Newer antidepressants not necessarily safest for older people

Selective serotonin reuptake inhibitors (SSRIs) are associated with an increased risk of several severe adverse outcomes in older people compared with older tricyclic antidepressants (TCAs), a study found.

To investigate the association between antidepressant treatment and the risk of a number of potentially life-threatening outcomes in older people, researchers identified from 570 general practices in the U.K. a total of 60,746 patients ages 65 and older with a newly diagnosed episode of depression between 1996 and 2007. Of that group, 54,038 (89%) received at least one prescription for an antidepressant. A total of 6,484 patients (10.7% of the entire cohort) received one prescription, and 6,624 (10.9%) received 60 prescriptions or more during follow-up. The median duration of treatment was 364 days (interquartile range, 91 to 1,029).Of the nearly 1.4 million prescriptions issued, 55% were for SSRIs, 32% for TCAs, 0.2% for monoamine oxidase inhibitors (MAOIs), and 13.5% for other classes.

Antidepressant use was then analyzed against several adverse outcomes, including all-cause mortality, attempted suicide or self-harm, heart attack, stroke, falls, fractures, epilepsy or seizures, and hyponatremia. Results were published online Aug. 2 by BMJ.

SSRIs were associated with an increased risk of all-cause mortality, epilepsy or seizures, stroke, falls, fracture and hyponatremia compared with TCAs. The group of other antidepressants was associated with an increased risk of all-cause mortality, attempted suicide or self-harm, stroke, fracture, and epilepsy or seizures compared to TCAs.

Depressed patients not taking antidepressants at all had a 7% risk of dying (absolute risk of all-cause mortality) in the next year, while the comparable risks were 8.1% for those taking TCAs, 10.6% for those taking SSRIs, and 11.4 % for those taking other antidepressants. For stroke, one-year risks were 2.3%, 2.6% and 3.0% for TCAs, SSRIs, and other antidepressants, respectively, compared to 2.2% for those not on antidepressants, and one-year risks for fracture were 2.2%, 2.7% and 2.8% compared to 1.8%.

The authors attempted to reduce indication bias by restricting the study cohort to patients with a recorded diagnosis of depression so all patients had the same indication for treatment, whether treated or not. They also attempted to adjust for channeling bias by adjusting at baseline for potential confounders, such as falls, and attempted to adjust for residual confounding through a self-controlled case series analysis. The authors pointed out that TCAs were prescribed at lower doses than SSRIs and other antidepressant drugs, which could in part explain the findings.

An accompanying editorial said “[T]he study has clear implications for more informed prescribing and enhanced clinical monitoring,” adding, “Given the potential harms, the decision to prescribe for an older person with depression should not be taken lightly.”