/or sedating medication exposure (any vs. none) during a 14-day hazard period preceding the fracture to exposure to these medications during an earlier 14-day control period. RESULTS: A total of 1,097,989 Medicare beneficiaries initiated medications with anticholinergic/sedating properties in the study period. The 1-year cumulative incidence of fall-related fracture, accounting for death as a competing risk, was 5.0% (95% CI: 5.0%-5.0%). Using the case-crossover design (n = 41,889), the adjusted OR for the association between anticholinergic/sedating medications and fractures was 1.03 (95% CI: 0.99, 1.08). Accounting for the noted temporal trend using the case-time-control design (n = 209,395), the adjusted OR was 1.60 (95% CI: 1.52, 1.69). CONCLUSION: Use of anticholinergic/sedating medication was temporally associated with an increased odds of fall-related fractures. Patients and their healthcare providers should consider pharmacologic and non-pharmacologic treatments for the target condition that are safer.

Health Care Use Among Latinx Children After 2017 Executive Actions on Immigration.

BACKGROUND: US immigration policy changes may affect health care use among Latinx children. We hypothesized that January 2017 restrictive immigration executive actions would lead to decreased health care use among Latinx children. METHODS: We used controlled interrupted time series to estimate the effect of executive actions on outpatient cancellation or no-show rates from October 2016 to March 2017 ("immigration action period") among Latinx children in 4 health care systems in North Carolina. We included control groups of (1) non-Latinx children and (2) Latinx children from the same period in the previous year ("control period") to account for natural trends such as seasonality. RESULTS: In the immigration action period, 114 627 children contributed 314 092 appointments. In the control period, 107 657 children contributed 295 993 appointments. Relative to the control period, there was an immediate 5.7% (95% confidence interval [CI]: 0.40%-10.9%) decrease in cancellation rates among all Latinx children, but no sustained change in trend of cancellations and no change in no-show rates after executive immigration actions. Among uninsured Latinx children, there was an immediate 12.7% (95% CI: 2.3%-23.1%) decrease in cancellations; however, cancellations then increased by 2.4% (95% CI: 0.89%-3.9%) per week after immigration actions, an absolute increase of 15.5 cancellations per 100 appointments made. CONCLUSIONS: There was a sustained increase in cancellations among uninsured Latinx children after immigration actions, suggesting decreased health care use among uninsured Latinx children. Continued monitoring of effects of immigration policy on child health is needed, along with measures to ensure that all children receive necessary health care.

Quantifying cumulative anticholinergic and sedative drug load among US Medicare Beneficiaries.

PURPOSE: Medications with anticholinergic and sedative properties are widely used among older adults despite strong evidence of harm. The drug burden index (DBI), a pharmacological screening tool, measures these properties across drug classes, and higher DBI drug exposure (DBI > 1) has been associated with certain physical function-related adverse events. Our aim was to quantify mean daily DBI drug exposure among older adults in the United States (US). METHODS: We screened medications for DBI properties and operationalized the DBI for US Medicare claims. We then conducted a retrospective cohort study of a 20% random, nationwide sample of 4 137 384 fee-for-service Medicare beneficiaries aged 66+ years (134 757 039 person-months) from January 2013 to December 2016. We measured the monthly distribution based on mean daily DBI, categorized as (a) >0 vs 0 (any use) and (b) 0, 0 < DBI ≤ 1, 1 < DBI ≤ 2, and DBI > 2, and examined temporal trends. We described patient-level factors (eg, demographics, healthcare use) associated with high (>2) vs low (0 < DBI≤1) DBI drug exposure. RESULTS: The distribution of the mean daily DBI, aggregated at the month-level, was: 58.1% DBI = 0, 29.0% 0 < DBI≤1, 9.3% 1 < DBI≤2, and 3.7% DBI > 2. Predictors of high monthly DBI drug exposure (DBI > 2) included certain indicators of increased healthcare use (eg, high number of drug claims), white race, younger age, frailty, and a psychosis diagnosis code. CONCLUSIONS: The predictors of high DBI drug exposure can inform discussions between patients and providers about medication appropriateness and potential de-prescribing. Future Medicare-based studies should assess the association between the DBI and adverse events.

Effectiveness of adjuvant FOLFOX vs 5FU/LV in adults over age 65 with stage II and III colon cancer using a novel hybrid approach.

PURPOSE: Estimates of cancer therapy effects can differ in clinical trials and clinical practice, partly due to underrepresentation of certain patient subgroups in trials. We utilize a hybrid approach, combining clinical trial and real-world data, to estimate the comparative effectiveness of two adjuvant chemotherapy regimens for colon cancer. METHODS: We identified patients aged 66 and older enrolled in the Multicenter International Study of Oxaliplatin/5FU-LV in the Adjuvant Treatment of Colon Cancer. Similar patients were identified in the Surveillance, Epidemiology, and End Results (SEER)-Medicare database, initiating adjuvant chemotherapy with either 5-fluorouracil (5FU) alone or in combination with oxaliplatin (FOLFOX). We used logistic regression to estimate the likelihood of trial enrollment as a function of age, sex, and substage. Using inverse odds of sampling weights (IOSW), we compared 5-year mortality in patients randomized to FOLFOX vs 5FU using weighted Cox proportional hazards regression, the Nelson-Aalen estimator for cumulative hazards, and bootstrapping for 95% confidence intervals (CIs). RESULTS: There were 690 trial participants and 3834 SEER-Medicare patients. The SEER-Medicare population was older and had a higher proportion of stage IIIB and IIIC patients than the trial. After controlling for differences between populations, the IOSW 5-year HR was 1.21 (0.89, 1.65), slightly farther from the null than the trial estimate (HR = 1.14, 95%CI: 0.87, 1.49). CONCLUSIONS: This study supports mounting evidence of little to no incremental reduction in 5-year mortality for FOLFOX vs 5FU in older adults with stage II-III colon cancer, emphasizing the importance of combining clinical trial and real-world data to support such conclusions.

Polypharmacy and Incident Frailty in a Longitudinal Community-Based Cohort Study.

OBJECTIVES: Polypharmacy may affect frailty, a common and costly condition among older adults. Frailty prevalence is elevated among racial/ethnic minorities and persons living in the US South, and research is needed to inform future pharmacologic interventions in these populations. Our aim was to quantify the prevalence of frailty and polypharmacy, and to estimate the association between polypharmacy and incident frailty. DESIGN: Prospective cohort study. SETTING: A community-based cohort study of adults residing in Johnston County, North Carolina. PARTICIPANTS: White and African American adults aged 50 to 95 years (n=1697). MEASUREMENTS: At each study visit, all prescription and over-the-counter medications were recorded. We calculated annual polypharmacy (5-9 medications) and excessive polypharmacy (≥10 medications) prevalence at the 2006-2010 visit (n = 1697) and operationalized the Fried frailty phenotype to describe prevalent and incident frailty at two consecutive visits (2006-2010 and 2013-2015). We estimated risk ratios (RRs) and 95% confidence intervals (CIs) for the association between polypharmacy and incident frailty using weighted log-binomial regression to account for measured confounding and attrition using inverse probability of treatment and attrition weights, respectively. RESULTS: At the 2006-2010 visit, 678 (41%) and 260 (16%) participants were exposed to polypharmacy and excessive polypharmacy, respectively. Overall, 353 (21%) participants and 180 (21%) participants were frail at the 2006-2010 and 2013-2015 visits, respectively. Frailty was more common among participants identifying as white, women, and having less educational attainment relative to those without these characteristics. Incident frailty at the 2013-2015 visit was 15% (mean follow-up = 5.5 years). Our results suggest that polypharmacy is positively associated with incident frailty (weighted RR = 1.4; 95% CI = .9-2.0), yet estimates are imprecise and should be interpreted with caution. CONCLUSION: Consistent with the current weight of evidence, our results suggest an association between polypharmacy and incident frailty. Prospective studies evaluating deprescribing interventions are needed to clarify whether reducing polypharmacy decreases frailty incidence. J Am Geriatr Soc 67:2482-2489, 2019.

Residential exposure to vehicular traffic-related air pollution during childhood and breast cancer risk.

BACKGROUND: Some studies have supported an association between traffic-related air pollution exposure and breast cancer risk. However, few studies have considered exposures in early life, which may be a period of increased susceptibility. OBJECTIVES: To examine the association of childhood residential exposure to traffic-related air pollution with breast cancer development. METHODS: The Sister Study is a prospective cohort of 50,884 initially breast cancer-free women, of whom 42,934 provided information at enrollment about roads and traffic near their primary childhood residence before age 14 as well as relevant covariates. Adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) for the association between traffic-related measures at childhood residence and adult incident breast cancer were estimated using Cox regression. RESULTS: During follow-up (mean = 6.3 years), 2,028 breast cancers were diagnosed. Traffic-related characteristics were not consistently associated with breast cancer risk. However, incidence was elevated among women who reported a median/barrier dividing either their primary childhood residential road (aHR = 1.2; 95% CI: 0.9-1.7) or the nearest cross-street (aHR = 1.3; 95% CI: 0.9-1.8, if the cross-street was within 100ft.), and among women whose nearest cross-street had the highest traffic, ≥3 lanes, and/or a median/barrier (aHR = 1.4; 95% CI: 1.0-1.9). CONCLUSIONS: Measures of potential exposure to vehicular traffic were not consistently associated with breast cancer risk. However, living during childhood on or near a road with a median or other barrier, which may be a more easily remembered road characteristic than the others assessed, was associated with increased breast cancer risk.