LAUSR

Question In adults, is initiation of benzodiazepines linked to increased risk for all-cause mortality? Methods Design Retrospective cohort study using administrative data from a commercial health care insurance database (Optum Clinformatics Datamart). Setting USA. Patients 1930159 patients 18 years of age who received a prescription for a benzodiazepine (alprazolam, chlordiazepoxide, clobazam, clonazepam, clorazepate, diazepam, estazolam, flurazepam, lorazepam, midazolam, oxazepam, quazepam, triazolam, temazepam) between Jul 2004 and Dec 2013 were identified for the user cohort (date of initial prescription = index date). For each patient in the user cohort, a patient with a physician visit within 14 days of the index date was randomly selected from the database (comparator cohort). Patients could be eligible for both the user and comparator cohorts if they started benzodiazepines after their randomly selected physician visit; these patients were analyzed in the comparator cohort. Patients in both cohorts met the following inclusion criteria: continuous insurance coverage for 6 months, no use of benzodiazepines in the past 6 months, 1 filled prescription of any medication at 90 days, and 1 filled prescription of any medication at 91 to 180 days. Through use of propensity scores, 1252988 patients (mean age 46 y, 65% women) in the user cohort were matched 1:1 with a patient in the comparator cohort and included in the intention-to-treat analyses. Risk factors Initiation of a benzodiazepine. Propensity scores used for matching users of benzodiazepines with comparator patients were based on >200 variables identified by investigators or an automated high-dimensional propensity score algorithm and included demographic variables (age, sex, index year), comorbid conditions, lifestyle factors, medications used, and health care use. Outcome All-cause mortality, obtained through linkage with the Social Security Administration Death Master File at 6 (primary outcome), 12, and 48 months after inclusion. Main results Propensity scorematched analyses showed that benzodiazepine use was not associated with all-cause mortality at 6 months; secondary analyses showed associations between benzodiazepine use and all-cause mortality at 12 and 48 months. In subgroup analyses, benzodiazepine use was associated with all-cause mortality {at 6 months}* in patients <65 years of age (4.3 vs 4.0 deaths/1000 person-y, hazard ratio [HR] 1.09, 95% CI 1.02 to 1.15) and those using short-acting benzodiazepines (11.3 vs 10.8 deaths/1000 person-y, HR 1.06, CI 1.02 to 1.10). Conclusions In adults, benzodiazepines were not linked to increased risk for all-cause mortality at 6 months. In secondary analyses, the drugs were linked to increased risk for mortality at 12 and 48 months. Association between benzodiazepines and all-cause mortality in adults Follow-up period Deaths/1000 person-y Hazard ratio (95% CI) Benzodiazepine user cohort Comparator cohort 6 mo 9.3 9.4 1.00 (0.96 to 1.04) 12 mo 7.8 7.7 1.04 (1.01 to 1.08) 48 mo 6.0 6.1 1.05 (1.02 to 1.07) CI defined in Glossary. Patients in the user and comparator cohorts (n =1252988 in each cohort) were matched 1:1 based on >200 variables (demographics, comorbid conditions, lifestyle factors, medications used, and health care use) identified by investigators or an automated high-dimensional propensity score algorithm. Commentary Patorno and colleagues address an important issue, citing unmeasured confounding bias as a threat to the validity of previous studies of sedativehypnotic exposure and mortality (1-3). Alas, null bias threatens the validity of their study. They defined users as those exposed to benzodiazepines, but omitted z-drugs, carbamates, barbiturates, ethanol, and valerenic acid. Like benzodiazepines, these drugs are ligands of positive allosteric modulator sites of -aminobutyric acid (GABA)A receptors and enhance GABAergic inhibition of central neurotransmission. Users were those with a new benzodiazepine prescription fill, excluding those with benzodiazepine fills in the previous 6 months. This is a weak criterion, as some users probably had negligible exposure to benzodiazepines; better would be 1 refill. Comparators had no benzodiazepine prescription fills 6 months before the match date, but some had benzodiazepine fills afterwards. Both users and comparators had exposure to other GABAergic agonists. Matching included fills for other sedating drugs but ignored dose and exposure duration of benzodiazepines, other GABAergic agonists, opioids, and other sedatives. These flaws blur differences between users and comparators, obscuring any effect on mortality. There is no administrative dataset so vast, nor propensity-matching algorithm so clever as to compensate for these design flaws. There is other compelling evidence that benzodiazepines increase mortality (4). Prudence dictates avoidance of sedativehypnotics for indications lacking evidence of safety.