Emerging models of care for individuals with multiple chronic conditions

Abstract

Objective

To characterize emerging and current practice models to more effectively treat and support patients with multiple chronic conditions (MCC).

Data Sources/Study Setting

We conducted a rapid literature scoping augmented by key informant interviews with clinicians knowledgeable about MCC care from a broad spectrum of US delivery systems and feedback from multidisciplinary experts at two virtual meetings.

Study Design

Literature findings were triangulated with data from semi‐structured interviews with clinical experts. Reflections on early results were obtained from policy, research, clinical, advocacy, and patient representatives at two virtual meetings sponsored by the Agency for Healthcare Research and Quality. Emergent themes addressed were as follows: (1) more timely strategies for MCC care; and (2) trends not previously represented in the peer‐reviewed literature.

Data Collection/Extraction Methods

The rapid literature scoping relied on Ovid MEDLINE(R) and Epub Ahead of Print databases for the most recent 5‐year period. Qualitative interviews were conducted by telephone. Virtual meetings provided oral and written (chat) captured inputs.

Principal Findings

Although the literature scoping did not identify a specific set of evidence‐based care models, key informant discussions identified eight themes reflecting emerging approaches to population‐based MCC care. For example, addressing the needs of individuals with MCC through a complexity lens by assessing and addressing social risk factors; extending the care continuum with home‐based care; understanding how to address ongoing patient and caregiver supports outside of clinical encounters; and engaging available community resources.

Conclusions

Integrating care for MCC patient populations requires processes for determining different subpopulation needs in various settings and lived experiences. Innovation should be anchored at the nexus of payment systems, social risks, medical needs, and community‐based resources. Our learnings suggest a need for an ongoing MCC care research agenda to inform new approaches to care delivery incorporating innovations in technology and home‐based supports for patients and caregivers.

What is known on this topic

• People with multiple chronic conditions (MCC) are more likely to have social risks worsening their health and complicating their health care needs. ¹ , ² , ³ , ⁴ , ⁵

• Interventions to improve MCC care have focused on bundled, care delivery‐ anchored interventions reflecting the heterogeneity and complexity of the patient population.

• Recommended elements of care for individuals with MCC include team‐based care, care coordination, community partnerships, shared decision making, and assessment of patient treatment burden including assessment of social risk factors.

What this study adds

• There is no single ideal care model or set of discrete MCC care components recommended to improve care for people with MCC.

• Emerging models of MCC care incorporate home‐based supports beyond traditional care delivery interactions to align system and community resources with patient needs.

• Gaps in knowledge include addressing social risk and health disparities within medical complexity, aligning financial incentives for delivery systems, and applying methods to address population heterogeneity.

1. INTRODUCTION

Delivering care to populations with multiple chronic conditions (MCC), typically defined as two or more chronic physical or mental health conditions experienced by the same individual, comprises a disproportionate share of health care spending. ⁶ , ⁷ , ⁸ One in every four Americans and two‐thirds of Medicare beneficiaries have MCC, and 75 cents of every dollar spent on health care services is spent managing chronic conditions. ⁹ Multimorbidity is more common among individuals with lower socioeconomic status, and disproportionately affects racial and ethnic minorities. ¹⁰

Health care systems originally designed to treat single conditions may not have well‐developed processes to support MCC care delivery. Individuals with MCC often require multiple providers, and MCC populations are more likely than lower morbidity populations to experience financial constraints and associated social needs. ¹¹ , ¹² Impaired communication between providers can result in contradictory advice for patients, insufficient chronic care coordination, and missed opportunities to leverage patient self‐management. ³ , ¹³ , ¹⁴ Interactions between patients' social and medical needs can further complicate care delivery and worsen health outcomes for MCC individuals. ¹ , ² , ³ , ⁴ , ⁵ Further, current reimbursement for care delivery is insufficient to support the longer visit times needed for patient‐centered approaches to complex care needs. ¹⁵ , ¹⁶ , ¹⁷

Individuals with MCC are under‐represented in randomized controlled trials (RCT) resulting in relatively little evidence to guide care for populations with MCC or for those with single diseases in the context of comorbid MCC. ⁷ , ¹⁸ , ¹⁹ Reviews of natural experiments and intervention evaluations have sought to identify and synthesize key features of successful MCC care. ⁴ , ⁵ , ⁷ , ²⁰ Recommendations from these reviews include targeting patients with specific risk factors (including the cost of care) or functional limitations, performing comprehensive individual assessments of care needs and goals, providing excellent care coordination including for transitions of care and within team communication, aligning care needs with specific settings (such as primary care), segmenting heterogeneous populations into more homogenous groups, and an organizational culture that embraces data‐based continuous learning. However, studies included in this background literature have limitations such as short‐term evaluation of populations with long‐term care needs, single‐site trials, or programs targeting individuals with specific conditions. Further, they do not focus on current approaches to individuals living with MCC and potentially other complex care needs. It is also unclear how recommendations from reviews apply to different populations in different settings. Care delivery organizations, payers, quality assessment organizations, funders, patient advocates, professional societies, and others have begun to recognize the knowledge gaps in meeting these patients needs and are encouraging research studies directed at this population. ¹⁹

To better characterize emerging models of MCC care delivery and new knowledge required to inform these emerging models, we conducted a systematic, iterative process to identify and describe high priority areas for further investigation that included the following:

1. A rapid literature scoping of MCC care models reported in the peer‐reviewed literature over the past 5 years;

2. Key informant interviews with a purposive sample of nine experts across different care delivery systems and settings; and

3. Triangulating literature findings, input from interviews, interactive conferences on MCC care, personal experience studying MCC care, and approaches within our own integrated delivery system.

2. METHODS

A rapid scoping of the peer‐reviewed literature was conducted in March–April 2020. Articles were restricted to studies of interventions with high‐cost, high‐need, or high health care utilization populations, or with individuals having multimorbidity or MCC (terms we use interchangeably as they reflect similar constructs and terminology differs primarily by geographic region or author preference). Results had to include outcomes related to care quality, patient or provider satisfaction, health care utilization, and/or cost. Details can be found in the Supplemental information (Appendix S1 A ).

To complement literature findings, we conducted semi‐structured interviews with a purposive sample of clinical experts from Accelerating Change and Transformation in Organizations and Network (ACTION) IV health systems across the United states with varying size, geography, and patient populations. Interviews were conducted by the authors using an established protocol ²¹ (Table 1 and Appendix S1 B). From this, responses from the first nine were contacted for scheduling to meet AHRQ's project timeline. Questions focused on perceived challenges facing populations living with MCC, organizational approaches to supporting MCC care, services, and resources available to specific subgroups and approaches to identifying these subgroups, technology or tools to support MCC care delivery, and reflections MCC initiatives or interventions (e.g., system or department) and resources to support them. We also explored learnings gleaned from changes in care delivery prompted by the SARS‐CoV‐2 (Covid) pandemic. Interviews lasted 30–45 min and were conducted by telephone by the authors between September and October, 2020.

TABLE 1.

Health systems with representatives participating in interviews from all US Census Regions

System name	Base location
Denver Health	Colorado
Hawaii Pacific Health	Hawaii
Intermountain Health care	Utah
Lehigh Valley	Pennsylvania
Knowledge and Evaluation Research Unit, Mayo Clinic	Minnesota
Military Health System	United States (Washington, DC) a
Northwell Health	New York
Southwestern Health Resources	Texas
Veterans Administration	United States (California) a

^aLocation of the key informant within the system.

Interview notes were thematically coded (i.e., themes were identified with ≥3 unique, individual mentions) to identify emerging trends. We additionally identified novel innovations and illustrative quotes. Both the Kaiser Permanente Northwest and Kaiser Permanente Colorado regional Institutional Review Boards exempted these interviews from informed consent.

Literature review and qualitative results were augmented with input from experts in the field of MCC research and care delivery who attended meetings on the topic in May 2020 and/or November 2020 (see Table 2). Meeting participants were identified by AHRQ and Academy Health staff and meetings were sponsored by AHRQ. Inputs from these three data sources were triangulated to guide recommendations for future research.

TABLE 2.

Soliciting broad‐based feedback and reactions to work in progress

AHRQ convening to solicit feedback & guidance	Transforming care for people living with multiple chronic conditions planning meeting	2020 Research summit on transforming care for people living with multiple chronic conditions (MCCs)
Virtual Meeting Date	May 19, 2020	November 17–18, 2020
Meeting Goals	Identify research priorities to improve care for people living with MCC. Help structure the Fall Research Summit to most effectively meet this goal. This is one of AHRQ's top priorities.	Conversations about living with MCCs. Cutting‐edge methods for MCC research. Feedback to authors on work in progress. Identifying research gaps.
Number of participants	75	133
Manuscript presentations & discussion topics for all papers—(1) patient‐centeredness, (2) models of care, (3) health care technology	What is the state of the evidence? What are the challenges/gaps? What are further research and policy needs?	What we know and where are the gaps: Authors' insights from their research

2.1. Literature review results

Fourteen articles met inclusion criteria for our rapid literature scoping. Of these, seven were prospective studies and seven were retrospective studies using administrative health system or Medicaid claims data. Two prospective studies evaluated the implementation of an intervention, ²² , ²³ one prospective study used a cluster RCT, ²⁴ one study was a randomized quality improvement trial, ²⁵ and two were prospective cohort studies, although one included an economic analysis of the prospective cohort study. ²⁶ , ²⁷ Finally, one prospective study, using data from a Chronic Care Model provided by primary care providers in Italy, collected phone survey data in patients with single conditions or MCC. ²⁸ The prospective implementation studies, as well as the cluster RCT, were considered mixed methods. ²² , ²³ , ²⁴ Of the seven retrospective studies, five were retrospective cohort designs, ²⁹ , ³⁰ , ³¹ , ³² one was an interrupted time series, ³³ and one was a retrospective review of longitudinal program data without any comparison group. ³⁴ One retrospective study from the Veterans Health Administration investigated a telehealth intervention. ³⁰ Interventions in four studies addressed behavioral or mental health concerns, whether through a care team member, formal intervention, or care coordination with behavioral health. ²³ , ³¹ , ³³ , ³⁵

Table 3 summarizes populations targeted and intervention elements for the 14 selected articles using a framework from a previous comprehensive literature synthesis on complex and MCC care needs. ³⁶ In addition to the care model features from the framework, eight studies incorporated a team‐based approach to MCC care. ²² , ²³ , ²⁵ , ²⁶ , ²⁸ , ³³ , ³⁴ , ³⁵ Most (12/14) articles focused on health care utilization and/or cost outcomes. Two studies instead examined patient experience or patient self‐management and satisfaction. ²⁸ , ²⁹ Four studies included some measure of physical function, quality of life, patient experience, clinical outcomes, or self‐rated health, in addition to utilization and cost outcomes. ²³ , ²⁴ , ²⁵ , ²⁷ Studies showed mixed results on all outcomes, although patients tended to rate their experience more highly if they felt they were supported and had well‐coordinated care. ²³ , ²⁴ There were also inconsistent findings on utilization outcomes with many programs noting decreases in emergent or hospital care utilization, though follow‐up time, specific outcomes, and methods for cost calculation varied widely across studies.

TABLE 3.

Target population and features of care models from the literature (January 2015–April 2020)

Care model	Target population	Patient age criteria	Assessment	Self‐care	Coordination/communication	Care transitions	Patient preferences	Social supports	Tailored care plan
Care Support 20	High utilizers, all payers	≥18	X		X				X
PACT Program 27	Adults with 2 + admissions in 6 months, including BH, all payers	≥18	X	X	X	X			X
CareOne 33	Top 1% of high utilizers, MM, all payers	≥18	X		X			X
Community Care (NC) Medical Home 29	Adults with SCZ & 1+ MM, Medicaid only	≥18	X			X
Chronic Care Model 26	1 chronic\condition or MM, non‐U.S.	≥18		X
CCHT (VHA) 28	Veterans living at home (≥1 chronic condition)	≥18			X
Chronic Care Management 30	Medicaid patients with functional disability receiving in home care	18–63	X	X	X		X		X
Integrated Care Pathway GeriRehab 24	OA DC'd to rehab post‐IP, non‐U.S.	>65			X	X	X
ECHO 31	Medicaid with chronic mental illness, high cost with two IP admissions or three ED visits	≥18			X	X		X
HouseCalls 32	Medicare Advantage & commercial pay, high‐risk, home‐bound	≥18	X	X	X	X			X
BHICCI 21	MediCal with 1+ medical & 1+ behavioral condition	≥18	X	X	X
ISCOPE 22	MM; non‐U.S.	≥75			X		X		X
CareMore 23	Top 5% TME; Medicaid only	≥18	X	X	X			X	X
LifeCourse 25	One chronic condition including dementia, all payers	≥18	X		X		X

Abbreviations: BH, behavioral health; BHICCI, behavioral health integration and complex care initiative; CCHT, care coordination home telehealth; CCI, Charlson Comorbidity Index; DC, discharged; ECHO, extension for community health outcomes; GeriRehab, geriatric rehabilitation; ISCOPE, integrated systematic care for older people; MM, multimorbidity; OA, older adults; PACT, preventable admissions care team; SCZ, schizophrenia; TME, total medical costs; VHA, Veterans Health Administration.

2.2. Interview results

The goal of the interviews was to solicit information about emerging improvements in models of care targeting patients with MCC prior to and during the COVID pandemic. In‐depth discussions with nine key informants from a broad spectrum of delivery system settings yielded eight themes itemized in Table 4 and expanded upon below.

TABLE 4.

Agreement among interviewees on MCC care themes [Color table can be viewed at wileyonlinelibrary.com]

Identified themes	Number of unique mentions
	1	2	3	4	5	6	7	8	9
Patient‐centered approach
Complex care, including social risk factors
Focus on transitions
Incorporating community
Population management
IT tools
Financial incentives/reimbursement/ROI
Importance of research/evidence‐based practice

Patient‐centered approach (seven out of nine mentioned): Most respondents noted the importance of the relationship between the patient and the entire care team. Multidisciplinary teams and integrative practice units augmented the patient–provider relationship and helped coordinate care—a key element in Table 3. Expanding access to home care was also seen as patient‐centered by meeting MCC patients where they are. One respondent commented: “Patients are more comfortable at home; [our system is] exploring what care elements can be accomplished at home,” also noting that COVID has provided increased uptake of home care. Examples include hospital at home, home visits, home‐based palliative care, and remote monitoring. Informants noted that increased uptake of virtual care during the COVID pandemic has reinvigorated primary care through new connections. However, health outcomes for individuals with MCC in an increasingly virtual world will ultimately depend on the complexity of their care needs and their personal abilities and resources.

Complex care needs, including social risk (nine out of nine mentioned): Our key informants were unanimous in describing the defined population as “complex” versus individuals with “MCC.” We discussed how “…complexity of care involves identifying and addressing social risk” together with medical risk. This distinction requires identifying social risks and addressing associated needs as part of holistic care for individuals with MCC. “This is not easy, patients are extremely variable—many are medically complex, socially, and psychologically complex. Those with psychological complexity often have impaired ability and separate social issues like unemployment, homelessness, domestic violence—all exacerbate their medical conditions.” Another expert commented, “sick care hasn't really helped people manage their lives outside the bounds of our care delivery system. People need to find the reserve to face the workload of their disease burden—resilience, social support, financial support.” Several respondents hoped that the pandemic could be potentially transformative, optimizing the use of virtual care to support enhanced self‐management. However, they also noted that it has amplified some problems including consequences of delays in care and social isolation—worsening social risks and amplifying social needs. One system leader explained, “The fear factor of COVID is a huge part of our problem for these MCC patients who are at high risk.” All respondents felt that until social risks are addressed, we cannot begin to address patients' medical conditions.

Focus on transitions (five out of nine mentioned): Traditionally, MCC interventions have focused on coordinating care and communications across the continuum of care (e.g., primary to specialty care, inpatient to medical home). “The fragmentation of care is a challenge—especially the transitions between primary care, specialty care, different locations, different hospitals, some with no access to our EMR—with poor communication across these.” Several experts distinguished between encounter‐based care and what one respondent referred to as “between visit care” in their delivery systems. This distinction markedly expands clinicians' abilities and available resources to meet the varied needs of MCC patients. Visit‐based care delivery models are unlikely to succeed as the point of care—especially primary care—is full. ³⁷ , ³⁸ , ³⁹ According to one respondent, “The biggest challenge are those patients we're not in touch with 300 or so odd days out of the year, making sure they have consistent services and supports—not the 5 days they are in the hospital or 20+ minutes during all other (visit) moments; we need to make progress in addressing that time. We are focusing on complex at‐risk and rising risk patients and aiming to decrease in hospital time. We use a transitional care model (TCM). You could say we are using a kitchen sink approach with TCM including a 24‐h and 72‐h follow up calls post discharge, polypharmacy assessment, and home visits.” Several respondents described extending the care continuum with home visits and hospital‐at‐home services. Such “between visit care” requires coordinated outreach efforts to minimize patient burden and improve system efficiency. ⁵ , ²⁵ , ⁴⁰ Learnings prompted by the COVID pandemic, such as how to optimize virtual care, are particularly relevant for individuals with MCC and could become incorporated into a care continuum.

Incorporating the community (five out of nine mentioned): Interviewees emphasized the need for health systems to work with communities. One mentioned jointly identifying areas for cross‐subsidization such as between‐care support or addressing social needs such as housing insecurity. “In this way, the community capacity is an active stakeholder and is not incapacitated.” Another system respondent noted that “We need to elevate health—identify needs, transfer solutions to the community, and push for public health support.” Informants noted several interventions bridging the gap between the care delivery system and community by using community‐based organization (CBO) navigators and community health workers. “We are expanding to incorporate community—closing the loop with CBOs through notifications that something happened, what is known—coupled together with an exception report; missed appointment notification, ED visit, and/or hospitalization note in our network. These data are then used to understand and predict patterns in utilization and optimal opportunities to proactively reach out to our patients.”

Population management (five out of nine mentioned) relies on segmenting heterogeneous populations into more homogeneous subgroups and tiering these groups by some measure of predicted risk to appropriately tailor care and meet addressable needs. Thus, creating more homogenous patient subgroups was considered a core patient‐centered strategy for MCC care management across delivery systems. ¹² , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ Approaches to segmentation vary with available data and reflect specific patient needs, available resources to address those needs, and system priorities. Examples range from targeting specific social needs or racial/ethnic disparities to identifying patients at risk of readmission. Tiering patients by level of predicted risk aims at “getting the right care to the right patient.” “Our IT system gives us the ability to find and monitor patients in the system using multiple criteria and regular reports.” Although terms such as “high utilizers” may be perceived as judgmental at the level of the individual, utilization or admission risk were seen by system leaders as essential to identifying patients for outreach. As noted by one key informant, “Care management teams' working on a risk stratified population to avoid unnecessary readmissions helps.” As our approaches to population management evolve with predictive analytics, one respondent noted that “The digital divide and disparities are under‐addressed issues. Clinical and technical solutions are important, but we need to address these risk factors, or our work will be frustratingly incomplete.”

Information technology (IT) tools (seven out of nine mentions): In‐person care can leverage specific tools such as EHR‐based best practice reminders to engage care teams during visits or hospitalizations. Between‐visit care can potentially be enhanced by other tools such as remote monitoring devices, which were greatly expanded during the COVID‐19 pandemic. One system reported using a chatbot to contact MCC patients between visits and alert clinical teams to deviations in clinical status. Standardized screening incorporating electronic referral tools allows for personalized outreach focused on specific needs such as connecting patients to community resources for social support or to coordinate addressing social risks. Electronic referral systems offer bi‐directional information sharing between health care delivery systems and CBOs. Lastly, the COVID pandemic has greatly expanded the use of virtual care for all patients, including those with MCC. Respondents believe that “telehealth and remote monitoring are important tools that will persist.”

Financial (three out of nine mentions): Reimbursement structures (e.g., fee‐for‐service or FFS, risk‐based or capitated payments) and financial incentives (e.g., quality penalties) varied by system ranging from the Accountable Care Organization model that drives a population‐based approach, to FFS that disincentivizes well care, to capitated models that incentivize wellness. These structures strongly influenced the development and evaluation of MCC care initiatives. One respondent noted how the reimbursement structure impeded its system's ability to demonstrate a return on investment (ROI) for targeted MCC services whereby the system made the investment in targeted care but the cost savings accrued to the insurer and patients. For system leaders, effective approaches to MCC care lay at the intersection of meeting patients' needs, addressing system needs, available community resources, and aligning financial incentives. Identifying and addressing patient social needs was a frequently cited example of this sweet spot (e.g., transportation): improving access through transportation support meets an essential patient need, increases access to necessary care, and thereby decreases the likelihood of unnecessary emergency or hospital utilization. One key informant put it well, noting that “Caring for these individuals is a system issue—care incentives are misaligned; we don't fund well care; but there are lots of incentives to care for the sick.”

Importance of research/evidence‐based practice (4): Nearly all the mission and vision statements from participating systems refer to their commitment to providing evidence‐based care. Several of our key informants specifically noted their desire for evidence‐based approaches to MCC care and a few described embedded research structures within their care delivery system to support literature review and evaluation. As one respondent noted, we have “work groups focusing on specific areas, going to the literature, developing an approach, implementing and scaling if needed.” In dynamic operating environments, there is quite a bit of rapid innovation in place, which has been heightened by rapid shifts in care delivery during the COVID pandemic but is at risk for insufficient evaluation due to the rapid pace of change. “A big concern is that I'm not sure we are evaluating…how we are holding each other responsible to determine what is working.” The changes needed to adapt to a contagious pandemic have led to rapid implementation without a disciplined evaluation approach—a call to learn and report on what is being learned through research. While there is overlap in care delivery intervention elements identified in the literature scoping (e.g., coordinated care, patient preferences, tailored care plans), there were clearly advances in how delivery systems are continuing to evolve in meeting the needs of patients with MCC.

3. DISCUSSION

There is a growing evidence base for MCC care, and health care systems are implementing processes to foster patient‐centered care delivery for this population. However, system‐ and patient‐level innovations are not necessarily based on published evidence, and the literature does not address many aspects of essential MCC care articulated by clinical leaders. Our literature review revealed several challenges related to uptake of available evidence due to mixed/weak reported results, lack of intervention fidelity, and heterogeneous patient populations that cannot be readily extrapolated to different contexts with varying resource constraints.

In addition to our scoping review, learnings from the geriatric and palliative care literature (e.g., deprescribing to minimize treatment risk and burden, and clearly defining goals of care) can inform MCC care across the age spectrum. However, other essential MCC care elements such as best practices for addressing social determinants of health and delivering accessible care to improve health equity are insufficiently studied. Many persons with MCC are in younger age groups and have complexities and health disparities that derive from social, medical, and mental health complexities. ⁴⁵ , ⁴⁶ These groups in particular will benefit from new approaches to MCC care.

Literature findings, interviewees, and meeting participants all highlighted the importance of a patient‐centered approach: Understanding and addressing patient preferences for how care is delivered and aligning the content of care with patients' goals. From a system standpoint, this might include more convenient care delivery options (e.g., virtual or home visits) or developing dedicated care teams to help patients achieve those goals. Such team‐based care was considered essential by interviewees for facilitating novel care delivery options (e.g., hospital at home) and augmenting the patient–provider relationship. Managing transitions of care was another area of mutual focus, though interviewees expanded the concept of care transitions to include those between primary and specialty care and between office visits and home care responsibilities (not just commonly recognized care transitions between hospital and skilled nursing facility and home).

Areas of discordance between literature sources and interviewees included identifying and managing social determinants of health, infrastructure support for care delivery, incorporating evidence generation into MCC care delivery, and the influence of financial incentives. Interviewees felt that identifying and addressing social determinants of health was essential to MCC care; while racial/ethnic disparities were not directly mentioned, the intersectionality of such disparities aligns directly with current efforts for social risk screening. Three reviewed studies highlighted social support, but only in the context of populations with known financial constraints (e.g., Medicaid beneficiaries) despite the growing recognition of isolation as a social risk. Infrastructure support—specifically IT—was mentioned multiple times by interviewees and conference participants, but not identified in the literature review. Interviewees saw IT as underpinning patient‐centered care, informational continuity, and population segmentation to tailor care. Although reviewed studies focused on evidence generation, interviewees mentioned evidence generation as an aspirational aspect of operational initiatives implemented to meet immediate patient and population needs. Finally, interviewees considered financial incentives as contextual imperatives driving much system change and emphasized the need for aligned incentives. Financial incentives were not addressed by reviewed studies, though reimbursement structures (e.g., Medicaid demonstration projects) did inform populations studied.

Successful emerging models of MCC care will reflect organizational processes supported by IT infrastructure and aligned financial incentives that discern patient goals and needs within and between settings and align those needs with system and community resources. ⁴⁰ Such a patient‐centered, process‐focused approach reflects the evolution of population‐based care to target resources to meet the needs of individuals with MCC: Early siloed, condition‐specific, specialty care and registries gave way to population management and coordinated care for patients living with MCC—which was further enhanced by a complexity lens incorporating social risk screening and links to community resources, more recently incorporating technological advances to extend to home and community‐based supports, virtual visits, and remote sensing.

Areas for further evidence generation that emerged from our information synthesis include the following:

• Develop and apply processes for understanding population heterogeneity to better match care and resources to patient needs. Identifying patients and populations for targeted care delivery have frequently focused on cost. However, this approach (often reflected in utilization outcomes) has not prompted effective care models or improved outcomes ⁴⁷ , ⁴⁸ , ⁴⁹ : Health care needs of MCC populations are often dynamic with fluctuating needs for services. ⁵⁰ , ⁵¹ Patient‐reported data on function and well‐being, coupled with newer dynamic modeling techniques may, better identify populations at risk for future health care utilization. ¹² , ⁵² Patient‐reported data may also reveal understudied MCC sub‐populations at risk for poor outcomes such as middle age adults who are faced with competing demands and logistical challenges. ¹ , ⁵³ , ⁵⁴ Refined predictive models may help stratify populations by risk level, segment individuals into homogeneous groups, and target those with actionable care needs. ⁵ , ⁵⁵

• Generate evidence on identifying and addressing social determinants of health to support a culture of health in the community served. Effective collaboration will likely require effective and ongoing data sharing between health care delivery and community entities, identifying individuals with MCC who may be interested in community referrals, and engaging the community in establishing priorities for shared efforts to support population health.

• Align financial incentives to support best practices. This remains elusive outside a risk‐based reimbursement environment and will require that those who invest in targeted care programming also benefit from it. It will also require assessing long‐term returns on care management investments made by systems, payers, communities, and informal caregivers.

• Couple short‐term utilization and quality metric outcomes with newly developed outcomes that reflect the priorities of individuals with MCC (e.g., quality of life, minimizing treatment burden, health literacy, digital connectivity), and can be used to evaluate new programs. ¹⁴ Patient‐centered metrics incorporated into system‐level performance evaluation and associated reimbursement strategies could further align financial incentives for improved MCC care.

• Employ implementation science strategies to determine elements of complex interventions that are essential (“nonadaptable”), which elements of an MCC care interventions can be adapted to local settings, and which interventions should (and should not) be taken to scale.

• Determine which MCC patient needs and preferences can be addressed remotely based on learnings about virtual care gleaned from the COVID pandemic and clarify essential IT support for this care. Emerging evidence on telehealth may be particularly useful for the many individuals with MCC who manage competing demands of work, family, and MCC self‐care although research must address issues of disparities and inequities that are associated with the digital divide and digital health literacy.

Addressing these questions in real‐world settings will align MCC care evidence generation with essential domains for primary care health services research in which effective organization and financing of care, coupled with personal preferences and social needs assessments result in high‐quality, equitable‐care using appropriate resources. ⁵⁶ Efficient implementation can be enhanced through dedicated dissemination opportunities at the intersection of quality improvement, implementation science, and health services research. ⁵⁷

4. LIMITATIONS

Our efforts to explore emerging directions in care redesign for patients with MCCs were based on recently published evidence and key informant interviews with clinicians in the field with expertise in treating patients with MCCs. To reveal current thinking on where health system resources were being invested, we limited the rapid literature scoping to the most recent 5 years and a purposeful sample of nine clinicians leading MCC care at their respective institutions. The commissioned paper, prepared under contract from the Agency for Healthcare Research and Quality (AHRQ), limited the number of key informants to be interviewed in compliance with Office of Management and Budget restrictions to nine interviews. While the number of in‐depth discussions was limited, much more broad‐based reaction and input were received from participants at the AHRQ planning meeting and research summit. Despite this broad input, many findings are familiar—illustrating the need for ongoing, rigorous MCC evidence generation. Our assessment of emerging and current practice models did not include a formal policy analysis. It is true that selected states and the Federal government have recognized issues of access and affordability in caring for people with MCCs for some time. In particular, the Department of Health and Human Services, the Centers for Medicare and Medicaid Services (CMS), and AHRQ have contributed in terms of strategic direction, large‐scale convenings, payment reforms, and sponsored evaluation of targeted programmatic demonstrations to advance care for people with MCCs. While we did not explicitly probe on the impact of these or more recent state/federal policy changes (e.g., revisions in use of supplemental benefits in the Medicare Advantage program) or regulatory relief during the pandemic (e.g., reimbursement for virtual care, hospital at home programs), payment policies related to reimbursement emerged as a theme.

5. CONCLUSIONS

Essential components of emerging MCC care models include team‐based care, care coordination, community partnerships, shared decision making, home‐based supports, and understanding patient treatment burden including assessment of social risk factors. We need to develop processes for determining the needs of different subpopulations and align incentives and IT infrastructure to ensure they are met. “Success” should be measured long‐term and with patient priorities in mind. Effective models of MCC care should be scalable and adaptable and have the capacity to handle complex interactions.

Evidence for innovative MCC care delivery strategies will likely emerge from the nexus between quality improvement, implementation science, and health services research. ⁵⁷ A greater reliance on pragmatic investigations from learning health systems along with mixed methods and observational studies is needed together with targeted journals willing to publish non‐RCT study designs. Such an approach could result in a toolbox of MCC model care elements that permit adaptable approaches to system‐specific programming, avoids a one‐size‐fits‐all approach, and still achieves improved patient‐centered care and optimal health outcomes.

Supporting information

Appendix S1 Supporting information

Savitz LA, Bayliss EA. Emerging models of care for individuals with multiple chronic conditions. Health Serv Res. 2021;56(S1):980‐989. 10.1111/1475-6773.13774