Women living with HIV in the UK continue to have poorer treatment outcomes

Social problems make adherence harder

Women have triple the rate of poor treatment outcomes of gay men and the gap has not narrowed over the last decade, according to an analysis from one of London’s leading HIV clinics presented at the recent 15th European AIDS Conference in Barcelona, Spain. The conference also heard from another UK study which suggests that the gender gap is driven by poverty, housing difficulties and other socio-economic factors which make adherence challenging.

Poor treatment outcomes

The first study looked at 1675 people beginning HIV treatment at the Royal Free Hospital between 2001 and 2013. Lisa Burch and colleagues were interested in those participants who had a detectable viral load above 200 copies/ml, seen as an indicator of difficulties with HIV treatment. They looked at the first viral load measure one year after starting treatment and also the first one two years afterwards.

Whereas just over half the cohort were gay men, 28% were women and 17% heterosexual men. Two thirds of women and half of heterosexual men were black African. Women joining the cohort tended to be younger than the men.

Reflecting patterns of late HIV diagnosis in the UK, women began HIV treatment with a median CD4 cell count of 205 cells/mm3 and heterosexual men with 166 cells/mm3. In contrast, gay men’s median CD4 cell count was 273 cells/mm3.


detectable viral load

When viral load is detectable, this indicates that HIV is replicating in the body. If the person is taking HIV treatment but their viral load is detectable, the treatment is not working properly. There may still be a risk of HIV transmission to sexual partners.


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

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.

One year after starting treatment, 6.5% of gay men had a detectable viral load, compared to 13.0% of heterosexual men and 20.4% of women.

After two years, the proportions were similar – 8.4% of gay men, 12.7% of heterosexual men and 19.4% of women.

Looking at changes between 2001 and 2013, the prevalence of detectable viral loads dropped for all groups, but the gap in outcomes did not narrow. Women had the poorest outcomes throughout the period of study.

Treatment interruptions explained most of the poorer outcomes. Whereas few men have taken treatment interruptions in recent years, around one in seven women have done so.

A separate study – from the large UK CHIC cohort – also showed that women, black Africans and younger people bear a disproportionate burden of the deaths from AIDS-defining illnesses in the UK. Typically, those who died of AIDS had been diagnosed late, began treatment with a low CD4 cell count, had difficulties with adherence and were unable to maintain an undetectable viral load.

Socio-economic factors

In order to assess whether socio-economic factors could explain these kinds of differences in treatment outcomes, Rebecca O'Connell examined data from another UK cohort, the ASTRA (Antiretrovirals, Sexual Transmission Risk and Attitudes) study.

Four years ago, 2445 people who had been taking HIV treatment for at least six months gave researchers more information about a wider range of social factors than is routinely collected. Of note, most of this information was based on self-report. The participants also gave permission for these data to be linked with their medical records.

While the study was more successful in recruiting gay men than members of other groups, the ethnic and age profile of the women and heterosexual men recruited was comparable to that of the Royal Free cohort in the first study. Both women and heterosexual men had been taking HIV treatment for an average of seven years.

At baseline, 8.3% of gay men, 14.0% of heterosexual men and 11.4% of women had a detectable viral load.

Socio-economic problems at baseline were more commonly reported by women and heterosexual men than by gay men. For example, 49% of heterosexual men and 54% of women said that they only sometimes or never had enough money for their basic needs, compared to 20% of gay men.

Furthermore, whereas 6% of people who always had enough money had a detectable viral load at baseline, this was the case for 15% of those who never had enough.

Having unstable housing (including staying with friends or homelessness) was reported by 18% of heterosexual men and 17% of women, compared to 6% of gay men. While 17% of people with unstable housing had a detectable viral load, 5% of homeowners did.

While levels of education and employment were quite high across all groups, they were somewhat lower in women. People with less education or no job were more likely to have a detectable viral load.

People who did not speak fluent English (17% of women, compared to 3% of gay men) and people with symptoms of depression (33% of women, compared to 25% of gay men) were more likely to have a detectable viral load.

In unadjusted analysis, women had a 38% higher risk of a detectable viral load, compared to gay men (prevalence ratio 1.38, 95% confidence interval 1.03 – 1.85). The greater risk was lessened in a series of analyses which made statistical adjustments for the impact of age, financial hardship, home ownership, employment and education.

When the figures were adjusted to take account of all socio-economic factors and depression, there was no difference in the risk of having a detectable viral load between women and gay men (prevalence ratio 0.96, 95% confidence interval 0.68 – 1.37).

The picture was similar for heterosexual men. In unadjusted analysis, they had a 69% greater risk of detectable viral load than gay men (prevalence ratio 1.69, 95% confidence interval 1.22 – 2.35). After taking all socio-economic factors and also depression into account, this was substantially but not entirely lessened (prevalence ratio 1.39, 95% confidence interval 0.95 – 2.02).

Of note, these analyses took depression into consideration alongside socio-economic problems. As www.aidsmap.com has previously reported, depression was highly prevalent in people with socio-economic problems in this cohort and was independently associated with poorer treatment outcomes.


“The differences in virological outcomes between men who have sex with men and women largely seem to be explained by socio-economic circumstances,” Rebecca O’Connell concluded. However, the explanation is less clear for heterosexual men.

“This suggests that interventions aimed at assisting those at socio-economic disadvantage may improve viral load outcomes in women and to some extent in men who have sex with women,” she continued.

Similarly, Lisa Burch – presenting the first study from the Royal Free – called for improved, tailored social support for women and heterosexual men living with HIV. This may help people maintain adherence to treatment and avoid treatment interruptions.


Burch L et al. Is the Gender Difference in Viral Load Response to ART Narrowing over Time? 15th European AIDS Conference, Barcelona, abstract PS6/3, 2015.

Jose S et al. Ongoing Contribution of AIDS Deaths in the HAART Era: Data from the UK CHIC Study. 15th European AIDS Conference, Barcelona, abstract PE12/7, 2015.

O'Connell R et al. Do Socio-economic Factors Explain Gender Differences in Virological Response to ART in the UK? 15th European AIDS Conference, Barcelona, abstract PS6/5, 2015.