Assessing the effectiveness of a diabetes group visit training for health center staff: a pilot study of five Midwestern community health centers

Table 1 summarizes the participants’ prior experience and the healthcare roles they served at their respective CHCs. Most participants (85%) were non-Hispanic White females. While they had been in practice an average of 11.9 years, only 11.5% of respondents reported having prior experience conducting group visits. The majority (76.9%) had prior training in lifestyle coaching and motivational interviewing techniques.

In general, the training increased participants’ awareness about what is needed for successful implementation of group visits at their health center. General awareness increased from 3.00 (SE = 0.24) prior to training, 4.38 (SE = 0.13) post-LS1, to 4.50 (SE = 0.13) post-LS2 [p = 0.023]. In addition, familiarity with billing for group visits increased from 2.94 post-LS1 to 3.59 post-LS2 [diff (SE) = 0.65 (0.26), p = 0.023] (Table 2). Overall, participants demonstrated high baseline knowledge of and motivation for implementing diabetes group visits. Although not statistically significant, participants believed their CHCs were prepared to conduct and continue diabetes group visits upon completion of the training. Moreover, participants felt that primary care providers in their CHCs would refer patients to diabetes group visits (Table 2).

As for perceived benefits, participants were aware of the benefits of the group visit model of care prior to beginning the training; however, awareness of these benefits significantly increased throughout the course of the training (Table 3). General awareness increased from 4.00 (SE = 0.17) prior to training, 4.60 (SE = 0.13) post-LS1, to 4.60 (SE = 0.13) post-LS2 [p = 0.001]. More specifically, diabetes group visits were perceived as improving efficiency in patient care (4.06 (SE = 0.18) pre-LS1, 4.59 (SE = 0.12) post-LS1, to 4.53 (SE = 0.15) post-LS2 [p = 0.033]). An additional benefit noted by participants was that group visits can increase patient satisfaction with diabetes care (4.18 (SE = 0.13) pre-LS1, 4.53 (SE =0.13) post-LS1, to 4.53 (SE =0.15) post-LS2 [p = 0.045]). It is important to note that participants did not agree or disagree that group visits improved provider productivity (3.94 (SE = 0.18) pre-LS1, 3.71 (SE =0.17) post-LS1, to 3.18 (SE =0.21) post-LS2 [p = 0.013]) (Table 3).

Participants became more aware of the barriers of the group visit model of care (Table 4). Awareness of potential implementation barriers increased from 3.75 (SE = 0.19) pre-LS1, to 4.31 (SE = 0.12) post LS1, to 4.88 post-LS2 (SE = 0.09) [p = 0.001] (Table 4). Participants reported creating and maintaining a billing mechanism (3.12 (SE = 0.17) pre-LS1, 2.59 (SE =0.15) post-LS1, to 2.82 (SE =0.15) post-LS2 [p = 0.013]) as a significant organizational barrier, particularly after attending the learning sessions. Concerns regarding lack of individual medical attention (3.00 (SE = 0.17) pre-LS1, 3.06 (SE =0.16) post-LS1, to 3.65 (SE =0.12) post-LS2 [p = 0.002]) and confidentiality and privacy (2.88 (SE = 0.22) pre-LS1, 2.81 (SE =0.14) post-LS1, to 3.50 (SE =0.13) post-LS2 [p = 0.012]) were seen as patient barriers when staff began training but less so post-LS2. It is important to note that participants reported recruiting at least 12–15 potential patients as somewhat of a barrier at pre-LS1 [3.13 (SE = 0.18] and post-LS1 [3.06 (SE = 0.17)]. Following post-LS2, recruitment was noted as a moderate barrier [2.31 (SE = 0.24)].

Overall, staff possessed high interest and knowledge about diabetes group visits. Access to a pre-designed resources and materials, a highly motivated team and supportive leadership were identified as key facilitators to early adoption of visits in their CHCs. Socioeconomic challenges experienced by patients, constraints on staff time dedicated to group visit implementation, and staff turnover, were obstacles to ongoing implementation.

In general, staff acknowledged that the CHC training serves as a collaborative ‘space’ for clinical, educational, and administrative staff to simultaneously work together to build an infrastructure that supports initial implementation of diabetes group visits. In this study, in-person learning sessions and monthly team updates on webinars provided peer support, encouragement, and networking during the process of taking on a demanding project. These sessions improved awareness of potential implementation barriers after discussions regarding the patient (client focused), provider, and health center (organizational) level challenges the staff members were likely to encounter. Of note, by the time CHC staff completed the post-LS2 surveys, all sites had initiated their six-monthly group visits, and thus had been able to directly observe the true role of the potential barriers listed in the survey, rather than making conjectures as they did prior to LS1. In-person learning sessions were designed to provide direction in navigating potential implementation barriers, but only LS2 included protected time for sites to lead discussions detailing their experiences, including how they overcame challenges. In a broad sense, this elucidates the role of camaraderie and peer support in the process of taking on a demanding project. More specifically, it demonstrates the importance of the exchange of advice and guidance between CHCs – institutions that by design share many traits in common – launching similar initiatives (in this case, group visit programs).

Similarly, increased awareness about the barriers of the group visit model improved after LS2, by which time participants attended monthly webinars and experienced the launch or completion of the group visit programs at their respective sites. The group check-in portions of the monthly webinars and the site-led reflection sessions at LS2 provided a support network for participants to share their own experiences with that of their peers at other sites, perhaps affirming observations they had already made regarding the benefits of group visits, but also allowing them to assess the program from new vantage points. In addition, it is conjectured from the change in perception from moderate to somewhat a barrier, specifically seen in confidentiality and privacy concerns as well as lack of individual medical attention, helped to improve administrative processes within CHCs.

The lack of change in the mean score of the remainder of the survey results may be attributed to a “ceiling effect”. The participants gave high pre-training scores to many measures of CHC capacity for group visits, team ability to handle the demands of group visits, and the perception of the specific benefits of group visits. This may have been a result of the site selection process, which led to enrollment of CHCs that had many of the resources necessary to start a group visit program and team members who valued the group visit model prior to training. The training, however, did impact CHC staff familiarity with billing diabetes group visits. This is an area that was new to participants and an unexplored opportunity of most CHCs.

The training raised awareness about obstacles encountered during the implementation process. Some participants adapted to these obstacles by expanding patient recruitment to provider panels other than that of the group visit provider; enlisting the help of additional health center staff and volunteers to make recruitment phone calls, check patients in, and measure vital signs; allowing one-on-one makeup sessions after the first group visit; offering transportation services; and using patient information cards to expedite preparation of the one-on-one component of the visit. In addition, participants credited our training materials with helping them initiate and sustain the group visits. As examples of resources, the training guided participants through the challenges of patient recruitment and staff time limitations. CHC teams were given telephone recruitment scripts, sample request letters for primary care physicians to approve the recruitment of their patients, and presentations on how to address staff time limitations. We did not, however, measure the use of these resources by staff who found themselves navigating the challenges these materials were meant to address.

Analysis elucidated the portions of the training that were most essential to the improvement of overall participant preparedness to conduct diabetes group visits as reflected by our survey results and qualitative analysis. Ongoing discussions about both anticipated and unforeseen challenges, with evidence-based advice for how to overcome them, were perhaps the most impactful of the educational tools we provided. The study team also believe the provision of detailed guidelines for group visit logistics and curricula were essential. The peer learning aspect of our training program also bolstered the success of each site, as participants were able to share solutions to common barriers. Finally, our support regarding the navigation of intra-CHC relationships (for example, in staff recruitment and retention and garnering leadership support) appeared to be one of the key components of our training intervention.

Success in diabetes self-management, as observed at the patient level in group visits, is not only predicated on information delivery about the disease. Peer support, especially among people with common lived experiences, has been identified as an important factor that facilitates intention to consider new attitudes or adopt new behaviors related to improved diabetes self-care [8, 25]. In addition, diabetes group visits serve as a supportive network for individuals to keep each other accountable on goals as well as celebrate short- and long-term successes. A similar dynamic can also occur at the organizational level through interactive training models for diabetes group visits. Several diabetes group visits models exist [14, 26, 27] (e.g., Cooperative Health Care Clinics [CHCCs], Drop In Group Medical Appointments [DIGMAs], American Diabetes Association’s [ADA] Adherence to Diabetes Guidelines), but it is unclear if the trainings of these models apply instructor or student centered curricula. This pilot study provides a framework for delivering interactive diabetes group trainings involving multiple CHCs that share common organizational and environmental challenges in addressing type 2 diabetes. In addition, this study points to the importance of creating experiential active trainings that empower staff to solve ‘real-time’ challenges. Future studies can explore the aspects of collaborative learning and problem solving using a co-learning collaborative approach that allows multiple CHCs to engage in the early implementation of diabetes group visit at one time as opposed to doing it on their own.

There were several study limitations. The inclusion of only Midwestern states may limit application of the findings to other geographic areas. Another potential source of bias was the study team’s selection of CHCs that were likely to be successful, possibly causing us to overestimate the value of the educational intervention. In addition, it is possible that the training intervention attracted CHCs that were motivated or already positioned to deliver diabetes group visits. With lack of a comparison group, it is difficult to assess whether the trainees’ strategies for overcoming barriers stemmed from our trainings. Also, the completion of the telephone interviews by only a portion of the participants may have introduced a selection bias. Moreover, the ability to determine the extent to which the improvements measured after LS2 could be attributed to the learning session itself was limited. The participants had also gained experience running the program and had likely gained confidence and knowledge by simple trial and error.

This is the first study to assess the effectiveness of a diabetes group visit training for CHCs and sets the stage for ongoing exploration of training models in this organizational setting. Of the health centers participating in this training intervention, six of seven started group visits and five of seven conducted monthly group visits within the study period. Interest in group visits has prompt expansion of this initiative to other health centers. Two additional cohorts of MWCN affiliated health centers have been trained as part of an ongoing cluster randomized controlled trial of diabetes group visits. The results of this diabetes group intervention have shown significant improvements in patients’ hemoglobin A1C compared to patients who received usual care at CHC sites [28].

Specifically, this study heightened awareness about the challenges related to staff turnover and billing. There are numerous potential approaches that might speculatively improve implementation and sustainability. These might include, for example, changes in administrative infrastructure such as changing billing systems or obtaining state and federal grants to support specific staff to conduct group visits. At the same time, definitive recommendations would likely be premature at this point. Instead, an important next step is to studies using dissemination and implementation models (i.e., Consolidated Framework for Implementation Research) to explore the feasibility and mechanisms through which the intervention can be integrated within CHCs. Future studies involve examining the application of dissemination and implementation models to evaluate strategies employed by multidisciplinary teams in these health centers to sustain group visits. Also, future investigations of CHCs outside the Midwest, as well as CHCs with fewer resources and less staff buy-in than those followed in our study, are expected to yield valuable additions to our findings and to the results of prior research.