Life expectancy now considerably exceeds the average in some people with HIV in the US

Early treatment is key, but huge disparities still remain
This article is more than 11 years old. Click here for more recent articles on this topic

A study from the US has found that some groups of people with HIV, especially those treated before their CD4 count falls below 350 cells/mm3, now have life expectancies equal to or even higher than the US general population.

However, it also finds that life expectancy for some other groups – most notably women and non-white people – is still considerably below comparable members of the general population and that for people who inject drugs, life expectancy in the era of antiretroviral therapy (ART) has not improved at all.

A second study, which looked at death rates among both HIV-positive and HIV-negative members of two cohorts of people with or at risk of HIV, has found that the death rate from non-AIDS-defining illnesses among people with HIV who started ART above the 350 cells/mm3 threshold was not, and never has been, any higher than among comparable HIV-negative people.


AIDS defining condition

Any HIV-related illness included in the list of diagnostic criteria for AIDS, which in the presence of HIV infection result in an AIDS diagnosis. They include opportunistic infections and cancers that are life-threatening in a person with HIV.


Relating to the heart and blood vessels.


A mental health problem causing long-lasting low mood that interferes with everyday life.

cardiovascular disease

Disease of the heart or blood vessels, such as heart attack (myocardial infarction) and stroke.


In a case-control study, a process to make the cases and the controls comparable with respect to extraneous factors. For example, each case is matched individually with a control subject on variables such as age, sex and HIV status. 

In other words, the sole contributor to the increased mortality in people who started ART early was AIDS. This was not, however, the case for people who started ART later, who had raised mortality due to non-AIDS-related causes as well as due to AIDS.

Life expectancy in people on therapy, 2000-2007

The first study looked at death rates among, and then computed life expectancy for, 22,937 people with HIV in the US and Canada who started ART between the beginning of 2000 and the end of 2007. It compared their life expectancy at age 20 with the general population and noted how it had changed in the study’s eight years.

Life expectancy at age 20 in the US population is approximately 57 years in men (i.e. on average, and in the absence of further change, 50% will die by the age of 77) and 62 years in women (i.e. 50% chance of death by 82). In Canada, men can expect to live nearly three years longer than this and women just over two.

The study found that for the group as a whole and over the full eight years, the average life expectancy in people with HIV was just under 43 years, i.e. 50% will die by the age of 63 – 15 years earlier than men and 19 years earlier than women in the general US population.

However, there were huge disparities in life expectancies between different groups. Whereas people who inject drugs only had a life expectancy of 29 more years at age 20, for white people it was 52 years, for those starting treatment with a CD4 count above 350 cells/mm3 it was 55 years and for gay men it was 57 years – the same (or slightly higher) than in US men in general.

Furthermore, life expectancy had improved dramatically between 2000 and 2008 for most groups. In non-white people, even though life expectancy for those on ART between 2005 and 2007 was still only 48 more years at age 20 – i.e. nine years behind US men and 14 years behind US women – this was a dramatic improvement since 2000-2002 when non-white people on ART could expect, on average, to die at 50 – a gain of 18 years.

Life expectancy at age 20 had gone up 17 years in men, 10 years in women (though notably, this had not improved since 2005), by 13 years in gay men, by 12.5 years in heterosexual people, and by 20 years in those starting ART at CD4 counts over 350 cells/mm3.

This means that average life expectancy at age 20 was now equal to US men in the general population, among heterosexual people with HIV and in white people. It was also a remarkable 69 years at age 20 in gay men and people starting ART before 350 cells/mm3 – meaning that, if nothing else changed, these groups, as long as they stay on ART, have a 50/50 chance of seeing their 89th birthday – a full seven years longer than women in the general US population.

In contrast, life expectancy at age 20 in people who inject drugs had not changed at all and was still 29 years at age 20 in 2007, as it was in 2000.

Another sobering finding was that only 28% of the cohort had started ART before their CD4 count fell below 350 cells/mm3, though this proportion had improved over time.

Mortality rates in HIV-positive and -negative people

One of the problems with this kind of study is that like is not being compared with like. People with HIV will have many differences other than their status and their medication from the average member of the public, so differences in mortality could be due to all sorts of other factors.

A second study of mortality tried to get around this by comparing death rates in people who, apart from their HIV status, were closely similar. By doing this, it was able to tease out the proportion of deaths that were due to AIDS and therefore whether deaths due to non-AIDS-defining illness were any higher in people with HIV or on ART than they are in the general population.

This study looked at mortality due to AIDS-defining and non-AIDS defining illness in two long-standing US cohort studies – the Multicenter AIDS Cohort Study (MACS) and the Women’s Interagency HIV Study (WIHS). These long-standing cohort studies were set up in 1985 and 1993 respectively. MACS has recruited 6972 gay men who are either HIV positive or at high risk of HIV infection (41% with HIV at enrolment) and WIHS has recruited 4137 women who are either HIV positive or closely matched to the HIV-positive women in terms of characteristics (38% with HIV at enrolment).

This study compared mortality rates between the HIV-negative cohort members and the ones with HIV who were on combination antiretroviral therapy (cART). Because there were not large numbers of cohort members on cART who were either young or very old, it only looked at mortality in the ‘middle years’, between 35 and 70. For the people with HIV it looked only at mortality subsequent to them starting cART if they were older than 35 when they started. The study looked at mortality up to the end of 2010, so some people could have been on cART of various kinds for 15 years or more, if they started in the mid-1990s and were aged 35 to 55 at the time. Average length of follow-up was in fact 10.2 years: 11.7 years in the HIV-negative people and 7.6 and 8.1 years (depending on CD4 count at cART initiation) in the HIV-positive people on cART.

A high proportion of the cohorts – 60% or 6699 individuals – were included in this study. The first and most obvious fact is that mortality was a lot higher in the people with HIV, as you might expect: over the years, 540 out of 2953 people with HIV died (18.2%) compared with 165 out of 3854 HIV-negative people (3.4%). In terms of annual mortality rates, this is 2.32% per year in the people with HIV and 0.37% per year in the HIV-negative people.

The researchers then divided deaths in the people with HIV into AIDS-related and non-AIDS-related causes: 11.5% of the people with HIV died of AIDS and 6.7% of other conditions.

In one specific group, namely people with HIV who started cART with a CD4 count over 350 cells/mm3, mortality due to non-AIDS illness was no higher than it was in the HIV-negative people. However, even in this group, AIDS deaths predominated, more than doubling mortality, so overall mortality in this group was approximately 1% per year compared with approximately 0.4% in the HIV-negative people. This probably reflects the fact that many people would have died in the early years of sub-standard cART.

This is reflected in the fact that if people died of AIDS-related illness, they tended to do so much younger. Models were done that, based on the mortality rates seen, projected the likely future mortality rates of people over 70. These showed that in people who started cART at a CD4 count above 350 cells/mm3 and who died of AIDS, there was a 50% chance of death by the age of 54: in those who died of non-AIDS-related illness, 50% was not reached till the age of 75, no different from HIV-negative people. Thus people starting ART early were living near-normal lifespans as long as they avoided early death from AIDS, probably reflecting the generally improved lifespan and vastly decreased AIDS incidence of those who survived beyond the early 2000s.

The non-AIDS-related mortality in people who started cART at lower CD4 counts, however, was higher than in HIV-negative people. It was 66% higher in people starting cART at CD4 counts between 200 and 350 cells/mm3 and 115% higher in people starting it at CD4 counts below 200 cells/mm3, reinforcing the message that starting ART early is generally better for the health, not only because it stops AIDS-related illness. Other factors that increased the chance of death for people on cART were smoking (50% higher AIDS mortality and 120% higher non-AIDS mortality in smokers); depression (65% more non-AIDS mortality and 58% more AIDS mortality); and high blood pressure (42% higher AIDS and 30% higher non-AIDS mortality).

The women in WIHS had 40% higher mortality due to non-AIDS illness than the men in MACS, but no higher AIDS mortality.

The biggest influence on non-AIDS mortality was hepatitis B or C co-infection. This more than doubled non-AIDS mortality. HIV-negative people with hepatitis B or C died on average eight years younger than those without, and people with co-infection on cART 15 years younger than those with HIV alone.

More comparative data needed

In a separate editorial on the second paper, researchers Veronica Miller and Sally Hodder commented that improvements in life expectancy might be expected to continue in MACS and WIHS. They added that the second paper adds considerably to the evidence for earlier initiation of antiretroviral therapy; noting that over 40% of non-AIDS and non-hepatitis deaths were due to cardiovascular disease, and that non-AIDS deaths were higher in people who started ART later. They add that the paper continues to beg the question of whether inflammatory processes in untreated people with HIV do add to the risk of cardiovascular disease at lower CD4 counts.

Pointing out that the robustness of the findings on life expectancy and cause of death in the study is due to the accumulation of 25 or more years of data, they make a plea for continued government support of large cohort studies, saying: “Continued public funding of cohorts such as MACS, WIHS and others will be even more important as we enter the fourth decade of antiretroviral treatment and seek to optimise strategies to improve individual and public health.”


Samji H et al. Closing the gap: increases in life expectancy among treated HIV-positive individuals in the United States and Canada. PLOS ONE 8(12): e81355. Doi:10.1371/journal.pone.0081355. 2014.

Wada N et al. Cause-specific mortality among HIV-infected individuals, by CD4+ cell count at HAART initiation, compared with HIV-uninfected individuals. AIDS 28:257-265. 2014.

Miller V and Hodder S Beneficial impact of antiretroviral therapy on non-AIDS mortality. AIDS 28:273-274. 2014.

This news report is also available in Russian.