Background: VA's highest-utilizing patients generally have complicated health care needs-including complex and multiple chronic conditions, comorbid mental health conditions, and social stressors-that contribute to high rates of hospitalization, emergency services, and specialty care use. Inspired by emerging intensive primary care models for high-utilizers, VA Palo Alto obtained FY2012-FY2013 Office of Specialty Care Transformation (OSCT) funding to launch a novel program that augments the existing PACT model with intensive care delivered by a multidisciplinary team. The Intensive management PACT (ImPACT) intervention encompasses a number of evidence-based strategies, including a comprehensive intake process, coordination of specialty care, chronic condition case management, provision of social services, rapid response to deteriorations in health, and facilitation of transitions after high-acuity events. Objectives: Through this HSR&D pilot, we will evaluate ImPACT's feasibility, implementation, and effectiveness, and lay the groundwork for a follow-up multi-center randomized controlled trial (RCT). Methods: We will partner with the implementation team of VA Palo Alto's ImPACT clinic to conduct a Hybrid Type 1 evaluation of the program's feasibility, implementation, and effectiveness. For the first phase of ImPACT, VA Palo Alto identified PACT patients who were among the most costly 5% for the facility but were not enrolled in other intensive management programs. A random sample of 150 patients were selected for immediate enrollment in ImPACT, and 272 were assigned to delayed-enrollment and will continue to receive usual PACT care, thereby serving as a control group for our analyses. Our evaluation of ImPACT will use data from VA Palo Alto's Decision Support System (DSS) office, as well as data from surveys that are being administered by the clinic for an internal evaluation, to meet the following objectives: ? Aim 1: Evaluate the feasibility and implementation of the pilot ImPACT intervention. Using semi- structured interviews with ImPACT and PACT team members and leadership, we will evaluate the success of intervention delivery, including patient identification, recruitment, and retention;provision and uptake of planned services;and monitoring of patient participation and key outcomes. ? Aim 2: Obtain preliminary estimates of ImPACT's effect on utilization and costs of care, and refine analytic techniques for a subsequent multi-center study. We will use a difference-in- differences approach, wherein we compare changes in VA health care costs (total, as well as inpatient, outpatient, and fee-basis) and utilization (including hospitalizations, emergency department visits, and specialty care) among ImPACT patients and high-utilizing patients who are receiving usual PACT care. Through these analyses, we will obtain point estimates for effect sizes of key outcomes and refine our analytic methods for a subsequent multi-center RCT. ? Aim 3: Examine the association between ImPACT participation and patient-centered outcomes, and refine survey instruments for future evaluations. Using data from surveys administered in the ImPACT clinic, we will assess patient satisfaction with the ImPACT intervention and overall care, as well as changes in patient-reported outcomes, including health status, symptom burden, and function. We will also evaluate the survey instruments in order to eliminate redundancy, improve discriminatory capacity, and reduce burden to patients in subsequent studies. After synthesizing pilot findings, we will derive a recommended sample size and sampling strategy, refine primary and secondary outcomes, and develop study protocols for a subsequent multi-site RCT of ImPACT. Our pilot efforts will also yield evaluation tools that can be disseminated to others who are designing and studying models of care for complex and high-utilizing patients within VA.
Nearly half of VA's total health care expenditures are generated by 5% of its highest-utilizing patients. In recent years, a number of promising intensive primary care models have emerged outside VA that focus on health systems'highest utilizers. These models, if integrated successfully with PACT and tailored to meet the needs of VA patients, have the potential to contain utilization and enhance the health and care experience of some of VA's most complex and vulnerable patients. VA Palo Alto is in the process of implementing a pilot intervention that employs strategies developed outside VA to provide intensive management for its most costly PACT patients. Evaluations of this program's effectiveness and implementation will inform efforts to improve care delivery for this patient population nationwide.