Expanding directly observed therapy: tuberculosis to human immunodeficiency virus

Abstract

The resurgence of tuberculosis and the association between poor treatment adherence and drug-resistant tuberculosis in the 1990s emphasized the importance of identifying ways to improve medication adherence. In particular, directly observed therapy improved adherence to tuberculosis therapy for several groups of patients, including those with frequent drug use, mental illness, and unstable housing, thereby contributing to the control of drug-resistant tuberculosis in New York City (1Chaulk C.P. Iseman M.D. Directly observed therapy.Lancet. 1997; 350: 666-667Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar, 2Curtis R. Friedman S.R. Neaigus A. et al.Implications of directly observed therapy in tuberculosis control measures among IDUs.Public Health Rep. 1994; 109: 319-327PubMed Google Scholar). This success was later extended to the delivery of isoniazid preventive therapy to similar patients (3Tulsky J.P. Pilote L. Hahn J.A. et al.Adherence to isoniazid prophylaxis in the homeless a randomized controlled trial.Arch Intern Med. 2000; 160: 697-702Crossref PubMed Google Scholar).Still, nonadherence remains the Achilles’ heel of medical treatment for many diseases, including tuberculosis. In this issue of The American Journal of Medicine, Chaisson and colleagues (4Chaisson R. Barnes G. Hackman J. et al.A randomized controlled trial of interventions to improve adherence with isoniazid therapy to prevent tuberculosis in injection drug users.Am J Med. 2001; 110: 610-615Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar) report on a three-arm, random assignment trial of isoniazid adherence. Injection drug users were randomly assigned to directly observed therapy, self-administered therapy with peer counseling and education, or self-administered therapy alone. All participants received either immediate or delayed cash incentives to promote adherence. The study adds two new contributions to the body of adherence research. First, directly observed therapy improved adherence among injection drug users. Second, injection drug users were successfully maintained in monthly care for more than 6 months, perhaps as a result of the cash incentives.This study raises several important questions for adherence strategists. First, why does directly observed therapy yield better results than psychoeducational interventions? Second, what is the role of incentives, especially cash, in improving adherence? Finally, should incentive-based directly observed therapy be used for patients with chronic diseases for which lifetime adherence is required, such as antiretroviral therapy for patients with human immunodeficiency virus (HIV) infection?Directly observed therapy versus psychoeducational approachesPsychoeducational interventions, including education and counseling, medication reminders, self-monitoring tools, and support groups, can improve adherence to therapy (5Haynes R.B. McKibbon K.A. Kanani R. Systematic review of randomised trials of interventions to assist patients to follow prescriptions for medications.Lancet. 1996; 348 ([published erratum appears in Lancet 1997;349:1180]): 383-386Abstract Full Text Full Text PDF PubMed Scopus (534) Google Scholar, 6Roter D.L. Hall J.A. Merisca R. et al.Effectiveness of interventions to improve patient compliance a meta-analysis.Med Care. 1998; 36: 1138-1161Crossref PubMed Scopus (814) Google Scholar), but their efficacy is modest compared with directly observed therapy. In the study by Chaisson, 77% of patients assigned to the directly observed therapy arm received 100% of their doses, compared with the 6% to 10% of patients in the other two groups. These results are consistent with preliminary findings among incarcerated patients with HIV infection, among whom directly observed therapy led to unprecedented levels of viral suppression (7Fischl M, Rodriguez A, Scerpella E, et al. Impact of directly observed therapy on outcomes in HIV clinical trials. Presented at the Seventh CROI, San Francisco, California, 2000.Google Scholar) while the effect of psychoeducational interventions has been more modest (8Tuldra A. Fumaz C.R. Ferrer M.J. et al.Prospective randomized two-arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy.J Acquir Immune Defic Syndr. 2000; 25: 221-228Crossref PubMed Scopus (213) Google Scholar, 9Gifford AL, Bormann JE, Shively MJ, et al. Effects of group HIV patient education on adherence to antiretrovirals. A randomized controlled trial. Presented at the Eighth CROI, Chicago, Illinois, 2001.Google Scholar).One explanation for the limited success of psychoeducational interventions is that they require a higher level of internal motivation on the part of the patient. A leading theory postulates that behavior change requires “motivational readiness” on the part of the patient (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar); in its absence, nonadherence to the adherence intervention is the rule. Using daily structure and financial incentives, directly observed therapy interventions place less reliance on a patient’s internal motivation to change behavior. As such, directly observed therapy programs are a strategic public health tool: the treatment of contagious tuberculosis cannot wait for the patient to become ready to adhere.While effective, directly observed therapy has limitations. Traditional programs do not promote patient-level commitment to the long-term skills required to sustain medication adherence. One study of modified directly observed therapy for HIV therapy, in which the first dose of the drug regimen was delivered 5 days a week, found that adherence dropped precipitously during the 2 nonobserved days (11Sorensen J.L. Mascovich A. Wall T.L. et al.Medication adherence strategies for drug abusers with HIV/AIDS.AIDS Care. 1998; 10: 297-312Crossref PubMed Scopus (86) Google Scholar). The lack of an exit strategy is a core problem for applying directly observed therapy to chronic diseases such as HIV. Psychoeducational interventions may have the advantage of promoting adherence skills that extend beyond the intervention. In practice, however, durable changes are uncommon (8Tuldra A. Fumaz C.R. Ferrer M.J. et al.Prospective randomized two-arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy.J Acquir Immune Defic Syndr. 2000; 25: 221-228Crossref PubMed Scopus (213) Google Scholar, 9Gifford AL, Bormann JE, Shively MJ, et al. Effects of group HIV patient education on adherence to antiretrovirals. A randomized controlled trial. Presented at the Eighth CROI, Chicago, Illinois, 2001.Google Scholar). Is indefinite adherence “life support” using directly observed therapy reasonable for patients who do not become adherent on their own?What is the role of incentives, especially cash, in facilitating adherence?Studies looking at incentives, including cash, have shown excellent results in facilitating directly observed therapy (12Giuffrida A. Torgerson D.J. Should we pay the patient? Review of financial incentives to enhance patient compliance.BMJ. 1997; 315: 703-707Crossref PubMed Scopus (267) Google Scholar). Financial incentives retain participants in care and are associated with reduced costs of outreach (13Tulsky J.P. White M.C. Considerations on the road to involuntary confinement.West J Med. 1999; 171: 48-49PubMed Google Scholar, 14Buchanan R.J. Compliance with tuberculosis drug regimens incentives and enablers offered by public health departments.Am J Public Health. 1997; 87: 2014-2017Crossref PubMed Scopus (19) Google Scholar, 15Gourevitch M.N. Alcabes P. Wasserman W.C. Arno P.S. Cost-effectiveness of directly observed chemoprophylaxis of tuberculosis among drug users at high risk for tuberculosis.Int J Tuberc Lung Dis. 1998; 2: 531-540PubMed Google Scholar). While incentives have gained acceptance for time-limited interventions, incentives for prolonged therapies have been more problematic. Financial incentives also approach the fine line between encouragement and coercion, particularly for low-income or vulnerable patients, and they may discourage the disclosure of toxicity to prevent the interruption of scheduled reimbursement. Lastly, some low-income patients may claim poor adherence to become eligible for incentives provided in these programs. Thus, for some patients, incentive-based programs may unintentionally promote poor adherence.Expanding directly observed therapy to HIV antiretroviral therapyThere is considerable overlap among groups of patients affected by tuberculosis and by HIV, including drug use, mental illness, and unstable housing. Should we start building incentive-based directly observed programs for HIV antiretroviral therapy? We suggest caution in extrapolating lessons learned in tuberculosis treatment and prevention to HIV. Although once-daily HIV antiretroviral therapy will make directly observed treatment more feasible in the future, it is unlikely to resemble therapy for tuberculosis, which can be administered two times weekly after an initial 2-week induction. More importantly, therapy for tuberculosis is time limited, whereas HIV therapy is life long. Because tuberculosis is transmitted through casual contact and treatment renders patients noninfectious quickly, there is a greater public health priority to deliver directly observed therapy for tuberculosis than for HIV infection, although similar arguments have been made that improved adherence and viral load suppression may not only delay disease progression, but also reduce HIV transmission (16Bangsberg DR, Perry S, Charlebois E, et al. Adherence to antiretroviral therapy predicts progression to AIDS. AIDS. 2001. In press.Google Scholar, 17Quinn T.C. Wawer M.J. Sewankambo N. et al.Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group.N Engl J Med. 2000; 342: 921-929Crossref PubMed Scopus (2477) Google Scholar).Questioning the effects of directly observed therapy on limiting HIV drug resistanceThe public health imperative to reduce transmission of multidrug-resistant tuberculosis generated much of the enthusiasm for the use of directly observed therapy, and the transmission of drug-resistant HIV is cited as an analogous concern. Yet, there is little empiric support for the hypothesis that improved adherence to HIV medications will prevent HIV drug resistance. In contrast, good adherence may be necessary for the selection of drug-resistant virus (18Bangsberg D.R. Hecht F.M. Charlebois E.D. et al.Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population.AIDS. 2000; 14: 357-366Crossref PubMed Scopus (874) Google Scholar, 19Walsh J, Hertogs K, Gazzard B. Viral resistance, adherence and pharmokinetic indices in patients on successful and failing PI based HAART. Presented at the 40th ICAAC, Toronto, Ontario, 2000.Google Scholar, 20Campo R. Suarez G. Miller N. et al.Efficacy of indinavir (IDV)/ritonavir (RTV)-based regimens (IRBR) among patients with prior protease inhibitor failure.Antiviral Ther. 2000; 5 (Abstract.): 23PubMed Google Scholar), possibly because resistant virus has decreased fitness and requires the sufficient drug pressure that is associated with good adherence (21Deeks S.G. Wrin T. Liegler T. et al.Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIV-infected patients with detectable viremia.N Engl J Med. 2001; 344: 472-480Crossref PubMed Scopus (599) Google Scholar). Thus, it is conceivable that adherence interventions that fall short of complete viral suppression may increase the development of drug resistance among poorly adherent patients. These considerations, however, should not impede efforts to improve adherence, delay disease progression, and improve survival: resistance may be a necessary price for these improved outcomes.The problem of selecting candidates for directly observed HIV therapyThe expectation that injection drug users, the homeless, or the mentally ill would be the only candidates for HIV directly observed therapy is not based on evidence that nonadherence is limited to these groups of patients. For example, recognizing that nonadherence extends beyond the poor, the New York City Department of Health recommends that all patients with active tuberculosis should receive directly observed therapy. In San Francisco, clinic staff prescribes directly observed therapy for all patients who are smear-positive for tuberculosis. While enforced attendance at infectious disease clinics to receive antituberculosis therapy provokes little outcry, a similar approach to HIV therapy would be politically unacceptable and would be perceived as burdensome or stigmatizing by many patients.We suggest that if directly observed therapy is proven effective outside of institutional settings, selection strategies should be based on the patient’s stage of disease and motivational commitment to medication adherence (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar). Objective measures of motivational readiness can predict success with tobacco cessation and adherence to exercise programs and hypertensive medication as well as HIV therapy (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar, 22Prochaska J.O. Velicer W.F. DiClemente C.C. Fava J. Measuring processes of change applications to the cessation of smoking.J Consult Clin Psychol. 1988; 56: 520-528Crossref PubMed Scopus (795) Google Scholar, 23Willey C. Redding C. Stafford J. et al.Stages of change for adherence with medication regimens for chronic disease development and validation of a measure.Clin Ther. 2000; 22: 858-871Abstract Full Text PDF PubMed Scopus (109) Google Scholar). For patients with a high motivational state, psychoeducational approaches should be attempted first, followed, if necessary, by directly observed therapy. Patients with a low motivational state and late stage-disease are the ideal candidates for directly observed therapy, which should be recommended regardless of socioeconomic status. Ultimately, directly observed therapy should not be required to receive antiretroviral therapy, which should not be postponed indefinitely if it or other forms of adherence support is declined (24Bangsberg D.R. Moss A. When should we delay highly active antiretroviral therapy?.J Gen Intern Med. 1999; 14: 446-448Crossref PubMed Scopus (40) Google Scholar).Recommendations and summaryDirectly observed therapy has contributed to the treatment and prevention of tuberculosis. Before implementing such programs for HIV care, we recommend a vigorous effort to identify effective strategies to improve treatment adherence and to fill the gap between pharmacotherapeutic studies and durable viral suppression in patients with HIV infection. Studies should compare the effectiveness of directly observed therapy and psychoeducational interventions. Effective exit strategies should be developed to extend the durability of time-limited programs, recognizing that some patients may need indefinite support. Rational and nonprejudicial selection strategies are needed to identify the best candidates for directly observed therapy. Ultimately, the value of directly observed therapy should be compared with other approaches to adherence support to measure their effects on HIV progression and even transmission. The resurgence of tuberculosis and the association between poor treatment adherence and drug-resistant tuberculosis in the 1990s emphasized the importance of identifying ways to improve medication adherence. In particular, directly observed therapy improved adherence to tuberculosis therapy for several groups of patients, including those with frequent drug use, mental illness, and unstable housing, thereby contributing to the control of drug-resistant tuberculosis in New York City (1Chaulk C.P. Iseman M.D. Directly observed therapy.Lancet. 1997; 350: 666-667Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar, 2Curtis R. Friedman S.R. Neaigus A. et al.Implications of directly observed therapy in tuberculosis control measures among IDUs.Public Health Rep. 1994; 109: 319-327PubMed Google Scholar). This success was later extended to the delivery of isoniazid preventive therapy to similar patients (3Tulsky J.P. Pilote L. Hahn J.A. et al.Adherence to isoniazid prophylaxis in the homeless a randomized controlled trial.Arch Intern Med. 2000; 160: 697-702Crossref PubMed Google Scholar). Still, nonadherence remains the Achilles’ heel of medical treatment for many diseases, including tuberculosis. In this issue of The American Journal of Medicine, Chaisson and colleagues (4Chaisson R. Barnes G. Hackman J. et al.A randomized controlled trial of interventions to improve adherence with isoniazid therapy to prevent tuberculosis in injection drug users.Am J Med. 2001; 110: 610-615Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar) report on a three-arm, random assignment trial of isoniazid adherence. Injection drug users were randomly assigned to directly observed therapy, self-administered therapy with peer counseling and education, or self-administered therapy alone. All participants received either immediate or delayed cash incentives to promote adherence. The study adds two new contributions to the body of adherence research. First, directly observed therapy improved adherence among injection drug users. Second, injection drug users were successfully maintained in monthly care for more than 6 months, perhaps as a result of the cash incentives. This study raises several important questions for adherence strategists. First, why does directly observed therapy yield better results than psychoeducational interventions? Second, what is the role of incentives, especially cash, in improving adherence? Finally, should incentive-based directly observed therapy be used for patients with chronic diseases for which lifetime adherence is required, such as antiretroviral therapy for patients with human immunodeficiency virus (HIV) infection? Directly observed therapy versus psychoeducational approachesPsychoeducational interventions, including education and counseling, medication reminders, self-monitoring tools, and support groups, can improve adherence to therapy (5Haynes R.B. McKibbon K.A. Kanani R. Systematic review of randomised trials of interventions to assist patients to follow prescriptions for medications.Lancet. 1996; 348 ([published erratum appears in Lancet 1997;349:1180]): 383-386Abstract Full Text Full Text PDF PubMed Scopus (534) Google Scholar, 6Roter D.L. Hall J.A. Merisca R. et al.Effectiveness of interventions to improve patient compliance a meta-analysis.Med Care. 1998; 36: 1138-1161Crossref PubMed Scopus (814) Google Scholar), but their efficacy is modest compared with directly observed therapy. In the study by Chaisson, 77% of patients assigned to the directly observed therapy arm received 100% of their doses, compared with the 6% to 10% of patients in the other two groups. These results are consistent with preliminary findings among incarcerated patients with HIV infection, among whom directly observed therapy led to unprecedented levels of viral suppression (7Fischl M, Rodriguez A, Scerpella E, et al. Impact of directly observed therapy on outcomes in HIV clinical trials. Presented at the Seventh CROI, San Francisco, California, 2000.Google Scholar) while the effect of psychoeducational interventions has been more modest (8Tuldra A. Fumaz C.R. Ferrer M.J. et al.Prospective randomized two-arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy.J Acquir Immune Defic Syndr. 2000; 25: 221-228Crossref PubMed Scopus (213) Google Scholar, 9Gifford AL, Bormann JE, Shively MJ, et al. Effects of group HIV patient education on adherence to antiretrovirals. A randomized controlled trial. Presented at the Eighth CROI, Chicago, Illinois, 2001.Google Scholar).One explanation for the limited success of psychoeducational interventions is that they require a higher level of internal motivation on the part of the patient. A leading theory postulates that behavior change requires “motivational readiness” on the part of the patient (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar); in its absence, nonadherence to the adherence intervention is the rule. Using daily structure and financial incentives, directly observed therapy interventions place less reliance on a patient’s internal motivation to change behavior. As such, directly observed therapy programs are a strategic public health tool: the treatment of contagious tuberculosis cannot wait for the patient to become ready to adhere.While effective, directly observed therapy has limitations. Traditional programs do not promote patient-level commitment to the long-term skills required to sustain medication adherence. One study of modified directly observed therapy for HIV therapy, in which the first dose of the drug regimen was delivered 5 days a week, found that adherence dropped precipitously during the 2 nonobserved days (11Sorensen J.L. Mascovich A. Wall T.L. et al.Medication adherence strategies for drug abusers with HIV/AIDS.AIDS Care. 1998; 10: 297-312Crossref PubMed Scopus (86) Google Scholar). The lack of an exit strategy is a core problem for applying directly observed therapy to chronic diseases such as HIV. Psychoeducational interventions may have the advantage of promoting adherence skills that extend beyond the intervention. In practice, however, durable changes are uncommon (8Tuldra A. Fumaz C.R. Ferrer M.J. et al.Prospective randomized two-arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy.J Acquir Immune Defic Syndr. 2000; 25: 221-228Crossref PubMed Scopus (213) Google Scholar, 9Gifford AL, Bormann JE, Shively MJ, et al. Effects of group HIV patient education on adherence to antiretrovirals. A randomized controlled trial. Presented at the Eighth CROI, Chicago, Illinois, 2001.Google Scholar). Is indefinite adherence “life support” using directly observed therapy reasonable for patients who do not become adherent on their own? Psychoeducational interventions, including education and counseling, medication reminders, self-monitoring tools, and support groups, can improve adherence to therapy (5Haynes R.B. McKibbon K.A. Kanani R. Systematic review of randomised trials of interventions to assist patients to follow prescriptions for medications.Lancet. 1996; 348 ([published erratum appears in Lancet 1997;349:1180]): 383-386Abstract Full Text Full Text PDF PubMed Scopus (534) Google Scholar, 6Roter D.L. Hall J.A. Merisca R. et al.Effectiveness of interventions to improve patient compliance a meta-analysis.Med Care. 1998; 36: 1138-1161Crossref PubMed Scopus (814) Google Scholar), but their efficacy is modest compared with directly observed therapy. In the study by Chaisson, 77% of patients assigned to the directly observed therapy arm received 100% of their doses, compared with the 6% to 10% of patients in the other two groups. These results are consistent with preliminary findings among incarcerated patients with HIV infection, among whom directly observed therapy led to unprecedented levels of viral suppression (7Fischl M, Rodriguez A, Scerpella E, et al. Impact of directly observed therapy on outcomes in HIV clinical trials. Presented at the Seventh CROI, San Francisco, California, 2000.Google Scholar) while the effect of psychoeducational interventions has been more modest (8Tuldra A. Fumaz C.R. Ferrer M.J. et al.Prospective randomized two-arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy.J Acquir Immune Defic Syndr. 2000; 25: 221-228Crossref PubMed Scopus (213) Google Scholar, 9Gifford AL, Bormann JE, Shively MJ, et al. Effects of group HIV patient education on adherence to antiretrovirals. A randomized controlled trial. Presented at the Eighth CROI, Chicago, Illinois, 2001.Google Scholar). One explanation for the limited success of psychoeducational interventions is that they require a higher level of internal motivation on the part of the patient. A leading theory postulates that behavior change requires “motivational readiness” on the part of the patient (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar); in its absence, nonadherence to the adherence intervention is the rule. Using daily structure and financial incentives, directly observed therapy interventions place less reliance on a patient’s internal motivation to change behavior. As such, directly observed therapy programs are a strategic public health tool: the treatment of contagious tuberculosis cannot wait for the patient to become ready to adhere. While effective, directly observed therapy has limitations. Traditional programs do not promote patient-level commitment to the long-term skills required to sustain medication adherence. One study of modified directly observed therapy for HIV therapy, in which the first dose of the drug regimen was delivered 5 days a week, found that adherence dropped precipitously during the 2 nonobserved days (11Sorensen J.L. Mascovich A. Wall T.L. et al.Medication adherence strategies for drug abusers with HIV/AIDS.AIDS Care. 1998; 10: 297-312Crossref PubMed Scopus (86) Google Scholar). The lack of an exit strategy is a core problem for applying directly observed therapy to chronic diseases such as HIV. Psychoeducational interventions may have the advantage of promoting adherence skills that extend beyond the intervention. In practice, however, durable changes are uncommon (8Tuldra A. Fumaz C.R. Ferrer M.J. et al.Prospective randomized two-arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy.J Acquir Immune Defic Syndr. 2000; 25: 221-228Crossref PubMed Scopus (213) Google Scholar, 9Gifford AL, Bormann JE, Shively MJ, et al. Effects of group HIV patient education on adherence to antiretrovirals. A randomized controlled trial. Presented at the Eighth CROI, Chicago, Illinois, 2001.Google Scholar). Is indefinite adherence “life support” using directly observed therapy reasonable for patients who do not become adherent on their own? What is the role of incentives, especially cash, in facilitating adherence?Studies looking at incentives, including cash, have shown excellent results in facilitating directly observed therapy (12Giuffrida A. Torgerson D.J. Should we pay the patient? Review of financial incentives to enhance patient compliance.BMJ. 1997; 315: 703-707Crossref PubMed Scopus (267) Google Scholar). Financial incentives retain participants in care and are associated with reduced costs of outreach (13Tulsky J.P. White M.C. Considerations on the road to involuntary confinement.West J Med. 1999; 171: 48-49PubMed Google Scholar, 14Buchanan R.J. Compliance with tuberculosis drug regimens incentives and enablers offered by public health departments.Am J Public Health. 1997; 87: 2014-2017Crossref PubMed Scopus (19) Google Scholar, 15Gourevitch M.N. Alcabes P. Wasserman W.C. Arno P.S. Cost-effectiveness of directly observed chemoprophylaxis of tuberculosis among drug users at high risk for tuberculosis.Int J Tuberc Lung Dis. 1998; 2: 531-540PubMed Google Scholar). While incentives have gained acceptance for time-limited interventions, incentives for prolonged therapies have been more problematic. Financial incentives also approach the fine line between encouragement and coercion, particularly for low-income or vulnerable patients, and they may discourage the disclosure of toxicity to prevent the interruption of scheduled reimbursement. Lastly, some low-income patients may claim poor adherence to become eligible for incentives provided in these programs. Thus, for some patients, incentive-based programs may unintentionally promote poor adherence. Studies looking at incentives, including cash, have shown excellent results in facilitating directly observed therapy (12Giuffrida A. Torgerson D.J. Should we pay the patient? Review of financial incentives to enhance patient compliance.BMJ. 1997; 315: 703-707Crossref PubMed Scopus (267) Google Scholar). Financial incentives retain participants in care and are associated with reduced costs of outreach (13Tulsky J.P. White M.C. Considerations on the road to involuntary confinement.West J Med. 1999; 171: 48-49PubMed Google Scholar, 14Buchanan R.J. Compliance with tuberculosis drug regimens incentives and enablers offered by public health departments.Am J Public Health. 1997; 87: 2014-2017Crossref PubMed Scopus (19) Google Scholar, 15Gourevitch M.N. Alcabes P. Wasserman W.C. Arno P.S. Cost-effectiveness of directly observed chemoprophylaxis of tuberculosis among drug users at high risk for tuberculosis.Int J Tuberc Lung Dis. 1998; 2: 531-540PubMed Google Scholar). While incentives have gained acceptance for time-limited interventions, incentives for prolonged therapies have been more problematic. Financial incentives also approach the fine line between encouragement and coercion, particularly for low-income or vulnerable patients, and they may discourage the disclosure of toxicity to prevent the interruption of scheduled reimbursement. Lastly, some low-income patients may claim poor adherence to become eligible for incentives provided in these programs. Thus, for some patients, incentive-based programs may unintentionally promote poor adherence. Expanding directly observed therapy to HIV antiretroviral therapyThere is considerable overlap among groups of patients affected by tuberculosis and by HIV, including drug use, mental illness, and unstable housing. Should we start building incentive-based directly observed programs for HIV antiretroviral therapy? We suggest caution in extrapolating lessons learned in tuberculosis treatment and prevention to HIV. Although once-daily HIV antiretroviral therapy will make directly observed treatment more feasible in the future, it is unlikely to resemble therapy for tuberculosis, which can be administered two times weekly after an initial 2-week induction. More importantly, therapy for tuberculosis is time limited, whereas HIV therapy is life long. Because tuberculosis is transmitted through casual contact and treatment renders patients noninfectious quickly, there is a greater public health priority to deliver directly observed therapy for tuberculosis than for HIV infection, although similar arguments have been made that improved adherence and viral load suppression may not only delay disease progression, but also reduce HIV transmission (16Bangsberg DR, Perry S, Charlebois E, et al. Adherence to antiretroviral therapy predicts progression to AIDS. AIDS. 2001. In press.Google Scholar, 17Quinn T.C. Wawer M.J. Sewankambo N. et al.Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group.N Engl J Med. 2000; 342: 921-929Crossref PubMed Scopus (2477) Google Scholar). There is considerable overlap among groups of patients affected by tuberculosis and by HIV, including drug use, mental illness, and unstable housing. Should we start building incentive-based directly observed programs for HIV antiretroviral therapy? We suggest caution in extrapolating lessons learned in tuberculosis treatment and prevention to HIV. Although once-daily HIV antiretroviral therapy will make directly observed treatment more feasible in the future, it is unlikely to resemble therapy for tuberculosis, which can be administered two times weekly after an initial 2-week induction. More importantly, therapy for tuberculosis is time limited, whereas HIV therapy is life long. Because tuberculosis is transmitted through casual contact and treatment renders patients noninfectious quickly, there is a greater public health priority to deliver directly observed therapy for tuberculosis than for HIV infection, although similar arguments have been made that improved adherence and viral load suppression may not only delay disease progression, but also reduce HIV transmission (16Bangsberg DR, Perry S, Charlebois E, et al. Adherence to antiretroviral therapy predicts progression to AIDS. AIDS. 2001. In press.Google Scholar, 17Quinn T.C. Wawer M.J. Sewankambo N. et al.Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group.N Engl J Med. 2000; 342: 921-929Crossref PubMed Scopus (2477) Google Scholar). Questioning the effects of directly observed therapy on limiting HIV drug resistanceThe public health imperative to reduce transmission of multidrug-resistant tuberculosis generated much of the enthusiasm for the use of directly observed therapy, and the transmission of drug-resistant HIV is cited as an analogous concern. Yet, there is little empiric support for the hypothesis that improved adherence to HIV medications will prevent HIV drug resistance. In contrast, good adherence may be necessary for the selection of drug-resistant virus (18Bangsberg D.R. Hecht F.M. Charlebois E.D. et al.Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population.AIDS. 2000; 14: 357-366Crossref PubMed Scopus (874) Google Scholar, 19Walsh J, Hertogs K, Gazzard B. Viral resistance, adherence and pharmokinetic indices in patients on successful and failing PI based HAART. Presented at the 40th ICAAC, Toronto, Ontario, 2000.Google Scholar, 20Campo R. Suarez G. Miller N. et al.Efficacy of indinavir (IDV)/ritonavir (RTV)-based regimens (IRBR) among patients with prior protease inhibitor failure.Antiviral Ther. 2000; 5 (Abstract.): 23PubMed Google Scholar), possibly because resistant virus has decreased fitness and requires the sufficient drug pressure that is associated with good adherence (21Deeks S.G. Wrin T. Liegler T. et al.Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIV-infected patients with detectable viremia.N Engl J Med. 2001; 344: 472-480Crossref PubMed Scopus (599) Google Scholar). Thus, it is conceivable that adherence interventions that fall short of complete viral suppression may increase the development of drug resistance among poorly adherent patients. These considerations, however, should not impede efforts to improve adherence, delay disease progression, and improve survival: resistance may be a necessary price for these improved outcomes. The public health imperative to reduce transmission of multidrug-resistant tuberculosis generated much of the enthusiasm for the use of directly observed therapy, and the transmission of drug-resistant HIV is cited as an analogous concern. Yet, there is little empiric support for the hypothesis that improved adherence to HIV medications will prevent HIV drug resistance. In contrast, good adherence may be necessary for the selection of drug-resistant virus (18Bangsberg D.R. Hecht F.M. Charlebois E.D. et al.Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population.AIDS. 2000; 14: 357-366Crossref PubMed Scopus (874) Google Scholar, 19Walsh J, Hertogs K, Gazzard B. Viral resistance, adherence and pharmokinetic indices in patients on successful and failing PI based HAART. Presented at the 40th ICAAC, Toronto, Ontario, 2000.Google Scholar, 20Campo R. Suarez G. Miller N. et al.Efficacy of indinavir (IDV)/ritonavir (RTV)-based regimens (IRBR) among patients with prior protease inhibitor failure.Antiviral Ther. 2000; 5 (Abstract.): 23PubMed Google Scholar), possibly because resistant virus has decreased fitness and requires the sufficient drug pressure that is associated with good adherence (21Deeks S.G. Wrin T. Liegler T. et al.Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIV-infected patients with detectable viremia.N Engl J Med. 2001; 344: 472-480Crossref PubMed Scopus (599) Google Scholar). Thus, it is conceivable that adherence interventions that fall short of complete viral suppression may increase the development of drug resistance among poorly adherent patients. These considerations, however, should not impede efforts to improve adherence, delay disease progression, and improve survival: resistance may be a necessary price for these improved outcomes. The problem of selecting candidates for directly observed HIV therapyThe expectation that injection drug users, the homeless, or the mentally ill would be the only candidates for HIV directly observed therapy is not based on evidence that nonadherence is limited to these groups of patients. For example, recognizing that nonadherence extends beyond the poor, the New York City Department of Health recommends that all patients with active tuberculosis should receive directly observed therapy. In San Francisco, clinic staff prescribes directly observed therapy for all patients who are smear-positive for tuberculosis. While enforced attendance at infectious disease clinics to receive antituberculosis therapy provokes little outcry, a similar approach to HIV therapy would be politically unacceptable and would be perceived as burdensome or stigmatizing by many patients.We suggest that if directly observed therapy is proven effective outside of institutional settings, selection strategies should be based on the patient’s stage of disease and motivational commitment to medication adherence (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar). Objective measures of motivational readiness can predict success with tobacco cessation and adherence to exercise programs and hypertensive medication as well as HIV therapy (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar, 22Prochaska J.O. Velicer W.F. DiClemente C.C. Fava J. Measuring processes of change applications to the cessation of smoking.J Consult Clin Psychol. 1988; 56: 520-528Crossref PubMed Scopus (795) Google Scholar, 23Willey C. Redding C. Stafford J. et al.Stages of change for adherence with medication regimens for chronic disease development and validation of a measure.Clin Ther. 2000; 22: 858-871Abstract Full Text PDF PubMed Scopus (109) Google Scholar). For patients with a high motivational state, psychoeducational approaches should be attempted first, followed, if necessary, by directly observed therapy. Patients with a low motivational state and late stage-disease are the ideal candidates for directly observed therapy, which should be recommended regardless of socioeconomic status. Ultimately, directly observed therapy should not be required to receive antiretroviral therapy, which should not be postponed indefinitely if it or other forms of adherence support is declined (24Bangsberg D.R. Moss A. When should we delay highly active antiretroviral therapy?.J Gen Intern Med. 1999; 14: 446-448Crossref PubMed Scopus (40) Google Scholar). The expectation that injection drug users, the homeless, or the mentally ill would be the only candidates for HIV directly observed therapy is not based on evidence that nonadherence is limited to these groups of patients. For example, recognizing that nonadherence extends beyond the poor, the New York City Department of Health recommends that all patients with active tuberculosis should receive directly observed therapy. In San Francisco, clinic staff prescribes directly observed therapy for all patients who are smear-positive for tuberculosis. While enforced attendance at infectious disease clinics to receive antituberculosis therapy provokes little outcry, a similar approach to HIV therapy would be politically unacceptable and would be perceived as burdensome or stigmatizing by many patients. We suggest that if directly observed therapy is proven effective outside of institutional settings, selection strategies should be based on the patient’s stage of disease and motivational commitment to medication adherence (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar). Objective measures of motivational readiness can predict success with tobacco cessation and adherence to exercise programs and hypertensive medication as well as HIV therapy (10Prochaska J.O. Velicer W.F. Rossi J.S. et al.Stages of change and decisional balance for 12 problem behaviors.Health Psychol. 1994; 13: 39-46Crossref PubMed Scopus (1872) Google Scholar, 22Prochaska J.O. Velicer W.F. DiClemente C.C. Fava J. Measuring processes of change applications to the cessation of smoking.J Consult Clin Psychol. 1988; 56: 520-528Crossref PubMed Scopus (795) Google Scholar, 23Willey C. Redding C. Stafford J. et al.Stages of change for adherence with medication regimens for chronic disease development and validation of a measure.Clin Ther. 2000; 22: 858-871Abstract Full Text PDF PubMed Scopus (109) Google Scholar). For patients with a high motivational state, psychoeducational approaches should be attempted first, followed, if necessary, by directly observed therapy. Patients with a low motivational state and late stage-disease are the ideal candidates for directly observed therapy, which should be recommended regardless of socioeconomic status. Ultimately, directly observed therapy should not be required to receive antiretroviral therapy, which should not be postponed indefinitely if it or other forms of adherence support is declined (24Bangsberg D.R. Moss A. When should we delay highly active antiretroviral therapy?.J Gen Intern Med. 1999; 14: 446-448Crossref PubMed Scopus (40) Google Scholar). Recommendations and summaryDirectly observed therapy has contributed to the treatment and prevention of tuberculosis. Before implementing such programs for HIV care, we recommend a vigorous effort to identify effective strategies to improve treatment adherence and to fill the gap between pharmacotherapeutic studies and durable viral suppression in patients with HIV infection. Studies should compare the effectiveness of directly observed therapy and psychoeducational interventions. Effective exit strategies should be developed to extend the durability of time-limited programs, recognizing that some patients may need indefinite support. Rational and nonprejudicial selection strategies are needed to identify the best candidates for directly observed therapy. Ultimately, the value of directly observed therapy should be compared with other approaches to adherence support to measure their effects on HIV progression and even transmission. Directly observed therapy has contributed to the treatment and prevention of tuberculosis. Before implementing such programs for HIV care, we recommend a vigorous effort to identify effective strategies to improve treatment adherence and to fill the gap between pharmacotherapeutic studies and durable viral suppression in patients with HIV infection. Studies should compare the effectiveness of directly observed therapy and psychoeducational interventions. Effective exit strategies should be developed to extend the durability of time-limited programs, recognizing that some patients may need indefinite support. Rational and nonprejudicial selection strategies are needed to identify the best candidates for directly observed therapy. Ultimately, the value of directly observed therapy should be compared with other approaches to adherence support to measure their effects on HIV progression and even transmission.