In this special issue of Medical Care, researchers from the Multiple Chronic Conditions (MCC) Research Network, funded by the Agency for Healthcare Research and Quality (AHRQ), share important findings and suggest topics for future research on persons with MCC. This issue of Medical Care includes a theoretical framework for conducting research on MCC, a review of existing MCC research methods and methodological gaps to be addressed, and the results of several research studies addressing MCC. Major themes addressed by authors in this special issue include factors affecting healthcare costs and utilization for patients with MCC, treatment guidelines and effects, and special considerations for patients whose comorbidities include behavioral health conditions and substance abuse. This compendium of articles both captures the state of research on MCC and serves as a model for future research in this field.
Articles selected for this special issue focus on MCC as defined by the Health and Human Services Initiative on Multiple Chronic Conditions and the National Quality Forum (NQF). The HHS Initiative on Multiple Chronic Conditions encompasses “Chronic conditions that last a year or more and require ongoing medical attention and/or limit activities of daily living. They include both physical conditions such as arthritis, cancer, and HIV infection. Also included are mental and cognitive disorders, such as ongoing depression, substance addiction, and dementia.”1 In articulating a measurement framework for MCC, the National Quality Forum noted that, “Persons with multiple chronic conditions are defined as having two or more concurrent chronic conditions that collectively have an adverse effect on health status, function, or quality of life and that require complex healthcare management, decision-making, or coordination.”2
This issue begins with 2 overview papers that provide a frame of reference for the remaining articles and for the field in general. Grembowski et al3 describe a conceptual model developed by members of the AHRQ MCC Research Network that may be helpful to researchers and policy-makers who focus on MCC. The conceptual model replaces the single-condition paradigm typically used to study chronic disease with a model that embraces complexity, defined as the misalignment between patient needs and the current health services structure. The model offers new insights into how to research and develop solutions for the care needs of patients with MCC. LeRoy et al4 then provide an overview of research findings published by investigators funded by both the 2008 and 2010 rounds of AHRQ MCC grant programs. Their summary provides context for the findings highlighted in this issue of Medical Care and suggests overarching principles for future MCC research. Authors state that patients with MCC should be understood as more than a collection of diseases; rather, the focus should be on holistic outcomes in humans who live meaningful lives in relationship with family and community. This means that research into MCC should: (1) include person-centered and person-driven measures and outcomes; (2) consider the person in the context of their relationships and community; and (3) include mental health as an integral part of health.
Maciejewski and Bayliss5 provide an excellent summary of the main methodological challenges that plague research into MCC and suggest possible remedies. For example, authors note that although randomized controlled trials are a highly valid design, they are cost prohibitive for most MCC research because of the large study samples needed to represent relevant combinations of chronic conditions. Maciejewski and Bayliss5 suggest that quasiexperimental designs using large databases will be the workhorse design of the MCC field. The authors highlight other central methodological issues such as the heterogeneity of study populations and the development of meaningful, patient-centered outcome measures.
Two articles provide approaches for addressing the heterogeneity and complexity of populations with MCC. Zeng et al6 show that a simpler methodological approach can work equally well or better than a more complex one. In their article, Zeng et al6 show that simple summary morbidity measures were as, or more, predictive than more complex longitudinal trajectory measures in models of health outcomes. Chrischilles et al,7 by contrast, found that more complicated approaches could be beneficial. In their study, the addition of a new class of function-related indicators derived from administrative claims data better captured heterogeneity than standard indicators and improved mortality and cardiac catheterization prediction models. These contradictory findings reflect the complex nature of research involving MCC, negating a one size fits all solution and highlighting the need to select the optimal approach for the particular study question, population, and data source.
Yoon et al8 provide advice for researchers and funders who wish to bend the cost curve for the care of persons with MCC. The authors argue that funders should focus their attention on both the most prevalent and the most costly combinations of conditions. In their study population of patients cared for by the Veterans Affairs (VA), the authors confirmed the commonly reported finding that patients with MCC account for approximately one third of patients but two thirds or more of total healthcare costs. Importantly, they showed that patients with the most common combinations of conditions (eg, diabetes, hyperlipidemia, and hypertension) are not the most costly. For patients receiving care by the VA, the costliest triads of comorbidities include combinations of conditions such as spinal injuries, heart failure, renal failure, depression, ischemic heart disease, peripheral vascular disease, and stroke. In addition to their suggestion to focus on high-prevalence and high-cost conditions, Yoon et al8 also recommend that research concentrate on guidelines for joint management of comorbid conditions, prevention of costly conditions, and case management for patients who are costly and high healthcare utilizers.
Freeman et al9 delve into the causes for high costs among patients with behavioral conditions among their multiple chronic health conditions. Their analysis of Maine Medicaid data shows that there are 3 components to the high costs incurred by patients with behavioral health conditions. First, the behavioral condition accounts for the largest share of most patients’ costs. Second, patients with behavioral health conditions also have higher costs for nonbehavioral services than persons without behavioral health conditions, both because they have more physical comorbidities and because treatment costs are higher even controlling for the number of physical comorbidities. Authors also report that patients with behavioral health conditions appear to receive more fragmented care than other patients, further adding to complexity and cost. The authors hypothesize that navigation difficulties and challenges in engaging with the healthcare system may partly account for both the fragmentation and the higher cost.9
Although the articles by Yoon and colleagues and Freeman and colleagues analyze costs at the individual patient level, the article by Hempstead et al10 approaches cost and utilization from a systems perspective, examining a population of hospitalized patients in New Jersey. This paper focuses on the cause of fragmentation of hospital use, namely the use of >1 hospital by individual patients. The authors note that although the use of multiple hospitals is not necessarily inappropriate, it may present barriers to effective care coordination for complex patients with multiple chronic conditions, leading to higher costs or worse outcomes. In their study, patients with a greater number of conditions were more likely than other patients to use multiple hospitals. Other determinants of hospital fragmentation included total number of admissions, lower hospital market concentration, injury, and behavioral health diagnoses.10
Several papers in the special issue focus on current treatment guidelines and assess their appropriateness for patients with MCC. Wyatt et al11 focused on the degree to which guideline-developing bodies considered issues relevant to patients with MCC during the development of treatment guidelines. In their review of 28 sets of diabetes clinical practice guidelines, the authors found that 8 guidelines did not consider the effect of MCC on treatment regimens. The remaining guideline developers addressed multimorbidity from a narrow biological standpoint. The authors recommend that guideline-developing bodies consider the effect of MCC on treatment complexity or burden. Further, they emphasize the need to focus on patient context and patient-centered outcomes to more practically guide the care of patients with multiple chronic conditions.11 These recommendations are beginning to be adopted by guideline developers as evidenced by the just-released 2013 American College of Cardiology/American Heart Association management guideline for heart failure.12 Unfortunately, current efforts at addressing MCC in guidelines are hampered by the lack of evidence to inform recommendations for optimal diagnostic and therapeutic approaches for patients with MCC. Research spurred by AHRQ, NIH, and other sources will hopefully provide the evidence to populate these guidelines over the coming years.
In 4 papers, investigators point out unexplained and potentially unwarranted variation in treatment patterns for patients with MCC. Variation was observed both geographically and based on patients’ specific combinations of conditions. These observations suggest a need for new treatment guidelines to address the variation and/or further research to explain it, as described below.
Brooks and colleagues updated territory covered by Wennberg in the 1970s and 1980s, when he identified unexplained geographic variation in prescribing patterns for single conditions, by exploring this variation in persons with MCC.13–15 In their study, Brooks and colleagues identified unexplained geographic variation in statin-prescribing patterns for patients with MCC after acute myocardial infarction. The authors conclude that patient and provider beliefs and preferences weigh heavily in statin prescribing decisions. As Wennberg’s research did for single conditions, this paper may spur interest in both the development of treatment guidelines and more shared decision making for patients with MCC.15
Jordan and colleagues identified variation of a different kind, based on specific comorbidities. In a sample of Veterans Administration (VA) patients with depression, some comorbidities (eg, cardio/cerebrovascular disease, peptic ulcer/GERD, or arthritis) increased the likelihood of receiving guideline-concordant care, whereas others (eg, substance or alcohol abuse) decreased the likelihood of receiving guideline-concordant care.16 Similarly, Domino et al17 showed that cancer screening and quality measures for single disease were generally lower among persons with MCC who also had mental illness, compared with those without mental illness. These findings emerged from an analysis of integrated Medicaid and mental health system data from the AHRQ-funded North Carolina Integrated Data for Researchers. Domino et al17 argue that quality of care for patients with behavioral conditions can serve as a benchmark for overall quality of care for all persons with MCC. Conversely, Tai-Seale et al,18 in a longitudinal study of 110,000 patients, found that patients with depression were more likely to be diagnosed as overweight or obese than persons without depression. The findings of these 3 studies suggest that specific morbidities such as behavioral health conditions and substance abuse may be sources of health disparity. It cannot be determined from these studies whether the variation in care was appropriate or inappropriate. As for other causes of health disparity, however, the reasons for and the outcomes associated with these variations warrant further investigation.
Although in many cases treatment variation is unwarranted, some papers in this special issue show that a different treatment may be appropriate for patients with certain combinations of MCC. The results of 2 studies address the appropriateness of extrapolating evidence of treatment effect from younger to older or from healthier to sicker populations. Zhang and colleagues review anemia management guidelines among dialysis [end-stage renal disease (ESRD)] patients. Prior research found that targeting normal hematocrit levels versus lower levels resulted in worse clinical outcomes in patients with ESRD. As a result, clinicians now typically aim for hematocrits in the 34.5%–39% range, although no evidence supports the use of this level versus lower hematocrit levels. Zhang and colleagues compared outcomes from ESRD patients with the middle-range hematocrit target versus outcomes for patients with hematocrits in the 30%–34.5% range. Their study showed no statistically significant difference between patients managed with the middle-range target versus the lower target.19 Their study provides important information for the development of future treatment guidelines for anemia management in ESRD patients.
Lee et al20 assessed the use of β-blockers after acute myocardial infarction in older adults with both cardiovascular and pulmonary conditions. Current American Heart Association/American College of Cardiology guidelines recommend that β-blockers be used for 3 years after a myocardial infarction or acute coronary syndrome. The authors found that, contrary to earlier studies showing that β-blockers helped patients who were younger with or without MCC, they neither helped nor harmed older adults with both cardiovascular and pulmonary conditions.20 These findings challenge the current practice of extrapolating results from younger and healthier populations to older persons with MCC.
Finally, one paper draws attention to the challenges for patients of following complex treatment regimens resulting from multiple conditions. Boyd et al21 validated a new psychometric instrument, the Healthcare Treatment Difficulty (HCTD) Scale, to identify patients who may be having difficulty implementing complex treatment regimens. Authors also showed that greater patient activation was associated with lower HCTD scores over time. This scale may be helpful in clinical practice for identifying patients who may not be able to follow their treatment regimen. For these patients, clinicians may want to identify ways to either simplify the treatment regimen or better engage the patient or caregiver.21
A cross-cutting issue in several papers was that of behavioral health or the coexistence of behavioral and physical health conditions. Patients with behavioral health comorbidities are particularly vulnerable. Strategies to address the needs of these most complex patients can potentially benefit all patients with MCC. Several of the articles touching upon behavioral health issues are summarized above.9,10,16–18 In addition, Lichstein et al22 show that patients with serious mental illness may not benefit as readily as other patients from current innovations in healthcare delivery. Their study showed that medical home use was relatively high for Medicaid enrollees with MCC and that patients with MCC including depression were comparable with patients with physical comorbidities only. However, use of medical homes was lower among children and adults with severe mental illness. The authors call for new strategies to increase participation in the medical home model of care among patients with severe mental illness.22
Although the field of MCC research is no longer in its infancy, much remains to be learned and uncovered. Currently available research has begun to delineate the gaps in our understanding of MCC that hopefully will be addressed by researchers in the near future. The articles presented in these pages outline at least 3 directions and priorities for future work.
First, methods for research in the MCC field need further development. We wholeheartedly agree with the suggestion by Maciejewski and Bayliss5 that an agenda for methods development be crafted collaboratively through dialogue with relevant stakeholders including clinicians, methodologists, patients, MCC content experts from a range of disciplines, research funders, and advocacy organizations.
Second, research is needed to improve treatment guidelines for patients with MCC. The prospect may seem daunting, as the myriad combinations of MCC will magnify the complexity of guidelines. To wit, Sorace et al23 showed that a cohort of over 32 million Medicare beneficiaries exhibited >2 million disease combinations. A focus on patient-centered outcomes, facilitated by the judicious use of electronic health records and decision aids to facilitate shared decision-making, may help manage the increased complexity that is likely to result from evidence-based treatment guidelines for patients with MCC. Further, Yoon et al8 offer helpful guidance on where to start. They suggest that research resources be allocated to both the most common combinations of conditions and also the most costly. Efforts related to those conditions should focus on prevention, guidelines for joint management of comorbid conditions, and case management of high utilization and high-cost patients.8
Third, health system issues remain a challenge for patients with MCC. Innovative healthcare delivery models have been developed and implemented in recent years to promote team-based care, care coordination, self-management support, integration of primary and specialty care, and integration of care with community resources. These models include Accountable Care Organizations, Patient-Centered Medical Homes, Health Homes for Medicaid enrollees with chronic conditions and meaningful use of electronic health records. A number of articles in this issue show, however, that these innovations are not equally reaching all patients with MCC. Further efforts are required to develop successful models of coordinated care that can reach these most vulnerable populations.
The authors thank Therese Miller, DrPH, the Project Officer at AHRQ, David Meyers, MD, the Director of the Center for Primary Care, Prevention, and Clinical Partnerships at AHRQ, and Anand Parekh, MD, MPH, the Deputy Assistant Secretary for Health, for their financial and intellectual support of this work and for their dedication to care for patients with MCC. They also thank the Medical Care staff, including the Deputy Editors, Amy Rosen, PhD, and Robert Weech-Maldonado, MBA, PhD, the Publisher, Druanne Martin, the Managing Editor, Karen Doyle, and the Abt Associates team, Lisa LeRoy, MBA, PhD, Melanie Wasserman, PhD, Meghan Woo, PhD, and Emma Oppenheim, for their technical contributions and logistical support.
3. Grembowski DE, Schaefer J, Johnson K, et al..A conceptual model of the role of complexity in the care of patients with multiple chronic conditions.Med Care.2014;523 suppl 2S7–S14.
4. LeRoy L, Bayliss E, Domino M, et al..The Agency for Healthcare Research and Quality Multiple Chronic Conditions Research Network: overview of research contributions and future priorities.Med Care.2014;523 suppl 2S15–S22.
5. Maciejewski ML, Bayliss EA.Approaches to comparative effectiveness research in multimorbid populations.Med Care.2014;523 suppl 2S23–S30.
6. Zeng C, Ellis JL, Steiner JF, et al..Assessment of morbidity over time in predicting health outcomes.Med Care.2014;523 suppl 2S52–S59.
7. Chrischilles E, Schneider K, Wilwert J, et al..Beyond comorbidity: expanding the definition and measurement of complexity among older adults using administrative claims data.Med Care.2014;523 suppl 2S75–S84.
8. Yoon J, Zulman D, Scott JY, et al..Costs associated with multimorbidity among VA patients.Med Care.2014;523 suppl 2S31–S36.
9. Freeman E, McGuire CA, Thomas JW, et al..Factors affecting costs in Medicaid populations with behavioral health disorders.Med Care.2014;523 suppl 2S60–S66.
10. Hempstead K, DeLia D, Cantor JC, et al..The fragmentation of hospital use among a cohort of high utilizers: implications for emerging care coordination strategies for patients with multiple chronic conditions.Med Care.2014;523 suppl 2S67–S74.
11. Wyatt KD, Stuart LM, Brito JP, et al..Out of context: clinical practice guidelines and patients with multiple chronic conditions. A systematic review.Med Care.2014;523 suppl 2S92–S100.
12. Yancy CW, Jessup M, Bozkurt B, et al..2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.J Am Coll Cardiol.2013;62:e147–e239.
13. Wennberg J, Gittelsohn A.Small area variations in health care delivery.Science.1973;182:1102–1108.
14. Wennberg J, Gittelsohn A.Variations in medical care among small areas.Sci Am.1982;246:120–134.
15. Brooks JM, Cook EA, Chapman CG, et al..Geographic variation in statin use for complex acute myocardial infarction patients: evidence of effective care?Med Care.2014;523 suppl 2S37–S44.
16. Jordan N, Sohn M-W, Bartle B, et al..Association between chronic illness complexity and receipt of evidence-based depression care.Med Care.2014;523 suppl 2S126–S131.
17. Domino M, Beadles CA, Lichstein JC, et al..Heterogeneity in the quality of care for patients with multiple chronic conditions by psychiatric comorbidity.Med Care.2014;523 suppl 2S101–S109.
18. Tai-Seale M, Wilson CJ, Stone A, et al..Patients’ body mass index and blood pressure over time: diagnoses, treatments and the effects of comorbidities.Med Care.2014;523 suppl 2S110–S117.
19. Zhang Y, Thamer M, Kaufman J, et al..Comparative effectiveness of two anemia management strategies for complex elderly dialysis patients.Med Care.2014;523 suppl 2S132–S138.
20. Lee DSH, Markwardt S, McAvay GJ, et al..Effect of β-blockers on cardiac and pulmonary events and death in older adults with cardiovascular disease and chronic obstructive pulmonary disease.Med Care.2014;523 suppl 2S45–S51.
21. Boyd CM, Wolff JL, Giovannetti E, et al..Healthcare task difficulty among older adults with multimorbidity.Med Care.2014;523 suppl 2S118–S125.
22. Lichstein J, Domino ME, Beadles CA, et al..Use of medical homes by patients with comorbid physical and severe mental illness.Med Care.2014;523 suppl 2S85–S91.
23. Sorace J, Wong H, Worrall C, et al..The complexity of disease combinations in the Medicare population.Popul Health Manag.2011;14:161–166.