Understanding the Context of Health for Persons With Multiple Chronic Conditions: Moving From What Is the Matter to What Matters

After 15 years of practice in rural family medicine, one of my daily work experiences is feeling the squeeze between the rock of practice guidelines and the hard place of how my patients want to live their lives. Often, my patients make health-related decisions based on personal considerations far more then the advice of expert panels or research results. I'm most unsuccessful at "treating to target" when my best advice contradicts a life story: "I'm doing OK, just a little bit of diabetes" or "I had that heart attack but my heart's fine now."

As the conversation currently plays out, I get to report the bad news of a rising A1C result or declining ejection fraction. But then I don't have the information or skills to discuss the actual burdens of adding daily insulin to one's life, nor the expected benefit of lowering a particular person's average blood sugar from 160 to 120.

In the course of a regular day, I frequently find myself overlaying 4-10 practice guidelines in order to capture the screening and treatment measures that apply to a single person. But which interventions are more important, and when recommendations conflict how do we choose? At what point does the sum of burdens from the interventions outweigh the benefit? I'm currently serving as an on-the-fly integrator, stitching together multiple disease-based recommendations to try to represent what's best for the real person in front of me.

Clinicians and the communities we serve will welcome a research approach that reframes the questions, as recommended in this article. I look forward to researchers learning what strategies are likely to achieve what goals for whole people across the course of their lives. Every day in clinic I work hard to channel our collective knowledge into meaningful positive change in a individual patient's life. My patients and I could really use some help navigating the complicated, multifaceted real world in which we find ourselves.

Competing interests: None declared

This article provides a refreshing view and brings a view of realism to a complex subject. A paradigm shift is needed as highlighted by these authors from a disease-based to a person-driven and goal-directed approach. This approach needs to drive research but it can't stop there. Policy and practice also need to shift. Achieving the triple aim is a subtext here but paramount in the policy arena.

Advancing the research agenda means changing the way that we view evidence. Narrowly specified questions with narrowly specified interventions and narrowly specified outcomes in unrealistic environments are far less useful than realist evaluations that incorporate context in the interpretation of results. We also need new designs to allow for and rigorously evaluate the effect of adaptive trials where interventions are not fixed in the protocol but can adapt to become more effective when gaps and failures are detected early in the trial. Rigourous evaluations of quality-improvement approaches are needed. This also requires changing accepted definitions of rigour from reductionist to more expansive realist approaches.

The research agenda proposed by the authors requires a broad network of researchers to address even a few of the questions thoroughly. Some of the most vexing proposed questions may be 'how can interacting effects of MCC be understood', 'how can patients [with MCC] be [effectively] supported in their self management', 'what can be learned from international comparisons', and 'how do multidirectional linkages that include community resources affect the health and health care of persons with MCC'. For all of these the authors make the poignant case that context is key, but answers will have to provide evidence to guide specific practice and policy change.

Competing interests: None declared

The report is very helpful in giving appropriate attention to social and economic factors. However, the people with high utilization with MCC are mostly living with increasing disability and will die without escaping these multiple conditions. This seems to be important to patients and to decisions, but is not mentioned in this report or in other reports about MCC. Certainly, there are some combinations that are not eventually fatal, but even these are progressively disabling.

Competing interests: None declared

Elizabeth Bayliss and colleagues should be commended for highlighting the significance of "contextual" factors to better understand the causes and consequences of multiple chronic conditions (MCC) in individuals (1). This work follows in large part from a set of recommendations included in a recent report by the U.S. Department of Health and Human Services (DHHS), "Multiple Chronic Conditions: A Strategic Framework" to improve the evidence base for MCC (2). Indeed, several of the co-authors of the Bayliss article were instrumental in the development and dissemination of that seminal DHHS report.

As noted by Bayliss and colleagues, a contextual approach shifts the focus from categorical disease to the individual and, most important, the range of factors that affect individuals as part of their everyday lives. The contextual approach is very much in keeping with an ecological perspective that is becoming the standard framework for research in population health. As described in an Institute of Medicine report, an ecological approach assumes that "patterns of health and well-being are due to a dynamic interplay of biological, behavioral, social, and environmental factors that unfolds over the life course of individuals, families, and communities" (3). Bayliss and colleagues make a compelling case that this contextual or ecological approach will advance the DHHS MCC research agenda to enhance: (a) the external validity of clinical trials; (b) the epidemiology of MCCs; (c) clinical, community, and patient- centered health research; and (d) research on disparities in MCCs within and among populations.

With that as background, I want to call attention to a subject, identified in the article, which I believe to be especially important - family health. Bayliss and colleagues highlight the significant ways in which the family, including the health and functioning of family members, may affect the causes and consequences of MCCs in individuals (1). This point is important for at least two reasons. First, it underscores the point that a consideration of "health" should not be restricted to the individual with MCCs, the main study outcome for most work in this area. If a family caregiver has chronic obstructive pulmonary disease, it may adversely affect or at least complicate the assistance she may provide in caring for a family member with diabetes and breast cancer. Second, a consideration of family health also suggests that the study of MCCs can lead to new areas of research, practice, and policy. Specifically, it should be possible to expand the unit of investigation and analysis from the "individual" to the "family" itself. The health of the family, then, can be characterized, at least in part, by the total number and severity of multiple chronic conditions facing the family as a unit. A family's health might be characterized, to use our previous example, as including chronic obstructive pulmonary disease, diabetes, and breast cancer. The questions and issues outlined in the Bayliss article (1) and, indeed, the DHHS report itself (2), could be applied to the study of family health. While the clustering of health conditions has been investigated in neighborhoods and communities (4), to my knowledge, it has not been examined systematically in families. The proposed family MCC agenda will be very challenging, e.g., the family itself may change form over time in terms of numbers of people and their relationships to each other, thus complicating longitudinal studies. Despite this and other challenges, a new family MCC research agenda could serve to expand our understanding of the epidemiology of MCCs, associated health disparities, and lead to new programs and policies to enhance the health and well-being of families.

1. Bayliss EA, Bonds DE, Boyd CM et al. Understanding the context of health for persons with multiple chronic conditions: Moving from what is the matter to what matters. Ann Fam Med 12 (3): 260-269.
2. U.S. Department of Health & Human Services. Multiple chronic conditions: A strategic framework. Optimum health and quality of life for individuals with multiple chronic conditions. Washington, DC: U.S. Department of Health & Human Services;2010. Available from: http://www.hhs.gov/ash/initiatives/mcc/mcc_framework.pdf [Last accessed Dec 13, 2013].
3. Smedley BD, Syme SL (Eds.). Promoting health: Intervention strategies from social and behavioral research. Washington DC: National Academy Press, 2001:2.
4. Singer M. Introduction to syndemics: A critical systems approach to public and community health. San Francisco, CA: Jossey-Bass, 2009.

Competing interests: None declared

An important contribution of the paper by Bayliss and colleagues[1] is the comprehensive list of contextual factors it identifies, which potentially affect the care and outcomes of persons with multiple chronic conditions (MCCs). These include: ethnicity, personal resources, such as housing, education and health literacy; the social and cultural context of families; and an array of other socio-demographic, cultural, and economic characteristics of communities. Such a wide-ranging suggested contextual framework stands in sharp contrast to the current paucity of research related to multimorbidity within socio-economically and ethnically diverse populations.[2] Indeed, the US Department of Health and Human Services' strategic framework[3] identifies the need to improve the understanding of the roles of disparities in persons and populations with MCCs. Yet, moving forward to addressing "what matters" for persons and populations facing numerous contextual factors, in tandem with multiple chronic conditions, may require a bold shift in how we classify morbidity.

First, re-thinking the definition of multimorbidity is called for. Multimorbidity is conventionally defined as the presence of two or more chronic illnesses. For ethnic and cultural minority populations, often from lower socioeconomic backgrounds, such a definition underestimates the burden of illness as the assignment of "chronic" diagnoses is, at least in part contingent on access to specialists.[4] Disparities in specialty care exist, and socially deprived populations often present to primary care with a wide-range of ill-defined symptoms and complaints. Thus, limiting the identification of high need populations based only on the number of "chronic" conditions may result in missed opportunities targeting those at highest risk for poor outcomes.

Furthermore, a classification based only on the number of "chronic" conditions may misrepresent need as it fails to account for severity, duration, and interactions between conditions of any type, including acute and time-limited conditions.[5] For example, socially deprived populations with extensive severe, time-limited "non-chronic" conditions may have a similar or even higher level of need than affluent older patients with two or three stable chronic conditions. Similarly, for persons with myriad contextual challenges, having one chronic condition may surpass the needs of those with several chronic conditions.

Thus, expanding the definition of multimorbidity to include the entire spectrum of conditions, whether, acute, recurrent, or of long-term duration, and examining patterns of health states rather than clusters of single diseases, would likely result in a better understanding of the characteristics and outcomes of multi-contextual factors in patients with multiple health needs.

References
1. Bayliss EA, Bonds DE, Boyd CM, et al. Understanding the context of health for persons with multiple chronic conditions: moving from what is the matter to what matters. Ann Fam Med. 2014;12(3):260-269.
2. Shadmi E. Disparities in multiple chronic conditions within populations. Journal of Comorbidity. 2013;3(2):45-50.
3. US Department of Health and Human Services. Multiple Chronic Conditions--A Strategic Framework: Optimum Health and Quality of Life for Individuals with Multiple Chronic Conditions. Washington, DC: US Department of Health and Human Services; 2010.
4. Starfield B, Chang HY, Lemke KW, Weiner JP. Ambulatory specialist use by nonhospitalized patients in us health plans: correlates and consequences. J Ambul Care Manage. 2009;32(3):216-225.
5. Starfield B. The hidden inequity in health care. Int J Equity Health. 2011;10:15.

Competing interests: None declared

What a refreshing article by Elizabeth Bayliss et al, which lays such a strong foundation for the work ahead - understanding the clinical and social complexity that is at the heart of caring for whole people, free-range humans living in families and neighborhoods and cultural tapestries. It makes one-disease-at-a-time RCT's and clinical guidelines seem almost quaint.

The authors rightly emphasize the multiple chronic conditions that can combine to form disease clusters and clinical complexity, as well as the myriad psychosocial and family/neighborhood/cultural contextual factors that generate social complexity.

Still, we must be cautious about generating lists of "conditions" and "factors". The authors acknowledge the complex, non-linear interactions that occur between these "multiple chronic conditions" and the "many contextual factors", but I wonder if that really captures the nuance and the flavor of these interactions. I can count chronic conditions and compute a comorbidity score, but am I holistically capturing the new physiologic equilibrium, the new rebalanced homeostasis, that is keeping this human being alive?

In the same way, it is a huge step forward to include family and neighborhood-level contextual factors in any research designed to improve health outcomes. However, we cannot stop at counting and measuring social "factors, or treating them as lists of independent covariates in a regression equation. Rather, these factors are merely ingredients in the complex (and flavorful!) stew that is borne of whole people in relationship with families and neighborhoods and culture.

Noticing flavors (nuanced interactions and subtleties) is even more important at the community level as we work to achieve optimal and equitable health outcomes for all. Some communities are making great progress toward overcoming racial-ethnic and socioeconomic health disparities - how are they doing it? What are the ingredients and flavors that create a distinctive taste of health equity at the community level? More importantly, what interventions could transform the historically-bitter ingredients and flavors of a high-disparity community into the savory boldness of a high-equity community?

This will require a very different approach, not only to clinical trials but also to community health research. Just as each person is a unique, complex, and integrated biopsychosocial organism, each community is also a unique, complex sociocultural, relational, and socioeconomic ecosystem interfacing with its own unique and complex healthcare system.

Therefore, meaningful research in the "what matters" space will need to test approaches more than interventions. We must standardize experimental approaches that include assessing and respecting and listening first, then customizing interventions to each community's specific challenges and strengths, and finally maintaining a dynamic posture of rapid-cycle improvement that modifies the intervention repeatedly based on timely and granular measurement of critical outcomes. We won't get it right the first time, so the faster we move through failure to finding what works, the faster we will move toward optimal and equitable health outcomes for all -- one community at a time.[1,2]

1: Rust G, Levine RS, Fry-Johnson Y, Baltrus P, Ye J, Mack D. Paths to success: optimal and equitable health outcomes for all. J Health Care Poor Underserved. 2012 May;23(2 Suppl):7-19.
2: Satcher D, Rust G. Achieving health equity in America. Ethn Dis. 2006 Spring;16(2 Suppl 3):S3-8-13.

Competing interests: None declared

The authors provide an ambitious agenda and challenges for MCC stakeholders. Outside of MCC-specific research, there may be good examples of contextual factor analysis. PACE evaluations by Mukamel and colleagues (2007 Milbank Q; Geront 2006) may be most relevant, but there is also important work on neighborhood effects by Kathleen Cagney and others that may represent possible starting points in applying lessons learned from non-MCC populations to MCC populations.

Competing interests: ?? I currently collaborate with Dr. Bayliss

This is an excellent piece on the levels at which work is needed to improve population health. The recognition of this systematic work is important alongside how we can approach this strategically. We need to continue the good work of promoting evidence-based self-management, but we also need to see how we can help this with different policy levers. This goes beyond payment- how can we effect the change needed on a scale larger than 40-60 people over an 8 week time span. This is a non-partisan question and the time is ripe for change.

Competing interests: ?? None declared

Older women are more likely to have 2 or more chronic conditions than men of the same age. Older women and men have differing proportions of chronic diseases and disorders. Medicare and other health insurance is focused on acute care so for the typical 75 year old man and a 75 year old women hospitalized for an acute heart condition are treated the same. In his case he has an acute heart condition but no other disease or disability. After the specified length of hospital stay for his condition he is sent home where his wife can take care of him and make sure that he takes his heart medication. He is up and around soon, can get to the bathroom safely on his own. If the house or apartment has stairs, she can help him go up and down. If a more serious problem arises, they are likely to have the economic resources to hire help in the home. In contrast, for her she is hospitalized for an acute heart condition, but she is also likely to have at least one chronic condition, typically arthritis. She lives alone and when she is sent home arrangements have to be made with neighbors and/or family members who often do not live nearby. She needs help getting to the bathroom because of her arthritis. Without help she risks falling. If there are stairs, the risk is even greater. There is no one in the home to make sure she takes her heart medication regularly. And she is not likely to have the economic resources to hire help.

The sex ratios of the elderly population become more and more skewed as the population examined is higher in age. This skewed gender ratio impacts other important factors such as the probability of living alone, especially with a disability, and the risk of poverty. The gender neutral approach is incomplete in the description of the situation of older adults as well as the process of aging.

Some of the changes reported for older adults between the ages 60-69 and the ages 80-89 may not be due to aging itself, but due to the population of older adults becoming predominantly female with increasing age. Inattention to this simple distinction can and does distort an understanding of the process of aging. In other words, what is attributed to aging, may more accurately be attributed to the higher mortality rates among men. If men were to live as long as women the risk of reported depression would be much lower and the risk of suicide much higher.

The risk of poverty is higher for older age groups but this result is due, in part, to the fact that older women who live alone predominate among the very old. One factor in lowering the poverty risk for men is that they usually have the advantage of sharing household expenses with a spouse. A gender analysis is necessary for policies and programs to be effectively developed and implemented. It must begin with data disaggregated by gender.

Competing interests: ?? None declared