Challenges to establishing successful partnerships in community health promotion programs: local experiences from the national implementation of healthy eating activity and lifestyle (HEAL™) program

Community-based health promotion programs addressing physical activity and diet are important strategies for improving the health of the population and reducing risk factors for chronic diseases such as diabetes.¹ Implementation of such programs is an ongoing challenge and there are few publications describing the experiences of implementing community-based programs in Australia.² The socio-ecological framework for health promotion emphasises the importance of developing and maintaining multidisciplinary partnerships and these relationships have been shown to be important to the success of community-based programs.³ NSW Health developed a ‘Framework for Building Capacity to Improve Health’ which has informed the Victorian Healthcare Association’s population health planning framework. There are five components of this framework but partnerships and leadership underpin the organisational change, workforce development and resource allocation.^4,5 Nutbeam et al. emphasise the importance of existing relationships and organisational capacity and support for the successful implementation of programs, and that if these are not in place action is unlikely.⁶ For policy makers, seeking to understand how community-based health programs are implemented is important to understand how such partnerships are formed and sustained so that lessons can be learnt and applied to the funding of new initiatives.⁷

The academic literature highlights the importance of partnerships to the success of health promotion interventions.⁸ Bors et al. report on the implementation of an active living program in several different communities in the USA.³ They used a socio-ecological framework and explored the differences and successes according to the 5P strategies (preparation, promotions, programs, policy and physical projects). Partnerships, especially existing ones, were important to the success of the projects in the different areas; in particular, successful projects had strong leadership and local champions that reflect the recommendations of frameworks such as those developed by NSW Health.^4,6 Leadership was challenging because of the time involved and the need to balance the competing interests of the different partners. Flexibility was important, as was establishing a sense of mutual accountability and benefit. There were similar findings from an evaluation of interventions in Australia,⁷ where large variations were found in the local partnerships and their ability and capacity to implement the program. In rural and remote areas, this was compounded by workforce capacity and the evaluation highlighted the need for flexibility in the way in which these community-based interventions are funded.

The partnership synergy framework proposed by Lasker et al.⁹ aims to better understand the concept of synergy as an outcome measure of the functioning of the partnership – if the partnership is effective then the synergy that is created is greater than the sum of its individual parts. Their framework for synergy includes five broad components. These components are resources, characteristics of the partner and partnership, relationships and the external environment. Resources include connections to groups, organisations and people and the ability to bring these groups together, and convening power. Partner characteristics are their perceptions of the benefits or disadvantages to them of being involved in the project. Relations include trust and respect among the partners, power and potential for conflict. The partnership characteristics focus on governance, leadership and efficiency and this exists in the context of the external environment. This framework draws heavily on the theoretical literature and has been tested in several partnerships where leadership and efficiency were found to be most related to partnership functioning.¹⁰

The Healthy Eating Activity and Lifestyle (HEAL™) program is an initiative of South Western Sydney Medicare Local (SWSML). The goal of HEAL™ is to reduce overweight and obesity and increase physical activity, particularly in those community members who are at risk of developing lifestyle diseases such as cardiovascular disease and type-2 diabetes. HEAL™ is facilitated in a group environment where people can learn and share experiences with the aim of empowering clients to take responsibility for their own health by focusing on behaviour modification principles in the community setting. The program focuses on combining nutrition, physical activity and psychology by addressing: behaviour modification and goal setting, benefits of being physically active and physical activity opportunities, nutrition education, label reading, recipe modification and low-fat cooking techniques, eating in a social environment (take-away meals and eating out), skills for maintaining a healthy lifestyle and psychosocial issues of eating.

The HEAL™ intervention model is that health professionals, usually exercise physiologists and dieticians, who have expertise in providing evidence-based advice on physical activity and healthy eating, provide an eight-week group program of support to HEAL™ participants. HEAL™ has been accredited as a lifestyle modification program by the Australian Government-funded Healthy Living Network, having met the quality standards of the Healthy Communities Quality Framework.

SWSML in collaboration with Exercise & Sports Science Australia (ESSA) received funding though the Australian Government’s Healthy Communities Initiative (HCI) to roll out the HEAL™ Program nationally, to 67 of the 92 HCI Local Government Areas (LGAs) between 1 July 2010 and 31 May 2013. The roll-out of HEAL™ in the LGAs was undertaken in collaboration with Healthy Communities Coordinators (HCC) employed by the local councils and was successful. The number of programs implemented (n = 297) exceeded the target number (n = 275) and those people who completed the program had significant improvements in all outcome measures.¹¹ The model for making HEAL™ available across these LGAs was to invite the HCC to identify interested tertiary-educated allied health professionals, usually accredited exercise physiologists (AEPs), dieticians or physiotherapists, to be trained as HEAL™ facilitators and to provide the program. As a result 310 allied health professionals were trained. Using the database held by ESSA the HEAL™ project team staff also offered to assist the HCCs to identify AEPs within their LGA if required. Across the LGAs there has been variation in the implementation of HEAL™, the number of programs delivered and their sustainability.

The aim of the qualitative study was to explore in more detail the experiences and opinions of key stakeholders that have taken part in the HEAL™ program. The study sought to explore the local adaptations required and the barriers and facilitators to the implementation of the program. In particular, we were interested in the experiences and views of key stakeholders about what influenced or challenged the development and maintenance of such partnerships.

A qualitative study was undertaken to collect information about the local implementation of the HEAL™ Program from those who had taken part; the HCC and the HEAL™ facilitators. Purposive sampling was used to select LGAs to take part in the qualitative interviews. The project team identified LGAs that represented a range of characteristics including socioeconomic indices, cultural and linguistic diversity, urban, regional and rural location. The LGAs were also chosen because there was local innovation of the HEAL™ program to meet the needs of the population. LGA performance data was returned to the national HEAL™ coordinators every six months for inclusion in their reporting to government against project milestones. Both quantitative and qualitative data were gathered and from this information the LGAs under study were selected. The LGAs were innovative because in most cases they had formed partnerships with key local stakeholders or organisations to implement the program and make it available to disadvantaged populations, for example, Aboriginal people through a local football team, people with intellectual disabilities and their carers, people with mental health problems who were also at high risk of developing type-2 diabetes. Purposive sampling identified four LGAs for this project in remote Western Australia, rural New South Wales, rural Victoria and the ACT.

Once identified, invitation letters to take part in semi-structured telephone interviews were sent to the HCCs (n = 4) and HEAL™ facilitators (n = 6) by the HEAL™ project team. Once the researchers had received the written informed consent, participants were contacted to make an appointment for a telephone interview. All agreed to take part in the interviews and were interviewed. The characteristics of the four LGAs are detailed in Table 1.

Table 1.  Characteristics of the HEAL™ sites

All interviews were conducted over the telephone by two of the researchers (SD, OH) and were digitally recorded with the participants’ informed consent, transcribed and analysed thematically. A copy of the interview schedule is included as an appendix. The research team held regular meetings to discuss every stage of the study. The interview schedule was slightly modified after the interviews with people from the first two LGAs, following preliminary thematic analysis and initial coding. Data were managed using NVivo 10 software (NVivo qualitative data analysis software; QSR International Pty Ltd Version 10, 2012). Initial coding was undertaken by one researcher (SD) and discussed with the research team to enable consensus with regards to the themes emerging.¹² As partnerships emerged as a key theme the literature on this topic was searched to further develop the framework for this core theme.

The study was approved by the Human Research Ethics Committee of the University of New South Wales (HREC 13072).

The key themes emerging from the analysis centred on the characteristics of partnerships including the time and effort required to develop and maintain these partnerships, the needs of the partners and key people in those partner organisations. The findings are organised according to Lasker’s determinants of partnership synergy.⁹ Their framework for synergy includes five broad components which are: resources, characteristics of the partner and partnership, relationships and the external environment.

Partnerships with several different stakeholders were important to the successful implementation of HEAL™ in the four LGAs. The four HCCs worked in very different areas and faced very diverse challenges. In all four areas, a variety of partnerships with health services, disability services and Aboriginal medical services were important in developing interest in the HEAL™ program and establishing it in their community. Differences between the LGAs were in the challenges they faced in establishing the partnerships and in the more remote areas these partnerships were slower to establish and more difficult to maintain. These partnerships sometimes developed from other programs of work or through key people where there were existing relationships:

a key partnership was with each of the Health Services and in the case of [the town] there was a very good relationship between the council and the health service and that is due to the fact that the manager of the health promotion unit and all of the functions that came under them and dietetics. She was extremely proactive in a whole range of council activity … (HCC 4)

a local champion certainly helped and in some of our towns’ community resource centres, we formed strong partnerships with them and they had better local knowledge. (HCC 1)

Meeting the needs of the partner organisation with the HEAL™ program was both an enabler and a barrier to the implementation of the program. Where the program was perceived to meet the needs of an organisation or contributed to their key performance indicators the program was much more likely to be implemented and have the support of the organisation. In rural NSW details of patients referred to community health are entered into a NSW Health computer system called CHIME. Once in the system, the health needs of these patients should be addressed and they saw HEAL™ as a way of meeting this requirement, so there was buy-in from the start. An additional advantage to the local health service was that their staff received the HEAL™ training and so this was seen as a capacity-building exercise that might result in the sustainability of the program.

Some of the partners approached the HCC because the HEAL™ program delivered what they were looking for with their clients. For example, in one LGA the providers of the mental health services were looking to improve the physical health of their clients as well as their mental health. So again this acted as a facilitator or opportunity as it enabled them to meet their service delivery requirements:

but they were the ones that asked us. They were looking for [a] nutrition and physical activities program and HEAL™, which was a program that had both of these components, was absolutely fantastic. (HCC 3)

We deliver a cardiac rehab session. But there was nothing else for them to go to after that. So the HEAL™ program really fitted the bill quite well there. (Facilitator 2)

In one of the areas the partnering health services did not initially see that the HEAL™ program could meet their service delivery requirements. It was seen more as a burden on their staff in an area with workforce shortages rather than an opportunity for capacity building. However, over time, this health service realised how the program could enable them to meet some of their service delivery goals and did finally engage much more with the program, although by this stage the Healthy Communities funding had come to an end.

The partnerships were strategic, gave them insight into local knowledge or the local community, and also provided an opportunity to advertise to increase recruitment of participants into the program. The HCCs all described building on existing capacity and partnering with organisations that were already providing services for people who were disadvantaged and were key target groups for the program. This included disability care services so that people with a disability and their carers could be recruited to HEAL™, mental health service providers, organisations providing training and support for unemployed people through which the HEAL™ program was able to access gymnasium spaces in higher education colleges, Aboriginal health service providers and organisations providing English language support for migrant populations:

What we’re doing is we’ve worked with the organisations and professionals who are already working with the target group and taking the programs to them, and that’s been a lot more successful rather than trying to advertise a program on its own. (HCC 2)

In spite of their success in delivering the program to specific population groups, they were not always as successful in capacity building of the local workforce. Several Aboriginal health workers in one LGA completed the HEAL™ training but did not feel confident to run the program locally without one of the other HEAL™ facilitators present.

The partnerships took time to develop and more time was required in the rural and remote areas compared with the less remote areas:

the partnerships were slow to develop and we probably started delivering before we’d really embedded our partnerships. (HCC 1)

It was a requirement of the funding that the LGAs set up an advisory committee with representatives from a range of organisations such as the Medicare Local, local health services, GPs and NGOs. In WA, this was challenging because of the distances involved which meant that opportunities to strengthen the partnerships or for informal information sharing were more limited:

when they have to travel 40 or 50 kms for a meeting at the end of a long work day might not take advantage of networking opportunities. (HCC 1)

This meant that the word-of-mouth discussion of the program and its successes were slower to permeate through the system. In less geographically challenging areas the advisory committees were useful in engaging with key partner organisations from several settings. Examples of key partner organisations included: Aboriginal medical services, local area health districts, drug and alcohol recovery therapeutic communities, sport and recreation centres, multicultural health services, refugee support centres and university health clinics:

We meet about every 4 months and it’s some in government and it’s non-government organisations as well and not exactly a community representative but a representative from one of the free meals programs that has a lot of contact with people and brings us down to earth and things but they’ve been fabulous in helping to keep us sort of targeted. (HCC 3)

In the more remote areas workforce capacity was a barrier to implementation of the program because it took longer to build the partnerships with health professionals who are already stretched trying to provide services to a large geographic area. By the time they were finally on-board the funding for the program had almost run out:

It took a while for people in the community – well, more like the health profession in the community like Allied Health staff to really embrace it. And it’s only now when I think they’ve seen, maybe, some of their clients and the positive effects, people are starting to ring us and refer to us or to find out how you can get involved and it’s almost like well, it’s two years down the track [laughs]. It’s too late. (HCC 1)

Barriers were often related to financial resources and included reluctance from private allied health providers to engage in the program or undertake the training so that they could provide the program. Allied health professionals working for some local health services delivering the program were easier to engage and train because it was seen as part of their health service role. However, in WA there were workforce shortages and it took some time for the health professionals to realise that this was an opportunity as opposed to yet another demand on their time. Allied health professionals in private practices such as exercise physiologists and dieticians who were eligible for the Medicare rebate to provide group services to people with type 2 diabetes did not always see HEAL™ as a viable business opportunity:

one of the EPs I’ve trained and he has tried to implement it but he says that unless you give up working hours and if you look at the Medicare rebate it is just not cost effective for the amount of time that he would need to put into running the program for what he would get back through those rebates. (Facilitator 3)

Physiotherapists were trained to deliver the HEAL™ program in some LGAs, especially those where there was a shortage of exercise physiologists, and they would not be eligible to claim the group services rebate from Medicare. This has major issues for the long-term sustainability of such programs and not all allied health professionals would be able to claim the Medicare rebates.

The partnerships were not easy to maintain and people talked about repeatedly meeting with certain groups or organisations and working hard to develop those relationships and trust in the HEAL™ program:

it is just constantly being working with all those other groups that are working specifically with those population groups and just reminding them what we were doing or when programs were coming up and using them to sort of help communicate how people could get involved and their benefits. (HCC 4)

This was particularly the case in engaging with the local Aboriginal community and this was further complicated by high staff turnover at the Aboriginal health service. One of the facilitators described how their HCC ‘went out of their way to go regularly to the local Aboriginal medical centre and brief the workers there on what was happening’ (Facilitator 2):

yeah so it really has to be more of a narrative and a discussion and a chat and a yarn about these things and you do have to establish a lot of rapport with the community … It does take time to establish and build it. (HCC 2)

Sometimes the timing of the partnerships coincided with another event, which then became the catalyst or opportunity for the program. An example was the timing of a NSW rugby league tournament, which provided a perfect opportunity to engage local Aboriginal men in the program:

It was a weight loss competition and it was eligible to communities who had teams participating in the New South Wales Aboriginal rugby league knock out. So that was just something we used to help to drive that within the community. (HCC 2)

All the HCCs reported a lack of engagement with local GPs in spite of GP membership on the local advisory groups. Referrals from GPs to the program were generally low and it was only once the HEAL™ programs had been running for some time that they finally started to refer their patients. Successful local implementation was dependant more on engagement with other local health professionals:

We found out fairly early that attempting to engage with GPs to set up a referral mechanism that way probably wasn’t going to be successful. (HCC 4)

We had one [GP] on the steering committee who made an appearance three times but hasn’t since then. (HCC 2)

Even if the program seemed to meet the service delivery needs of the organisation it was dependent on a key person driving the engagement, but when this person left the process stalled. This was the case with a church-based organisation in the ACT:

we got quite close to being able to deliver something within but with the change of staff and the new person there, just put up some barriers. (HCC 3)

Key people were important to drive the program and build enthusiasm and bring others with them:

Once the facilitator had been trained she was an incredible advocate for the program and actually pushed to have more staff trained and to expand the HEAL™ program not only delimiting to [the town] but also some of the outreach communities that they serviced. (HCC 4)

This also extended to participants in the program. In NSW one of the participants had a lot of influence in the community and brought more people to the program:

she has a following like there’s no tomorrow. So she just brings this massive number of people with her like the Pied Piper. (HCC 2)

Partnerships with key local stakeholders were vital to the success of the local implementation of the HEAL™ program in the LGAs studied. Successful partnerships occurred where the program met the needs of the partnering organisation, or could be adapted to do so, and where there were local champions. Partnerships take time to develop and in more remote areas it can take even longer; the distances involved and workforce shortage mean that there are fewer opportunities to meet stakeholders face to face. They are also dependent on leadership within a partnering organisation and if this leadership changes the partnership does not always continue or needs to be re-established. This has implications for program funding in these areas because by the time the partnerships were functioning effectively the funding for the local implementation had run out. Building on existing partnerships and relationships is a useful strategy for successful implementation.⁶

In the health promotion literature, the importance of partnerships to the success of the implementation of community health promotion is emphasised. The Victorian Healthcare Association have produced guidelines for the development of partnerships in their toolkit.⁵ This includes finding partners with shared goals and the importance of developing trust over time. Several groups have explored partnerships in health promotion in terms of checklists¹³ or frameworks⁹ that can be used as tools to guide the development and evaluation of partnership approaches.⁵ Wagemakers et al.¹³ have developed a checklist to enable those involved to monitor their strengths and weaknesses in the partnership and suggest the use of interviews to further understanding if a partnership is not working. We chose to use the partnership synergy framework to organise the themes emerging from the analysis.⁹

So how do the results from this study relate to the partnership synergy framework proposed by Lasker et al.⁹? In terms of resources, the HEAL™ program had funding through the Healthy Communities Initiative, had a program that provided training for local health professionals, and was seen in some areas as an opportunity for capacity building. In those partnerships that worked well the partners also contributed resources to the collaboration in the form of skills and expertise with population groups and connections or even a physical space in which to run the program. The partners in the local implementation of the HEAL™ program were heterogeneous, bringing together a combination of stakeholders with expertise working with particular population groups such as people with mental health problems, Aboriginal men or people with a disability. The HCC and HEAL™ facilitators brought expertise in lifestyle risk factor modification. The partners saw the opportunities the program had to offer as beneficial to both organisations. Leadership was important, and if a person who was championing the program in an organisation moved on then it took some time to re-establish new partnerships.

One of the key issues raised by all those interviewed were the challenges to the development of the partnership, in particular the length of time required, and interestingly this is not mentioned in the Lasker et al. framework⁹ but is mentioned but not quantified in the guidelines from the Victorian Healthcare Association.⁵ Relationships are developed over time and there are specific challenges for those trying to build relationships in rural and remote areas, complicated by workforce shortages and distance and the need to gain the support of GPs. Both Haby et al.⁷ and Minkler et al.⁸ in their reviews of the implementation of community-based programs in Australia and the USA, noted the importance of time in developing the partnerships and its impact on program implementation within available funding cycles. They both suggest flexibility in funding arrangements to allow time for these partnerships to be established and linking funding to meeting project milestones rather than implementation in fixed periods of time.

None of the LGAs reported engagement with local GPs. In rural NSW there was even a GP on the local implementation advisory group but still this did not lead to GP engagement with the program. Referrals from the local GPs were low until the program had been running for some time and in some areas they only started to refer as the funding was coming to an end. This reflects findings from an evaluation of implementation of a chronic disease self-management program in South Western Sydney. In that study GPs were reluctant to refer patients to programs that may not be sustainable in the long-term.¹⁴

There are limitations to this study: only four LGAs were included and were chosen because they had innovative local solutions to the implementation of the HEAL™ program and may not be representative of all the LGAs that took part in the national roll out. In spite of the small number of LGAs and the small number of stakeholders interviewed in each LGA there were very similar themes emerging from the analysis. Future research could investigate the impact of flexible funding approaches on the implementation of community health promotion interventions, to evaluate whether allowing time for the partnerships to develop results in improved engagement with, and uptake of, the programs with potential for sustainability beyond the program funding.

Partnerships are important for the successful implementation of community preventive health programs. They take time to develop, are dependent on the needs of the stakeholders and are facilitated by consistent leadership and a stable workforce. Building on existing successful partnerships is a useful strategy to start the implementation of a community health promotion intervention. Funding for community health programs should be sufficiently flexible with an initial phase dedicated to the establishment of successful partnerships and then a release of further funding for the program once the partnerships are in place.