Designing an integrated, nurse-driven and home-based digital intervention to improve insulin management in under-resourced settings

Nurse and midwife involvement in task-sharing and telehealth service delivery models in primary care: A scoping review

AIM

To synthesise and map current evidence on nurse and midwife involvement in task-sharing service delivery, including both face-to-face and telehealth models, in primary care.

DESIGN

This scoping review was informed by the Joanna Briggs Institute (JBI) Methodology for Scoping Reviews.

DATA SOURCE/REVIEW METHODS

Five databases (Ovid MEDLINE, Embase, PubMed, CINAHL and Cochrane Library) were searched from inception to 16 January 2024, and articles were screened for inclusion in Covidence by three authors. Findings were mapped according to the research questions and review outcomes such as characteristics of models, health and economic outcomes, and the feasibility and acceptability of nurse-led models.

RESULTS

One hundred peer-reviewed articles (as 99 studies) were deemed eligible for inclusion. Task-sharing models existed for a range of conditions, particularly diabetes and hypertension. Nurse-led models allowed nurses to work to the extent of their practice scope, were acceptable to patients and providers, and improved health outcomes. Models can be cost-effective, and increase system efficiencies with supportive training, clinical set-up and regulatory systems. Some limitations to telehealth models are described, including technological issues, time burden and concerns around accessibility for patients with lower technological literacy.

CONCLUSION

Nurse-led models can improve health, economic and service delivery outcomes in primary care and are acceptable to patients and providers. Appropriate training, funding and regulatory systems are essential for task-sharing models with nurses to be feasible and effective.

IMPACT

Nurse-led models are one strategy to improve health equity and access; however, there is a scarcity of literature on what these models look like and how they work in the primary care setting. Evidence suggests these models can also improve health outcomes, are perceived to be feasible and acceptable, and can be cost-effective. Increased utilisation of nurse-led models should be considered to address health system challenges and improve access to essential primary healthcare services globally.

REPORTING METHOD

This review is reported against the PRISMA-ScR criteria.

PATIENT OR PUBLIC CONTRIBUTION

No patient or public contribution.

PROTOCOL REGISTRATION

The study protocol is published in BJGP Open (Moulton et al., 2022).

Optimized glycemic control of type 2 diabetes with reinforcement learning: a proof-of-concept trial

The personalized titration and optimization of insulin regimens for treatment of type 2 diabetes (T2D) are resource-demanding healthcare tasks. Here we propose a model-based reinforcement learning (RL) framework (called RL-DITR), which learns the optimal insulin regimen by analyzing glycemic state rewards through patient model interactions. When evaluated during the development phase for managing hospitalized patients with T2D, RL-DITR achieved superior insulin titration optimization (mean absolute error (MAE) of 1.10 ± 0.03 U) compared to other deep learning models and standard clinical methods. We performed a stepwise clinical validation of the artificial intelligence system from simulation to deployment, demonstrating better performance in glycemic control in inpatients compared to junior and intermediate-level physicians through quantitative (MAE of 1.18 ± 0.09 U) and qualitative metrics from a blinded review. Additionally, we conducted a single-arm, patient-blinded, proof-of-concept feasibility trial in 16 patients with T2D. The primary outcome was difference in mean daily capillary blood glucose during the trial, which decreased from 11.1 (±3.6) to 8.6 (±2.4) mmol L-1 (P < 0.01), meeting the pre-specified endpoint. No episodes of severe hypoglycemia or hyperglycemia with ketosis occurred. These preliminary results warrant further investigation in larger, more diverse clinical studies. ClinicalTrials.gov registration: NCT05409391 .

Assessing barriers to insulin therapy among people with type 2 diabetes in South Africa using the Insulin Treatment Appraisal Scale: A cross-sectional survey

AIMS

To assess barriers to insulin therapy among people with type 2 diabetes after adapting the Insulin Treatment Appraisal Scale (ITAS) to the South African context.

METHODS

A panel of experts reviewed the original ITAS for clarity and relevance to the South African context. The ITAS was administered to 253 adults with type 2 diabetes attending diabetes outpatient clinics in the Tshwane Metropolitan Municipality. Internal consistency (Cronbach's alpha) was tested and construct validity was examined using exploratory factor analysis (EFA). PIR was appraised in insulin users and non-users.

RESULTS

The EFA revealed that the adapted ITAS had a two-factor structure, similar to the original scale, with acceptable internal consistency (α = 0.85). Insulin-using participants had significantly less negative attitudes to insulin therapy than non-users (40.7 ± 7.1 vs. 51.5 ± 11.2, p < 0.001). Compared to participants who used insulin, participants who did not use insulin were afraid of injecting themselves with a needle (71% vs. 11%, p < 0.001) and saw insulin treatment as a sign of worsening diabetes (63% vs. 29%, p < 0.001).

CONCLUSIONS

Consistent with previous studies, participants who were not using insulin had more negative beliefs and attitudes towards insulin treatment than those who were already using insulin. South African clinicians should use the ITAS to assess positive and negative perceptions regarding insulin therapy in both insulin-naïve and insulin-treated people, to evaluate interventions to reduce PIR and improve treatment outcomes.