Background: Diabetes is a major contributor to the burden of disease in South Africa, but glycaemic control is poor. Group empowerment and training (GREAT) for people with type 2 diabetes is a cost-effective intervention in our setting. The aim was to evaluate the implementation of GREAT for diabetes at a primary care facility in Cape Town.

Methods: A convergent mixed-method study evaluated implementation outcomes over 6 months. A programme theory model was used to prospectively plan implementation. Semi-structured individual interviews were held with healthcare workers, focus group interviews with patients and sessions were observed to evaluate adoption, feasibility, fidelity and sustainability. Quantitative data evaluated reach, cost and aspects of fidelity. Qualitative data analysis used ATLAS.ti and the framework method. Data were analysed deductively according to the pre-determined implementation outcomes.

Results: Key stakeholders agreed that GREAT was acceptable and appropriate and encouraged adoption. Initial implementation reached 35 patients in four groups and 65% of those invited attended. Only 29% attended all four sessions. Fidelity of the intervention to the training manual was good. Several contextual factors influenced the feasibility of implementation (e.g. support of management, space for group sessions, integration with appointment system, effect on number of walk-in patients, streamlined referral system). Incremental operational costs were trivial (R30/month), while opportunity costs were higher (R26 252/month). Sustainability will be related to ongoing managerial and staff buy-in and adjustment of the implementation strategies to overcome some of the barriers.

Conclusion: The study identified 30 determinants of successful implementation outcomes.

Contribution: Can guide future implementation in similar contexts as GREAT for diabetes is scaled-up in Cape Town and scaled-out to other provinces in South Africa.

Keywords: type 2 diabetes; group empowerment; health promotion; self-management; self-care; health education.

Introduction

Diabetes is a global health problem, and in 2021, there were 537 million adults living with diabetes, which is expected to rise to 783 million by the year 2045.¹ In 2021, diabetes was responsible for 12.2% of all global deaths in adults.¹ Global health expenditure on diabetes is expected to reach USD 1.05 trillion by 2045.¹ In South Africa, diabetes is the second overall cause of death and the first for women.² South Africa has one of the worst records for glycaemic control when compared to similar low- and middle-income countries.³ Consequently, diabetes is a substantial burden on patients and families, our health system and our economy.

Most of the population in the Cape Metropole rely on public healthcare. Our primary care facilities serve low socio-economic groups with a lack of education and health literacy.⁴ In the Western Cape, the Practical Approach to Care Kit (PACK) guidelines are used to manage people with diabetes, but glycaemic control remains poor.⁵ The majority of people have uncontrolled diabetes (58% with HbA1c > 8%), and one-third have an HbA1c > 10%.³ Good control requires adherence to appropriate treatment and commitment to self-management.⁴ People living with diabetes find self-care challenging because of a lack of knowledge about their condition, poor family support and symptoms such as fatigue, caused by their poorly controlled diabetes or co-morbidities.⁵ This may disempower and demotivate them in terms of self-care and treatment adherence.⁵

Group education for people with type 2 diabetes is effective at improving glycaemic control and is cost-effective.^4,5,6,7,8,9 The benefit of these structured group programmes is that they allow healthcare professionals to focus their time and energy on consultations and other aspects of diabetes care.¹⁰ In the Cape Town public sector, the workload and time constraints make it unlikely that comprehensive individual education will take place in the consultation. These group interventions focus on addressing self-care, health literacy, lifestyle changes and treatment adherence while acknowledging the personal, social and economic challenges patients face. Studies that show the effectiveness of group empowerment in diabetic management are mostly from highly resourced settings;⁴ however, the Group Empowerment and Training (GREAT) programme has established its cost-effectiveness in low socio-economic settings in South Africa.^4,9,11 The implementation of GREAT was previously evaluated in primary care facilities in five provinces in South Africa.¹²

The focus now is on how to implement GREAT for diabetes in various South African primary care contexts and ensure its sustainability and further scale-up. The GREAT programme has never been implemented in the Southern-Western Substructure (SWSS) of Cape Town. The key lessons learnt from previous implementation were considered in this study and were incorporated into a programme theory to guide implementation.¹² Key lessons included the need for a suitable venue, appropriately trained and sufficient facilitators, support from the whole facility and especially the facility manager, inclusion in clinical governance and health information systems and integration with patient flow and appointment systems.⁹ This study will add further evidence on how to implement GREAT for diabetes in a new substructure with its own contextual issues. The aim was to evaluate the implementation of GREAT for diabetes at Grassy Park Community Day Centre (CDC) in SWSS.

Research methods and design

Study design

This study used a convergent mixed methods design to evaluate the following implementation outcomes:

Qualitative data were collected to explore adoption, feasibility, fidelity and sustainability in the form of semi-structured individual and focus group interviews (FGIs). Quantitative data were collected to evaluate reach, cost and aspects of fidelity.

Study setting

Grassy Park CDC is in the southern suburbs of Cape Town and serves a community of 82 199 people of mixed ancestry who mainly speak English and Afrikaans.^13,14 In 2023, data from the Western Cape Single Patient Viewer (SPV) dataset indicated that 56% of patients seen at this facility with diabetes had uncontrolled glucose (HbA1C of > 8%). The CDC did not use a separate ‘club’ system to manage their patients with chronic diseases, and all such patients were seen by appointment within the general practice population. Patients were seen by clinicians, who were either clinical nurse practitioners (CNPs) or doctors. Controlled and uncontrolled patients were seen by both CNPs and doctors and managed according to the PACK guidelines.¹⁵ Clinicians offered brief health education during the consultation and patients that were referred to the dietician had a 2-month to 3-month waiting period. Diet sheets were given to patients if needed.

The intervention

Group empowerment and training was a group empowerment programme, training was focused on people with type 2 diabetes.¹⁴

The implementation strategies

The study focused on the first 6 months of implementation. The programme theory is shown in Figure 1 and was used to prospectively plan implementation.¹² This theory model guided the implementation strategies and anticipated contextual factors that could hinder successful implementation.

Health system structures

Support was provided by the clinical director of the substructure and facility manager of Grassy Park CDC as well as the local family physician. Grassy Park CDC had a substantial number of patients with type 2 diabetes and could easily identify people needing GREAT for diabetes. No additional financing was made available.

Health system inputs

The mosque opposite the facility was used as a venue at no extra cost. The CDC had a multidisciplinary team that was sufficient to implement GREAT with no extra staff members needed. Resource materials were obtained for free from another substructure in collaboration with Stellenbosch University. The implementation of GREAT was monitored by use of a registry for each patient attending each session. Referral to GREAT was documented in the medical record, and attendance was also noted in the folder in the form of a sticker.

Activities

Patients with type 2 diabetes, who had a HbA1c > 7%, and those newly diagnosed were referred to the programme. This programme was not intended for people with type 1 diabetes or gestational diabetes. Prior to training, the organisational aspects were discussed using a 10-question planning guide with facility managers and the clinical team responsible for diabetes care.¹⁷ Various members of the clinical team assisted with identifying suitable patients, drawing folders, providing the medication at the end of sessions and directing patients to the venue. The sessions were delivered monthly, which coincided with the medication collection dates.

Five facilitators (a CNP, a doctor, a nurse, a dietician and a post-basic pharmacy assistant) attended the 3-day training offered by the neighbouring substructure in June 2023 at no extra cost.

Feedback on the implementation of GREAT was given at the clinical governance meetings. The staff members responsible for implementation were encouraged to include their new responsibilities in their performance management by the the facility managers.

Qualitative data

A descriptive exploratory qualitative approach using semi-structured interviews was followed.

Healthcare workers sampling and sample size

This study used purposive sampling to select healthcare workers. The following 10 key informants were selected: (1) the clinical manager of the substructure; (2) the facility manager; (3) two clinicians; (4) the dietician; and (5) facilitators. The final sample size was determined by data saturation when the last two individuals did not identify new issues or potential themes.

Healthcare workers’ data collection

Semi-structured individual interviews were conducted with healthcare workers and lasted 30 min - 60 min. The interview guide explored the experience of the intervention, adoption, feasibility, fidelity, opportunity costs and sustainability. The guide was also informed by the Collaboration Framework of Implementation Research and previous interview guides from similar studies.^9,18

The interviews were conducted 6 months after the implementation of GREAT in the boardroom or consulting room at a mutually convenient time by a research assistant (RA). The interviews were audio-recorded, and field notes were made immediately after. The interviews were conducted in English, the official language of the health services.

Patients sampling and sample size

Patients were purposively selected for FGIs. Three groups were interviewed when they completed session four of the programme. Each focus group had between 9 and 12 patients. The final sample size was determined by data saturation when the last focus group did not identify new issues or potential themes.

Patients’ data collection

Semi-structured FGIs were used to explore patients’ experiences with the GREAT intervention using an interview guide. The FGIs explored the adoption, feasibility, fidelity, effect and overall patient experience of the programme. The guide was developed using the Collaboration Framework of Implementation Research and a previous study.^9,18 Interviews were conducted by the RA after the last session and lasted 30 min to 60 min in the same venue. These interviews were audio-recorded, and field notes were made. The interviews were mainly conducted in English with some parts in Afrikaans.

Qualitative data analysis

All interviews were transcribed verbatim by a professional transcriber and checked for accuracy by the researcher. The analysis was performed by the first author (ZS), under supervision. ATLAS.ti software assisted with the analysis using the framework method:¹⁹

Trustworthiness

Data were triangulated from healthcare workers and patients to improve credibility, and respondent validation was conducted with the staff. In terms of reflexivity, ZS was a medical officer at the facility and a novice researcher. She managed people with diabetes at the CDC and initially did not facilitate the sessions. Because of a change in the work profile of one of the facilitators, ZS had to step in and facilitate two out of the four sessions so that implementation could continue. She collected data via an RA to enable openness and honesty in the participants’ responses. Her analysis of the data was informed by her own experience, and her coherence with the data during interpretation was improved by external supervision. The RA held a B.Sc. in Food Science at Stellenbosch University and was a trained health coach. She was not affiliated with the CDC or its staff members.

Quantitative data

Fidelity

Each group of patients was observed once, and each type of session was observed twice. A structured tool was used to observe the GREAT sessions and evaluate fidelity to the content, group facilitation, communication style and skills (Likert scale from 1 rarely, 2 sometimes, 3 mostly, 4 always). This observation tool was validated for use in the original clinical trial¹⁴ and later adapted for use in the evaluation of GREAT in other provinces.¹² The RA was a non-participant trained observer, who completed the tool, audio-recorded the sessions and made field notes. The Likert scale was used to calculate mean scores for the various components of the guiding style and communication skills.

Reach

Data were collected from the session registers to evaluate the reach in terms of the number referred to GREAT, the number of groups, number of patients that attended and the number of sessions. The GREAT register was cross-checked with the facility’s records to validate the numbers and correct any omissions.

Costs

Eligible costs were identified by the researcher with the help of the key informant interviews. Costs included incremental set-up and operational costs to ascertain the monthly and annual costs for the facility. The time spent by staff in implementation was estimated from the interviews and these opportunity costs were calculated by determining the hourly rate per staff category that was provided by the human resources department. Information on unit costs was obtained from both the University and the Department of Health and Wellness.

Ethical considerations

The study was approved by the Health Research Ethics Committee at Stellenbosch University (S23/06/133), and permission was granted by the Department of Health and Wellness, Western Cape Government.

Findings

The characteristics of the 10 healthcare workers (HCWs) and 20 patients who participated in the interviews are summarised in Table 1 and Table 2. The themes and subthemes are summarised in Table 3.

Adoption, acceptability and appropriateness of the intervention

Burden of disease and need for the intervention

Many of the respondents were concerned at the increase of people with poorly controlled diabetes. One of the managers highlighted that the poor outcomes were because of a lack of innovation in managing diabetes:

Challenges included a high patient load and limited time, combined with stretched human resources and complex patients that needed skilled clinicians. Most providers were nurse practitioners with support from doctors. The volume of patients with complications, multimorbidity and poor glycaemic control were a major cost driver. Several respondents highlighted that if patients were not empowered then poor glycaemic control would continue. All the respondents affirmed the need to implement an empowerment programme that was cost-effective and feasible:

Alignment with health system goals

A few respondents commented that GREAT was aligned with the facility’s, substructure’s and health department’s goals:

Many healthcare workers reiterated that this programme would allow patients to take control of their health and make better-informed decisions:

Several respondents noted that if patients were better controlled, they could be followed up more in the community and reduce the workload in the facility. A more person-centred community-orientated approach was also aligned with policy goals:

Addressing gaps in management of diabetes

Many respondents noted that the lack of time to deliver effective education was one of the main issues that GREAT addressed. Several also mentioned that the group approach allowed more time to deliver education in a non-punitive manner. Some respondents commented that this programme would alleviate pressure on clinicians and fill a gap because of the lack of a health promoter. It was believed that GREAT could make use of the limited resources in an efficient manner:

Manager and staff buy-in

The leadership was aware that this programme was implemented successfully at other facilities and commented on the eagerness of clinicians for this programme:

However, there was some resistance because it was introduced as a research project for one person’s degree and not as an intervention driven by the management. Feedback was mixed on how much support there was among the staff. Some felt, including the managers, that the staff were positive and displayed great interest, while others said they were not well informed or motivated. Some respondents felt that there was management support, while others did not:

A few criticised the programme for only focusing on people with diabetes and not their family members. However, most respondents encouraged implementation. The facility manager highlighted positive feedback from patients and was happy that they benefited from the programme.

Perceptions of acceptability and appropriateness

Many respondents commented how GREAT was an appropriate intervention that promoted peer support, learning together, shared experiences and allowed participants to ask more questions and gain comprehensive information in a supportive environment. This programme was seen as more acceptable to deliver health education than traditional approaches:

The patients were very positive towards the intervention and highlighted how they found the sessions informative, comprehensive, in-depth and enjoyable. Overall, they found the programme acceptable and appropriate to help them manage their diabetes. They expressed their appreciation for the knowledge they gained and expressed a strong need for earlier intervention:

Reach

During the first 6 months, 54 patients were invited and 35 attended GREAT (64.8%) in four groups (Table 4). The majority attended more than half of the sessions although only 28.6% attended all four sessions. Most were pensioners (51.5%) with an average age of 59.5 (s.d. 12.3) years and had type 2 diabetes for more than 6 months (82.9%).

Feasibility and fidelity of the intervention

Coordination and integration of implementation processes: Management highlighted that even with a motivated champion, significant coordination between staff members and entities in the facility was required for implementation to be successful:

All the CNPs and doctors referred patients to the programme. Initially, it was a complex process to refer patients, and this was simplified over time. Even with the streamlining, it took extra time to refer, inform and motivate patients to attend:

In reception, a designated person was responsible for GREAT, and folders were kept in a special box for the duration of the sessions. The pharmacy played a crucial role in dispensing medication on the day of the session. They advised the clinicians to rewrite all the chronic scripts at the first session so that pre-packed medication would be ready at subsequent sessions. This decreased waiting time and was an incentive for people to attend. Dates for the sessions and for collecting medication needed to be aligned:

Management highlighted the importance of choosing a suitable and motivated facilitator with good communication skills. There were different views on whether the facilitator should be a clinician. On the one hand, clinicians were more knowledgeable and were able to address some of the participant’s individual clinical needs during or after the sessions. A clinician might have a greater impact on the participants and control of their diabetes. However, taking a clinician away from the queue of patients added pressure on other clinicians and might reduce the number of walk-in patients that could be seen. People debated the benefits of empowering patients versus the negative effects on the queue and consultations that day. Facilitators were put under some pressure to return as quickly as possible to their usual duties:

The venue was a community mosque directly opposite the facility that was previously used to provide services during the coronavirus disease 2019 (COVID-19) pandemic. The venue was free, acceptable and convenient for patients, and the health centre did not have a suitable space. Staff were concerned about being off-site and having to move medication and folders across the road. Security guards helped to deliver medication and direct participants. This choice of venue also aligned with the goal of providing more care in the community:

Patient factors

Reasons for declining the programme were lack of awareness, work commitments, transport issues, out-of-area patients, low perceived benefit, acute illnesses, other commitments and family responsibilities:

Impact on staff, time and material resources

Two clinicians, one nurse, the dietician and a pharmacy representative attended a 3-day training to be facilitators. Several respondents commented that it would benefit the facility if all suitable staff were trained, as this would allow flexibility for managers in the allocation of staff on the days the programme was conducted. Implementation required a multidisciplinary approach with inputs from clinicians, receptionists, pharmacy and facilitators. This had opportunity costs although the dietician and pharmacy respondent felt that this was part of their usual responsibilities:

Although five people were trained, one was rotated away to other duties and could not assist, one often had duties elsewhere and was not always available, one did not fully embody the guiding style and one was not able to deliver all the sessions. It was necessary to train additional facilitators to ensure the sustainability of the initiative. However, no one felt that additional staff should be employed:

There was some debate about combining the sessions into 2 days, although this would place more cognitive load on the participants and take facilitators away from clinical work for longer. However, it would reduce the number of visits by patients and might improve attendance. GREAT was still seen as an addition to routine care, rather than as part of such care:

All the other resource materials were provided during training and so there was no impact on consumables, human resources were the main requirement.

Session delivery

Facilitators generally took 2 h - 3 h to organise and deliver a session. The facilitators all thought that the structure and content of the sessions were easy for the staff and patients to understand and relate to. One of the clinicians had some issues with the nutritional section in terms of alternative dietary guidelines, and some patients found the identification of the different food groups challenging. Overall, both patients and facilitators appreciated the guiding style, the content that was delivered and the learning environment that was created:

Different sessions were allocated to different facilitators depending on their background. For example, the dietician conducted the nutrition session, the pharmacy representative conducted the medication session, and the clinicians conducted the sessions on understanding diabetes and complications:

The facilitators and patients appreciated the flipcharts and materials used during the sessions, which were relatable and easy to understand. The facilitators found their manual to be comprehensive and easy to understand. The facilitators added fun icebreakers to the sessions to enhance the learning environment:

Facilitators were flexible when patients missed their sessions and booked them for other groups. However, despite the positive feedback from patients and flexibility of facilitators, there were patients who missed sessions because of administrative errors, family responsibilities, work, weather conditions and acute illnesses. Some patients received medication collection dates that did not coincide with their session dates:

Fidelity to the intervention design

Eight sessions were observed by the RA, and Table 5 shows the scores for the facilitators’ communication style and skills. Facilitators performed well in adhering to the guiding style and communication skills. The highest scores were for empathy in understanding the perspective and context of participants and for the use of open questions. Facilitators mostly demonstrated respect for participants’ choice and control, evoked ideas and solutions from the group, used active listening and enabled collaboration. The lowest score was for exchanging information using the elicit-provide-elicit model.

Facilitators adjusted the time needed to deliver the content depending on the group, for example older adults might need longer. In the observed sessions, 80% of the intended content was delivered. Omitting content could be because of time constraints or a lack of preparation:

The clinical and social effects of the intervention

A few participants noted significant reductions in their sugar levels, and some mentioned how their symptoms improved. Some noted that they were more conscious of their dietary habits, particularly portion size, incorporating more vegetables in their diet and less sugar-containing foods. They were also more adherent to their medication and increased physical activity. Some of them mentioned how they advocated for a healthier lifestyle among their families and friends. Overall, the intervention empowered the participants to adopt healthier lifestyle behaviours:

Managers wanted to document improvement in HbA1c to ensure that the programme was achieving the expected effects. Facilitators observed that some of the patients progressed, and improvements were more evident among those who attended more sessions. One of the senior clinicians indicated that there was already feedback from patients that the programme was effective.

Cost

Incremental and opportunity costs of the intervention: Box 1 summarises the incremental and opportunity costs. During the start-up phase, the incremental costs were R6683 and the opportunity costs R38 561 for the time spent training. The incremental operational costs were trivial at R30/month, and the opportunity costs for the staff time were estimated at R26 252/month or R315 024/year. According to most of the respondents, the short- and long-term benefits of this programme outweighed the opportunity costs. If glycaemic control improved, then patients could attend the facility less often, and there would be a substantial saving on resources and staff time at the facility. Overall, most of the respondents agreed that this implementation was cost-effective:

The sustainability of group empowerment and training

Ongoing commitment to the initiative would depend on confirming the benefits to patients in their HbA1c results. Respondents felt there was a need to monitor the quality and fidelity of the initiative over time and that it would be necessary to train more staff:

There was some debate on the ongoing accessibility and convenience of the community venue. More awareness of the initiative was needed among patients to increase demand and commitment. This could be assisted by community-based services and the patients who attended the sessions. More flexible session times might be needed to improve reach and exploration of the reasons for non-attendance. Messaging services could be used to remind patients:

The GREAT programme needed to be integrated into routine care and the logistical and administrative processes streamlined. Ongoing managerial support would be essential and maintaining awareness among the staff. The programme also needed to scale-up and improve its reach. It should be embedded within a comprehensive approach to preventing diabetes and treating people with diabetes within the community and facility. The GREAT programme could also be extended to non-government organisations that run support groups in different venues and offer services via community health workers:

Discussion

Summary of the key findings

The key findings are summarised in an Implementation Research Logic Model (IRLM) as shown in Figure 2.²⁰ The barriers and enablers to implementation are summarised as contextual factors as well as the extent to which the implementation outcomes were attained. The implementation strategies and intervention itself, as outlined in the methods, are also included to give a complete picture of implementation. The discussion of the key findings focuses on the implementation outcomes and contextual factors.

Discussion of key findings

The key determinant of implementation was the support and commitment of management at the facility and substructure levels. Even though attempts were made to enable this by informing them timeously, discussing the 10 planning questions for management,¹⁷ and attaining buy-in, there was still ambivalence. Although the senior leadership of the health system promotes a model of disseminated authority and decision making within a conceptual model of a complex adaptative system, the reality may be different.²¹ The organisational culture is still quite hierarchical and reluctant to innovate without a clear mandate from above.

How do we change the organisational environment? This seems to be key in supporting the implementation of innovations.¹⁷ Leaders in primary care may need to transform their leadership style so that the organisational culture is more collaborative and open to experiment.²² Leadership development is key to transforming and inviting change.²³ This is a crucial factor to be cognisant of when considering upscaling at other facilities. This could alleviate barriers related to teamwork, logistics, staff allocation and commitment. The management team with the researcher looked at lessons learned from previous studies with the implementation of GREAT and found similar barriers despite trying to mitigate these factors. This suggests that the underlying factor remains the leadership and team dynamics in implementing new programmes in primary care.

The support of the family physician was evident, but they were not actively involved in the implementation process. The role of family physicians and senior clinicians is integral to leadership and governance in the district health system, and utilising their skills and expertise in the implementation of GREAT has been noted elsewhere.^17,24

The reach of the programme is related to the awareness and understanding of staff members, their motivation to promote it and the number of groups that can be conducted.¹⁵ The correlation between healthcare worker awareness and the uptake of patients to group empowerment programmes was noted elsewhere.²⁵ Clinicians are the preferred choice to refer patients to group diabetes programmes.^25,26 On average, 65% of referred patients attended the programme, which could be attributed to work commitments, unemployment, family responsibilities, low perceived benefit and other health issues.^25,26

Several studies indicated that peer support from other patients or community health workers can increase retention, especially among the middle-aged population.^27,28,29 The positive impact of integrating group diabetes education into community-based services was evident in other studies.^17,27 Community health workers could be trained as facilitators for people in the community and decrease the burden of utilising health professionals, especially in low-resourced settings. Although community health workers would have less expertise in diabetes and might struggle to address all the questions in the group. In addition, community-based services tend to target the people with better glycaemic control. Similar reasons for poor retention of patients were found in other group programmes for diabetes.²⁷ One barrier to patient retention is the misalignment of sessions and medication collection dates.⁹

Suitable facilitators were key to ensure session quality and retention of patients into the programme. The patients who attended the GREAT sessions highlighted how they valued the communication style and skills, the non-judgemental approach and the engaged environment. Using a clinician as a facilitator in this study allowed them to answer complex questions and adjust the prescriptions, if needed, to prevent clinical inertia. This showed a greater impact on glycaemic control, empowering patients with knowledge to elicit self-care and through the regular review of their glucose readings and medication adjustments.

Many facilities do not have space for group education.¹⁷ The problem was solved by using a community venue, although this created some logistical issues for staff to work outside of the facility. Managers saw the collaboration with the mosque as a positive example of community participation that was aligned with the broader policy on community-orientated primary care.^30,31 The venue also appeared acceptable to patients, and the use of more accessible community venues can improve retention.^25,32

This study evaluated implementation in a resource-constrained context, although studies in more developed countries and other ethnic groups observed similar barriers despite having more resources at their disposal.^25,27 The impact on staff resources and the opportunity costs might be a deterrent for lower-income and resource-constrained settings. In South Africa, in 2009, it was estimated that diabetes caused 2000 amputations annually.³³ According to a previous study, the estimated mean cost per patient for lower limb amputation in South Africa in 2020 was about $21441.³⁴ The total cost per patient that attended the GREAT programme was estimated to be $42 per month or $11 per session and about $72 per person for the entire programme (1$ = ZAR17.9). The GREAT programme therefore is a cost-effective strategy in the South African context,^4,13 but managers must still decide whether the longer-term benefits warrant the immediate opportunity costs. Better glycaemic control translates into fewer complications and hospitalisations, better quality of life and less workload for the facility in the longer term as patients can be seen less frequently and supported more via community-based services. It may be difficult, however, for managers to take such a long-term view as the opportunity costs are more immediately tangible.

Ongoing management and staff buy-in will depend on seeing the benefits of GREAT reflected in improved HbA1c results. Awareness among patients, a suitable venue and integration of the programme into the model of care are also important. Increased uptake and retention of patients in the programme could be enhanced by peer support from previous participants²⁷ and use of electronic reminders.²⁵ The effectiveness could be reinforced using digital solutions, such as a WhatsApp Chatbot.³⁵ Training additional staff members to facilitate will help spread the load and ensure sustainability. Training could also be reinforced using videos to model and refresh what is expected as well as ongoing monitoring and feedback.

Strengths and limitations

The mixed-method study design provided a comprehensive evaluation of the implementation outcomes and determinants. This evaluation was strengthened by the triangulation of quantitative and qualitative data, including data from patients and healthcare workers as well as non-participant observations. The IRLM provided a useful structure to make sense of the findings. The detailed description of the context and participants should enable the transferability of the findings to similar contexts in the Western Cape and other parts of South Africa.

The researcher had to facilitate some of the sessions because of facility needs and thus became an embedded researcher. On the one hand, this experience enhanced the interpretation of data, but on the other hand, the researcher had to strengthen her reflexivity to prevent her own experiences unduly influencing the interpretation of data. Interpretation of the data was supervised by an experienced researcher in qualitative research, which mitigated this limitation. Observations were also conducted by an independent RA.

Recommendations

Conclusion

Group empowerment and training was adopted by all key stakeholders as an intervention that was valued and would decrease the burden of diabetes at the facility. They viewed GREAT as a programme that would have a favourable return on investment. Both staff and patients reported that the programme significantly impacted understanding of diabetes and self-management and provided a supportive and engaging environment. The group dynamics, communication style, incentives, easy access to the programme, utilising suitable facilitators from a multi-disciplinary team, the pivotal roles of management and staff buy-in, were key factors in the success of its implementation. Despite the small number of patients that this programme reached over the initial 6 months, the value of empowering patients and the cost per patient was seen as a justifiable investment and reason to sustain and scale-up the intervention.

Acknowledgements

The authors would like to thank the participants of this study, the Grassy Park CDC facility, the Southern Western Substructure and the University of Stellenbosch in the implementation of GREAT. The authors would like to thank Ms Mariaan von Pressentin who conducted the observations and collected qualitative data. While their contribution was important, it did not meet the criteria for authorship.

Robert Mash and Darcelle Schouw are leading the scale-up of GREAT for diabetes in the Western Cape and other provinces of South Africa.

Zahraa Saban: Conceptualisation, Formal analysis, Funding acquisition, Investigation, Methodology, Visualisation, Writing – original draft, Writing – review & editing. Darcelle Schouw: Co-supervised all aspects of the work. Robert J. Mash: Supervised all steps in the research process. All authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication and take responsibility for the integrity of its findings.

The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article. This work was supported by funds in the Division of Family Medicine and Primary Care at Stellenbosch University.

Data can be made available by the corresponding author, Robert J. Mash, on reasonable request if participant consent allows.

The views and opinions expressed in this article are those of the authors and are the product of professional research. They do not necessarily reflect the official policy or position of any affiliated institution, funder, agency or that of the publisher. The authors are responsible for this article’s results, findings and content.

References

IDF Diabetes Atlas 10th edition [homepage on the Internet]. [cited 2023 Jan 28]. Available from: www.diabetesatlas.org
P0309. Statistical release mortality and causes of death in South Africa: Findings from death notification 2020 [homepage on the Internet] [cited 2025 Jul 04]. Available from: www.statssa.gov.za,info@statssa.gov.za
Boake M, Mash R. Diabetes in the Western Cape, South Africa: A secondary analysis of the diabetes cascade database 2015–2020. Prim Care Diabetes. 2022;16(4):555–561. https://doi.org/10.1016/j.pcd.2022.05.011
Mash R, Kroukamp R, Gaziano T, Levitt N. Cost-effectiveness of a diabetes group education program delivered by health promoters with a guiding style in underserved communities in Cape Town, South Africa. Patient Educ Couns. 2015;98(5):622–626. https://doi.org/10.1016/j.pec.2015.01.005
Mogueo A, Oga-Omenka C, Hatem M, Kuate Defo B. Effectiveness of interventions based on patient empowerment in the control of type 2 diabetes in sub-Saharan Africa: A review of randomized controlled trials. Endocrinol Diabetes Metab. 2021;4(1):e00174. https://doi.org/10.1002/edm2.174
Mash B, Levitt N, Steyn K, Zwarenstein M, Rollnick S. Effectiveness of a group diabetes education programme in underserved communities in South Africa: Pragmatic cluster randomized control trial. BMC Fam Pract. 2012;13:126. https://doi.org/10.1186/1471-2296-13-126
Ağralı H, Akyar İ. The effect of health literacy-based, health belief-constructed education on glycated hemoglobin (HbA1c) in people with type 2 diabetes: A randomized controlled study. Prim Care Diabetes. 2022;16(1):173–178. https://doi.org/10.1016/j.pcd.2021.12.010
Duke SAS, Colagiuri S, Colagiuri R. Individual patient education for people with type 2 diabetes mellitus. Cochrane Database Syst Rev. 2009;1:CD005268. https://doi.org/10.1002/14651858.CD005268.pub2
Serfontein S, Mash RJ. Views of patients on a group diabetes education programme using motivational interviewing in South African primary care: A qualitative study. S Afr Fam Pract. 2013;55(5):453–458. https://doi.org/10.1080/20786204.2013.10874395
Skinner TC, Carey ME, Cradock S, et al. Diabetes education and self-management for ongoing and newly diagnosed (DESMOND): Process modelling of pilot study. Patient Educ Couns. 2006;64(1–3):369–377. https://doi.org/10.1016/j.pec.2006.04.007
Allerton J, Mash R. The impact of intensified clinical care on glycaemic control in patients with type 2 diabetes at Khayelitsha Community Health Centre, South Africa: Quasi-experimental study. Prim Care Diabetes. 2020;14(2):97–103. https://doi.org/10.1016/j.pcd.2019.08.006
Mash R, Schouw D. Evaluating the implementation of Group Empowerment and Training (GREAT) for diabetes in South Africa: Convergent mixed methods. 2023. https://doi.org/10.21203/rs.3.rs-2529013/v1
Grassy Park population [homepage on the Internet]. [cited 2023 Mar 04]. Available from: https://www.citypopulation.de/en/southafrica/cityofcapetown/199040_grassy_park/
Botes AS, Majikela-Dlangamandla B, Mash R. The ability of health promoters to deliver group diabetes education in South African primary care. Afr J Prim Health Care Fam Med. 2013;5(1):a484. https://doi.org/10.4102/phcfm.v5i1.484
Knowledge Translation Unit. Practical approach to care kit (PACK) – Primary care guide for the adult. University of Cape Town and Department of Health and Wellness: Cape Town: Western Cape Edition; 2021.
Buyelwa Majikela-Dlangamandla. The diabetes toolkit. Cape Town: SunMedia; 2017.
Mash RJ, Schouw D. The sustainability of group empowerment and training for people with diabetes in South Africa. S Afr Fam Pract. 2024;66(1):a5918. https://doi.org/10.4102/safp.v66i1.5918
Collaboration Framework in Implementation Research [homepage on the Internet]. [cited 2023 Apr 08]. Available from: https://cfirguide.org/guide/app/#/
Pope C. Qualitative research in health care: Analysing qualitative data. BMJ. 2000;320(7227):114–116. https://doi.org/10.1136/bmj.320.7227.114
Smith JD, Li DH, Rafferty MR. The implementation research logic model: A method for planning, executing, reporting, and synthesizing implementation projects. Implement Sci. 2020;15(1):84. https://doi.org/10.1186/s13012-020-01041-8
Welcome-Family Medicine in the Western Cape SAAFP 26th Annual Conference Dr K Cloete Overview 2. 2024 [cited 2025 Nov 28]. Available from: https://saafp.org/wp-content/uploads/2024/09/Dr-Keith-Cloete-_SAAFP-Welcome-Address-8-September-2024.pdf
Mash R, De Sá A, Christodoulou M. How to change organisational culture: Action research in a South African public sector primary care facility. Afr J Prim Health Care Fam Med. 2016;8(1):a1184. https://doi.org/10.4102/phcfm.v8i1.1184
Gilson L, Daire J. Leadership and governance within the South African Health System. S Afr Health Rev. 2011;1:70–80.
Mash R, Blitz J, Malan Z, Von Pressentin K. Leadership and governance: Learning outcomes and competencies required of the family physician in the district health system. S Afr Fam Pract. 2016;58(6):232–235. https://doi.org/10.1080/20786190.2016.1148338
Dallosso H, Khunti K, Gray LJ, et al. Implementation of a diabetes prevention programme in a multi-ethnic community in primary care in England: An evaluation using constructs from the RE-AIM Framework. Prim Care Diabetes. 2023;17(4):309–313. https://doi.org/10.1016/j.pcd.2023.05.001
Harris SM, Joyce H, Miller A, Connor C, Amiel SA, Mulnier H. The attitude of healthcare professionals plays an important role in the uptake of diabetes self-management education: Analysis of the barriers to uptake of type 1 diabetes education (BUD1E) study survey. Diabetic Med. 2018;35(9):1189–1196. https://doi.org/10.1111/dme.13704
Sinclair KA, Zamora-Kapoor A, Townsend-Ing C, McElfish PA, Kaholokula JK. Implementation outcomes of a culturally adapted diabetes self-management education intervention for Native Hawaiians and Pacific islanders. BMC Public Health. 2020;20(1): 1579. https://doi.org/10.1186/s12889-020-09690-6
Werfalli M, Werfalli M, Raubenheimer PJ, et al. The effectiveness of peer and community health worker-led self-management support programs for improving diabetes health-related outcomes in adults in low- and-middle-income countries: A systematic review. Syst Rev. 2020;9(1):133. https://doi.org/10.1186/s13643-020-01377-8
Community Preventive Services Task Force. Diabetes management: Interventions engaging community health workers 2017 [cited 2025 Nov 28]. Available from: https://www.thecommunityguide.org/findings/diabetes-management-interventions-engaging-community-health-workers.html
Western Cape Government: Health and Wellness. Position statement on community-oriented primary care (COPC) towards universal health coverage (Circular H11/2023). Cape Town: Western Cape Government Health; 2023.
Mash R, Goliath C, Mahomed H, Reid S, Hellenberg D, Perez G. A framework for implementation of community-orientated primary care in the Metro Health Services, Cape Town, South Africa. Afr J Prim Health Care Fam Med. 2020;12(1):1–5. https://doi.org/10.4102/phcfm.v12i1.2632
Willis A, Roshan M, Patel N, et al. A community faith centre based screening and educational intervention to reduce the risk of type 2 diabetes: A feasibility study. Diabetes Res Clin Pract. 2016;120:73–80. https://doi.org/10.1016/j.diabres.2016.07.025
Pheiffer C, Pillay-Van Wyk V, Joubert JD, Levitt N, Nglazi MD, Bradshaw D. The prevalence of type 2 diabetes in South Africa: A systematic review protocol. BMJ Open. 2018;8(7):e021029. https://doi.org/10.1136/bmjopen-2017-021029
Mtshali S, Hongoro C, Mahomed O. Hospitalization costs for diabetes related lower extremity amputation at a referral hospital in KwaZulu Natal, South Africa. Curr Diabetes Rev. 2022;18(9):e020222200776. https://doi.org/10.2174/1573399818666220202153336
Mash R, Schouw D, Fischer AE. Evaluating the implementation of the GREAT4Diabetes WhatsApp Chatbot to educate people with type 2 diabetes during the COVID-19 pandemic: Convergent mixed methods study. JMIR Diabetes. 2022;7(2):e37882. https://doi.org/10.2196/37882
Gagliardi AR, Alhabib S. Trends in guideline implementation: A scoping systematic review. Implement Sci. 2015;10:54. https://doi.org/10.1186/s13012-015-0247-8

Original Research

Evaluating the implementation of group empowerment for people with type 2 diabetes at Grassy Park Community Day Centre, Cape Town: A convergent mixed-methods study

Zahraa Saban, Darcelle Schouw, Robert J. Mash

Abstract

Introduction

Research methods and design

Study design

Study setting

The intervention

The implementation strategies

Health system structures

Health system inputs

Activities

Qualitative data

Healthcare workers sampling and sample size

Healthcare workers’ data collection

Patients sampling and sample size

Patients’ data collection

Qualitative data analysis

Trustworthiness

Quantitative data

Fidelity

Reach

Costs

Ethical considerations

Findings

Adoption, acceptability and appropriateness of the intervention

Burden of disease and need for the intervention

Alignment with health system goals

Addressing gaps in management of diabetes

Manager and staff buy-in

Perceptions of acceptability and appropriateness

Reach

Feasibility and fidelity of the intervention

Patient factors

Impact on staff, time and material resources

Session delivery

Fidelity to the intervention design

The clinical and social effects of the intervention

Cost

The sustainability of group empowerment and training

Discussion

Summary of the key findings

Discussion of key findings

Strengths and limitations

Recommendations

Conclusion

Acknowledgements

Competing interests

Authorship contribution

Funding information

Data availability

Disclaimer

References