Interventions to Improve Adherence to Antiretroviral Therapy: a Rapid Systematic Review

Interventions to improve adherence to antiretroviral therapy: a rapid systematic review

Krisda H. Chaiyachatia, Osondu Ogbuojib, Matthew Priceb, Amitabh B. Sutharc, Eyerusalem K. Negussiec and Till Ba¨rnighausenb,d

Introduction: Access to antiretroviral treatment (ART) has substantially improved over the past decade. In this new era of HIV as a chronic disease, the continued success of ART will depend critically on sustained high ART adherence. The objective of this review was to systematically review interventions that can improve adherence to ART, including individual-level interventions and changes to the structure of ART delivery, to inform the evidence base for the 2013 WHO consolidated antiretroviral guidelines. Design: A rapid systematic review. Methods: We conducted a rapid systematic review of the global evidence on interventions to improve adherence to ART, utilizing pre-existing systematic reviews to identify relevant research evidence complemented by screening of databases for articles published over the past 2 years on evidence from randomized controlled trials (RCTs). We searched five databases for both systematic reviews and primary RCT studies (Cochrane Library, EMBASE, MEDLINE, Web of Science, and WHO Global Health Library); we additionally searched ClinicalTrials.gov for RCT studies. We examined intervention effectiveness by different study characteristics, in particular, the specific populations who received the intervention. Results: A total of 124 studies met our selection criteria. Eighty-six studies were RCTs. More than 20 studies have tested the effectiveness of each of the following interventions, either singly or in combination with other interventions: cognitive-behavioural interventions, education, treatment supporters, directly observed therapy, and active adherence reminder devices (such as mobile phone text messages). Although there is strong evidence that all five of these interventions can significantly increase ART adherence in some settings, each intervention has also been found not to produce significant effects in several studies. Almost half (55) of the 124 studies investigated the effectiveness of combination interventions. Combination interventions tended to have effects that were similar to those of single interventions. The evidence base on interventions in key populations was weak, with the exception of interventions for people who inject drugs. Conclusion: Tested and effective adherence-enhancing interventions should be increasingly moved into implementation in routine programme and care settings, accompanied by rigorous evaluation of implementation impact and performance. Major evidence gaps on adherence-enhancing interventions remain, in particular, on the cost-effectiveness of interventions in different settings, long-term effectiveness, and effectiveness of interventions in specific populations, such as pregnant and breastfeeding women. ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins

AIDS 2014, 28 (Suppl 2):S187–S204

Keywords: antiretroviral adherence, interventions, randomized controlled trials, systematic review aYale School of Medicine, New Haven, bDepartment of Global Health and Population, Harvard School of Public Health, Boston, USA, cHIV Department, World Health Organization, Geneva, Switzerland, and dWellcome Trust Africa Centre for Health and Population Science, University of KwaZulu-Natal, Mtubatuba, South Africa. Correspondence to Till Ba¨rnighausen, 665 Huntington Avenue, Boston 02115, Boston, MA. Tel: +1 617 379 0372; fax: +1 617 432 6733; e-mail: [email protected]

DOI:10.1097/QAD.0000000000000252

ISSN 0269-9370 Q 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins S187 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. S188 AIDS 2014, Vol 28 (Suppl 2)

Background adhere to ART. In addition to the contribution to the WHO 2013 consolidated guidelines, our review aims Antiretroviral treatment (ART) has converted a highly to provide a guide for ART programme managers, fatal HIV infection into a chronic condition that requires policy makers, and researchers to the portfolio of ART lifelong care [1]. Within the past decade, worldwide adherence-enhancing interventions for practice, policy access to ART has improved significantly, with almost and further study. 10 million people receiving ART by the end of 2012 [2]. In addition to its life-prolonging effects, ART can also reduce HIV transmission to uninfected people [3,4]. In this new era of HIV treatment, the continued Methods success of ART will depend on improving our under- standing of when to initiate therapy, creating continuity of General methodology of rapid systematic care, and ensuring high treatment adherence. Adherence reviews is the extent to which a person uses a medication We conducted a rapid systematic review of the global according to medical recommendations, inclusive of evidence on interventions to improve ART medication timing, dosing, and consistency. Arguably, adherence is adherence. Rapid systematic reviews differ from the most critical factor in ensuring ART success, because traditional systematic reviews in that they utilize pre- without good adherence, treatment failure is likely, existing systematic reviews to identify relevant research leading to avoidable HIV-related morbidity and mortality. evidence in addition to screening databases for recent Additionally, imperfect adherence increases the risk primary studies [18–21]. This practice is useful for of developing resistant HIV strains and transmitting the making health policy decisions, because it allows virus to others [5–7]. Because adherence behaviours examination of the evidence while ensuring that and patterns can profoundly affect an individual’s information is assimilated as fast as possible given prior treatment response and potentially narrow future work [18–24]. therapeutic options, improving and sustaining ART adherence is a critical component and priority of public Using pre-existing systematic reviews to identify relevant health efforts. primary articles reduces the time needed to identify the relevant body of evidence on a particular topic. However, People living with HIV and their care providers often given that the time required to conduct, complete, and face challenges in ensuring good adherence. A 2011 publish a systematic review typically ranges from 1 to meta-analysis, which pooled ART adherence of 33 199 2 years [20,22], synthesis solely based on pre-existing adults in 84 observational studies, reports that only 62% systematic reviews runs the danger of failing to of individuals took at least 90% of their prescribed incorporate evidence that has accrued over the most ART doses [8]. Given these adherence difficulties, recent few years. We thus supplement our systematic effective, feasible and acceptable interventions to review of systematic reviews, with a complete screening enhance ART adherence are urgently needed to ensure of databases of primary evidence, but – in order to the continued success and clinical and financial sustain- maintain rapidity in the identification of primary studies ability of the global ART scale-up [9–11]. Multiple – we constrained these searches to the past 2 years (2010– systematic reviews and meta-analysis of ART adherence- 2012) and to randomized controlled trials (RCTs). enhancing interventions have been conducted over the past few years, but these studies have often been Search strategies limited to particular interventions, populations, or To identify systematic reviews, we conducted searches settings [12–16]. in the Cochrane Library, EMBASE, MEDLINE, Web of Science, and WHO Global Health Library (which To inform the evidence base for the 2013 WHO includes both regional and global indices). The search consolidated guidelines on the Use Antiretroviral algorithms are shown in Boxes A1 and A2 in the appendix Drugs for Treating and Preventing HIV Infection [17], (http://links.lww.com/QAD/A499). Abstracts from we conducted a rapid systematic review synthesizing conferences and meetings were excluded because they the research results on ART adherence-enhancing do not undergo the same level of peer review as published interventions across intervention types, populations, full-text articles and they do not provide the necessary and settings. Our review advances the existing literature references for extracting study-level data. Publications in three ways: first, it is the most comprehensive on adherence interventions were excluded if they compilation of the evidence on adherence-enhancing were letters to the editor, editorials, commentaries, or interventions to date; second, it allows evaluation of opinion articles. We further excluded systematic robustness of interventions across settings; and third, reviews of interventions studying programme retention, we indicate studies that focus on specific populations of efficacy of combination antiretrovirals (fixed or multiple particular interest because of comorbidities and other medications), dosing strategies, or use of antiretrovirals vulnerabilities that may interfere with their ability to for pre-existing or post-exposure prophylaxis. We did not

limit our search to particular times, locations, or have been found to be sufficient to carry out a high- languages. Additionally, we searched ClinicalTrials.gov, quality systematic review [26]. The same reviewers used Cochrane Central Register of Controlled Trials, the inclusion and exclusion criteria to independently EMBASE, MEDLINE, Web of Science, and WHO assess the full eligibility of studies identified in the Global Health Library for RCTs published between 1 databases. Reviewers were not blinded to study authors, September 2010 and 31 August 2012 that investigated conclusions, or outcomes, because blinding is compli- interventions targeted towards improving ART adher- cated to implement and has been shown to have little ence. To be included in this review, RCTs could report effect on the quality of systematic reviews [27]. Once all an adherence intervention as the primary or secondary potentially relevant full-text articles and abstracts were aim or simply report adherence measurements in the identified, the three reviewers achieved consensus presence of an intervention. Studies comparing or regarding eligibility and extracted data onto a standar- validating adherence measurement approaches without dized extraction form. Where consensus was not possible, reporting on an adherence-enhancing intervention were a fourth reviewer (T.B.) served as arbiter. After relevant excluded. We followed the reporting standards described systematic reviews were identified, we searched for the in the Preferred Reporting Items for Systematic Reviews primary studies featured in these reviews and extracted and Meta-Analyses (PRISMA) statement [25]. the data from the studies. Data entry was compared, and discordant information was resolved by consensus Study selection through data checks and discussion between the data Three investigators (K.C., M.P., and O.O.) worked extractors. When necessary, the further reviewer (T.B.), independently, completing separate screenings of the who guided but was not directly involved in the primary literature. We screened titles and abstracts of studies data extraction process, was asked to mediate. Figures 1 that were identified in previous systematic reviews on and 2 show flowcharts of the study selection processes. the effectiveness of interventions aimed at increasing antiretroviral adherence; as well as titles and abstracts of Data extraction records identified in the search of databases for RCTs We organized the synthesis of results by adherence investigating adherence interventions. All records were intervention type, that is, the actual intervention activity, screened by two of the three reviewers; two reviewers such as directly observed therapy (DOT) or depression

923 systematic reviews identified 773 reviews excluded based on by database searches screening titles

150 reviews screened 60 duplicate reviews excluded

55 reviews excluded based on 90 reviews after duplicates removed screening abstracts

3 reviews excluded 35 full-text reviews assessed for • 1 Not a systematic review eligibility • 2 Not about HIV/AIDS

32 full-text systematic reviews for study extraction 383 studies excluded • 138 Duplicates • 86 No described intervention • 50 No comparison group • 42 Conference abstracts 488 studies available for full-text • 36 No adherence measure review and extraction • 14 Descriptive reports • 10 Letters or magazine articles • 3 Studies on drug effects • 3 Other 105 full studies included in final • 1 Non-ART intervention review

Fig. 1. Flowchart of study selection process based on systematic reviews of ART adherence-enhancing interventions. ART, antiretroviral therapy.

825 RCTs identified by database 690 RCTs excluded based on searches screening titles

57 RCTs excluded based on review 135 RCTs after screening of abstracts

40 RCTs excluded based on review 78 RCTs after screening of abstracts

19 RCTs excluded after review • 10 Duplicates 38 full-text RCTs assessed for • 6 Descriptive reports eligibility • 2 No comparison group • 1 Non-ART intervention

19 RCTs for data extraction

Fig. 2. Flowchart of study selection process of randomized controlled trials of ART adherence-enhancing interventions. RCTs, randomized controlled trials. ART, antiretroviral therapy.

treatment. In addition to the intervention types, we adherence-enhancing interventions increased over time; extracted from the studies the following data: author and each year before 2003 three or fewer articles were year of publication, study period, study design, country of published, whereas in 2003 and thereafter, at least six study, population, source of information, and healthcare articles were published each year and in many years more setting, in which the study took place; study duration, than 10 articles (Web Appendix, http://links.lww.com/ sample size, loss to follow-up, intervention, control QAD/A506). group, adherence measure, and study results. Web Appendix, http://links.lww.com/QAD/A506 shows Almost half (55) of the 124 studies investigated the the study characteristics; Table 1 provides an overview effectiveness of combination interventions, that is, of the different adherence-enhancing interventions that interventions that were composed of several clearly were tested in the studies and reports the results by identifiable components. The most commonly tested outcome measure. We report on results for subjective interventions were cognitive-behavioural therapy (CBT) adherence measures (self-report by patients), objective (60), followed by education (28), treatment supporters adherence measures (pill count, pharmacy refill, and (26), DOT (20) and active reminder devices (20). The less electronic monitoring), and the biological correlates of commonly tested intervention types included structural adherence (viral load, CD4þ cell count, and change in interventions (such as changes in the person delivering body weight). A few studies report composite adherence ART, or in the location where ARTwere provided) (10), indices incorporating information from several outcome counselling (8), nutritional support (7), financial incen- measures. We do not include the results in terms of tives (5), passive reminder devices (5), and drug use these outcome measures in our review, because the use treatment (4). Active reminder devices included both of these indices is usually particular to one study, and telephone reminders and other technologies, such as all studies using indices also report results in terms of pagers and pillboxes with in-built timers and alarms. outcome based on individual measures. Passive reminder devices included pillboxes and diary cards. Detailed information on intervention types and the interventions are shown in Table 1. Commonly (in 29 studies), CBT, education or counselling were combined Results with other interventions. DOT, passive reminder devices, treatment supporters, nutritional support, and financial A total of 124 studies met our selection criteria (Figures 1 incentives were combined with other interventions and 2). These studies included 86 RCTs, 6 non- in more than two-fifths of the studies, whereas the randomized controlled trials (NRCT), 19 before-after other interventions were less likely to be investigated in studies, 8 cohort studies, 4 case-control studies, and 1 combination. cross-sectional study. Seventy-five studies were carried out in North America, 30 in Africa, 11 in Europe, The synthetic picture that emerges becomes even more 4 in Asia, 3 in Central and South America, and 2 in complex when the success of particular interventions is Australia. Publication intensity in studies testing ART considered across different outcomes. Table 2 shows the

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Improving adherence to antiretroviral therapy Chaiyachati et al. S191 ) continued ( WC O þ Results –––YY – –– –––YY N –– –Y– –N – –– ––––– – –N Y–– –N – –– Y–– –Y N –– Y–– –Y – –– Y–– –N N –– Y – – – (Y) (Y) – – N–––– – –– N–N–N N – – N–––– – –– N–––– – –– SR PC PR EM VL CD4 (Y) – – – N N – N EM results symptoms and stress related to ART side effects programme consisting of educational materials,and goal targeted selection, interventions focusedbehavioral on strategies motivational for and improving adherence day-to-day experiences, identifying barriers toexploring motivations adherence, and adherence strategies and developing skills to avoid drugand use, improve unsafe adherence sexual practices, that provides consultation, telepharmacy, accessresources, to and a a library community of of other individuals withpsychiatrist, HIV a case manager, and pharmacist adherence and use of skills neededenhancing self-efﬁcacy, for and reinforcing proper positive treatment-taking attitudes toward behaviour, treatment-taking in the participant anonymous videotape, education materials planning cards, pill boxes training, adherence training, andadherence exploration of barriers to support; medication planners, SMS textbox, messages, and medication medication alarms treatments Home care monitoring through an internet-based clinical system Peer-led treatment with baseline psycho-educational component N – – – N – – – Nurse-led education about HIV and adherence, telephone-based Education on medication choices, side effects, and nutrition intervention education education nutritional support CBT Nurse-led counselling sessions seeking toCBT, stimulate education, development other Nurse counsellor-led motivational interview sessions, alcoholics CBT, PRD, education Personalized educational diagnoses made forCBT each patient, Manual-guided group therapy sessions with harm reduction skills ARD, PRD, CBT, CBT, education, Other Future writing N – – – – – – – Intervention type Intervention et al. et al. [48]) [48]) , 2011a [37] CBT, education Counselling about effective decision-making, providing education, , 2011 [38] Counselling Counselling sessions via mobile phones Y Y – – – – – – , [48]) , 2003 [53] , 2012 [31] ARD SMS messages before last scheduled medication for the day (Y) (Y) – (Y) – – – – , 2010 [32] Counselling, CBT Adherence counselling, brief motivational interviewing utilizing , 2012 [46] CBT Motivational interviews with information to promote motivation for , 2003 [50] , 2012 [33] CBT Mindfulness-based stress reduction strategies for reducing ART , 2011 [36] CBT Motivational interviewing group sessions involving exploring , 2003 [47] , 2011 [30] Counselling, ARD Intensive adherence counselling, pocket alarm device, or both – Y Y – Y N – – , 2011 [35] ARD Personalized mobile phone reminder system for adherence Y Y – N – – – – , 2012 [44] CBT Cognitive-behavioural therapy for adherence and depression – – – Y N N – – , 2011 [45] ARD Phone call reminders as memory aids Y – – – Y – – Y , 2011 [34] CBT, education Interactive computer-based antiretroviral adherence promotion et al. et al. , 2011 [28] Structural intervention Advanced practice nurse for monitoring and managing ART – – – – Y N – – , 2010 [43] ARD EM feedback – – – Y N Y – – , 2003 [49] (Amico et al. et al. et al. , 2011 [39], 2011 [40] Counselling, structural Depression treatment Depression treatment through a clinical team consisting of a , 2011 [29] DOT, other ART DOT and methadone maintenance therapy – Y – Y Y – – – , 2010 [42] Treatment supporters, et al. et al. et al. et al. et al. et al. [48]) et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. 2012 [41] (Amico et al. (Amico [48]) [48]) (Amico Blank Berg Chung da Costa de Bruin Hardy Holstad Duncan Fisher Kalichman Zubaran Kalichman Leon Pyne Ramirez-Garcia and Cote Ruiz Sabin Safren Uzma DiIorio Fairly Goujard Lyon 2003 [51] (Amico Mann, 2001 [52] (Amico Margolin Table 1. Adherence-enhancing interventions and results. Authors and year (review authors)

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. S192 AIDS 2014, Vol 28 (Suppl 2) WC O þ Results –––YN – –– –––Y– – –– –––NN N –– –––YN – –– ––Y–N N –Y ––Y–– N –– ––––N N –– Y–––– – –– Y–– –Y – –– Y–– –N Y –Y Y–– –Y – –– YY––– N –– N–––– – –– N–Y – Y N – – N– – –N N – – NN – – Y N – – SR PC PR EM VL CD4 (Y) – – – (Y) (Y) – – education and assistance withskills medication self-management medimom.com pathogenesis and treatment, and medicationadherence; management and or education modulesphysical focused description on of names medications, and dosageMEMS, instructions, importance use of of adherence, and side-effects cash incentives behavioural, problem-solving, and motivational interviewing techniques to enhance motivation, rehearsebehaviours, and adherence-related solve problems that interfereHIV with medications, adherence with to one follow-up telephone review nurse psychiatric nurse strategies side effects, weekly counselling, and follow-up phone calls relayed to physician for follow-up barriers for intervention group fears and attitudes about ART earner in his/her family;members rations if for patient a is total primary of income seven earner) household questions and pill count,side and effects to discuss treatment beneﬁts and supporters (from the community or the patient’s family) response within 48 h community-based model of ART care Customized medication schedules, daily reminders, with or without Medication counselling, pill boxes, education on problem solving Nurse-led DOT and adherence support, side effects information Psycho-education, education material, counselling supportStructured home-based support for education and identifying NMedication diaries and counselling – – – Y N – – – – – – N N – – Peer support sessions, in which ART patients discuss their feelings, Treatment supporter and/or nutritional support within a Calendar for record-keeping of dose intake, or treatment supporters N – – – – N N – other ARD education counselling education CBT CBT nutritional support other CBT, educationDOT, counselling, Feedback on adherence, rooted in social cognitive theory, ARD Daily pill diary, paged electronic reminders through www. CBT, educationFinancial incentives, Education modules, focusedCBT on patient empowerment, HIV Life-Steps protocol, a single-session intervention utilizing cognitive- Counselling, education Individual educational and counselling sessions with a trained CBTARD Intervention sessions by cognitive-behavioural therapist and Timer, pager, or pillbox with timer integrated into box N – – – – – – – CBT, PRD, counselling, CBT, ARD, education Individual education sessions about antiretroviral medication and CBT, education, Treatment supporters, Treatment supporters, Treatment supporters, Nutritional support Food rations (individual rations if patient is not the primary income Treatment supportersDOT Home visits by treatment supporters to promote adherence through ARD DAOT, TWOT, or WOT; provided by patient-selected treatment SMS from study clinicians asking ‘How are you?’ requiring a Intervention type Intervention Treatment supporters, Treatment supporters, ) [71]) [71]) [71]) [71]) [71]) [71]) [67]) [67]) , 2000 et al. , 2005 [48]) , 2003 [57] et al. et al. et al. et al. et al. et al. et al. et al. et al. , 2003 [55] [48]) [48]) [48]) [48]) [48]) et al. , 2009 [69] et al. , 2003 [59] et al. , 2008 [70] , 2002 [56] , 2009 [76] , 2001 [64] , 2004 [66] , 2010 [74] , 2010 [72] , 2003 [58] (Amico , 2000 [60] (Amico et al. , 2000 [65] (Amico continued , 2003 [62] (Amico , 2001 [61] (Amico , 2010 [75] , 2007 [73] , 2003 [63] (Amico et al. et al. et al. et al. et al. et al. et al. et al. [48]) [48]) [48]) [48]) [48]) [48]) et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen ¨ ¨ ¨ ¨ ¨ ¨ et al. (Amico et al. (Amico et al. et al. et al. (Amico (Amico [54] (Amico (Amico et al. (Bain-Brickley [68] (Bain-Brickley [67]) (Bain-Brickley (Ba (Ba (Ba (Ba (Ba (Ba Stenzel Safren Smith Rawlings Rigsby Safren Murphy Powell-Cope Pradier McPherson-Baker Molassiotis Tuldra Berrien Funck-Brentano Cantrell Wamalwa Chang Idoko Kabore Lester Table 1 ( Authors and year (review authors) Mugusi

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Improving adherence to antiretroviral therapy Chaiyachati et al. S193 ) continued ( –N– –N Y – – ––Y–– – –– ––Y–Y N –– ––––Y Y –– –––Y– – –– Y––– – N –– Y––– – – –– Y–––Y Y –– YY– –N – – – N– – – Y Y – – N– – –N N – – N– – –N N – – N– – – – – – – N– – – – Y – – (Y) – – – – – – – inventory instrument, one baselinetreatment education supporter, session four for additional baseline educationand sessions refresher course everytreatment three supporter months for patients and with on-site clinician, drug treatment coordinator, case manager, outreach workers, methadone co-management lipid paste provided psycho-social support managing adverse effects, and reminded patients of drug pick-up lipid paste linked patients to community service providers, andcare coordinated with medical staff and community health workers and provided support to improve adherence non-DOT days treatment, identification and mitigation of adherence barriers treatment coping skills and medication adherence and educational material community-based models (doctor-led primary carenurse-led primary clinic, care clinic, integrated primary caretwo clinic) hospital-based and models (rural district hospital, hospital-based specialist service) short weekly reminders, or long weekly reminders) lifestyle issues, by pharmacist at first ART dispensing interaction DAOT by a treatment supporter chosen using a personal network Enhanced community-based DOT, beeper reminders, mobile vans Treatment supporters provided DOT, assisted in reporting and Treatment supporter-delivered DAOT, patient education about Cognitive stress management with expressive-supportive therapy treatment supporters supporters, CBT DOT supporters, education education DOT, education, DOT, treatment Nutritional support Supplementary feeding with ready-to-use fortified, energy-dense, Treatment supportersTreatment supporters, Treatment supporters (from the community or the patient’s family) DOT DOT by a healthcare professional – – – N N N – – Nutritional support Supplementary feeding with ready-to-use fortified, energy-dense, Structural interventionTreatment supporters Case managers (nurses or socialCBT workers) identified patients’ needs, CBT Treatment supporters followed up with patients in the community Cognitive medication adherence and management training A mindfulness-based stress reduction meditation programme N – – – – – N – N N N N – – – – DOT, drug use treatment DOT by nurse or medical assistant and prepackaged doses on DOT, treatment CBTDrug use treatment, CBT, Individually tailored CBT sessions designed to improve HIV DOT DOT by trained outreach worker, prepackaged pills Y – – – Y Y – – Structural intervention Five different models of ART delivery; three Structural intervention Home-based ART delivery N – – – N N – – ARD Different types of SMS (short daily reminders, long daily messages, DOTCBT, education Individual advise and education on Nurse-supervised ART DAOT adherence, addressing N N – N – – – Y PRD Diary cards with calendars showing medication dosing schemes N – – – – – – – DOT TWOT at nearby clinics, pill counting, and treatment support Y N – – N N Y – Structural intervention Assignment to non-physician clinicians – – Y – – Y – Y et al. [71]) [71]) [71]) [71]) [71]) [71]) [71]) [71]) [71]) [71]) [71]) [71]) [71]) [16]) , 2011 [82] et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. [91]) [91]) , 2007 [104] , 2010 [88] , 2010 [77] [101]) , 2009 [92] , 2011 [93] , 2009 [78] , 2009 [79] , 2007 [80] , 2006 [81] et al. , 1999 [98] , 2008 [89] , 2009 [86] et al. , 2010 [87] , 2006 [90] (Brown , 2007 [99], Maru , 2011 [94] (Brown , 2009 [102] (Hart , 2006 [103] (Hart , 2008 [84] , 2004 [83] , 2010 [85] et al. et al. , 2009 [95] (Brown , 1995 [96] (Fogarty et al. et al. et al. et al. et al. et al. et al. et al. [101]) [91]) [101]) et al. [91]) [91]) [97]) et al. et al. et al. , 2009 [100] (Hart et al. et al. et al. et al. et al. et al. et al. et al. rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen rnighausen et al. et al. et al. et al. ¨ ¨ ¨ ¨ ¨ ¨ ¨ ¨ ¨ ¨ ¨ ¨ ¨ (Ba et al. [101]) (Ba (Ba (Ba et al. (Ba (Ba (Ba et al. (Brown et al. (Ba (Brown et al. (Hart (Ba et al. (Ba et al. (Ba (Haddad (Ba (Ba Nachega Ndekha Stubbs Gross Ndekha Taiwo Thurman Torpey Lucas Pearson Antoni Creswell Johnson Macalino Pienaar Weiss Jaffar Pop-Eleches Wall Roux Knobel Altice Sarna Sherr

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. S194 AIDS 2014, Vol 28 (Suppl 2) WC O þ Results ––––Y Y –– –––YN N –– ––––Y N –– –––YN N –– Y––YN – –– Y–– –Y N –– N– –NN N – – N – – – Y (N) – – N– – –N N – – N– –NY N – – N– – –N – – – N– –NN – – – NN – NN – – – SR PC PR EM VL CD4 therapy enhancement of perceived medicationHIV efficacy, individualized counselling and exploration interviewing, problem-solving skills strategy, and support groups provided social support; financial aidtransportation; nutritional for support tests, medication, provided social support; financial aidtransportation; nutritional for support tests, medication, problems with the patient,adherence financial case incentives, or management intensive reminder message system to remind patients to take medications management education about coping strategiesadherence and the medical impact of peer-directed phone calls caseworker, and pharmacist, or peer DAOT and social support barriers to adherence and propose solutions audiotape and booklet, one-on-one sessionseducator, mail with follow-up a after health each session Educational sessions with nurse practitioners, motivational DOT by lay healthcare worker who also monitored side effects and DOT by community health worker who also discussed adherence A couple-focused adherence programme to provide support and Peer-led sessions on HIV care, adherence, and risk behaviours N – – – – – – – Care management team consisting of social worker, peer Home visits by nurses and community support workers to discuss CBT, education supporters, nutritional support supporters, financial incentive CBT, education CBT supporters, CBT CBT CBTCBT Psychotherapy sessions based on concepts of cognitive-behavioural Motivational interviewing – – – N N N – – CBTTreatment supporters, Assessment of alcohol use and readiness for behaviour change, DOT, treatment DOTDOT, treatment ARDARD DOT byTreatment lay healthcare supporters, worker who also monitored side effects and Disease Management Assistance system (DMAS), an electronic Serial, supportive phone calls using a standardized script, side effect CBTDOT Motivational interviewing and cognitive-behavioural skills training DAOT by peer outreach workers Y – – – – Y – Y – – – – (Y) (Y) – – Intervention type Intervention Treatment supporters, Treatment supporters Peer-led sessions to identify barriers, create coping strategies, and DOT, treatment Treatment supporters, CBT Motivational interviewing focused on adherence, including DOT DAOT by outreach worker – – – – (Y) (Y) – – ) [109]) , 2005 , 2006 [122] et al. [125]) [125]) [109]) [109]) [109]) [109]) [109]) [119]) [119]) [119]) [119]) [119]) et al. , 2006 [123] , 2005 [108] et al. , 2003 [126] , 2005 [111] , 2007 [117] (Hill , 2008 [115] (Hill , 2007 [120] , 2007 [121] , 2010 [105] (Hart , 2005 [110] , 2004 [114] , 2004 [106] (Hart et al. et al. , 2005 [112] et al. et al. et al. et al. et al. continued , 2006 [124] , 2006 [107] (Hart , 2005 [118] et al. et al. et al. et al. et al. et al. et al. et al. [101]) [101]) [101]) et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. [113] (Haynes (Haynes and Kavookjian [116]) and Kavookjian [116]) et al. et al. et al. (Haynes (Haynes (Haynes (Haynes (Kenya (Kenya (Kenya (Kenya (Kenya (Leeman (Leeman DiIorio Parsons Munoz Tinoco Van Servellen Weber Wohl Andrade Collier Remien Samet Mitty Purcell Simoni Table 1 ( Authors and year (review authors) Visnegarwala Williams Golin Harwell

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Improving adherence to antiretroviral therapy Chaiyachati et al. S195 ) continued ( Y–– a ––––Y Y –– –––YY N –– Y––NN N –– Y (Y) – (Y) – – – – Y–– –N Y–– –N N –– Y–– –N N –– N–––– – –– N– –NY N – – NNNNN N – – N–––– – –– N–––– – –– N– – –N N – – N– –NN – – – N–––– – –– practice ART supportive therapy MEMS devices on time restrictions, adverse-event management strategies, and monitoring of patient progress after therapy initiation programme with a range of interventions, includingdiscussions teaching and about adherence, self-care managementperceived of side-effects, role performance, andthe client-provider improvement relationship in expressive supportive therapy intervention withhealthier education lifestyle on a follow-up with case management tailored adherence support inelectronic a medication protocol-guided reminder manner, system or usingportable a alarm small for all antiretroviral doses, or both monetary reinforcement generate possible solutions, selectbarriers, strategies and to evaluate overcome how the supporter, strategies are MEMS providing working) treatment electronic adherence cues bimonthly postal reminders building, behavioural cues rewards education adherence tools (dosette boxeselectronic for alarms) antiretroviral pills and Individualized case management with treatment supporters and Structured interviews (to help patients identify adherence barriers, Reinforcement of medication taking with prizes or monetary use treatment, treatment supporters CBT use treatment, CBT CBTCBT Cognitive-behavioural adherence intervention with or without Cognitive-behavioural stress management and expressive Financial incentives Medication coaching and voucher reinforcement for opening CBT, education Visit and phone follow-up to provide education about ART, food CBTFinancial incentives, drug CBT IndividuallyCBT, tailored, education nurse-delivered adherence intervention Treatment supporters, Group cognitive-behavioural stress management sessions and Individually cognitive-behavioural interventionsCBTARDFinancial incentives, drug Y – Motivational interviewing and cognitive-behavioural therapy – Structured telephone calls by speciﬁcally trained nurse – (Y) – – Y – – – – – – – – – – N – – – – – CBTCBT Instructional support programme to enhance health literacy with Medication manager involving research staff member providing ARD, PRDDOTCBT Printed cards with information about each drug, pill boxes, and DOT Printed and verbal adherence information, self-efﬁcacy and skill (Y) – – – (Y) – – – CBT, educationCBT, education Sessions guided by cognitive-behavioural principles, psycho- Adherence education programme, individualized counselling, , 2003 [143] , 2006 [146] [125]) [125]) [125]) [125]) [125]) [125]) [125]) [125]) [125]) [125]) [125]) [125]) , 2006 [128] , 2006 [127] , 2008 [136] , 2007 [138] , 2005 [142] et al. , 2007 [129] , 2006 [139] et al. , 2005 [135] , 2008 [131] , 2005 [134] et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. , 2007 [137] et al. , 2004 [144] , 2006 [132] , 2003 [140] , 2007 [130] et al. , 2004 [145] (Rueda et al. et al. et al. , 2008 [133] (Leeman et al. et al. [125]) [15]) et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. (Manias and Williams [141]) (Leeman (Manias and Williams [141]) (Leeman (Manias and Williams [141]) (Saberi and Johnson, 2011 [147]) (Leeman (Leeman (Leeman (Leeman (Leeman (Leeman (Leeman (Leeman (Manias and Williams [141]) (Leeman et al. (Leeman et al. Jones Wagner Rathbun von Servellen Holzemer Javanbakht Jones Koenig Parsons Reynolds Rosen Wyatt Mannheimer Johnson Levin Ma Milam Sorenson Levy

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. S196 AIDS 2014, Vol 28 (Suppl 2) l WC O þ Results ––NN N –– eight change; WOT, weekly DOT. a –Y–Y– – –– ––Y–Y – –– –() – –– –––(Y)– ––N–Y N –– – – – – (Y) – – – – – – – (Y) (Y) (Y) – –N––– – –– Y––NN N –– Y–– –N – –– N–––– – –– N–––– – –– SR PC PR EM VL CD4 N n group; results (Y, N) are shown in parentheses if an effect size is reported and the authors draw a cell count; DAOT, daily DOT; DOT, directly observed therapy; EM, electronic monitoring; MEMS, medication event monitoring system; O, other; PC, pil þ CD4 , reinforcement strategies programme provided by adherence supporters with group meetings, or both behavioural and problem-solving approaches, or both and electronically records doses, adherence education session worker focused on ARTidentification education, and ART mitigation readiness, of and adherence barriers counsellors reducing sexual and drug use acts, and improving mental health patient’s interaction with her family and other socialhand-outs groups support, financial assistance, patienttransportation support groups, and adherence diaries, and education þ Pager messaging with a reminder device or phone, or peer support Treatment supporter initiative designed to improve access, supporters education CBTARD, treatment Sessions focused on role-playing, problem-solving, coaching, Treatment supporters Patient advocates, a community-based adherence support ARD, CBTARD, CBT, educationCBT, education Prompting device thatStructural verbally Electronic reminds intervention patients reminder at device, medication times or counselling on cognitive- CBT One-on-one sessions withStructural a intervention Pharmacist-provided pharmacist, ART dietician, medication and management social DOTARD, CBT Pharmacist-provided ART medication management CareCBT, at other clinics employing an – HIV clinical pharmacistCBT, education DOT withStructural Phone weekly intervention or follow-up in-person Y visits sessions from focused – on pharmacists improving or physical adherence health, Y Family-based HIV interventions Motivational services therapy, – interviewing provided emphasizing and by the – counselling general female sessions, practitioners educational – – – – – Y N – Y – – Y Y – – – N – – – – N – N – – Nutritional support Food support programme (Y) – – – – (Y) – – Treatment supporters, Structural intervention Community-based treatment programme providing nutritional Intervention type Intervention ) , [159]) , 2004 [163]) [163]) [168]) [168]) [168]) et al. et al. et al. et al. , 2012 [153]) , 2012 [153]) , 2012 [153]) , 2011 [169] , 2011 [164] , 2011 [167] , 2007 [155] , 2007 [148] , 2009 [149] , 2011 [150] , 2010 [162] et al. et al. et al. , 2011 [154] , 2007 [156] continued , 2008 [160] , 2009 [157] (Saberi , 2006 [152] (Saberi et al. et al. et al. , 2003 [165] (Wong , 2012 [170] et al. , 2006 [151] (Saberi et al. et al. et al. [166]) et al. , 2012 [153]) , 2012 [153]) et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. et al. (Saberi and Johnson, 2011 [147]) (Saberi and Johnson, 2011 [147]) (Saberi and Johnson, 2011 [147]) and Johnson, 2011 [147]) et al. (Saberi (Saberi (Saberi et al. [158] (Simoni et al. (Wouters (Tirivayi and Groot [161]) (Wechsberg (Wechsberg (Wouters (Wouters The study showed that the intervention decreased adherence as assessed by this outcome measure. Murphy Simoni Simoni Wu Frick Hirsch Horberg March Pirkle Rotheram-Borus Page Igumbor Byron Feaster Ingersoll count; PR, pharmacy refill;Y PRD, means passive significantly better reminder outcome in devices; theconclusion RD, intervention group as active (at least to reminder at whether devices; one time the SR, point); N intervention self-report; means has not TWOT, significantly improved twice better outcome adherence weekly in the or DOT; interventio not VL, but viral without load; reporting WC, significance w levels. ART, antiretroviral therapy; CBT, cognitive and/or behavioural therapy; CD4 Table 1 ( Authors and year (review authors) Kunutsor Rich a

distribution of outcome measures used across the only to initiate patients on ART once we are able to 124 studies. Two-fifths of studies followed the general ensure good ARTadherence. It could also reflect the fact recommendation to use both outcomes that capture that ART adherence is more easily conducted than adherence (subjective measures-self-reported adherence research into other aspects of ART services, because levels, or objective measures – pill count, pharmacy refill, unlike studies of access, linkage, and retention, it only etc.), as well as those that capture the biological outcomes requires data collection in clinical cohorts and not in determined by adherence behaviour (viral load, CD4þ HIV-infected populations in communities. Whatever cell count, body weight). However, 16% of the studies the reason for the intensity of the research on ART measured adherence using only subjective outcomes. adherence-enhancing interventions, the speed of study Overall, 72 of the 124 studies were found to generate implementation, analysis, and publication means that significant positive effects as assessed by at least one evidence syntheses will rapidly grow out of date. Our outcome measure. But only 24 studies (or one-fifth) review provides an updated synthesis on the body of found significant positive effects in at least one knowledge on the effectiveness of ART adherence- biological and one (objective or subjective) ART- enhancing interventions. adherence measure. Combination interventions were not more or less likely to succeed in significantly Each of the following interventions has been tested improving outcomes than single interventions (P ¼ 0.80 in more than 20, mostly rigorous studies, either singly for having at least one positive effect across all outcomes; or in combination with other interventions: CBT, P ¼ 0.55 for having at least one positive effect each for education, treatment supporters, DOT, and active a biological and a subjective or objective adherence adherence reminder devices (such as mobile phone text outcome). messages). Whereas there is strong evidence that all five of these interventions can significantly increase Table 3 shows a synthesis of the study results by ART adherence in some settings, each intervention intervention type. In the case of combination interven- has also been found not to produce significant effects in tions, each component intervention is counted separately. several studies. The table shows that for most interventions, at least three- fifths of the studies found a positive result for at least The 2013 WHO consolidated guidelines on the use of one outcome, but the proportion of studies finding antiretroviral drugs for treating and preventing HIV positive results for both at least one biological and one infection describe the portfolio of adherence-enhancing subjective or objective adherence outcome is less than interventions and recommends that ‘[M]obile phone text 50%. messages could be considered as a reminder tool for promoting adherence to ART as part of a package of Most studies (87) investigated adherence-enhancing adherence interventions’. This recommendation, as well interventions in the general population; the remainder as the descriptions of the evidence on other adherence- focused on particular sub-populations. The most enhancing interventions in the guidelines, have been commonly researched sub-populations were persons informed and are broadly supported by this systematic who use drugs (PWUD), with 22 studies, followed by review. In addition – and with the caveats regarding women (8 studies), children (4 studies), and persons with context-specificity of findings discussed below – our mental health disorders (2 studies). It is an important review suggests that the other four interventions which finding that despite overall small sample sizes, there have been widely tested in rigorous studies – CBT, were significant effects in 12 out of the 22 studies in education, treatment supporters, and DOT – warrant PWUD. Syntheses of results by outcome measure are consideration by ART programme managers. Given the presented in Table 1. critical importance of adherence for the long-term individual and population-level success of ART, routine implementation of adherence-enhancing interventions should be considered. Discussion Whereas the current evidence base provides a portfolio of A large global evidence base on ART adherence- interventions that have been shown to be effective in enhancing interventions – a total of 124 studies including high-quality studies at least in some settings, adherence is 86 RCTs – provides important information for ART a behaviour and as such is affected by culture and programming and planning. The field of ARTadherence circumstance. The standard approaches to synthesizing intervention research is developing rapidly and relatively evidence on effectiveness take on a different meaning more rapidly than research into ART access, linkage to when considering behavioural interventions as opposed care, and retention. The reason for this differential to biological interventions. For behavioural inter- in research intensity within the overall field of HIV ventions, consistency of causal effects across studies operations and health services research plausibly reflects is an indicator of the degree of generalizability of an the importance of ART adherence – we would prefer intervention effect to other settings rather than a measure

Table 2. Distribution of outcome measures. as it is introduced into routine ART services. Quasi- % of studies experimental designs, such as stepped wedge scale-up Type of outcome measure (N ¼ 124) of adherence interventions across HIV clinics, might offer ‘natural’ opportunities for rigorous confirmation of Subjective adherence measure only 16 effectiveness of the five interventions that the currently Objective adherence measure only 6 Subjective and objective adherence measure 4 available body of evidence can increase adherence. Biological measure only 10 Biological measure and subjective and/or 63 Whereas the global evidence on effectiveness of objective adherence measure Other 1 adherence-enhancing interventions is rich, our review has identified several important knowledge gaps that will be relevant for implementation decisions and should of the degree to which an effect is ‘true’ as in the case of increasingly be filled with evidence from implementation biological interventions. science research. First, more evidence is needed to examine interventions that have shown promise in a few We would expect that behavioural interventions that studies, but have only been tested in a limited range have been truly successful in one setting may not be of settings. Our review finds that these interventions effective in another one with different economic, social include the following: alternative health system structures and behavioural barriers to adherence. Thus, health for ART delivery, nutrition support, financial incentives, policy makers and programme planners need to carefully passive reminder devices (such as diary cards and consider which adherence intervention to choose for compartmentalized pill boxes), drug use treatment, and routine implementation in a particular setting based on anti-depressive treatment. socio-cultural context, feasibility, acceptability, and health systems organization. The adherence-enhancing inter- Second, comparative information on costs and cost- ventions identified in this review are likely to differ effectiveness of different effective adherence interventions widely in implementation-relevant aspects, such as costs, is largely lacking, and when it is available, it is unclear in human resources requirements, and scalability. Thus, how far the costs assessed in a research setting are other factors than the effectiveness evidence covered in transferable to routine implementation situations. More this review will likely guide implementation decisions. cost-benefit studies as part of routine care are needed For instance, DOT is labour-intensive and expensive, but to provide this critical component for deciding between it may be a good strategy for particular settings, for alternative effective adherence-enhancing interventions. example, where patients can be easily reached, such as in Whereas several studies investigated combination hospitals or prisons. In contrast, some types of mobile interventions (see Table 1), differential effectiveness of phone text messaging interventions are comparatively alternative combination portfolios and interaction effects inexpensive and do not require substantial human between different intervention components were rarely resources investment. As such, they may be a good examined. It would seem plausible that combination option for general populations with high individual adherence interventions will be particularly successful mobile phone coverage. Future meta-analyses of the in increasing ART adherence because they commonly contextual predictors of success of particular types of work through different pathways. However, our synthesis ART adherence interventions can further inform these shows that combination interventions tend to be similarly choices. Additionally, it will be critical to monitor the likely to succeed in increasing ART adherence as single performance of an adherence-enhancing intervention interventions. One reason for this finding could be that

Table 3. Summary of effects of adherence-enhancing interventions.

% with positive % with positive results for at least one Number of results for at least positive effect each for a biological and Intervention component studies one outcome measure a subjective or objective adherence outcome

CBT 60 67 20 Education 28 79 21 Treatment supporter 26 62 19 DOT 20 85 30 ARD 20 75 25 Structural 10 70 10 Counselling 8 88 63 Nutrition support 7 71 43 PRD 5 60 0 Financial incentives 5 60 0 Drug use treatment 5 80 40 Depression treatment 1 0 0

ARD, active reminder device; CBT, cognitive-behavioural therapy; DOT, directly observe therapy; PRD, passive reminder device.

there is usually one dominant intervention within the whose primary aim was to enhance ART adherence. combination, and the other interventions merely These selection criteria may have led to the exclusion moderately enhance the effectiveness of the dominant of some interventions that can be of use in enhancing intervention. Another reason could be that combination ART adherence, in particular, approaches to optimize interventions are more difficult to implement than ARTregimens [177]. One example of such an intervention single interventions, and the achieved effects reflect is single-tablet ART regimens, which have not been these implementation difficulties. Future experimental included in our review. Recently published reviews research should increasingly use factorial designs that concluded that single-tablet regimens improve adherence allow precise determinations of component intervention and quality of life among ART patients in comparison to and interaction effects. multi-tablet regimens [178,179].

Third, the majority of studies establishing the effective- Another unavoidable limitation of a systematic review ness of adherence-enhancing interventions have lasted based on formally published studies in a fast moving 2 years or less. Antiretroviral therapy, however, requires research field is that evidence that is emerging life-long adherence, spanning several decades for many informally but has not yet been formally published patients. Long-term studies of ART adherence are will likely have been ignored, because academic urgently needed, and several teams are currently writing, review and publication times in global health conducting follow-up studies, which will generate these can last several years. These delays would have been important results [171–174]. Fourth, many studies are particularly limiting if they led to the exclusion of internally inconsistent in their findings, establishing completely novel interventions, for example, based on significant effects on some outcomes (e.g. self-reported new technologies. adherence), but not on other, related outcomes (e.g. immunological recovery). Technological improvements Although some studies were identified related to PWUD, in capturing ART adherence could substantially improve data on other key populations were scarce. Given the strength of the evidence regarding adherence that these populations are disproportionately affected behaviours, which tend to be unreliably reported [175] by the HIV epidemic and commonly face multiple and may also not be accurately measureable with objective challenges in ARTadherence, future research focused on approaches, such medication event monitoring systems ART adherence-enhancing interventions tailored to (MEMS), pill counts, or observation of pharmacy refill. key populations will be important, in particular, in Finally, as ART initiation is moving into earlier disease sub-Saharan Africa, where such focused studies have been stages, average effects of ART adherence-enhancing especially scarce. interventions may change, because the population composition of people on ART changes. For instance, In conclusion, we find a large and overall strong evidence people initiating in earlier stages of HIV infection are base to support the claim that five interventions – CBT, less likely to have experienced recovery from advanced education, treatment supporters, DOT, and active HIV-related disease and may thus require different reminder devices – can improve ART adherence at least cognitive and behavioural strategies and different in some settings. These tested and effective adherence- technological support to ensure good adherence than enhancing interventions should increasingly be con- people who initiated in late stages of the infection [176]. sidered for routine implementation in ART programmes and health systems. However, rigorous on-going Our study has several limitations. Although it was a evaluation of the impact and performance of these systematic review, it was ‘rapid’ in the methodological interventions will be critical, because all interventions sense that it utilized existing systematic reviews to identify that proved effective in at least one setting were also found studies on adherence-enhancing interventions. Some of not to significantly increase adherence in at least one these systematic reviews may have missed relevant studies other setting. Significant evidence gaps on adherence- related to their objective and timeframe, and these studies enhancing interventions need to be closed, including could have also been missed in our review. In particular, on cost-effectiveness, long-term effectiveness, and effec- the reliance on previous systematic reviews and our tiveness in specific key populations. focused search of recent published results from RCTs imply that our synthesis is largely based on experimental studies, and that an additional review of quasi- experimental and non-experimental evidence may Acknowledgements provide important additional insights. Additionally, our selection of reviews to identify primary studies under Conflicts of interest the rapid review methodology we employed excluded There are no conflicts of interest. reviews that were not systematic, for example, narrative reviews; and our identification of records reporting TB and KC were the lead authors, designing the study in primary RCT-based results was limited to studies close collaboration with EN and AS. KC, OO and MP

scrutinized identified studies for eligibility and extracted 17. World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infec- data. TB and KC wrote the first draft of the manuscript; all tion. Geneva: World Health Organization; 2013, www.who.int/ authors contributed to the interpretation of the extracted hiv/pub/guidelines/arv2013. [Accessed 26 January 2014] data and critically reviewed the manuscript before 18. Smith V, Devane D, Begley CM, Clarke M. Methodology in conducting a systematic review of systematic reviews of health- submission. care interventions. BMC Med Res Methodol 2011; 11:15. 19. Whitlock EP, Lin JS, Chou R, Shekelle P, Robinson KA. Using existing systematic reviews in complex systematic reviews. Ann Intern Med 2008; 148:776–782. 20. Ganann R, Ciliska D, Thomas H. Expediting systematic References reviews: methods and implications of rapid reviews. Imple- ment Sci 2010; 5:56. 1. Swendeman D, Ingram BL, Rotheram-Borus MJ. Common 21. Ba¨rnighausen T, Tanser F, Dabis F, Newell ML. Interventions elements in self-management of HIV and other chronic ill- to improve the performance of HIV health systems for treat- nesses: an integrative framework. AIDS Care 2009; 21:1321– ment-as-prevention in sub-Saharan Africa: the experimental 1334. evidence. Curr Opin HIV AIDS 2012; 7:140–150. 2. WHO, UNICEF, UNAIDS. Global update on HIV treatment 22. Khangura S, Konnyu K, Cushman R, Grimshaw J, Moher D. 2013: results, impact and opportunities. www.who.int/hiv/ Evidence summaries: the evolution of a rapid review pub/progressreports/update2013/en. [Accessed 26 January approach. Syst Rev 2012; 1:10. 2014] 23. Bambra C, Joyce KE, Bellis MA, Greatley A, Greengross S, 3. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour Hughes S, et al. Reducing health inequalities in priority public MC, Kumarasamy N, et al. Prevention of HIV-1 infection with health conditions: using rapid review to develop proposals for early antiretroviral therapy. N Engl J Med 2011; 365:493–505. evidence-based policy. JPublicHealth(Oxf)2010; 32:496–505. 4. Tanser F, Ba¨rnighausen T, Grapsa E, Zaidi J, Newell ML. High 24. Greenhalgh T, Peacock R. Effectiveness and efficiency of coverage of ART associated with decline in risk of HIV search methods in systematic reviews of complex evidence: acquisition in rural KwaZulu-Natal, South Africa. Science audit of primary sources. Br Med J 2005; 331:1064–1065. 2013; 339:966–971. 25. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, 5. Press N, Tyndall MW, Wood E, Hogg RS, Montaner JS. Ioannidis JP, et al. The PRISMA statement for reporting Virologic and immunologic response, clinical progression, systematic reviews and meta-analyses of studies that evaluate and highly active antiretroviral therapy adherence. J Acquir healthcare interventions: explanation and elaboration. J Clin Immune Defic Syndr 2002; 31 (Suppl 3):S112–117. Epidemiol 2009; 62:e1–e34. 6. Wood E, Hogg RS, Yip B, Harrigan PR, O’Shaughnessy MV, 26. Buscemi N, Hartling L, Vandermeer B, Tjosvold L, Klassen TP. Montaner JS. Effect of medication adherence on survival of Single data extraction generated more errors than double data HIV-infected adults who start highly active antiretroviral extraction in systematic reviews. J Clin Epidemiol 2006; therapy when the CD4R cell count is 0.200 to 0.350 x 59:697–703. 10(9) cells/L. Ann Intern Med 2003; 139:810–816. 27. Berlin JA. Does blinding of readers affect the results of meta- 7. Bangsberg DR, Perry S, Charlebois ED, Clark RA, Roberston M, analyses? University of Pennsylvania Meta-analysis Blinding Zolopa AR, et al. Nonadherence to highly active antiretroviral Study Group. Lancet 1997; 350:185–186. therapy predicts progression to AIDS. AIDS 2001; 15:1181– 28. Blank MB, Hanrahan NP, Fishbein M, Wu ES, Tennille JA, Ten 1183. Have TR, et al. A randomized trial of a nursing intervention for 8. Ortego C, Huedo-Medina TB, Llorca J, Sevilla L, Santos P, HIV disease management among persons with serious mental Rodriguez E, et al. Adherence to highly active antiretroviral illness. Psychiatr Serv 2011; 62:1318–1324. therapy (HAART): a meta-analysis. AIDS Behav 2011; 15: 29. Berg KM, Litwin A, Li X, Heo M, Arnsten JH. Directly observed 1381–1396. antiretroviral therapy improves adherence and viral load in drug 9. Altice FL, Kamarulzaman A, Soriano VV, Schechter M, users attending methadone maintenance clinics: a randomized Friedland GH. Treatment of medical, psychiatric, and sub- controlled trial. Drug Alcohol Depend 2011; 113:192–199. stance-use comorbidities in people infected with HIV who use 30. Chung MH, Richardson BA, Tapia K, Benki-Nugent S, Kiarie drugs. Lancet 2010; 376:367–387. JN, Simoni JM, et al. A randomized controlled trial comparing 10. Atkinson MJ, Petrozzino JJ. An evidence-based review of the effects of counseling and alarm device on HAART adher- treatment-related determinants of patients’ nonadherence to ence and virologic outcomes. PLoS Med 2011; 8:e1000422. HIV medications. AIDS Patient Care STDS 2009; 23:903–914. 31. da Costa TM, Barbosa BJ, Gomes e Costa DA, Sigulem D, de 11. Gordon CM. Commentary on meta-analysis of randomized Fatima Marin H, Filho AC, et al. Results of a randomized controlled trials for HIV treatment adherence interventions. controlled trial to assess the effects of a mobile SMS-based Research directions and implications for practice. J Acquir intervention on treatment adherence in HIV/AIDS-infected Immune Defic Syndr 2006; 43 (Suppl 1):S36–40. Brazilian women and impressions and satisfaction with respect 12. Simoni JM, Pearson CR, Pantalone DW, Marks G, Crepaz N. to incoming messages. Int J Med Inform 2012; 81:257–269. Efficacy of interventions in improving highly active 32. de Bruin M, Hospers HJ, van Breukelen GJP, Kok G, Koevoets antiretroviral therapy adherence and HIV-1 RNA viral load. WM, Prins JM. Electronic monitoring-based counseling to en- A meta-analytic review of randomized controlled trials. hance adherence among HIV-infected patients: a randomized J Acquir Immune Defic Syndr 2006; 43 (Suppl 1):S23–35. controlled trial. 4th ed. United States: American Psychological 13. Ford N, Nachega JB, Engel ME, Mills EJ. Directly observed Association Inc; 2010. pp. 421–428. antiretroviral therapy: a systematic review and meta-analysis 33. Duncan LG, Moskowitz JT, Neilands TB, Dilworth SE, Hecht of randomised clinical trials. Lancet 2009; 374:2064–2071. FM, Johnson MO. Mindfulness-based stress reduction for HIV 14. Hart JE, Jeon CY, Ivers LC, Behforouz HL, Caldas A, Drobac PC, treatment side effects: a randomized, wait-list controlled et al. Effect of directly observed therapy for highly active trial. J Pain Symptom Manage 2012; 43:161–171. antiretroviral therapy on virologic, immunologic, and 34. Fisher JD, Amico KR, Fisher WA, Cornman DH, Shuper PA, adherence outcomes: a meta-analysis and systematic review. Trayling C, et al. Computer-based intervention in HIV clinical J Acquir Immune Defic Syndr 2010; 54:167–179. care setting improves antiretroviral adherence: the LifeWin- 15. Rueda S, Park-Wyllie Laura Y, Bayoumi A, Tynan A-M, dows Project. AIDS Behav 2011; 15:1635–1646. Antoniou T, Rourke S, et al. Patient support and education 35. Hardy H, Kumar V, Doros G, Farmer E, Drainoni ML, Rybin D, for promoting adherence to highly active antiretroviral et al. Randomized controlled trial of a personalized cellular therapy for HIV/AIDS. Cochrane Database of Systematic phone reminder system to enhance adherence to antiretro- Reviews: John Wiley & Sons, Ltd; 2006. viral therapy. AIDS Patient Care STDS 2011; 25:153–161. 16. Haddad M, Inch C, Glazier RH, Wilkins AL, Urbshott G, 36. Holstad MM, DiIorio C, Kelley ME, Resnicow K, Sharma S. Bayoumi A, et al. Patient support and education for promoting Group motivational interviewing to promote adherence adherence to highly active antiretroviral therapy for HIV/ to antiretroviral medications and risk reduction behaviors AIDS. Cochrane Database Syst Rev 2000; 3:CD001442. in HIV infected women. AIDS Behav 2011; 15:885–896.

37. Kalichman SC, Cherry C, Kalichman MO, Amaral CM, 56. Murphy DA, Lu MC, Martin D, Hoffman D, Marelich WD. White D, Pope H, et al. Integrated behavioral intervention Results of a pilot intervention trial to improve antiretroviral to improve HIV/AIDS treatment adherence and reduce HIV adherence among HIV-positive patients. J Assoc Nurses AIDS transmission. Am J Public Health 2011; 101:531–538. Care 2002; 13:57–69. 38. Kalichman SC, Kalichman MO, Cherry C, Swetzes C, Amaral 57. Powell-Cope GM, White J, Henkelman EJ, Turner BJ. CM, White D, et al. Brief behavioral self-regulation counseling Qualitative and quantitative assessments of HAART adher- for HIV treatment adherence delivered by cell phone: an ence of substance-abusing women. AIDS Care 2003; 15:239– initial test of concept trial. AIDS Patient Care STDS 2011; 249. 25:303–310. 58. Pradier C, Bentz L, Spire B, Tourette-Turgis C, Morin M, 39. Leon A, Caceres C, Fernandez E, Chausa P, Martin M, Codina Souville M, et al. Efficacy of an educational and counseling C, et al. A new multidisciplinary home care telemedicine intervention on adherence to highly active antiretroviral system to monitor stable chronic human immunodeficiency therapy: French prospective controlled study. HIV Clin Trials virus-infected patients: a randomized study. PLoS One 2003; 4:121–131. 2011:6. 59. Rawlings MK, Thompson MA, Farthing CF, Brown LS, Racine J, 40. Pyne JM, Fortney JC, Curran GM, Tripathi S, Atkinson JH, Scott RC, et al. Impact of an educational program on efficacy Kilbourne AM, et al. Effectiveness of collaborative care for and adherence with a twice-daily lamivudine/zidovudine/ depression in human immunodeficiency virus clinics. Arch abacavir regimen in underrepresented HIV-infected patients. Intern Med 2011; 171:23–31. J Acquir Immune Defic Syndr 2003; 34:174–183. 41. Ramirez-Garcia P, Cote J. An individualized intervention 60. Rigsby MO, Rosen MI, Beauvais JE, Cramer JA, Rainey PM, to foster optimal antiretroviral treatment-taking behavior O’Malley SS, et al. Cue-dose training with monetary among persons living with HIV: a pilot randomized controlled reinforcement: pilot study of an antiretroviral adherence trial. J Assoc Nurses AIDS Care 2012; 23:220–232. intervention. J Gen Intern Med 2000; 15:841–847. 42. Ruiz I, Olry A, Lopez MA, Prada JL, Causse M. Prospective, 61. Safren SA, Otto MW, Worth JL, Salomon E, Johnson W, Mayer randomized, two-arm controlled study to evaluate two interven- K, et al. Two strategies to increase adherence to HIV anti- tions to improve adherence to antiretroviral therapy in Spain. 7th retroviral medication: life-steps and medication monitoring. ed. Spain: Ediciones Doyma, S.L., Spain; 2010. pp. 409–415. Behav Res Ther 2001; 39:1151–1162. 43. Sabin LL, DeSilva MB, Hamer DH, Xu K, Zhang J, Li T, et al. 62. Safren SA, Hendriksen ES, Desousa N, Boswell SL, Mayer KH. Using electronic drug monitor feedback to improve adher- Use of an on-line pager system to increase adherence to ence to antiretroviral therapy among HIV-positive patients in antiretroviral medications. AIDS Care 2003; 15:787–793. China. AIDS Behav 2010; 14:580–589. 63. Smith SR, Rublein JC, Marcus C, Brock TP, Chesney MA. 44. Safren SA, O’Cleirigh CM, Bullis JR, Otto MW, Stein MD, A medication self-management program to improve adher- Pollack MH. Cognitive behavioral therapy for adherence and ence to HIV therapy regimens. Patient Educ Couns 2003; depression (CBT-AD) in HIV-infected injection drug users: 50:187–199. a randomized controlled trial. J Consult Clin Psychol 2012; 64. Stenzel MS, McKenzie M, Mitty JA, Flanigan TP. Enhancing 80:404–415. adherence to HAART: a pilot program of modified directly 45. Uzma Q, Emmanuel F, Ather U, Zaman S. Efficacy of inter- observed therapy. AIDS Reader 2001; 11:317–319. ventions for improving antiretroviral therapy adherence in HIV/ 65. Tuldra A, Fumaz CR, Ferrer MJ, Bayes R, Arno A, Balague M, AIDS cases at PIMS, Islamabad. 6th ed. United States: SAGE et al. Prospective randomized two-arm controlled study to Publications Inc; 2011. pp. 373–383. determine the efficacy of a specific intervention to improve 46. Zubaran C, Michelim L, Medeiros G, May W, Foresti K, Madi long-term adherence to highly active antiretroviral therapy. JM. A randomized controlled trial of a protocol of interviews J Acquir Immune Defic Syndr 2000; 25:221–228. designed to improve adherence to antiretroviral medications 66. Berrien VM, Salazar JC, Reynolds E, McKay K, Group HIVMAI. in Southern Brazil. 6th ed. United Kingdom: Royal Society of Adherence to antiretroviral therapy in HIV-infected Medicine Press Ltd, UK; 2012. pp. 429–434. pediatric patients improves with home-based intensive 47. DiIorio C, Resnicow K, McDonnell M, Soet J, McCarty F, nursing intervention. AIDS Patient Care Stds 2004; 18:355– Yeager K. Using motivational interviewing to promote 363. adherence to antiretroviral medications: a pilot study. J Assoc 67. Bain-Brickley D, Butler LM, Kennedy GE, Rutherford GW. Nurses AIDS Care 2003; 14:52–62. Interventions to improve adherence to antiretroviral therapy 48. Amico KR, Harman JJ, Johnson BT. Efficacy of antiretroviral in children with HIV infection. Cochrane Database Syst Rev therapy adherence interventions: a research synthesis of 2011; 12:CD009513. trials, 1996 to 2004. J Acquir Immune Defic Syndr 2006; 68. Funck-Brentano I, Dalban C, Veber F, Quartier P, Hefez S, 41:285–297. Costagliola D, et al. Evaluation of a peer support group 49. Fairley CK, Levy R, Rayner CR, Allardice K, Costello K, Thomas therapy for HIV-infected adolescents. AIDS 2005; 19:1501– C, et al. Randomized trial of an adherence programme for 1508. clients with HIV. Int J STD AIDS 2003; 14:805–809. 69. Wamalwa DC, Farquhar C, Obimbo EM, Selig S, Mbori- 50. Goujard C, Bernard N, Sohier N, Peyramond D, Lancon F, Ngacha DA, Richardson BA, et al. Medication diaries Chwalow J, et al. Impact of a patient education program do not improve outcomes with highly active antiretroviral on adherence to HIV medication: a randomized clinical trial. therapy in Kenyan children: a randomized clinical trial. J Int J Acquir Immune Defic Syndr 2003; 34:191–194. AIDS Soc 2009; 12:8. 51. Lyon ME, Trexler C, Akpan-Townsend C, Pao M, Selden K, 70. Cantrell RA, Sinkala M, Megazinni K, Lawson-Marriott S, Fletcher J, et al. A family group approach to increasing Washington S, Chi BH, et al. A pilot study of food adherence to therapy in HIV-infected youths: results of a supplementation to improve adherence to antiretroviral ther- pilot project. AIDS Patient Care STDS 2003; 17:299–308. apy among food-insecure adults in Lusaka, Zambia. J Acquir 52. Mann T. Effects of future writing and optimism on health Immune Defic Syndr 2008; 49:190–195. behaviors in HIV-infected women. Ann Behav Med 2001; 71. Barnighausen T, Chaiyachati K, Chimbindi N, Peoples A, 23:26–33. Haberer J, Newell ML. Interventions to increase antiretro- 53. Margolin A, Avants SK, Warburton LA, Hawkins KA, Shi J. viral adherence in sub-Saharan Africa: a systematic review A randomized clinical trial of a manual-guided risk reduction of evaluation studies. Lancet Infect Dis 2011; 11:942– intervention for HIV-positive injection drug users. Health 951. Psychol 2003; 22:223–228. 72. Chang LW, Kagaayi J, Nakigozi G, Ssempijja V, Packer AH, 54. McPherson-Baker S, Malow RM, Penedo F, Jones DL, Serwadda D, et al. Effect of peer health workers on AIDS care Schneiderman N, Klimas NG. Enhancing adherence to com- in Rakai, Uganda: a cluster-randomized trial. PLoS One 2010; bination antiretroviral therapy in nonadherent HIV-positive 5:e10923. men. AIDS Care 2000; 12:399–404. 73. Idoko JA, Agbaji O, Agaba P, Akolo C, Inuwa B, Hassan Z, et al. 55. Molassiotis A, Lopez-Nahas V, Chung WY, Lam SW. A pilot Direct observation therapy-highly active antiretroviral study of the effects of a behavioural intervention on treatment therapy in a resource-limited setting: the use of community adherence in HIV-infected patients. AIDS Care 2003; 15:125– treatment support can be effective. Int J STD AIDS 2007; 135. 18:760–763.

74. Kabore I, Bloem J, Etheredge G, Obiero W, Wanless S, Doykos 93. Johnson MO, Dilworth SE, Taylor JM, Neilands TB. Improv- P, et al. The effect of community-based support services on ing coping skills for self-management of treatment side clinical efficacy and health-related quality of life in HIV/AIDS effects can reduce antiretroviral medication nonadherence patients in resource-limited settings in sub-Saharan Africa. among people living with HIV. Ann Behav Med 2011; 41:83– AIDS Patient Care STDS 2010; 24:581–594. 91. 75. Lester RT, Ritvo P, Mills EJ, Kariri A, Karanja S, Chung MH, 94. Weiss SM, Tobin JN, Antoni M, Ironson G, Ishii M, Vaughn A, et al. Effects of a mobile phone short message service et al. Enhancing the health of women living with HIV: the on antiretroviral treatment adherence in Kenya (WelTel SMART/EST Women’s Project. Int J Womens Health 2011; Kenya1): a randomised trial. Lancet 2010; 376:1838–1845. 3:63–77. 76. Mugusi F, Mugusi S, Bakari M, Hejdemann B, Josiah R, Janabi 95. Jaffar S, Amuron B, Foster S, Birungi J, Levin J, Namara G, et al. M, et al. Enhancing adherence to antiretroviral therapy at the Rates of virological failure in patients treated in a home-based HIV clinic in resource constrained countries; the Tanzanian versus a facility-based HIV-care model in Jinja, southeast experience. Trop Med Int Health 2009; 14:1226–1232. Uganda: a cluster-randomised equivalence trial. Lancet 77. Nachega JB, Chaisson RE, Goliath R, Efron A, Chaudhary MA, 2009; 374:2080–2089. Ram M, et al. Randomized controlled trial of trained patient- 96. Wall TL, Sorensen JL, Batki SL, Delucchi KL, London JA, nominated treatment supporters providing partial directly Chesney MA. Adherence to zidovudine (AZT) among HIV- observed antiretroviral therapy. AIDS 2010; 24:1273–1280. infected methadone patients: a pilot study of supervised 78. Ndekha M, van Oosterhout JJ, Saloojee H, Pettifor J, Manary M. therapy and dispensing compared to usual care. Drug Alcohol Nutritional status of Malawian adults on antiretroviral ther- Depend 1995; 37:261–269. apy 1 year after supplementary feeding in the first 3 months of 97. Fogarty L, Roter D, Larson S, Burke J, Gillespie J, Levy R. therapy. Trop Med Int Health 2009; 14:1059–1063. Patient adherence to HIV medication regimens: a review 79. Ndekha MJ, van Oosterhout JJ, Zijlstra EE, Manary M, Saloojee of published and abstract reports. Patient Educ Couns H, Manary MJ. Supplementary feeding with either ready-to- 2002; 46:93–108. use fortified spread or corn-soy blend in wasted adults starting 98. Knobel H, Carmona A, Lopez JL, Gimeno JL, Saballs P, antiretroviral therapy in Malawi: randomised, investigator Gonzalez A, et al. [Adherence to very active antiretro- blinded, controlled trial. Br Med J 2009; 338:b1867. viral treatment: impact of individualized assessment]. Enferm 80. Pearson CR, Micek MA, Simoni JM, Hoff PD, Matediana E, Infecc Microbiol Clin 1999; 17:78–81. Martin DP, et al. Randomized control trial of peer-delivered, 99. Altice FL, Maru DS, Bruce RD, Springer SA, Friedland GH. modified directly observed therapy for HAART in Mozambique. Superiority of directly administered antiretroviral therapy J Acquir Immune Defic Syndr 2007; 46:238–244. over self-administered therapy among HIV-infected drug 81. Pienaar DH, Myer L, Cleary S, Coetzee D, Michaels D, Cloete users: a prospective, randomized, controlled trial. Clin Infect K, et al. Models of Care for Antiretroviral Service Delivery. Cape Dis 2007; 45:770–778. Town: University of Capetown; 2006. 100. Maru DS, Bruce RD, Walton M, Springer SA, Altice FL. 82. Pop-Eleches C, Thirumurthy H, Habyarimana JP, Zivin JG, Persistence of virological benefits following directly adminis- Goldstein MP, de Walque D, et al. Mobile phone technologies tered antiretroviral therapy among drug users: results from a improve adherence to antiretroviral treatment in a resource- randomized controlled trial. J Acquir Immune Defic Syndr limited setting: a randomized controlled trial of text message 2009; 50:176–181. reminders. AIDS 2011; 25:825–834. 101. Hart JE, Jeon CY, Ivers LC, Behforouz HL, Caldas A, Drobac 83. Roux SM. Diary cards: preliminary evaluation of an interven- PC, et al. Effect of directly observed therapy for highly tion tool for improving adherence to antiretroviral therapy and active antiretroviral therapy on virologic, immunologic, TB preventive therapy in people living with HIV/AIDS [MPH and adherence outcomes: a meta-analysis and systematic thesis]. University of the Western Cape; 2004. review. J Acquir Immune Defic Syndr 2010; 54:167–179. 84. Sarna A, Luchters S, Geibel S, Chersich MF, Munyao P, Kaai S, 102. Gross R, Tierney C, Andrade A, Lalama C, Rosenkranz S, et al. Short- and long-term efficacy of modified directly Eshleman SH, et al. Modified directly observed antiretroviral observed antiretroviral treatment in Mombasa, Kenya: a rando- therapy compared with self-administered therapy in treat- mized trial. J Acquir Immune Defic Syndr 2008; 48:611–619. ment-naive HIV-1-infected patients: a randomized trial. Arch 85. Sherr KH, Micek MA, Gimbel SO, Gloyd SS, Hughes JP, John- Intern Med 2009; 169:1224–1232. Stewart GC, et al. Quality of HIV care provided by nonphy- 103. Lucas GM, Mullen BA, Weidle PJ, Hader S, McCaul ME, sician clinicians and physicians in Mozambique: a retrospec- Moore RD. Directly administered antiretroviral therapy in tive cohort study. AIDS 2010; 24 (Suppl 1):S59–66. methadone clinics is associated with improved HIV treatment 86. Stubbs BA, Micek MA, Pfeiffer JT, Montoya P, Gloyd S. outcomes, compared with outcomes among concurrent Treatment partners and adherence to HAART in central comparison groups. Clin Infect Dis 2006; 42:1628–1635. Mozambique. AIDS Care 2009; 21:1412–1419. 104. Macalino GE, Hogan JW, Mitty JA, Bazerman LB, Delong AK, 87. Taiwo BO, Idoko JA, Welty LJ, Otoh I, Job G, Iyaji PG, et al. Loewenthal H, et al. A randomized clinical trial of Assessing the viorologic and adherence benefits of patient- community-based directly observed therapy as an adherence selected HIV treatment partners in a resource-limited setting. intervention for HAART among substance users. AIDS 2007; J Acquir Immune Defic Syndr 2010; 54:85–92. 21:1473–1477. 88. Thurman TR, Haas LJ, Dushimimana A, Lavin B, Mock N. 105. Munoz M, Finnegan K, Zeladita J, Caldas A, Sanchez E, Evaluation of a case management program for HIV clients in Callacna M, et al. Community-based DOT-HAART accompa- Rwanda. AIDS Care 2010; 22:759–765. niment in an urban resource-poor setting. AIDS Behav 2010; 89. Torpey KE, Kabaso ME, Mutale LN, Kamanga MK, Mwango AJ, 14:721–730. Simpungwe J, et al. Adherence support workers: a way to 106. Tinoco I, Giron-Gonzalez JA, Gonzalez-Gonzalez MT, address human resource constraints in antiretroviral treat- Vergara de Campos A, Rodriguez-Felix L, Serrano A, et al. ment programs in the public health setting in Zambia. PLoS Efficacy of directly observed treatment of HIV infection: One 2008; 3:e2204. experience in AIDS welfare homes. Eur J Clin Microbiol Infect 90. Antoni MH, Carrico AW, Duran RE, Spitzer S, Penedo F, Dis 2004; 23:331–335. Ironson G, et al. Randomized clinical trial of cognitive 107. Wohl AR, Garland WH, Valencia R, Squires K, Witt MD, behavioral stress management on human immunodeficiency Kovacs A, et al. A randomized trial of directly administered virus viral load in gay men treated with highly active anti- antiretroviral therapy and adherence case management inter- retroviral therapy. Psychosom Med 2006; 68:143–151. vention. Clin Infect Dis 2006; 42:1619–1627. 91. Brown JL, Vanable PA. Stress management interventions for 108. Andrade AS, McGruder HF, Wu AW, Celano SA, Skolasky RL HIV-infected individuals: review of recent intervention appro- Jr, Selnes OA, et al. A programmable prompting device aches and directions for future research. 1st ed. New Zealand: improves adherence to highly active antiretroviral therapy DOVE Medical Press Ltd., New Zealand; 2011. pp. 95–106. in HIV-infected subjects with memory impairment. Clin Infect 92. Creswell JD, Myers HF, Cole SW, Irwin MR. Mindfulness Dis 2005; 41:875–882. meditation training effects on CD4R T lymphocytes in 109. Haynes RB, Ackloo E, Sahota N, McDonald HP, Yao X. HIV-1 infected adults: a small randomized controlled trial. Interventions for enhancing medication adherence. Cochrane Brain Behav Immun 2009; 23:184–188. Database Syst Rev 2008; 2:CD000011.

110. Collier AC, Ribaudo H, Mukherjee AL, Feinberg J, Fischl MA, 127. Holzemer WL, Bakken S, Portillo CJ, Grimes R, Welch J, Chesney M. A randomized study of serial telephone call Wantland D, et al. Testing a nurse-tailored HIV medication support to increase adherence and thereby improve virologic adherence intervention. Nurs Res 2006; 55:189–197. outcome in persons initiating antiretroviral therapy. J Infect 128. Javanbakht M, Prosser P, Grimes T, Weinstein M, Farthing C. Dis 2005; 192:1398–1406. Efficacy of an individualized adherence support program with 111. Remien RH, Stirratt MJ, Dolezal C, Dognin JS, Wagner GJ, contingent reinforcement among nonadherent HIV-positive Carballo-Dieguez A, et al. Couple-focused support to improve patients: results from a randomized trial. J Int Assoc Phys AIDS HIV medication adherence: a randomized controlled trial. Care (Chic) 2006; 5:143–150. AIDS 2005; 19:807–814. 129. Johnson MO, Charlebois E, Morin SF, Remien RH, Chesney 112. Samet JH, Horton NJ, Meli S, Dukes K, Tripps T, Sullivan L, MA. Effects of a behavioral intervention on antiretroviral et al. A randomized controlled trial to enhance antiretroviral medication adherence among people living with HIV: the therapy adherence in patients with a history of alcohol healthy living project randomized controlled study. J Acquir problems. Antivir Ther 2005; 10:83–93. Immune Defic Syndr 2007; 46:574–580. 113. van Servellen G, Nyamathi A, Carpio F, Pearce D, Garcia- 130. Jones DL, McPherson-Baker S, Lydston D, Camille J, Brondolo Teague L, Herrera G, et al. Effects of a treatment adherence E, Tobin JN, et al. Efficacy of a group medication adherence enhancement program on health literacy, patient-provider intervention among HIV positive women: the SMART/EST relationships, and adherence to HAART among low-income Women’s Project. AIDS Behav 2007; 11:79–86. HIV-positive Spanish-speaking Latinos. AIDS Patient Care 131. Koenig LJ, Pals SL, Bush T, Pratt Palmore M, Stratford D, STDS 2005; 19:745–759. Ellerbrock TV. Randomized controlled trial of an intervention 114. Weber R, Christen L, Christen S, Tschopp S, Znoj H, to prevent adherence failure among HIV-infected patients Schneider C, et al. Effect of individual cognitive behaviour initiating antiretroviral therapy. Health Psychol 2008; 27: intervention on adherence to antiretroviral therapy: prospec- 159–169. tive randomized trial. Antivir Ther 2004; 9:85–95. 132. Levin TR, Klibanov OM, Axelrod P, Finley GL, Gray A, 115. DiIorio C, Mccarty F, Resnicow K, Holstad MM, Soet J, Holdsworth C, et al. A randomized trial of educational Yeager K, et al. Using motivational interviewing to promote materials, pillboxes, and mailings to improve adherence with adherence to antiretroviral medications: a randomized antiretroviral therapy in an inner city HIV clinic. J Clin Out- controlled study. AIDS Care 2008; 20:273–283. comes Manag 2006; 13:217–221. 116. Hill S, Kavookjian J. Motivational interviewing as a behavioral 133. Ma M, Brown BR, Coleman M, Kibler JL, Loewenthal H, intervention to increase HAART adherence in patients Mitty JA. The feasibility of modified directly observed therapy who are HIV-positive: a systematic review of the literature. for HIV-seropositive African American substance users. AIDS AIDS Care 2012; 24:583–592. Patient Care STDS 2008; 22:139–146. 117. Parsons JT, Golub SA, Rosof E, Holder C. Motivational 134. Milam J, Richardson JL, McCutchan A, Stoyanoff S, Weiss J, interviewing and cognitive-behavioral intervention to im- Kemper C, et al. Effect of a brief antiretroviral adherence prove HIV medication adherence among hazardous drinkers: intervention delivered by HIV care providers. J Acquir Im- a randomized controlled trial. J Acquir Immune Defic Syndr mune Defic Syndr 2005; 40:356–363. 2007; 46:443–450. 135. Parsons JT, Rosof E, Punzalan JC, Di Maria L. Integration 118. Mitty JA, Macalino GE, Bazerman LB, Loewenthal HG, Hogan of motivational interviewing and cognitive behavioral JW, MacLeod CJ, et al. The use of community-based modified therapy to improve HIV medication adherence and directly observed therapy for the treatment of HIV-infected reduce substance use among HIV-positive men and women: persons. JAcquirImmuneDeficSyndr2005; 39:545–550. results of a pilot project. AIDS Patient Care STDS 2005; 19: 119. Kenya S, Chida N, Symes S, Shor-Posner G. Can community 31–39. health workers improve adherence to highly active anti- 136. Reynolds NR, Testa MA, Su M, Chesney MA, Neidig JL, Frank I, retroviral therapy in the USA? A review of the literature. et al. Telephone support to improve antiretroviral medication HIV Med 2011; 12:525–534. adherence: a multisite, randomized controlled trial. J Acquir 120. Purcell DW, Latka MH, Metsch LR, Latkin CA, Gomez CA, Immune Defic Syndr 2008; 47:62–68. Mizuno Y, et al. Results from a randomized controlled trial of 137. Rosen MI, Dieckhaus K, McMahon TJ, Valdes B, Petry NM, a peer-mentoring intervention to reduce HIV transmission Cramer J, et al. Improved adherence with contingency and increase access to care and adherence to HIV medica- management. AIDS Patient Care STDS 2007; 21:30–40. tions among HIV-seropositive injection drug users. J Acquir 138. Sorensen JL, Haug NA, Delucchi KL, Gruber V, Kletter E, Immune Defic Syndr 2007; 46 (Suppl 2):S35–47. Batki SL, et al. Voucher reinforcement improves medication 121. Simoni JM, Pantalone DW, Plummer MD, Huang B. A rando- adherence in HIV-positive methadone patients: a randomized mized controlled trial of a peer support intervention targeting trial. Drug Alcohol Depend 2007; 88:54–63. antiretroviral medication adherence and depressive sympto- 139. Wagner GJ, Kanouse DE, Golinelli D, Miller LG, Daar ES, Witt matology in HIV-positive men and women. Health Psychol MD, et al. Cognitive-behavioral intervention to enhance 2007; 26:488–495. adherence to antiretroviral therapy: a randomized controlled 122. Visnegarwala F, Rodriguez-Barradass MC, Graviss EA, Caprio trial (CCTG 578). AIDS 2006; 20:1295–1302. M, Nykyforchyn M, Laufman L. Community outreach with 140. Jones DL, Ishii M, LaPerriere A, Stanley H, Antoni M, Ironson weekly delivery of antiretroviral drugs compared to cogni- G, et al. Influencing medication adherence among women tive-behavioural healthcare team-based approach to improve with AIDS. AIDS Care 2003; 15:463–474. adherence among indigent women newly starting HAART. 141. Manias E, Williams A. Medication adherence in people of AIDS Care 2006; 18:332–338. culturally and linguistically diverse backgrounds: a meta- 123. Williams AB, Fennie KP, Bova CA, Burgess JD, Danvers KA, analysis. Ann Pharmacother 2010; 44:964–982. Dieckhaus KD. Home visits to improve adherence to highly 142. Rathbun RC, Farmer KC, Stephens JR, Lockhart SM. Impact of active antiretroviral therapy: a randomized controlled trial. an adherence clinic on behavioral outcomes and virologic J Acquir Immune Defic Syndr 2006; 42:314–321. response in treatment of HIV infection: a prospective, 124. Golin CE, Earp J, Tien HC, Stewart P, Porter C, Howie L. randomized, controlled pilot study. Clin Ther 2005; 27: A 2-arm, randomized, controlled trial of a motivational 199–209. interviewing-based intervention to improve adherence to 143. van Servellen G, Carpio F, Lopez M, Garcia-Teague L, Herrera antiretroviral therapy (ART) among patients failing or initiat- G, Monterrosa F, et al. Program to enhance health literacy and ing ART. J Acquir Immune Defic Syndr 2006; 42:42–51. treatment adherence in low-income HIV-infected Latino men 125. Leeman J, Chang YK, Lee EJ, Voils CI, Crandell J, Sandelowski and women. AIDS Patient Care STDS 2003; 17:581–594. M. Implementation of antiretroviral therapy adherence inter- 144. Wyatt GE, Longshore D, Chin D, Carmona JV, Loeb TB, ventions: a realist synthesis of evidence. J Adv Nurs 2010; Myers HF, et al. The efficacy of an integrated risk reduction 66:1915–1930. intervention for HIV-positive women with child sexual abuse 126. Harwell JI, Flanigan TP, Mitty JA, Macalino GE, Caliendo AM, histories. AIDS Behav 2004; 8:453–462. Ingersoll J, et al. Directly observed antiretroviral therapy to 145. Levy RW, Rayner CR, Fairley CK, Kong DC, Mijch A, Costello K, reduce genital tract and plasma HIV-1 RNA in women with et al. Multidisciplinary HIV adherence intervention: a rando- poor adherence. AIDS 2003; 17:1990–1993. mized study. AIDS Patient Care STDS 2004; 18:728–735.

146. Mannheimer SB, Morse E, Matts JP, Andrews L, Child C, 162. Feaster DJ, Mitrani VB, Burns MJ, McCabe BE, Brincks AM, Schmetter B, et al. Sustained benefit from a long-term Rodriguez AE, et al. A randomized controlled trial of struc- antiretroviral adherence intervention. Results of a large tural ecosystems therapy for HIV medication adherence and randomized clinical trial. J Acquir Immune Defic Syndr substance abuse relapse prevention. Drug Alcohol Depend 2006; 43 (Suppl 1):S41–47. 2010; 111:227–234. 147. Saberi P, Johnson MO. Technology-based self-care methods 163. Wechsberg WM, Golin C, El-Bassel N, Hopkins J, Zule W. of improving antiretroviral adherence: a systematic review. Current interventions to reduce sexual risk behaviors and PLoS One 2011:6. crack cocaine use among HIV-infected individuals. Curr HIV/ 148. Murphy DA, Marelich WD, Rappaport NB, Hoffman D, AIDS Rep 2012. Farthing C. Results of an antiretroviral adherence inter- 164. Ingersoll KS, Farrell-Carnahan L, Cohen-Filipic J, Heckman CJ, vention: STAR (Staying Healthy: Taking Antiretrovirals Ceperich SD, Hettema J, et al. A pilot randomized clinical trial Regularly). J Int Assoc Physicians AIDS Care (Chic) 2007; of two medication adherence and drug use interventions 6:113–124. for HIVR crack cocaine users. Drug Alcohol Depend 2011; 149. Simoni JM, Huh D, Frick PA, Pearson CR, Andrasik MP, 116:177–187. Dunbar PJ, et al. Peer support and pager messaging to promote 165. Page J, Weber R, Somaini B, Nostlinger C, Donath K, Jaccard R. antiretroviral modifying therapy in Seattle: a randomized Quality of generalist vs. specialty care for people with HIV on controlled trial. J Acquir Immune Defic Syndr 2009; 52: antiretroviral treatment: a prospective cohort study. HIV Med 465–473. 2003; 4:276–286. 150. Simoni JM, Chen WT, Huh D, Fredriksen-Goldsen KI, 166. Wong WC, Luk CW, Kidd MR. Is there a role for primary care Pearson C, Zhao H, et al. A preliminary randomized clinicians in providing shared care in HIV treatment? A syste- controlled trial of a nurse-delivered medication adherence matic literature review. Sex Transm Infect 2012; 88:125–131. intervention among HIV-positive outpatients initiating anti- 167. Igumbor JO, Scheepers E, Ebrahim R, Jason A, Grimwood A. retroviral therapy in Beijing, China. AIDS Behav 2011; An evaluation of the impact of a community-based adherence 15:919–929. support programme on ART outcomes in selected government 151. Wu AW, Snyder CF, Huang IC, Skolasky R, McGruder HF, HIV treatment sites in South Africa. AIDS Care 2011; 23:231– Celano SA, et al. A randomized trial of the impact of a 236. programmable medication reminder device on quality of life 168. Wouters E, Van Damme W, van Rensburg D, Masquillier C, in patients with AIDS. AIDS Patient Care STDS 2006; 20:773– Meulemans H. Impact of community-based support services 781. on antiretroviral treatment programme delivery and out- 152. Frick P, Tapia K, Grant P, Novotny M, Kerzee J. The effect of a comes in resource-limited countries: a synthetic review. multidisciplinary program on HAART adherence. AIDS Pa- BMC Health Serv Res 2012; 12:194. tient Care STDS 2006; 20:511–524. 169. Kunutsor S, Walley J, Katabira E, Muchuro S, Balidawa H, 153. Saberi P, Dong BJ, Johnson MO, Greenblatt RM, Cocohoba JM. Namagala E, et al. Improving clinic attendance and adherence The impact of HIV clinical pharmacists on HIV treatment to antiretroviral therapy through a treatment supporter outcomes: a systematic review. Patient Prefer Adherence intervention in Uganda: a randomized controlled trial. AIDS 2012; 6:297–322. Behav 2011; 15:1795–1802. 154. Hirsch JD, Gonzales M, Rosenquist A, Miller TA, Gilmer TP, 170. Rich ML, Miller AC, Niyigena P, Franke MF, Niyonzima JB, Best BM. Antiretroviral therapy adherence, medication use, Socci A, et al. Excellent clinical outcomes and high retention and healthcare costs during 3 years of a community pharmacy in care among adults in a community-based HIV treatment medication therapy management program for Medi-Cal program in rural Rwanda. J Acquir Immune Defic Syndr 2012; beneficiaries with HIV/AIDS. J Manag Care Pharm 2011; 59:e35–e42. 17:213–223. 171. Bangsberg DR, Mills EJ. Long-term adherence to antiretroviral 155. Horberg MA, Hurley LB, Silverberg MJ, Kinsman CJ, therapy in resource-limited settings: a bitter pill to swallow. Quesenberry CP. Effect of clinical pharmacists on utilization Antivir Ther 2013; 18:25–28. of and clinical response to antiretroviral therapy. J Acquir 172. Mills EJ, Lester R, Ford N. Promoting long term adherence Immune Defic Syndr 2007; 44:531–539. to antiretroviral treatment. Br Med J 2012; 344:e4173. 156. March K, Mak M, Louie SG. Effects of pharmacists’ inter- 173. Bastard M, Fall MB, Laniece I, Taverne B, Desclaux A, ventions on patient outcomes in an HIV primary care clinic. Ecochard R, et al. Revisiting long-term adherence to highly Am J Health Syst Pharm 2007; 64:2574–2578. active antiretroviral therapy in Senegal using latent class 157. Pirkle CM, Boileau C, Nguyen VK, Machouf N, Ag-Abouba- analysis. J Acquir Immune Defic Syndr 2011; 57:55–61. crine S, Niamba PA, et al. Impact of a modified directly 174. Cambiano V, Lampe FC, Rodger AJ, Smith CJ, Geretti AM, administered antiretroviral treatment intervention on viro- Lodwick RK, et al. Long-term trends in adherence to antire- logical outcome in HIV-infected patients treated in Burkina troviral therapy from start of HAART. AIDS 2010; 24:1153– Faso and Mali. HIV Med 2009; 10:152–156. 1162. 158. Rotheram-Borus MJ, Swendeman D, Comulada WS, Weiss RE, 175. Chaiyachati K, Hirschhorn LR, Tanser F, Newell ML, Lee M, Lightfoot M. Prevention for substance-using Barnighausen T. Validating five questions of antiretroviral HIV-positive young people: telephone and in-person delivery. nonadherence in a public-sector treatment program in rural J Acquir Immune Defic Syndr 2004; 37 (Suppl 2):S68– South Africa. AIDS Patient Care STDS 2011; 25:163–170. 77. 176. Ba¨rnighausen T, Salomon JA, Sangrujee N. HIV treatment as 159. Simoni JM, Pearson CR, Pantalone DW, Marks G, Crepaz N. prevention: issues in economic evaluation. PLoS Med 2012; Efficacy of interventions in improving highly active antire- 9:e1001263. troviral therapy adherence and HIV-1 RNA viral load: a meta- 177. Flexner C, Plumley B, Brown Ripin D. Treatment optimiza- analytic review of randomized controlled trials. J Acquir tion: an outline for future success. Curr Opin HIV AIDS 2013; Immune Defic Syndr 2006; 43:S23–S35. 8:523–527. 160. Byron E, Gillespie S, Nangami M. Integrating nutrition secur- 178. Aldir I, Horta A, Serrado M. Single-tablet regimens in HIV: ity with treatment of people living with HIV: lessons from does it really make a difference? Curr Med Res Opin 2014; Kenya. Food Nutr Bull 2008; 29:87–97. 30:89–97. 161. Tirivayi N, Groot W. Health and welfare effects of integrating 179. Nachega JB, Parienti JJ, Uthman OA, Gross R, Dowdy DW, Sax AIDS treatment with food assistance in resource constrained PE, et al. Lower Pill Burden and Once-Daily Antiretroviral settings: a systematic review of theory and evidence. Soc Sci Treatment Regimens for HIV Infection: A Meta-Analysis of Med 2011; 73:685–692. Randomized Controlled Trials. Clin Infectious Dis 2014. [Epub].