Suicide not associated with efavirenz use in the D:A:D cohort study

Simon Collins, HIV i-Base

An analysis from the D:A:D cohort study, presented as an oral abstract by Colette Smith at the 2014 HIV Drug Therapy Glasgow Congress, was notable for reporting no association between suicide in HIV positive people taking efavirenz in European cohorts. [1]

This is important because last year a combined analysis of four studies from the US ACTG network reported a three-fold higher suicide rate in people randomised to efavirenz (n = 3241) compared to non-efavirenz (n = 2091) combinations, even though a relatively low number of actual or attempted suicides were reported (n=22). Suicidality incidence in the ACTG study was 8.08 vs 3.66 per 1000 person-years (47 vs 15 events), in the efavirenz vs efavirenz-free group respectively (hazard ratio, 2.28 [95% CI, 1.27 to 4.10]; p=0.006). [2]

The side effects profile for efavirenz is complex in that efavirenz can affect mood changes, and both suicidal ideation and suicide have been reported. A history of depression and anxiety is a contraindication that makes efavirenz unsuitable for approximately 10% of HIV positive people. A further 10-20% who start with efavirenz, additionally switch to an alternative within the first year.

The potential for underestimating the impact of severe mood changes in clinical management prompted the D:A:D cohort to see whether similar findings to the ACTG study were recorded in this large European cohort collaboration. The analysis included data from almost 49,717 patients followed prospectively, using CoDe methodology that allows for multiple causes of death to be recorded, including one underlying cause, one immediate cause, and up to four contributory causes. The primary outcome in this analysis, included suicide or psychiatric illness listed as any cause, analysed by current ARV use.

All cause mortality included 4420 deaths from 371,333 person years of follow-up (rate 11.9 per 1000 PY). Suicide and psychiatric illness was the underlying cause in 133 cases (0.52 per 1000 PY) and underlying, immediate or contributory cause in 482 cases (1.30 per 1000 PY).

In analysis adjusted for age, gender, nadir and current CD4 count, time since diagnosis, cohort and HIV acquisition risk, neither use of efavirenz or other ARVs or being on ART were associated with suicide.

However, the association between HIV treatment and the risk of suicide being an underlying cause of death was significantly higher in treatment-experienced patients who had discontinued ARVs compared to treatment naive patients: adjusted IRR 3.24; 95%CI 1.95, 5.38. Comparable results were reported for suicide to be any contributory factor: adjusted IRR 2.29; 95%CI 1.63, 3.21.

Death from suicide/psychiatric disease was more common for lower current CD4 count, men, older age and if IDU was the HIV transmission risk.


An important part of the conclusion to this talk was a statement that the results do not support a lack of association between efavirenz and suicide.

Instead, the analysis provides reassurance that the way efavirenz is being used in these European cohorts is not leading to increased suicide rates.


  1. Smith C et al. Lack of association between use of efavirenz and death from suicide: the D:A:D Study. HIV Drug Therapy Glasgow Congress, 2-6 November 2014. Oral abstract O315. Journal of the International AIDS Society 2014, 17(Suppl 3):19512. (PDF)
  2. Mollan et al. Association between efavirenz as initial therapy for HIV-1 infection and increased risk for suicidal ideation or attempted or completed suicide: an analysis of trial data. Ann Intern Med 2014.161(1):1-10. doi:10.7326/M14-0293.

Links to other websites are current at date of posting but not maintained.