Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Observational Study
. 2019 Dec;21(12):2704-2711.
doi: 10.1111/dom.13860. Epub 2019 Sep 30.

Predicting short- and long-term glycated haemoglobin response after insulin initiation in patients with type 2 diabetes mellitus using machine-learning algorithms

Sunil B Nagaraj  1 , Grigory Sidorenkov  1   2 , Job F M van Boven  1 , Petra Denig  1
Affiliations
Observational Study

Predicting short- and long-term glycated haemoglobin response after insulin initiation in patients with type 2 diabetes mellitus using machine-learning algorithms

Sunil B Nagaraj et al. Diabetes Obes Metab. 2019 Dec.

Abstract

Aim: To assess the potential of supervised machine-learning techniques to identify clinical variables for predicting short-term and long-term glycated haemoglobin (HbA1c) response after insulin treatment initiation in patients with type 2 diabetes mellitus (T2DM).

Materials and methods: We included patients with T2DM from the Groningen Initiative to Analyse Type 2 diabetes Treatment (GIANTT) database who started insulin treatment between 2007 and 2013 and had a minimum follow-up of 2 years. Short- and long-term responses at 6 (±2) and 24 (±2) months after insulin initiation, respectively, were assessed. Patients were defined as good responders if they had a decrease in HbA1c ≥ 5 mmol/mol or reached the recommended level of HbA1c ≤ 53 mmol/mol. Twenty-four baseline clinical variables were used for the analysis and an elastic net regularization technique was used for variable selection. The performance of three traditional machine-learning algorithms was compared for the prediction of short- and long-term responses and the area under the receiver-operating characteristic curve (AUC) was used to assess the performance of the prediction models.

Results: The elastic net regularization-based generalized linear model, which included baseline HbA1c and estimated glomerular filtration rate, correctly classified short- and long-term HbA1c response after treatment initiation, with AUCs of 0.80 (95% CI 0.78-0.83) and 0.81 (95% CI 0.79-0.84), respectively, and outperformed the other machine-learning algorithms. Using baseline HbA1c alone, an AUC = 0.71 (95% CI 0.65-0.73) and 0.72 (95% CI 0.66-0.75) was obtained for predicting short-term and long-term response, respectively.

Conclusions: Machine-learning algorithm performed well in the prediction of an individual's short-term and long-term HbA1c response using baseline clinical variables.

Keywords: cohort study; database research; insulin therapy; observational study; primary care; type 2 diabetes.

PubMed Disclaimer

Conflict of interest statement

The authors have no conflicts of interest.

Figures

Figure 1
Figure 1
Architecture of the proposed supervised machine learning based HbA1c response prediction system. LT, long‐term; MICE, multiple imputation by chained equations; ST, short‐term
Figure 2
Figure 2
Flow chart illustrating patient inclusion and exclusion criteria used in this study to select patients for the final analysis. GIANTT, Groningen initiative to analyse type 2 diabetes treatment; HbA1c, glycated haemoglobin
Figure 3
Figure 3
Heatmap illustrating the of set of clinical variables selected (in columns) by the elastic net regularization techniques across different leave one out iterations (in rows) for A, short‐term and B, long‐term HbA1c response prediction, respectively. The colour bar indicates the weightage assigned by elastic net to discriminate between responders and non‐responders. Higher intensity in the colormap indicates variables that are more robustly informative (selected more consistently across different iterations of model training). ACR, albumin‐to‐creatinine ratio; AU, acarbose use; BMI, body mass index; CV, history of cardiovascular disease; DMD, type 2 diabetes melitus duration (≥2 years); DPP‐4, dipeptidyl‐peptidase‐4‐inhibitors use; eGFR, estimated glomerular filtration rate, GLD, other oral glucose‐lowering drugs use;HbA1c, glycated haemoglobin; HDL, HDL cholesterol; MA, micro/macro‐albuminuria; MU, metformin use; ML, malignancy; PSC, psychological conditions PV, peripheral vascular disease; SBP, systolic blood pressure; TC, total cholesterol; TG, triglycerides; LDL, LDL cholesterol; SM, smoking status; SU, sulphonylurea use; TZD, thiazolidines use
Figure 4
Figure 4
Comparison of glycated haemoglobin (HbA1c; mmol/Mol) levels in four groups against time predicted by the elastic net regularization‐based generalized linear model. The graph shows mean values with 95% confidence interval. The time axis is divided into 6‐month intervals. Here group 1, short‐ and long‐term responders; group 2, short‐term responders; group 3, no change in response; group 4, long‐term responders. To obtain distinct subgroups of patients, we obtained a median probability outputs of the generalized linear model. HbA1c, glycated haemoglobin
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Rutten G, Grauw WD, Nijpels G, et al. NHG‐Standaard Diabetes mellitus type 2 (derde herziening). Huisarts Wet. 2013;56:512‐525.
    1. Lee SJ, Stijacic‐Cenzer I, Barnhart C, McClymont K, Steinman MA. Changing patterns of glucose‐lowering medication use in VA nursing home residents with diabetes, 2005 to 2011. J Am Med Dir Assoc. 2015;16(10):898.e9‐898.e14. - PMC - PubMed
    1. Sharma M, Nazareth I, Petersen I. Trends in incidence, prevalence and prescribing in type 2 diabetes mellitus between 2000 and 2013 in primary care: a retrospective cohort study. BMJ Open. 2016;6(1):e010210. - PMC - PubMed
    1. van HKJ, Drion I, Kleefstra N, et al. A prospective observational study of quality of diabetes care in a shared care setting: trends and age differences (ZODIAC‐19). BMJ Open. 2012;2(4):e001387. - PMC - PubMed
    1. Cantrell RA, Alatorre CI, Davis EJ, et al. A review of treatment response in type 2 diabetes: assessing the role of patient heterogeneity. Diabetes Obes Metab. 2010;12(10):845‐857. - PubMed

Publication types