Maintaining long-term adherence to HIV medications is difficult, especially for youth3
for whom barriers to adherence are multiple.5
We wanted to enroll such youth with demonstrated adherence problems to study the feasibility and challenges of delivering a complex, resource-demanding, community-based DOT intervention as well as to refine the model to deliver it. Although excellent (>90%) adherence to the DOT visit requirements was seen in only about a third of participants, the acceptability of this intervention based on participant feedback was overwhelmingly positive.
Despite the choice to receive DOT anywhere in the community and at anytime in the day, most participants selected their home and the earlier part of the day to receive DOT. In light of these expressed preferences, the significantly higher DOT interaction “no-show” rate for visits scheduled earlier in the day is surprising and needs further exploration in future studies. For youth, the majority of whom were unemployed, not in school, and stably housed, we incorrectly anticipated the earlier part of the day as providing more predictable structure and likelihood for DOT success for what ended up being a predominantly home-based intervention.
DOT was clearly not for all, and even among the 14 participants who completed exit interviews and were overall very positive about the intervention, the ability to meet with the DOT facilitator at the prearranged time was quite variable. Perhaps the >90% adherence to DOT interactions to be deemed a DOT success was too high an expectation given the unpredictability of youth schedules and motivations. Substituting some of the DOT interaction requirements with cell phone-based support from DOT facilitators19
may be one strategy worth exploring to reduce the time commitment from the in-person meetings without compromising the frequency of participant contact.
Although this study was not designed/powered to identify who is best suited for DOT, a higher baseline depressive symptom score was observed for participants who were adherent to DOT as prescribed, raising the possibility these participants found the interactions with the DOT facilitators supportive, and for that reason kept up with them. Similar findings were noted in a DOT versus standard of care randomized clinical trial for HIV-infected adults starting HAART therapy in Mombasa, Kenya.20
Among participants who had moderate to severe depression at study entry, those who received DOT were seven times as likely to be adherent in the initial 24-week period and more likely to have viral suppression at 48 weeks than similar controls. Decreased depression scores also were noted in these Kenyan participants after receiving DOT. This finding could be particularly relevant since depression is one of the commonly observed clinical barriers to medication adherence,21
suggesting DOT potentially would be suited to participants with this condition.
A lack of durable benefit post-DOT was clearly seen in this study, with only three of six participants who had >90% adherence at week 12 able to sustain this level after DOT was discontinued. A similar observation has been reported for other adherence improvement interventions19,22,23
suggesting the core barriers to nonadherence were not resolved during or as a result of the intervention. The optimal duration of initial DOT and the timing and duration of DOT “boosters” for waning adherence need to be further studied.
As seen from participant exit interview responses DOT is much more than merely the act of observing persons take their medications. A participant quote from one of the exit interviews with reference to missing the facilitator when DOT ended is particularly poignant and emphasizes the desire for support and investment in the lives of these youth. The participant states “[I] liked having the support and someone to see everyday—liked having someone to care about me and who I also care about
.” One anticipates the facilitator–participant interactions create a relationship whereby the participant views the facilitator as a support person, a motivator, and a person who helps triage the participant's needs. Although we incorporated many of these elements when defining the scope of the DOT facilitators' role in this study and designing their training program, there is an opportunity to do more. For example, DOT facilitators could be trained in motivational interviewing to use these techniques in the field, thus taking one additional adherence intervention of interest24
to the patient's doorstep. In addition, our study results and the youth participant feedback underscore the need to develop community-based support for HIV-positive youth that is innovative and may utilize “nontraditional” support systems such as the church,25
Such community-based support rather than the mere act of being observed while taking medications, in the long run, may be the answer to sustaining the benefits from short-term DOT.
Finally, among the various interventions studied to improve adherence to medications, DOT has been particularly contentious given the resources it requires. While a recent systematic review and meta-analysis using virologic suppression as the primary outcome measure concluded no benefit from DOT over self-administered therapy in the general HIV patient population, it noted a marginal benefit of DOT in groups that were judged to be at high risk of nonadherence and in trials of short duration (<6 months).28
Similarly, based on simulation models, in patients with lower baseline levels of adherence or advanced disease, even very expensive, moderately effective adherence interventions are likely to confer cost-effectiveness benefits that compare favorably with other interventions.29
The cost of DOT is in large part driven by the salary of the DOT staff respective to case load with the cost distributed over the number of participants receiving this intervention. With limited patient accrual at each site, our pilot study cannot provide accurate cost estimates of this intervention. We anticipate when offered to a subpopulation of high-risk patients who are failing clinic-based interventions to improve adherence, and with creative approaches to improve program flexibility (start, stop, and reinitiation of DOT), cost-effective DOT is feasible.
In summary, although this study shows community-based DOT is safe, feasible, and as per participant feedback, acceptable to youth, it also highlights that this intervention is not for all and the benefits for those who improve their medication adherence are not sustained after DOT is discontinued. Depressed youth with poor social support appear to be one subgroup that would benefit from this intervention. As we recognize the limits of clinic-based interventions to improve adherence, DOT and other community-based interventions are the next frontier for adherence research. Given the cost and resources involved in delivering DOT and the need to get youth-specific feasibility and logistics information in this regards, we chose a small sample size and a pilot study design. The experience gained from this pilot study should help inform the design of future community-based intervention studies in youth. Clearly poor adherence is more than just “forgetfulness” and expectations from DOT for HIV-infected patients is more than just another reminder intervention. We see tremendous potential for and envision customizing community-based DOT “plus” interventions that address youth-specific barriers to adherence, by packaging elements such as case management, motivational interventions, and directly observed therapy and delivering them at the doorstep.