Targeting Candidates for Directly Administered Highly Active Antiretroviral Therapy: Benefits Observed in HIV-Infected Injecting Drug Users in Residential Drug-Rehabilitation Facilities

Abstract

The purpose of this study was to evaluate retrospectively the potential benefits of directly administered antiretroviral therapy (DAART) in HIV-infected former injecting drug users (ex-IDUs) admitted to residential drug rehabilitation facilities. We compared 106 of these patients consecutively admitted in 12 communities where DAART was administered (DAART group) to two matched control groups of ex-IDUs undergoing self-administered ART: 106 subjects in other 10 communities (SAT group) and 106 outpatients at hospital infectious-disease wards where community patients were referred after discharge (OUT group). We estimated the proportion of patients with high adherence and the hazard ratio (HR) of 20% or more increase in the CD4⁺ cell count and of reaching an undetectable viral load. The proportion of patients with high adherence to treatment was highest in the DAART group. The probability of 20% or more increase in the CD4⁺ cell count was significantly lower in the two control groups versus the DAART group (SAT group HR=0.32; OUT group HR=0.43). The HR of observing an undetectable HIV-RNA level versus DAART was significantly lower in the OUT group (HR: 0.71; 95% confidence interval [CI]: 0.52–0.97) but did not reach statistical significance for the SAT group (HR: 0.99; 95% CI: 0.74–1.33). Our findings after a 24-month follow-up, suggest that DAART in HIV-infected patients of drug-rehabilitation communities improves adherence, immunologic, and virologic outcome toward free outpatients. Even if our retrospective 36-month data do not show a prolonged viral suppression in these patients, DAART may be considered a valuable therapeutic and educational strategy in this particular target group.

Introduction

T he use of highly active antiretroviral therapy (HAART) has resulted in an impressive reduction in morbidity and mortality among people living with HIV/AIDS.¹ Considering the lifelong duration of HAART, the main challenge remains ensuring proper adherence to therapy in the long term. Although the previously accepted level of adherence of 90–95% might not be required with newly developed protease inhibitor (PI)-boosted and non-nucleoside reverse transcription inhibitor (NNRTI)-based regimens,^2,3 long-term adherence to therapy remains key to ensuring optimal individual outcomes.^4,5 In addition, better individual disease control might provide community benefits through some degree of containment of virus circulation and resistance.

To address adherence issues, directly administered antiretroviral therapy (DAART) has been proposed,^6,7 yet questions have been raised as to its effectiveness and economic sustainability for HIV care.⁸ Previous studies have shown mixed results and this might be due to different populations, settings, and methodologies.^9

–12 A recent meta-analysis of randomized clinical trials showed no benefit of DAART over self-administered therapy (SAT) with regard to virologic outcome.¹³ However, when broadening the analysis to include controlled studies published in peer-reviewed journals, a positive effect of DAART was observed on virologic, immunologic, and adherence outcomes.¹⁴

Since contextual factors could influence significantly the conclusions of the different studies previously conducted, some efforts have been made to identify specific target groups more likely to benefit from an enhanced adherence strategy, such as injecting drug users (IDUs) and homeless people. IDUs generally have less favorable therapy outcome and although full agreement between studies has not been achieved, a number of observational^12,15 and randomized controlled trials¹⁶ have shown better virologic and/or immunologic outcomes in DAART over SAT IDUs patients. Furthemore, additional virologic benefits were observed when onsite medical and case management services were provided to DAART IDUs recipients.¹⁷ Nevertheless, considering that directly administered therapy cannot be recommended lifelong, follow-up interventions are required to sustain any benefit reached with DAART, which needs to be integrated in a comprehensive treatment scheme in the long term to avoid a rapid waning effect, as already observed in other studies.¹⁸ In addition, the coercive nature of the intervention might not be applicable to all settings and this further emphasizes the need to characterize target groups and acceptable delivery modalities of DAART.

Overall, DAART has triggered great interest as a potential tool in the management of patients with HIV, however, questions remain unanswered as to the acceptability of this intervention and its long-term effects in terms of retention, attendance, and participant's satisfaction.¹⁹

The objective of the present study was to evaluate the effectiveness of DAART among HIV-infected ex-IDUs in residential drug-rehabilitation communities, conditions that, to the best of our knowledge, have not been previously explored. To this end, adherence, immunologic, and virologic responses to DAART were evaluated in these subjects compared to IDUs undergoing SAT either in the same setting or in outpatient HIV units.

Materials and Methods

Study population

We retrospectively enrolled 318 HIV-infected ex-IDUs receiving antiretroviral drugs. One hundred six of these patients had been consecutively admitted to 12 residential drug-rehabilitation communities, in which antiretroviral drugs are always administered by nonmedical community staff who observe the patient taking every dose and receiving therapies for the entire period (referred to as the DAART group). The communities were located throughout Italy and constituted a convenience sample.

Often ex-IDUs show weak character profiles that can contribute to low levels of adherence; we wanted to study if their adherence, and the subsequent therapeutic effectiveness during SAT, changed among those patients still forced to a residential control and those who live in an outside community and are totally free-living. For this reason, we considered two comparison groups matched for age and gender with the DAART group.

One group consisted of 106 consecutive ex-IDUs undergoing self-administered HAART, who were admitted in the same period to 10 other communities in which staff left all drugs to the patients once a week with no directly observed control (referred to as the SAT group). The choice of therapeutic strategy (DAART or SAT) has been based on the place where the patient casually went to reside.

The second comparison group consisted of 106 ex-IDUs who consecutively started self-administered HAART in the same period at 7 infectious-disease outpatient units where community patients are followed-up after discharge (referred to as the OUT group). With the latter group, we compared the adherence levels and the outcomes in outpatient versus community care. Each therapeutic group (DAART, SAT, OUT) comprised individuals who spontaneously presented themselves to either outpatient units or residential communities based on personal preference. None of the subjects enrolled were using intravenous drugs or receiving methadone or buprenorphine during the study period. Due to the specific setting of residential drug-rehabilitation communities, it is also expected that this subpopulation is free from any significant psychiatric diseases.

Methodology

DAART was administered by nonmedical community staff, whereas SAT subjects were provided with their packaged pills once a week along with a written detailed schedule of administration. The individuals in the OUT group picked up their pills at the outpatient department at the time of their quarterly visit. Visits in all settings were opportunities for patient counseling.

CD4⁺ cell count and HIV plasma viral load were routinely recorded at study entry and at 3, 6, 12, 18, 24, 30, and 36 months. In all participating institutions, CD4+ cell count was measured by BD FACScan™ II flow cytometer (Becton Dickinson Immunocytometry Systems, Erembodegem, Belgium]. Plasma HIV load was measured by Nuclisens Easy Q HIV-1 v1.2 assay (bioMerieux bv, Boxtel, The Netherlands; detection limit, 50 copies per milliliter).

Adherence was assessed by the infectious disease specialists who interviewed participants at regular time intervals. Adherence was unblinded and considered as high if more than 95% of the prescribed pills were taken during the observed period (at the various time points). A semiquantitative adherence scale was defined as follows: a score of 1 was recorded for subjects with over 95% adherence, while those with 80–95% adherence had a score of 2, those with 50–80% adherence a score of 3 and those with adherence less than 50% a score of 4.

Patients who started HAART for the first time were defined as naïve; those who had already undergone some form of antiretroviral therapy before the study were defined as experienced.

Statistical analysis

The baseline characteristics of participants in the DAART group were compared with each of the two groups (SAT versus DAART and OUT versus DAART) with the use of Mann-Whitney test for continuous variables and χ² test for categorical data.

A Cox model was applied to estimate the hazard ratio (HR) of an increment in the CD4⁺ cell count of at least 20% with respect to baseline, in the SAT and OUT groups versus the DAART group. Similarly, a Cox model was applied to estimate the HR of observing an undetectable viral load. The following variables were evaluated as possible confounders (variables available at baseline and possibly associated with both treatment approaches and response to therapy): AIDS diagnosis before enrolment in the study, CD4 and viral load at baseline, type of HAART regimen, ART dose frequency, and being HAART naïve versus antiretroviral therapy-experienced. Only variables that resulted associated with the time to the event (with a p value<0.10) in the univariable models were thereafter considered in the multiple analyses. Given that data were matched for gender and age, the analyses were repeated applying Cox models stratified by gender and age. In particular, time to event or censoring was set to a constant value of 36 months and Breslow methods are used to account for tied survival times.²⁰

A logistic model was applied to evaluate the adherence in the SAT and OUT group compared to the DAART group at different times: 6, 12, 24, and 36 months. In particular the odds ratio (OR) of observing a high adherence (≥95%) was estimated for SAT and OUT versus DAART group.

Results

The baseline characteristics of the 318 ex-IDUs enrolled in the study are shown in Table 1. The two control groups (SAT or OUT) did not significantly differ with DAART group in terms of age, gender, viral load, antiretroviral experience, or therapeutic regimen, while a significant higher proportion of participants in the DAART group (versus both comparison groups) had been previously diagnosed with AIDS. A significantly higher CD4 level at baseline was reported in the SAT group.

Table 1.

Baseline Characteristics of Participants in the DAART Group and the Two Comparison Groups

Characteristics	DAART n=106	SAT n=106	OUT n=106	p Values ^a (SAT vs. DAART)	p Values ^a (OUT vs. DAART)
Age, median (IQR)	38 (33–42)	38 (34–42)	39 (35–42)	0.768	0.335
Male (n, %)	82 (77.4)	87 (82.1)	89 (84.0)	0.393	0.223
CD4 cell count, median cells/mL (IQR)	203 (87–296)	239 (157–380)	215 (117–322)	0.021	0.648
HIV-RNA level, median log₁₀ (IQR)	4.9 (4.2–5.7)	4.9 (4.3–5.4)	5.0 (4.4–5.5)	0.553	0.861
Previous AIDS diagnosis (n, %)	34 (32.0)	19 (18.0)	19 (18.0)	0.017	0.017
Antiretroviral experienced (n, %)	81 (76.4)	76 (71.7)	76 (71.7)	0.433	0.433
Type of HAART regimens				0.461	0.151
2 NRTIs+IPI boosted (n, %)	13 (12.3)	17 (16.0)	6 (5.7)
2 NRTIs+1 PI unboosted	67 (63.2)	61 (57.7)	80 (75.4)
2 NRTI+NNRTI	18 (16.9)	24 (22.6)	13 (12.3)
NRTI+NNRTI+PI	4 (3.8)	1 (0.9)	6 (5.7)
3 NNRTIs	4 (3.8)	3 (2.8)	1 (0.9)
ART dose frequency:				0.888	0.890
Twice-daily regimen (n, %)	64 (60.38)	65 (60.95)	63 (59.43)
Three-daily regimen (n, %)	42 (39.62)	41 (39.05)	43 (40.57)

p Values by Mann-Whitney test for continuous variables and by χ² test for categorical variables.

DAART; patients housed in community and treated with a directly-observed therapy approach; SAT; patients housed in community and treated with a self-administered therapy approach; OUT; patients followed as outpatient and treated with a a self-administered therapy approach; IQR, interquartile range; NRTIs, nucleoside reverse transcription inhibitor; PI, protease inhibitor; NNRTI, non-nucleoside reverse transcription inhibitor; ART, antiretroviral therapy.

Nadir CD4⁺ cell counts were comparable between groups for naive patients (DAART n=25, SAT n=30, OUT n=27). The median follow-up was 24 months (interquartile range [IQR]: 12–30; range, 3–36).

The proportion of patients with high level (>95%) of adherence to ART, as measured at the four follow-up visits, ranged from 87% to 90% in the DAART group, 56–65% in the SAT group, and 38–53% in the OUT group. The OR of observing high adherence (>95%) for SAT and OUT versus DAART group are shown in Table 2.

Table 2.

Odds Ratio of Observing a High Adherence (>95%) of SAT and OUT Groups Versus DAART at Different Times: 6, 12, 24, and 36 Months

	OR of high adherence at different times
	6 months		12 months		24 months		36 months
	OR	(95% CI)	OR	(95% CI)	OR	(95% CI)	OR	(95% CI)
SAT group versus DAART group	0.22	(0.12–0.39)	0.30	(0.17–0.54)	0.35	(0.20–0.62)	0.65	(0.34–1.27)
OUT group versus DAART group	0.19	(0.10–0.34)	0.19	(0.11–0.35)	0.39	(0.22–0.68)	1.70	(0.94–3.07)

SAT, self-administered treatment approach; OUT, outpatients; DAART, directly administered antiretroviral therapy; OR, odds ratio; CI, confidence interval.

Significantly lower adherence was observed in the SAT and OUT groups compared to the DAART group at 6, 12, and 24 months from the start of HAART, but not at the 36-month time point (Table 2).

The proportion of subjects with an increase of at least 20% in the CD4⁺ cell count over the entire follow-up period is shown in Table 3. A significantly lower proportion of subjects in the comparison groups (SAT and OUT) presented with an increase of at least 20% in the CD4⁺ cell count with respect to baseline, compared to the DAART group (crude HR 0.34 for the SAT and 0.44 for the OUT group). HAART-naïve patients were significantly more likely to have an increase in the CD4⁺ cell count compared to HAART-experienced patients. After adjustment, RH tended to decrease but remained statistically significant (Table 3).

Table 3.

Hazard Ratios of Immunologic and Virologic Response of SAT and OUT Groups Versus DAART

Variable	cHR	95% CI	aHR	95% CI
CD4+ cell increase > 20% respect to baseline (N∘events=45)
SAT group versus DAART group	0.34	0.15–0.77	0.32	0.14–0.73
OUT group versus DAART group	0.44	0.23–0.87	0.43	0.22–0.83
Naïve versus experienced	1.99	1.08–3.68	2.14	1.16–3.95
Undetectable HIV-RNA level (<50 copies/mL) (n∘events=258)
SAT group versus DAART group	0.99	0.74–1.33	0.98	0.73–1.32
OUT group versus DAART group	0.71	0.52–0.97	0.70	0.52–0.96
Naïve versus experienced	1.17	0.90–1.55	1.20	0.90–1.55

DAART, patients housed in community and treated with a directly-observed therapy approach; SAT, patients housed in community and treated with a self-administered therapy approach; OUT, patients followed as outpatient and treated with a self-administered therapy approach; cHR, crude hazard ratio; CI, confidence interval; aHR, adjusted hazard ratio.

The likelihood of reaching undetectable HIV-RNA level was lower in the OUT and SAT groups compared to DAART, although the difference was not statistically significant for the SAT group (Table 3).

No association was found between both end points (immunologic and virologic) and the other potential confounders (i.e., AIDS diagnosis before enrolment in the study, CD4 and viral load at baseline, type of HAART regimen, ART dose frequency). Furthermore, considering that data were matched for gender and age, analyses have been repeated applying stratified, by gender and age Cox models, obtaining similar results (results not shown).

For both immunologic and virologic end points, Cox models were repeated considering naïve (n=85) and experienced individuals (n=232), showing similar magnitude of HR compared to that shown in Table 2 (in particular we observed lower risks in naïve subjects, results not shown). Likewise, the analysis was restricted to subjects with high adherence during the entire study period (n=182, 57% of the entire population), showing 50% lower likelihood of immunologic response in both the OUT and the SAT group compared to the DAART group, whereas the virologic response did not differ among the three groups.

Discussion

Our study showed higher adherence to therapy in community patients either in the DAART and the SAT group compared to the OUT group, after a 24-month follow-up, suggesting that being in a residential community has a positive effect per se. DAART patients had greater adherence than any other group, suggesting additional effect of directly administered therapy even in a favorable environment such as residential communities.

The DAART group also showed a better immunologic response than the two comparison groups. When limiting the analysis to individuals with high adherence during the entire study period, the proportion of subjects with good immunologic response was still higher in the DAART group than in the other two groups. Considering self-reported adherence in the SAT and OUT groups, this observation suggests possible complacent responses to health care workers conducting the interview, as already shown in other studies,²¹ leading to overestimated adherence figures for these latter patients whereas compliance assessment was more accurate in DAART patients.

With regard to the virologic response, the DAART group had a significantly better response than the OUT group, but not when compared to the SAT group, although a slightly favorable trend to DAART versus SAT seemed to be present. This finding can be related to a more “protected” setting such as residential communities where adherence is expected to be higher than in an outpatient setting.⁴

Overall, our study suggests that DAART administered in ex-IDUs subjects living in residential rehabilitation communities, might be a useful tool ensuring better adherence and disease outcomes over a median follow-up of 2 years. Our findings add further insight to the complex picture of studies conducted among IDUs in methadone clinics,²² AIDS welfare homes,²³ residential treatment facilities,²⁴ and prisons.^25
–27 One strength of the evaluation is the availability of a large sample of IDUs, who generally have poorer adherence and outcome, suggesting benefits of DAART over SAT particularly versus outpatient care.

Issues concerning the coercive nature of DAART and its economic sustainability have been raised by other authors.²⁸ In our study, no extra costs were incurred since DAART was administered by the regular community staff. With regard to acceptance, the peer to peer approach selected in our DAART program could be more easily accepted with potentially positive effects on overall outcomes. In addition, therapy administration by peer community members might represent a unique occasion for a specific counseling about the illness, the potential adverse events of anti-HIV drugs with their control, and the higher quality of life when the patient maintains high adherence levels after release, as already documented in prison setting.²⁹ The acquisition of these information can be key to maintaining higher levels of adherence after returning to a SAT regimen, when every patient returned to his or her external community. These situations could be assimilated to the problems already studied in HIV-positive inmates after their release.^30,31

The few articles about randomized controlled trials concerning the DAART efficacy have produced mixed results.^{11,13,16,32,33} In the most recent of these, conducted on naïve AIDS Clinical Trials Group (ACTG) patients, Gross et al.³³ conclude that in their cohort the potential benefit of modified DOT was marginal and not sustained after DOT was discontinued, suggesting that DOT should not be incorporated routinely for care of treatment-naïve HIV-1–infected patients, but also that their findings do not exclude the possible utility of DOT in populations enriched for nonadherence.

This last is the central point of DOT strategy for antiretroviral therapy. It is necessary to propose DAART to high-risk populations for not-assumption or low-adherence to anti-HIV schedules, focusing only on patients who are likely to derive benefit.³⁴ Long-term maintenance of improved outcomes remains one of the greatest issues in these special populations. In our study, median follow-up was 2 years, with a significant proportion of community patients lost to follow-up at 36 months, resulting in the loss of statistical significance between groups for the main variables studied at this time point. These data are not unexpected since in Italy, patients tend to stay for only 1 year in communities. In any case, DAART is not sustainable much beyond 1 or 2 years and other studies should be undertaken to identify proper follow-up measures in the long term.

Our study has several limitations. First of all, its retrospective nonrandomized design does not allow exclusion of selection bias, although it should be stated that the baseline demographic and clinical characteristics were very similar for the three groups. In addition, admission to communities tends to select a rather homogeneous population. Self-reported adherence by the SAT and OUT group might also be artificially increased with respect to directly observed adherence in the DAART group, thus maximizing the differences observed for the main evaluated outcomes. With regard to treatment management and strategies, bias resulting from different approaches are expected to be minimal, since antiretroviral management was defined according to common protocols in all groups. Overall, we assume that control groups are largely similar demographically and with regard to treatment approach, providing a fair picture of midterm outcomes in three different strategies for antiretroviral therapy in ex-IDUs in Italy.

In conclusion, we found that DAART strategy is associated with immunologic improvement and to an higher adherence to highly active ART as compared to self-administered therapy in ex-IDUs living in drug-rehabilitation communities. We also found a benefit in viral suppression compared to outpatients. This subpopulation of HIV patients is known to be less likely to show adherence and consequently, good immunologic and virologic outcomes. Better control of these patients can bring individual and overall benefits by contributing to the reduction of virus circulation and resistance. Improved outcomes were observed without additional costs, given that only resources already present in the communities were used. Our results are very promising considering that most patients had complicated treatment schedules, consisting of a large number of pills administered two or three times per day. The progressive availability and access to once-daily therapies may facilitate and simplify the application of DAART. In addition, directly administered therapy might have other benefits beyond the sole adherence to retroviral therapy with improved compliance to nonantiretroviral medications as well as enhanced community involvement and better daily living functioning.³⁵

Questions remain as to the transition from DAART to SAT and the maintenance of benefits observed in the long term. Overall, counseling is essential to reinforce a positive attitude toward adherence to ART and to create the conditions for keeping high levels of adherence among ex-IDUs who leave the therapeutic community. In further studies on DAART in low-adherence patients, it will be necessary to also evaluate long-term individual benefits, such as the maintenance of viral suppression after months from the transition to SAT.

Footnotes

Acknowledgments

The authors are grateful to the general staff of the drug-rehabilitation communities and the physicians in these communities: Gian Piero D'Offizi (Roma), Saverio Parisi (Verona), Walter Sarais (Macomer), Manuaela Heichen (Bari), Lucia Spanu (Sassari), and Paola Piano (Cagliari), and the infectious-disease specialists of the University of Sassari (Giovanna Calia, Carla Lovigu, and Marco Mannazzu).

Author Disclosure Statement

No competing financial interests exist.

References

Palella

, Delaney

, Moorman

et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med, 1998; 338:853–880.

Bangsberg

. Less than 95% adherence to nonnucleoside reverse-transcriptase inhibitor therapy can lead to viral suppression. Clin Infect Dis, 2006; 43:939–941.

Markowitz

, Slater

, Schwartz

et al. Long-term efficacy and safety of tipranavir boosted with ritonavir in HIV-1-infected patients failing multiple protease inhibitor regimens: 80-week data from a phase 2 study. J Acquire Immune Defic Syndrome, 2007; 45:401–410.

Paterson

, Swindells

, Mohr

et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med, 2000; 133:21–30.

Paris

, Lederberger

, Weber

et al. Incidence and predictors of virologic failure of antiretroviral triple-drug therapy in a community-based cohort. AIDS Res Hum Retroviruses, 1999; 151:1631–1638.

Woodward

. Should directly observed therapy be considered for treatment of HIV [letter]? JAMA, 1996; 276:1956.

Stenzel

, McKenzie

, Mitty

, Flanigan

. Enhancing adherence to HAART: a pilot program of modified directly observed therapy. AIDS Reader, 2001; 11:317–328.

Mitty

, Stone

, Sands

, Macalino

, Flanigan

. Directly observed therapy for the treatment of people with human immunodeficiency virus infection: A work in progress. Clin Infect Dis, 2002; 34:984–990.

Babudieri

, Aceti

, D'Offizi

, Carbonara

, Starnini

. Directly observed therapy to treat HIV infection in prisoners [letter] JAMA, 2000; 284:179–180.

10.

Wohl

, Stephenson

, Golin

et al. Adherence to directly observed antiretroviral therapy among human immunodeficiency virus-infected prison inmates. Clin Infect Dis, 2003; 36:1572–1576.

11.

Wohl

, Garland

, Valencia

et al. A randomized trial of Directly Administered Antiretroviral Therapy and adherence case management intervention. Clin Infect Dis, 2006; 42:1619–1627.

12.

, Brown

, Coleman

, Kibler

, Loewenthal

, Mitty

. The feasibility of modified directly observed therapy for HIV-seropositive African American substance users. AIDS Patient Care STDs, 2008; 22:139–146.

13.

Ford

, Nachega

, Engel

, Mills

. Directly observed antiretroviral therapy: A systematic review and meta-analysis of randomised clinical trials. Lancet, 2009; 374:2064–2071.

14.

Hart

, Jeon

, Ivers

et al. Effect of directly observed therapy for highly active antiretroviral therapy on virologic, immunologic, and adherence outcomes: A meta-analysis and systematic review. J Acquir Immune Defic Syndr, 2010; 54:167–179.

15.

Lucas

, Mullen

, Weidle

, Hader

, McCaul

, Moore

. Directly administered antiretroviral therapy in methadone clinics is associated with improved HIV treatment outcomes, compared with outcomes among concurrent comparison groups. Clin Infect Dis, 2006; 42:1628–1635.

16.

Altice

, Maru

, Bruce

, Springer

, Friedland

. Superiority of directly administered antiretroviral therapy over self-administered antiretroviral therapy among HIV-infected drug users: A prospective, randomized, controlled trial. Clin Infect Dis, 2007; 45:770–778.

17.

Smith-Rohrberg

, Mezger

, Walton

, Bruce

, Altice

. Impact of enhanced services on virologic outcomes in a directly administered antiretroviral therapy trial for HIV-infected drug users. J Acquire Immune Defic Syndrome, 2006; 43,Suppl 1:s48–53.

18.

Maru

, Bruce

, Walton

, Springer

, Altice

. Persistance of virological benefits following directly administered antiretroviral therapy among drug users: Results from a randomized controlled trial. J Acquir Immune Defic Syndr, 2009; 50:176–181.

19.

Garland

, Wohl

, Valencia

et al. The acceptability of a directly-administered antiretroviral therapy (DAART) intervention among patients in public HIV clinics in Los Angeles, California. AIDS Care, 2007; 19:159–167.

20.

Cummings

, McNight

, Greenland

. Matched cohort methods for injury research. Epidemiol Rev, 2003; 25:43–50.

21.

Arnsten

, Demas

, Farzadegan

et al. Antiretroviral therapy adherence and viral suppression in HIV-infected drug users: Comparison of self-report and electronic monitoring. Clin Infect Dis, 2001; 33:1417–1423.

22.

Conway

, Prasad

, Raynolds

et al. Directly observed therapy for the management of HIV-infected patients in a methadone program. Clin Infect Dis, 2004; 38,Suppl 5:S402–408.

23.

Tinoco

, Giron-Gonzalez

, Gonzalez-Gonzalez

et al. Efficacy of directly observed treatment of HIV infection: Experience in AIDS welfare homes. Eur J Clin Microbiol Infect Dis, 2004; 23:331–335.

24.

Hader

, Weidle

, Cergnul

et al.

Directly observed antiretroviral therapy (DART) in residential treatment facilities in New York City [abstract 476]

Program and Abstracts of the 8th Conference on Retroviruses and Opportunistic Infections (Chicago), Alexandria, VA. Foundation for Retrovirology and Human Health. 2001; 185

25.

Fischl

, Castro

, Mondroig

et al.

Impact of directly observed therapy on long-term outcomes in HIV clinical trials [abstract 528]

Program and Abstracts of the 8th Conference on Retroviruses and Opportunistic Infections (Chicago), Alexandria, VA. Foundation for Retrovirology and Human Health. 2001; 202.

26.

Soto Blanco

, Pérez

, March

. Adherence to antiretroviral therapy among HIV-infected prison inmates (Spain) Int J STD AIDS, 2005; 16:133–138.

27.

Babudieri

, Pintus

, Maida

, Starnini

, Rezza

. Does counseling increase sustained benefit of HAART among prison inmates after release to the community? [letter] Clin Infect Dis, 2005; 40:321–322.

28.

Sansom

, Anthony

, Garland

et al. The costs of HIV antiretroviral therapy adherence programs and impact on health care utilization. AIDS Patients Care STDs, 2008; 22:131–138.

29.

Springer

, Altice

. Improving the care for HIV-infected prisoners. An integrated prison release health model. Public Health behind Bars. New York: Springer, 2007; 535–555.

30.

Baillargeon

, Giordano

, Harzke

et al. Predictors of reicarcaration and disease progression among released HIV-infected inmates. AIDS Patients Care STDs, 2010; 24:389–394.

31.

Springer

, Pesanti

, Hodges

, Macura

, Doros

, Altice

. Effectiveness of antiretroviral therapy among HIV-infected prisoners: Reincarceration and the lack of sustained benefit after release to the community. Clin Infect Dis, 2004; 38:1754–1760.

32.

Macalino

, Hogan

, Mitty

et al. A randomized clinical trial of community-based directly observed therapy as an adherence intervention for HAART among substance users. AIDS, 2007; 21:1473–1477.

33.

Gross

, Tierney

, Andrade

et al. Modified directly observed antiretroviral therapy compared with self-administered therapy in treatment-naïve HIV-1 infected patients: A randomized trial. Arch Intern Med, 2009; 169:1224–1232.

34.

Altice

, Springer

. DAART for human immunodeficiency virus–infected patients: Studying subjects not at risk for nonadherence and use of untested interventions. Arch Intern Med, 2010; 170:109–110.

35.

Bradley-Ewing

, Thomson

, Pinkston

, Goggin

. A qualitative examination of the indirect effects of modified directly observed therapy on health behaviours other than adherence. AIDS Patient Care STDs, 2008; 22:663–668.