Potential selection bias associated with using geocoded birth records for epidemiologic research

Industry and occupation in California birth certificates (1998-2019): Reporting disparities and classification codability

BACKGROUND

Missing and noncodable parental industry and occupation (I/O) information on birth certificates (BCs) can bias analyses informing parental worksite exposures and family economic stability.

METHODS

We used the National Institute for Occupational Safety and Health (NIOSH) software to code parental I/O in 1989-2019 California BC data (N = 21,739,406). We assessed I/O missingness and codability by reporting period, parental sex, race/ethnicity, age, and education.

RESULTS

During 1989-2019, records missing I/O increased from 4.4% to 9.4%. I/O was missing more frequently from parents who were male (7.8% vs. 4.4%), Black or American Indian/Alaska Native (AIAN) (9.3% and 8.9% vs. 3.2%-4.7% in others), and had high school or less education (4.0%-5.9% vs. 1.4%-2.6% in others). Of records with I/O, less than 2% were noncodable by NIOSH software. Noncodable entries were more common for parents who were male (industry (1.9% vs. 1.0%); occupation (1.5% vs. 0.7%)), Asian/Pacific Islander (industry (2.4% vs. 1.2%-1.6% in other groups); occupation (1.7% vs. 0.7%-1.5% in other groups)), age 40 and older (industry (2.1% vs. 0.4%-1.7% in younger groups); occupation (1.7% vs. 0.3%-1.3% in younger groups)), and 4-year college graduates (industry (2.0% vs. 1.0%-1.9% in other groups); occupation (1.7% vs. 0.5%-1.4%)).

CONCLUSIONS

In California BC, I/O missingness was systematically higher among parents who are male, Black, AIAN, less than 20 years old, and report no college education. I/O codability is high when information is reported, with small percentage disparities. Improving data collection is vital to equitably describe economic contexts that determine important family outcomes.

The Relation Between Travel Distance and Overall Survival for HPV-Associated Cancers in a High-Burden State

Purpose

To assess the association between travel distance to an academic health system and overall survival for patients with human papillomavirus (HPV)-associated cancers.

Methods

Using hospital-based cancer registry data from 2005-2019, we calculated unidirectional travel distance from each patient's geocoded address to our academic health center through network analysis. We categorized distance as short (<25 miles), intermediate (25-74.9 miles), or long (≥75 miles). The primary outcome was time from the date of initial diagnosis to the date of death or last contact. We used multivariable Cox proportional hazards regression to evaluate the association between travel distance and overall survival. We also estimated the adjusted observed 5-year survival rate.

Results

Patients with HPV-associated cancers traveling distances that were intermediate (hazard ratio [HR], 1.23; 95% CI, 1.06-1.43) and long (HR, 1.15; 95% CI, 1.01-1.32) had a higher hazard of death than the short-distance group. The adjusted 5-year observed survival rates for HPV-associated cancers were lowest in the intermediate-distance group (60.4%) compared with the long-(62.6%) and short-distance (66.2%) groups.

Conclusions

Our findings indicate that travel distance to an academic health center was associated with overall survival for patients with HPV-associated cancers, reflecting the importance of considering travel burden in improving patient outcomes.

Geocoding precision of birth records from 2008 to 2017 in Kentucky, USA

Maternal address information captured on birth records is increasingly used to estimate residential environmental exposures during pregnancy. However, there has been limited assessment of the geocoding precision of birth records, particularly since the adoption of the 2003 standard birth certificate in 2015. To address this gap, this study evaluated the geocoding precision of live and stillbirth records of Kentucky residents over ten years, from 2008 through 2017. This study summarized the demographic characteristics of imprecisely geocoded records and, using a bivariate logistic regression, identified covariates associated with poor geocoding precision among three population density designations-metro, non-metro, and rural. We found that in metro areas, after adjusting for area deprivation, education, and the race, age and education of both parents, records for Black mothers had 48% lower odds of imprecise geocoding (aOR=0.52, 95% CI: 0.48, 0.56), while Black women in rural areas had 96% higher odds of imprecise geocoding (aOr=1.96, 95% CI: 1.68, 2.28). This study also found that over the study period, rural and non-metro areas began with a high proportion of imprecisely geocoded records (38% in rural areas, 19% in non-metro), but both experienced an 8% decline in imprecisely geocoded records over the study period (aOr=0.92, 95% CI: 0.92, 0.94). This study shows that, while geocoding precision has improved in Kentucky, further work is needed to improve geocoding in rural areas and address racial and ethnic disparities.

Biases arising from linked administrative data for epidemiological research: a conceptual framework from registration to analyses

Linked administrative data offer a rich source of information that can be harnessed to describe patterns of disease, understand their causes and evaluate interventions. However, administrative data are primarily collected for operational reasons such as recording vital events for legal purposes, and planning, provision and monitoring of services. The processes involved in generating and linking administrative datasets may generate sources of bias that are often not adequately considered by researchers. We provide a framework describing these biases, drawing on our experiences of using the 100 Million Brazilian Cohort (100MCohort) which contains records of more than 131 million people whose families applied for social assistance between 2001 and 2018. Datasets for epidemiological research were derived by linking the 100MCohort to health-related databases such as the Mortality Information System and the Hospital Information System. Using the framework, we demonstrate how selection and misclassification biases may be introduced in three different stages: registering and recording of people's life events and use of services, linkage across administrative databases, and cleaning and coding of variables from derived datasets. Finally, we suggest eight recommendations which may reduce biases when analysing data from administrative sources.

Uncertainty in geospatial health: challenges and opportunities ahead

PURPOSE

Uncertainty is not always well captured, understood, or modeled properly, and can bias the robustness of complex relationships, such as the association between the environment and public health through exposure, estimates of geographic accessibility and cluster detection, to name a few.

METHODS

We review current challenges and future opportunities as geospatial data and analyses are applied to the field of public health. We are particularly interested in the sources of uncertainty in geospatial data and how this uncertainty may propagate in spatial analysis.

RESULTS

We present opportunities to reduce the magnitude and impact of uncertainty. Specifically, we focus on (1) the use of multiple reference data sources to reduce geocoding errors, (2) the validity of online geocoders and how confidentiality (e.g., HIPAA) may be breached, (3) use of multiple reference data sources to reduce geocoding errors, (4) the impact of geoimputation techniques on travel estimates, (5) residential mobility and how it affects accessibility metrics and clustering, and (6) modeling errors in the American Community Survey. Our paper discusses how to communicate spatial and spatiotemporal uncertainty, and high-performance computing to conduct large amounts of simulations to ultimately increase statistical robustness for studies in public health.

CONCLUSIONS

Our paper contributes to recent efforts to fill in knowledge gaps at the intersection of spatial uncertainty and public health.

Traffic-related metrics and adverse birth outcomes: A systematic review and meta-analysis

Given the inconsistency of epidemiologic evidence for associations between maternal exposures to traffic-related metrics and adverse birth outcomes, this manuscript aims to provide clarity on this topic. Pooled meta-estimates were calculated using random-effects analyses. Subgroup analyses were conducted by study area, study design, and Newcastle-Ottawa quality score (NOS). Funnel plots and Egger's test were conducted to evaluate the publication bias, and Fail-safe Numbers (Fail-safe N) were measured to evaluate the robustness of models. From the initial 740 studies (last search, July 11, 2019), 26 studies were included in our analysis. The pooled odds ratio for the change in small for gestational age associated with per 500 m decrease in the distance to roads was 1.016 (95% CI: 1.004, 1.029). Subgroup analyses revealed significant positive associations between term low birth weight and traffic density in higher-quality literatures with higher NOS [1.060 (95% CI: 1.002, 1.121)], cohort studies [1.020 (95% CI: 1.006, 1.033)], and studies in North America [1.018 (95% CI: 1.005, 1.131)]. The buffer of traffic density made no difference in the effect size. Traffic density seemed to be a better indicator of traffic pollution than the distance to roads.

Impact of geo-imputation on epidemiologic associations in a study of outdoor air pollution and respiratory hospitalization

Imputation of missing spatial attributes in health records may facilitate linkages to geo-referenced environmental exposures, but few studies have assessed geo-imputation impacts on epidemiologic inference. We imputed patient Census tracts in a case-crossover analysis of fine particulate matter (PM_2.5) and respiratory hospitalizations in New York State (2000-2005). We observed non-significantly higher PM_2.5 exposures, high accuracy of binary exposure assignment (89 to 99%), and marginally different hazard ratios (HRs) (-0.2 to 0.7%). HR differences were greater in urban versus rural areas. Given its efficiency and nominal influence on accuracy of exposure classification and measures of association, geo-imputation is a candidate method to address missing spatial attributes for health studies.

Strong upward neighborhood mobility and preterm birth: a matched-sibling design approach

PURPOSE

Much cross-sectional work reports an association between neighborhood disadvantage and adverse birth outcomes. Limitations of this work include that mothers with pre-existing unmeasured morbidity may "select" into, or out of, certain neighborhoods. This selection issue remains a key rival explanation for work concerned with place-based disparities in birth outcomes. We move beyond a cross-sectional approach and exploit a sibling-linked data set in California to test whether upwardly mobile mothers, who move from a very high to a very low disadvantaged neighborhood, exhibit a lower than expected risk of preterm birth (PTB) (i.e., delivery <37 weeks).

METHODS

We used a matched-sibling design for 461,061 sibling pairs (i.e., 922,122 births total) to mothers in urban areas in who gave birth in California from 2005 to 2010. We linked mother's address at two time points (i.e., two sibling birth dates) to a census-derived composite indicator of neighborhood disadvantage. Conditional logistic regression methods controlled for mother's risk of PTB in the sibling delivered before the move when estimating the relation between strong upward mobility and preterm of the subsequent birth after the move.

RESULTS

As hypothesized, strong upward mobility (relative to no mobility) varies inversely with the odds of PTB of the second sibling (odds ratio [OR] for PTB = 0.83, 95% confidence interval: 0.74, 0.93).

CONCLUSIONS

Mothers moving from very high to very low disadvantaged neighborhoods show a reduced odds of PTB. Our findings, if replicated, raise the possibility that improvements in neighborhood quality may improve perinatal health in a relatively short time span.

Public and Population Health Informatics: The Bridging of Big Data to Benefit Communities

Objective:  To summarize the recent public and population health informatics literature with a focus on the synergistic “bridging” of electronic data to benefit communities and other populations.

Methods:  The review was primarily driven by a search of the literature from July 1, 2016 to September 30, 2017. The search included articles indexed in PubMed using subject headings with (MeSH) keywords “public health informatics” and “social determinants of health”. The “social determinants of health” search was refined to include articles that contained the keywords “public health”, “population health” or “surveillance”.

Results:  Several categories were observed in the review focusing on public health's socio-technical infrastructure: evaluation of surveillance practices, surveillance methods, interoperable health information infrastructure, mobile health, social media, and population health. Common trends discussing socio-technical infrastructure included big data platforms, social determinants of health, geographical information systems, novel data sources, and new visualization techniques. A common thread connected these categories of workforce, governance, and sustainability: using clinical resources and data to bridge public and population health.

Conclusions:  Both medical care providers and public health agencies are increasingly using informatics and big data tools to create and share digital information. The intent of this “bridging” is to proactively identify, monitor, and improve a range of medical, environmental, and social factors relevant to the health of communities. These efforts show a significant growth in a range of population health-centric information exchange and analytics activities.

Retirements of Coal and Oil Power Plants in California: Association With Reduced Preterm Birth Among Populations Nearby

Coal and oil power plant retirements reduce air pollution nearby, but few studies have leveraged these natural experiments for public health research. We used California Department of Public Health birth records and US Energy Information Administration data from 2001-2011 to evaluate the relationship between the retirements of 8 coal and oil power plants and nearby preterm (gestational age of <37 weeks) birth. We conducted a difference-in-differences analysis using adjusted linear mixed models that included 57,005 births-6.3% of which were preterm-to compare the probability of preterm birth before and after power plant retirement among mothers residing within 0-5 km and 5-10 km of the 8 power plants. We found that power plant retirements were associated with a decrease in the proportion of preterm birth within 5 km (-0.019, 95% CI: -0.031, -0.008) and 5-10 km (-0.015, 95% CI: -0.024, -0.007), controlling for secular trends with mothers living 10-20 km away. For the 0-5-km area, this corresponds to a reduction in preterm birth from 7.0% to 5.1%. Subgroup analyses indicated a potentially larger association among non-Hispanic black and Asian mothers than among non-Hispanic white and Hispanic mothers and no differences in educational attainment. Future coal and oil power plant retirements may reduce preterm birth among nearby populations.

Retirements of Coal and Oil Power Plants in California: Association With Reduced Preterm Birth Among Populations Nearby

Coal and oil power plant retirements reduce air pollution nearby, but few studies have leveraged these natural experiments for public health research. We used California Department of Public Health birth records and US Energy Information Administration data from 2001-2011 to evaluate the relationship between the retirements of 8 coal and oil power plants and nearby preterm (gestational age of <37 weeks) birth. We conducted a difference-in-differences analysis using adjusted linear mixed models that included 57,005 births-6.3% of which were preterm-to compare the probability of preterm birth before and after power plant retirement among mothers residing within 0-5 km and 5-10 km of the 8 power plants. We found that power plant retirements were associated with a decrease in the proportion of preterm birth within 5 km (-0.019, 95% CI: -0.031, -0.008) and 5-10 km (-0.015, 95% CI: -0.024, -0.007), controlling for secular trends with mothers living 10-20 km away. For the 0-5-km area, this corresponds to a reduction in preterm birth from 7.0% to 5.1%. Subgroup analyses indicated a potentially larger association among non-Hispanic black and Asian mothers than among non-Hispanic white and Hispanic mothers and no differences in educational attainment. Future coal and oil power plant retirements may reduce preterm birth among nearby populations.

Decentralized and reproducible geocoding and characterization of community and environmental exposures for multisite studies

Objective

Geocoding and characterizing geographic, community, and environmental characteristics of study participants is frequently done in epidemiological studies. However, participant addresses are identifiable protected health information (PHI) and geocoding must be conducted in a Health Insurance Portability and Accountability Act–compliant manner. Our objective was to create a software application for this process that addresses limitations in current approaches.

Materials and Methods

We used a containerization platform to create DeGAUSS (Decentralized Geomarker Assessment for Multi-Site Studies), a software application that facilitates reproducible geocoding and geomarker assessment while maintaining the confidentiality of PHI. To validate the software, 215 350 addresses in Hamilton County, Ohio, were geocoded using DeGAUSS, ArcGIS, Google, and SAS and compared to a gold-standard approach. We distributed the DeGAUSS software to sites in an ongoing multisite study (Electronic Medical Records and Genomics, or eMERGE), and individual sites independently geocoded and assigned median census tract–level income and distance to nearest major roadway to their participants’ addresses, removed associated PHI, and returned deidentified data.

Results

Within a multisite study, 52 244 study participants’ addresses across 5 sites were geocoded with a median distance to roadway of 10 022m and a median census tract income of $57 266, demonstrating the feasibility of DeGAUSS within a multisite study. Compared to other commonly used geocoding platforms, DeGAUSS had similar geocoding and geomarker assessment accuracies.

Conclusion

The open source DeGAUSS software overcomes multiple challenges in the use of address data in multisite studies and also serves as a more general reproducible research tool for geocoding and geomarker assessment.