Patients' Perspectives of Care Management: A Qualitative Study
Ann S. O’Malley, MD, MPH; Deborah Peikes, PhD, MPA; Claire Wilson, PhD; Rachel Gaddes, MPH; Victoria Peebles, MSW; Timothy J. Day, MSPH; and Janel Jin, MSPH
Risk-stratified care management is a cornerstone of the patient-centered medical home (PCMH) model, but studies on patients’ perspectives of care management are scarce. PCMHs are intended to focus on patients’ needs, so knowing patients’ perspectives about care management is critical. To help bridge this gap in the literature, we conducted semi-structured interviews with patients or their caregivers in practices participating in a large initiative to understand their experiences with care management, what they found most useful, and what might be improved.
Care management is a “set of activities designed to assist patients and their support systems in managing medical conditions and related psychosocial problems more effectively.”1 It is also associated with a lower risk of complications.2,3 A common care management model in primary care includes the use of “care managers,” a term we use to refer to the person who works with high-risk patients and primary care teams to develop care plans, improve patients’ understanding and self-management of their chronic conditions, monitor chronic conditions between visits, and follow up after transitions in care, such as a discharge from a hospital or an emergency department (ED).1,4 Care managers are considered most effective when they work within a primary care practice, meet patients in person, communicate with them by phone between visits, and coordinate with their primary care physician.1,5 Care management can be provided by a care manager embedded within or external to the practice who works closely with the primary care clinician or by a primary care physician with the help of a medical assistant or other staff.1,4
Through the Comprehensive Primary Care (CPC) initiative, CMS collaborated with 39 private and public payers to test care management, alongside other interventions, in nearly 500 primary care practices in 7 US regions. CPC ran from October 2012 through December 2016. Designed to reduce costs and improve primary care delivery, patients’ and providers’ experiences, and health outcomes, CPC gave participating practices non–visit-based payments and the opportunity to share in cost savings, regular feedback on patient outcomes, and a learning network.6 The CPC initiative emphasized patient and caregiver engagement, and CMS held a number of technical advisory panels, which included patients, when they designed the initiative.
CPC required participating practices to risk-stratify patients and provide care management to those at highest risk by: 1) linking each active patient to a provider or care team, 2) defining each patient’s level of need, and 3) managing each high-risk patient’s care according to their needs.
Most CPC practices used embedded nurse care managers, either newly hired or trained “on the job” to provide care management. A few of these had been certified in case management.7-9 In some small physician-owned practices, the primary care physician performed the care management tasks, often with the help of a medical assistant.
Most literature on care management focuses on provider or delivery system, rather than patient, perspectives.1 Two studies from Canada and the United Kingdom examined the patient’s perspective of care management models and found that patients receiving care management perceived their care, psychosocial support, access to services, advocacy, and communication with providers favorably.10-12 These studies called for more studies on patient perspectives on care management.
As part of the CPC evaluation, we collected qualitative data annually from 21 CPC “deep-dive” practices (3 from each of the 7 CPC regions) about CPC implementation. We selected these deep-dive practices because they had characteristics similar to those of all CPC practices. Building on our relationships with these practices, we asked 11 of them ranging in size, ownership, and location to recruit high-risk patients for semi-structured interviews about care management. We interviewed patients in the middle of the third year of the 4-year initiative.
We gave a script and responses to frequently asked questions to care managers at the 11 practices who described the study to patients (or their caregivers) with whom they had contact over the next 2 weeks and asked if each would participate in 1 telephone interview. Each care manager compiled a list of patients who consented, their contact information, and whether they were discharged from a hospital in the previous month. The combined lists included 159 patients, all of whom were mailed an advance letter.
We explained to patients that their participation in the study was voluntary and would not affect their insurance coverage or healthcare and that comments would not be shared with their primary care practice. Patients were offered $25 to complete the interview.
Because our research examined a public benefit, the New England Institutional Review Board (NEIRB# 13-174) exempted the study. Respondents provided verbal consent first to their care manager to be included on the list of volunteers and again to the interviewer to participate in this 1-time interview.
We conducted semi-structured telephone interviews from March to May 2015, starting with the first patient on each list and continuing until we completed 7 interviews per CPC region. Of the 159 patients, we attempted calls to the 138 for whom we had working phone numbers. These patients were distributed across regions. Forty-two patients did not answer the phone or return voicemails, 38 refused to be interviewed, and 15 had hearing, health, or time issues that made it difficult to interview them. There were no differences in mean age or disability status between responders and nonresponders, but responders were slightly more likely to be female (74% vs 63%). We completed 43 interviews (18 with patients who had recently been hospitalized). After completing 40 interviews, we received no new information, and all responses fit into the existing codes.The respondents were fairly evenly distributed across regions and practices per region (Table 1).
A service transcribed the interviews verbatim.
We based the interview protocol (see eAppendix [eAppendices available at ajmc.com]) on Bodenheimer’s conceptual framework on care management,1 the Chronic Care Model,13,14 and CPC requirements.6 We pilot tested and refined the protocol with CPC patients receiving care management.
We first asked about the patients’ general experience with their primary care practice to see if patients in practices with nurse care managers mentioned the care manager without prompting. We then asked about these topics: 1) care management—how the practice introduced patients to its care manager, frequency of communication with the care manager, and kinds of support the care manager provided; 2) care planning and patient engagement—patients’ understanding of and level of involvement in developing a care plan; the practices’ understanding and incorporation of patients’ needs, values, and goals for care into the plan; 3) care coordination—the practice’s referral procedures for specialists, how the physician or care manager communicated with specialists, and whether the physician and/or care manager discussed the specialist’s recommendations with the patient; and 4) care transitions—patients discharged from a hospital or an ED in the previous month were asked about their practices’ follow-up.
Coding and Analysis
Using the aforementioned conceptual framework, we developed a coding rubric and dictionary15 to capture key themes.1,6,13,14,16 During the first weeks of data collection, the interviewers met weekly to discuss emerging themes, refine the coding rubric, and assess interrater agreement on a set of transcripts coded simultaneously. Two interviewers coded the transcripts using the qualitative analysis software NVivo10 (QSR International; Burlington, Massachusetts). We built frequent debriefings and peer review by the research team into the coding and analysis to maximize the reliability of coding and to reduce researcher bias.17,18
Practice and Patient Characteristics
Based on data they reported to CMS, 6 of the 11 practices we studied had a full-time nurse care manager and 3 had a part-time nurse care manager. In the 2 solo practices, the physician, helped by a medical assistant or other staff, was the sole care manager. This distribution of staff was, according to data that practices report to CMS annually, similar to the 2015 distribution of care manager staffing across all CPC practices.
Thirty-seven of the 43 respondents who completed interviews were patients and 6 were caregivers who responded for patients who lacked the cognitive or physical ability to respond for themselves (Table 1). Thirty-one of the 43 were in system-owned practices, 28 were 65 years or older, and 15 had disabilities. Eighteen patients were discharged from a hospital in the previous month.
About half the patients were with the same practice for 10 or more years (ranging from 6 months to 32 years). More than half reported visiting their practices 2 to 6 times per year; 10 reported visiting at least monthly.
Patients’ Perceptions of the Care Team
When asked about the composition of their primary care team, patients typically identified their physician and the nurse or medical assistant with whom the physician worked most closely during their routine, comprehensive, and acute care visits. All interviewees reported that their lead clinician was a physician. Most patients who said that they had a care manager felt that the care manager was an important part of their primary care team.
In many of the practices that had a nurse care manager, the primary care physician or another person in the office introduced the patient to the care manager during an office visit. Echoing the experience of others, a patient said, “She came in [during my visit with my physician] and introduced herself and told me if I had any problems I should call her...She gave me her card...[and] then she would call every couple days and make sure that my blood pressure’s down, and she coordinated with [my physician].” A few patients first met their care manager after a hospitalization, through a phone call from the care manager or during a visit by the care manager in the hospital.
One-fifth of patients were not familiar with the concept of a care manager and had trouble identifying who played this role although we intentionally sampled patients who had recent contact with their care manager and interviewers defined a care manager’s role for the patients, including examples of what that person typically does. (This proportion does not include the patients in the 2 solo practices whose physicians performed care management tasks.)
Outreach from CPC practice care managers, health plans, and hospitals to patients at the time of hospital discharge contributed to patient confusion about the role each person played in their care and about who their care manager was. This duplicate outreach was an issue for approximately one-third of the patients with a hospitalization in the previous month. One patient who was confused about the identity of his care manager noted, “To be truthful, my Medicare supplements [supplementary insurance plan] has tried to take on the role of a care manager.”
A few patients preferred not to interact with a care manager, particularly older patients who had a longstanding relationship with their primary care physician. One, for example, preferred to rely on her primary care physician of 20 years: “[He] is the one I can talk to the easiest. I am a firm believer in primary care doctors because I totally think that they are the ones who know a patient the best.”
Development of Care Plans
CPC guidance to practices19 notes that an essential feature of care management is “a mutually agreed upon and documented plan of care, based on the patient’s goals and the best available medical evidence; it is accessible to all team members...and addresses all major and significant ongoing health problems and risks.” Although the initiative did not require practices to give patients a copy of their care plan, the patient’s involvement in goal setting was implicit in the CPC definition of a care plan. Few patients, however, had heard the terms “plan of care” or “care plan,” and many did not understand this concept even after we described it. After probing by the interviewers, about a quarter of the patients described formal care plans and goals (including steps for achieving them), which they had set with their physician and/or care manager.
Patients varied widely in the degree to which they wanted to contribute to their care plan. About one-third of respondents mentioned that they were too old or sick to consider their goals or to develop such a plan and preferred to rely on their doctors’ opinion. Another one-fourth of respondents reported being very engaged in their care planning; of these, a handful had only recently become more engaged in their healthcare either because of failing health or their physician’s or care manager’s encouragement. The remainder of respondents did not understand the concept of care plans after we described it and thus could not comment on the extent to which they would want to contribute.
Only 1 patient reported having a copy of her care plan. Patients typically described a discussion or a verbal agreement with their primary care physician: “He didn’t write all that down. We just had a good discussion about my age, getting older, and stuff like that, and what I need to be aware of and start doing to help stay healthy.” For patients who could not articulate their goals and did not recall discussing them with their primary care physician or care manager, it is unclear whether someone in their practice created a care plan for them but did not share it or whether a plan was not created.
Care Management Activities
Care managers communicated with patients primarily by phone and during office visits; a few used e-mail or text messaging. Patients who reported regular contact with a care manager said that they primarily received follow-up after hospitalizations, help with the ongoing management of chronic conditions, medication monitoring, navigating the health system and community services, and other forms of assistance (Table 2). These activities are consistent with those required by CPC.
Several care managers helped patients and caregivers navigate the healthcare system and community-based resources when they needed medical equipment, home health care, or affordable medications. As one patient noted, “[The care manager] was a go-between with the [staff at the Agency] on Aging, who were case managing my home modification, and between some of the providers, and then also [for] anything that we needed from [my primary care physician], like a prescription for grab bars or a wheelchair.”
Care managers also helped patients and caregivers prepare for being discharged from hospitals and EDs. A spouse caregiver noted, “[The care manager] helped us go through some of the parameters of what would be a good [rehabilitation program] for him. Then she just kept in touch with us about things that we might need [and] resources in the community...so that [my husband] could come home rather than have to go to a nursing home indefinitely.”
Patients in most practices reported that their care manager played less of a role in coordinating with specialists. In general, clerical staff in the practices helped patients make appointments with specialists and tracked down the specialists’ notes; physicians then explained the specialists’ recommendations to patients.
Patient Satisfaction With Care Managers
Over half of patients reported having regular contact with their care manager and were willing to work with their care manager to manage their health conditions; such patients felt that the care manager was an asset to their team. In addition, there were patients who needed less frequent contact, but still valued help from care managers.
Many patients valued their care manager’s help in managing and monitoring their chronic conditions. As one patient reported, “[My care manager] calls on a regular basis to check in on me and see...if I need anything or basically how I’m feeling and if I think [my physical therapy] is helping me.” Some of these patients noted that their care manager offered them practical advice on how to meet their health goals. One said, “They know that it’s hard for me to exercise—but they still encourage me to do what I can. [My care manager] told me to take 2 cans of soup and put ’em on the top of my feet and just raise ’em up…, but that’s a suggestion that I wouldn’t have thought of.” They appreciated that their care manager took the time to help them understand how to implement their physicians’ recommendations and how to stay well after a hospitalization.
Patients were also pleased with their care manager’s assistance with managing medications. One patient said that his care manager was particularly valuable when his medications were adjusted: “When we add one [medication], or we change one, it changes everything else. Because I take so many, it’s a complicated thing.”
Patients who reported having regular contact with their care manager or who were open to working with their care manager felt that the care manager was an asset to their team. Patients particularly valued care managers who listened to them and explained things in lay terms, helped to manage medications and chronic conditions, followed up after a hospitalization, and helped to navigate the healthcare delivery system and community resources.
Although we purposely selected patients who were, according to their primary care practices, receiving care management services, one-fifth of patients (or their caregivers) could not identify a person at the practice other than the physician whom they felt fit the role of a care manager as we described it to them. This may reflect several factors, including inadequate integration of nurse care managers into primary care teams, lack of patient interest in engaging with a care manager, and repeated interactions with many individuals, such as home health aides and care managers from insurers and hospitals. We found variation among deep-dive practice clinicians’ efforts to introduce care managers to patients in person.8 Others have noted challenges to integration of care managers into the practice team, including a lack of experience with and knowledge of the best ways to design and implement effective care management; the limited availability of high-quality, standardized training for care managers; and a need to train practices to collaborate effectively with care managers.20 Other patients preferred not to interact with a care manager.
It is likely that most patients barely recognized the term “care plan” because practices did not develop the plans systematically. Based on our site visits to the deep-dive practices,8 there is variation in the extent to which care plans are developed, used by care team members in a practice, and shared with patients.
Given that we interviewed selected patients in selected practices in 1 initiative, our findings cannot be generalized to all CPC practices or all patients receiving care management. In addition, CPC practices received funding and learning support for care management, and other practices—particularly small, independent, or rural ones—may not have the financial resources or enough local nurses to hire even part-time care managers.21
Nonetheless, the patient and caregiver feedback identifies ways primary care practices can improve care management: 1) Enhance the initial connection between patients and care managers by having physicians introduce care managers to patients in person and describing the benefits of care management; 2) Consider testing whether risk-stratification could also identify patients who might be amenable to changing health behaviors. Some patients who reported having limited contact with their care manager did not feel that they needed care management. Refining risk-stratification to identify patients willing to engage with care managers could focus care management resources on patients who need and are willing to use the services; 3) Train care managers to engage patients. Care managers, particularly those who lack formal case management training, may need guidance on gaining buy-in from disinterested patients. Because a few patients reported that their care manager did not initiate regular communication but advised them to call with questions or concerns, a more proactive approach may be beneficial; 4) Provide clinicians with technical assistance on how to explain the care manager’s role and how to delegate aspects of care to care managers in a way that is congruent with each care manager’s training, experience, licensure, and skills; 5) Coordinate care management resources across providers and health plans. Given the confusion among patients who also received outreach from “care managers” from health plans and hospitals, better coordination across these entities is necessary22,23; 6) Improve the use and understanding of care plans among clinicians and patients.Practices may need to better engage willing patients in care plan development to make the plans more recognizable and useful to patients; and 7) Use consistent terminology for “care manager” and “care plan”and develop broader agreement on the elements of these plans. This might help to promote a common understanding among providers and patients about care management and solidify provider and patient buy-in. Standardized terminology might also enhance efforts to evaluate patients’ experience with care management.
Patients who reported regular contact with their care manager or a willingness to engage in care management noted its numerous benefits, but challenges to engaging patients in care management remain. Future research on their experiences with care management might include how researchers and practices could best gather patient feedback, particularly when patients’ frailty or lack of time make this challenging. Research on the best language to use with patients and providers to refer to care management and care plans and effective ways to gain and maintain patient buy-in would also be helpful.
Ongoing assessment of patients’ experiences and engagement with care management is essential. As active partners in care management, patients’ feedback can help inform both internal practice processes as well as how care managers, clinicians, and staff engage with patients.