Details

HPM student Donald Bourne won the Diversity, Equity and Inclusion Award, Doctoral Level for the project, “Changes in Medication Utilization and Adherence Associated with Homeless Adults’ Entry into Permanent Supportive Housing”.

Dean’s Day is an annual student research competition. Students present their research during multiple poster sessions while faculty members judge presentations for prizes and students evaluate posters for Grand Rounds credit.

Visit publichealth.pitt.edu/deansday to learn more about the competition and see a full list of winners.

Abstract

Research Objective
Permanent Supportive Housing (PSH) links affordable, non-time-limited housing with voluntary support services designed to address the needs of those who are unstably housed or chronically homeless. These individuals often suffer from multiple chronic health conditions that require prescription medications for proper management, and housing instability can complicate medication adherence. PSH programs have grown over the last decade, but evidence of their long-term impacts is lacking. Policymakers need such evidence in order to make informed decisions about the future of these services. Our objective is to evaluate the effects of PSH on medication utilization and adherence across the most common medication classes among Pennsylvania Medicaid enrollees (Antidepressants, Antipsychotics, Asthma, Diabetes).

Study Design
Retrospective cohort study using Medicaid enrollment and claims data from the Pennsylvania Medicaid program linked to Homeless Management Information Systems records and a difference-in-differences approach. Main measures included proportion with prescription fill, mean proportion of days covered (PDC), and percent adherent (PDC > 80%). PDC was calculated by the number of days in a period covered divided by the number of days in treatment period. We created sub-groups based on therapeutic class (i.e., NDC numbers) for medications and the presence of diagnosis codes (i.e., ICD-9-CM and ICD-10-CM codes) as additional inclusion criteria.

Population Studied
Pennsylvania Medicaid enrollees who entered PSH between January 1, 2011 and December 31, 2016, and a propensity-matched comparison cohort of enrollees experiencing housing instability who did not receive PSH.

Principal Findings
The analysis included 1375 participants in the PSH cohort and 5405 in the control cohort. The PSH cohort saw an 3.2% (95% CI 0.4% to 6.0%) increase in the proportion of PSH recipients who had filled any prescription, compared to the comparison cohort. Percent Adherent for antidepressant medication in the PSH cohort rose 9% (95% CI 1.4% to 16.5%), compared to the comparison cohort (Table 1). The PSH cohort also saw significant increases in proportion of filled prescriptions for antidepressants; mean PDC for antidepressants, antipsychotics, asthma medications; and percent adherent for antidepressants and asthma medications (Table 2). However, the increases in these categories were similar to those observed in the comparison cohort, and thus were not statistically significant in the difference-in-differences analyses.

Conclusions
PSH participation is associated with improved access to prescription drugs overall and improved adherence to antidepressant medications. These results can inform emerging state efforts to incorporate PSH services through Medicaid funds.

Implications for Policy or Practice
This study fills important gaps in understanding medication utilization among persons who are chronically homeless or at risk for homelessness and the impacts of PSH on this population. Our findings suggest that states looking to use waivers to provide housing services to Medicaid enrollees may see improved medication utilization and adherence among certain medication classes.

Table 1. Results for Primary Outcome

Therapeutic Class (n = PSH, Control)	PSH Cohort Pre vs. Post Difference % (95% CI)*	Control Cohort Pre vs. Post Difference % (95% CI)*	Difference in Differences % (95% CI)**	P-value1
Any Medication (n = 1375, 5405)	8.2% (5.9% - 10.6%)	5%  (3.5% - 6.5%)	3.2% (0.4% - 6.0%)	0.02
Antidepressants (n = 640, 1490)	6.3% (1.7% - 10.8%)	6.3% (3.3% - 9.3%)	-0.1% (-5.5% - 5.4%)	0.98
Antipsychotics (n = 170, 351)	1.2% (-7.7% - 10%)	3.4% (-2.8% - 9.6%)	-2.2% (-13.0% - 8.6%)	0.68
Asthma (n = 195, 485)	1.5% (-7.5% - 10.6%)	6.6% (0.8% - 12.4%)	-5.1% (-15.8% - 5.7%)	0.36
Diabetes (n = 138, 463)	6.5% (-5.1% - 18.1%)	6.7% (0.3% - 13.1%)	-0.2% (-13.4% - 13.1%)	0.98

*Pre-period is the reference; **The control cohort is the reference

¹P-value for PDC using paired sample t-test.

Abbreviations: PSH = permanent supportive housing; SD = standard deviation; PDC = percent of days covered

Table 2. Results for Secondary Outcomes

Therapeutic Class (n = PSH, Control)	PSH Cohort Pre vs. Post Difference % (95% CI)*	Control Cohort Pre vs. Post Difference % (95% CI)*	Difference in Differences % (95% CI)**	P-value1
Antidepressants (n = 359, 746)
PDC Mean (SD)	11% (8% - 14%)	7% (5% - 9%)	4% (-1% - 8%)	0.12
Percent Adherent (PDC > 80%)	13.9% (7.6% - 20.2%)	5% (0.8% - 9.2%)	9.0% (1.4% - 16.5%)	0.02
Antipsychotics (n = 97, 186)
PDC Mean (SD)	7% (1% – 13%)	5% (1% - 10%)	2% (8% - 11%)	0.74
Percent Adherent (PDC > 80%)	8.2% (-5.7% - 22.2%)	3.8% (-5.6% - 13.1%)	4.5% (-12.3% - 21.3%)	0.60
Asthma (n = 73, 167)
PDC Mean (SD)	10% (2% – 17%)	10% (6% – 14%)	0% (-10% – 10%)	0.94
Percent Adherent (PDC > 80%)	15.1% (4.4% - 25.7%)	8.4% (1.8% - 15%)	6.7% (-5.8% - 19.2%)	0.30
Diabetes (n = 59, 181)
PDC Mean (SD)	0% (-6% - 6%)	9% (4% – 14%)	-9% (-20% - 2%)	0.10
Percent Adherent (PDC > 80%)	1.7% (-13.5% - 16.9%)	9.9% (1.1% - 18.8%)	-8.2% (-25.8% - 9.3%)	0.36

*Pre-period is the reference; **The control cohort is the reference

¹P-value for PDC using paired sample t-test. P-value for Percent Adherent using Chi-square tests for categorical variables.

Abbreviations: PSH = permanent supportive housing; SD = standard deviation; PDC = percent of days covered

5/09/2022