Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status

G Model PEC 5782 No. of Pages 8

Patient Education and Counseling xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou

Research Paper

Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status Miriam Komaromya,f,* , Erin Fanning Maddenb , Andrea Zurawskia , Summers Kalishmana,c , Kristin Barkerd , Patricia O’Sullivane , Martin Juradoa , Sanjeev Aroraa,f a

The ECHO Institute at the University of New Mexico Health Sciences Center, Albuquerque, NM, USA Department of Sociology, University of Texas, San Antonio, TX, USA c Department of Family and Community Medicine, Office of Education at the University of New Mexico Health Sciences Center, Albuquerque, NM, USA d Department of Sociology, University of New Mexico, Albuquerque, NM, USA e The University of California, San Francisco CA, USA f Department of Internal Medicine, University of New Mexico Health Sciences Center, Albuquerque, NM, USA b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 12 April 2017 Received in revised form 6 August 2017 Accepted 30 August 2017

Objective: Elicit patients’ perceptions of factors that facilitate their engagement in care Methods: In-depth interviews with 20 adult Medicaid patients who had complex health problems, frequent hospitalizations/emergency department use, and who were enrolled in an intensive, teambased care program designed to address medical, behavioral, and social needs. Results: Prior to engaging in the program, participants described weak relationships with primary care providers, frequent hospitalizations and emergency visits, poor adherence to medications and severe social barriers to care. After participating in the program, participants identified key factors that enabled them to develop trust and engage with care including: availability for extended intensive interactions, a non-judgmental approach, addressing patients' material needs, and providing social contact for isolated patients. After developing relationships with their care team, participants described changes such as sustained interactions with their primary care team and incremental improvements in health behaviors. Conclusion: These findings illuminate factors promoting “contingent engagement” for low socioeconomic status patients with complex health problems, which allow them to become proactive in ways commensurate with their circumstances, and offers insights for designing interventions to improve patient outcomes. Practice implications: For these patients, engagement is contingent on healthcare providers’ efforts to develop trust and address patients’ material needs. © 2017 Elsevier B.V. All rights reserved.

Keywords: Patient engagement Contingent engagement Complex care Primary healthcare Medicaid Traditionally underserved Socioeconomic status Team based care ECHO model Hospitalization

1. Introduction Individuals with complex health conditions, including physical, mental, and substance use problems, often lack primary care that meets their healthcare needs [1]. When they do not receive this care, they frequently (and repeatedly) end up in emergency departments (EDs) and experience avoidable hospitalizations [2,3]. Their chronic medical problems progress, and they experience earlier deaths compared to individuals with access to adequate care [4]. Patients with complex health problems account

* Corresponding author at: ECHO Institute, 1 University of New Mexico, MSC 07 4245, Albuquerque, NM 87131, USA. E-mail address: [email protected] (M. Komaromy).

for substantial healthcare spending [5]. Many of these patients also experience adverse social conditions, such as poverty, low-literacy, and homelessness that make it difficult to access and benefit from care [6]. Because of these barriers and high rates of behavioral health disorders, these patients are less likely to engage in healthpromoting actions, including establishing ties to the primary healthcare system, than are other patients. Juxtapose this situation to the mounting body of research demonstrating the importance of patient engagement for improving health outcomes and healthcare experiences [7–9]. There are different definitions of patient engagement, but the construct generally includes behaviors and cognitive-emotional states reflective of patients’ pro-active stance vis-à-vis their health and healthcare [10–12]. These include “understanding and acting on health information (health literacy), working together with

http://dx.doi.org/10.1016/j.pec.2017.08.019 0738-3991/© 2017 Elsevier B.V. All rights reserved.

Please cite this article in press as: M. Komaromy, et al., Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status, Patient Educ Couns (2017), http://dx.doi.org/10.1016/j.pec.2017.08.019

G Model PEC 5782 No. of Pages 8

2

M. Komaromy et al. / Patient Education and Counseling xxx (2017) xxx–xxx

clinicians to select appropriate treatments or management options (shared decision making), and providing feedback on healthcare processes and outcomes (quality improvement)” [13]. Patient activation, defined as “having the knowledge, skill, and confidence to manage one’s health and healthcare” [9], is a prevailing framework of patient engagement. The self-reported Patient Activation Measure (PAM) quantifies aspects of patient engagement. Patients with high PAM scores adhere to medications, are hospitalized or use EDs less, and generate lower healthcare costs [14]. There are reasons to believe existing conceptualizations and measurements of patient engagement in general, and patient activation as one aspect of patient engagement in particular [15], are more salient for some patient groups than for others [16,17]. For example, on average patients with low SES have lower PAM scores [14]. This is unsurprising given some PAM items, such as “I am confident that I can follow through on recommendations my healthcare provider makes, such as changing my diet or doing regular exercise.” Complying with provider recommendations requires resources that are unequally distributed by SES. This and other indicators of patient engagement and activation may also be less salient to individuals facing multiple, complex health problems, including behavioral disorders, which can collectively constrain proactivity. Moreover, there is evidence that low-SES patients are disinclined to participate in shared medical decisionmaking; and yet doing so is considered demonstrative of patient engagement [18,19]. On one hand there is compelling evidence that patient engagement improves myriad health outcomes and the quality of care. On the other hand, as it is often conceptualized and measured, patient engagement may not be fully applicable to patients who stand to benefit the most from being engaged in some capacity. Leading scholars in the area fully acknowledge these tensions. Whereas existing efforts to enhance engagement largely assume a ‘one size fits all’ approach, there is growing awareness that the ‘next generation’ of engagement promotion will require targeted interventions [20,21] customized for different

patient populations in an effort to “meet the patient where they are” [17]. In this paper we answer the question, “What do we learn about patient engagement from listening to a population of hard-toreach patients with complex health problems?” Based on interviews, we examine the experience of engagement among low-SES patients with multiple comorbidities as they participate in an intensive team-based primary care program. We characterize engagement among this marginalized patient population and outline aspects of care enabling activation in ways commensurate with patient circumstances. We call this contingent engagement, which we define as actions and attitudes that are conditional on high levels of care and support that incrementally enhances health management and the benefits of healthcare. We offer insights into the mechanisms that promote contingent engagement and concomitant improvements in health and healthcare outcomes. 2. Methods This study investigates the experiences of patients and their close personal contacts with an intensive primary care program for low-income patients with multiple diagnoses called ECHO Care. The patient experience is increasingly recognized as important in the engagement literature and thus the use of in-depth semistructured interviews is appropriate for examining the meaning of and processes behind patient engagement [22]. This research was approved by the University of New Mexico Human Research Review Committee, HRRC #12-617. 2.1. Intervention ECHO Care (henceforth ‘the program’) is an intervention designed to give individuals with complex medical, behavioral and social needs improved access to outpatient care. Patients are referred to the program by inpatient and outpatient providers, Medicaid care coordinators, and social service agencies. The program uses an outpatient intensive team (OIT) that provides

Fig. 1. The ECHO Care Model. (ECHO: Extension for Community Healthcare Outcomes) Outpatient Intensive Team = OIT (referred to as “care team”) CHW = Community Health Worker NP = Nurse Practitioner PA = Physician Assistant

Please cite this article in press as: M. Komaromy, et al., Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status, Patient Educ Couns (2017), http://dx.doi.org/10.1016/j.pec.2017.08.019

G Model PEC 5782 No. of Pages 8

M. Komaromy et al. / Patient Education and Counseling xxx (2017) xxx–xxx

primary care services in the office or the patient’s home. OITs, or care teams, are comprised of a nurse practitioner or physician assistant, a nurse, two community health workers (CHWs), and a counselor. They serve up to 130 patients, allowing a high level of individualized care (See Fig. 1). Patients have direct access to the team via phone and access to care outside of scheduled visits, including after-hours and weekends. Care teams have access to specialty consultation through weekly videoconferencing, in which they join a dozen physical and mental health specialists, along with the other ECHO Care teams. Weekly teleECHO videoconferences are designed to foster case-based learning [23,24]. 2.2. Recruitment Twenty-six patients were selected from two of six ECHO Care clinics, including one urban and one rural site. Four of these patients were excluded due to incarceration or cognitive impairment, and two refused to participate. The twenty patients in our sample generally reflect the medical and social characteristics of the population of patients served by the program (See Table 1). Of particular note, 80% have chronic substance use and/or serious mental illness. All participants meet program eligibility criteria, which include:  Medicaid enrollee   2 chronic conditions  High utilization, defined as 3 ED visits in the past 6 months, or 2 hospitalizations in the past year. During interviews patients were asked to nominate close friends and family to be interviewed about the patient’s experiences. Ten contacts agreed to be interviewed. Five patients did not nominate a contact and five nominated contacts refused participation. Contact interviews provide a type of data

3

corroboration [25]. Recruitment ceased when saturation was achieved vis-à-vis capturing the features of patient engagement presented in this paper [22,26]. 2.3. Data collection and analysis Interviews with patients and contacts took place in their homes. Interviews were conducted by five trained interviewers, lasted 30– 90 min, were recorded with a digital device, and were professionally transcribed. The patient interview guide included openended questions eliciting patients’ perception of their health and healthcare before and after enrollment in the program. Patient contacts were asked to assess the patient’s health and healthcare. Both patients and contacts are referred to by pseudonym. Our analysis followed Charmaz’s [27] constructivist grounded theory methodology, which emphasizes building theory from data [28]. Our research team used NVivo 10 software [29] to code the interview transcriptions. First, 6 members of the research team collaboratively open-coded several interviews to develop codes describing pieces of data in abstract terms [30]. Next, the team created a code list in which we defined and arranged the most frequent and substantively relevant codes into a hierarchical web describing patient relationships to healthcare [27]. Interviews were then double-coded, which resulted in robust coding concurrency [30]. Finally, we “memoed,” or wrote theoretical understandings of the coded interviews [27], leading to the interpretations of the data presented in this paper. 3. Findings 3.1. Pre-engagement High healthcare utilization was a prerequisite for eligibility for the program, and is common among patients with complex healthcare problems [31,32]. Prior to enrollment participants had

Table 1 Population and sample characteristics. All data self-reported by patients except problem type, which is extracted from Medicaid claims.

Gender M F Av. Age (years) Race White American Indian Black Asian American Other response Ethnicity Hispanic Non-Hispanic Education less than HS diploma or less Some college or higher Income Source Full or part time employment Not employed Housing problem Feel afraid of partner or other family Self-rated health Excellent or very good Good Fair Poor Problem type (not self-reported) Medical Behavioral health (mental health and substance use disorders) Social barriers/needs

All patients ever enrolled in ECHO Care (n = 767)

Sample patients (n = 20)

50% 50% 47

45% 55% 50

64% 4% 4% 0% 28%

85% 5% 10% – –

68% 30%

55% 45%

69% 29%

45% 40%

5% 95% 43% 19%

10% 90% 20% 20%

3% 19% 39% 39%

0% 25% 40% 25%

97% 84% 67%

100% 80% 60%

Please cite this article in press as: M. Komaromy, et al., Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status, Patient Educ Couns (2017), http://dx.doi.org/10.1016/j.pec.2017.08.019

G Model PEC 5782 No. of Pages 8

4

M. Komaromy et al. / Patient Education and Counseling xxx (2017) xxx–xxx

little engagement with a regular provider but numerous ED visits and hospitalizations. Some patients, including Pilar, explained the relationship between their underutilization of primary care and the overutilization of hospital care as the outcome of ineffective communication with providers who they felt did not really care about them. “With my previous [doctor] I’d go in with my pancreas, but he never offered to tell me, ‘you need to go to counseling.’ He would just see me, give me my medication, and tell me I’m going to be sick again if I continue to drink. That’s about it, but not ever sat down and talked to me . . . [I’d] go to the hospital every week, but it just costs money. And it’s for anxiety [over] my pancreas . . . Whenever I get scared, I think the worst, so that I go right to the hospital” (Pilar) Patients also recalled missing outpatient appointments due to anxiety or depression, addiction relapse, and deterioration of physical health. Patients widely reported non-adherence to medications. Patients attributed their inability to adhere to medication recommendations to complex dosage instructions, confusion over the purpose of a drug, and substance use. The intersection of SES and chronic illness poses additional challenges to engagement. Because participation in Medicaid is a prerequisite for ECHO Care eligibility, the patients are low-income and describe financial challenges to engagement such as difficulties adhering to dietary restrictions on food stamps, lack of transportation, or worries about paying for housing and utilities. In general the pre-engagement period is a time when patients describe health-related actions and attitudes, as well as relationships with healthcare, as sub-optimal. This is largely because they face many obstacles to engagement and encounter a healthcare system that is ill-equipped to address their complex needs [3,33,34]. 3.2. Mechanisms increasing engagement Patients’ most commonly described reason for increased engagement is development of a strong rapport with team members. High levels of interaction with team members are described as crucial to rapport and in turn engagement. Interactions take the form of home visits, extended clinic visits, after-hours accessibility, and follow-ups during and after hospitalizations. Keith’s wife explains the importance of home visits for improving medication adherence, “Before he would say he takes the medicine, but because nobody was following him . . . he thought he did, but he didn't. Now [the team] checks in once a week. . . . They come to the home to show him how to do things instead of just telling him and sending him home.” Echoing his wife, Keith explains the value of home visits: “I’m getting more quality care than by just going to the hospital. [The team] keeps me from going to the hospitals . . . and helps with taking my medications . . . and that’s been very helpful. It makes me feel better, much healthier.” Patients describe how attentive interaction with team members in longer appointments contributes to feeling listened to: “[Nurse practitioner] listens to me... He explains everything . . . he makes time for me” (Teresa). “It makes me feel like I’m the only one [physician assistant] sees, like he’s paying attention just to me” (Carol). Dennis explains in detail how visit length affects his engagement, “With my diabetes they are really helping me try to get [my blood sugar] down. It’s hard because, being diabetic . . . I’m struggling to understand it, but I’m understanding it better now that I’ve been in ECHO Care. The visits take an hour and they

pretty much take their time . . . When I walk away, I feel satisfied that I’m being cared for. I feel important. Before, doctors would say, ‘yeah, he’s another number.”' Offering primary care services outside of the office and allowing extended visits results in deeper patient understanding and builds rapport with their care team. Team member willingness to help patients with resource issues and social support also fosters rapport. With respect to the former, Rebecca explains the logistical and emotional benefits received when her team’s CHW assisted with her disability application, “She was real vigilant following up with them and saying, ‘what stage is it in?’ . . . She did the whole round up of all the paperwork and it went through . . . I feel a lot less stress in my life . . . It’s a money thing, and being on disability has taken pressure off me because . . . I feel like I’m contributing again, and it boosted my self-esteem.” Alma described the emotional benefits of the team’s support in the face of social isolation: “It makes me feel good because I can count on them and that’s really hard when you don't have nobody to count on” and “It makes me feel safe because I know I could call them, and they’ll answer.” Patients also emphasized non-judgmental dispositions among team members as contributing to their feelings of acceptance. Jose struggles with substance use and explains “they’d never judged me . . . and they didn’t give up on me.” Pilar recalls the nurse’s reaction to her relapse: “Don’t be afraid. Everybody messes up . . . . [we’ll] just try and get you back on track.” Jose explains the importance of non-judgmental support at crucial times: “[The team] went up to bat for me when it counted . . . I violated [probation] once. They kept me [in jail] like 48 days . . . I asked my mom to get with ECHO staff to get letters about my progress and I believe those letters were part of the big reason why I’m not sitting in prison . . . I feel like I have someplace to go if shit hits the fan.” Patients and contacts describe forging emotional connections with the team, using language related to trust and safety. “I don't trust doctors . . . but I got a great team because they’re real” (Rhonda). “I know I could call and they’ll answer, and that makes me feel safe . . . and that’s important to me because I pass out a lot” (Alma). Care teams are even compared to familial relationships, “They’re not strangers, you know, they’re family” (Carol). Carol’s partner explains the problem of high turnover with previous providers, “when she was getting used to one person, then all of a sudden, ‘oh, I’m going to be leaving.’ . . . But now . . . [nurse practitioner] says he’s not leaving and I think she feels more secure.” Patients connect the strong relationship with the primary care team with increased confidence and motivation to improve health: “I have a team I can depend on . . . it makes me feel confident” (Ricardo). “Every time we meet, we’re strengthening that bond . . . I would feel very uncomfortable if I decided to sit down and drink, and if [CHW] walked in . . . I would probably feel like I wish I had a hole to crawl off into” (Jose). “Before, I was on drugs and all kinds of stuff, and then once I got into this program, I just quit the stuff because I have support” (Javier). Strengthened provider-patient relationships support the forms of engagement described in the next section. 3.3. Contingent engagement Engagement among these patients includes skills, behaviors, and relationships that may be considered “low” or “pre-engagement” traits among healthier and more affluent groups. We find engagement, as described by participants, is characterized by

Please cite this article in press as: M. Komaromy, et al., Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status, Patient Educ Couns (2017), http://dx.doi.org/10.1016/j.pec.2017.08.019

G Model PEC 5782 No. of Pages 8

M. Komaromy et al. / Patient Education and Counseling xxx (2017) xxx–xxx

reliance on intensive primary care efforts, and by incremental, non-linear engagement trajectories. Patients describe significant improvement in understanding chronic conditions, and more confidence in their ability to work with care providers to evaluate and manage symptoms. Keith explains, “I’m getting more quality care than by going to the hospital . . . They’re quick and they respond to my call even after hours.” Keith’s wife agrees, “He’s not wondering if he should go to the hospital when something is going on . . . He knows [to] call first, and then they tell him what to do.” Many patients in the sample replaced ED overuse by turning to the team’s advice for assessing health issue severity. While this may seem highly dependent, given the complexity of these patients’ conditions, it is unrealistic to expect sustained improvement without ongoing support. Engagement is displayed when patients use the team’s advice to improve health behaviors. This is especially apparent in medication adherence improvements, but like reducing ED use, adherence is contingent upon continued support. Manuel, who has multiple chronic conditions including diabetes, explains, “The team comes here once a week and they make all my appointments, they fix me up with transportation, and at the pharmacy they get the medication for me . . . [The team] has been working on getting my sugars down . . . they’re coming down compared to what they used to be, I feel better.” Manuel’s brother explains these improvements are slow and incremental in nature, “I’ve noticed there’s been changes . . . It’s not a complete big change, it’s little by little. It took him a long time to get that way, so I know it’s going to take a long time to get him better.” Some patients displayed conflicting health behaviors. Javier identifies as someone with ongoing substance use problems, but also discusses health-promoting changes, “I used to just go on an [alcohol] binge for a week, not eat anything, but now I’m eating . . . [CHW] comes to check on me once a week. They just talk to me about eating habits and how I should change them . . . he tells me to eat a lot of greens and bake . . . but I haven't got into that yet . . . they suggested I stop drinking. I haven't yet . . . But I recognized I wouldn't feel sick in the morning [if I ate].” Javier illustrates that some patients maintain dangerous behaviors while improving others. Often, however, patients make progress on all health behaviors. Fernando illustrates this, “[The team] helps with staying healthy, eating right, and making me think positive. This time last year I was drinking my life away . . . ” He explains that since working with the team “I completely stopped drinking. I haven't had a sip.” Health behavior improvements may or may not follow a linear engagement path. 4. Discussion and conclusion 4.1. Discussion Providers and policy makers have come to emphasize patient engagement as a primary means by which to improve the therapeutic and cost effectiveness of healthcare [8,9]. Yet, 20% of all healthcare expenditures are allocated to 1% of the population – so-called “super-utilizers” [3,8,35,36] – comprised of individuals with complex physical, behavioral, and social needs that can easily thwart their health engagement [17]. We describe the challenges and successes of patients in this 1% and chart their engagement trajectories. This patient population represents a group that does not fit well into existing notions of patient engagement, including the popular activation [17,37,38] and self-management [39]

5

frameworks. Although these frameworks are highly valuable, as conceptualized they are less applicable to patients like those served by ECHO Care (for a notable exception see [40]). By studying how these patients experience a program designed to address their needs, we offer an addendum to engagement concepts for patients who experience a web of severe and mutually reinforcing health and social challenges [41,42]. 4.2. Conclusions We summarize our findings in Fig. 2, which depicts these patients’ engagement trajectory. We find that patients begin with low engagement and trust, which fuel pre-engagement characteristics. Through efforts such as addressing material needs and being available to patients, an emotional connection between the team and the patient is built and fosters increased trust, which then translates into post-engagement behaviors like sustained interaction with healthcare. While some may critique such an intensive primary care program as enabling or fostering extreme dependence, our findings indicate team efforts do result in improved engagement. Additionally, we found that ECHO Care enrollment was associated with lower ED use and hospitalizations, which decreased overall healthcare costs [43]. Because patients with complex problems are particularly challenging to treat, both the process and the goals of engagement must be adjusted to realistic expectations. We suggest that the notion of contingent engagement captures both these important qualifications. One way in which we found engagement to be contingent is that engagement changes are realized when elements of unmet needs are addressed, including basics such as food, shelter, and financial security. With these needs addressed, patients describe greater focus on symptom management, medication adherence, and keeping medical appointments. As noted by others [17], addressing basic material needs as an essential step to promoting engagement was a common pattern in our data. Here we also draw and build on insights from the transtheoretical model [44,45]: lowincome patients with complex health problems may be unable to move beyond the contemplation stage of change if their basic needs are unaddressed. A second way contingent engagement was observed is that engagement improvements are dependent upon intensive primary care support. Engagement skills described in the concept of selfmanagement, emphasize patients developing independent control of their health and healthcare. Skills such as “resource utilization”—wherein patients use multiple health information sources [39] – are almost completely absent in our sample. Patients rely heavily on teams for all health and healthcare information almost without exception. Patients do not cite getting multiple expert opinions or seeking internet advice. The exclusive and intense reliance on care teams illustrates how these patients have fundamental differences in the ways they manifest engagement compared to patients with higher SES and fewer or less serious illnesses. Other researchers have likewise noted that intensive support may be needed for interventions targeting the least engaged patient populations [20,21]. Third, engagement improvement trajectories are also contingent in that they are often non-linear. Patients who oscillate in and out of substance use behaviors while also improving diet, exercise, or appropriate healthcare utilization challenge the image of a straightforward engagement improvement path. Recognition of this non-linearity is also described in a study examining “patient engagement at the margins” [17]. Safety-net providers, understanding that engagement would necessarily wax and wane with patients’ unpredictably precarious lives, calibrate their thinking about engagement accordingly.

Please cite this article in press as: M. Komaromy, et al., Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status, Patient Educ Couns (2017), http://dx.doi.org/10.1016/j.pec.2017.08.019

G Model PEC 5782 No. of Pages 8

6

M. Komaromy et al. / Patient Education and Counseling xxx (2017) xxx–xxx

Fig. 2. Contingent engagement within ECHO Care complex care program. Note: Although progress was not always linear, when teams addressed material needs, provided social support, were available for extended interactions, and took a nonjudgmental approach, patients generally moved in the direction of increased engagement with care.

In sum, insofar as engagement is contingent on structural adaptions that are external to individual patients, our findings contribute to thinking about patient engagement in ways that take into consideration contextual factors and the structure of healthcare systems [11,15]. Such thinking is essential in that it eschews the notion that marginalized patients are fundamentally deficient when it comes to acting on their own behalf. As for limitations of our study, this research used a small number of subjects and the findings should be confirmed with a larger sample. Although we did not have any previous contact with patients in our sample, they knew the research was being

conducted by ECHO. Even as our open-ended questions avoided leading them to do so, interviewees may have overstated the intervention’s impact in an effort to be polite [22]. Relatedly, we have been mindful of the ways in which our positions as researchers, with varying affiliations to the program under study (See Table 2), have informed our choices and interpretations [46]. 4.3. Practical implications Because engagement or activation appears to underlie “most health related behaviors” [20] and is a strong predictor of positive

Table 2 Additional Methodological Details. Sample recruitment and method of approach

Informed Consent Interviewer characteristics

Additional research team members’ qualifications and positions

Interview guide

Data Coding and Analysis

Participants were recruited with up to three telephone calls, followed by a home visit by a research team member. During the interview patients identified close personal friends and family for the research team to interview about the patient’s progress. These contacts were recruited with up to 3 phone calls, but home visits were only made if contacts agreed to be interviewed. All participants received a $30 gift card for participation and were only interviewed once. Verbal and written consent procedures were followed prior to conducting patient and contact interviews. Two authors and 3 additional research assistants conducted the interviews. The interviewing team consisted of 4 women and 1 man. Three interviewers are Hispanic and 2 are non-Hispanic white. The methodologist leading the team holds a PhD and an MS in sociology, 2 team members hold Bachelor’s degrees, and 2 are undergraduate students. One member of the interview team worked for the ECHO Care program and was employed full time by Project ECHO (the parent organization to ECHO Care), and the remaining 4 interviewers were employed part time for the purposes of this research only. Additional members of the research team included 2 full-time employees of Project ECHO (MDs), 1 part time volunteer from Project ECHO (PhD), 1 research consultant (PhD) from the Sociology Department at University of New Mexico, and 1 research consultant (EdD) from University of California, San Francisco. Including several researchers with relatively inconsequential ties to Project ECHO enhanced our ability to reflect on, address, and where possible avert partiality. Two interview guides were used in the study. Initial guide development drew on themes associated with patient engagement and activation, but these sensitizing concepts were not directly inquired about in either guide. The patient interview guide used open-ended questions eliciting patients’ perception of their health and healthcare before and after ECHO Care enrollment. The contact guide asked patient contacts to evaluate the patient’s health and healthcare from their perspective. We made adjustments as data were collected to focus in on elements of patient engagement that required further elaboration. The team worked together on open coding, which entails generating a list possible themes occurring in the data. The same team refined open codes into a “focused code” tree, or a constellation of related major themes. Two researchers doubled-coded the interviews using NVivo 10 software. Theoretical interpretations of the data were written up in memos and all team members read and discussed memos in order to identify significant themes.

Please cite this article in press as: M. Komaromy, et al., Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status, Patient Educ Couns (2017), http://dx.doi.org/10.1016/j.pec.2017.08.019

G Model PEC 5782 No. of Pages 8

M. Komaromy et al. / Patient Education and Counseling xxx (2017) xxx–xxx

health outcomes, it is urgent to understand how to promote engagement across diverse patient groups [9]. We offer several recommendations for improving engagement among low-income complex patients. We found that the primary mechanisms promoting engagement were strong rapport between patients and healthcare teams and the forging of trust. Rapport and trust are built by providing frequent and extended interaction, addressing material needs, offering social support and maintaining a nonjudgmental approach. Of course there are institutional barriers to the implementation of these suggestions. Engagement changes like reduced ED use and better chronic illness management offer compelling justifications for healthcare organizations to consider extending encounter times, offering home visits, and providing comprehensive social service support to complex patients. These results have implications for primary care systems attempting to address the needs of patients who have limited resources and complex healthcare problems. The most important is the need to integrate behavioral, medical and social interventions. When these foundations of care are not integrated, the effectiveness of any isolated intervention is undercut. Second, this integrated approach requires a team of providers including those beyond the conventional medical workforce (e.g., counselors and CHWs). Third, it is critical to recognize that these patients are typically distrustful of a healthcare system they feel has treated them with disregard. Overcoming distrust using the team strategies outlined above is essential for engagement. Finally, finding ways to foster a non-judgmental approach among care team members is vital. These patients have high rates of substance use, mental health disorders and poor self-care, and are accustomed to being judged harshly. The suspension of such judgement to allow the forging of a “therapeutic alliance” [21] between patients and care teams can promote patient engagement and the resulting improvements in health and healthcare outcomes. Funding The project described was supported by Grant Number 1C1CMS330973-01-00 from the U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services. The content of this paper is solely the responsibility of the authors and does not necessarily represent the official views of the U.S. Department of Health and Human Services or any of its agencies. Research disclaimer The research presented was conducted by the awardee. Findings may or may not be consistent with or confirmed by the findings of the independent evaluation contractor. Acknowledgements Thanks to Judy Bartlett for assistance in developing the manuscript, Xi Sun for data analysis, and Andrea Bradford for editorial assistances. References [1] L.J. Harris, I. Graetz, P.S.B. Podila, J. Wan, T.M. Waters, J.E. Bailey, Characteristics of hospital and emergency care super-utilizers with multiple chronic conditions, J. Emerg. Med. 50 (2016) e203–214, doi:http://dx.doi.org/10.1016/j. jemermed.2015.09.002. [2] H.J. Jiang, C.A. Russo, M.L. Barrett, Nationwide frequency and costs of potentially preventable hospitalizations, 2006: statistical brief #72, Healthc. Cost Util. Proj. HCUP Stat. Briefs, Agency for Healthcare Research and Quality (US), Rockville (MD), 2006 http://www.ncbi.nlm.nih.gov/books/NBK53971/ (Accessed 18 July 2016).

7

[3] D. Hasselman, Super-Utilizer Summit. https://www.chcs.org/resource/superutilizer-summit-common-themes-from-innovative-complex-caremanagement-programs/ 2013 (Accessed 14 July, 2017). [4] E.R. Cheng, D.A. Kindig, Disparities in premature mortality between high- and low-income US counties, Prev. Chronic. Dis. 9 (2012) E75. [5] Informational Bulletin, Targeting Medicaid Super-Utilizers to Decrease Costs and Improve Quality, Centers for Medicaid and Medicare Services, Baltimore, MD, 2013. https://www.medicaid.gov/federal-policy-guidance/downloads/ cib-07-24-2013.pdf. [6] F. Sirotich, A. Durbin, J. Durbin, Examining the need profiles of patients with multiple emergency department visits for mental health reasons: a crosssectional study, Soc. Psychiatry Psychiatr. Epidemiol. 51 (2016) 777–786, doi: http://dx.doi.org/10.1007/s00127-016-1188-5. [7] J.H. Hibbard, P.J. Cunningham, How engaged are consumers in their health and healthcare, and why does it matter, Res. Brief 8 (2008) 1–9. [8] D.M. Cosgrove, M. Fisher, P. Gabow, G. Gottlieb, G.C. Halvorson, B.C. James, G.S. Kaplan, J.B. Perlin, R. Petzel, G.D. Steele, J.S. Toussaint, Ten strategies to lower costs, improve quality, and engage patients: the view from leading health system CEOs, Health Aff. (Millwood) 32 (2013) 321–327, doi:http://dx.doi.org/ 10.1377/hlthaff.2012.1074. [9] J. Protheroe, A. Rogers, A.P. Kennedy, W. Macdonald, V. Lee, Promoting patient engagement with self-management support information: a qualitative metasynthesis of processes influencing uptake, Implement. Sci. 3 (2008), doi:http:// dx.doi.org/10.1186/1748-5908-3-44. [10] G. Graffigna, S. Barello, A. Bonanomi, G. Riva, Factors affecting patients’ online health information-seeking behaviours: the role of the patient health engagement (PHE) model, Patient Educ. Couns. (2017), doi:http://dx.doi.org/ 10.1016/j.pec.2017.05.033. [11] A. Finset, Patient participation, engagement and activation: increased emphasis on the role of patients in healthcare, Patient Educ. Couns. 100 (2017) 1245–1246, doi:http://dx.doi.org/10.1016/j.pec.2017.05.011. [12] Center for Advancing Health, Engagement, (2014). http://www.cfah.org/ engagement/ (Accessed 18 July 2016). [13] A. Coulter, Patient engagement–what works? J. Ambulatory Care Manage. 35 (2012) 80–89, doi:http://dx.doi.org/10.1097/JAC.0b013e318249e0fd. [14] J. Greene, J.H. Hibbard, Why does patient activation matter? An examination of the relationships between patient activation and health-related outcomes, J. Gen. Intern. Med. 27 (2012) 520–526, doi:http://dx.doi.org/10.1007/s11606011-1931-2. [15] K.L. Carman, P. Dardess, M. Maurer, S. Sofaer, K. Adams, C. Bechtel, J. Sweeney, Patient and family engagement: a framework for understanding the elements and developing interventions and policies, Health Aff. (Millwood) 32 (2013) 223–231, doi:http://dx.doi.org/10.1377/hlthaff.2012.1133. [16] T.M. Napoles, N.J. Burke, J.K. Shim, E. Davis, D. Moskowitz, I.H. Yen, Assessing patient activation among high-need, high-cost patients in urban safety net care settings, J. Urban Health (2017), doi:http://dx.doi.org/10.1007/s11524017-0159-9. [17] M.D. Fleming, J.K. Shim, I.H. Yen, A. Thompson-Lastad, S. Rubin, M. Van Natta, N.J. Burke, Patient engagement at the margins: healthcare providers’ assessments of engagement and the structural determinants of health in the safety-net, Soc. Sci. Med. 183 (2017) 11–18, doi:http://dx.doi.org/10.1016/j. socscimed.2017.04.028. [18] S. Garfield, F. Smith, S.A. Francis, C. Chalmers, Can patients’ preferences for involvement in decision-making regarding the use of medicines be predicted? Patient Educ. Couns. 66 (2007) 361–367, doi:http://dx.doi.org/10.1016/j. pec.2007.01.012. [19] R. Say, M. Murtagh, R. Thomson, Patients’ preference for involvement in medical decision making: a narrative review, Patient Educ. Couns. 60 (2006) 102–114, doi:http://dx.doi.org/10.1016/j.pec.2005.02.003. [20] J.H. Hibbard, Patient activation and the use of information to support informed health decisions, Patient Educ. Couns. 100 (2017) 5–7, doi:http://dx.doi.org/ 10.1016/j.pec.2016.07.006. [21] T. Higgins, E. Larson, R. Schnall, Unraveling the meaning of patient engagement: a concept analysis, Patient Educ. Couns. 100 (2017) 30–36, doi:http://dx.doi.org/10.1016/j.pec.2016.09.002. [22] P. Salmon, Assessing the quality of qualitative research, Patient Educ. Couns. 90 (2013) 1–3, doi:http://dx.doi.org/10.1016/j.pec.2012.11.018. [23] M. Hostetter, S. Klein, D. McCarthy, Project ECHO’s complex care initiative: building capacity to help superutilizers in underserved communities, Common Fund (2016) http://www.commonwealthfund.org/publications/ case-studies/2016/aug/project-echo (Accessed 22 February 2017).. [24] S. Arora, K. Thornton, G. Murata, P. Deming, S. Kalishman, D. Dion, B. Parish, T. Burke, W. Pak, J. Dunkelberg, M. Kistin, J. Brown, S. Jenkusky, M. Komaromy, C. Qualls, Outcomes of treatment for hepatitis C virus infection by primary care providers, N. Engl. J. Med. 364 (2011) 2199–2207, doi:http://dx.doi.org/ 10.1056/NEJMoa1009370. [25] L.C. Sobell, T.P. Ellingstad, M.B. Sobell, Natural recovery from alcohol and drug problems: methodological review of the research with suggestions for future directions, Addiction 95 (2000) 749–764, doi:http://dx.doi.org/10.1046/j.13600443.2000.95574911.x. [26] A. Finset, Qualitative methods in communication and patient education research, Patient Educ. Couns. 73 (2008) 1–2, doi:http://dx.doi.org/10.1016/j. pec.2008.08.004. [27] K. Charmaz, Constructing Grounded Theory, SAGE, 2014. [28] A. Tong, P. Sainsbury, J. Craig, Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups, Int. J.

Please cite this article in press as: M. Komaromy, et al., Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status, Patient Educ Couns (2017), http://dx.doi.org/10.1016/j.pec.2017.08.019

G Model PEC 5782 No. of Pages 8

8

M. Komaromy et al. / Patient Education and Counseling xxx (2017) xxx–xxx

[29] [30]

[31] [32]

[33] [34]

[35]

[36] [37]

[38]

Qual. HealthCare 19 (2007) 349–357, doi:http://dx.doi.org/10.1093/intqhc/ mzm042. Q.S.R. International, NVivo Qualitative Data Analysis Software, QSR International Pty Ltd., 2012 2012. C. Weston, T. Gandell, J. Beauchamp, L. McAlpine, C. Wiseman, C. Beauchamp, Analyzing interview data: the development and evolution of a coding system, Qual. Sociol. 24 (2001) 381–400, doi:http://dx.doi.org/10.1023/ A:1010690908200. T. Bodenheimer, R. Berry-Millett, Care management of patients with complex healthcare needs, Policy 1 (2009) 6. P.B. Smulowitz, L. Honigman, B.E. Landon, A novel approach to identifying targets for cost reduction in the emergency department, Ann. Emerg. Med. 61 (2013) 293–300, doi:http://dx.doi.org/10.1016/j.annemergmed.2012.05.042. D. Blumenthal, M.K. Abrams, Tailoring complex care management for highneed, high-cost patients, JAMA 316 (2016) 1657–1658. D. Blumenthal, B. Chernof, T. Fulmer, J. Lumpkin, J. Selberg, Caring for highneed, high-cost patients—an urgent priority, N. Engl. J. Med. 375 (2016) 909– 911. S.B. Cohen, STATISTICAL BRIEF# 481: The Concentration and Persistence in the Level of Health Expenditures over Time: Estimates for the US Population, 2012–2013, (2015). A. Gawande, Finding Medicine’s Hot Spots | The New Yorker, New Yorker. January 24 (2011) 40–51. J.H. Hibbard, J. Greene, What the evidence shows about patient activation: better health outcomes and care experiences; fewer data on costs, Health Aff. (Millwood) 32 (2013) 207–214, doi:http://dx.doi.org/10.1377/ hlthaff.2012.1061. J.H. Hibbard, J. Stockard, E.R. Mahoney, M. Tusler, Development of the patient activation measure (PAM): conceptualizing and measuring activation in

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

patients and consumers, Health Serv. Res. 39 (2004) 1005–1026, doi:http://dx. doi.org/10.1111/j.1475-6773.2004.00269.x. K.R. Lorig, H. Holman, Self-management education: history, definition, outcomes, and mechanisms, Ann. Behav. Med. Publ. Soc. Behav. Med. 26 (2003) 1–7. R.R. Weaver, M. Lemonde, N. Payman, W.M. Goodman, Health capabilities and diabetes self-management: the impact of economic, social, and cultural resources, Soc. Sci. Med. 102 (2014) 58–68, doi:http://dx.doi.org/10.1016/j. socscimed.2013.11.033. K. Barnett, S.W. Mercer, M. Norbury, G. Watt, S. Wyke, B. Guthrie, Epidemiology of multimorbidity and implications for healthcare, research, and medical education: a cross-sectional study, Lancet 380 (2012) 37–43, doi:http://dx.doi. org/10.1016/S0140-6736(12)60240-2. R.L. Morris, C. Sanders, A.P. Kennedy, A. Rogers, Shifting priorities in multimorbidity: a longitudinal qualitative study of patient’s prioritization of multiple conditions, Chronic Illn. 7 (2011) 147–161, doi:http://dx.doi.org/ 10.1177/1742395310393365. M. Komaromy, J. Barlett, A. Zurawski, et al., A Novel Intervention for High-Cost, High-Need Medicaid Patients: A Study of ECHO Care, (submitted for publication). J.O. Prochaska, W.F. Velicer, The transtheoretical model of health behavior change, Am. J. Health Promot. (2016), doi:http://dx.doi.org/10.4278/08901171-12.1.38. H. Han, K.P. Gabriel, H.W.K. Iii, Application of the transtheoretical model to sedentary behaviors and its association with physical activity status, PLoS One 12 (2017) e0176330, doi:http://dx.doi.org/10.1371/journal.pone.0176330. K. Malterud, Qualitative research: standards, challenges, and guidelines, Lancet 358 (2001) 483–488.

Please cite this article in press as: M. Komaromy, et al., Contingent engagement: What we learn from patients with complex health problems and low socioeconomic status, Patient Educ Couns (2017), http://dx.doi.org/10.1016/j.pec.2017.08.019