Predicting health-related social needs in Medicaid and Medicare populations using machine learning

Understanding individual health-related social needs in the context of area-level social determinants of health: The case for granularity

Introduction

Screening for health-related social needs (HRSNs) within health systems is a widely accepted recommendation, however challenging to implement. Aggregate area-level metrics of social determinants of health (SDoH) are easily accessible and have been used as proxies in the interim. However, gaps remain in our understanding of the relationships between these measurement methodologies. This study assesses the relationships between three area-level SDoH measures, Area Deprivation Index (ADI), Social Deprivation Index (SDI) and Social Vulnerability Index (SVI), and individual HRSNs among patients within one large urban health system.

Methods

Patients screened for HRSNs between 2018 and 2019 (N = 45,312) were included in the analysis. Multivariable logistic regression models assessed the association between area-level SDoH scores and individual HRSNs. Bivariate choropleth maps displayed the intersection of area-level SDoH and individual HRSNs, and the sensitivity, specificity, and positive and negative predictive values of the three area-level metrics were assessed in relation to individual HRSNs.

Results

The SDI and SVI were significantly associated with HRSNs in areas with high SDoH scores, with strong specificity and positive predictive values (∼83% and ∼78%) but poor sensitivity and negative predictive values (∼54% and 62%). The strength of these associations and predictive values was poor in areas with low SDoH scores.

Conclusions

While limitations exist in utilizing area-level SDoH metrics as proxies for individual social risk, understanding where and how these data can be useful in combination is critical both for meeting the immediate needs of individuals and for strengthening the advocacy platform needed for resource allocation across communities.

Development of a Social Risk Score in the Electronic Health Record to Identify Social Needs Among Underserved Populations: Retrospective Study

BACKGROUND

Patients with unmet social needs and social determinants of health (SDOH) challenges continue to face a disproportionate risk of increased prevalence of disease, health care use, higher health care costs, and worse outcomes. Some existing predictive models have used the available data on social needs and SDOH challenges to predict health-related social needs or the need for various social service referrals. Despite these one-off efforts, the work to date suggests that many technical and organizational challenges must be surmounted before SDOH-integrated solutions can be implemented on an ongoing, wide-scale basis within most US-based health care organizations.

OBJECTIVE

We aimed to retrieve available information in the electronic health record (EHR) relevant to the identification of persons with social needs and to develop a social risk score for use within clinical practice to better identify patients at risk of having future social needs.

METHODS

We conducted a retrospective study using EHR data (2016-2021) and data from the US Census American Community Survey. We developed a prospective model using current year-1 risk factors to predict future year-2 outcomes within four 2-year cohorts. Predictors of interest included demographics, previous health care use, comorbidity, previously identified social needs, and neighborhood characteristics as reflected by the area deprivation index. The outcome variable was a binary indicator reflecting the likelihood of the presence of a patient with social needs. We applied a generalized estimating equation approach, adjusting for patient-level risk factors, the possible effect of geographically clustered data, and the effect of multiple visits for each patient.

RESULTS

The study population of 1,852,228 patients included middle-aged (mean age range 53.76-55.95 years), White (range 324,279/510,770, 63.49% to 290,688/488,666, 64.79%), and female (range 314,741/510,770, 61.62% to 278,488/448,666, 62.07%) patients from neighborhoods with high socioeconomic status (mean area deprivation index percentile range 28.76-30.31). Between 8.28% (37,137/448,666) and 11.55% (52,037/450,426) of patients across the study cohorts had at least 1 social need documented in their EHR, with safety issues and economic challenges (ie, financial resource strain, employment, and food insecurity) being the most common documented social needs (87,152/1,852,228, 4.71% and 58,242/1,852,228, 3.14% of overall patients, respectively). The model had an area under the curve of 0.702 (95% CI 0.699-0.705) in predicting prospective social needs in the overall study population. Previous social needs (odds ratio 3.285, 95% CI 3.237-3.335) and emergency department visits (odds ratio 1.659, 95% CI 1.634-1.684) were the strongest predictors of future social needs.

CONCLUSIONS

Our model provides an opportunity to make use of available EHR data to help identify patients with high social needs. Our proposed social risk score could help identify the subset of patients who would most benefit from further social needs screening and data collection to avoid potentially more burdensome primary data collection on all patients in a target population of interest.

Toward standardization, harmonization, and integration of social determinants of health data: A Texas Clinical and Translational Science Award institutions collaboration

Introduction

The focus on social determinants of health (SDOH) and their impact on health outcomes is evident in U.S. federal actions by Centers for Medicare & Medicaid Services and Office of National Coordinator for Health Information Technology. The disproportionate impact of COVID-19 on minorities and communities of color heightened awareness of health inequities and the need for more robust SDOH data collection. Four Clinical and Translational Science Award (CTSA) hubs comprising the Texas Regional CTSA Consortium (TRCC) undertook an inventory to understand what contextual-level SDOH datasets are offered centrally and which individual-level SDOH are collected in structured fields in each electronic health record (EHR) system potentially for all patients.

Methods

Hub teams identified American Community Survey (ACS) datasets available via their enterprise data warehouses for research. Each hub's EHR analyst team identified structured fields available in their EHR for SDOH using a collection instrument based on a 2021 PCORnet survey and conducted an SDOH field completion rate analysis.

Results

One hub offered ACS datasets centrally. All hubs collected eleven SDOH elements in structured EHR fields. Two collected Homeless and Veteran statuses. Completeness at four hubs was 80%-98%: Ethnicity, Race; < 10%: Education, Financial Strain, Food Insecurity, Housing Security/Stability, Interpersonal Violence, Social Isolation, Stress, Transportation.

Conclusion

Completeness levels for SDOH data in EHR at TRCC hubs varied and were low for most measures. Multiple system-level discussions may be necessary to increase standardized SDOH EHR-based data collection and harmonization to drive effective value-based care, health disparities research, translational interventions, and evidence-based policy.

Predicting Hospitalization among Medicaid Home- and Community-Based Services Users Using Machine Learning Methods

We compare multiple machine learning algorithms and develop models to predict future hospitalization among Home- and Community-Based Services (HCBS) Users. Furthermore, we calculate feature importance, the score of input variables based on their importance to predict the outcome, to identify the most relevant variables to predict hospitalization. We use the 2012 national Medicaid Analytic eXtract data and Medicare Provider Analysis and Review data. Predicting any hospitalization, Random Forest appears to be the most robust approach, though XGBoost achieved similar predictive performance. While the importance of features varies by algorithm, chronic conditions, previous hospitalizations, as well as use of services for ambulance, personal care, and durable medical equipment were generally found to be important predictors of hospitalization. Utilizing prediction models to identify those who are prone to hospitalization could be useful in developing early interventions to improve outcomes among HCBS users.

Toward informatics-enabled preparedness for natural hazards to minimize health impacts of climate change

Natural hazards (NHs) associated with climate change have been increasing in frequency and intensity. These acute events impact humans both directly and through their effects on social and environmental determinants of health. Rather than relying on a fully reactive incident response disposition, it is crucial to ramp up preparedness initiatives for worsening case scenarios. In this perspective, we review the landscape of NH effects for human health and explore the potential of health informatics to address associated challenges, specifically from a preparedness angle. We outline important components in a health informatics agenda for hazard preparedness involving hazard-disease associations, social determinants of health, and hazard forecasting models, and call for novel methods to integrate them toward projecting healthcare needs in the wake of a hazard. We describe potential gaps and barriers in implementing these components and propose some high-level ideas to address them.

Individual-Level and Neighborhood-Level Factors Associated with Longitudinal Changes in Cardiometabolic Measures in Participants of a Clinic-Based Care Coordination Program: A Secondary Data Analysis

Background: Identifying individual and neighborhood-level factors associated with worsening cardiometabolic risks despite clinic-based care coordination may help identify candidates for supplementary team-based care. Methods: Secondary data analysis of data from a two-year nurse-led care coordination program cohort of Medicare, Medicaid, dual-eligible adults, Leveraging Information Technology to Guide High Tech, High Touch Care (LIGHT2), from ten Midwestern primary care clinics in the U.S. Outcome Measures: Hemoglobin A1C, low-density-lipoprotein (LDL) cholesterol, and blood pressure. Multivariable generalized linear regression models assessed individual and neighborhood-level factors associated with changes in outcome measures from before to after completion of the LIGHT2 program. Results: 6378 participants had pre-and post-intervention levels reported for at least one outcome measure. In adjusted models, higher pre-intervention cardiometabolic measures were associated with worsening of all cardiometabolic measures. Women had worsening LDL-cholesterol compared with men. Women with pre-intervention HbA1c > 6.8% and systolic blood pressure > 131 mm of Hg had worse post-intervention HbA1c and systolic blood pressure compared with men. Adding individual’s neighborhood-level risks did not change effect sizes significantly. Conclusions: Increased cardiometabolic risks and gender were associated with worsening cardiometabolic outcomes. Understanding unresolved gender-specific needs and preferences of patients with increased cardiometabolic risks may aid in tailoring clinic-community-linked care planning.