Implementation and 12-Month Health Service Utilization and Cost Outcomes from a Managed Care Health Plan's Permanent Supportive Housing Program

Addressing a Social Determinant of Health Through Permanent Supportive Housing

Permanent supportive housing (PSH) is an intervention that combines a long-term housing subsidy with supportive services and has been shown to increase housing stability among the formerly homeless. Health plans, hospitals, and large health systems have expressed interest in this intervention to address one of the key social determinants of health, stable housing, but little is known about the associated program costs and potential impacts on health care utilization among insured populations. We sought to understand the program's implementation, the program's impact on member health care utilization, and associated costs. We also examined the PSH programmatic expenditures to determine the overall costs and benefits of the intervention for the health plan. The PSH program is operated by a not-for-profit Medicaid and Medicare managed care plan in Southern California.

Approach

Using program administrative and health care service data (including health care claims and encounters) provided by the health plan, we conducted a regression analysis with propensity score weighting to examine health care utilization and associated costs for 162 health plan members 12 months prior to and 12 months following enrollment into a PSH program relative to a comparable cohort of plan members (N = 356) who were not enrolled in the PSH program. We also differentiated the program's effect during the transitional period—that is, after program enrollment but prior to housing placement—from its effects during the period after members are housed. In addition, we present participant flow through the key program milestones (e.g., referral, enrollment, housing placement, program exit) and describe health care utilization and associated costs for members who exited the program. Finally, we report the PSH programmatic expenditures relative to the changes in health care costs to provide an overall picture of the intervention's benefits and costs to the health plan.

Key Findings

• Similar to the findings presented in the baseline (McBain et al., 2020) and six-month outcome (Hunter et al., 2021) studies, our updated findings indicate that members who were enrolled in the PSH program used a significant amount of health services in the 12-month period prior to program entry: roughly $70,000 per person. This demonstrates that the health plan was successful in identifying and enrolling high health care utilizers into its PSH program.
• PSH enrollees’ average health service utilization increased in the 12 months following program enrollment, while associated costs decreased from $70,136 to $48,718—a total reduction of $21,418 (about 31 percent). Members in the comparison group also observed a reduction in utilization and costs: from $65,700 to $40,427—a total reduction of $25,273 (about 39 percent).
• Overall, health care costs declined further among program enrollees once they were placed into housing, relative to the transition period (i.e., after enrollment but before PSH placement): from an average of $5,542 per member per month (PMPM) in the transitional period to $2,919 PMPM once in PSH placement, an amount lower than the comparable value in the comparison group. Enrollees who exited the PSH program prior to 12 months had higher health care utilization and associated costs in the pre-enrollment and post-enrollment periods relative to enrollees who remained in stable housing through the PSH program.
• The PSH program was associated with increased health care engagement and fewer high-cost health care events. Members had an increased probability of having a health care service when enrolled in the PSH program relative to not being enrolled. Among program enrollees with at least one health care service, the total health care service costs while housed were significantly lower ($2,201 PMPM) relative to the transitional period.
• The estimated PSH program costs were $30,540 per enrollee over the 12-month period, or $2,545 PMPM.
• When accounting for the PSH program costs and health care service utilization among enrollees, the PSH program has yet to achieve any net cost offset from program participation.

In sum, our findings are similar to those of previous studies, which have shown that supportive housing interventions are associated with decreases in inpatient and emergency care. In this study, enrollees were more likely to remain engaged with health care providers and used more primary and home health care after program enrollment, relative to the comparison group. Health care among program enrollees may have shifted to early intervention or more regular disease management, as observed by the increase in visits for many of the most common chronic health conditions (e.g., diabetes and hypertension).

Study Implications and Recommendations

• Our study is one of the first of its kind to describe PSH program implementation, including documentation of the timing between the different program milestones and associated health care utilization and costs. These findings help to elucidate the care and costs associated with identifying and enrolling high health care utilizers into a PSH program. More specifically, individuals experiencing homelessness who are discharged from emergent or inpatient settings may require temporary housing with a variety of medical and nonmedical supports before PSH placement—resulting in a continuation of elevated costs. Payers of such programs should be aware of these initial cost drivers.
• Although we observed patterns in health care utilization and associated costs consistent with other PSH studies—including that health care costs among those engaged with health care providers were reduced once individuals were placed in housing—reductions in costs did not entirely offset the PSH programmatic expenses. Also, members were engaging more consistently with health care providers following program enrollment and obtaining less intensive and expensive forms of care. More research is needed to determine whether these changes in health care utilization are leading to better chronic condition management and quality of life.
• We observed that enrollees who exited the program before reaching 12 months of enrollment had higher utilization patterns and associated costs both before and after enrollment, relative to enrollees who were stably housed at 12 months post-enrollment. These findings suggest that the program might not have been appropriate, or might have been insufficient, for some of the highest-cost health care utilizers. Increased targeted recruitment and identification of successful program candidates could result in more-consistent findings and larger decreases in health care costs among enrollees. For higher-risk candidates, a higher level of care or an increased direct and targeted intervention may be necessary.
• Health care utilization and associated costs were reduced among members once they were placed in PSH. These findings might help payers better prepare and plan for PSH program implementation in terms of initial and longer-term impacts on health care utilization and associated costs.