People with HIV who selected treatment partners to support their adherence were more likely to return to the clinic to collect further doses of antiretrovirals, and showed a higher rate of viral suppression after six months of treatment, but showed no longer-lasting advantage in terms of viral suppression, CD4 cell counts or mortality, Nigerian and American researchers report in the Journal of Acquired Immune Deficiency Syndromes.
In contrast to other studies the authors found that living closer to the clinic was associated with better adherence and virological outcomes. They suggest this may be an argument in support of decentralisation of HIV services in resource-poor settings, but say that they are unable to establish a causal association, and suggest further study.
Adherence is vital to successful antiretroviral treatment. Drug resistance and treatment failure are associated with non-adherence.
While there are high rates of adherence in resource-poor settings, there is always a need to develop improved adherence strategies for better health outcomes.
Health care workers and volunteers play an important role in ensuring adherence. Disclosure of HIV status is central to this, and is independently associated with the prevention of virologic failure. However, concern over confidentiality as well as the sustainability of such an approach led the authors to look at the little studied role of pre-existing social networks, in particular patient-selected treatment partners, on health outcomes.
Over a period of 18 months (June 2006-December 2007) 499 HIV-infected treatment naïve adults (over 15 years of age) attending the Jos University Teaching Hospital (JUTH) ART clinic in Nigeria were randomised to standard-of-care (SOC) or patient-selected treatment partner-assisted therapy (TPA).
Each patient was followed for 48 weeks. Outcomes studied included virologic control (viral load reduction of greater than 1 log at week 12 and viral load below 400 copies at weeks 24 and 48), adherence to drug pick-up, CD4 cell counts and mortality.
All antiretroviral drugs were given out at the clinic pharmacy. Before starting ART all patients took part in a two-hour interactive adherence education session. Led by openly HIV-infected registered nurses trained in HIV adherence counselling, the sessions, conducted in the local language (Hausa) as well as English, took place in a dedicated room in the clinic.
At each drug pick-up visit (28 days) the study pharmacist, unaware of the treatment arm, gave further support as well as targeted information related to self-reports of adherence and side effects. The study pharmacists are trained in adherence counselling and have a minimum of three years experience at the JUTH ART clinic. Patients at 24 weeks with a viral load above 400 copies/ml had intensive adherence re-training.
Those assigned to the TPA arm received the interventions previously described (the standard of care) and chose a treatment partner aware of the patient’s status and who lived in the same house or very close by. TPAs attended one adherence education session. TPAs received no compensation but received travel allowances if needed. They were asked to watch participants taking their drugs at least once a day, help with reporting of, and management of side effects, as well as reminding participants when it was time to pick up drugs.
At week 24 more participants with TPAs had viral suppression compared to those with SOC alone ( 61.7% versus 50.2% (odds ratios (OR)=1.58, 95% CI 1.11 to 2.226, p<0.05). However at week 48 there was no significant difference in viral load suppression (65.3% (TPA) compared to 59.4% (SOC) OR=3.06, 95% CI:0.89 to 1.84. p>0.05). There was no significant difference in CD4 count or mortality rate in either arm at week 48.
The authors suggest that the absence of a long-term impact of TPAs on viral suppression is explained by the fact that all patients who had a viral load above 400 copies/ml at week 24 received focused adherence retraining. However the authors refer to another randomised study that also found no long-term virologic benefit with treatment partners.
Other explanations they note include the possibility that even a relatively low level of adherence (>54%) may be sufficient to ensure viral suppression with an NNRTI-based regimen. As a consequence, modest increases in longer-term adherence in the TPA group may not result in a superior virologic outcomes in the TPA group.
The high level of adherence seen in both arms is consistent with previous studies in sub-Saharan Africa, note the authors.
The TPA arm were three times more likely to demonstrate 95% adherence as measured by drug pick-up rates at week 24 (89% versus 72% in the SOC group, OR = 3.06. 95% CI: 1.89-4.94, P<0.01), and twice as likely to demonstrate 95% adherence at week 48 (80% versus 67% for the SOC arm, OR=1.95. 95% CI: 1.29-2.93, P<0.01).
The authors note, as have other investigators, a disparity between drug pick-up rates and virologic suppression when using pharmacy records as a measure.
They stress drug pick-up adherence does not mean that patients will take the drugs as prescribed, highlighting the limitation of drug pick-up as a measure of adherence. The authors note while this is the first randomised study to show a link between improved drug pick-up adherence and treatment partner-associated therapy, it did not show lasting virologic, immunologic or mortality benefits.
They suggest their findings point to the importance of looking at non-adherence endpoints in the evaluation of adherence interventions and to the evaluation of interventions not focused on drug pick-up alone but which include the taking of the drug.
The authors conclude that since improved drug pick-up adherence is associated with the use of patient-selected treatment partners but not on lasting viral suppression, CD4 count increase or mortality, strategies are needed to adapt improved drug pick-up adherence that will result in lasting changes to health outcomes.
Referente
Babafemi, O Taiwo et al. Assessing the virologic and adherence benefits of patient-selected HIV treatment partners in a resource-limited setting. J Acquire Immune Defic Syndr Vol (advance online publication), 2009.