Risk of thiazide-induced metabolic adverse events in older adults.

objectivesTo evaluate the risk and predictors of thiazide-induced adverse events (AEs) in multimorbid older adults in real-world clinical settings.

designObservational cohort study.

settingNational Veterans Affairs data from 2007 to 2008.

participantsVeterans aged 65 and older newly prescribed a thiazide (N = 1,060) compared with propensity-matched nonusers of antihypertensive medications (N = 1,060). MEASUREMENTS: The primary outcome was a composite of metabolic AEs defined as sodium less than 135 mEq/L, potassium less than 3.5 mEq/L, or a decrease in the estimated glomerular filtration rate (eGFR) of more than 25% from the baseline rate. Secondary outcomes included sev-ere AEs (sodium <130 mEq/L, potassium <3.0 mEq/L, or a decrease in eGFR of more than 50%).

resultsOver 9 months of follow-up, 14.3% of new thiazide users developed an AE, compared with 6.0% of nonusers (number needed to harm (NNH) 12, 95% confidence interval (CI) = 9-17, P < .001); 1.8% of new users developed a severe AE, compared with 0.6% of nonusers (NNH = 82, P = .008), and 3.8% of new users had an emergency department visit or hospitalization with an AE, compared with 2.0% of nonusers (NNH = 56, P = .02). Risk of AEs did not vary according to age, but having five or more comorbidities was associated with 3.0 times the odds (95% CI = 1.4-6.2) of developing an AE as having one comorbidity (hypertension). Low-normal and unmeasured baseline sodium and potassium values were among the strongest predictors of hyponatremia and hypokalemia, respectively. Only 42% of thiazide users had laboratory monitoring within 90 days after initiation.

conclusionThiazide-induced AEs are common in older adults. Greater attention should be paid to potential complications in prescribing thiazides to older adults, including closer laboratory monitoring before and after initiation of thiazides.