icon-    folder.gif   Conference Reports for NATAP  
  19th Conference on Retroviruses and
Opportunistic Infections
Seattle, WA March 5 - 8, 2012
Back grey_arrow_rt.gif
Untreated Depression and Refusing ART Common in Seattle-Area HIV+ Deaths
  19th Conference on Retroviruses and Opportunistic Infections, March 5-8, 2012, Seattle

Mark Mascolini

More than half of HIV-positive people who died in King County, Washington in 2009 had a depression diagnosis, but only 57% of those people got antidepressant therapy [1]. One in 5 people who died refused antiretroviral therapy, and over 70% of deaths had a non-AIDS primary cause.

This descriptive analysis compared HIV-positive people who died in the Seattle/King County area in 2008-2010 and HIV-positive people who did not die in those years. In-depth medical record reviews and provider interviews focused on deaths in 2009. Multivariate analysis to identify death risk factors accounted for gender, race/ethnicity, homeless status, late HIV diagnosis, timeliness of establishing HIV care, and the specific variables being assessed.

Of the 6193 people studied throughout 2008-2010, 240 (3.9%) died. Nonwhites accounted for 35% of deaths and 34% of survivors, while men accounted for 88% of deaths and 89% of survivors. Men who have sex with men accounted for 52% of deaths compared with 69% of survivors, while injection drug users (IDUs) plus MSM/IDUs accounted for 29% of deaths versus 13% of survivors.

People who died were older than survivors (median 52 versus 46 years), had HIV infection longer (median 12 versus 10 years), had lower CD4 counts (59% versus 10% under 200), had higher viral loads (50% versus 75% undetectable), and were more likely to have an AIDS diagnosis within 6 months of their HIV diagnosis (43% versus 28%) (P < 0.01 for all comparisons).

For key death risk factors, the researchers figured an attributable risk percent, defined as the percent of risk associated with each factor among those exposed to that factor. They also determined the population-attributable risk, defined as the percent of all HIV-positive people with that risk factor. Attributable risk was highest for a recent CD4 count below 200 (86%), followed by an AIDS diagnosis (65%), detectable viral load or age over 40 (each 50%), injection drug use (44%), and age over 50 (33%). Comparing attributable risk with population-attributable risk yielded an odds ratio of 7.3 (95% confidence interval [CI] 5.2 to 10.3) for a recent CD4 count under 200, 2.8 (95% CI 1.7 to 4.8) for an AIDS diagnosis, and 2.0 or lower for all other risk factors.

In-depth analysis of causes of death among 87 people who died in 2009 determined that the primary cause was one or more AIDS illnesses in 29%, non-AIDS cancers in 23%, pneumonia or other pulmonary disease in 10%, self-harm in 10%, heart disease in 8%, liver disease in 6%, and cerebrovascular disease in 3%. Among those who died, 21% had refused antiretroviral therapy.

In the year before death, clinicians diagnosed depression as a new or ongoing condition in 55% of people who died, but only 57% of those diagnosed with depression were prescribed antidepressants. For comparison, 45% of people who did not die had a diagnosis of depression, and 78% of them got antidepressants. Among people who died, 47% had a diagnosis of hypertension, and 78% of them were taking antihypertensives. Among people who did not die, 23% had hypertension and 74% of them were on antihypertensives.

A depression diagnosis raised the risk of death 50%, but that risk fell short of statistical significance (odds ratio [OR] 1.5, 95% CI 0.8 to 2.6). Untreated depression significantly tripled the risk of death (OR 2.9, 95% CI 1.3 to 6.3). A hypertension diagnosis significantly tripled the risk of death (OR 3.0, 95% CI 1.6 to 5.6), but untreated hypertension did not significantly inflate the death risk.

Three factors related to substance use independently raised the risk of death: tobacco use (OR 2.0, 95% CI 1.0 to 4.0), cocaine or crack cocaine use (OR 2.7, 95% CI 1.2 to 6.3), and any illicit drug use (OR 2.8, 95% CI 1.4 to 5.6). There were trends toward a higher risk of death with alcoholism (OR 1.9, 95% CI 0.8 to 4.4) and methamphetamine use (OR 2.4, 95% CI 0.8 to 6.9).

Higher proportions of people who died than people who did not die made 0 or 1 outpatient visit in the past year, and a lower proportion of people who died made 4 or more visits. Making 0 or 1 outpatient visit versus 2 or more visits raised the risk of death 5 times (OR 5.2, 95% CI 1.9 to 14.2).

The researchers believe that "immunosuppression, advanced age, substance use, HIV viremia, refusal of antiretrovirals, and perhaps under-use of antidepressants suggest that potentially treatable illnesses contribute to premature mortality" in people with HIV. At the same time they observed that AIDS remains a major cause of death in people with HIV, despite availability of potent antiretroviral combinations.


1. Buskin S, Kent J, Barash E, Aboulafia D, Golden M. Inevitable vs preventable deaths among HIV+ people in King County: Washington, 2008-2010. 19th Conference on Retroviruses and Opportunistic Infections. March 5-8, 2012. Seattle. Abstract 1128. http://www.retroconference.org/2012b/PDFs/1128.pdf.