Suicide rates tumble in people with HIV in Canadian study but psychiatric drug use remains high in Danish people with HIV

This article is more than 11 years old. Click here for more recent articles on this topic

A study presented at the 14th European AIDS Conference last week found that the suicide rate in people with HIV in British Columbia, Canada, had fallen 35-fold since 1996 and was now close to that in the general population. However, a second study from Denmark found rates of psychiatric drug use two to three times higher than in the general population, especially of sleeping pills and tranquillisers, and a marked rise in the use of anti-psychotic drugs after seven or eight years of diagnosis.

Suicide rates in British Columbia

Rates of depression, anxiety and suicide are higher in people with HIV than in the general population.  But a longitudinal survey, of 5229 people on antiretroviral therapy in British Columbia in Canada between 1996 and 2012, found that the suicide rate fell from a rate of nearly 1% a year (961 suicides per 100,000 patients a year) to 28 per 100,000 (below 0.03%) in 2010 – a nearly 35-fold fall. In 2011, the last full year of data, there were actually no suicides recorded in the British Columbian cohort at all. In total, 82 people (2%) of patients killed themselves during the study period.

The 28/100,000 rate is still three times that seen in the general population, but the rarer suicide rates become among people with HIV, the larger the confidence interval becomes around the actual rate so the ‘true’ rate could differ considerably from this.

Seventeen per cent of people (911) died of causes other than suicide during the study period. Among the other 4318 patients, in univariate analysis, those with less than 95% self-reported adherence were 6.25 times more likely to commit suicide. The suicide rate declined by 23% for each 100 cells/mm3 increase in people’s CD4 count and multiplied by more than three per every ten-fold increase in viral load. This seems to support the idea that people in treatment failure may have a higher suicide risk. Suicide was also nearly four times higher among injecting drug users even in multivariate analysis though, as presenter Jasmine Gurm remarked, overdoses can sometimes look like suicides and vice versa.

Other findings are more counterintuitive, though. Among the whole patient group, having never had an AIDS-defining illness was associated with a 6.6-times-higher suicide rate – though this could be partly explained by there being more non-suicide deaths in people with AIDS. Among the 4318 who did not die of other causes, however, in multivariate analysis, suicide was 4.45 times more likely among people who had never had an AIDS-defining illness than in ones who had.

Jasmine Gurm commented that people needed a certain amount of energy to commit suicide, so people who were very sick might be less likely to be able to kill themselves, but this does not explain why having had an AIDS diagnosis was persistently associated with a 77% reduction in the risk of suicide. Maybe people who have had AIDS-defining diagnoses get more intensive monitoring and support – or maybe, in this population, being threatened with death paradoxically gives them something to live for.

Psychiatric prescriptions in Denmark

The Danish study, also presented at the 14th European AIDS Conference, looked at the use of psychotropic prescription drugs – antidepressants, tranquillisers, sleeping pills, anxiety drugs and antipsychotics – among 3615 of the country’s HIV-positive adults between 1995 and 2009. It compared prescriptions for these drugs with 32,535 members of the general population, matching each person with HIV with nine age- and gender-matched members of the general population.

People with HIV were certainly prescribed more psychotropic drugs. During the observation period,  just under twice as many people with HIV had ever been prescribed antipsychotics and antidepressants as the general population controls; twice as many had taken anxiety drugs; and nearly three times as many had taken sedatives and sleeping pills.

In terms of how long people spent on various drugs, rather than just whether they were prescribed them, people with HIV were hardly any more likely to spend time on antipsychotics than the general population. But they spent 76% more time on anxiety drugs, 2.28 more time on antidepressants and 4.42 times more time on sedatives and sleeping pills. Antidepressants were almost exclusively taken by gay men, who used them over three times more than the general population; use in heterosexuals was no higher than average.

The researchers looked at drug use over time, starting two years before diagnosis and continuing up to ten years after diagnosis (Denmark, uniquely, can do this because they keep central records of prescriptions for all patients).

The HIV-positive people already had higher rates of psychotropic drug use than the general population before their HIV diagnosis: relative to the general population, in the year before diagnosis they took 22% more antidepressants, 68% more anxiety drugs, and twice as many sedatives and sleeping pills.

In the two years after diagnosis, antidepressant use soared to twice that in the general population and stayed twice that in the general population thereafter,  and sedative/sleeping pill use doubled in the year after diagnosis and then stayed at about three times higher than that in the general population. In anti-anxiety drugs, use rose in the first year after diagnosis but then started to fall until, at five to six years after diagnosis, their use was no higher than in the general population. However, their use started to rise again eight years after diagnosis.

The HIV-positive cohort was using fewer antipsychotics than the general population before diagnosis and continued to use lower or similar levels up to six years after diagnosis. After this, their usage rate rose steeply, to twice that in the general population. The researchers speculate that this, and the similar late rise in anti-anxiety drugs, may accompany neurocognitive problems, but though antipsychotic usage rises with time since diagnosis, it is not related to age.

There was no strong association between taking antiretroviral therapy and psychotropic drugs. Use of sedatives and antidepressants was slightly higher in people on ART, and anti-anxiety drug use was about 25% lower, but differences were not significant. There was no relationship between the use of efavirenz and psychotropic drugs, though the researchers speculate that doctors might specifically exclude people with psychological problems from efavirenz use.

The researchers criticised the overuse of sleeping pills and sedatives in HIV-positive people – especially as there is an association between overuse of these drugs and the development of dementia.


Gurm J et al. Declining suicide rates among people living with HIV (PLHIV) initiating HAART between August 1996 – June 2012 in the HAART Observational Medical Evaluation and Research cohort in British Columbia, Canada. 14th European AIDS Conference, Brussels, abstract PS 5/3, 2013. View the abstract on the conference website.

Rasmussen LD et al. Utilization of psychotropic drugs prescribed to persons with and without HIV infection: a Danish nationwide population-based cohort study. 14th European AIDS Conference, Brussels, abstract PS 5/4, 2013. View the abstract on the conference website.