Evaluating the impact of the Piki te Ora extended primary care team on clinical outcomes in type 2 diabetes patients: a retrospective study

Introduction

Type 2 diabetes (T2D) represents a significant health challenge globally,¹ with New Zealand (NZ) experiencing a high prevalence of this chronic condition, in particular a disproportionate prevalence for Māori.² In response to the need to better assist patients living with T2D and their whānau,³ Te Korowai Hauora o Hauraki, an Iwi Māori health provider servicing the tribal boundaries of Pare Hauraki, NZ, established the Piki te Ora service to support its five clinics. The Piki te Ora service adopts a multi-disciplinary approach, with cultural safety as a key element, aiming to provide comprehensive support and care to whānau (enrolled patients) living with long-term conditions, including T2D.⁴

Central to the Piki te Ora service is the provision of wrap-around care characterised by frequent patient follow-ups and personalised interventions delivered by a team comprising nurses, dietitians, occupational therapists, kaiāwhina (support workers) and social workers. All members of the team work part-time, allowing for a flexible, coordinated approach to patient care. Furthermore, referrals to the service can be made by professionals within Te Korowai Hauora o Hauraki, external health care providers, or through patient self-referral. A key aspect of the Piki te Ora service is its accessibility and flexibility as consultations are provided both face-to-face and over the phone, enhancing accessibility for individuals residing across Pare Hauraki. Importantly, the service incorporates frequent patient follow-up, often conducted by multiple team members as necessary, with no time constraints or associated costs to the patients. This approach enables ongoing monitoring and adjustment of care plans based on individual patient needs and progress.⁵

The overarching goal of the Piki te Ora service is to achieve measurable improvements in clinical outcomes for patients, including those with T2D. As patients demonstrate progress and achieve clinical targets, they may transition out of the service, reflecting the service’s emphasis on empowering patients to self-manage T2D effectively. However, although the programme has been running for several years, it has not been evaluated with regard to measurable improvements in clinical outcomes. Thus, this review aimed to assess the efficacy of the Piki te Ora service for supporting T2D management in primary care.

Methods

All patients who were enrolled in the Piki te Ora Service (at March 2024) and had been included in the service since January 2021 (combined n = 112) were invited to opt in to this study (by text initially, followed by phone call and/or face-to-face if needed) and to have their data included in the data review/analysis. Patients could be enrolled at any of the five clinics, including Whitianga, Paeroa, Thames, Coromandel and Te Aroha.

Patient sociodemographic and clinical data were extracted from the Te Korowai Hauora o Hauraki database from their time of enrolment in the Piki te Ora service to 6 months post enrolment. Sociodemographic data included age, gender, ethnicity, socioeconomic deprivation and community service card use. Clinical data included glycated haemoglobin (Hba1c), total cholesterol, high-density lipoprotein cholesterol (HDLc), low-density lipoprotein cholesterol (LDLc) and blood pressure measurements. All measures were collected between January 2021 and March 2024, with measures at 3 and 6 months then compared to baseline for each participant.

Clinical notes for each consented patient were then manually searched to extract information regarding the frequency of contact made by staff, and the type of contact (phone call, text message, in clinic, or outside clinic [at home, work, etc]). The frequency of contact refers to how often the staff member reached out to the patient, irrespective of patient response, thereby focusing on system-related factors rather than patient-related factors.

Descriptive statistics, including means, standard deviations, frequencies and percentages were conducted to summarise the sociodemographic characteristics of the participants, clinical outcomes and clinic–patient contact. Wilcoxon signed-rank tests were conducted to explore associations between changes in clinical outcomes at different time periods. Missing data were excluded to ensure valid pairs for the Wilcoxon signed-rank test. All analyses were conducted using SPSS (v29.0). Ethics approval for this study was provided by the Health and Disability Ethics Committee (HDEC 2022 AM 8309).

Results

Overall, 64 of the 112 patients enrolled/previously enrolled in the Piki te Ora service provided consent to have their data included for review. The mean age of included (consented) participants was 58 ± 13 years with the youngest 34 years and the oldest 83 years. Over half of participants were Māori (62.5% vs European 37.5%) and were male (59.4%), which was representative of the full cohort (n = 112, including those who did not consent whose data are therefore not included below). Most consented participants were living in the most deprived areas of Waikato (76.5%) and half were community service card users (Table 1).

Overall, median HbA1c levels decreased significantly by 22.0 mmol/mol at 3 months (66.5 (50.5, 90.5) mmol/mol) compared to baseline (88.5 (72.0, 104.8) mmol/mol; P < 0.001) and this reduction was maintained at 6 months (68.0 (54.5, 83.0) mmol/mol; Table 2). In contrast, no significant reductions were seen in mean total cholesterol, HDLc or LDLc, and systolic/diastolic blood pressure (BP) remained relatively constant over the study period. Further sub-analysis showed no change in any of the clinical measures at 6 months by age, sex, ethnicity, New Zealand Deprivation 2018 (NZDep18) or community service card use (P > 0.05). However, follow-up 3- and 6-month data were lacking across this cohort, with only 10.9–40.6% having lipids data recorded at 3 and/or 6 months.

Over 6 months, patients in the Piki te Ora service were contacted an average of 9.5 ± 9.0 times. Phone calls were the most common mode of contact throughout the study period (mean 5.0 ± 5.7 over 6 months). Face-to-face interactions outside the clinic (mean 1.8 ± 3.6 over 6 months), in-clinic consultations (mean 1.1 ± 1.9) and text messages (mean 0.5 ± 1.1) were also used modes of contact with patients.

Discussion

Multidisciplinary, extended primary care teams offer great potential for supporting patients with T2D in NZ which agrees with international literature.⁶ The Piki te Ora service significantly improved glycaemic control in only 3 months, with this being likely due to higher engagement with the health care system and patient-specific education,⁷ and importantly these improvements were maintained in the medium term unlike previous studies.⁸ While the observed drop in HbA1c is promising and could be credible, we must consider potential limitations such as sample size and recruitment biases, which may affect generalisability and accuracy of the results. However, the emphasis on whakawhanaungatanga may have created an environment built on trust and meaningful connections and relationships between patients, their whānau and health care providers.⁹ This sense of support for patients where they feel valued and understood may lead to better health outcomes and sustained engagement.¹⁰ The increased frequency of free health care/patient interactions is likely also a strong contributor to the success of this programme compared to standard primary care where interactions are fee-for-service periodic check-ups, visits and prescription refills.

Inadequate outreach from clinicians to patients coupled with the limited access patients have to health care, particularly in rural areas, is a well-documented barrier to T2D management.^3,11 The Piki te Ora team has addressed this issue by engaging with patients through various outreach methods including visits to patients’ homes and workplaces. This approach has resulted in higher engagement in our study compared to the traditional in-clinic consultations, highlighting the need for more flexible and accessible health care delivery models. Health care professionals often face the challenge of making multiple attempts to contact patients due to various factors such as busy schedules, changes in contact information or unforeseen circumstances. This challenge can increase operational costs and strain resources, ultimately affecting the sustainability of health care services. However, addressing this issue effectively can enhance patient engagement and improve overall care outcomes. Furthermore, by eliminating financial barriers, the service ensures that all patients, regardless of their socioeconomic status, have equitable access to comprehensive care. This inclusivity is essential for addressing health disparities and improving overall community health outcomes.

Conclusion

In conclusion, the multidisciplinary Piki te Ora service may help patients with high-risk T2D manage their glycaemic control, although we currently lack the statistical power to confirm its effectiveness. Additionally, while measurable data on lipids and other clinical markers are insufficient, with further research and funding, these data gaps can be addressed, along with an evaluation of long-term outcomes. We suggest that there is a need to more fully explore the role that this and other extended primary care teams may have for supporting T2D management – particularly the engagement of the patient and their whānau into the clinic and the health system as a factor in the seen improvement in HbA1c. A wider review could also include patient experience and how the different members of these extended teams each contribute to patient success. The findings can inform health care policy and practice, with implications for improving diabetes care and advancing health equity.