Matching participant address with public records database in a US national longitudinal cohort study

Lifetime residential data collection protocol for the Adolescent Brain Cognitive Development (ABCD) Study

Understanding the impacts of environmental exposures on health outcomes during development is an important area of research for plenty of reasons. Collecting retrospective and prospective residential history can enrich observational studies through eventual linkages to external sources. Augmenting participant health outcome data with environmental data can better inform on the role of the environment, thereby enhancing prevention and intervention efforts. However, collecting the geospatial information needed for this type of research can be difficult, especially when data are collected directly from participants. Participants' residential histories are unique and often complex. Collecting residential history data often involves capturing precise spatial locations along specific timeframes as well as contending with recall bias and unique, complex living arrangements. When trying to assess lifetime environmental exposures, researchers must consider the many changes in location a person goes through and the timeframes in which these changes occur, ultimately creating a multidimensional and dynamic dataset. Creating data collection protocols that are feasible to administer, result in accurate data, and minimize data missingness is a major challenge to undertake. Here, we provide an overview of the protocol developed to collect the lifetime residential address information of participants in the Adolescent Brain Cognitive Development (ABCD) Study.

Lifetime residential history collection and processing for environmental data linkages in the ABCD study

By using geospatial information such as participants' residential history along with external datasets of environmental exposures, ongoing studies can enrich their cohorts to investigate the role of the environment on brain-behavior health outcomes. However, challenges may arise if clear guidance and key quality control steps are not taken at the outset of data collection of residential information. Here, we detail the protocol development aimed at improving the collection of lifetime residential address information from the Adolescent Brain Cognitive Development (ABCD) Study. This protocol generates a workflow for minimizing gaps in residential information, improving data collection processes, and reducing misclassification error in exposure estimates.

Construction of residential histories to estimate long-term environmental exposures in the California Teachers Study cohort

Environmental epidemiologic studies using geospatial data often estimate exposure at a participant's residence upon enrollment, but mobility during the exposure period can lead to misclassification. We aimed to mitigate this issue by constructing residential histories for participants in the California Teachers Study through follow-up (1995-2018). Address records have been collected from the US Postal Service, LexisNexis, Experian, and California Cancer Registry. We identified records of the same address based on geo-coordinate distance (≤250 m) and street name similarity. We consolidated addresses, prioritizing those confirmed by participants during follow-up questionnaires, and estimating the duration lived at each address using dates associated with records (e.g., date-first-seen). During 23 years of follow-up, about half of participants moved (48%, including 14% out-of-state). We observed greater mobility among younger women, Hispanic/Latino women, and those in metropolitan and lower socioeconomic status areas. The cumulative proportion of in-state movers remaining eligible for analysis was 21%, 32%, and 41% at 5, 10, and 20 years post enrollment, respectively. Using self-reported information collected 10 years after enrollment, we correctly identified 94% of movers and 95% of non-movers as having moved or not moved from their enrollment address. This dataset provides a foundation for estimating long-term environmental exposures in diverse epidemiologic studies in this cohort. IMPACT: Our efforts in constructing residential histories for California Teachers Study participants through follow-up (1995-2018) benefit future environmental epidemiologic studies. Address availability during the exposure period can mitigate misclassification due to residential changes, especially when evaluating long-term exposures and chronic health outcomes. This can reduce differential misclassification among more mobile subgroups, including younger women and those from lower socioeconomic and urban areas. Our approach to consolidating addresses from multiple sources showed high accuracy in comparison to self-reported residential information. The residential dataset produced from this analysis provides a valuable tool for future studies, ultimately enhancing our understanding of environmental health impacts.

Utilizing residential histories to assess environmental exposure and socioeconomic status over the life course among mesothelioma patients

Background

Exposure misclassification based solely on the address at cancer diagnosis has been widely recognized though not commonly assessed.

Methods

We linked 1,015 mesothelioma cases diagnosed during 2011-2015 from the New York State Cancer Registry to inpatient claims and LexisNexis administrative data and constructed residential histories. Percentile ranking of exposure to ambient air toxics and socioeconomic status (SES) were based on the National Air Toxic Assessment and United States Census data, respectively. To facilitate comparisons over time, relative exposures (REs) were calculated by dividing the percentile ranking at individual census tract by the state-level average and subtracting one. We used generalized linear regression models to compare the RE in the past with that at cancer diagnosis, adjusting for patient-level characteristics.

Results

Approximately 43.7% of patients had residential information available for up to 30 years, and 96.0% up to 5 years. The median number of unique places lived was 4 [interquartile range (IQR), 2-6]. The time-weighted-average RE from all addresses available had a median of -0.11 (IQR, -0.50 to 0.30) for air toxics and -0.28 (IQR, -0.65 to 0.25) for SES. RE associated with air toxics (but not SES) was significantly higher for earlier addresses than addresses at cancer diagnosis for the 5-year [annual increase =1.24%; 95% confidence interval (CI): 0.71-1.77%; n=974] and 30-year (annual increase =0.36%; 95% CI: 0.25-0.48%; n=444) look-back windows, respectively.

Conclusions

Environmental exposure to non-asbestos air toxics among mesothelioma patients may be underestimated if based solely on the address at diagnosis. With geospatial data becoming more readily available, incorporating cancer patients' residential history would lead to reduced exposure misclassification and accurate health risk estimates.

The association of baseline depressive symptoms and stress on withdrawal in a national longitudinal cohort: the REGARDS study

PURPOSE

To measure the association of baseline psychological symptoms (depressive symptoms and perceived stress) with withdrawal from a cohort study.

METHODS

Depressive symptoms and perceived stress were obtained using validated measures during the baseline computer-assisted telephonic interview for the REasons for Geographic and Racial Differences in Stroke study a national longitudinal cohort (≥45 years, 42% Black, 55% women) recruited between 2003 and 2007. Participants who completed follow-up after September 1, 2019, were considered active. Primary outcome was time to study withdrawal. The association of psychological symptoms and time-to-withdrawal was measured using Cox proportional hazard regression models with incremental adjustments by demographic and clinical factors.

RESULTS

Out of 29,964 participants included in the analysis, 11,111 (37.1%) participants withdrew over the follow-up period (median: 11 years). Compared to participants with low depressive symptoms, those with moderate symptoms had 5% higher risk (aHR= 1.05; 95% CI= 1.00-1.10) and those with high level of depressive had 19% higher risk (aHR= 1.19; 95% CI= 1.11-1.27) of withdrawal in fully adjusted models. No significant association between perceived stress and withdrawal risk was observed.

CONCLUSIONS

Depressive symptoms were significantly associated with withdrawal. Prevalence of depressive symptoms at baseline is an important indicator of participant retention in large prospective cohorts.

A longitudinal examination of objective neighborhood walkability, body mass index, and waist circumference: the REasons for Geographic And Racial Differences in Stroke study

Background

Studies have shown neighborhood walkability is associated with obesity. To advance this research, study designs involving longer follow-up, broader geographic regions, appropriate neighborhood characterization, assessment of exposure length and severity, and consideration of stayers and movers are needed. Using a cohort spanning the conterminous United States, this study examines the longitudinal relationship between a network buffer-derived, duration-weighted neighborhood walkability measure and two adiposity-related outcomes.

Methods

This study included 12,846 Black/African American and White adults in the REasons for Geographic And Racial Differences in Stroke study. Body mass index (BMI) and waist circumference (WC) were assessed at baseline and up to 13.3 years later (M (SD) = 9.4 (1.0) years). BMI and WC were dichotomized. Walk Score® was duration-weighted based on time at each address and categorized as Very Car-Dependent, Car-Dependent, Somewhat Walkable, Very Walkable, and Walker’s Paradise. Unadjusted and adjusted logistic regression models tested each neighborhood walkability-adiposity association. Adjusted models controlled for demographics, health factors, neighborhood socioeconomic status, follow-up time, and either baseline BMI or baseline WC. Adjusted models also tested for interactions. Post-estimation Wald tests examined whether categorical variables had coefficients jointly equal to zero. Orthogonal polynomial contrasts tested for a linear trend in the neighborhood walkability-adiposity relationships.

Results

The odds of being overweight/obese at follow-up were lower for residents with duration-weighted Walk Score® values in the Walker’s Paradise range and residents with values in the Very Walkable range compared to residents with values in the Very Car-Dependent range. Residents with duration-weighted Walk Score® values classified as Very Walkable had significantly lower odds of having a moderate-to-high risk WC at follow-up relative to those in the Very Car-Dependent range. For both outcomes, the effects were small but meaningful. The negative linear trend was significant for BMI but not WC.

Conclusion

People with cumulative neighborhood walkability scores in the Walker’s Paradise range were less likely to be overweight/obese independent of other factors, while people with scores in the Very Walkable range were less likely to be overweight/obese and less likely to have a moderate-to-high risk WC. Addressing neighborhood walkability is one approach to combating obesity.

Health care experiences during the COVID-19 pandemic by race and social determinants of health among adults age ≥ 58 years in the REGARDS study

BACKGROUND

Understanding health care experiences during the COVID-19 pandemic may provide insights into patient needs and inform policy. The objective of this study was to describe health care experiences by race and social determinants of health.

METHODS

We conducted a telephone survey (July 6, 2020-September 4, 2021) among 9492 Black and White participants in the longitudinal REasons for Geographic And Racial Differences in Stroke cohort study, age 58-105 years, from the continental United States. Among participants with symptoms of COVID-19, outcomes were: 1. Sought care or advice for the illness; 2. Received a SARS-CoV-2 test for the illness; and 3. Tested positive. Among participants without symptoms of COVID-19, outcomes were: 1. Wanted a test; 2. Wanted and received a test; 3. Did not want but received a test; and 4. Tested positive. We examined these outcomes overall and in subgroups defined by race, household income, marital status, education, area-level poverty, rural residence, Medicaid expansion, public health infrastructure ranking, and residential segregation.

RESULTS

The average age of participants was 76.8 years, 36% were Black, and 57% were female. Among participants with COVID-19 symptoms (n = 697), 74% sought care or advice for the illness, 50% received a SARS-CoV-2 test, and 25% had a positive test (50% of those tested). Among participants without potential COVID-19 symptoms (n = 8795), 29% wanted a SARS-CoV-2 test, 22% wanted and received a test, 8% did not want but received a test, and 1% tested positive; a greater percentage of participants who were Black compared to White wanted (38% vs 23%, p < 0.001) and received tests (30% vs 18%, p < 0.001) and tested positive (1.4% vs 0.8%, p = 0.005).

CONCLUSIONS

In this national study of older US adults, many participants with potential COVID-19 symptoms and asymptomatic participants who desired testing did not receive COVID-19 testing.