Abstract:Background: In South Africa, initiating insulin for people with type 2 diabetes and subsequent titration is a major challenge for the resource-constrained healthcare system. Inadequate support systems in primary care, including not being able to access blood glucose monitors and test strips for self-monitoring of blood glucose, results in patients with type 2 diabetes being referred to higher levels of care. In primary care, initiation of insulin may be delayed due to a shortage of healthcare workers. The dela… Show more
“…Although some algorithms have been developed to assist physicians in insulin titration, only a few have been validated in clinical trials 31 , 32 . We conducted a proof-of-concept feasibility trial demonstrating the viability of the RL-DITR system in inpatients with T2D.…”
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.
“…Although some algorithms have been developed to assist physicians in insulin titration, only a few have been validated in clinical trials 31 , 32 . We conducted a proof-of-concept feasibility trial demonstrating the viability of the RL-DITR system in inpatients with T2D.…”
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.
“…29 In Tshwane, a nurse-driven application has supported people as they start insulin with some success. 30 Guidelines should specify the timing and broad content of counselling for patients regarding the use of insulin. The likelihood of requiring insulin at a later stage should be discussed soon after diagnosis.…”
Background: Type 2 diabetes (T2DM) is a leading cause of mortality in South Africa and resistance to the use of insulin is common. This study aimed to explore factors that influence the initiation of insulin in patients with T2DM in primary care facilities in Cape Town, South Africa.Methods: An exploratory descriptive qualitative study was conducted. Seventeen semi-structured interviews were held with patients eligible for insulin, on insulin and primary care providers. Participants were selected by maximum variation purposive sampling. Data were analysed using the framework method in Atlas-ti.Results: Factors related to the health system, service delivery, clinical care and patients. Systemic issues related to the required inputs of workforce, educational materials, and supplies. Service delivery issues related to workload, poor continuity and parallel coordination of care. Clinical issues related to adequate counselling. Patient factors included a lack of trust, concerns about injections, impact on lifestyle and disposal of needles.Conclusion: Although resource constraints are likely to remain, district and facility managers can improve supplies, educational materials, continuity and coordination. Counselling must be improved and may require innovative alternative approaches to support clinicians who face high number of patients. Alternative approaches using group education, telehealth and digital solutions should be considered.Contribution: This study identified key factors influencing insulin initiation in patients with T2DM in primary care. These can be addressed by those responsible for clinical governance, service delivery and in further research.
“…11 To address these challenges related to insulin management in primary care, a multi-disciplinary team from the University of Pretoria in South Africa developed a nursedriven and home-based telehealth intervention named the Tshwane Insulin project (TIP). 12 The TIP intervention was developed using the Integrated Chronic Disease Management (ICDM) framework 13 and comprises a facilitylevel intervention, where professional nurses evaluate PLWD and initiate insulin, an individual-level intervention where community healthcare workers (CHWs) monitor patients at their homes and provide educational information, whilst using telehealth to enable physician-directed insulin titration if needed, and a community-level intervention aimed at empowering CHWs to support PLWD and raise awareness of diabetes. 12 The TIP intervention was developed in four sequential phases following the guidance of the United Kingdom (UK) Medical Research Council (MRC) for designing and evaluating complex interventions.…”
Section: Introductionmentioning
confidence: 99%
“… 12 The TIP intervention was developed using the Integrated Chronic Disease Management (ICDM) framework 13 and comprises a facility-level intervention, where professional nurses evaluate PLWD and initiate insulin, an individual-level intervention where community healthcare workers (CHWs) monitor patients at their homes and provide educational information, whilst using telehealth to enable physician-directed insulin titration if needed, and a community-level intervention aimed at empowering CHWs to support PLWD and raise awareness of diabetes. 12 …”
Section: Introductionmentioning
confidence: 99%
“… 14 The four phases were (I) planning, (II) design, (III) implementation and (IV) evaluation. 12 During phase IV, the TIP intervention was piloted at 10 primary care facilities in the Tshwane District. Patients with T2DM and primary health care providers including doctors, nurses and CHWs were involved in the trial.…”
Background: In South Africa, initiating and managing insulin in primary care for people living with type 2 diabetes (PLWD) is a major challenge. To address these challenges, a multidisciplinary team from the University of Pretoria (South Africa) developed the Tshwane Insulin project (TIP) intervention.Aim: To determine internal and external factors, either facilitators or barriers, that could influence the implementation of the TIP intervention and propose strategies to ensure sustainability.Setting: Tshwane District, Gauteng province, South Africa.Methods: We used the SWOT framework to qualitatively analyse the strengths, weaknesses, opportunities, and threats influencing the implementation of the TIP intervention. Four field researchers and three managers from the TIP team participated in an online group discussion. We also conducted semi-structured interviews with healthcare providers (HCPs) (seven nurses, five doctors) and patients with type 2 diabetes (n = 13).Results: Regardless of the identified weaknesses, the TIP intervention was accepted by PLWD and HCPs. Participants identified strengths including app-enabled insulin initiation and titration, pro-active patient follow-up, patient empowerment and provision of glucose monitoring devices. Participants viewed insulin resistance and the attitudes of HCPs as potential threats. Participants suggested that weaknesses and threats could be mitigated by translating education material into local languages and using the lived experiences of insulin-treated patients to address insulin resistance. The procurement of glucose monitoring devices by national authorities would promote the sustainability of the intervention.Conclusion: Our findings may help decision-makers and health researchers to improve insulin management for PLWD in resource-constrained settings by using telehealth interventions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.