Background

Improving health outcomes for people living with obesity is paramount to healthcare providers and policymakers. This is in part because the annual total costs of obesity in Canada ranges up to $11.08 billion Canadian dollars [1]. Studies suggest that a primary care-based obesity treatment model could be cost-effective over the long term [2]. However, there is a paucity of evidence on the effectiveness of the current obesity management services provided through primary care [3, 4]. The Canadian Obesity Network—Réseau canadien en obésité (CON-RCO) has developed the “5As of obesity management” framework [5], which incorporates the conceptual structure of the best practices in obesity management in a step-wise approach (ask, assess, advise, agree and assist) to facilitate obesity management in primary care [5]. The aim of the 5As Team (5AsT) study is to examine the impact of a team-based intervention on the frequency and quality of obesity management encounters in a primary care setting. [6]

Recently there has been increased awareness on the need for improved reporting of the details of complex innovations being testing in real-world settings in pragmatic study designs [7, 8]. This has led to the international panel from the EQUATOR network creating the TIDieR guide, with the intent to have sufficient detail to permit more nuanced understanding of the context, and content of the intervention [9]. 5AsT is a pragmatic study that seeks to work in real world context, and to create an intervention that works in this setting. Thus, context, and the end-user’s input is crucial in creating the intervention [10]. The focus of this paper is to provide a detailed overview of the 5AsT intervention to support complete reporting and replication.

Methods

The intervention was informed by the conceptual framework of Complex Innovation Implementation (CII) [11] and by the Theoretical Domains Framework (TDF) [12], illustrated in Figs. 1 and 2. CII is important because ensuring good alignment with the care organizations’ visions and business plan, increases the likelihood for ongoing stable partnership for the duration of the intervention. The detailed negotiation of the study question, and mode of delivery of the intervention was important as it led to a strong innovations-values fit with the organization and supported the implementation climate. A key insight from CII was the need for a clinical champion, a trusted clinical member of the team, who could act as a liaison between the care organization and the research team. This individual was provided by the partner organization as an in kind contribution, and was crucial for the intervention implementation. The TDF was important as it informed the nature of the intervention as having to include not only knowledge elements, but also deliberate efforts to promote social/professional role identity, and social influences, peer support, practice, and the setting of individual provider goals. This led to the structure of the intervention having a content element, and a learning collaborative element.

Fig. 1
figure 1

Conceptual framework of complex innovation implementation (adapted from [11])

Full size image
Fig. 2
figure 2

Theoretical domains framework for behaviour change of the provider (adapted from [12])

Full size image

This intervention was designed to be tested in a pragmatic randomized control trial with a longitudinal convergent mixed-method design, which has been described in detail in the protocol elsewhere [6]. Briefly, 5AsT is an allocation concealed; pragmatic randomized controlled trial with longitudinal convergent mixed-method evaluation aimed at increasing the number and quality of weight management visits conducted by primary care providers [6]. Of note, there was ongoing monitoring of the intervention delivery, the context and the impact of the intervention using interviews, log books, and field notes [6]. We present here only data pertinent to provider views of the intervention itself.

Participants in the intervention design were team members from primary care clinics randomized to the 5AsT intervention (Registered Nurses/Nurse Practitioners, Mental health workers, Registered Dieticians), and the researcher team (family physicians, obesity specialist, anthropologist, epidemiologist, public health). In this paper, we describe the derivation of the 5AsT intervention, including the co-creation with the community partners of the research questions, and the process of eliciting practitioners’ self-identified knowledge gaps to inform the curricula for the 5AsT intervention. Themes and topics were identified through facilitated group discussion and a curriculum relevant to this group of practitioners was developed and delivered in a series of 12 workshops. The intervention commenced with a kick-off session October 21, 2013, 12 × 2-h workshop sessions held biweekly for 6 months (November 2013–April 2014); and, an evaluation session post-intervention in May 2014, and 6-months after the end of the intervention (October 2014). See Fig. 3 for a schematic diagram of the 5AsT intervention.

Fig. 3
figure 3

Schematic diagram illustration the 5Ast intervention and its components

Full size image

Study setting

The 5AsT study was conducted in a primary care network (PCN) in Alberta, which employs dedicated multidisciplinary healthcare providers (nurses, nurse practitioners, mental health workers, dieticians, exercise physiologists, respiratory therapists) embedded in 67 family practices with over 170 family physician members serving 192,655 Albertans. This PCN is an extension of the primary care services, which provides a comprehensive family medicine through multi-disciplinary teams that include physicians, nurses, dietitians, social workers, respiratory therapists and exercise specialists. These extended teams are embedded in community family practices and provide support for chronic disease management. As the physicians are fee for service, and the interdisciplinary team members are salaried, it was easier for the team members to participate in this initial intervention. Ongoing work external to this project is ongoing for physicians, evaluating more condensed training formats.

Intervention group

The multidisciplinary providers in the clinics randomized to the 5AsT intervention group (n = 29) were consented at each stage of our evaluation (in order to give them the chance to decline participation at any point). All providers were age ≥18 years, one provider was male and all others were female. Six of the providers were registered dieticians, with a seventh new hire joining 1 month into the intervention; seven mental health workers; and 15 registered nurses/nurse practitioners (one withdrew post-randomization). All providers contributed to the design of the intervention.

Control group

Providers from the control group were not consented as only de-identified, routinely collected data was used from this group. The control group received standard training in the 5As, as well as other obesity training from the regional health authority, as part of their orientation and development through their employer. They did not receive the 5AsT intervention program; we expected them to continue their standard practice. As they practice in geographically dispersed locations from the intervention team members, contamination was minimized.

5AsT intervention

The content of the 5AsT intervention was derived by asking primary care practitioners (n = 29) attending the 3-h kick-off session with an introductory teaching session on the 5As of Obesity Management™, followed by an interactive workshop to determine the content for the intervention. The providers were asked the following question: “What do you think would help in your patient care around weight management?”

Providers identified topics, which were related to themes around obesity assessment, prevention, and management from which they felt the need for further education and training. The 5AsT members then categorized the materials into intervention sessions from the topics [two members (DCS and AAO) initially did the categorization of the topics, which was debated and approved by other team members]. The team, with strong prior relationships with the obesity community, then coordinated with regional experts and resources to find speakers to support each of the 12 intervention sessions.

In the 5AsT intervention sessions an invited speaker presented for about 1 h. They were encouraged to be interactive and to bring useful tools and resources on the topic. The presentation was then followed by a learning collaborative session for an hour, as described below.

As it was expected that not all providers could make each session, eleven sessions were videotaped and posted to You Tube (with presenters’ written consent) immediately after each session. The purpose was to allow for providers to watch the talk if they were not able to make the session. The twelfth sessions was an interactive team communications session for the PCN, so was not videotaped.

Table 1 provides an overview of the intervention content based on the users’ needs assessment, providers/speakers, their expertise and the summaries of the session content. The attendance at each session, by discipline is provided. The intervention materials have been compiled into learning modules and are available at http://www.obesitynetwork.ca/5As_Team.

Table 1 Intervention sessions, speaker’s designation and a session summaries of the content
Full size table

Learning collaborations

The advantages of learning collaboration in primary care practice have been highlighted previously [13–15]. Briefly, learning collaboration is a learning process centered on sharing among participants. In other words it is a shared learning process in which participants are responsible for their own learning as well as for one another [16]. It can be a good strategy to leverage resources [17], and also, an important advantage of collaborative learning is to facilitate group learning in order to achieve a particular goal.

The providers were divided into two groups for the learning collaborative, with colleagues working in the same clinic teams grouped together. The learning collaboratives had facilitated discussion of the presentation content of the day, tools and materials shared with them prior to the session, and reflection from their practice experience. At the goal setting element of the session, providers also had the chance to share with the rest of their group the goals they set for themselves and the resources they found useful in their practices.

Some elements of our collaborative learning include: learning about newer research knowledge, practices on weight management and patient goal setting sessions, team-driven small tests of change, collaborative resource sharing among providers, experience sharing teach-backs, and the sessions being led by an experienced facilitator.

Practice and group facilitators

It is important also to note that we employ the use of practice facilitators and group facilitators in the 5AsT intervention. The use of practice facilitators has been previously described as an effective strategy to improve primary care processes, outcomes, and the delivery of services [18]. Two kinds of practice facilitation were employed in the study: internal (clinical champion) and external practice facilitators. The internal practice facilitator, or clinical champion as informed by the complex innovations framework, was the person designated by the PCN 1 day per week to support the intervention. This was a trusted clinical colleague (dietician) and leader who was able to support the providers in their context, and liaise with the research team to support creating space, climate, and time for the intervention. The external practice facilitators in the 5AsT study acted as a link between providers and evidence or resources that may be used to facilitate weight management encounter with patients as illustrated in Fig. 3. They identified and liaised with speakers, and implemented the planning and execution of the intervention and evaluation session.

Following each session, the external practice facilitators compiled a summary of the materials, and circulated them to the members of the group. In addition, each time a participant identified that it would be useful to have a tool or resource, the external practice facilitators identified one and provided it. Where none existed, they were created with the assistance of a graphic design team, and iteratively reviewed with the participants. This has been described in detail elsewhere, and the tools compiled are available for use [19].

In addition to the practice facilitators, the learning collaboratives had facilitated discussions by the internal practice facilitator, and another trusted internal PCN expert. The group learning collaborative facilitator’s roles was to prompt the conversation among providers and to lead the goal setting sessions. The two group facilitators were rotated on two occasions during the early aspect of the intervention to improve discussion and sharing among providers in the separate groups. This modification was deemed necessary so that the two goal setting groups would experience both group facilitators with their different personal attributes.

Evaluation of the 5AsT Intervention

The evaluation of the 5AsT intervention was done in three ways: (1) real time monitoring with field notes as described above; (2) individual semi-structured interviews with all participants and (3) questionnaires presented to the participants following the 6-month intervention at the evaluation session.

For the qualitative portion, three researchers took field notes during all sessions. Semi-structured interviews were conducted with all intervention participants (N = 29). The field notes and interviews focused on key aspects of: Theoretical Domains Framework (knowledge, skill, beliefs about capabilities, goals, beliefs about consequences, intentions, emotion, optimism, and role identity) [12], Complex Innovations Implementation (CII) [11], a framework developed to locate and build upon factors that may influence intervention success, and questions pertaining to their views of the intervention, the 5AsT approach and their work environment. We used a thematic analysis approach to determine themes from within the qualitative data [20, 21]. Transcripts were inductively coded line by line according to subject. Data was managed using NVIVO 10 software (QSR International, Burlington, Mass.) Research team members and an independent third party cross-checked all analysis and key findings were shared with participants after the intervention, at which point an opportunity for comment was provided. This paper presents only the results relevant to the evaluation of the intervention.

For the quantitative evaluation we used an intervention specific questionnaire to evaluate the sessions, and a Likert scale to rate each of the intervention sessions and exact data from the providers regarding the intervention. The questionnaire reports a 7-item Likert scale (1-Excellent, 2-very good, 3- good, 4-satisfactory, 5-poor, 6-very poor and 7- unable to comment), at the evaluation session on May 8, 2014. Quantitative data was managed in Microsoft Excel and analyzed in SPSS software.

This study is approved by the University of Alberta ethics committee and was registered at Trials.gov (NCT01967797). It is funded by an Alberta Innovates Health Solutions grant, with significant in kind support from the Edmonton Southside Primary Care Network.

Results

Providers identified 43 topics that they thought would be helpful in their patient conversations about weight management at the kick-off session (“Appendix 1”). These topics were grouped into 13 themes, which facilitated the choice of 5AsT intervention speakers and the content of the 12 sessions Table 1). The topics for the 12 sessions (“Appendix 2”) are related obesity assessment and weight management in which practitioners felt the need for further education and training. These included issues related to cultural identity and body image, emotional and mental health, motivation, setting goals, managing expectations, weight-bias, caregiver fatigue, clinic dynamics and team-based care. Participants also identified a need for greater understanding of physiology and the use of a systematic framework for obesity assessment (the “4Ms” of obesity).

The attendance sheet was used as a proxy to measure adherence of the participants to the intervention. Detailed attendance by session is reported in Table 1. Fifteen providers attended ≥10 sessions of the intervention, including five who attended all sessions. Nine providers attended 5–9 sessions. Five providers attended fewer than 5 sessions including: one who withdrew from the study at the beginning (no data), two mental health workers did not attend the any sessions, and two who only attended a few sessions. All providers contributed to the interviews.

At the final evaluation session on May 8, 2014, 21 providers (9 = RN, 3 = NP, 2 = MHP, 7 = RD) were present on the day and two addition providers filled the questionnaire and returned it on a subsequent date.

On the 7-item Likert scale, 83 % of respondents rate the intervention as either very good or excellent, with the remaining 17 % rating it as good. Overall, 86 % of the providers responding also said they were either strongly comfortable or somewhat comfortable with the 5As of Obesity Management™ [5] following the 5AsT intervention, and 91 % reported they felt more comfortable discussing weight issues with their patients as a result of the intervention. Of the 23 respondents, 21 reported they would recommend the intervention to others, and 2 respondents felt they were not able to comment.

In terms of the structure of the intervention, overall, 18 of the 23 respondents (82 %) felt that biweekly (once in 2 weeks) learning collaboration format was suitable for them. Table 1 provides the proportion of the 23 respondents that scored each session excellent, very good, or good (1–3) on the Likert scale.

In terms of the learning collaborative groups, 73 % (16) of the respondents rated them as excellent/very good/good. Of the respondents, 64 % felt the goal setting in the learning collaborative sessions was helpful, with 39 % reporting that they often/always met their goals.

The Youtube videos were used by 64 % of respondents, and among those who viewed them 87 % rated the videos as very good or good. The main challenge was the sound quality of the videos.

Overall, the intervention was very well received, Interview and field note data reveal strong intervention values fit and self-reported behavior change. Table 2 provides some representative quotes of positive views of the intervention, while Table 3 provides some examples of challenges from provider views of the intervention. The overall results are summarized below.

Table 2 Examples of representative positive provider views on the intervention
Full size table
Table 3 Examples of challenges from provider views of the intervention
Full size table

Positive themes that stood out included: variety, it was collaborative, multidisciplinary, long-term and sustainable in that it leveraged the internal practice facilitator as a change agent with the task of ongoing training of new staff in the organization. Comments included appreciating the insights of multidisciplinary teams, hearing their “clinical peers”, sharing ideas, hearing from diverse speakers, and collaboratively discussing issues. One nurse suggested the intervention provides options of where to start the conversation and has changed how in general she thinks about weight management. The format was generally considered positively; providers stated that the recurrent sessions helped the information sink in and gave them time to adapt it to their practice. Participants felt this lead to increase in confidence and comfort with the material. A provider also, suggested that the format of the sessions allowed for self-reflection, with another stating that the structure of the sessions allowed new information to become part of the practice.

Some providers, however, felt that either the sessions were too long, or that it was difficult to get the time away from their clinical practice. The perceived usefulness of the learning collaborative was mixed, many participants feeling that it was both useful to have space to share their clinical experience with peers while also stating that at times the conversation was difficult. However, the structure did lead to increased collaboration between multidisciplinary team members. Active monitoring of the field notes of the intervention meant that the research team was aware of the concerns for the imbalance between the two learning collaborative groups, with one group with more quiet individuals. This was then purposefully reviewed with the group and solutions were obtained from the participants. This led to a rebalancing of the teams between the groups to have more balance, as well as periodic rotation of facilitators.

Discussion

Through the 5AsT study we were able to identify obesity management related topics and learning that may help providers change behavior, improve their practices and refine obesity encounter for patients. Here we highlight the 5AsT method and intervention content. The intervention sessions, video links and the tools co-created with providers are available on the web (http://www.obesitynetwork.ca/5As_Team). The purpose of these modules is to create a living repository of tools and resources to support primary care teams in the community who would like to improve obesity management in their context. From a research perspective, they serve as a record of the content of the intervention, supporting transparency of reporting [7–9]. Our overarching aim is not only to improve the quantity of obesity management in primary care setting, but also to improve its encounter quality. Through the kick-off of the 5AsT intervention, we identified primary care providers’ barriers and knowledge gaps to weight management in their practices. We envisage that a participatory provider engagement, such as 5AsT intervention, may increase the frequency, quality of weight management encounters in family practices and the quality of life of the patients.

Interventions aimed at changing provider behavior in the real world are best informed by the active engagement of the end-user to ensure applicability and context-appropriateness [10], as was amply observed in this study. The engagement with the end users resulted in many pragmatic solutions to challenges in implementation, which proved crucial. Both the complex innovation framework [11] and the theoretical domains framework informed this intervention [12], with core elements such as practice facilitation (internal [11], and external [18]), proving crucial, and learning collaboratives [13–15] proving more mixed. Overall, the intervention proved positive for the majority of the participants, resulting in self-reported practice change. Challenges frequently revolved around scheduling and time constraints, which were partly mitigated by providing an asynchronous video option for catching up on missed material.

Previous studies suggest that providers experience barriers in obesity management [22, 23] and lack adequate weight management knowledge [24, 25]. We also know the frequency of obesity management in the PCN is low (Unpublished data from routine continuous administrative monitoring), leading to the premise that if we reduce the knowledge gaps in providers we may improve the quality and frequency of obesity management visits by patients and also improve weight management consultations.

Most behavioral weight loss interventions have failed to demonstrate long-term effectiveness and sustainability of weight management. It may therefore be important to encourage more emphasis on other non-weight related outcomes of obesity management intervention as this unrealistic concentration on weight loss by providers, was a key learning point in the course of our intervention. Providers may need to look beyond the anthropometric changes following an intervention and mindful on the quality of life of the patient as well [26]. A key finding was the providers’ choice of topics around caregiver fatigue, relapse prevention, emotional eating, and mental health concerns; daily challenges in their practice.

There are several limitations to this study. The 5AsT intervention can be generalized to other similar populations to a certain extent. Similar to the finding of other studies [23, 27, 28], the knowledge gaps highlighted by the providers involved in this study are common. However, one challenge in our context was it was not possible to include fee for service busy family physicians in the intensive intervention. We were able to have two family physicians participate on the research team. Our future research will focus on interventions on family physicians, and on other aspects of provider’s consultations that may indirectly affect weight management. A primary care system, with a multidisciplinary team, similar to that of 5AsT study is likely to share the same issues as our practitioners have highlighted. However, given the diversity of contexts in which primary care is practiced, future work will need to consider how the intervention may need to be modified for different settings. A rich description of the intervention is a necessary first starting point in synthesizing what works in diverse settings.

Conclusion

Primary care practitioners perceive important knowledge gaps across a wide range of topics relevant to obesity assessment and management. The 5AsT intervention was designed to respond to the identified needs of front line providers in terms of content, and the structure promoted interaction and collaboration, emphasizing practice opportunities and innovation.

Further work should focus on how these knowledge gaps can be addressed and whether increased knowledge and competencies in these areas will translate into better health outcomes for overweight/obese clients. Furthermore, 5AsT intervention’s goal is improved weight management by improving provider’s knowledge and patients experience. Ultimately, the 5AsT intervention is a promising primary care-based approach co-created with end users to achieve better management of obesity. The 5AsT web resources can support community primary care teams in practice-based learning to improve obesity management.