Directly observed therapy (DOT) refers to the taking of medications in the presence of a medical provider, either in the provider's office or at the patient's home, so the provider can actually watch the pills being swallowed. To date, the use of DOT has been used mainly to help control the tuberculosis (TB) epidemic. DOT programs are an effort to increase medication compliance to achieve better treatment successes. High treatment completion rates, substantial increases in cure rates, and improved survival associated with TB DOT have been documented in the United States. DOT programs have also made great strides decreasing the TB burden in many developing countries. A growing number of projects in Asia, South America, and Africa demonstrate that population-based DOT is feasible in countries without an otherwise well developed public health infrastructure.
The strategy of the World Health Organization (WHO) for TB includes DOT as one of its components. Given the difficulty in predicting which patients will adhere to the anti-TB regimen prescribed, the WHO supports "universal" DOT for TB. DOT programs frequently consist of more than just observed therapy. They include various additional interventions to increase compliance through incentives, aggressive outreach methods to re-engage persons, medication reminders, motivated and supportive staff, improved access to health services, and additional external funds. It is likely that it is the range of interventions associated with DOT that makes it so successful; DOT may simply be a marker of a more intensive effort towards TB control. Participants identified the access to good medical care and staff as the most important aspects of the DOT strategy. DOT alone without other supportive interventions is unlikely to be effective. This is corroborated by a study that found no difference in the treatment success rates between self-observed therapy (SOT) and DOT when other supportive measures were not used. However, when combined with these other interventions to improve adherence, DOT has increased the success rate of TB treatment when other non-DOT strategies have failed. Acceptance of DOT by both pill takers and providers has generally been high, particularly when DOT is addressing a medical concern, such as TB control. Lastly, in resource-poor settings, DOT overall has been shown to be more cost efficient than SOT for TB.
Since DOT for TB has been largely successful in many settings worldwide, and since similarities exist between HIV and TB, it is reasonable to consider using the DOT strategy for highly active antiretroviral therapy (HAART). TB and HIV are similar in the development of multi-drug resistance due to exposure to inadequate or intermittent treatment. They are both global health threats and have complex treatment regimens with multiple toxicities. Several pilot programs have had some preliminary success using DOT and modified DOT to increase adherence to HAART, but a large randomized controlled trial has yet to be done. (For people who take medications multiple times a day, only the pills that are taken in the morning are observed in "modified DOT.") Plus, most studies to date have concentrated on special populations of people with HIV, such as prisoners, people in methadone maintenance programs, children, people with TB, and people with a past record of poor adherence.
Rather than enrolling prisoners, S.L. Hader enrolled 148 people admitted to two residential AIDS treatment facilities in New York. In an antiretroviral-experienced population, although DOT achieved an impressive 99 percent adherence rate, only two-thirds of the residents achieved an undetectable viral load and 41 percent had a treatment failure at least once, regardless of their regimen.
Modified DOT has been easily incorporated into methadone maintenance and TB treatment programs, since the patients are already observed swallowing their morning medications during the week. B. Conway found that after a median follow-up of 11 months, 9 of 13 people on a once-a-day HAART regimen and 8 of 12 on a twice-a-day regimen had a viral load below 400 copies/mL. During an 8-week intervention at a methadone maintenance program in San Francisco, participants receiving modified DOT showed slightly better adherence rates on all outcome measures when compared to usual care participants, but there were no significant group differences in any measures at the 1-month follow-up. In HIV-positive people with TB, it was found that supervision of the morning dose of HAART led to better adherence with evening medications and more awareness of potential side effects of the medication regimen.
Research using DOT has also focused on populations with documented or anticipated non-adherence. M.S. Stenzel was referred 49 high-risk people, of whom 37 agreed to participate. Of the 12 who refused to participate, 5 reported the ability to take their own medications without assistance, 4 felt the regular home visits were intrusive, 2 were homeless and unable to arrange regular contact, and 1 was unwilling to initiate HAART. Modified DOT for HAART was given at the patient's home every morning during the week and prepackaged medication was given for the afternoon and weekend doses. Six participants (16 percent) missed visits for 2 weeks and were disenrolled, 3 (8 percent) discontinued HAART, 5 (14 percent) opted to take medications without modified DOT because they didn't like home visits, and 17 (46 percent) completed 12 months of modified DOT. There was a significant virologic response on DOT, but there was no control group as a comparison. At enrollment, 4 of 37 people (11 percent) had a viral load below 400 copies/mL; at 3 months, 14 of 30 (47 percent) people had achieved that level; and at 12 months, 10 out of 18 (56 percent) had. There was also an immunologic advantage to DOT. The mean CD4+ cell count increased to 245 cells ug/L at 3 months and 339 cells ug/L at 12 months.
DOT for HAART may be more cost-effective and ultimately more successful in resource-poor settings where the cost of antiretrovirals is relatively high, especially when compared to the cost of personnel. Since 1998 in Haiti, one of the poorest countries in the world, Dr. Paul Farmer has been conducting a preliminary demonstration project of DOT for HAART. The program is modeled after the existing TB-control program in the country and involves a community health worker observing the ingestion of pills and providing support and assistance with social services. An initial cohort of 60 people has been enrolled and although outcomes have not been systematically analyzed, 18 of 20 had undetectable viral load after 4 to 30 months of follow-up.
Unlike TB, however, HIV requires long-term, potentially life-long, therapy. The goal of DOT programs for HAART is to improve education and foster medication self-administration over a limited time period, leading to long-term improvement in medication compliance. However, the data on long-term results associated with DOT is very limited. In the study by M. Fischl, HIV-positive individuals who received DOT had both greater short- and long-term virologic responses compared with those that received SOT. After 80 weeks, the number of patients with an HIV viral load below 400 copies/mL was 95 percent for those receiving DOT and 75 percent for those receiving SOT. M.S. Stenzel found that 47 percent of people understood how to take their medications at baseline and 80 percent understood at 3 months. At 12 months, most people felt prepared to take their HAART with minimal assistance. J.L. Sorenson, on the other hand, found that although modified DOT distributed at a methadone maintenance program enhanced medication adherence, it did not produce effects enduring beyond the time of the intervention.
It is still too early to judge long-term adherence following completion of DOT. The optimal duration of DOT for HAART, the most appropriate candidates to target and its cost effectiveness are all questions that still need to be answered. The AIDS Clinical Trial Group has begun a study to compare DOT to standard, self-administered medications. This study is discussed in more detail in this issue of the STEP Perspective.