Kenya: Despite free ARVs, delayed treatment often due to poverty

Over a third of participants with an eligible CD4 cell count in a cohort of HIV serodiscordant couples in Nairobi, Kenya had not started ART within a year in spite of regular CD4 testing, referrals to local treatment programmes, counselling on the importance of ART and free access to drugs.

Brandon L. Guthrie and colleagues report their findings from a two-year prospective study in the advance online edition of the Journal of Acquired Immune Deficiency Syndromes.

While there were no direct costs associated with treatment, indirect costs including transportation delayed ART start. The lower the socioeconomic status among participants, the greater the delay in starting ART. After adjusting for CD4 cell count those paying higher rents started ART at half the rate of homeowners and those paying lower rents did so at a third of the rate of homeowners.

The benefits of antiretroviral treatment are well known. In sub-Saharan Africa close to 50% of those in need of ART are not getting it.  Approximately half of all couples affected by HIV are in discordant relationships.

Timely start of ART in such couples benefits both the infected partner as well as reducing the risk of transmission to the uninfected partner. A couples-centred approach with the uninfected partner involved in the decision-making process of starting ART and providing support once treatment starts can be an effective strategy to ultimately improve treatment outcomes.

Attention has been focused on system-level barriers to ART, notably on making antiretroviral drugs more affordable. Little attention has been directed to what happens at the individual level when drugs are freely available.

The authors cite studies in South Africa that have shown that within a year of receiving an HIV diagnosis and eligible CD4 cell count only 39% started ART.

Poor retention in pre-ART programmes is one of the primary obstacles to improving ART coverage.

The authors chose to follow a cohort of HIV discordant couples where both partners were aware of each other’s HIV status to look at the time to starting ART and those factors that delayed the start and so identify potential interventions.

From September 2007 to December 2009 439 discordant couples were recruited from voluntary counselling and testing centres in Nairobi, Kenya. Eligible couples reported sex three or more times in the three months before screening and planned to stay together for the length of the study. They agreed to be followed on a quarterly basis for up to two years.

Questionnaires were administered and CD4 cell counts and viral loads taken at each visit. Those with eligible CD4 cell counts were referred to a local PEPFAR-funded treatment centre approximately 500 metres from the study clinic where CD4 cell counts were done independently of the study.

Two-thirds of the HIV infected partners were female. Median CD4 cell count at enrolment was 405 cells/mm³ (IQR: 280-586 cells/mm³). Females were younger with a median age of 28 years compared to 36 years for males and had been in the relationship for a shorter time than the men, median length of 4.8 years compared to 6 years.

During follow-up 33% (146) had eligible CD4 cell counts to start ART, of which 81 (55%) were female. The median time to starting ART from CD4 eligibility was 8.9 months (95% CI: 6.0-10.0). By six months from CD4 eligibility 42% had started ART and by one year 63.4% were on ART.

CD4 cell count at the time of eligibility was the strongest predictor for starting ART. The lower the CD4 cell count the shorter the time to starting ART; 56.3%, 42.6% and 32.2% by six months for those with eligible CD4 cell counts of under 100, 100-200 and 200-250, respectively.

The authors note such findings highlight potential future challenges as guidelines recommend starting ART at higher CD4 cell counts. Those with lower CD4 cell counts are sicker and more likely to want to start treatment, whereas those with higher counts may not feel unwell, are worried about side effects and so delay starting treatment.

Contrary to other studies the authors did not find that men started ART more slowly than women and suggested this was because of the nature of the study population who willingly engaged in care by agreeing to participate.

The authors undertook a secondary analysis looking at ART start and its association with a woman becoming pregnant. They found that women on ART were less likely to become pregnant compared to those not yet on ART (4.5 per 100 woman years and 10.7 per 100 woman years, respectively).

Additionally there was evidence to suggest that women starting ART chose not to become pregnant. Women were 81% more likely to report using hormonal contraceptives after starting ART than those not yet on ART (RR=1.81, 95% CI: 1.02-3.19). These findings, they note, contradict previous findings of higher rates of fertility of women on ART.

While they suggest differences may be explained by differences in the study populations and follow-up time, they stress the critical need in the management of women with HIV for access to family planning resources as well as a better understanding of the decision-making process among women and their reproductive wishes and starting ART.

Strengths of the study include a relatively large sample size and long period of follow-up.

Limitations that preclude generalising these findings: this study is a research cohort where both partners were aware of each other’s status so leading to faster start of ART, especially among men. Regular and consistent counselling over a two-year period probably resulted in a higher uptake so underestimating the true extent of the problem in the general population.

The authors conclude even where treatment is accessible and free, delays in starting ART persist. “Factors of lower socioeconomic status may slow starting ART and targeted approaches are needed to avoid such delays."