Search

The discourse amongst diabetes specialists and academics regarding technology and artificial intelligence (AI) typically centres around the 10% of people with diabetes who have type 1 diabetes, focusing on glucose sensors, insulin pumps and, increasingly, closed-loop systems. This focus is reflected in conference topics, strategy documents, technology appraisals and funding streams. What is often overlooked is the wider application of data and AI, as demonstrated through published literature and emerging marketplace products, that offers promising avenues for enhanced clinical care, health-service efficiency and cost-effectiveness. This review provides an overview of AI techniques and explores the use and potential of AI and data-driven systems in a broad context, covering all diabetes types, encompassing: (1) patient education and self-management; (2) clinical decision support systems and predictive analytics, including diagnostic support, treatment and screening advice, complications prediction; and (3) the use of multimodal data, such as imaging or genetic data. The review provides a perspective on how data- and AI-driven systems could transform diabetes care in the coming years and how they could be integrated into daily clinical practice. We discuss evidence for benefits and potential harms, and consider existing barriers to scalable adoption, including challenges related to data availability and exchange, health inequality, clinician hesitancy and regulation. Stakeholders, including clinicians, academics, commissioners, policymakers and those with lived experience, must proactively collaborate to realise the potential benefits that AI-supported diabetes care could bring, whilst mitigating risk and navigating the challenges along the way., (© 2023. The Author(s).)

Glucocorticoids, also known as steroids, are a class of anti-inflammatory drugs utilised widely in clinical practice for a variety of conditions. They are associated with a range of side effects including abnormalities of glucose metabolism. Multiple guidelines have been published to illustrate best management of glucocorticoid-induced hyperglycaemia and diabetes in a variety of settings. This article discusses current best clinical practice including diagnosis, investigations and ongoing management of glucocorticoid-induced dysglycaemia in both in- and outpatient settings., (© 2023. The Author(s).)

Aims: We investigated a point of admission metric of glycemia, the Admission Glucose Number (AGN), and its relationship with both high risk inpatient glucose patterns and mortality in hospital inpatients with type 2 diabetes (T2DM)., Methods: Inpatient capillary blood glucose (CBG) data for patients with T2DM in our health board were identified for a 5-year period and associated with most recent preadmission HbA1c. AGN was calculated as first CBG measured during admission (mmol/L), subtracted from most recent preadmission HbA1c (converted to estimated median glucose mmol/l) within 15 months preadmission. The association between AGN and CBG variability (interquartile range), hypoglycemia free survival (HR) and both inpatient and 100-day mortality (HR) were investigated., Results: A total of 21 045 first admissions with available HbA1c data were identified. A positive correlation between AGN and glycemic variability was described (partial correlation coefficient 0.25, P < .001), which was stronger than the correlation of either of AGNs' individual components: adjusted CBG1 = 0.07 ( P < .001), eAG = 0.08 ( P < .001). The hazard ratio for time to first recorded CBG < 3 mmol/L for high AGN versus low AGN was 1.74 (95% CI 1.55-1.96), P < .001. A high AGN was associated with increased 100-day mortality (HR 1.26, P = .005), however not with in-hospital mortality (HR = 1.31, P = .08)., Conclusion: AGN is a simple metric that combines 2 readily available measures associated with adverse outcome in T2DM. AGN may be a useful tool to stratify patients for risk of hypoglycemia and postdischarge death.

Aims: Previous evidence has demonstrated that participation in the Dose Adjustment for Normal Eating (DAFNE) education programme can reduce HbA 1c and severe hypoglycaemia in people with Type 1 diabetes. In a number of studies, increased HbA 1c variability has been associated with higher diabetic morbidity and mortality. No studies have examined the impact of structured education on HbA 1c variability in Type 1 diabetes., Methods: People with Type 1 diabetes who had attended DAFNE were identified for inclusion from the Scottish Care Information-Diabetes dataset. HbA 1c median and variability, expressed as coefficient of variation (CV) before and after DAFNE was calculated., Results: Some 1061 individuals participated in DAFNE education and 687 met the inclusion criteria. A significant median reduction in HbA 1c [-3.5 mmol/mol (-0.3%)] was seen at 12 months with a significant reduction [-1.5 mmol/mol (-0.1%)] still seen at 60 months of follow-up. HbA 1c variability as measured by CV was significantly lower during the post-DAFNE period: 0.08 (IQR 0.05-0.12) reduced to 0.07 (IQR 0.05-0.10); P = 0.002., Conclusion: The data confirm that DAFNE participation improves glycaemic control in Type 1 diabetes with benefits being sustained for 5 years. This study is the first to demonstrate reduced HbA 1c variability after completion of structured education. This is new evidence of the beneficial impact of DAFNE on glycaemic profile., (© 2018 Diabetes UK.)

Aim: To investigate the relationship between variability in both visit-to-visit HbA1c and SBP and mortality in individuals with type 1 diabetes., Methods: The Scottish Care Information (SCI) Diabetes dataset was used to identify 5952 individuals with type 1 diabetes for inclusion in this observational study. The SCI-Diabetes dataset allowed access to blood pressure values, HbA1c readings, demographic information and mortality rates for all study participants. Participants were dichotomized to above and below median values for both HbA1c coefficient of variation (CV) and SBP CV, thus dividing participants into 4 cohorts for survival analysis. Survival analysis was carried out over 1430 days. A Cox proportional hazard model was used to allow comparison of mortality between the 4 cohorts., Results: Of the 5952 patients, death occurred in 416. CV for both HbA1c and SBP were significantly associated with mortality. The median values for HbA1c CV and SBP CV were 8.0 and 8.1, respectively. The hazard ratio for high HbA1c CV only (P = .0015) was 1.78 ± 0.36. The hazard ratio for high SBP CV only (P = .0018) was 1.69 ± 0.33. The hazard ratio for both high HbA1c CV and high SBP CV (P < .00001) was 2.37 ± 0.32., Conclusions: Our findings demonstrate that variability of both HbA1c and SBP is significantly and additively associated with mortality in individuals with type 1 diabetes. The variability of these parameters might be useful for risk stratification and is a potential target for future interventional studies., (© 2017 John Wiley & Sons Ltd.)

Aim: Despite evidence of morbidity, no evidence exists on the relationship between HbA 1c variability and mortality in Type 1 diabetes. We performed an observational study to investigate whether the association between HbA 1c variability and mortality exists in a population of people with Type 1 diabetes. As a secondary outcome, we compared onset of first hospital admission between groups., Methods: People with Type 1 diabetes were identified for inclusion from the Scottish Care Information - Diabetes data set. This database includes data of all people known to have diabetes who live within Scotland. A survival analysis was carried out over a 47-month period comparing two groups; group 1 with a HbA 1c coefficient of variation (CV) above the median CV value, and group 2 with a CV below the median value. Time to death or first admission was also analysed. A Cox proportional hazard model was used to compare time to death, adjusting for appropriate covariables., Results: Some 6048 individuals with Type 1 diabetes were included in the analysis. Median HbA 1c CV was 7.9. The hazard ratio (HR) for mortality for those with an HbA 1c CV above the median value is 1.5 over 47 months of follow-up (P < 0.001). HR for survival to either the first admission to hospital or death for those with an HbA 1c CV above the median value was 1.35 (95% confidence interval 1.25-1.45) over 730 days of follow-up (P < 0.001)., Conclusion: Our results show that people with greater HbA 1c variability have a higher rate of mortality and earlier hospital admission in Type 1 diabetes., (© 2017 Diabetes UK.)

Background: Hypoglycemia is associated with increased length of stay in hospital patients, but previous studies have not considered the confounding effect of increased hypoglycemia detection associated with increased capillary blood glucose (CBG) measurement in prolonged admissions. We aimed to determine the effect of recorded hypoglycemia on length of stay of hospital inpatients (LOS) when this mathematical association is subtracted., Methods: CBG data were analyzed for inpatients within our health board area (01/2009-01/2015). A simulated CBG data set was generated for each patient with an identical sampling frequency to the measured CBG data set. The mathematical component of increased LOS was determined using the simulated data set. Subtraction of this confounding mathematical association was used to provide measurement of the true clinical association between recorded hypoglycemia (CBG < 4 mmol [< 72mg/dl]) and LOS., Results: A total of 196 962 admissions of 52 475 individuals with known diabetes were analyzed. 68 809 admissions of 29 551 individuals had >4 CBG measurements made and were included in analysis. After subtraction of the mathematical association of increased sample number, the clinical effect of recorded hypoglycemia is reduced-but persists-compared to previous studies. 1-2 days with recorded hypoglycemia has a relatively minor effect on LOS. LOS increases rapidly if there are ≥3 days with recorded hypoglycemia, with an increase of 0.75 days LOS per additional day with hypoglycemia., Conclusions: This technique increases accuracy of economic modeling of the impact of hypoglycemia on health care systems. This could assist study of the impact of hypoglycemia on other outcomes by factoring for bias of increased sample numbers.

Aims: Inpatient hypoglycaemia is common and associated with adverse outcomes. There is often increased vigilance of hypoglycaemia in inpatients with type 1 diabetes (T1DM) compared to type 2 diabetes (T2DM). We aimed to investigate this apparent discrepancy, utilising the time to repeat (TTR) capillary blood glucose (CBG) measurement as a surrogate for engagement with guidelines stating that CBG should be rechecked following intervention within 15 min of an initial CBG of <4 mmol/L., Methods: This is an observational study of inpatient CBG data from 8 hospitals over a 7-year period. A national diabetes registry allowed identification of individual's diagnosis and diabetes therapy. For each initial (index) CBG, the TTR for individuals with T2DM-on insulin or sulphonylurea-was compared with the TTR for individuals with T1DM, using a t test for significance performed on log(TTR). The median TTR was plotted for each group per index CBG., Results: In total, 1480,335 CBG measurements were obtained. A total of 26,664 were <4 mmol/L. The TTR in T2DM individuals on sulphonylurea was significantly greater than in T1DM individuals where index CBG was ≥2.3 mmol/L (except index CBG 2.6 mmol/L). For T2DM patients receiving insulin significance exists for index CBGs of ≥3.2 mmol/L., Conclusions: This analysis suggests that quality of care of hypoglycaemia varies according to diagnosis and medication. The group with the highest TTR (T2DM sulphonylurea treated) are possibly the clinical group in whom hypoglycaemia is most concerning. These data therefore suggest a need for education and raising awareness within the inpatient nursing staff.