Background:

Since 2013, SC’s Healthy Outcome Plan (HOP) has incentivized hospitals to partner to serve the high-risk, high-cost uninsured. HOP care plans address social determinants, behavioral health, and patient engagement measured through the Patient Activation Measure® (PAM). The objective of this study was to determine whether increased patient activation was associated with decreased emergency department (ED) visits.

Methods:

We examined ED data from 2013-2020 for a cohort of 4,699 participants ages 18-64 who had at least 18 months of HOP enrollment and multiple PAM administrations. Total PAM scores were categorized as Level 1 (lowest activation) through Level 4 (highest activation), and participants were grouped as increased level, no-change level, and decreased level based on their repeated PAM administration scores. The primary outcomes were ED visits and charges measured one year pre-intervention through three years post-intervention. Generalized Estimation Equations (GEEs) with Poisson distribution and robust standard errors estimation was used to control for confounding and mild dispersion.

Results:

Of the 4,699 participants, 1,307 (27.8 %) were increased level, 1,749 (37.2 %) were no-change level, and 1,643 (35.0%) were decreased level. Post-intervention, a higher rate of ED visits was observed for participants with decreased level compared to increased level (incident rate ratio [IRR] 1.05, 95 % CI, 1.07 to 1.09). The rate ratio of ED charges was also higher for decreased level (1.08 [1.01 to 1.16]).

Conclusion:

Care coordination that concurrently addresses patient self-efficacy and the social determinants of health is a beneficial strategy to reduce avoidable ED visits for the uninsured.

Background: Healthcare settings are adopting interventions to screen patients for social needs (e.g., food, housing) and refer to relevant community resources. However, the extent to which such interventions connect patients with needed resources is unclear. Methods: This systematic mixed studies review (a) assessed the extent to which patients participating in U.S. healthcare-based social needs screening and referral interventions connect with resources and (b) explored barriers and facilitators to resource connections. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We completed the search in PubMed and CINAHL for publications between October 2015 and December 2020. We used dual review to assess articles for exclusion or inclusion based on pre-determined criteria. We abstracted data within the following areas: study design, setting, intervention, and outcomes, and then assessed articles’ quality using the Mixed Methods Appraisal Tool (MMAT). We used thematic analysis to synthesize the data. Preliminary Results: The search identified 1,824 articles, 32 of which met the inclusion criteria. We excluded 1,708 articles based on title and abstract review and 84 during full text review. Thematic analysis revealed considerable heterogeneity across study designs, settings, interventions, and the extent to which patients accessed resources. Connection barriers included resource eligibility criteria, inadequacy of resources, and language barriers. Facilitators included timely follow-ups, tailored referrals, and trusting patient-healthcare personnel relationships. Conclusions: Ameliorating social needs through resource connections is critical for patient health. Our review identified patient-, practice-, and system-level factors that may enhance social needs interventions’ ability to connect patients to resources.

BACKGROUND

Standardized tracking of health-related social needs (HRSNs) is critical to improve the health of Medicare beneficiaries. One tool available to capture these data are Z-codes: a set (Z55-Z65) of ICD-10-CM codes for identifying hazardous social and environmental conditions that negatively affect health and health-related outcomes.

METHODS

2019 Medicare fee-for-service (FFS) claims and enrollment data were used to produce descriptives for Z-code utilization, overall and by claims, place of service, and provider types. Sociodemographic, clinical, and geographic characteristics statistics for continuously enrolled beneficiaries with Z-code claims were also reported.

RESULTS

Beneficiaries with Z-code claims represented 1.59% of all FFS beneficiaries. The most common Z-codes were Z59.0 (homelessness), Z63.4 (disappearance/ death of family), Z60.2 (problems living alone), Z59.3 (problems living in residential institution) and Z63.0 (relationship problems with spouse/partner). Half (49.6%) of Z-codes were billed on Part B Non-institutional claims, and family practice physicians submitted the largest share of Z-code claims (15.3%).

Relative to their shares of the overall FFS population, beneficiaries who were Black appeared overrepresented (24.8% vs. 8.8%) in Z59.0 claims, as were males (67.1% vs. 45.4%) and dually eligible and younger (<65 years) beneficiaries (66.0% and 67.6% vs. 14.3% and 14.2%). White (83.0% vs. 79.5%), female (69.1% vs. 54.6%), and older beneficiaries (≥85 years) (30.9% vs. 26.7%) appeared overrepresented in Z60.2 claims. Rural beneficiaries appeared overrepresented in Z59.3 claims (39.7% vs. 21.7%).

CONCLUSIONS

Z-code use remains low; standardized and consistent use of these codes is needed for better estimates of HRSNs burdens and tracking progress toward addressing them.

Background: Optional Medicaid expansion under the Affordable Care Act improved hospital finances by reducing bad debt and charity care for uninsured patients. But finances of hospitals in expansion states could also improve through increased Medicare Disproportionate Share Hospital (DSH) payments, which support care for low-income populations and increase as hospitals treat more Medicaid patients.

Methods: Using 2011-2017 national MedPAR data from hospitals in 45 states, 26 of which expanded as of 2014 plus 19 which had not expanded by 2017, we tested the impact of Medicaid expansion on Medicare DSH payments with a discharge-level difference-in-differences (DID) approach. We used generalized linear models with a log link, cluster-robust standard errors and state fixed effects, adjusting for state-level, patient, and hospital factors. We also examined late-expander and excluded early-expander states, year-specific changes in DSH, and uncompensated care payments introduced to address reduced DSH payments.

Results: Medicaid expansion was associated with a 22.5% (95% CI: 22.4%, 22.6%) increase in per-discharge DSH payments, which translates to an annual increase in DSH revenue of ~$200,000 for expansion-state hospitals, or $900 million for all expansion state hospitals during the three-year post-ACA period. Results were robust to the inclusion of states that had later or earlier expansion dates and when uncompensated care payments were included.

Conclusions: States’ decisions to not expand Medicaid resulted in millions of foregone DSH payments to hospitals, due to a quirk in the technical formula for DSH. Medicare should ensure that DSH subsidies are directed to hospitals with greatest patient need.