Incidence of Diabetic Ketoacidosis among Patients with Type 2 Diabetes Mellitus Treated with SGLT2 inhibitors or Other Antihyperglycemic Agents- A Retrospective, Observational, New-User Cohort Study Using 4 Administrative Claims Databases in the US

Study type

• Study topic:
  Disease /health condition

  Human medicinal product
• Study type:
  Non-interventional study
• Scope of the study:
  Assessment of risk minimisation measure implementation or effectiveness

  Disease epidemiology

  Safety study (incl. comparative)
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Name of medicine:
  INVOKANA
• Study drug International non-proprietary name (INN) or common name:
  CANAGLIFLOZIN
• Anatomical Therapeutic Chemical (ATC) code:
  (A10BK02) canagliflozin

  canagliflozin
• Medical condition to be studied:
  Diabetic ketoacidosis

Population studied

• Short description of the study population:
  Patients who had a diagnosis of T2DM preceding new use of an SGLT2i or at least one pre-specified comparator AHAs during the study period and had at least 365 days of continuous enrollment prior to the first day of new drug exposure (index date).
• 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)
• Special population of interest:
  Other
• Special population of interest, other:
  Type 2 diabetes mellitus patients
• Estimated number of subjects:
  409360

Study design details

• Main study objective:
  The main objective of this study is to compare diabetic ketoacidosis (DKA) incidence between new users of sodium-glucose co-transporter 2 inhibitors (SGLT2i, combined and separate) and new users of other antihyperglycemic agents (AHAs) among patients diagnosed with type 2 diabetes (T2DM). This study will also identify precipitating events and evaluate risk factors for incident DKA.
• Outcomes:
  Diabetic ketoacidosis
• Data analysis plan:
  The crude incidence rates of DKA in the different AHA new-user groups will be estimated as the number of incident DKA cases divided by the total follow-up time at risk. Baseline patient characteristics including risk factors for DKA will be summarized for those treated with SGLT2i versus other AHAs. A conditional Cox proportional hazards model based on time-to-first event approach will be used to estimate Hazard Ratio (HR) associated with SGLT2i (combined and separate) versus other AHAs, with each exposure propensity-score matched set treated as a separate stratum in Cox model. The exposure propensity score will be estimated through large-scale regularized regression, with demographics, all prior conditions/drugs/procedures, risk scores, utilization density as baseline covariates. An empirical p-value calibration using negative control outcomes will be conducted to address potential systematic bias. HRs, 95% CIs, pre- and post- calibration p values will all be reported.