An Observational/Non-interventional Evaluation of Subject Outcomes for Type 2 Diabetes Mellitus (T2DM) Subjects Prescribed Dipeptidyl Peptidase-4 Inhibitors (DPP4i), Sodium-glucose Cotransporter-2 Inhibitors (SGLT2i) or Sulphonylureas (SUs) at First Intensification

Study type

• Study topic:
  Disease /health condition
• Study type:
  Non-interventional study
• Scope of the study:
  Disease epidemiology
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Medical condition to be studied:
  Type 2 diabetes mellitus

Population studied

• Short description of the study population:
  Adult patients with Type 2 diabetes mellitus (T2DM) on the CPRD GOLD database between 01 January 2002 to 31 December 2017.
• Age groups:
  • Adult and elderly population (≥18 years)

    • Adults (18 to < 65 years)

      • Adults (18 to < 46 years)
      • Adults (46 to < 65 years)
    • Elderly (≥ 65 years)

      • Adults (65 to < 75 years)
      • Adults (75 to < 85 years)
      • Adults (85 years and over)
• Special population of interest:
  Other
• Special population of interest, other:
  Type 2 diabetes mellitus patients
• Estimated number of subjects:
  40000

Study design details

• Main study objective:
  The primary objective is to estimate the change in HbA1c post-baseline (that is addition of an SU, DPP-4i or SGLT2i to metformin) stratified by baseline HbA1c, at Months 6 and 12.
• Outcomes:
  There is one primary outcome which evaluates HbA1c change from baseline to Months 6 and 12. The secondary outcomes include: evaluation of change from baseline in body weight at Months 6 and 12, discontinuation rates at Months 6 and 12 post-baseline, treatment intensification at Months 6 and 12 post-baseline, mean medicines possession ratio (MRP) across medication, medication persistence, urinary tract infection (UTI) rates, hypoglycaemic episode counts and rates of adverse events.
• Data analysis plan:
  Mixed-effects regression modelling will be used to test for significant differences in changes in HbA1c levels from baseline at 6 and 12 months for 3 different medications. Models will be fitted to data to adjust for the effect of observed covariates including:demographic, clinical factors, other prescriptions, comorbidities and centre effects. Appropriate regression techniques will be used in accordance with the assumed distribution of outcome variable. Descriptive statistics will be used to characterise patterns and interrelationships between different factors. If appropriate, models of rates will be estimated using survival analysis,and binary outcomes will be estimated using logistic or probit regression. Transformation of outcome, or alternatives such as Poisson and negative binomial regression may be used in case of non-normally distributed outcomes. Akaike’s Information Criteria may be used to inform variable inclusion.