Abstract
Purpose: While experts suggest that primary care needs far-reaching transformation that includes adding or reconfiguring roles to improve patient care, little is known about how role change occurs in practice settings. Methods This was a cross-case comparative analysis of 3 projects designed to improve health behavior counseling in primary care practices by adding to or changing clinical support staff roles. Qualitative data (site visits notes, grantee reports, interviews with grantees, and online diary entries) were analyzed to examine instances of role change in depth, using role change theory as an organizing framework. Results Practice team members had greater success taking on new roles when patients valued the services provided. Often, it was easier to a hire a new person into a new role rather than have an existing practice member shift responsibilities. This was because new personnel had the structural autonomy, credibility, and organizational support needed to develop new responsibilities and routines. Conclusion: As primary care delivery systems are redesigned in ways that rely on new roles to deliver care, understanding how to effectively add or change staff roles is essential and requires attention to patients’, practice members’, and institutions’ support for new roles.
New clinical roles such as health coaches, patient navigators, care coordinators, and counselors hold promise to enhance practices’ capability to provide high-quality and efficient preventive and chronic care.1-5 Integrating these new roles into the delivery of primary care may involve hiring new staff or modifying the roles of existing practice members. Changes in the structure of delivery systems and the roles of health care workers have implications for how practice members work together and may expose training deficits and/or role stressors among current and newly hired staff. In the extensive national dialogue around primary care transformation, the challenges of changing care delivery roles and responsibilities have not been well examined.6-12
Social roles are constructed by the rules and norms that guide behavior and attitudes in a particular context. 13 Role change is the shift in definition, rules, or norms for performance of usual behaviors. 14 Adding or altering a role and the behaviors that support it is not an easy task, although there are conditions that can foster role change. A helpful framework for understanding the conditions for successful role change is role change theory. The theory predicts that role change is influenced by (1) relative structural autonomy of those in the roles, (2) level of institutional support for the role change, (3) resources and mobilization of the new role occupants, (4) relative cultural credibility of the proposed new role, and (5) the extent to which client demand and trust are mobilized. 14 Table 1 contains the definitions of these dimensions.
Elements of Role Change in Primary Care Practice: Based on Role Change Theory
Drawing on role change theory, we examined how health care workers adapted to changes in their roles and identified the barriers and facilitators of this process. We accomplished this by examining 3 innovative interventions to redesign practice and promote patient health behavior change, each of which either added to or modified the roles of practice staff to meet the needs of a redesigned care delivery process.
Methods
Data Collection
The 3 projects were funded by the Prescription for Health (P4H) program. P4H was a collaboration between the Robert Wood Johnson Foundation and the Agency for Healthcare Research and Quality and funded 10 practice-based research networks to test innovative approaches for changing health behaviors (ie, risky alcohol use, unhealthy diet, sedentary lifestyle, and tobacco use) in primary care.15,16 The projects we examined varied on the type of role change the intervention prompted, and included: 1) adding a new person in the practice; 2) adding new staff members in the community rather than the practice, and 3) extending the roles of existing practice staff. Table 2 describes each intervention and the type of role change involved. More detailed descriptions of the projects have been published.17-19 Although only three projects were considered, these projects were implemented in 27 practices that represented a diverse set of practices in terms of geographic region, size, community setting, and patient population. An independent evaluation team collected data across the projects. The data collection strategy is described in detail elsewhere.16,20-21 Briefly, for each intervention in the P4H program, project- and practice-level data were gathered using surveys, site visits, and interviews with research team members. In addition, project and research team members participated in online diaries to record confidential biweekly reflections on implementation experiences. 20 We selected the 3 projects for analysis because their core strategies involved considerable change in staff roles. Multiple assessment methods produced rich insights into the implementation process and the barriers and facilitators associated with integrating a new role into practices. 22 The University of Medicine and Dentistry of New Jersey–Robert Wood Johnson Medical School Institutional Review Board approved this study.
Project Descriptions
Analyses
Real-time process analysis
All data were deidentified and entered into ATLAS.ti for analysis. An iterative group analysis process of reading and reflection was used to analyze all of the P4H qualitative data. 23 Diary data, site visits notes, and grantee reports and interviews were discussed weekly as data were collected. The purpose was to monitor, as the projects unfolded, the implementation experiences across the funded projects, including barriers and facilitators, so that emerging lessons could be identified and shared. In the early stages of the P4H evaluation, role change was not a specific analytic focus, but it emerged as an important issue in 3 projects.
Role change analysis
We intensively examined the 3 projects using an immersion-crystallization approach. 24 We reviewed relevant text tagged during the ongoing analysis to identify categories, concepts, properties, and their interrelationships, 11 creating codes to capture the variations and subthemes observed. In response to emerging findings regarding role change, we conducted a review of the literature to begin making connections to relevant literature. 25
In a second immersion-crystallization cycle, role change theory was used as a theoretical framework to inform our observations and analysis. Role change theory sensitized the team to important concepts but did not limit our insights during this phase of the analysis. 26 Emerging findings were challenged by seeking disconfirming evidence and by sharing findings with the principal investigators implementing interventions that involved role change. We modified our findings as needed. These steps, in addition to triangulating findings from multiple data sources, enhanced the validity and trustworthiness of the results.
Results
Structural autonomy of role
When newly introduced personnel had structural autonomy, they had the flexibility and independence needed to make changes to meet practice and patient needs. This was observed in projects 1 and 2, where new personnel (a lifestyle counselor and a health educator/community liaison, respectively) were introduced to the practices as part of the interventions. One lifestyle counselor reported that in one practice,
a nurse practitioner asked me if I would see a family where the child (10 years old) is 125 lbs and 4 ft and 6 inches tall. Although, she doesn’t qualify . . . based on age, I . . . wanted to see the family because obviously there is a problem. . . . The nurse . . . wanted me to meet with her and the mom, so I did. [Diary entry]
In contrast, project 3 attempted to transition medical assistants (MAs) into a new role in which they assessed and made referrals for unhealthy behaviors. MAs had little structural autonomy; they were answerable to clinicians, supervisors, practice leaders, and administrators and had little control over their work.
And sometimes, [the administrator] he’s not down here, so he doesn’t understand everything we do have to do . . . to us, it’s like, “You’re being unfair . . . because you don’t know what we do down here . . . and for you to force it, it’s, . . . hard with no help from the providers.” . . . One time . . . I said [to another MA] “Why aren’t you doing it? [the intervention]” She’s like, “Because my provider said she don’t have time and she ain’t gonna mess with it.” [Interview 1601]
As a result, MAs were often caught between competing demands from administrators to do the assessments and referrals for the intervention and pressure from clinicians to maintain the flow of patients through patient rooms (the usual MA role), making it more difficult to fulfill their new responsibilities.
Institutional support
Even when there is structural autonomy for a new role, lack of institutional support could impair role change. For example, in project 1, financial constraints prevented the lifestyle counselor from working in each practice more than 1 half-day each week. As a result, the lifestyle counselor struggled to create a sense of continuity in each practice, and integrating the new role into the practice team was therefore more difficult. Other institutional constraints such as lack of adequate space for counseling patients, lack of easy access to computer systems for scheduling and reviewing patient records, and malfunctioning equipment were observed barriers to role transitions, regardless of whether new or existing personnel were asked to transition into a new counseling role in the practice. The lifestyle counselor reported,
The administrator saw me and wanted to put me in my old room which has been taken over . . . by the social worker . . . he has not been able to arrange the phone to work yet or my own login and password for the computer. . . . I am . . . still working out of the doctor’s lounge. [Diary entry]
In particular, lack of time to conduct the work, including lack of institutional assistance with reallocating existing tasks, could make taking on new tasks difficult (project 3), even when strong leadership support is evident.
The clinic administrator actively tracks MA productivity in terms of enrolling participants. . . . The medical director is also strongly supportive . . . the MAs know what is expected of them in terms of the goal of one P4H enrollment per day, per MA. The top-down leadership emphasis seems to be the greatest component of motivation for the MAs. [Diary entry]
[The intervention] is too much for us. It really, really is. Only because we have so much we have to do and that . . . stops us from doing what we have to do. . . . Because . . . it does take time. [MA interview 1623]
Resources, mobilization, and cultural credibility of new role occupants
Institutional support and role occupant expertise influenced the new role’s credibility in the practice culture. That is, integration of new responsibilities and behaviors relied, in part, on credibility and acceptance of the role within the practice culture. Roles were more credible when institutional support was present and role occupants demonstrated the expertise needed to adopt a new role.
Mobilization entails building the capacity of role occupants to take on new tasks. Without appropriate skills and training, staff is unlikely to succeed. Projects 1 and 2 hired new individuals in the roles of liaison or lifestyle counselor. Both projects identified educationally prepared and skilled professionals capable of taking initiative in new roles.
We . . . report that a health educator is hired. She is a . . . bilingual, bicultural woman. . . . She has lots of experience with motivational interviewing and is working on a masters in health ed. Her nutrition background is not as strong, so we are working on ways to get her basic training and resources. She will have lots of support from our nutrition experts. [Diary entry]
Mobilization of the MAs in project 3 was more challenging. Although the project designed the counseling as limited in scope, MAs voiced discomfort with evaluating patients’ health behavior and did not think that they had the appropriate skills to provide brief counseling and referral services. One MA said,
I find it hard to preach about losing weight when I’m sitting there and I’m 80 pounds overweight. And I’m not in a program. [Interview 1583]
Projects 1 and 2 identified appropriately qualified individuals and provided necessary support to effectively become new role occupants. Conversely, a poor fit between skills and job demands and a lack of additional training and support impeded progress in project 3. 19
Demand for a new service
Demand for the new role influenced the extent to which it could be implemented. When practice members and patients valued what a new role occupant was offering, it was easier for the role occupant to succeed. A project 2 team member reported,
. . . response to the [liaison] has been excellent . . . we receive reports like “my patient was thrilled with what she was getting from her [liaison].” [Diary entry]
On the other hand, many MAs in project 3 reported that attempts to offer brief counseling and referral were frequently met with resistance and resignation from patients:
“you’re getting after me.” . . . and we’re not getting after them, we’re trying to just make them aware but they’re like, “Oh, here we go again. Like it’s not enough to hear the doctor.” That’s what they tell us . . . that’s what one patient told me one day when I tried explain . . . and was like, “Oh, I already know. I’m gonna die of something anyways.” [Interview 1601]
In addition, trust influenced patient demand for a new service. Patients were sometimes skeptical of entrusting information to people they were unaccustomed to providing it to and not always accepting of services provided by new staff. For example, in project 1, the lifestyle counselors reported that parents or guardians expressed concern that the health risk assessment was a form of government surveillance and worried about the government finding out and taking their child away or being considered “bad parents.” In project 3, lack of patient trust was connected to perceptions about the new role occupant’s lack of expertise. MAs varied in the extent to which they perceived patients to be comfortable sharing health behavior information with them. Some MAs believed that many of the patients do not trust them enough and think that MAs don’t know enough or can’t do much for them. Yet other MAs felt that they had established relationships with patients that helped them accept the new services, suggesting that many times the patients would open up to them in ways that they would not to the physician.
MAs also recognized the value of doing something for themselves, to help themselves. How MAs perceived their own expertise and the level of trust patients had in them influenced the extent to which they were able to transition into their new role in the practice. Clinicians who did not support their MAs in this new role (with the MAs or with patients), as described above, eroded MAs’ confidence and impaired successful role change.
Discussion
We identified several factors that can facilitate role change in primary care practice. Success in taking on a new role may be enhanced when the following conditions are met: there is patient demand for services provided via the new role, occupants of the new role have the structural autonomy to take on new responsibilities and behave in new ways, and the new role has cultural credibility and institutional support in the practice setting. In the interventions we examined, projects with greater flexibility to hire people in new roles faced fewer barriers mobilizing role occupants than the project relying primarily on the skills, background, and training of existing practice staff. For project 3, strong institutional support at the administrative and practice leader level was not enough to overcome lack of client demand, limited credibility within the practice culture for the role change, and difficulty mobilizing the MAs to embrace their new role responsibilities. In addition, clinicians who did not support their MAs in the new role may have eroded MAs’ confidence and impaired successful role change. Project 3 conceptualized the MA role change as a simple and low-intensity intervention; however, MAs did not experience it this way. In contrast, projects 1 and 2 had role occupants with higher skill levels and cultural credibility within the practice; nonetheless, when patient demand was low and institutional support limited, even these staff found it more difficult to effectively provide health behavior change counseling.
These findings suggest that if existing staff are asked to assume new roles, leadership should carefully consider what background is sufficient, what training is needed, what old responsibilities could be removed, and how support in the environment can be optimized. Significant change in one role on a team is also likely to stress other team members, making change difficult to implement and sustain without investment in professional development and support for other team members. Thus, adequate resources, including adequate financial reimbursement, to support integration of new responsibilities are also necessary.7,27
This research is limited in several ways. We have a small sample of 3 projects; however, these projects included 27 primary care practices that are reflective of many typical practice settings. Because there are no direct data from patients, information about patient demand for services is inferred from diary and interview data. Another limitation is that these interventions were implemented as research studies, which could have reduced practices’ investment in the outcomes. Last, there may be other explanations for the success or failure of the role transitions described that were not brought forth for examination. While this study has limitations, these limitations are outweighed by the novel contribution this research makes to an important, highly relevant, and understudied area of primary care research that identifies factors that influence role change in the primary care setting.
This study illustrates how role change theory is one theory that may inform future practice redesign efforts. Adding preventive and chronic care services such as health behavior change counseling often requires highly skilled roles and adequate financial resources, as well as structural autonomy for skilled practitioners to deliver these services. Regardless of the decision to either expand the roles of existing staff or to hire new staff careful attention to the 5 factors examined here may facilitate improved role transition in either case. Understanding how these conditions affect role change in primary care practice is important for planning and redesigning primary care delivery systems to provide health behavior services. Although this study did not specifically examine teamwork—how the people in the new roles worked together (or not) with other members of the clinics—understanding the elements needed to facilitate role adoption in practice is an important step in furthering efforts toward practice transformation to new models of team-based care.
Footnotes
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Prescription for Health (P4H) program funded by the Robert Wood Johnson Foundation (RWJF; grant 053221) in collaboration with the Agency for Healthcare Research and Quality (AHRQ). The authors report no conflicts ofinterest.