Association between glycated hemoglobin (HbA1c), weight and hypoglycemia to treatment choice and medication adherence in Type 2 Diabetes Mellitus (T2DM)

Application Number
Lay Summary

Type 2 diabetes patients who have difficulty taking their medications as prescribed are at risk of developing other diabetes-related health problems, which may lead to increased public healthcare spending. Medications prescribed by doctors are often tailored to individual patient needs, in order to balance how well a specific drug works against the risk of side effects, including weight gain and low blood sugar levels (hypoglycaemia).1 However, the precise relationships that exist between treatment choice, achieving normal blood sugar levels (glycaemic control), treatment-related side effects and how well patients comply with their prescriptions (medication adherence) are unclear.

Recent studies have investigated links between medication adherence and glycaemic control2; and links between hypoglycaemia and the risks of heart disease and death.3 However, these studies do not assess all possible relationships between treatment choice, medication adherence, and treatment side effects. Furthermore, these studies do not investigate the treatments that are frequently prescribed by doctors, such as combinations of different blood sugar-lowering drugs.

By analysing clinical data that is routinely collected in the UK, this study aims to relate changes in blood sugar levels, weight and incidence of hypoglycaemia, to treatment choice and medication adherence in type 2 diabetes patients.

Technical Summary

The objectives of this retrospective cohort study are to characterise associations between HbA1c, weight and hypoglycemia to treatment choice and medication adherence in T2DM. Patient-level data will be extracted from the CPRD, to retrospectively analyse a cohort of T2DM patients (>18 years of age) who were prescribed one of five treatment regimens between 2008 and 2016: OHA monotherapy; OHA dual therapy; OHA triple therapy; GLP-1-based therapy; insulin-based therapy. Outcomes of interest in this study are: 1-year change in HbA1c from baseline; 1-year change in weight from baseline; hypoglycemia incidence (1-year total events and rate); medication adherence, including medication compliance (defined by patients' calculated medication possession ratio, MPR) and medication persistence (defined by patients' proportion of days covered, PDC; and time to treatment intensification).

Univariate and multivariate methods of statistical analysis will characterise the associations between HbA1c, weight and hypoglycemia to treatment choice and medication adherence. Summary descriptive statistics will be generated, characterising patient demographics, clinical and treatment characteristics, and medication use. Stratified descriptive statistics and regression models for medication adherence and HbA1c will be fitted to the data adjusting for the influence of observed covariates (potential confounders) including demographic, clinical and socioeconomic factors, other prescriptions, comorbidities and centre effects.

Health Outcomes to be Measured

Health outcomes: The following outcomes will be derived and analysed:
- 1-year HbA1c change from baseline
- 1-year weight change from baseline
- 1-year hypoglycemia incidence (total and rate per patient year)
- Medication compliance (MPR)
- Medication persistence (PDC; time to first treatment intensification including predictors of persistence)


Jason Gordon - Chief Investigator - Health Economics & Outcomes Research Ltd ( HEOR Ltd )
Jason Gordon - Corresponding Applicant - Health Economics & Outcomes Research Ltd ( HEOR Ltd )
Iskandar Idris - Collaborator - University of Nottingham
Jorge Puelles - Collaborator - Takeda Europe Ltd
Phil McEwan - Collaborator - Health Economics & Outcomes Research Ltd ( HEOR Ltd )