Back HIV Treatment Approved HIV Drugs IDWeek 2014: Efavirenz Not Linked to Suicide in Analysis of Insurance Records

IDWeek 2014: Efavirenz Not Linked to Suicide in Analysis of Insurance Records


The non-nucleoside reverse transcriptase inhibitor efavirenz (Sustiva, also in the Atripla single-tablet regimen) was not associated with a higher rate of suicidal thoughts or attempts in an analysis conducted by manufacturer Bristol-Myers Squibb (BMS), researchers reported at the 2014 IDWeek meeting last week in Philadelphia.

Efavirenz, one of the most widely used HIV medications, is highly effective but can cause neuropsychiatric or central nervous system side effects such as insomnia, abnormal dreams, anxiety, and dizziness.

At last year's IDWeek, Katie Mollan from the University of North Carolina at Chapel Hill reported findings from a combined analysis of 4 AIDS Clinical Trials Group (ACTG) studies showing that suicidal thoughts, attempted suicide, and death by suicide were more than twice as common among people with HIV who started treatment with efavirenz, although the overall or absolute risk was quite low. The study was later published in the July 1, 2014, Annals of Internal Medicine

At this year's meeting, Ella Nkhoma from BMS presented results from another multi-study analysis indicating that people taking efavirenz did not have a greater risk of suicide.

Nkhoma and colleagues compare the incidence of suicidal ideas or attempts among patients starting antiretroviral therapy (ART) using efavirenz-containing versus efavirenz-free regimens, after adjusting for confounding factors such as demographic factors, comorbid conditions including depression, and use of other medications.

This cohort analysis used administrative claims data from a U.S. commercial insurance database and a Medicaid database covering the period 2006-2013. Eligible participants were age 12 or older, starting ART for the first time, and enrolled in the database for at least 6 months before starting ART.

The commercial database included 19,983 patients, of whom 11,187 started efavirenz, while the Medicaid database included 5154 people, of whom 2224 started efavirenz. The mean age in both groups was about 40 years, and a higher proportion of patients in the commercial database were men (83% vs 53%).

Suicidality was defined based on the World Health Organization's International Classification of Diseases (ICD-9-CM) diagnostic codes for suicidal ideation or attempts. A suicide attempt was defined as an inpatient or emergency department claim with a diagnostic code for suicide attempt, while an expanded definition also included codes for injuries consistent with suicide attempts in people who had a psychiatric diagnosis recorded during the same visit.


  • In the commercial insurance database, suicidality rates were low in both patients taking and those not taking efavirenz (0.2% vs 0.4%, respectively).
  • In the Medicaid database, absolute rates of suicidality were a bit higher but still low overall for patients taking and not taking efavirenz (1.3% vs 2.9%, respectively).
  • Incidences of suicidality and suicide attempts by both definitions were higher for the Medicaid population compared with the commercial insurance group.
  • In both databases, patients starting efavirenz-free regimens were more likely to have a history of depression and other psychiatric conditions -- not surprising, as many doctors consider this a contraindication for efavirenz.
  • In the commercial insurance population, incidence of suicidality and suicide attempts according to the expanded definition did not differ significantly between efavirenz users and non-users in an adjusted analysis (adjusted hazard ratio [HR] 1.03 and 1.00).
  • Using the narrower definition of suicide attempts, the rate was higher among efavirenz users, but the difference did not reach statistical significance (adjusted HR 5.70).
  • In the Medicaid population, incidence of suicidality and suicide attempts according to the expanded definition were slightly lower among efavirenz users compared with non-users, but the difference was not significant (adjusted HR 0.90 and 0.71).
  • Using the narrower definition, the rate of suicide attempts was significantly lower among efavirenz users even after adjusting for other factors (adjusted HR 0.11).

"In this analysis of 2 large real world databases, HIV patients with depression and psychiatric conditions were less likely to be prescribed efavirenz," the researchers concluded. "Despite [propensity score]-adjustment, we did not find conclusive evidence of an increased risk of suicidality or suicide attempt among patients initiating an efavirenz-containing regimen."

It is unclear why suicide rates were higher in the ACTG study than in the BMS analysis, both overall and among efavirenz users compared to non-users. As a limitation of this study, the researchers noted that the insurance and Medicaid databases may not have captured all events, especially suicidal ideas.



E Nkhoma, J Coumbis, A Farr, et al. Using Real World Data to Assess the Risk of Suicidality among Patients Initiating an Efavirenz-containing Regimen versus an Efavirenz-free Antiretroviral Regimen. IDWeek 2014. Philadelphia, October 8-12, 2014. Abstract 646.