Assessment of cardiovascular effects of non-insulin glucose-lowering agents. Major cardiovascular events in new users of non-insulin glucose-lowering agents: observational longitudinal study in the Catalan population-based electronic health record database, SIDIAP, 2010-2015

Study type

• Study type:
  Non-interventional study
• Scope of the study:
  Effectiveness study (incl. comparative)

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Anatomical Therapeutic Chemical (ATC) code:
  (A10B) BLOOD GLUCOSE LOWERING DRUGS, EXCL. INSULINS

  BLOOD GLUCOSE LOWERING DRUGS, EXCL. INSULINS

Population studied

• Age groups:
  Adults (18 to < 46 years)

  Adults (46 to < 65 years)

  Adults (65 to < 75 years)

  Adults (75 to < 85 years)

  Adults (85 years and over)
• Estimated number of subjects:
  270000

Study design details

• Main study objective:
  To evaluate the effect of currently marketed non-insulin glucose-lowering agents on major CV outcomes in cohorts of Spanish population based on records of population-based EMR SIDIAP
• Outcomes:
  Composite of three-components of mayor cardiovascular events (MACE): cardiovascular death, non-fatal myocardial infarction (MI) and non-fatal stroke. Secondary Composite Outcome is a MACE of four components: CV death, non-fatal MI, non-fatal stroke and hospitalization due to unstable angina or coronary revascularization procedures
• Data analysis plan:
  Incidence rates of primary and secondary composite outcomes events and secondary outcomes events will be estimated for each cohort during follow-up. Incident rates will be presented per 1000 patient-years and their corresponding 95% confidence intervals (CIs). Hazard ratios of PCO, SCO and SO will be calculated between cohorts (treated vs. non-treated) for each therapeutic group and, secondarily, for each given agent. Data will be analysed with multivariate Cox proportional-hazard regression models, once verified proporcionality assumptions. To control potential biases for confounding factors, the differences between exposed and non-exposed populations to the different hypoglycemic agents will be adjusted by estimating a propensity index using a logistic regression model. In order to control the effect of time-dependent confounders the use of marginal structural models will be also considered.