Refining Intensive Care Management Programs: Can Providing the Neediest Patients the Right Care in the Right Setting Improve Health and Reduce Expenditures?

Am J Manag Care. 2022;28(9):425-426. https://doi.org/10.37765/ajmc.2022.89217

Advocates often assume that intensive care management programs—including services such as integrated behavioral health care, social supports coordination, and home-based medical visits—delivered to very-high-need individuals will improve health and reduce health care spending. However, most rigorous studies examining the clinical and fiscal impact of these interventions, most notably the randomized Camden Coalition trial,¹ have found that they failed to achieve these ambitious dual goals of “save lives and save dollars.” Similarly, the recently published evaluation of the California Whole Person Care case management program reported reductions in utilization but an overall increase in spending.² In this issue of The American Journal of Managed Care^®, the study by Rowe and colleagues is a positive outlier and provides cautious optimism that desired clinical and fiscal outcomes may both be achieved by a carefully designed and implemented program.³

Compared with usual care, intensive care management programs proactively aim to meet patient needs through a collection of outreach, coordination, and other interventions. For care management programs to be sustainable, however, they need to demonstrate a return on investment, which in many circumstances requires a reduction in medical expenditures that offsets the program costs. Upon failing to meet this key financial metric, internally developed programs will often be discontinued; vendor-based programs are typically not purchased or the contracts are not renewed. No margin, no mission.

Health system leaders who must make difficult resource allocation decisions are often circumspect of perceived results from uncontrolled studies or modeling studies. Moreover, skeptics frequently intimate that any positive findings produced by less rigorous evaluations may result without the care management intervention (ie, regression to the mean). Rowe and colleagues highlight this concern in their introduction.

Interestingly, this randomized controlled trial reported the opposite of regression to the mean. The control group, which received “usual care,” showed a 30% rise in cost over the 12-month observation period. Meanwhile, the spending by intervention groups stayed essentially unchanged over the same period (as shown in Figure 2 on page 433). This calls into question the notion that costs decline spontaneously without intervention in high-risk Medicaid patients.

Why did spending by the control group in this study increase when in other studies it tends to regress over time? There are at least 2 possible explanations. First, unlike many studies of high-risk individuals, the intervention group was not based on a recent sentinel health event such as a hospital admission or emergency department (ED) visit. Instead, individuals in the intervention group were selected based on prolonged prior history of illness and health care utilization. Hence, these patients were not in the middle of a clinical episode but instead were established high utilizers. Second, the authors did not solely look at prior utilization patterns, but further refined the study cohort based on clinical judgment for high medical, behavioral, or social needs. This 2-tiered screening method identified the patients with the highest needs—representing not the typically cited top 5%, but approximately the top 0.5% of utilizers. These 2 factors are important to keep in mind when designing programs, evaluating future studies, and attempting to assess magnitude of impact.

In addition to the patient selection process, were there specific aspects of type or number of services provided by this program that led to these somewhat surprising outcomes? Care management programs offer an assortment of interventions that vary depending on the individual. This care management program was rich in several features not typically incorporated, including (1) integrated teams with low patient census that allowed intensive and personalized care, (2) integrated medical and social services, and (3) use of prehospital systems to avoid the need to admit patients who may be starting to decompensate. The ED avoidance strategy used by this program included personnel to address acute problems in the field without transporting patients to the ED. This likely reduced “reflex” admissions by providing a patient-specific care plan. In addition, the psychiatric crisis stabilization units were available at all hours to address acute behavioral health needs without a hospital admission.

Based on this single study, we can be cautiously optimistic about the important (and welcome) finding that a well-designed care management program delivering personalized care to carefully selected, high-risk Medicaid beneficiaries can result in substantially better clinical and financial outcomes. Given the severe unmet needs of the most vulnerable members of our communities, geographic expansion and rigorous evaluations of comparable, highly personalized interventions that provide the right care, provided by the right clinician, in the right setting, are warranted.

Author Affiliations: MedZed (NAS), San Francisco, CA; Department of Internal Medicine, University of Michigan School of Medicine (AMF), Ann Arbor, MI.

Source of Funding: None.

Author Disclosures: Dr Solomon is cofounder, board member, and chief medical officer of MedZed, a home care practice for high-risk patients like those seen in the referenced study. Dr Fendrick reports consulting fees from AbbVie, Bayer, Centivo, Covered California, Emblem Health, Exact Sciences, GRAIL, Harvard University, Health & Wellness Innovations, Health at Scale Technologies, HealthCorum, Hygieia, MedZed, Merck, Mother Goose Health, Phathom Pharmaceuticals, Sempre Health, Silverfern Health, State of Minnesota, Teledoc Health, US Department of Defense, Virginia Center for Health Innovation, Wellth, Wildflower Health, Yale–New Haven Health System, and Zansors; research support from the Agency for Healthcare Research and Quality, Arnold Ventures, Boehringer Ingelheim, Gary and Mary West Health Policy Center, National Pharmaceutical Council, Patient-Centered Outcomes Research Institute, PhRMA, Robert Wood Johnson Foundation, and State of Michigan/CMS; and serving as coeditor in chief of The American Journal of Managed Care^®, member of the Medicare Evidence Development & Coverage Advisory Committee, and partner of V-BID Health, LLC.

Authorship Information: Drafting of the manuscript (NAS, AMF); critical revision of the manuscript for important intellectual content (NAS, AMF); and administrative, technical, or logistic support (NAS, AMF).

Address Correspondence to: A. Mark Fendrick, MD, University of Michigan, 2800 Plymouth Rd, Bldg 16, Floor 4, 016-400S-25, Ann Arbor, MI 48109-2800. Email: amfen@med.umich.edu.

REFERENCES

1. Finkelstein A, Zhou A, Taubman S, Doyle J. Health care hotspotting—a randomized, controlled trial. N Engl J Med. 2020;382(2):152-162. doi:10.1056/NEJMsa1906848

2. Brown DM, Hernandez EA, Levin S, et al. Effect of social needs case management on hospital use among adult Medicaid beneficiaries: a randomized study. Ann Intern Med. 2022;175(8):1109-1117. doi:10.7326/M22-0074

3. Rowe JS, Gulla J, Vienneau M, et al. Intensive care management of a complex Medicaid population: a randomized evaluation. Am J Manag Care. 2022;28(9):430-435. doi:10.37765/ajmc.2022.89219