Dataset Information

Social Determinants of Health Improve Predictive Accuracy of Clinical Risk Models for Cardiovascular Hospitalization, Annual Cost, and Death.

ABSTRACT:

Background

Risk models in the private insurance setting may systematically underpredict in the socially disadvantaged. In this study, we sought to determine whether US minority Medicare beneficiaries had disproportionately low costs compared with their clinical outcomes and whether adding social determinants of health (SDOH) into risk prediction models improves prediction accuracy.

Methods and results

Retrospective observational cohort study of 2016 to 2017 Medicare Current Beneficiary Survey data (n=3614) linked to Medicare fee-for-service claims. Logistic and linear regressions were used to determine the relationship between race/ethnicity and annual costs of care, all-cause hospitalization, cardiovascular hospitalization, and death. We calculated the observed-to-expected (O:E) ratios for all outcomes under 4 risk models: (1) age+sex, (2) model 1+clinical comorbidity adjustment, (3) model 2+SDOH, and (4) SDOH alone. Our sample was 44% male and 11% black or Hispanic. Among minorities, adverse clinical outcomes were inversely related to cost. After multivariable adjustment, blacks/Hispanics had higher rates of cardiovascular hospitalization (incidence rate ratio, 1.78; P=0.012) but similar annual costs ($-336, P=0.77) compared with whites. Among whites, models 1 to 4 all showed similar O:E ratios, suggesting high accuracy in risk prediction using current models. Among minorities, adjustment for age, sex, and comorbidities underpredicted all-cause hospitalization by 20% (O:E, 1.20) and cardiovascular hospitalization by 70% (O:E, 1.70) and overpredicted death by 21% (O:E, 0.79); adding SDOH brought O:E near 1 for all outcomes. Among both groups, the SDOH risk model alone performed with equal or superior accuracy to the model based on clinical comorbidities.

Conclusions

A paradoxical relationship was observed between clinical outcomes and costs among racial and ethnic minorities. Because of systematic differences in access to care, cost may not be an appropriate surrogate for predicting clinical risk among vulnerable populations. Adjustment for SDOH improves the accuracy of risk models among racial and ethnic minorities and could guide use of prevention strategies.

SUBMITTER: Hammond G

PROVIDER: S-EPMC7299736 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background Although there is research on the impact of social determinants of health (SDOHs) on cardiovascular health, most existing evidence is based on individual SDOH components. We evaluated the impact of cumulative SDOH burden on cardiovascular risk factors, subclinical atherosclerosis, and incident cardiovascular disease events. Methods and Results We included 6479 participants from the MESA (Multi-Ethnic Study of Atherosclerosis). A weighted aggregate SDOH score representing the cumulative number of unfavorable SDOHs, identified from 14 components across 5 domains (economic stability, neighborhood and physical environment, community and social context, education, and health care system access) was calculated and divided into quartiles (quartile 4 being the least favorable). The impact of cumulative SDOH burden on cardiovascular risk factors (hypertension, diabetes, dyslipidemia, smoking, and obesity), systemic inflammation, subclinical atherosclerosis, and incident cardiovascular disease was evaluated. Increasing social disadvantage was associated with increased odds of all cardiovascular risk factors except dyslipidemia. Smoking was the risk factor most strongly associated with worse SDOH (odds ratio [OR], 2.67 for quartile 4 versus quartile 1 [95% CI, 2.13-3.34]). Participants within SDOH quartile 4 had 33% higher odds of increased high-sensitivity C-reactive protein (OR, 1.33 [95% CI, 1.11-1.60]) and 31% higher risk of all cardiovascular disease (hazard ratio, 1.31 [95% CI, 1.03-1.67]), yet no greater burden of subclinical atherosclerosis (OR, 1.01 [95% CI, 0.79-1.29]), when compared with those in quartile 1. Conclusions Increasing social disadvantage was associated with more prevalent cardiovascular risk factors, inflammation, and incident cardiovascular disease. These findings call for better identification of SDOHs in clinical practice and stronger measures to mitigate the higher SDOH burden among the socially disadvantaged to improve cardiovascular outcomes.

Project description:ObjectivesEarly hospital readmissions or deaths are key healthcare quality measures in pay-for-performance programs. Predictive models could identify patients at higher risk of readmission or death and target interventions. However, existing models usually do not incorporate social determinants of health (SDH) information, although this information is of great importance to address health disparities related to social risk factors. The objective of this study is to examine the impact of social determinants of health on predictive models for potentially avoidable 30-day readmission.MethodsWe extracted electronic health record data for 19,941 hospital admissions between January 2015 and November 2017 at an academic medical center in New York City. We applied the Simplified HOSPITAL score model to predict potentially avoidable 30-day readmission or death and examined if incorporating individual- and community-level SDH could improve the prediction using cross-validation. We calculated the C-statistic for discrimination, Brier score for accuracy, and Hosmer-Lemeshow test for calibration for each model using logistic regression. Analysis was conducted for all patients and three subgroups that may be disproportionately affected by social risk factors, namely Medicaid patients, patients who are 65 or older, and obese patients.ResultsThe Simplified HOSPITAL score model achieved similar performance in our sample compared to previous studies. Adding SDH did not improve the prediction among all patients. However, adding individual- and community-level SDH at the US census tract level significantly improved the prediction for all three subgroups. Specifically, C-statistics improved from 0.70 to 0.73 for Medicaid patients, from 0.66 to 0.68 for patients 65 or older, and from 0.70 to 0.73 for obese patients.ConclusionsPatients from certain subgroups may be more likely to be affected by social risk factors. Incorporating SDH into predictive models may be helpful to identify these patients and reduce health disparities associated with vulnerable social conditions.

Project description:ImportanceScreening for breast, cervical, and colorectal cancers in the United States has remained below the Healthy People 2020 goals, with evidence indicating that persistent screening disparities still exist. The US Department of Health and Human Services has emphasized cross-sectoral collaboration in aligning social determinants of health with public health and medical services. Examining the economics of intervening through these novel methods in the realm of cancer screening can inform program planners, health care providers, implementers, and policy makers.ObjectiveTo conduct a systematic review of economic evaluations of interventions leveraging social determinants of health to improve screening for breast, cervical, and colorectal cancer to guide implementation.Evidence reviewA systematic literature search for economic evidence was performed in MEDLINE, Embase, PsycINFO, Cochrane Library, Global Health, Scopus, Academic Search Complete, Business Source Complete, EconLit, CINAHL (Cumulative Index to Nursing and Allied Health Literature), ERIC (Education Resources Information Center), and Sociological Abstracts from January 1, 2004, to November 25, 2019. Included studies intervened on social determinants of health to improve breast, cervical, and colorectal cancer screening in the United States and reported intervention cost, incremental cost per additional person screened, and/or incremental cost per quality-adjusted life-year (QALY). Risk of bias was assessed along with qualitative assessment of quality to ensure complete reporting of economic measures, data sources, and analytic methods. In addition, included studies with modeled outcomes had to define structural elements and sources for input parameters, distinguish between programmatic and literature-derived data, and assess uncertainty.FindingsThirty unique articles with 94 706 real and 4.21 million simulated participants satisfied our inclusion criteria and were included in the analysis. The median intervention cost per participant was $123.87 (interquartile interval [IQI], $24.44-$313.19; 34 estimates). The median incremental cost per additional person screened was $250.37 (IQI, $44.67-$609.38; 17 estimates). Studies that modeled final economic outcomes had a median incremental cost per person of $122.96 (IQI, $46.96-$124.80; 5 estimates), a median incremental screening rate of 15% (IQI, 14%-20%; 5 estimates), and a median incremental QALY per person of 0.04 years (IQI, 0.006-0.06 year; 5 estimates). The median incremental cost per QALY gained of $3120.00 (IQI, $782.59-$33 600.00; 5 estimates) was lower than $50 000, an established, conservative threshold of cost-effectiveness.Conclusions and relevanceInterventions focused on social determinants of health to improve breast, cervical, and colorectal cancer screening appear to be cost-effective for underserved, vulnerable populations in the United States. The increased screening rates were associated with earlier diagnosis and treatment and in improved health outcomes with significant gains in QALYs. These findings represent the latest economic evidence to guide implementation of these interventions, which serve the dual purpose of enhancing health equity and economic efficiency.

Project description:Background & aimsWe estimated the annual burden and costs of hospitalization in patients with chronic gastrointestinal and liver diseases, and identified characteristics of high-need, high-cost patients, in a nationally representative sample.MethodsUsing Nationwide Readmissions Database 2013, we identified patients with at least 1 hospitalization between January and June 2013, and a diagnosis of inflammatory bowel diseases (IBDs), chronic liver diseases (CLDs), functional gastrointestinal disorders (FGIDs), gastrointestinal hemorrhage, or pancreatic diseases, with 6 months or more of follow up. We calculated days spent in hospital/month and estimated costs of the entire cohort, and identified characteristics of high-need, high-cost patients (top decile of days spent in hospital/month).ResultsPatients with IBD (n = 47,402), CLDs (n = 376,810), FGIDs (n = 351,583), gastrointestinal hemorrhage (n = 190,881), or pancreatic diseases (n = 98,432), hospitalized at least once, spent a median of 6 to 7 days (interquartile range, 3-14 d) in the hospital each year (total for all diseases). Compared to patients in the lowest decile (median, 0.13-0.14 d/mo spent in the hospital), patients in the highest decile spent a median 3.7-4.1 days/month in hospital (total for all diseases), with hospitalization costs ranging from $7502/month to $8925/month and 1 hospitalization every 2 months. Gastrointestinal diseases, infections, and cardiopulmonary causes were leading reasons for hospitalization of these patients. Based on multivariate logistic regression, high-need, high-cost patients were more likely to have Medicare/Medicaid insurance, lower income status, index hospitalization in a large rural hospital, high comorbidity burden, obesity, and infection-related hospitalization.ConclusionsIn a nationwide database analysis of patients with IBD, CLD, FGID, gastrointestinal hemorrhage, or pancreatic diseases hospitalized at least once, we found that a small fraction of high-need, high-cost patients contribute disproportionately to hospitalization costs. Population health management directed toward these patients would facilitate high-value care.