Table 1

Cohort groups and study methods

Cohort descriptionCovariates included in different analysesNumber of statin non-usersNumber of statin users
Primary analysis
PS-matched cohort
 Pairs of statin users and non-users were matched based on PS1. Unadjusted model: no adjustment was necessary since treatment groups were fully matched67286728
2. Adjusted for medications that are associated with obesity or diabetes during the follow-up period*67286728
3. Adjusted for medications that are associated with obesity or diabetes and undergoing non-invasive cardiac or revascularization procedures during follow-up period†67286728
4. Adjusted for medications that are associated with obesity or diabetes, undergoing non-invasive cardiac or revascularization procedures, and number of inpatient admissions and outpatients encounters during follow-up period‡67286728
Sensitivity analysis
PS-matched cohort
 Pairs of statin users and non-users were matched based on PSDefinition of follow-up period was changed to start 360 days after the index date.
Unadjusted model: no adjustment was necessary since treatment groups were fully matched.
67286728
Secondary analyses
Diabetes incident cohort
 Excluded patients who had diabetes at baseline1. PS-adjusted model§46,8107852
2. Fully adjusted model¶46,8107852
Diabetes and obesity incident cohort
 Excluded patients who had diabetes or obesity at baseline1. PS-adjusted model§41,4245793
2. Fully adjusted model¶41,4245793
Healthy cohort
 Excluded patients with any Charlson comorbidity index or severe chronic disease1. PS-adjusted model§35,7834563
2. Fully adjusted model¶35,7834563
Non-obese cohort
 Excluded patients with obesity at baseline or at follow-up1. PS-adjusted model§29,8215340
2. Fully adjusted model¶29,8215340
Overall cohort: non-users vs high-intensity statin users
 Compared non-users vs high-intensity statin users in the overall cohort1. PS-adjusted model§49,5452470
2. Fully adjusted model¶49,5452470
Statin users only: high-intensity statin users vs moderate-intensity/low-intensity statin users
 Included all statin users only, comparing outcomes between high-intensity statin users to moderate-intensity/low-intensity statin users1. PS-adjusted model§None included10,910
2. Fully adjusted model¶None included10,910
  • *Adjustment for use of medications that are associated with obesity or diabetes (β-blockers, diuretics, systemic corticosteroids, selective serotonin reuptake inhibitors, antipsychotics, hormone replacement therapy, smoking cessation therapy, and cytochrome P450 inhibitors53).

  • †Adjustment for use of medications that are associated with obesity or diabetes and undergoing an non-invasive cardiac procedures (cardiac stress test, echocardiography, and cardiac catheterization) or invasive revascularization procedures (percutaneous coronary intervention, coronary artery bypass graft surgery, and peripheral arterial revascularization procedures).

  • ‡Adjustment for use of medications that are associated with obesity or diabetes, undergoing non-invasive cardiac or revascularization procedures, and number of inpatient admissions and outpatients encounters during the follow-up period.

  • §Adjusted for PS.

  • ¶Adjusted for PS, medications that are associated with obesity or diabetes, undergoing non-invasive cardiac and revascularization procedures, and number of inpatient admissions and outpatients encounters during the follow-up period.

  • PS, propensity score.