Tailoring Complex Care to Patients’ Needs: Myths, Realities, and Best Next Steps

,

The authors of this editorial highlight some of the myths surrounding complex care management, identify areas where research could be most informative, and recommend best next steps in developing effective and efficient complex care management programs.

Am J Manag Care. 2022;28(2):In Press

_____

The term “complex care management” encompasses a wide range of health care services with some common themes. Individuals are enrolled for intense follow-up and care management by a provider or a team of providers over varying periods of time. The roots of complex care management trace to the 19th century, and increasingly sophisticated models have been proposed for decades. Investment in and growth of complex care has exploded in the United States since the early 2000s, driven mainly by the aim of reducing health care costs but also motivated by the goals of increasing the quality of care, enhancing equity, and ultimately improving patient outcomes.

Arguably, complex care management is now at an inflection point, as the results of recent research have shown that the prospects of these high-value interventions leading to substantially lower health care costs are dimmer than previously believed.1 However, new opportunities to identify and deliver the benefits of complex care are emerging as the primary focus of complex care undergoes a transition from cost reduction to better patient-centered outcomes. This evolution, however, is hindered by pervasive myths that muddy the discussion and obscure the way forward. Here we highlight some of these myths, identify areas where research could be most informative, and recommend best next steps in developing effective and efficient complex care management programs.

Myths and Realities

As a road map, we offer a 5-part statement that, while deliberately evocative, illuminates core presumptions and elements of the current discussion around complex care management:

“High utilization is a problem…caused by high utilizers…who can be classified into groups using medical records…to which complex care management is the solution to reduce unnecessary spending…and improve patient outcomes.”

High utilization is a problem…

“Five percent of patients account for 50% of costs” is the standard introduction to discussions of complex care programs, as if it were prima facie evidence of their need. Consider a parallel statement: “Less than 1% of homeowners account for 100% of the costs of house fires.”2 In a well-functioning insurance system, including health care, a high disproportion between payers and users might indicate success. Typically, this introductory statement is offered to provide self-evident justification that expenditures on this small group of high utilizers could be reduced and a “better” distribution of spending achieved if we “fix the patient” through better management of chronic medical and/or behavioral conditions. One might wonder how patients and families receiving an organ transplant or with high-cost medical conditions such as cancer, multiple sclerosis, and rare, treatable genetic disorders might consider the desire to more evenly allocate health care spending across a population. Regardless of the shape of the medical expenditure distribution curve, a 5% tail will always exist. In an ideally functioning health care system, the allocation of health resources across the population is necessarily highly disproportionate.

Statements of disproportion between health care users and medical spending add nothing to the substantive discussion while fueling misconceptions about the causes of avoidable health care costs. Of more concern, advocating for “smoother” spending distribution likely fosters a victim-blaming mentality that suggests that the problem rests exclusively with high-spending individuals. This argument holds the delivery system blameless and becomes particularly insidious when the patient subgroups are people whose socioeconomic circumstances and access to health care are the results of centuries of systematic racism and oppression.3

…caused by high utilizers…

Complex care management programs are patient-focused care models that provide support to individuals to navigate an increasingly complex and fragmented delivery system, promote self-management capacity, and address barriers to care. Common case scenarios include creating stable in-home support for frail seniors; securing provider-provider communication, funding, and housing transitions for persons with chronic mental illness; establishing a stable primary outpatient provider for patients with frequent hospitalizations from complications of medical conditions; and securing postacute care for persons experiencing homelessness. Although vitally important, concerns arise that building ever-more-refined complex care systems can foster complacency about the system-based causes of care discontinuity and avoidable health care costs. For example, in our health system, after nearly 15 years of program experience, chasms remain between inpatient and outpatient settings and between primary and specialty care. Moreover, rapid access to primary care to avoid or follow up after emergency department visits or hospitalization remains problematic.

Complex care management programs can and should serve as diagnostic tools for health systems, revealing gaps in care, identifying specific patient populations negatively affected, and providing insight into root causes of health system dysfunction leading to fragmented, discontinuous care and redundant health care costs. We envision health systems that routinely integrate complex care management as a core informant and guide to enhance continuity and care coordination across separate clinical units and distinct revenue centers.

…who can be classified into groups using medical records…

Experience in the past 20 years has made it abundantly clear that patients who are likely to benefit from intensive care management are heterogeneous and that complex care programs must be tailored to specific individuals to function efficiently and effectively. The range of skills and resources to address the broad needs of this diverse patient population is unlikely to be incorporated into an “off-the-shelf” program. Thus, grouping patients—and therefore the structure, goals, and outcomes of different types of complex care management programs—is essential.

Patient classification schemes to guide enrollment criteria and program design have been proposed.4-7 However, nearly all are based primarily on patients’ past health care utilization derived from medical records. Importantly, these classification systems often lack information on behavioral health and social factors that influence care patterns and determine the most appropriate type of complex care management interventions.8,9

Two articles in this issue of The American Journal of Managed Care® (AJMC®) add to our knowledge of using patient classification to tailor complex care programs. In the first, Pourat and colleagues use latent class analysis to examine the intersection of clinical characteristics and utilization among adult patients aged 18 to 64 years in federally supported health centers.10 By combining data from patient surveys and medical records, the authors demonstrate that individuals with and without behavioral health conditions are in separable classes and that similar patient types based on clinical criteria can be identified among low utilizers, high utilizers, and superutilizers. The data did not, however, include many potentially important factors such as precarious housing, food, or transportation that may render the classification more useful for care management purposes. The second article, by Arnold and colleagues, provides a review of the literature on segmentation schemes for high-risk patient populations.11 Among the key conclusions is that studies were largely based on medical records and lacked data on patients’ function or on social conditions that critically inform complex care management.

It is important to note, however, that although improved classification schemes that include behavioral and social factors are needed, data-driven eligibility criteria are insufficient to determine whether a patient is suited for complex care. Certain risk scoring systems will not include conditions or circumstances that require complex care management. In contrast, providing an untargeted, standard set of services regardless of need to all patients deemed eligible will increase costs of care. At a deeper level, in many cases the interactions among a varying set of physical and behavioral health conditions and nonmedical circumstances that differ across individuals limit the utility of most multivariate models. To illustrate, imagine two 60-year-old patients with poorly controlled diabetes and hypertension, both living on limited income in stable housing with a spouse and adult child nearby. Only an intake interview will reveal that one patient’s spouse has dementia and is slowly declining, requiring a substantial amount of caregiving from the patient, whereas the other patient has a healthy, supportive spouse but is nearing the end of their life savings. Individualized care management plans may differ substantially for patients who may seem similar based on their personal medical record.

…to which complex care management is the solution to reduce unnecessary spending…

Since the early 2000s, findings from nonrandomized (pre-post comparisons, historical control, or concurrent cohort) trials have been routinely cited to claim that complex care management reduces health care costs.12,13 Randomized trials, however, have produced mixed results at best.1,14,15 As reported by many commentaries,3,16 the results of nonrandomized trials are heavily influenced by regression to the mean,17,18 as the circumstances and conditions driving high utilization for most patients subside over a few weeks to months. Other contributing factors are that many high-utilizing patients are already receiving high levels of care, limiting the incremental effects of complex care management on utilization, and that for some patients the interactions among behavioral, social, and health system factors resist short-term fixes.

…and improve patient outcomes.

Increasingly, and appropriately, attention is turning from the cost impact to the clinical benefits of complex care management. Among the emerging lessons, well known to complex care providers but seldom the subject of study, is that high-value complex care is based on the establishment of relationships and continuity. One of us vividly remembers a “fishbowl” exercise in which a group of persons experiencing homelessness and their care managers were speaking with providers not involved in their care. When the patients were asked to describe the value of the care management program, none mentioned doctor visits, control of chronic medical conditions, or self-management capacity. Instead, each patient revealed that the value was the care manager, who was available for and interested in addressing any problem that might arise. It is likely that such a trusting continuity relationship with a skilled care manager is essential to other positive effects in patients’ health behavior, health care utilization, and well-being. Thus, elements of trust and continuity are measurable factors that should be included in future studies of complex care management.

Recommendations to Guide the Next Generation of Complex Care Management Programs

First, reducing avoidable health costs will be likely driven more by health system and finance reforms and less so by the direct impact of complex care on utilization. It is critical that future complex care management programs are aligned with, and promote, these systemic reforms. For example, complex care participation could demonstrate increases in the use of clinical services that serve as health system metrics that are the basis for quality-driven alternative payment models.

Second, to be useful, patient classification models informing complex care must include more information on behavioral health, mental health and substance use disorders, functional status, and social influences of health such as housing, income, transportation, food insecurity, and social support. Although methods to systematically include behavioral and social factors to identify patients most likely to benefit from complex care management have been proposed,17 more work is needed.8

Third, even with improved data-driven classification systems, clinical assessment of individual patients and their support systems will remain essential. Consequently, patient classification schemes and risk scores should be considered a starting point, not a final determinant, of eligibility. Referral systems should be built with flexibility at early stages of patient identification and assessment to allow early review of patients’ individual situations to drive matches between patients and the best-fitting programs, not just whether a patient is appropriate for a given program.

Finally, by understanding complex care first as a patient-centered intervention rather than a cost-saving strategy, we can design complex care programs for purpose—such as reducing fragmented care and promoting chronic disease self-management through relationship-based continuity relationships—rather than as solutions to reduce expenditures. Research should focus on identifying the components of complex care management that are most useful for particular patients. Hadeed and Fendrick recently advocated in AJMC® that the effects of continuity and relationships should be more carefully examined and should eventually become a measure of program success.19 Although there has been much attention focused on overutilizers, substantial gaps in care exist in underserved populations. Therefore, health systems might also consider looking for underutilizers—community residents whose access is limited and/or care so fragmented that they are invisible to programs focusing on utilization-based enrollment criteria.

Conclusions

Current patient classification systems to promote tailored complex care are based largely on medical information and are inadequate, in and of themselves, to match patients to specific program types that are likely to be effective and efficient. We suggest that, based on the current literature and our experience, 5 basic types of complex care programs and related populations have emerged (Text Box).

Although complex care management has evolved enormously in recent decades, the legacy of its origins in attempts to control overall health care costs hinders our view of its potential to benefit patients and health care systems. Priorities should include research into ways to better tailor complex care and to more effectively integrate complex care programs into health care systems, with goals of reducing fragmentation, enhancing efficiency, and promoting the health of individual patients and surrounding communities.

Author Affiliations: Department of Internal Medicine, University of Michigan School of Medicine (BCW, AMF), Ann Arbor, MI.

Source of Funding: None.

Author Disclosures: Dr Williams is employed as the medical director of the Michigan Medicine Complex Care Management Program. Dr Fendrick has been a consultant for AbbVie, Amgen, Centivo, Community Oncology Alliance, Covered California, EmblemHealth, Exact Sciences, Freedman Health, GRAIL, Harvard University, Health & Wellness Innovations, Health at Scale Technologies, MedZed, Merck, Montana Health Cooperative, Penguin Pay, Risalto, Sempre Health, State of Minnesota, US Department of Defense, Virginia Center for Health Innovation, Wellth, Yale–New Haven Health System, and Zansors; has performed research for 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 holds outside positions as co-editor-in-chief of The American Journal of Managed Care®, member of the Medicare Evidence Development & Coverage Advisory Committee, and partner in V-BID Health, LLC.

Authorship Information: Concept and design (BCW, AMF); drafting of the manuscript (BCW, AMF); and critical revision of the manuscript for important intellectual content (BCW, AMF).

Address Correspondence to: Brent C. Williams, MD, MPH, University of Michigan, 2800 Plymouth Rd, North Campus Research Complex, Bldg 16, Room 447C, Ann Arbor, MI 48109. Email: bwilliam@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. Ahrens M, Evarts B. Fire loss in the United States during 2019. National Fire Protection Association. September 2020. Accessed June 5, 2021. https://www.nfpa.org/News-and-Research/Data-research-and-tools/US-Fire-Problem/Fire-loss-in-the-United-States

3. Lantz PM. “Super-utilizer” interventions: what they reveal about evaluation research, wishful thinking, and health equity. Milbank Q. 2020;98(1):31-34. doi:10.1111/1468-0009.12449

4. Rinehart DJ, Oronce C, Durfee MJ, et al. Identifying subgroups of adult superutilizers in an urban safety-net system using latent class analysis: implications for clinical practice. Med Care. 2016;56(1):e1-e9. doi:10.1097/MLR.0000000000000628

5. Davis AC, Shen E, Shah NR, et al. Segmentation of high-cost adults in an integrated healthcare system based on empirical clustering of acute and chronic conditions. J Gen Intern Med. 2018;33(12):2171-2179. doi:10.1007/s11606-018-4626-0

6. Kansagara D, Englander H, Salanitro A, et al. Risk prediction models for hospital readmission: a systematic review. JAMA. 2011;306(15):1688-1698. doi:10.1001/jama.2011.1515

7. Haas LR, Takahashi PY, Shah ND, et al. Risk-stratification methods for identifying patients for care coordination. Am J Manag Care. 2013;19(9):725-732.

8. Figueroa JF, Frakt AB, Jha AK. Addressing social determinants of health: time for a polysocial risk score. JAMA. 2020;323(16):1553-1554. doi:10.1001/jama.2020.2436

9. Risk adjustment for socioeconomic status or other sociodemographic factors. National Quality Forum. August 2014. Accessed June 5, 2021. https://www.qualityforum.org/Publications/2014/08/
Risk_Adjustment_for_Socioeconomic_Status_or_Other_Sociodemographic_Factors.aspx

10. Pourat N, Chen X, Tsugawa Y, et al. Intersection of complexity and high utilization among health center patients aged 18 to 64 years. Am J Manag Care. Published online September 27, 2021. doi:10.37765/ajmc.2022.88751

11. Arnold J, Thorpe J, Mains-Mason J, Rosland AM. Empiric segmentation of high-risk patients: a structured literature review. Am J Manag Care. Published online September 27, 2021. doi:10.37765/ajmc.2022.88752

12. Mann C. Targeting Medicaid super-utilizers to decrease costs and improve quality. Medicaid.gov. July 24, 2013. Accessed June 5, 2021. https://www.medicaid.gov/federal-policy-guidance/downloads/cib-07-24-2013.pdf

13. Hong CS, Siegel AL, Ferris TG. Caring for high-need, high-cost patients: what makes for a successful care management program? Issue Brief (Commonw Fund). 2014;19:1-19.

14. Baker JM, Grant RW, Gopalan A. A systematic review of care management interventions targeting multimorbidity and high care utilization. BMC Health Serv Res. 2018;18(1):65. doi:10.1186/s12913-018-2881-8

15. Kangovi S, Mitra N, Grande D, Huo H, Smith RA, Long JA. Community health worker support for disadvantaged patients with multiple chronic diseases: a randomized clinical trial. Am J Public Health. 2017;107(10):1660-1667. doi:10.2105/AJPH.2017.303985

16. Williams BC. Limited effects of care management for high utilizers on total healthcare costs. Am J Manag Care. 2015;21(4):e244-e246.

17. Lewis GH. “Impactability models”: identifying the subgroup of high-risk patients most amenable to hospital-avoidance programs. Milbank Q. 2010;88(2):240-255. doi:10.1111/j.1468-0009.2010.00597.x

18. Johnson TL, Rinehart DJ, Durfee J, et al. For many patients who use large amounts of health care services, the need is intense yet temporary. Health Aff (Millwood). 2015;34(8):1312-1319. doi:10.1377/hlthaff.2014.1186

19. Hadeed N, Fendrick AM. Enhance care continuity post COVID-19. Am J Manag Care. 2021;27(4):135-136. doi:10.37765/ajmc.​2021.88508