A realist review of shared medical appointments: How, for whom, and under what circumstances do they work?

Realist Review methodology was chosen to uncover how and for whom and under what circumstances SMAs work. The general pattern of work in realist review involves a set of iterative (i.e., non-linear) steps involving: (a) establishing research questions; (b) identifying candidate middle-range theory(ies) (MRT) for the program; (c) searching the literature to select, appraise, and retain primary studies and scan for new candidate MRTs; (d) synthesizing evidence using context-mechanism-outcome configurations (CMOc); and (e) refining the evidence-based middle-ranged theories to answer questions related to ‘what works, for whom, under what circumstances and how.’[69, 70] Middle-range theory should embody the underlying logic of the program in question (across many cases), and may also include causal theories on the key elements of context, mechanism and outcome. The synthesis stage involves examining the MRT in relation to the evidence captured, through a heuristic process of constructing, exploring, and refining context-mechanism-outcome configurations. Data to include in CMO configurations can involve primary outcomes (quantitative or qualitative) as well as program and setting descriptions and interpretation of outcomes by study authors [71].

Context can include social as well as physical settings, cultural/psychological norms, demographic factors, institutional practices and so on. Our definition of mechanism relies on the description laid out by Pawson in 2006, to include the cognitive and/or emotional responses of participants to resources offered [71]. For this review, we define ‘mechanisms’ as the way in which patients react/respond to the group clinical encounter, with specific attention paid to the range of resources offered through SMAs. Thus the CMO configurations portray how SMA programs works and under what contextual conditions. Further detail about realist synthesis methodology has been published [68, 72, 73]. We note that in contrast to Cochrane reviews which involve identification of the review question(s) and search for primary studies using pre-defined inclusion and exclusion criteria, realist review is an iterative process that involves clarification and refinement of the scope of the review and search for relevant evidence, refining inclusion criteria in the light of emerging data [69, 70]. In contrast, to the focus of Cochrance reviews on relevance to the specific research question and methodological rigor, a realist approach considers relevance and rigor from a broader fitness for purpose perspective.

The review team represented a variety of disciplines and professions (medicine, nursing, and social sciences) as well as experience in implementation research (DCA, SK, LDS), management research (DCA, JO), program evaluation (KJ, LES, LDS), systematic reviews (KJ,KRJ), chronic care delivery (DCA, SK, KJ, LES, LDS), and realist synthesis (JJ). SK is the national VA subject matter expert and along with DCA has been conducting and studying SMAs since 2006. SK and DCA formed the initial review questions and led the search, selection and appraisal of papers. LDS, LES, KDJ, and KRJ contributed initial study design and provided feedback on iteratively developed protocols. JJ joined at a later stage to further develop the middle-range theory and lead an in depth examination of 20 SMA studies, using CMO configurations.

The literature search was assisted by a librarian at the Health Sciences Library at CWRU. An initial search completed in September 2011 included Medline, CINAHL, and Pub Med using the following search terms: 1) shared medical appointments, 2) shared medical visits, 3) group appointments 4) group medical visits, 5) cluster visits, 6) chronic care clinics, 7) primary care group visit, 8) group medical appointment, and 9) group office visits. The search time frame was designed to coincide with a parallel evidence based synthesis using a traditional approach to systematic review that involved meta-analysis and statistical regression; both projects supported by funding provided by VA HSR&D, but were conducted independently [7]. The results of a search of Pub Med, Medline, and CINAHL are diagramed in Fig. 1. Eight hundred seventy one citations were reviewed with 205 meeting criteria (based on inclusion of terms relevant to SMAs in the article title). Of 205 articles, each was assessed by two independent abstractors using an adapted data collection tool based on the instrument of Zaza et al. [74] This standardized abstraction form and procedure was chosen to enhance consistency, reduce bias, and improve validity and reliability. Discrepancies between reviewers were resolved by consensus (The abstraction form and grid are available from the authors). Through this rigorous selection and appraisal process, 71 high quality, relevant SMAs articles were retained and examined in relation to the conceptual models of SMAs that were developed prior to the review [3] and which evolved during its initial stages; our final model is shown in Fig. 2. At this point, co-author (JJ) was invited to join the team as a realist review expert and to continue with the analysis of 71 articles. We then used 20 articles, purposively sampled to achieve a representative sample of types of sites, patients, and geography to conduct an in depth analysis involving CMO configurations. (See Table 2) The CMO configuration analysis followed the construction of candidate middle-range program theories – the latter being constructed from the expertise of the co-investigative team and a general reading of 71 articles. In constructing the CMO configurations, a series of middle-range program theories were hypothesized to suggest how SMAs work to improve service delivery and self-efficacy. The purposively selected 20 articles were read and re-read with an eye to coding data that provided support to these theories. After coding, exemplar quotes (i.e., that had the most explanatory power) were retained for synthesis using CMO configuration. To ensure transparency and rigor of the findings, these exemplar quotes were extracted and a CMO configuration was formulated in relation to each quotation. In most cases when authors reported an explanatory claim about how SMAs function to produce outcomes, an overt connection between context, mechanism and outcome was not made. Thus the CMO configurations represent our work in theorizing the connection between context, mechanism, and outcome, based on the quotes. The CMO configurations were thus inferred both from the literature and subject area expertise of the co-investigators, to become a set of evidence-informed theoretical claims for explaining how SMAs work.

The findings are organized across 9 sections describing the key mechanisms and contexts to explain how SMAs work and for whom. These are presented below and depicted in summary form in Table 3. These CMOcs were constructed by the research team, based on an interpretation of the key explanatory quotations in the context of evidence from other papers within the twenty reviewed. Within each section, one or more quotations from the primary literature are presented, followed by a context-mechanism-outcome configuration that provides an explanatory interpretation of the quotation. References from the other most relevant papers are cited. The sections are titled as followed:

Typically, patients are isolated in their illness experience. This often leads to a number of problems including cognitive dissonance and misperception of one’s situation and abilities. The SMA created social contact amongst people with similar background or illness, which relieved these misperceptions:

‘you have a feeling of not being capable enough, you feel you are the only one who is not able to manage it, everyone else is capable for sure, but I do not know anyone else. Therefore you get relaxed when you meet others who feel the same way.’ [29] (p. 253)

CMO configuration (CMOc): Feelings of inadequacy prevailed coupled sometimes with isolation (context). The SMA created social contact amongst a group of people with similar illness experiences. This exposure helped to correct misperceptions about their capabilities and the capabilities of others in self-efficacy (mechanism). The social contact combined with people sharing similar experience contributed to esprit de corps which promoted self-efficacy (outcome).

Patients found it eye-opening to witness the illness experience of others. This had an impact on their comfort-level, knowing that they are not alone, and learning about the illness through the experience of others:

‘Many patients had no inkling that others were experiencing the same or similar symptoms and stated that they were often comforted knowing that they were not alone in dealing with their post-cardiac surgery problems…Having others present in the conference room who had similar surgeries or surgeons sparked a lot of lively discourse among the participants, as they were curious to know if they were ambulating as well or had similar amounts of pain, wound issues, or musculoskeletal symptoms after discharge.’ [35] (p. 128)

‘all patients highlighted the knowledge they had gained from other patients who shared similar issues and concerns.’ [44] (p. 88)

CMOc: Patients typically experience their illness in isolation (context). The group visit offered patients the opportunity to compare their illness experience with the experience of others who are in similar situations. Patients were curious to know how other patients were managing, because such comparisons help to understand one’s own illness experience (mechanism). This most likely led to improved self-efficacy and reduction of stress, supporting disease management and healing (outcome).

The mix of patients in SMAs brought out key group dynamics which supported self-efficacy. Some patients were doing better than others, which triggered two separate mechanisms. For those who were doing better, their role modeling served to keep them motivated to stay on track. For those who were doing less well, witnessing others who were progressing well served as motivation and something to strive for:

‘…Some participants set high goals for physical activity and this encouraged other members. They also had the opportunity to share their knowledge and had a feeling of usefulness or altruism’ [25]. (p. 5)

“There used to be a guy here…he lost like 40 some pounds and we really clapped for that fellow cause he really worked hard for that. And it gave us something to try for.” [21] (p. 1289)

CMOc: SMA brought patients together, some of whom had lower motivation for self- management behavior and others who had relatively higher motivation (context). This led to feelings of usefulness and altruism (mechanism) for the former group, and motivation to do better for the latter (mechanism). Either way, these mechanisms most likely led to improved self-efficacy and disease management (outcome).

There was some evidence that SMAs alter patient-provider relationships such that friendship and partnerships were formed between patients and providers. Although there was not much detail on the inner mechanisms of this effect, we speculate that having multiple patients in a room, and increased duration of the clinical encounter helped to balance the power dynamic between providers and patients and allowed the patient to get to know the provider. One patient noted that the SMA helped in better understanding the doctor. This was perhaps due to the fact of witnessing the doctor in action with other patients:

‘A number of the study participants expressed the concept of becoming friends with the GMV provider. The informal structure of the GMV may have allowed the patient and provider to develop a partnership rather than having a more traditional active-passive relationship.’ [45] (p. 223).

‘I got to meet new people and really get a feel for my doctor.’ [33] (p. 47)

CMOc: Because of the structure of group visits with many patients involved, the clinical encounters are more informal, relaxed and friendly than one-on-one clinical encounters (context). This allowed new patient-provider relationships, including friendships, which fostered trust amongst all parties (mechanism). Stronger, more trusting relationships between patients and provider was a result (outcome)

In some SMAs, a number of providers worked together to provide clinical care. There was some evidence that the experience of providers witnessing other providers with patients, and working together in SMAs increased team cohesion and coordination of service. This led to increased work satisfaction and enhanced working relationships:

‘We have discovered a high level of satisfaction and enhanced team-building among the transplantation care providers. This type of appointment is a team-based approach to care, with each member of our bone marrow transplantation team playing an important role in the overall success of the program. This not only led to increased efficiency and team rapport but also to a mutual appreciation for the helpful role that each team member plays in the process …Because this type of visit is so novel, there is increased physician satisfaction. The team approach to health care also offers care providers with the help, encouragement, and assistance of an entire care delivery team, which rapidly evolves into a cohesive, goal oriented health-care team’ [44] (p. 89–90)

CMOc: Typically providers work in isolation with patients (context). The SMA, in which team members could interact with each other and gain mutual appreciation of their roles increased bonding, and mutual appreciation (mechanism). This led to improved service delivery and work satisfaction. Staff felt supported and collectively created a more cohesive team (outcome)

SMAs were described by some service providers as an environment for increased learning on how to improve service delivery, due to the fact that patients in the room were sharing and brainstorming ideas about illness management, from which the service provider gleaned new ideas. In the following quotation, the service provider suggested that this would not have been possible in a one-on-one visit. This may be due to a variety of reasons, including the fact that the one-on-one visit doesn’t as easily open up such kinds of brainstorming and information sharing:

“I think that it [the GMV {sic – equivalent to SMA}] has helped me to be more creative in looking at ways to meet people’s needs. Some of that just comes from the patients themselves because they often have some really neat ideas about how to overcome challenges or difficulties in dealing with the diabetes. So I think that, not only have I become more aware but I’ve also, they’ve given me some really good tips and ideas. I think there’s stuff I learned that I wouldn’t have learned if I had done it on an individual basis. There’s a lot of value that comes out of that, that kind of impromptu patient teaching of each other” Provider #28 [40]. (p. 5–6)

CMOc: The SMA created an environment in which patients with a common illness experience shared information and brainstormed ideas about self-management (context). This environment allowed the service provider to appreciate new ways of thinking about how to serve the patient population (mechanism). This in turn most likely led to improved service provision (outcome)

SMAs are typically longer than one-on-one appointments. The increased time that provider and patients spend together was reported as having a positive impact on patient perceptions:

‘We have discovered that patients are often encouraged and more hopeful when they spend time with other patients with similar or worse conditions’. Patients spend more time with their care providers and the specialty team in this setting. This provides a sense of comfort to each patient [44]. (p. 89)

CMOc: Due to having multiple patients in the clinical encounter, SMAs require substantially longer timeframes than regular visits (context). This added time allows patients and providers the opportunity to really get to know each other and have patients open up and relax in the clinical setting. This has been identified as comforting to the patient (mechanism). This relaxation and comfort presumably leads to better self-management (outcome).

Patients were able to absorb information from the provider in a more effective way due to the fact that such information was complemented by first-hand experience of other patients. This relieved stress and resolved questions that the provider was not always able to address:

During one of the GMAs (sic. Group Medical Appointments are equivalent to SMAs), a patient asked about travelling with oxygen and another patient explained how easy it was to travel. The NP may have been able to produce the same information but not the firsthand experience and the reassurance that it would not be complicated.’ [25] (p. 5)

CMOc: The SMA allows health information to be shared not only by the health professional, but also by other patients through firsthand experience (context). Patients receiving such knowledge felt reassured by the experience of other patients (mechanism). This likely led to decreased stressed, and improved application of the information given during sessions (outcome).

Some patients found that witnessing the physician in a group setting interacting with other patients help them to feel increased trust toward the physician and the healthcare system in general. This may be due to the fact that the group patient setting encourages the physician to remain open, willing and transparent in the face of many eyes watching. Even with the best of intentions, the one-on-one visit does not typically support such physician attitudes and behaviors:

“I’ve learned to trust him. I trust him more than I used to and that’s important, that bond of trust has to be there. I trust him more when I see that he’s open to learning and figuring out new things that are only happening in group dynamics” Patient #8 [40]. (p. 6)

“Do you know…what [SMA] helps me to see is what the physician, his devotion of trying to solve a health problem and trying to correct it. That actually reestablishes my faith in the medical system because you can see that they’re really devoted to trying to figure out really what is ailing you” Patient #13 [40]. (p. 5–6)

CMOc: Mistrust of physicians is pervasive in the health care setting (context). The SMA created an environment in which patients and providers are on more equal grounding. This led patients to feel that the physician is trustworthy (mechanism) leading to improved doctor-patient relationships and presumably improved health outcomes (outcome).