Experiences of overweight and obese patients with diabetes and practice nurses during implementation of a brief weight management intervention in general practice settings serving Culturally and Linguistically Diverse disadvantaged populations

Keywords: community health: nursing, culturally and linguistically diverse, diabetes, disease management, general practice, practise nurses, primary health care, self-regulation, weight management.

Type 2 diabetes mellitus (T2DM) is one of the most common chronic health conditions in Australia (AIHW 2020). There is evidence that disadvantaged groups have significantly higher prevalence of diabetes (Cunningham 2010; Abouzeid et al. 2013) and vulnerability to being overweight and obese (Abouzeid et al. 2013; Menigoz et al. 2016; Dao et al. 2019). Diabetes is one of the most common chronic illnesses managed by Australian GPs (Britt et al. 2016).

Lifestyle change, particularly entailing weight reduction, is a well-evidenced intervention for the management of T2DM (Lean et al. 2018), and GPs have identified a lack of time, reimbursement issues and limited nutrition training in weight management as barriers to engaging patients in lifestyle change (Britt et al. 2016). GPs have advocated for practice nurses (PNs; Hegney et al. 2013; Boyle et al. 2016; Whitehead et al. 2020), who are registered or enrolled nurses who work collaboratively with GPs within a general practice setting (Desborough et al. 2016), to manage implementation of weight management interventions for their patients. There is potential to task shift the initiation, delivery and follow up of lifestyle interventions to PNs to ensure that overweight and obese patients with diabetes can access cost-effective, supportive interventions in general practice settings (Afzali et al. 2014).

Self-care activities undertaken for diabetes self-management include adhering to a healthy diet, exercise, medication use and self-monitoring activities, such as blood glucose testing (Hampson et al. 1990). Although critical to diabetes self-management, dietary self-regulation remains a challenge for disadvantaged groups (Wermeling et al. 2014), who often have low self-efficacy and may be less proactive in making changes in their lifestyles because of competing priorities (Mathur et al. 2005). Self-regulation interventions, including motivational interviewing (Rosenbek Minet et al. 2011) and goal setting (Ammerman et al. 2000), have been shown to increase motivation, confidence and decision-making skills to sustain healthy eating habits among people with diabetes.

Mental contrasting and implementation intentions (MCII) combines two established self-regulation strategies: mental contrasting (contrasting their image of a positive goal and negative reality to identify obstacles to reaching this desired change) and implementation intentions (simple action plans specifying when, where and how a goal should be acted upon). MCII has also been tested as a strategy to promote healthier eating behaviours (Adriaanse et al. 2009, 2011, 2013; Stadler et al. 2010).

The aim of this paper was to test whether it is feasible and acceptable to motivate patients from a disadvantaged background to initiate and maintain dietary regulation, by task shifting to PNs within a general practice setting. We present the perspectives of providers (PNs) and patients with diabetes during the implementation of a brief MCII-based dietary self-regulation approach for weight management in general practice settings serving disadvantaged populations.

The study was conducted in South Western Sydney from August 2015 to November 2016, in a population characterised by a higher proportion of culturally and linguistically diverse (CALD) groups, and a higher rate of socioeconomic disadvantage than the rest of Sydney (Dao et al. 2019). It is estimated that 60% of residents in South Western Sydney are at risk of developing T2DM (SWSPHN 2018).

General practices were approached by the researchers with support from local Primary Healthcare Network liaison officers. General practices and PNs were provided with information about the study, including the research protocol, ethics approval, consent form and the patient information statement. Practices and PNs who agreed to participate constructed lists of potential participants who met the inclusion criteria (see Table 1). Lists were then reviewed by GPs to remove potential participants who met the exclusion criteria outlined in Table 1. GPs sent an invitation letter explaining the study to all patients who satisfied the inclusion and exclusion criteria. Potential participants were encouraged to make an appointment with the PNs, who then obtained consent for patients who agreed to participate.

Table 1.  Inclusion and exclusion criteria.

The intervention was based on a MCII approach to dietary self-regulation (Adriaanse et al. 2009; Hagger and Luszcznska 2014). The intervention was delivered over 24 weeks, and aimed to assist patients in identifying dietary goals and behavioural changes they would undertake to reach these goals. PNs were trained in T2DM-related nutrition and the MCII technique for a total of two 2-h sessions (4 h in total). This training included instruction in: guiding participants to complete a 3-day food diary to aid in the identification of opportunities to improve patients’ diets; understanding the ‘mental contrasting’ steps of imagining the positive outcome of desired change; identifying and reflecting on obstacles to making the change; and developing plans of when and how to overcome identified obstacles. PNs assisted patients to set dietary change goals during the first session, which lasted for approximately 1 h. This step-wise, personalised, patient-driven goal setting and selection of one behaviour to change is key for MCII, and each patient was encouraged to initially focus on a behaviour change that they were most motivated to achieve. To strengthen patients’ motivation for behaviour change, the MCII approach encourages people to imagine the positive outcomes they expect to attain. Subsequently, to strengthen self-efficacy, patients were asked to reflect on obstacles to behaviour change, and to develop plans to overcome these barriers. The PNs directed patients to write down the intended behaviour modifications that would support the change, and these statements were used to elicit behavioural-specific ‘Action Plans’ that specified what was to be done and by when. After the first face-to-face session, the PNs called patients every 2 weeks for 8 weeks to provide encouragement and to reaffirm commitment. The design of the intervention assumed that the intensive and continuous 8 weeks’ support (1-h phone calls each time) from the PNs would enable patients to develop understanding and skills to implement the changes required. At 12 weeks and 24 weeks, the patients made an appointment to meet the PNs at the general practice to record their weight, discuss how they had progressed and if they wanted to alter their dietary change goals.

Semi-structured interviews were conducted with patients and PNs. The research team developed two interview guides (separate tools used for patients (Appendix 1 in Supplementary Material) and PNs (Appendix 2 in Supplementary Material)) to ask patients and PNs about their expectations and experiences of the intervention, and their respective roles. Patients were interviewed at 12 weeks to pilot test the interview guide and to capture their initial experiences with the intervention. Patients were interviewed again at 24 weeks (final interview guide included in Appendix 1 in Supplementary Material) to capture their experiences after completing the intervention. Patients were interviewed either face-to-face or via telephone, depending on their availability. All PNs were interviewed face-to-face. Interviews were conducted by one of four researchers who were allocated to patients and PNs based on availability.

Interviews were audio-taped and transcribed, and analysed thematically (Braun and Clarke 2006). Ezzy (2002) recommends that multiple researchers should be engaged in the process of open coding and refined coding to limit the influence of any one perspective on the identification of key themes (Ezzy 2002). During the open coding process, two researchers (KY and SG) separately coded the same several transcripts, noting recurring ideas. Three researchers (KY, SG and RJ) then compared initial codes to develop a consistent and coherent coding framework. This framework was then used by two researchers (KY and SG) to code the remaining transcripts during the refined coding process. Following this stage of the analysis process, researchers KY, SG and RJ met to deliberate which codes provided the most useful and relevant insights for the research, resulting in the themes reported below. The qualitative data analysis tool, Nvivo (QSR International), was used to assist in the management of data.

Ethics approval for this study was obtained from University of New South Wales (HC14070).

Of the 29 participants who initially enrolled in the study, 22 were followed up at 12 weeks and 12 at 24 weeks. Approximately half (n = 10) of the 22 participants who were followed up at 12 weeks dropped out by the end of the intervention. Therefore, the findings below only reflect the experiences of the 12 patients who completed the intervention at 24 weeks. The main reasons given for dropout were competing priorities and the cessation of on-going phone calls after 8 weeks. Of the 12 patients interviewed via telephone at 24 weeks, seven were women and 10 were under the age of 65 years. All, except one, were born overseas, with 10 from the Middle East. On average, they had lived in Australia for 10 years, and only two patients spoke English at home (Table 2). Four PNs out of the six participating were interviewed. Due to competing schedules, the research team was unable to interview the other two PNs. Two overarching themes were uncovered: (1) challenges related to patient and PNs’ expectations, and (2) challenges of the context.

Table 2.  Sociodemographic characteristics of interviewed patients and practise nurses at 24 weeks.

For many patients, the goal of the intervention was to make dietary changes, including food substitution, reduction or elimination to achieve weight reduction. Many patients, therefore, perceived the intervention as a diet plan to: ‘increase vegetables […] decrease bread’ (Patient 1); no longer drinking ‘a lot of tea with sugar’ (Patient 1); changing the quantity of food eaten (Patient 1); eating less bread, rice and fried food (Patient 4); and eating more brown bread and vegetables (Patient 4).

Importantly, patients indicated that these diet change goals were not self-initiated, but directed by PNs. The following exchange from Patient 17’s interview is typical:

Interviewer: Why did you choose this goal?

Patient 17: Because the nurse said [emphasis added] it will help me control my blood sugar levels better.

The various intentions for change stated by patients were usually framed in terms of what the nurse had advised them to do and not self-directed, as required by the intervention theory. As Patient 10 said, ‘I followed her (the PN’s) diet. Sometimes I ate, sometimes I didn’t eat. I don’t know. I went down around 3 kg’. Similarly, Patient 2 also indicated that the PNs provided her with a diet plan: ‘She (the PN) recommended I eat all diet food. So that is what I have been eating’.

PNs reported that a challenge in implementing a patient-led intervention stemmed from patients’ attitude towards the ‘authority’ of PNs. For example, Nurse 4 noted that ‘they were like […] you know what’s best for me, you know the right thing to do […] just tell me how I can fix my diet or my way of eating’. One nurse claimed that their CALD patients in particular often requested a diet plan ‘rather than doing things for themselves’ [Nurse 2]. Nurse 1 explained that when some patients ‘realised it was all them, they lost interest’.

One PN commented that ‘you literally have to plan it for them’ [Nurse 2], while another explained that it was ‘all on me to do the work […] To put words in their mouth to keep pushing them to think of their goal’. Nurse 3 further elaborated that […] They couldn’t understand [the process]’. Nurse 4, who was from a ‘similar cultural background’, to their patients, explained that in that culture, patients were accustomed to accepting official information unquestioningly:

I think they look up to doctors and nurses […] [who] give them instructions and [patients] just follow it. They lived like that their whole life. There is no discussion, doctors do not discuss with them […] They expect you to give them direction, so that they can just follow it.

Nurse 3 characterised their patients as ‘challenging’, while Nurse 4 described the intervention itself as not appropriate for ‘the kind of patients who attend our practice’. She speculated that the intervention was not suited to migrants from CALD background or of a lower socioeconomic status, who prefer to receive instruction from a trusted authority, such as the PNs. Due to these factors, PNs reported they had to do much of the cognitive work on behalf of patients.

Patients generally described their experiences with PNs in positive terms. One patient explained that the ‘nurse was very, very good […] She encouraged me a lot about many things. She said I had to look after my health […] and […] watch my diet. Her advices were very good. I was very much encouraged by [the] way she talked’ (Patient 4). However, it is not clear how accurately patients self-reported their progress to PNs. Nurse 3 alluded to the possibility of patients falsely attributing change for the sake of this positive relationship with the nurse:

Since [the patient] started, every time when I [called] her […] she would say that she is keeping her goal very well, ‘I have lost weight’ […] But, when she came for her Week 12, she has actually put on weight […] At some point, I was thinking maybe she wanted to make me happy [emphasis added].

Personal and family commitments, and existing complexities of daily life presented challenges that affected patients’ engagement with the intervention. Patient 10 noted how stressed she was due to balancing long working hours with caring for an elderly parent, explaining that she was ‘busy from the morning at 8 am till 10–11 at night […] My mother is sick […] I always check up on her. I don’t have time for myself […]. It must be stress as well […] I am working all day […] I am also sick. I have been through several operations […] high sugar levels and cholesterol […]. It was a very depressing time in my life’.

Family and social events presented similar challenges. Patients reported that they were exposed to an ‘abundance of food’ (Patient 12) during lunches with family and friends, and during religious celebrations.

We had a number of occasions; for example, birthdays. I have a big family. My son has five children, so we would get invited to their birthdays. So during that time my eating would change.

PNs in this study provided a useful source for understanding the GP settings within which the interventions unfolded. As one nurse explained:

[At] the time the study was introduced to us, we were about to start a diabetic clinic […] it got so overwhelming […] the doctors [dropped] from that study […] when this study was introduced, the practice was [reluctant], but we [nurses] wanted to, as we […] thought it will be beneficial […] I explained it to the practice manager and they said ok. But it was overwhelming at times because of the paperwork [Nurse 3].

It was reported that the way the practice was operated complicated the delivery of this type of intervention. PNs generally explained that implementing the intervention as part of their regular work practices was challenging due to interruptions and concurrent commitments. As Nurse 2 explained:

That was the only bad thing, I didn’t have one set room […] when I am asking questions and I get called over to assist in something else. That was the hardest thing. […] I was expected to do my regular work around it [emphasis added]. […] you have doctors coming in and saying that [they need help] and […] you say excuse me [to the patient]. So […] this […] took longer [than patients expected], as I had to run off and do things.

Overall, the findings suggest a story of nurse–patient relationships mediating differing expectations of the MCII intervention and challenges faced by PNs to accommodate change in practice. Many of the challenges appear to be shaped by the cultural backgrounds and expectations of patients on the one hand, and intervention requirements on the other hand. The intervention was designed to be patient-driven, and required individuals to identify weight-loss goals and areas in which they saw fit to make changes in dietary behaviour. This was inherently challenging, as it was at odds with patients’ expectations that they would be given clear direction on what they needed to do by PNs.

The everyday complexities of lives and social commitments of disadvantaged patients emerged as disruptors to action plans for weight reduction. These issues have been reported among CALD migrants in Australia engaged in obesity prevention initiatives (Cyril et al. 2017). In the current study context, the intervention seemed to add another layer of complexity to patients’ lives, which may have been reflected back onto nurses by way of expectations of direct instruction and planning. This points to the need to set the appropriate expectation around the role of patient, provider and expected outcome throughout the intervention.

PNs in the current study, particularly those who themselves were migrants, reported their perception that self-regulatory interventions were inappropriate for lower socioeconomic status patients from their own cultural backgrounds. PNs interviewed by Kelly (2016) described their beliefs that migrants from their community who remain disadvantaged must lack motivation and ability to take charge of their lives. It is also possible to infer that PNs decided what works and is appropriate for their people (patients), informed by their shared cultural background. A study that aimed to adapt chronic disease self-management programmes to CALD groups found that there was shared understanding of concepts by different groups of participants, but a clear need to address issues of low literacy (Walker et al. 2005). However, more specific cognitive strategies, similar to MCII have not been tested with CALD groups. Overall, this reflects challenges in directly transferring self-regulatory interventions, such as MCII, that have been tested in high literacy and non-CALD groups (Adriaanse et al. 2009, 2011) without adequate adjustment for context.

An interesting finding was that many of the participants expressed a strong sense of satisfaction with the PNs and their competencies. However, a few PNs questioned the objective reality of patient-reported outcomes, of weight loss conveyed during monitoring phone calls. It is possible that this was influenced by the cultural need to demonstrate appreciation of the PN by keeping the provider happy, or a perception that they needed to maintain a positive relationship to ensure continued access to health services (Williams et al. 2015).

It was unclear whether the main challenge faced by PNs was to provide tailored advice around diet that appropriately addresses cultural and intervention expectations in a brief interaction, or because of the time and resource constraints that PNs experienced while juggling a busy clinic schedule, or a combination of both factors (Parker and Fuller 2016; Lenzen et al. 2017). Although, many mentioned competing priorities, PNs in this study were able to engage with the intervention and their patients, providing continuity of care through a service that was valued. These are positive findings on the use of PNs (Afzali et al. 2014; Parker and Fuller 2016) for this and associated tasks.

There were several limitations of this study. One of the general practices had a high number of CALD migrants and some of the findings may reflect issues of that sub-population that limits transferability of these findings to wider groups. The study sample was derived from a low number of general practices and did not include all CALD groups that exist in South Western Sydney, further limiting the transferability of the results. The study had a high dropout rate of patients, and those who remained may have been a select population. As we did not measure the baseline level of knowledge in nutrition and health literacy of our patients, it is unclear how these factors could have contributed to the challenges they faced in implementing self-regulation strategies.

The experiences of patients and providers (PNs) in a brief PN-led MCII-based self-regulation intervention to promote healthy eating highlight challenges for patients in setting and meeting self-directed dietary change goals due to their expectations that health professionals set goals and behaviours for them. Implementation of the intervention faced other obstacles, such as competing demands on the time and resources of PNs in a general practice setting and nurses’ own perceptions of the suitability of a patient-driven intervention for overweight and obese patients with T2DM from disadvantaged and CALD backgrounds.

Most of the participants expressed high levels of satisfaction and trust with PNs. It is evident that trust, rapport and support are key to ensuring continuity of engagement. This study provides some evidence to support a brief MCII-based dietary intervention, delivered by PNs in the general practice setting, that aims to reduce weight in disadvantaged CALD populations with T2DM. Future studies should address PN training and availability for intervention delivery. A prospective study is required to assess the effectiveness of tailoring this intervention more purposefully for disadvantaged CALD populations, and whether it may be sustained in general practice settings.

Supplementary material is available online.

Anonymised data that support this study can be made available on legitimate request to the corresponding author.

The authors have no conflicts of interest.

This study was funded by Diabetes Australia Research Program Grant number Y14G-ROSV.