Impact of VHA’s Primary Intensive Care Management Program on Dual System Use

Abstract

Research ObjectiveLike other health systems, the Veterans Health Administration (VHA) faces rising healthcare costs, which are disproportionately incurred by highly complex patients. VHA piloted the PACT Intensive Management (PIM) initiative to reduce acute care utilization and improve clinical outcomes among complex veterans through innovative intensive case management strategies in primary care. Prior work showed PIM increased VHA primary care and home visits but did not significantly affect inpatient stays and emergency department (ED) visits. However, this prior work did not consider non‐VHA utilization despite >80% of VHA enrollees having non‐VHA insurance including Medicare. Thus, the goal of this study was to examine the effect of PIM on veterans’ use of VHA and Medicare Fee‐for‐Service (FFS) inpatient and outpatient services.Study DesignPIM was a randomized, quality improvement trial of intensive primary care at five VHA facilities. Between July 2014 and December 2016, patients with a recent ED visit or hospitalization, and a risk score above the 90th percentile, as calculated by a previously validated algorithm, were randomized to receive intensive management or usual care. Intensive management consisted of interdisciplinary teams that performed comprehensive assessments, care coordination, medication management, and case management. Usual care consisted of primary care in VHA patient‐centered medical homes. Data were extracted from VHA’s Corporate Data Warehouse and Medicare claims. Primary outcomes were combined VHA and Medicare FFS utilization of inpatient and outpatient services. Inpatient utilization was measured as the number of hospitalizations in seven categories: total acute, medical, surgical, psychiatric, rehabilitation, hospice, and ambulatory care sensitive. Outpatient utilization was measured in seven categories: primary care, specialty care, mental health, surgical, diagnostic, ED, and home health. Prepost changes in 12‐month utilization were compared between PIM and usual care groups using Wilcoxon rank sum test.Population StudiedA total of 3489 complex VHA patients randomized to receive intensive management or usual care.Principal FindingsThe proportion of combined inpatient utilization through Medicare was <25% for all categories except for rehabilitation and hospice. The proportion of combined outpatient visits obtained through Medicare was <18% for all categories. Pre‐post changes in combined inpatient utilization were not significantly different between treatment and control groups. PIM did not affect combined specialty care, surgical, diagnostic, and ED visits. However, PIM increased combined 12‐month utilization of primary care, mental health, and home health services by 0.30, 1.47, and 0.55 visits, respectively, which were significantly different from control group (all P < 0.01). The inclusion of Medicare utilization slightly attenuated the magnitude of the difference of pre‐post changes between treatment and control groups. Subgroup analyses among the 875 veterans continuously enrolled in Medicare FFS produced similar results.ConclusionsVeterans at the highest risk for hospitalizations had little utilization of Medicare inpatient and outpatient services. Overall, PIM increased combined utilization in three of seven outpatient categories. However, PIM had no effect on inpatient and ED services, even after including Medicare utilization.Implications for Policy or PracticeDespite its intention, intensive management did not reduce acute care utilization, consistent with prior studies. Potential avenues for future research include improvement in program design to identify and target patients most likely to benefit from intensive management.Primary Funding SourceDepartment of Veterans Affairs.