Faculty Bibliography

OBJECTIVE:To develop a natural language processing (NLP) algorithm that identifies social determinants of health (SDoH), including housing, transportation, food, and medication insecurities, social isolation, abuse, neglect, or exploitation, and financial difficulties for patients with Alzheimer's disease and related dementias (ADRD) from unstructured electronic health records (EHRs). DATA SOURCES AND STUDY SETTING:We leveraged 1000 medical notes randomly selected from 7401 emergency department and inpatient social worker notes generated between 2015 and 2019 for 231 unique patients diagnosed with ADRD at Michigan Medicine. STUDY DESIGN:We developed a rule-based NLP algorithm for the identification of seven domains of SDoH noted above. We also compared the rule-based algorithm with deep learning and regularized logistic regression approaches. These models were compared using accuracy, sensitivity, specificity, F1 score, and the area under the receiver operating characteristic curve (AUC). All notes were split into 700 notes for training NLP algorithms, and 300 notes for validation. DATA COLLECTION/EXTRACTION METHODS:Social worker notes used in this study were extracted from the Michigan Medicine EHR database. PRINCIPAL FINDINGS:Of the 700 notes for training, F1 and AUC for the rule-based algorithm were at least 0.94 and 0.95, respectively, for all SDoH categories. Of the 300 notes for validation, F1 and AUC were at least 0.80 and 0.97, respectively, for all SDoH except housing and medication insecurities. The deep learning and regularized logistic regression algorithms had unsatisfactory performance. CONCLUSIONS:The rule-based algorithm can accurately extract SDoH information in all seven domains of SDoH except housing and medication insecurities. Findings from the algorithm can be used by clinicians and social workers to proactively address social needs of patients with ADRD and other vulnerable patient populations.

BACKGROUND:Podcasts, or episodic digital audio recordings, represent a novel way to reach large audiences for public education. Genetic evaluation has important implications for prostate cancer (PCa) care but is underutilized. We created a series of five podcasts about PCa genetics and tested their usefulness in raising awareness and providing education to lay audiences. METHODS:We recruited 157 men and women from the general public and 100 patients with PCa from across the U.S., who listened to a podcast and completed an online survey. The primary outcome was the perceived usefulness of the podcast (score ≥5 on a published 7-point Likert scale). Secondary outcomes were relevance to informational needs, satisfaction and ease of use, as well as genetic knowledge and attitudes toward genetic testing after listening to the podcasts. RESULTS:The podcasts were associated with high mean scores for perceived usefulness (5.6/7), relevance to informational needs (5.6/7), satisfaction (5.8/7), and ease of use (5.9/7). After listening to the podcasts, 80-100% correctly answered most key knowledge questions about PCa genetics, and 85% had a positive attitude toward genetic testing. On multivariable analysis, the perceived usefulness of the podcasts was higher among Black/Hispanic adults (p = 0.05) and those with a family history of PCa (p = 0.01). CONCLUSIONS:A podcast series on PCa genetics was perceived as useful and associated with high rates of knowledge for patients with PCa and the general public. Podcasts represent a promising new educational tool to raise awareness about PCa genetic evaluation, particularly for high-risk groups.

Stakeholders need data on health and drivers of health parsed to the boundaries of essential policy-relevant geographies. US Congressional Districts are an example of a policy-relevant geography which generally lack health data. One strategy to generate Congressional District heath data metric estimates is to aggregate estimates from other geographies, for example, from counties or census tracts to Congressional Districts. Doing so requires several methodological decisions. We refine a method to aggregate health metric estimates from one geography to another, using a population weighted approach. The method's accuracy is evaluated by comparing three aggregated metric estimates to metric estimates from the US Census American Community Survey for the same years: Broadband Access, High School Completion, and Unemployment. We then conducted four sensitivity analyses testing: the effect of aggregating counts vs. percentages; impacts of component geography size and data missingness; and extent of population overlap between component and target geographies. Aggregated estimates were very similar to estimates for identical metrics drawn directly from the data source. Sensitivity analyses suggest the following best practices for Congressional district-based metrics: utilizing smaller, more plentiful geographies like census tracts as opposed to larger, less plentiful geographies like counties, despite potential for less stable estimates in smaller geographies; favoring geographies with higher percentage population overlap.

BACKGROUND/UNASSIGNED:Nightclub/festival attendees are a population with high rates of party drug use, but research is needed to determine whether there have been shifts in unintended drug exposure in this population (e.g., via adulterants) to inform prevention and harm reduction efforts. METHODS/UNASSIGNED:Adults entering nightclubs and festivals in New York City were asked about past-year drug use in 2016 through 2022, with a subset providing a hair sample for testing. We focused on the 1943 who reported ecstasy use (of which 247 had a hair sample analyzed) and compared trends in self-reported drug use, drug positivity, and adjusted prevalence (adjusting for unreported use). RESULTS/UNASSIGNED:MDMA positivity decreased from 74.4 % to 42.3 %, and decreases occurred regarding detection of synthetic cathinones ("bath salts"; a 100.0 % decrease), MDA (a 76.9 % decrease), amphetamine (an 81.3 % decrease), methamphetamine (a 64.2 % decrease), and ketamine (a 33.4 % decrease) (ps < .05). Although prevalence of MDA and synthetic cathinone use was comparable between self-report and adjusted report in 2022, gaps in prevalence were wider in 2016 (ps < .01). Adjusted prevalence of synthetic cathinone use decreased more across time than prevalence based on self-report (a 79.4 % vs. 69.1 % decrease) and adjusted report for MDA use decreased more than prevalence based on self-report (a 50.6 % vs. 38.9 % decrease). CONCLUSIONS/UNASSIGNED:Combining self-report and toxicology tests helped us determine that decreases in drug use/exposure were steeper regarding adjusted prevalence. Underreported drug exposure-possibly due to exposure to adulterants-appears to have had less of an effect on prevalence in 2022 than it did in 2016.

BACKGROUND:In children with chronic kidney disease (CKD), certain risk factors are associated with faster eGFR decline and earlier kidney failure. Whether these factors have lingering effects on post-transplant eGFR trajectory remains unclear. We characterized pre- and post-transplant eGFR trajectories in pediatric kidney transplant recipients by their pre-kidney replacement therapy (KRT) risk factors. METHODS:We studied eGFR trajectories before KRT initiation and after transplantation among Chronic Kidney Disease in Children (CKiD) Study participants. We used mixed-effects models to compare pre-KRT versus post-transplant eGFR trajectories within individual participants by 7 pre-KRT risk factors: glomerular/non-glomerular etiology, race, preemptive transplant, proteinuria, albuminuria, and systolic/diastolic blood pressure (SBP/DBP). RESULTS:We analyzed 1602 pre-KRT and 592 post-transplant eGFR measurements from 246 transplant recipients. Mean annual eGFR decline was decreased from 18.0% pre-KRT (95%CI, 16.1-19.8) to 5.0% post-transplant (95%CI, 3.3-6.7). All 7 pre-KRT risk factors showed strong associations with faster pre-KRT eGFR decline, but not with post-transplant eGFR decline; only albuminuria, high SBP, and high DBP reached statistical significance with notably attenuated associations. In our multivariable model of the pre-KRT risk factors, post-transplant eGFR decline was more rapid only when albuminuria and high SBP were both present. CONCLUSIONS:eGFR decline substantially slows down after transplant even among children with rapidly progressing forms of CKD. Nonetheless, those who had albuminuria and high SBP before KRT might continue to show faster eGFR decline after transplant, specifically when both risk factors were present. This subgroup might benefit from intensive pre-transplant management for at least one of the two risk factors. A higher resolution version of the Graphical abstract is available as Supplementary information.

A more granular donor kidney grading scale, the kidney donor profile index (KDPI), has recently emerged in contradistinction to the standard criteria donor/expanded criteria donor framework. In this paper, we built a Markov decision process model to evaluate the survival, quality-adjusted life years (QALY), and cost advantages of using high-KDPI kidneys based on multiple KDPI strata over a 60-month time horizon as opposed to remaining on the waiting list waiting for a lower-KDPI kidney. Data for the model were gathered from the Scientific Registry of Transplant Recipients and the United States Renal Data System Medicare parts A, B, and D databases. Of the 129,024 phenotypes delineated in this model, 65% of them would experience a survival benefit, 81% would experience an increase in QALYs, 87% would see cost-savings, and 76% would experience cost-savings per QALY from accepting a high-KDPI kidney rather than remaining on the waiting list waiting for a kidney of lower-KDPI. Classification and regression tree analysis (CART) revealed the main drivers of increased survival in accepting high-KDPI kidneys were wait time ≥30 months, panel reactive antibody (PRA) <90, age ≥45 to 65, diagnosis leading to renal failure, and prior transplantation. The CART analysis showed the main drivers of increased QALYs in accepting high-kidneys were wait time ≥30 months, PRA <90, and age ≥55 to 65.

Research has linked spatial concentrations of incarceration with racial disparities in adverse birth outcomes. However, little is known about the specific mechanisms of this association. This represents an important knowledge gap in terms of intervention. We theorize two pathways that may account for the association between county-level prison rates and adverse birth outcomes: (1) community-level mental distress and (2) reduced health care access. Examining these mechanisms, we conducted a cross-sectional study of county-level prison rates, community-level mental distress, health insurance, availability of primary care physicians (PCP) and mental health providers (MHP), and adverse birth outcomes (preterm birth, low birth weight, infant mortality). Our data set included 475 counties and represented 2,677,840 live U.S. births in 2016. Main analyses involved between 170 and 326 counties. All data came from publicly available sources, including the U.S. Census and the Centers for Disease Control and Prevention. Descriptive and regression results confirmed the link between prison rates and adverse birth outcomes and highlighted Black-White inequities in this association. Further, bootstrap mediation analyses indicated that the impact of spatially concentrated prison rates on preterm birth was mediated by PCP, MHP, community-level mental distress, and health insurance in both crude and adjusted models. Community-level mental distress and health insurance (but not PCP or MHP) similarly mediated low birthweight in both models. Mediators were less stable in the effect on infant mortality with only MHP mediating consistently across models. We conclude that mass incarceration, health care access, and community mental distress represent actionable and urgent targets for structural-, community-, and individual-level interventions targeting population inequities in birth outcomes.

OBJECTIVE:To evaluate the association between census-tract level measures of social vulnerability and residential segregation and IPN follow up. METHODS:This retrospective cohort study included patients with IPN ≥6 mm in size or multiple subsolid/ground-glass IPNs <6 mm (with non-optional follow-up recommendations) diagnosed between January 1, 2018 and December 30, 2019 at a large urban tertiary center and followed ≥two years. Geographic sociodemographic context was characterized by 2018 U.S. Centers for Disease Control and Prevention Social Vulnerability Index (SVI) and the Index of Concentration at the Extreme (ICE), categorized in quartiles. Multivariable binomial regression models were utilized with a primary outcome of inappropriate IPN follow up (late or no follow up). Models were also stratified by nodule risk. RESULTS:The study consisted of 2,492 patients (mean age 65.6 years +/- 12.6 years; 1,361 women). Top-quartile SVI patients were more likely to have inappropriate follow up (Risk Ratio [RR]: 1.24, 95% Confidence Interval [95% CI], 1.12-1.36]), compared with the bottom quartile; risk was also elevated in top-quartile SVI subcategories of Socioeconomic Status (RR: 1.23, 95% CI, 1.13-1.34), Minority Status and Language (RR: 1.24, 95% CI, 1.03-1.48), Housing and Transportation (RR: 1.13, 95% CI, 1.02-1.26), and ICE (RR: 1.20, 95% CI, 1.11-1.30). Further, top-quartile ICE was associated with greater risk of inappropriate follow up among high-risk vs. lower-risk IPN (1.33 [1.18-1.50] vs. 1.13 [1.02-1.25]), respectively, P for interaction= 0.017). DISCUSSION/CONCLUSIONS:Local social vulnerability and residential segregation are associated with inappropriate IPN follow up and may inform policy or interventions tailored for neighborhoods.

Access to and utilization of consumer credit remains an understudied social determinant of health. We examined associations between a novel, small-area, multidimensional credit insecurity index (CII), and the prevalence of self-reported frequent mental distress across US cities in 2020. The census tract-level CII was developed by the Federal Reserve Bank of New York using Census population information and a nationally representative sample of anonymized Equifax credit report data. The CII was calculated for tracts in 766 cities displayed on the City Health Dashboard at the time of analysis, predominantly representing cities with over 50,000 residents. The CII combined data on tract-level participation in the formal credit economy with information on the percent of individuals without revolving credit, percent with high credit utilization, and percent with deep subprime credit scores. Tracts were classified as credit-assured, credit-likely, mid-tier, at-risk, or credit-insecure. We used linear regression to examine associations between the CII and a modeled tract-level measure of frequent mental distress, obtained from the CDC PLACES project. Regression models were adjusted for neighborhood economic and demographic characteristics. We examined effect modification by US region by including two-way interaction terms in regression models. In adjusted models, credit-insecure tracts had a modestly higher prevalence of frequent mental distress (prevalence difference = 0.38 percentage points; 95% CI = 0.32, 0.44), compared to credit-assured tracts. Associations were most pronounced in the Midwest. Local factors impacting credit access and utilization are often modifiable. The CII, a novel indicator of community financial well-being, may be an independent predictor of neighborhood health in US cities and could illuminate policy targets to improve access to desirable credit products and downstream health outcomes.