Interventions to increase antiretroviral adherence in sub-Saharan Africa: a systematic review of evaluation studies

Review

Interventions to increase antiretroviral adherence in sub-Saharan Africa: a systematic review of evaluation studies Till Bärnighausen, Krisda Chaiyachati, Natsayi Chimbindi, Ashleigh Peoples, Jessica Haberer, Marie-Louise Newell Lancet Infect Dis 2011; 11: 942–51 Published Online October 25, 2011 DOI:10.1016/S14733099(11)70181-5 Africa Centre for Health and Population Studies, University of KwaZulu-Natal, Mtubatuba, South Africa (T Bärnighausen MD, N Chimbindi MSc, Prof M-L Newell PhD); Department of Global Health and Population, Harvard School of Public Health, Boston, MA, USA (T Bärnighausen, K Chaiyachati MD); University of Michigan Medical School, Ann Arbor, MI, USA (K Chaiyachati); Department of Epidemiology, University of Michigan School of Public Health, Ann Arbor, MI, USA (A Peoples MSc); Massachusetts General Hospital, Center for Global Health, Boston, MA, USA (J Haberer MD); and Centre for Paediatric Epidemiology and Biostatistics, UCL Institute of Child Health, London, UK (Prof M-L Newell) Correspondence to: Dr T Bärnighausen, Department of Global Health and Population, Harvard School of Public Heath, Boston, MA 02115, USA [email protected]

942

The success of potent antiretroviral treatment for HIV infection is primarily determined by adherence. We systematically review the evidence of effectiveness of interventions to increase adherence to antiretroviral treatment in sub-Saharan Africa. We identified 27 relevant reports from 26 studies of behavioural, cognitive, biological, structural, and combination interventions done between 2003 and 2010. Despite study diversity and limitations, evidence suggests that treatment supporters, directly observed therapy, mobile-phone text messages, diary cards, and food rations can effectively increase adherence in sub-Saharan Africa. However, some interventions are unlikely to have large or lasting effects, and others are effective only in specific settings. These findings emphasise the need for more research, particularly for randomised controlled trials, to examine the effect of context and specific features of intervention content on effectiveness. Future work should assess intervention targeting and selection of interventions based on behavioural theories relevant to sub-Saharan Africa.

Introduction Antiretroviral treatment (ART) can substantially reduce morbidity and mortality in people with HIV.1–6 However, the clinical effectiveness of ART depends crucially on adherence.7–10 Studies done before 2005 showed that the greatest effectiveness could only be achieved if patients took at least 95% of prescribed antiretroviral doses.7,11,12 More recent studies suggest that treatment with some potent ART regimens, such as those based on ritonavirboosted protease inhibitors and non-nucleoside reversetranscriptase inhibitors, can achieve viral suppression at lower adherence,13–15 and that the adherence needed to prevent viral rebound decreases with duration of suppression.15 However, only sustained high levels of adherence will ensure that life-extending benefits of ART are maximised and risk of viral resistance is minimised,16 and low adherence is the most common reason why potential treatment benefits are not achieved or sustained.17 Poor adherence also substantially increases the health-care costs associated with treatment of HIV in both developing and developed countries.18,19 Additionally, high adherence is essential for the reduction of HIV transmission in treatment-as-prevention approaches.20 Many national governments in sub-Saharan Africa, with support from international agencies and donors, are striving to provide ART to all people in need.21 Of the more than 5 million people worldwide who were receiving ART at the end of 2009, almost four-fifths were in subSaharan Africa,22 where ART adherence might be low.23 Results of a 2006 meta-analysis24 of ART adherence investigations showed that, on average, 23% of patients in studies from sub-Saharan Africa did not achieve adequate adherence, with the proportion of non-adherent patients ranging from 2% to 70% across the primary studies included in the analysis. These findings suggest a need to improve adherence substantially in many settings in sub-Saharan Africa.25 Moreover, most treatment programmes in sub-Saharan Africa have only been enrolling patients for a few years. Experience from developed countries has shown that adherence falls with time on

ART,26–28 and recent studies suggest similar trends in subSaharan Africa.29 Interventions to increase ART adherence in sub-Saharan Africa are clearly needed. Most studies of adherence interventions are from developed countries, and many have been previously reviewed.25,30–38 Table 1 provides an overview of categories of interventions to improve ART adherence that have been investigated in the developed world.30,31,39 The evidence from these investigations might, however, have limited relevance to sub-Saharan Africa because the effectiveness of interventions is likely to depend on the context in which they are implemented. For example, adherence interventions in developed countries are usually provided by general nurses, pharmacists, or physicians,30,34 while in sub-Saharan Africa health workers specifically trained to care for patients with HIV are commonly responsible for the monitoring and support of ART patients, often with little involvement from physicians.40 Further, interventions in developed countries might be based on theories of behaviour that are not valid in sub-Saharan Africa, and those that have been specifically tailored to adherence support of specific subpopulations (such as men who have sex with men41 or injection drug users30) might be of little use in the generalised HIV epidemics of subSaharan Africa. Resource-intensive interventions directed towards the individual (eg, cognitive behavioural therapy42) could be difficult to implement in sub-Saharan Africa because of large numbers of patients, restricted resources, and the public health approach to treatment. Therefore, evidence from sub-Saharan Africa is important to inform the design and implementation of ART adherence interventions in the region. We present findings from a systematic review of studies investigating the effectiveness of ART adherence interventions in subSaharan Africa.

Methods Search strategy We systematically searched PubMed for studies that were published before 31 Jan, 2011, and evaluated www.thelancet.com/infection Vol 11 December 2011

Review

interventions to improve ART adherence in subSaharan Africa. To identify articles for this Review, we combined two broad search themes with the Boolean operator “and”. The first search theme—ART— combined the Medical Subject Headings (MeSH) terms “antiretroviral therapy, highly active” and “anti-HIV agents”, and the free-text word “antiretroviral”, with the Boolean operator “or”. The second theme—adherence— combined the MeSH term “patient compliance” and the free-text word “adherence”, with “or”. All narrower MeSH terms that are categorised below the selected term in the MeSH hierarchy were included in the PubMed search “((“antiretroviral therapy, highly active”[MeSH] OR “anti-HIV agents”[MeSH] OR antiretroviral) AND (“patient compliance”[MeSH] OR adherence))”. Additionally, we searched the reference lists of all reports included in the final Review and all articles that were excluded because they were review articles, editorials, or commentaries. We further searched the following conference databases: the Conference on Retroviruses and Opportunistic Infections (up to San Francisco, CA, USA, in February 2010),43 the International AIDS Society (up to Cape Town, South Africa, in July, 2009),44 the International Conference on HIV Treatment Adherence (up to Miami, FL, USA, in May, 2010),45 and the mHealth Summit (up to Washington, DC, USA, in November, 2010).46 Furthermore, relevant publications in the grey literature47 (research reports, working papers, and dissertations)48 were searched by typing of the terms “antiretroviral”, “therapy”, “adherence”, and “intervention” into Google’s internet search engine. Our Review conformed to the PRISMA checklist for systematic reviews.49

Study selection and eligibility criteria Three investigators (KC, NC, and AP) independently screened every abstract identified by the PubMed search. The same reviewers used inclusion and exclusion criteria to assess the full eligibility of conference abstracts, grey literature, and non-peer reviewed articles, as well as the studies identified through the screening of abstracts in PubMed. To be included in the final Review, studies could report adherence as the primary or secondary outcome, or simply report adherence measurements in the presence of an intervention. For the purposes of this Review, we restricted the definition of adherence to medication adherence—ie, the extent to which a patient takes a drug in the way intended by the health-care provider. We did not include studies that investigated the related but distinct concepts of adherence to ART appointments50 or retention within ART programmes.51,52 We did not exclude studies on the basis of the type of measurement used to assess ART adherence, but restricted our Review to studies with a control or comparison group (ie, a group that did not receive the intervention). We did not apply any other study-design exclusion criteria; www.thelancet.com/infection Vol 11 December 2011

Definition

Examples

Behavioural

Affect ART adherence through direct behaviour modification

Reminder devices (eg, 7 day pill boxes, alarms, mobile-phone text messages or pager messages) Cues for remembering dose times Cash incentives DOT

Cognitive

Affect ART adherence through teaching, clarification, or instruction

Media education materials (eg, audio, video, or reading materials) Group education Education of individual patients

Affective

Affect ART adherence through emotional support

Peer support Treatment with antidepressants Counselling

Biological

Affect ART adherence through improved physical ability to take ART

Food rations Vitamin or micronutrient supplements

Structural

Affect ART adherence through changes in the delivery structure or through additional service structures

Delivering ART in community centres Income-generating activities for ART patients Community mobilisation

Combination Use of a combination of one or more of the above intervention categories

Patient information, behavioural adherence strategies, and peer support

ART=antiretroviral treatment. DOT=directly observed therapy.

Table 1: Categories of interventions to improve ART adherence

randomised controlled trials, prospective and retrospective cohort studies, and before-and-after studies were acceptable. Ongoing trials without interim analyses were excluded from the final selection. We did not use any language exclusion criteria. Only results from sub-Saharan Africa were included. For multisite studies, data from locations in sub-Saharan Africa were included, unless the data could not be separately extracted. The exclusion criteria were applied in the order shown in the figure. Specifically, the criteria that led to the exclusion of 113 studies during full-text review were applied in the order from top to bottom as shown in the figure. The investigators who selected studies for review were not blinded to study authors, conclusions, or outcomes, because blinding has little effect on systematic review results.53

Data extraction Once all potentially relevant full-text articles and abstracts were identified, three of the investigators (TB, KC, and NC) achieved consensus regarding eligibility and extracted data using a standardised extraction form. All measures of adherence were noted, including subjective and objective measures and biological correlates (such as CD4 cell count and viral load). Two investigators independently extracted the data from each study and entered the data into an electronic database. When the two data entries did not match, consensus was reached through data checks nd discussion, and if necessary consultation with another investigator (JH) who was not involved in the data extraction.

Assessment of risk of bias We followed the PRISMA guidelines to assess the risk of bias in individual studies and across studies.49 As 943

Review

3930 records identified through Medline

3815 excluded on the basis of title or abstract review

115 articles identified with inclusion criteria

17 articles identified in grey literature

See Online for webappendix

132 included in full-text review of eligibility

113 excluded 8 reviews, letters, or editorials 1 not adults 12 outside of sub-Saharan Africa 50 no intervention 16 no control 16 no adherence results 10 not about ART adherence

19 full-text reports included for review

7 conference abstracts 1 from reference lists

27 records included

Figure: Study selection

recommended, we assessed the risk separately for different components rather than in an overall scale. For individual randomised controlled trials, we used the method of the Cochrane Collaboration, with checks for selection bias (random generation of sequences for trial allocation and concealment of allocation before and during trial enrolment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (completeness of data for each reported outcome), and reporting bias (completeness of reporting of assessed outcomes).54 For individual observational studies, we followed the GRADE recommendations.55 We also compared reporting bias across studies and assessed the risk of publication bias.54

Results Study selection process 27 reports from 26 studies met all inclusion and exclusion criteria (figure): 17 journal articles,56–72 seven conference abstracts,73–76,77–79 one master’s thesis,80 one research report,81 and one research letter.82 Five reports were from Nigeria,57,66,73,75,76 five from South Africa,71,77,78,80,81 four from Kenya,61,67,70,79 three from Mozambique,60,62,63 three from Uganda,68,74,82 two from Malawi,58,59 two from Zambia,56,64 one from Rwanda,72 one from Tanzania,65 and one from several countries in sub-Saharan Africa.69 All the records were published in English. A research letter82 and a conference abstract74 were from the same study in 944

Uganda; data from both sources were therefore combined to create one record. We extracted information separately—despite overlap of authors, study setting, and enrolled patients—if studies differed in sample size,58,59,68,71,74,76,78,82 primary aims,58,59 length of followup58,59,68,74,82 intervention type,71,78 or adherence 68,71,73,74,76,78,82 measures. For example, two of the full-text articles were from the same setting in Malawi, but the first study reported an initial analysis of the overall results of a randomised controlled trial with 3 months of followup,59 and the second included a subsequent subgroup analysis with 9 months of follow-up (webappendix).58

Study characteristics Studies were carried out between 2003 and 2009, and reported between 2004 and 2010; over two-thirds were reported in 2009–10 (table 2, webappendix). The median sample size was 433 individuals (IQR 274–620) and the median length of follow-up was 12 months (7–18 months). 12 studies took place exclusively in urban hospital outpatient clinics, five only in rural community-based clinics, and the remaining studies either took place in several settings or did not provide sufficient information for classification (table 2).

Adherence interventions 14 studies used interventions combining behavioural, cognitive, and affective components (table 1). These interventions included treatment supporters that provided both emotional and instrumental adherence support.60,63–66,68,74,75,82 One intervention incorporated behavioural, cognitive, affective, and biological interventions through combinations of treatment supporters, nutritional support, financial support, psychosocial support, and education sessions.69 Purely behavioural interventions used directly observed therapy,57,61,78 diary cards,80 and mobile-phone short message services (text messages) to remind patients to take their ART drugs.67,70 Several interventions used directly observed therapy in addition to other adherence support; in table 2, we classified these interventions separately, because the recorded effects can be attributed only to the package of interventions. Purely biological interventions used various food supplements.56,58,59 Structural interventions included several models of delivery,62,81 which differed in the type of health worker providing routine adherence support or the type of health-care setting. Studies of the same intervention category differed in the precise intervention characteristics, length of followup, study design, and setting. For example, in studies with directly observed therapy,57,60,61,66,71,78 the frequency of observed therapy varied from every day to every week. In studies with treatment supporters,60,63–66,68,69,74,75,82 the person providing the support was a family member or a friend, neighbour, community health worker, or HIV-infected community members. Treatment supporters had different tasks in different studies, including psychosocial support,63,69 education about ART,60 identification of www.thelancet.com/infection Vol 11 December 2011

Review

Source type

Intervention category

Busari et al (2009)76

Conference abstract

Busari et al (2010)73

Conference abstract

Intervention type

Study period

Study country (city or region)

Health-care setting

Cognitive and affective Structured teaching programme vs informal patient education

Not reported

Nigeria (Abuja)

Intervention: hospital-inpatient (urban); effect observation: hospital-outpatient and community-based (urban)

Cognitive and affective Structured teaching programme vs informal patient education

Not reported

Nigeria (Abuja)

Intervention: hospital-inpatient (urban); effect observation: hospital-outpatient and community-based (urban)

Cantrell et al (2008)56

Journal article

Biological

2004

Zambia (Lusaka)

Community-based (urban)

Chang et al (2008)74,82

Conference abstract and research letter

Behavioural, cognitive, Treatment supporter and affective

2006–08

Uganda (Rakai)

Community-based (rural)

Chang et al (2010)68

Journal article

Behavioural, cognitive, Treatment supporter and affective

2006–08

Uganda (Rakai)

Community-based (rural)

Chung et al (2009)79

Conference abstract

Behavioural and cognitive

Educational counselling vs alarm device vs combination vs neither

2006–08

Kenya (Nairobi)

Hospital-outpatient (urban)

DOT (DAOT, TWOT, and WOT)

Hospital-outpatient (urban)

Food rations

Idoko et al (2007)57

Journal article

Behavioural

2003

Nigeria (Jos)

Kabore et al (2010)69

Journal article

Behavioural, cognitive, Treatment supporter, nutritional support, financial support, psychosocial support, affective, biological, and educational counselling and structural

2005–07

Community-based (unknown) Lesotho (Maseru), South Africa (Ladysmith), Namibia (Katima-Mulilo), and Botswana (Bobonong)

Lester et al (2010)70

Journal article

Behavioural

2007–08

Kenya (Nairobi)

Community-based (urban and rural)

Mugusi et al (2009)65

Journal article

Behavioural, cognitive, Treatment supporter, calendar with and affective reminders, and educational counselling

2004–07

Tanzania (Dar es Salaam)

Hospital-outpatient (urban)

Nachega et al (2009)78

Conference abstract

Behavioural

DOT (DAOT)

2005–07

South Africa (Cape Town)

Hospital-outpatient (urban)

Nachega et al (2010)71

Journal article

Behavioural and cognitive

DOT (DAOT) and educational counselling

2005–07

South Africa (Cape Town)

Hospital-outpatient (urban)

Ndekha et al (2009a)58

Journal article

Biological

Lipid paste vs flour supplement

2006–07

Malawi (Blantyre)

Hospital-outpatient (urban)

Ndekha et al (2009b)59

Journal article

Biological

Lipid paste vs flour supplement

2006–07

Malawi (Blantyre)

Hospital-outpatient (urban)

Pearson et al (2007)60

Journal article

Behavioural, cognitive, Treatment supporters, DOT (DOAT), and and affective education

2004

Mozambique (Beira)

Hospital-outpatient (urban)

Pienaar et al (2006)81

Research report

Structural

Differing models of ART delivery

2004–05

South Africa (Western Cape)

Hospital-outpatient and community-based (urban)

Pop-Eleches et al (2011)67 Journal article

Mobile-phone text messages

Behavioural

Mobile-phone text messages

2007–08

Kenya

Community-based (rural)

Roux (2004)80

Master’s thesis

Behavioural

Diary cards

2004

South Africa (North West province)

Community-based (rural)

Sarna et al (2008)61

Journal article

Behavioural

DOT (TWOT)

2003–04

Kenya (Mombasa)

Hospital-outpatient (urban)

Sherr et al (2010)62

Journal article

Structural

Non-physician provider

2004–07

Mozambique (Central Region)

Hospital-outpatient (urban)

Stubbs et al (2009)63

Journal article

Behavioural, cognitive, Treatment supporter and affective

2004–06

Mozambique (Central Region)

Hospital-outpatient (urban)

Taiwo et al (2010)66

Journal article

Behavioural, cognitive, Treatment supporter and DOT (DAOT) and affective

2006–08

Nigeria (Jos)

Hospital-outpatient (urban)

Thurman et al (2010)72

Journal article

Cognitive, affective, and biological

2006

Rwanda (entire country)

Community-based (unknown)

Torpey et al (2008)64

Journal article

Behavioural, cognitive, Treatment supporter and affective

2007

Zambia (Northern, Luapula, Central, and Copperbelt provinces)

Hospital-outpatient (unknown)

Udo et al (2007)75

Conference abstract

Behavioural, cognitive, Treatment supporter and affective

2007

Nigeria (Lower Cross River Basin)

Community-based (rural)

van Loggerenberg et al (2010)77

Conference abstract

Cognitive

2007–09

South Africa (Durban)

Community-based (urban)

Case manager

Individual vs group education counselling

ART=antiretroviral treatment. DOT=directly observed therapy. DAOT=DOT every day. TWOT=DOT every 2 weeks. WOT=DOT every week.

Table 2: Characteristics of interventions for ART adherence in sub-Saharan Africa

barriers to adherence,60 adherence measurement,68,74,82 reminding patients to pick up or take their drugs,66,75 directly observed therapy,60,66 assessment of adverse events,68,74,82 and triage to other health-care providers.68,74,82 www.thelancet.com/infection Vol 11 December 2011

Adherence assessment Adherence was assessed by subjective and objective measurement instruments (webappendix). Subjective instruments included patient’s recall of the number of 945

Review

missed doses in the past 3 days, 7 days, or 30 days (11 studies);58–61,64,65,68–70,80,81 focus groups in which participants were asked if they were able to take their drugs as and when due (one study);75 and health-worker opinions about whether an intervention had improved adherence (two studies).72,74,82 Objective instruments included pharmacy refill rates (four studies),56,62,63,66 pill counts in clinics (five studies),61,68,69,74,78,82 the Medication Event Monitoring System (one study),67 viral-load measurement or CD4 cell count (17 studies),56–61,65,66,68–71,76–79,81 and body-mass index (three studies).58,61,78 17 studies used several methods to assess adherence.56–61,65,66,68–71,73,76,78,79,81 For Current Controlled Trials please see http://www. controlled-trials.com For ClinicalTrials.gov please see http://www.clinicaltrials.gov

Intervention outcomes 16 of the 26 studies showed significantly better adherence in the intervention group than in the comparison group for at least one outcome measurement and one timepoint (webappendix).56,60–63,66–71,73,75,76,79,81 The interventions tested in the studies with significant effects were structured teaching programmes,73,76 food rations,56 directly observed therapy,61,71 treatment supporters with directly observed therapy,60,66 treatment supporters without directly observed therapy,63,68,69,75 non-physician providers,62 different models of ART delivery,81 and mobile-phone text messages.67,70 Two additional studies showed an improvement in adherence, one with treatment supporters and another with case managers;72,74,82 however, assessments were based on the subjective opinion of health-care providers and the significance of the effect was not reported. The remaining eight studies57–59,64,65,77,78,80 reported non-significant results across all timepoints and adherence measures used. Ten60,61,66–68,70,71,73,76,79 of 15 randomised controlled trials, six62,63,68,69,71,81,82 of seven reports of cohort studies, and three56,72,75 of five before-and-after studies showed improved adherence, according to at least one outcome measure (webappendix). With three exceptions,70,73,76 all studies that found an improvement in adherence did not find an improvement consistently over time61,66,71 or on self-reported adherence,68 did not assess any biological outcome,62,63,67,72,75,76,80,81 or did not find a significant effect on biological outcomes.56,60,61,66,79

Risk of bias In most hierarchies of evidence ordered by risk of bias, randomised controlled trials rank above observational studies.83 However, this ranking could be reversed for specific studies, because randomised controlled trials can involve biases, which limit the strength of evidence derived from their results, and observational studies can have characteristics that increase the evidence strength, such as good control of confounding factors or large effect size.55 Five randomised controlled trials used both sequence generation for randomisation and allocation concealment as a safeguard against selection bias,58–61,70 two used only sequence generation67 and one only allocation concealment.66 Only two studies, comparing nutritional interventions, described the masking of study participants 946

and health-care providers to minimise performance bias.58,59 With four exceptions,58,59,61,70 no trial studies reported whether outcome assessors had been blinded to prevent detection bias. However, descriptions of outcome assessments suggested that assessors of self-reported adherence were mostly aware of the trial assignment, while laboratory personnel measuring biological outcomes and analysts were blinded. Finally, all investigators of randomised controlled trials completely reported the results of the outcomes they described in the methods sections. We also compared the reported outcomes of randomised controlled trials with those registered with either Current Controlled Trials or ClinicalTrials.gov and detected no evidence of selective reporting of primary or secondary adherence outcomes. In all observational studies, appropriate eligibility criteria were defined and applied equally to all participants, and outcome assessment did not differ by intervention assignment. However, only one of the five before-and-after studies used a control group to remove secular time changes in the comparison of adherence before and after the intervention,56 and only three of six cohort studies controlled for different distributions of potential confounding factors between intervention and comparison groups.62,63,69 In the 17 publications that reported loss to follow-up, the loss was always less than 20%—ie, less than the threshold proposed for potentially significant selection bias when individuals are not missing at random.84 In more than half of these studies, loss to follow-up was less than 10%.56,58,59,61,63,64,68,70,71 Most investigations that did not report loss to follow-up were conference abstracts.73,75–78,82

Discussion Several important insights have emerged from this systematic review of interventions to increase ART adherence in sub-Saharan Africa. The reviewed studies investigated six types of adherence-enhancing interventions: text messages and other reminder devices, treatment supporters, directly observed therapy, education and counselling, food supplements, and different health-systems approaches to ART delivery. For each intervention type there is at least one randomised controlled trial60,61,66,67,70,73,76 or one observational study56,62,63,69,72,74,75,81,82 showing effectiveness. However, before policy conclusions can be made, a more detailed examination of the evidence is necessary. First, the size of an intervention’s effect needs to be considered, not merely the statistical significance. In studies that showed significant effects, the results constituted an improvement of 10% or less in four,60,62,79,81 10–20% in two,69,70 and an improvement of more than 20% in 12 investigations.56,61,63,66–68,71–76,82 Although small effect sizes might indicate that investment in the intervention is not worthwhile, such decisions should be made on the basis of economic assessment comparing achievable effects with costs. As yet, such studies have not been done. www.thelancet.com/infection Vol 11 December 2011

Review

Second, improved adherence might not persist over time. Three randomised controlled trials showed a significant intervention effect by some measure of adherence in the first 6 months of the trial, but not in later phases. These results suggest that evidence of effectiveness of adherence-enhancing interventions from studies of short duration might not be generalisable to the longterm. Because ART is lifelong, adherence-enhancing interventions for which effectiveness is limited to a few months would contribute little to overall treatment success. Future studies should thus attempt to assess effectiveness for at least 1 year. Moreover, continuation of some of the studies with significant short-term effects for several more years would be valuable, because long-term effectiveness could be established. It is important to note, however, that high adherence to ART might be less essential later in treatment than earlier. Lima and colleagues15 reported that the risk of viral rebound caused by imperfect adherence decreases with duration of viral suppression. Therefore, interventions that increase adherence in the first months after ART initiation could improve long-term biological outcomes, even if their effect vanishes over time. While these findings suggest that high adherence might not always be necessary for viral suppression, other results from subSaharan Africa indicate that high adherence might not be sufficient for viral suppression in some patients.19,85 Adherence is merely a means to achieve good health outcomes of ART, rather than a final outcome in its own right. The relation between adherence and ART success is clearly strong; however, better understanding is needed of whether, when, and why near-perfect adherence might not be necessary or sufficient to attain good health outcomes in all patients. Specifically, more evidence is needed about the factors that allow some patients to maintain good virological and immunological outcomes despite imperfect adherence, and about the factors that lead to treatment failure despite near-perfect adherence. We also need to know whether in the two groups of patients in which the close association between adherence and biological measures does not seem to hold, the relation between biological measures and health outcomes (morbidity and mortality) conforms with or deviates from what is observed in the overall population of patients receiving ART. Third, in examination of the collective evidence on adherence-enhancing interventions in sub-Saharan Africa it is important to consider discrepant findings on intervention effectiveness. For five intervention types shown to be effective in at least one study, other studies did not detect a significant effect (non-text-message reminder devices,65,79,80 treatment supporters,64,65 directly observed therapy,57,86 education and counselling,65,77,79 and food supplements58,59). Because of the small number of studies of each intervention type, identification of exact sources of the discrepancies is difficult. One reason for the discrepancies is that for each intervention type, the specific content can differ. For example, non-text-message reminder www.thelancet.com/infection Vol 11 December 2011

devices included alarm devices and calendars to support adherence, directly observed therapy ranged from pill taking observed every day to once every week, and treatment-supporter interventions differed in the selection of the supporter and the intensity and content of the support provided. Further, several factors might limit the ability of some intervention studies to show an effect. For example, the standard of care used in the control group could already produce high adherence in one study, but not in another.87 Ceiling effects can occur when baseline adherence is high, which limits the potential to show improvement. For example, adherence in the study by Roux and colleagues80 was 100% before the intervention and 99% afterwards. A case in point for discrepant findings is the adherence effect of directly observed therapy. Of studies in subSaharan Africa, one high-quality randomised controlled trial66 shows significant adherence improvement with directly observed therapy in both self-reported adherence and biological correlates of adherence. A few other studies suggest that directly observed therapy could improve adherence but this evidence is comparatively weak, because it is based on subjective or objective adherence measures alone, without demonstrated effects on biological correlates,60,61 or because it shows that initial biological effects do not last over time.71 In a 2009 metaanalysis88 of randomised controlled trials in developed and developing countries, Ford and colleagues concluded that directly observed therapy did not improve adherence compared with self-administered treatment, but emphasised that “the fact that individual trials found opposing results with respect to benefit of directly observed therapy underscores the importance of considering contextual factors in assessment of adherence interventions”. Conversely, in a 2010 meta-analysis38 of both randomised controlled trials and observational studies, Hart and colleagues found that directly observed therapy “had a significant effect on virological, immunological, and adherence outcomes, although its efficacy was not supported when restricting analysis to randomized controlled trials”. The investigators of this study further noted that “interventions varied widely” and identified features of intervention content that improved the effect on adherence (targeting individuals at high risk of non-adherence; maximising convenience for patients participating in the intervention; and providing adherence support in addition to directly observed therapy, such as other behavioural interventions).38 Although intervention context and the precise content probably affect intervention effect on adherence, the robustness of conclusions drawn from analyses of study heterogeneity depends crucially on the number of available studies.89 The worldwide evidence is still limited (ten randomised controlled trials in the study by Ford and colleagues,88 and 11 randomised controlled trials and six observational studies in the investigation by Hart and 947

Review

co-workers38), and more data on directly observed ART in sub-Saharan Africa is clearly needed. Discrepant findings both between and within studies also arise because of differences in outcome measures. Many studies used subjective measures to assess adherence outcomes, including several that did not use any other measure, and a few of the studies that used objective measures did not assess any biological correlates of adherence. Subjective adherence measures are prone to social desirability biases90 and have not correlated well with plasma drug concentrations, CD4 cell counts, or viral loads in several studies.91–93 Health workers (including those involved in the care of the control group in an intervention of directly observed therapy) usually do not have direct knowledge of patients’ pill-taking behaviour and thus might not accurately estimate adherence. Objective measurement instruments, such as pharmacy refill data, pill count or the Medication Events Monitoring System, might not reflect true pill taking if patients discard pills or obtain ART from sources that are not included in the measurement of the refill data, such as friends, family members, or other pharmacies.94 In the absence of a goldstandard measure of adherence, future studies should assess many outcomes. Because adherence affects CD4 cell count and viral load, which in turn will affect morbidity and mortality, the robustness and relevance of a result will increase with the number of distinct outcome measures that show the same result. Conversely, within-study heterogeneity of findings across outcome measures will reduce study robustness and relevance. Fourth, it is important to consider how far the studies included in this Review differ from those done in other settings. Two important insights gained in two decades of ART adherence research in developed countries are not reflected in the studies in sub-Saharan Africa. Interventions derived from a substantive theory of behaviour change tend to be more effective than are those based on intuition.95,96 However, unlike many studies in developed countries,30,36,97,98 none of the studies in this Review reported a theoretical basis for the investigated intervention. The absence of theoretical foundation might be due to insufficient understanding of how far established theories of behaviour change are applicable to subSaharan Africa,, even though some models have been investigated in the region (eg, the Information-MotivationBehavioral Skills model).99,100 Further research on both established and new theories of adherence behaviour is needed in sub-Saharan Africa and could contribute to the design of successful interventions. Additionally, interventions targeted at patients known to be at risk often have better results than do untargeted interventions.101 Unlike many studies in developed countries,97,102 the studies in this Review did not target sub-populations of patients at high risk of nonadherence, which might have limited their effectiveness. Although some at-risk populations might be uncommon or difficult to identify in sub-Saharan Africa (eg, 948

intravenous drug users or men who have sex with men), other targeting strategies used in developed countries could be feasible, effective, and cost effective in the region. Directly observed therapy, for example, could be appropriate for specific populations,103 such as individuals with depression,104 but not for others, such as individuals who are especially sensitive to stigma.105 A focus on individuals who have had an unsuccessful ART regimen106,107 or on people who have been identified as non-adherent through an adherence screening might also be appropriate in this region.108–111 Fifth, cost is an important consideration for adherenceenhancing interventions in sub-Saharan Africa. A 2010 cost analysis19 in South Africa showed that high ART adherence is associated with substantially lower mean monthly direct health-care costs. This finding suggests that effective adherence support could decrease the cost of ART per patient, which would emphasise the importance of adherence-enhancing interventions at a time when the number of individuals requiring ART continues to grow while ART funding might not increase.112,113 Future assessments should include costing studies to test the hypothesis that adherence-enhancing interventions can lead to a net decrease in the cost of ART delivery. In sum, evidence is emerging that mobile-phone text messages and other reminder devices, treatment supporters, directly observed therapy, education and counselling, food supplements, and different organisations of ART can be effective approaches to increase ART adherence in some settings in subSaharan Africa. However, with the exception of textmessage reminders, either the evidence is based only on observational studies, which are unlikely to control completely for unmeasured confounding factors, or discrepant findings suggest that intervention effect is strongly affected by context or the precise intervention content. Future research efforts should include randomised controlled trials of interventions to improve adherence in routine ART delivery in sub-Saharan Africa, be based on substantive theories of behaviour, and incorporate costing studies to allow economic assessment of interventions. Additionally, health workers in programmes that have integrated adherence interventions into routine ART delivery should report available evidence on intervention effectiveness. Sustained, high adherence to ART will be crucial for the long-term success of treatment programmes in subSaharan Africa, where most of the people currently receiving and needing ART live and where the options for second-line therapy after first-line failure are often limited. Initial evidence has accrued as to which interventions are likely to substantially improve adherence, but further studies are needed to confirm intervention effects, assess effect duration, identify the modifying effects of the intervention context, and establish intervention cost-effectiveness. www.thelancet.com/infection Vol 11 December 2011

Review

Contributors TB and M-LN were jointly responsible for the design of the study. KC, NC, and AP did the search of the reported work and data extraction, with support from TB and JH. TB wrote the initial draft of the paper, all authors contributed to the writing of the report.

15

16

Conflicts of interest We declare that we have no conflicts of interest. Acknowledgments TB is supported by grant 1R01-HD058482-01 from the National Institute of Child Health and Human Development (NICHD) and by grant 1R01MG083539-01 from the National Institute of Mental Health and the Wellcome Trust. The funding organisation had no role in the design and conduct of the study, in the collection, analysis, and interpretation of the data, or in the preparation, review, or approval of the report. TB has been paid by UNITAID for consultancy work and has received grant funding for HIV modelling from the World Bank. JH is supported by National Institute of Mental Health grant K23087228. M-LN is supported by the Wellcome Trust. References 1 Jerene D, Naess A, Lindtjorn B. Antiretroviral therapy at a district hospital in Ethiopia prevents death and tuberculosis in a cohort of HIV patients. AIDS Res Ther 2006; 3: 10. 2 Chan KC, Wong KH, Lee SS. Universal decline in mortality in patients with advanced HIV-1 disease in various demographic subpopulations after the introduction of HAART in Hong Kong, from 1993 to 2002. HIV Med 2006; 7: 186–92. 3 Hogg RS, Yip B, Kully C, et al. Improved survival among HIV-infected patients after initiation of triple-drug antiretroviral regimens. CMAJ 1999; 160: 659–65. 4 Hammer SM, Squires KE, Hughes MD, et al, for the AIDS Clinical Trials Group 320 Study Team. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. N Engl J Med 1997; 337: 725–33. 5 Palella FJ Jr, Delaney KM, Moorman AC, et al, for the HIV Outpatient Study Investigators. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med 1998; 338: 853–60. 6 Herbst AJ, Cooke GS, Bärnighausen T, KanyKany A, Tanser F, Newell ML. Adult mortality and antiretroviral treatment roll-out in rural KwaZulu-Natal, South Africa. Bull World Health Organ 2009; 87: 754–62. 7 Paterson DL, Swindells S, Mohr J, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med 2000; 133: 21–30. 8 Moore DM, Hogg RS, Yip B, et al. Discordant immunologic and virologic responses to highly active antiretroviral therapy are associated with increased mortality and poor adherence to therapy. J Acquir Immune Defic Syndr 2005; 40: 288–93. 9 Wood E, Hogg RS, Yip B, Harrigan PR, O’Shaughnessy MV, Montaner JS. Effect of medication adherence on survival of HIV-infected adults who start highly active antiretroviral therapy when the CD4+ cell count is 0·200 to 0·350 × 10(9) cells/L. Ann Intern Med 2003; 139: 810–16. 10 Wood E, Hogg RS, Yip B, Harrigan PR, O’Shaughnessy MV, Montaner JS. The impact of adherence on CD4 cell count responses among HIV-infected patients. J Acquir Immune Defic Syndr 2004; 35: 261–68. 11 Arnsten JH, Demas PA, Grant RW, et al. Impact of active drug use on antiretroviral therapy adherence and viral suppression in HIV-infected drug users. J Gen Intern Med 2002; 17: 377–81. 12 Bangsberg DR, Hecht FM, Charlebois ED, et al. Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population. AIDS 2000; 14: 357–66. 13 Shuter J, Sarlo JA, Kanmaz TJ, Rode RA, Zingman BS. HIV-infected patients receiving lopinavir/ritonavir-based antiretroviral therapy achieve high rates of virologic suppression despite adherence rates less than 95%. J Acquir Immune Defic Syndr 2007; 45: 4–8. 14 Bangsberg DR. Less than 95% adherence to nonnucleoside reverse-transcriptase inhibitor therapy can lead to viral suppression. Clin Infect Dis 2006; 43: 939–41.

www.thelancet.com/infection Vol 11 December 2011

17

18

19

20

21

22

23 24

25

26

27

28

29

30

31

32

33

34

35

36 37

Lima VD, Bangsberg DR, Harrigan PR, et al. Risk of viral failure declines with duration of suppression on highly active antiretroviral therapy irrespective of adherence level. J Acquir Immune Defic Syndr 2010; 55: 460–65. Gardner EM, Burman WJ, Steiner JF, Anderson PL, Bangsberg DR. Antiretroviral medication adherence and the development of class-specific antiretroviral resistance. AIDS 2009; 23: 1035–46. Conway B. The role of adherence to antiretroviral therapy in the management of HIV infection. J Acquir Immune Defic Syndr 2007; 45 (suppl 1): S14–18. Goldie SJ, Paltiel AD, Weinstein MC, et al. Projecting the cost-effectiveness of adherence interventions in persons with human immunodeficiency virus infection. Am J Med 2003; 115: 632–41. Nachega JB, Leisegang R, Bishai D, et al. Association of antiretroviral therapy adherence and health care costs. Ann Intern Med 2010; 152: 18–25. Kalichman SC, Cherry C, Amaral CM, et al. Adherence to antiretroviral therapy and HIV transmission risks: implications for test-and-treat approaches to HIV prevention. AIDS Patient Care STDS 2010; 24: 271–77. Bärnighausen T. Access to antiretroviral treatment in the developing world: a framework, review and health systems research agenda. Therapy 2007; 4: 753–66. WHO, UNAIDS, UNICEF. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector. Geneva: World Health Organization, 2010. Wakabi W. Low ART adherence in Africa. Lancet Infect Dis 2008; 8: 94. Mills EJ, Nachega JB, Buchan I, et al. Adherence to antiretroviral therapy in sub-Saharan Africa and North America: a meta-analysis. JAMA 2006; 296: 679–90. Mills EJ, Nachega JB, Bangsberg DR, et al. Adherence to HAART: a systematic review of developed and developing nation patient-reported barriers and facilitators. PLoS Med 2006; 3: e438. Parruti G, Manzoli L, Toro PM, et al. Long-term adherence to first-line highly active antiretroviral therapy in a hospital-based cohort: predictors and impact on virologic response and relapse. AIDS Patient Care STDS 2006; 20: 48–56. Mannheimer S, Friedland G, Matts J, Child C, Chesney M. The consistency of adherence to antiretroviral therapy predicts biologic outcomes for human immunodeficiency virus-infected persons in clinical trials. Clin Infect Dis 2002; 34: 1115–21. Liu H, Miller LG, Hays RD, et al. Repeated measures longitudinal analyses of HIV virologic response as a function of percent adherence, dose timing, genotypic sensitivity, and other factors. J Acquir Immune Defic Syndr 2006; 41: 315–22. Byakika-Tusiime J, Crane J, Oyugi JH, et al. Longitudinal antiretroviral adherence in HIV+ Ugandan parents and their children initiating HAART in the MTCT-Plus family treatment model: role of depression in declining adherence over time. AIDS Behav 2009; 13 (suppl 1): 82–91. Simoni JM, Frick PA, Pantalone DW, Turner BJ. Antiretroviral adherence interventions: a review of current literature and ongoing studies. Top HIV Med 2003; 11: 185–98. Fogarty L, Roter D, Larson S, Burke J, Gillespie J, Levy R. Patient adherence to HIV medication regimens: a review of published and abstract reports. Patient Educ Couns 2002; 46: 93–108. Haddad M, Inch C, Glazier RH, et al. Patient support and education for promoting adherence to highly active antiretroviral therapy for HIV/AIDS. Cochrane Database Syst Rev 2000; 3: CD001442. Bigler S, Nicca D, Spirig R. Interventions to enhance adherence of patients with HIV on ART: a literature review. Pflege 2007; 20: 268–77 (in German). Rueda S, Park-Wyllie LY, Bayoumi AM, et al. Patient support and education for promoting adherence to highly active antiretroviral therapy for HIV/AIDS. Cochrane Database Syst Rev 2006; 3: CD001442. Machouf N, Lalonde RG. Directly observed therapy (DOT): from tuberculosis to HIV. Rev Epidemiol Sante Publique 2006; 54: 73–89 (in French). Cote JK, Godin G. Efficacy of interventions in improving adherence to antiretroviral therapy. Int J STD AIDS 2005; 16: 335–43. Wise J, Operario D. Use of electronic reminder devices to improve adherence to antiretroviral therapy: a systematic review. AIDS Patient Care STDS 2008; 22: 495–504.

949

Review

38

39

40

41

42

43

44 45

46 47 48

49

50

51

52

53

54

55

56

57

58

59

950

Hart JE, Jeon CY, Ivers LC, et al. Effect of directly observed therapy for highly active antiretroviral therapy on virologic, immunologic, and adherence outcomes: a meta-analysis and systematic review. J Acquir Immune Defic Syndr 2010; 54: 167–79. Heyer A, Ogunbanjo GA. Adherence to HIV antiretroviral therapy part II: which interventions are effective in improving adherence? S Afr Fam Pract 2006; 48: 6–10. Hirschhorn LR, Oguda L, Fullem A, Dreesch N, Wilson P. Estimating health workforce needs for antiretroviral therapy in resource-limited settings. Hum Resour Health 2006; 4: 1. Zaric GS, Bayoumi AM, Brandeau ML, Owens DK. The costeffectiveness of counseling strategies to improve adherence to highly active antiretroviral therapy among men who have sex with men. Med Decis Making 2008; 28: 359–76. Safren SA, O’Cleirigh C, Tan JY, et al. A randomized controlled trial of cognitive behavioral therapy for adherence and depression (CBT-AD) in HIV-infected individuals. Health Psychol 2009; 28: 1–10. Conference on Retroviruses and Opportunistic Infections. San Francisco, CA, USA; Feb 16–19, 2010. http://retroconference. org/2010/display.asp?page=493 (accessed Jan 27, 2011). International AIDS Society. Conferences. 2011. http://www. iasociety.org/Default.aspx?pageId=3 (accessed Jan 27, 2011). International Association of Physicians in AIDS Care. 5th International Conference on HIV Treatment Adherence. 2010. http://www.iapac.org/AdherenceConference/AdherenceConf2008info.html (accessed Jan 27, 2011). mHealthSummit. Overview. 2011. http://www.mhealthsummit.org/ program_details.php (accessed Jan 31, 2011). Auger CP. Information sources in grey literature, 4th edn. London: Bowker-Saur, 1998. Hopewell S, McDonald S, Clarke M, Egger M. Grey literature in meta-analyses of randomized trials of health care interventions. Cochrane Database Syst Rev 2007; 2: MR000010. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 2009; 6: e1000100. Sanne I, Orrell C, Fox MP, et al. Nurse versus doctor management of HIV-infected patients receiving antiretroviral therapy (CIPRA-SA): a randomised non-inferiority trial. Lancet 2010; 376: 33–40. Chan AK, Mateyu G, Jahn A, et al. Outcome assessment of decentralization of antiretroviral therapy provision in a rural district of Malawi using an integrated primary care model. Trop Med Int Health 2010; 15 (suppl 1): 90–97. Rosen S, Fox MP, Gill CJ. Patient retention in antiretroviral therapy programs in sub-Saharan Africa: a systematic review. PLoS Med 2007; 4: e298. Berlin JA, on behalf of University of Pennsylvania Meta-analysis Blinding Study Group. Does blinding of readers affect the results of meta-analyses?. Lancet 1997; 350: 185–86. Higgins JPT, Altman DG, Sterne JAC. Chapter 8: assessing risk of bias in included studies. In: Higgins JPT, Green S, eds. Cochrane handbook for systematic reviews of interventions. London: The Cochrane Collaboration, 2011: 187–235. Guyatt GH, Oxman AD, Vist G, et al. GRADE guidelines: 4. Rating the quality of evidence—study limitations (risk of bias). J Clin Epidemiol 2011; 64: 407–15. Cantrell RA, Sinkala M, Megazinni K, et al. A pilot study of food supplementation to improve adherence to antiretroviral therapy among food-insecure adults in Lusaka, Zambia. J Acquir Immune Defic Syndr 2008; 49: 190–95. Idoko JA, Agbaji O, Agaba P, et al. Direct observation therapy-highly active antiretroviral therapy in a resource-limited setting: the use of community treatment support can be effective. Int J STD AIDS 2007; 18: 760–63. Ndekha M, van Oosterhout JJ, Saloojee H, Pettifor J, Manary M. Nutritional status of Malawian adults on antiretroviral therapy 1 year after supplementary feeding in the first 3 months of therapy. Trop Med Int Health 2009; 14: 1059–63. Ndekha MJ, van Oosterhout JJ, Zijlstra EE, Manary M, Saloojee H, Manary MJ. Supplementary feeding with either ready-to-use fortified spread or corn-soy blend in wasted adults starting antiretroviral therapy in Malawi: randomised, investigator blinded, controlled trial. BMJ 2009; 338: b1867.

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

Pearson CR, Micek MA, Simoni JM, et al. Randomized control trial of peer-delivered, modified directly observed therapy for HAART in Mozambique. J Acquir Immune Defic Syndr 2007; 46: 238–44. Sarna A, Luchters S, Geibel S, et al. Short- and long-term efficacy of modified directly observed antiretroviral treatment in Mombasa, Kenya: a randomized trial. J Acquir Immune Defic Syndr 2008; 48: 611–19. Sherr KH, Micek MA, Gimbel SO, et al. Quality of HIV care provided by non-physician clinicians and physicians in Mozambique: a retrospective cohort study. AIDS 2010; 24 (suppl 1): S59–66. Stubbs BA, Micek MA, Pfeiffer JT, Montoya P, Gloyd S. Treatment partners and adherence to HAART in Central Mozambique. AIDS Care 2009; 21: 1412–19. Torpey KE, Kabaso ME, Mutale LN, et al. Adherence support workers: a way to address human resource constraints in antiretroviral treatment programs in the public health setting in Zambia. PLoS One 2008; 3: e2204. Mugusi F, Mugusi S, Bakari M, et al. Enhancing adherence to antiretroviral therapy at the HIV clinic in resource constrained countries; the Tanzanian experience. Trop Med Int Health 2009; 14: 1226–32. Taiwo BO, Idoko JA, Welty LJ, et al. Assessing the viorologic and adherence benefits of patient-selected HIV treatment partners in a resource-limited setting. J Acquir Immune Defic Syndr 2010; 54: 85–92. Pop-Eleches C, Thirumurthy H, Habyarimana JP, et al. Mobile phone technologies improve adherence to antiretroviral treatment in a resource-limited setting: a randomized controlled trial of text message reminders. AIDS 2011; 25: 825–34. Chang LW, Kagaayi J, Nakigozi G, et al. Effect of peer health workers on AIDS care in Rakai, Uganda: a cluster-randomized trial. PLoS One 2010; 5: e10923. Kabore I, Bloem J, Etheredge G, et al. The effect of communitybased support services on clinical efficacy and health-related quality of life in HIV/AIDS patients in resource-limited settings in subSaharan Africa. AIDS Patient Care STDS 2010; 24: 581–94. Lester RT, Ritvo P, Mills EJ, et al. Effects of a mobile phone short message service on antiretroviral treatment adherence in Kenya (WelTel Kenya1): a randomised trial. Lancet 2010; 376: 1838–45. Nachega JB, Chaisson RE, Goliath R, et al. Randomized controlled trial of trained patient-nominated treatment supporters providing partial directly observed antiretroviral therapy. AIDS 2010; 24: 1273–80. Thurman TR, Haas LJ, Dushimimana A, Lavin B, Mock N. Evaluation of a case management program for HIV clients in Rwanda. AIDS Care 2010; 22: 759–65. Busari O, Busari O, Oligbu G, Ashom D, Adeyemi A, HIV Study Group. Novel structured teaching of HIV patients in resourcelimiting setting: effect of learning outcomes on adherence to HAART and immunological recovery. 17th Conference on Retroviruses and Opportunistic Infections; San Francisco, CA, USA; Feb 16–19, 2010. 820. Chang LW, Kagaayi J, Nakigozi G, et al. Peer health workers, mobile phones, and HIV care in Rakai, Uganda. 3rd International Conference on HIV Treatment Adherence; Jersey City, NJ, USA; March 17–18, 2008: 26. Udo DU, Itah A, Udofia UA, The Lutheran AIDS Research Unit. Promoting ART adherence and compliance through family and peer group support initiative. 4th IAS Conference on HIV Pathogenesis, Treatment and Prevention; Sydney, Australia; July 22–25, 2007. B35. Busari O, Oligbu G, Busari O, Asham D, Adeyemi A. Effect of structured systematic (modular) teaching of HIV/AIDS patients on adherence to antiretroviral therapy (HAART): report from a national HIV/AIDS treatment center in sub-Saharan Africa. 4th Annual HIV Treatment Adherence Conference of the International Association of Providers in AIDS Care; Miami, FL, USA; April 5–7, 2009. 0037. van Loggerenberg F, Grant A, Naidoo K, et al. Group counseling achieves high adherence to antiretroviral therapy: results of the CAPRISA 058 randomized controlled trial comparing group versus individualized adherence counseling strategies in Durban, South Africa. 5th International Conference on HIV Treatment Adherence; Miami, FL, USA; May 23–25, 2010. 62157.

www.thelancet.com/infection Vol 11 December 2011

Review

78

79

80

81 82

83 84 85

86

87

88

89

90

91

92

93

94

95

Nachega J, Goliath R, Efron A, et al. Randomized trial of trained patient-nominated treatment supporters providing partial directly observed ART in South African adults initiating HIV therapy. Conference on Retroviruses and Opportunistic Infections; Montreal, QC; Feb 8–11, 2009: 143. Chung M, Benki-Nugent S, Richardson B, et al. Randomized controlled trial comparing educational counseling and alarm device on adherence to antiretretroviral medications in Nairobi, Kenya: over 18 months follow-up. 4th Annual HIV Treatment Adherence Conference of the International Association of Providers in AIDS Care; Miami, FL, USA; April 5–7, 2009. 0270. Roux SM. Diary cards: preliminary evaluation of an intervention tool for improving adherence to antiretroviral therapy and TB preventive therapy in people living with HIV/AIDS. MPH thesis, University of the Western Cape, 2004. Pienaar D, Myer L, Cleary S, et al. Models of Care for Antiretroviral Service Delivery. Cape Town: University of Cape Town, 2006. Chang LW, Kagaayi J, Nakigozi G, et al. Responding to the human resource crisis: peer health workers, mobile phones, and HIV care in Rakai, Uganda. AIDS Patient Care STDS 2008; 22: 173–74. Gugiu PC, Gugiu MR. A critical appraisal of standard guidelines for grading levels of evidence. Eval Health Prof 2010; 33: 233–55. Kristman V, Manno M, Cote P. Loss to follow-up in cohort studies: how much is too much? Eur J Epidemiol 2004; 19: 751–60. Bisson GP, Rowh A, Weinstein R, Gaolathe T, Frank I, Gross R. Antiretroviral failure despite high levels of adherence: discordant adherence-response relationship in Botswana. J Acquir Immune Defic Syndr 2008; 49: 107–10. Nachega JB, Hislop M, Nguyen H, et al. Antiretroviral therapy adherence, virologic and immunologic outcomes in adolescents compared with adults in southern Africa. J Acquir Immune Defic Syndr 2009; 51: 65–71. de Bruin M, Viechtbauer W, Hospers HJ, Schaalma HP, Kok G. Standard care quality determines treatment outcomes in control groups of HAART-adherence intervention studies: implications for the interpretation and comparison of intervention effects. Health Psychol 2009; 28: 668–74. Ford N, Nachega JB, Engel ME, Mills EJ. Directly observed antiretroviral therapy: a systematic review and meta-analysis of randomised clinical trials. Lancet 2009; 374: 2064–71. Higgins J, Thompson S, Deeks J, Altman D. Statistical heterogeneity in systematic reviews of clinical trials: a critical appraisal of guidelines and practice. J Health Serv Res Policy. 2002; 7: 51–61. Bangsberg DR. Preventing HIV antiretroviral resistance through better monitoring of treatment adherence. J Infect Dis 2008; 197 (suppl 3): S272–78. Hugen PW, Burger DM, Aarnoutse RE, et al. Therapeutic drug monitoring of HIV-protease inhibitors to assess noncompliance. Ther Drug Monit 2002; 24: 579–87. Kouanfack C, Laurent C, Peytavin G, et al. Adherence to antiretroviral therapy assessed by drug level monitoring and self-report in cameroon. J Acquir Immune Defic Syndr 2008; 48: 216–19. Chaiyachati K, Hirschhorn LR, Tanser F, Newell ML, Bärnighausen T. Validating five questions of antiretroviral nonadherence in a public-sector treatment program in rural South Africa. AIDS Patient Care STDS 2011; 25: 163–70. Berg KM, Arnsten JH. Practical and conceptual challenges in measuring antiretroviral adherence. J Acquir Immune Defic Syndr 2006; 43 (suppl 1): S79–87. Fisher JD, Cornman DH, Norton WE, Fisher WA. Involving behavioral scientists, health care providers, and HIV-infected patients as collaborators in theory-based HIV prevention and antiretroviral adherence interventions. J Acquir Immune Defic Syndr 2006; 43 (suppl 1): S10–17.

www.thelancet.com/infection Vol 11 December 2011

96

97

98

99 100

101

102

103

104

105 106

107

108

109

110

111

112

113

Elder JP, Ayala GX, Harris S. Theories and intervention approaches to health-behavior change in primary care. Am J Prev Med 1999; 17: 275–84. Simoni JM, Rivet de Amico K, Smith L, Nelson K. Antiretroviral adherence interventions: translating research findings to the real world clinic. Curr HIV/AIDS Rep 2010; 7: 44–51. Fisher JD, Fisher WA, Amico KR, Harman JJ. An informationmotivation-behavioral skills model of adherence to antiretroviral therapy. Health Psychol 2006; 25: 462–73. Fisher JD, Fisher WA. Changing AIDS-risk behavior. Psychol Bull 1992; 111: 455–74. Simbayi LC, Kalichman SC, Skinner D, et al. Theory-based HIV risk reduction counseling for sexually transmitted infection clinic patients in Cape Town, South Africa. Sex Transm Dis 2004; 31: 727–33. Amico KR, Harman JJ, Johnson BT. Efficacy of antiretroviral therapy adherence interventions: a research synthesis of trials, 1996 to 2004. J Acquir Immune Defic Syndr 2006; 41: 285–97. Simoni JM, Amico KR, Pearson CR, Malow R. Strategies for promoting adherence to antiretroviral therapy: a review of the literature. Curr Infect Dis Rep 2008; 10: 515–21. Bangsberg DR. Modified directly observed therapy to improve HIV treatment outcomes: little impact with potent, once-daily therapy in unselected antiretroviral-naive patients. Curr HIV/AIDS Rep 2009; 6: 173–74. Sarna A, Pujari S, Sengar AK, Garg R, Gupta I, Dam J. Adherence to antiretroviral therapy and its determinants amongst HIV patients in India. Indian J Med Res 2008; 127: 28–36. Liechty CA, Bangsberg DR. Doubts about DOT: antiretroviral therapy for resource-poor countries. AIDS 2003; 17: 1383–87. Smith SR, Rublein JC, Marcus C, Brock TP, Chesney MA. A medication self-management program to improve adherence to HIV therapy regimens. Patient Educ Couns 2003; 50: 187–99. Wohl AR, Garland WH, Valencia R, et al. A randomized trial of directly administered antiretroviral therapy and adherence case management intervention. Clin Infect Dis 2006; 42: 1619–27. Javanbakht M, Prosser P, Grimes T, Weinstein M, Farthing C. Efficacy of an individualized adherence support program with contingent reinforcement among nonadherent HIV-positive patients: results from a randomized trial. J Int Assoc Physicians AIDS Care (Chic) 2006; 5: 143–50. Murphy DA, Lu MC, Martin D, Hoffman D, Marelich WD. Results of a pilot intervention trial to improve antiretroviral adherence among HIV-positive patients. J Assoc Nurses AIDS Care 2002; 13: 57–69. Murphy DA, Marelich WD, Rappaport NB, Hoffman D, Farthing C. Results of an antiretroviral adherence intervention: STAR (Staying Healthy: taking antiretrovirals regularly). J Int Assoc Physicians AIDS Care (Chic) 2007; 6: 113–24. Parry MF, Wright P, Stewart J, McLeod GX, Tucker J, Weinberg AR. Impact of an adherence program on the health and outlook of HIV-infected patients failing antiretroviral therapy. J Int Assoc Physicians AIDS Care (Chic) 2005; 4: 59–65. Bärnighausen T, Bloom DE, Humair S. Human resources for treating HIV/AIDS: needs, capacities, and gaps. AIDS Patient Care STDS 2007; 21: 799–f812. WHO, UNAIDS, UNICEF. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector. Geneva: World Health Organization, 2009.

951