Objective: Thiazide diuretics are recommended as first line blood pressure lowering therapy. Previously conducted trials demonstrating cardiovascular benefit (like ALLHAT and SHEP), used Chlorthalidone and not Hydrochlorothiazide as the diuretic. Despite this, Hydrochlorothiazide is commonly prescribed and is available in many combination preparations. Little is known about the comparative effectiveness of the 2 drugs, as estimated glomerular filtration rate (eGFR) declines. We set out to examine the association of hydrochlorothiazide and cardiac, kidney and electrolyte-related adverse events compared to chlorthalidone by eGFR level. Design and method: Population-based, retrospective cohort study of adults of advanced age (> or = 66 years) with a diagnosis of hypertension in Ontario, Canada (2009–2016). Fine and Grey sub-distribution hazards models examining the association of matched hydrochlorothiazide vs. chlorthalidone pairs and outcome events in an intention to treat analysis. Study outcomes were a cardiac event, kidney (egfr decline> 30%) event, electrolyte disorder (hypo- or hyperkalemia, hyponatremia) or all-cause mortality. We examined the effect of eGFR groups (> 60, 45–59, < 45 ml/min/1.73m2) on outcomes using an interaction term. Results: A total of 9,786 hydrochlorothiazide users were matched 4:1 to 2,936 chlorthalidone users (mean age 73.6, 55% female, 41.6% diabetes). Baseline eGFR distributions were 73.4, 17.4 and 9.2% for > 60, 45–59, and <45 ml/min, respectively. In individuals with an eGFR > 60 ml/min, hydrochlorothiazide was associated with a lower risk of eGFR decline (HR 0.78 95%CI 0.70–0.88), CV events (HR 0.83 95%CI 0.76–0.92), hypokalemia (HR 0.54 95CI 0.48–0.60) and all-cause mortality (HR 0.79 95%CI 0.63–0.98) compared to chlorthalidone. These effects attenuated with lower eGFR levels and no differences were observed in hyperkalemia or hyponatremia. Conclusions: Among individuals with an eGFR > 60 ml/min, hydrochlorothiazide use was associated with a lower risk of eGFR decline, CV events, hypokalemia and all-cause mortality. These differential effects were not observed at lower eGFR levels.