Care managers’ experiences in a collaborative care program for high risk mothers with depression

Abstract

Objectives

The aim of this study is to understand care managers’ experiences in caring for depressed mothers in an integrated behavioral health program.

Methods

As part of a quality improvement project, we conducted a focus group interview with six care managers caring for low income mothers with behavioral health needs in a safety net program in King County, WA. Using thematic analysis, codes were organized into themes that describe the care managers’ experiences.

Results

Two organizing themes along with associated themes emerged: 1) Assets for improving depression outcomes: patient-provider interactions including the importance of engagement; program resources such as care coordination and access to a consulting psychiatrist and 2) Barriers to improved depression outcomes: patient-provider interactions including difficulty engaging patient; patient-related factors such as multiple stressors; program resources such as need for more psychiatric support; and difficulty accessing outside resources.

Conclusions

Numerous potentially modifiable factors including levels of engagement, motivational interviewing, and increased psychiatric support were identified by care managers as affecting depression care and outcomes. Implications for care management training and approaches to psychiatric consultations are discussed.

Introduction

Integrated behavioral health care models are increasingly recognized as way of providing needed care on a population level (1–4). Although many such models exist, the model with the most evidence is the collaborative care model initially developed to treat and manage depressive disorders in primary care (5). In most collaborative care programs, behavioral health care is provided by a team that includes a primary care provider (PCP), a behavioral care manager (CM) based in the primary care clinic, and a team psychiatrist. CMs have regularly scheduled consultations with the team psychiatrist to conduct psychiatric case reviews and develop a treatment plan which may include medication recommendations, brief psychotherapeutic interventions by the CM, and referrals to specialty services.

Although the number of collaborative care programs is increasing, few studies have examined the experiences of CMs caring for patients in this model of care (6). Documenting the experiences of CMs has the potential to provide valuable perspectives of how real-world implementation may differ from randomized controlled trials of collaborative care, suggesting opportunities for optimizing quality improvement and workforce development/training efforts. The primary aim of this qualitative study was to explore aspects of the collaborative care program associated with successful treatment of depressed mothers served in a collaborative care program as well as barriers to such successes.

Methods

Study setting

Funded by King County and administered by Public Health-Seattle & King County in collaboration with Community Health Plan of Washington, the High Risk Mothers program of the Mental Health Integration Program (MHIP) is a collaborative care program that provides behavioral health services for a number of safety net populations (7) including a group of high risk mothers (women who have a mental health need and who are either pregnant or parenting a child under 18 years old and low income) (8). Between 2008 and 2010, 1,244 high risk mothers were enrolled in this program.

Sample and procedures

This study included a focus group interview with 6 CM (4 females) working in the High Risk Mothers Program of MHIP across 6 community health center organizations. CMs were interviewed during a focus group as part of a regularly scheduled bimonthly care manager meeting at one of the community health centers in King County. One of the authors (JKW) facilitated the focus group. CMs were oriented to the format of the focus group and informed that a previous quality improvement project had identified certain care processes (e.g. number of in-person visits with the CM) to be associated with improved depression outcomes in this patient population (9). The CMs were also presented with data describing substantial variation in the number of follow-up visits and depression outcomes across the six participating organizations. The interview guide consisted of the following two open-ended questions: 1) “What helps you be successful with the many moms who do well in the program?” 2) “What about the moms who don’t do well? What could be done to help them do well?” The focus group moderator ensured that all participants were given the opportunity to discuss their experiences and beliefs as related to these two questions. This project was conducted as part of a routine quality improvement program and was not considered research requiring individual consent by the University of Washington’s Institutional Review Board.

Data analysis

The focus group interview was analyzed using a thematic analysis approach (10, 11). The interview transcript was reviewed and codes were identified and agreed upon by two raters (HH and JKW) using Atlas.ti 6.2 for data management. Overlapping and redundant codes were collapsed. The codes were then categorized into themes that describe the care managers’ experiences in working with depressed maternal clinic patients.

Results

Two organizing themes emerged from the focus group: Assets for improving and barriers to improvement in depression outcomes. In addition, we identified 2 unique assets and 4 unique barriers (summarized in Table 1).

Table 1.

Themes from care managers’ focus group

Organizing theme: Assets for improving depression outcomes	Organizing theme: Barriers to improved depression outcomes
Theme: Patient-provider interactions (e.g. provider-patient racial/ethnic/language concordance and good engagement)	Theme: Patient-provider interactions (e.g. difficulty engaging with patients)
Theme: Program resources (e.g. care coordination, benefits of seeing patients in primary care, and access to consulting psychiatrist)	Theme: Patient-related factors (e.g. multiple stressors)
	Theme: Program resources (e.g. need for more psychiatric support)
	Theme: Outside resources (e.g. difficulty connecting with services)

Assets for improving depression outcomes

Provider-patient interactions

Racial/ethnic/language concordance was helpful in engaging with primarily non-English speaking patients:

“..the biggest population is Hispanic/Latinos. And so, coming in, and seeing that I speak Spanish. And also, I’m bi-cultural, I think that in itself is huge for them. So, I think that’s another good reason for them to come. “

Good engagement with patients is a strong predictor of successful depression treatment:

“..the biggest indicator if they’re going to get well, is if they’re engaged with me, if they’re motivated, if they’re willing to come in and see me in the clinic, then I notice their (depression) scores going down a lot faster. “

and

“So, using motivational interviewing skills with them to figure out what stage of readiness they might be. What they’re willing to engage in at that time. It’s providing recommendations and seeing if they’re willing to engage in one of those recommendations”

Program resources

Many benefits were seen in providing care for patients in the primary care setting:

“I think that they feel safe with me because I’m part of their primary care clinic. And so, I think they see me as part of the family. A lot of these women have had their babies with the clinic. They have been seeing the same PCP for 10–20 years sometimes. So that makes them feel more comfortable to engage with me.”

and

“If they’re coming in to see primary care, we will try to piggy back on some of their appointments or get the primary care provider to refer them or nudge them back in our direction.”

Care coordination plays a role in depression improvement:

“I think also what helps is being able to communicate really easily with the PCP. The moms are busy, they don’t have time to be running around, or the transportation, to be able to go get their prescription and sometimes forget they needed to refill their prescriptions or forget their doctor’s appointments. So a lot of times, I’ve been able to send that to the PCPs through our EHR (electronic health record) system, asking for a refill at least for a week until they can reschedule their appointment for the next week. Or letting them know, letting them know, someone is having a problem with the medication is an important, medication issues or some kind of medical problem..”

Barriers to improving depression outcomes

Patient-provider interactions

Difficulty engaging patients:

“What could still be a concern for those moms who are not choosing to follow-up with us, or just are not ready, I should say not choosing, they’re just not ready, at that point of readiness to engage, to make changes.”

Patient-related factors

Patients experience multiple and concurrent stressors:

“..we have some of the DV, where, um, cases too, where it’s hard to come because of the domestic violence. “

Program resources

Although care managers feel comfortable seeing patients with low/moderate levels of psychiatric severity, they find that some patients are beyond their expertise:

“I certainly feel that we’re not that effective for moms that are that ill, other than maybe trying to get them to a place that would be more effective for treatment.”

Need for more psychiatric support:

“..the other thing that I think would be helpful, is if the patients were able to see the psychiatric provider directly. Like I said, I spend a lot of my time talking about their symptoms and the side effects of meds and why you should take meds and why meds are helpful. But, you know, I’m not a psych nurse practitioner or a psychiatrist. So I think, you know, if they’re able to see somebody face to face, it’s, it would be very helpful. A lot of times when, the consultant, my psych provider, he would bring up a lot of really good questions that I hadn’t thought about asking patients that would definitely be helpful, in terms of prescribing medications to them.”

Outside resources

Many care managers had difficulty connecting their patients with needed clinical and social services:

“..they can’t access services dependably in community mental health. Because of the red tape, the cost of the co-visits, they end up just staying with me even though they’re willing to go somewhere else.”

and

“Those external referrals, it’s hard for them to go through as was mentioned before, it’s like, we’re lucky enough that they are willing to come in and talk to anyone at all about what’s going on and then to have them to go somewhere else to, it’s really hard for that to happen, but it is successful sometimes.”

Discussion

This study sought to explore behavioral care managers’ perspectives and experiences regarding factors that contribute to successful depression treatment as well as barriers to depression improvement with the ultimate goal of improving the effectiveness of care for low income mothers served by the program. Several themes emerged under the organizing themes of “assets” and “barriers.”

Care managers believe that effective engagement with patients is critical in both retaining patients in treatment and having patients participate in recommended treatments. Many CMs reported using motivational interviewing techniques (12) in order assess patients’ readiness to change and to engage patients in behavior change (13). In motivational interviewing, once the provider has determined where along the spectrum of change (i.e. pre-contemplative, contemplative, preparation, action, and maintenance) a patient lies, s/he then proceeds to assist the patient in resolving their ambivalence towards the acceptance of recommended health treatments. A recent randomized controlled trial has shown that among depressed low income perinatal patients, an engagement session incorporating principles of motivational interviewing as part of interpersonal psychotherapy was more effective in reducing depressive symptoms than enhanced usual care (14).

In a prior program evaluation of the High Risk Mothers program, the number of in-person visits with the CM and phone calls by the CM were each independently associated with improved depression outcomes (9). Although it is plausible that more intensive care manager follow-up is responsible for these improved outcomes, it is likewise plausible that patients who receive more intensive follow-up are those most engaged in treatment. Patient engagement is a crucial element for improving outcomes for patients with any chronic conditions, including depression, which requires treatment for months to years. Nevertheless, most people who access mental health services discontinue treatment prematurely. A key intervention that care managers can offer is the facilitation of engagement through highly individualized education and treatment planning, motivational enhancement, and proactive outreach. Findings from our focus group interview suggest that motivational interviewing may be an important skill for adequately preparing CMs for working with this patient population. Ongoing and advanced motivational enhancement training for experienced CMs could have the potential to enhance their ability to engage an even wider array of patient groups that have historically underutilized mental health services, notably racial/ethnic minorities.

Provider-patient racial/ethnic/language concordance was also identified by CMs as a factor in helping them to work with patients in the program. Although research suggests that provider-patient ethnic or language concordance may not be necessary to improve health outcomes (15–17), the ability of the CM to elicit the patient’s understanding of depression may help to enhance the engagement process (18). Despite the CMs’ perception that their cultural awareness was a facilitator of positive outcomes, the collaborative care model itself was not explicitly culturally-tailored. However, the key features of the model emphasize the importance of patient-centered care that addresses patients’ preferences and concerns, takes into account barriers and limitations, and intensifies care in response to patient outcomes, all features that likely promote the provision of services in culturally-appropriate manners that are reflected in successful engagement. It is worthwhile noting that the delivery of patient-centered care in the primary care setting and the coordination of care with the PCP were also identified by CMs as contributors to treatment success.

CMs highly valued the psychiatric expertise provided by the consulting team psychiatrist, but felt that more support would be helpful in the management of patients with higher levels of complexity. In this collaborative care program, once treatment plans are implemented, select cases are reviewed by psychiatric consultants using a stepped care model. That is, patients who are not experiencing improvements in depressive symptoms are provided treatment changes which may include an addition of an antidepressant, an increase of antidepressant dosages, changes in antidepressants, additional behavioral interventions, or referral to higher levels of care. One way of providing more psychiatric support may be to increase the frequency of case reviews for complex cases. In addition, telemedicine offers the potential to extend the psychiatric consultants’ role and expand their capacity to provide direct patient evaluations. At this time, team psychiatrists and CMs are piloting the use of Skype for both case reviews and ‘face to face’ evaluations of patients not improving as expected. Lastly, group sessions co-led by the care manager and psychiatric consultant may be helpful in augmenting the care provided to patients in this program.

Conclusion

Behavioral care mangers who work with depressed low-income mothers report that their ability to successfully engage patients in care is a key factor in bringing about positive outcomes. Motivational interviewing training is one way to increase the care manager’s ability to engage this patient population. These findings inform quality improvement efforts in this program and may also be of interest to systems implementing similar models of care.