Secondary Logo

Journal Logo

Implementation Science

Exploring Sustainability in the Era of Differentiated HIV Service Delivery in Sub-Saharan Africa: A Systematic Review

Okere, Nwanneka E. MSca; Lennox, Laura PhDb; Urlings, Lisa BScc; Ford, Nathan PhDd,e; Naniche, Denise PhDf; Rinke de Wit, Tobias F. PhDa; Hermans, Sabine PhDa; Gomez, Gabriela B. PhDg,h

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: August 1, 2021 - Volume 87 - Issue 4 - p 1055-1071
doi: 10.1097/QAI.0000000000002688



Traditional clinic-based care to test, treat, and retain all people living with HIV (PLHIV) poses a challenge to constrained health systems, especially in sub-Saharan Africa (SSA).1 Innovative service delivery options are necessary to scale up and support favorable long-term outcomes of antiretroviral therapy (ART). By tailoring services according to client clinical profile, differentiated service delivery (DSD) offers a practical alternative.2 Overall, the goal of DSD is to decrease barriers in access to care and to guarantee the quality of services at reasonable costs to the health care system. Several DSD interventions have been implemented since the 2000s and show encouraging programmatic and clinical outcomes.3–5 DSD anchors on 4 pillars: the person who provides care [“who”; health care workers (HCWs), doctors, nurses, community health workers (CHWs), peers, etc.], the location of care (“where”; clinic or community), the frequency of care (“when”; monthly or multimonthly), and which HIV services are provided (“what”; ART refill, counseling, health screenings, etc.), respectively. DSD models are defined by a combination of one or more of these pillars which are adapted to the local context. The simplest DSD model is one that includes multiple (3–6) months prescriptions and task shifting of ART dispensing tasks from doctors to other HCWs. Currently, 4 main DSD intervention types exist including.6

  • (1) HCW managed groups, eg, adherence clubs in clinics or communities.
  • (2) Facility-based individual models delivered by HCWs eg, fast track refills, six-month appointments, and multimonth scripting.
  • (3) Client-managed groups in communities, eg, community ART groups.
  • (4) Community-based individual models, eg, community drug distribution points, mobile outreaches, and home delivery.

These DSD interventions focus on individuals who are established (stable) on ART, as complicated cases require facility-based individualized care. This group became the priority for service delivery innovations such as adherence clubs and down-referrals from hospitals to clinics in South Africa, to community ART groups in Mozambique.3,7–10 The defining criteria for being established on ART are in constant evolution and vary by setting, adapted by national HIV programs. WHO currently defines being established on ART as receiving ART for at least 6 months, no current illness (which does not include well-controlled chronic health conditions), good understanding of lifelong adherence, adequate adherence counseling provided, and evidence of treatment success (preferably at least one suppressed viral load result within the past 6 months).

As countries increasingly adopt DSD models that show encouraging results, it is necessary to assess where to focus efforts to enhance the sustainability of these models. In this review, we aimed to evaluate the sustainability of DSD interventions using a comprehensive framework and to assess whether variations in the definition of individuals who are established on ART influence outcomes and sustainability of DSD interventions.


The methods for this review have been published elsewhere.11 In brief, we systematically searched and identified studies in the English language about ART delivery interventions to individuals established on ART in SSA conducted between January 2000 and November 2019. The search was conducted in PubMed and EMBASE using terms including differentiated care, decentralized care, community ART, task shifting, SSA, and HIV program descriptors. Reference lists of included articles were also searched. Inclusion and exclusion criteria are summarized in Box 1.

Box 1.

Eligibility Criteria


  • Observational, qualitative, experimental, or quasiexperimental studies.
  • Studies involving stable adult ART clients accessing HIV care in SSA.
  • Studies describing or assessing HIV services delivered through models other than standard clinic-based care
  • Studies which compare the performance of these other service delivery models with standard clinic-based HIV service delivery accessed by other clients. Although, lack of this comparison is not an exclusion criterion.


  • Reviews, editorials, protocol studies, and clinical guidelines
  • Studies describing or assessing interventions focussed on special population groups eg, adolescents, children, pregnant women, men who have sex with men, commercial sex workers, etc
  • Studies using data retrospectively collected in electronic databases with little description of the actual intervention

Sustainability Definition and Framework

We used the following comprehensive definition of sustainability: “after a defined period—especially after initial funding, the program, clinical intervention, and/or implementation strategies continue to be delivered and/or; individual behavior change (ie, clinician, client) is maintained; the program and individual behavior change may evolve or adapt while continuing to produce benefits for individuals/systems.”12 We adopted the consolidated framework for sustainability constructs in health care to structure our evaluation. The framework was developed in 2018 based on a systematic review of sustainability frameworks, tools, and models in health care. It is the first framework to consolidate a comprehensive list of sustainability constructs which are organized into 6 broad domains with 40 individual constructs which may influence sustainability outcomes. The framework provides a mechanism to conceptualize and analyze sustainability data. To test validity and understand how representative the framework is of diverse health care interventions and settings, it has been tested within various health care studies.13–19 We seek to add to this literature by testing it within this work, with the constructs and domains providing a simple set of evaluation questions that formed the evaluation benchmark (see File 1, Supplemental Digital Content,

Study Outcomes

The primary outcome was an overall “sustainability score.” This was calculated by summing the scores assigned to the 40 consolidated framework for sustainability constructs for each study and DSD intervention type (see calculation details below). In addition, the median of the scores was estimated for the 6 domains per intervention. All construct and domain names are present in this article in italics. The secondary outcomes included (1) a descriptive summary of the main study outcome measures (eg, retention in care, viral suppression, loss-to-follow-up, and client-related or provider-related costs), (2) narrative synthesis of qualitative outcomes (eg, clients and HCW perspectives about DSD interventions and challenges), and (3) sensitivity of sustainability scores per intervention to the cut-off points (see data synthesis section below) with variations in definitions of individuals established on ART used across studies.

Quality Appraisal

The risk of bias was assessed using the Downs and Black checklist for quantitative studies20 and the Joanna Briggs Institute (JBI) checklist for qualitative studies.21

Data Extraction and Synthesis

We adapted a scoring method using a pretested checklist to assign scores per construct ranging from 1 to 3: 1, little to no evidence; 2, some or moderate evidence; and 3, sufficient evidence that the construct was realized.11,22

The scores assigned aimed to facilitate the prioritization of constructs and domains based on frequencies.23 There are no standards published for ranking sustainability. Studies measuring similar complex constructs in health interventions used mean scores or percentage scores, eg, >75% as cut-offs to determine performance and trends.24–26 We considered percentages an appropriate measure to rank constructs since we calculated total scores, and median to rank domains since the 3-point scores were non-normally distributed. Per intervention, we calculated an overall percentage of the total score possible across all constructs (3 × 40 = 120). We derived 3 cut-offs to indicate sufficient evidence for at least 75%, 50%, and <50% of all constructs which we deemed set a high standard for our evaluation. This translated into total scores falling (1) within the highest percentile ≥108, (2) between the eighth and highest percentile (≥99.6–107), and (3) below the eighth percentile (<99.6). Consequently, we regarded percentage total scores as follows: ≥90%: likely sustainable, ≥83%: potentially sustainable, and <83%: less likely to sustain. Similarly, we regarded a median domain score ≥2.7 for each unique DSD intervention as suggestive of being likely sustainable, ≥2.2 as suggestive of potentially sustainable, whereas <2.2. was regarded as less likely sustainable. R version 4.0.3 was used for analysis and visualized using the ggplot2 package.27,28 A narrative synthesis was conducted by using thematic analysis to summarize sustainability facilitators and challenges.

Sensitivity Analysis

To determine the impact of varying definitions of individuals established on ART, we assumed a minimum set of criteria (specifying CD4 count or VL, and months on ART or adherence status) as the base definition for individuals established on ART. Studies specifying additional criteria (about, eg, opportunistic infections, weight, adherence, residence) were categorized as base+, and studies with no definition as base−. The sensitivity of sustainability scores to these 3 categories was analyzed by constructing a forest plot setting the cut-off score of 83% as indicative of the potential for sustainability to assess for trends.


This systematic review was registered on the PROSPERO database; number CRD42019120891.


Characteristics of Included Articles

Of 3088 publications identified by our search, 34 articles were included reporting 39 different DSD interventions across 10 SSA countries (Fig. 1). South Africa, Malawi, and Mozambique contributed about 75% of included articles. Characteristics of included studies are summarized in Table 1, and full data extraction details are available in File 2, Supplemental Digital Content, Articles were published between 2010 and the end of 2019, describing interventions started between 2006 and 2018, except for one intervention that started in 2001.29 Most studies were observational cohorts 41% (16/39), followed by qualitative studies (including realist evaluations, 10/39) and experimental studies (3 cluster-randomized studies, 1 quasi-experimental study, and 1 pragmatic open-label study, 5/39). The remainder were mixed-methods studies, program evaluations, and cost-effectiveness studies. Studies with a comparison group comprised 44% (17/39), whereas the remainder was descriptive. Finally, 51% of studies were conducted in single sites.

PRISMA diagram. Description of search strategy and article retrieval.
TABLE 1. - Characteristics of Articles Included in Review
Author/Year (Ref) Intervention Site/Town/Country DSD Type DSD Start Study Design Study Aim
Bango F, 20169 Ubuntu clinic, Khayelitsha, Cape Town, South Africa AC 2007 Cost-effectiveness analysis (CEA) and access analysis (AA) From a provider's perspective, (i) to assess the cost-effectiveness of clubs in comparison with standard of care and (ii) to present perceived accessibility differences associated with each model of care.
Bekolo C, 201730 Matam out-patient clinic, Conakry, Guinea SMA 2013 Comparative cohort study Report a 6-monthly appointment for clinic and drug refill adapted locally as Rendezvous de Six Mois (R6M) for stable HIV patients receiving ART, as a decongestion scheme to relieve pressure on its overstretched referral Centre of Matam in Conakry and to improve retention in care during the Ebola outbreak
Bemelmans M, 201431 Chiradzulu Malawi; Khayelitsha, South Africa; Kinshasa, Congo; Tete, Mozambique. SMA 2008 Retrospective cohort study Describe several community-supported models of ART delivery developed by Medecins Sans Frontieres (MSF) together with Ministries of Health (MoH) in public health facilities in sub-Saharan Africa
Bochner AF, 201932 10 facilities—2 rural hospitals, 6 rural clinics, and 2 urban clinics in 5 provinces of Zimbabwe CARG 2018 A qualitative evaluation Evaluate the perceived effects of the CARG model for both HCWs and ART clients.
Bock P, 201933 1 PHC and 3 CAC in Cape Winelands district, South Africa AC 2014 A retrospective cohort analysis Determine clinical outcomes among ART clients attending adherence clubs and client experiences and health care worker perceptions of factors key to successful adherence club implementation in the Cape Winelands District, South Africa.
Brennan A, 201134 Themba Lethu Clinic/Crosby Clinic, Johannesburg, South Africa DR 2007 Comparative Cohort study Compare 1-year treatment outcomes among individuals down-referred for treatment maintenance at a nurse-managed PHC to patient's eligible for down-referral who remained at the doctor-managed treatment-initiation site
Decroo T, 201135 12 facilities in 6 districts of Tete Province, Mozambique CAG 2008 Observational cohort study Describe the implementation of the community ART group (CAG) model and report preliminary outcomes
Decroo T, 201436 Peri-urban, district, and rural clinics in Tete Province, Mozambique CAG 2008 Retrospective program evaluation Analyze long-term retention in CAG, estimate individual-level and CAG-level risk factors associated with attrition, and describe the circumstances in which CAG members died.
De Jager GA, 201837 14 PHCs in Eden district, Western Cape, South Africa AC 2013 Analytical cross-sectional study Investigate treatment adherence and patient satisfaction of stable patients living with HIV on ART in ART adherence clubs and clinics
Fox MP, 201938 AC 24 PHCs in 4 provinces (Gauteng, North West, Limpopo, and KwaZulu Natal), South Africa AC 2015 Unblinded cluster-randomized evaluation for AC; Evaluate retention and viral suppression in AC and DMD compared with standard clinic-based care
Fox DMD DMD Observational study for DMD
Geldsetzer P, 201839 18, 16, and 14 facilities in Temeke, Kinondoni, and Ilala municipalities, Dar es Salam, Tanzania HD 2016 Cluster randomized trial Determine whether an ARV community delivery model (lay health workers deliver ARVs to the homes of patients who are clinically stable on ART and nurses and physicians deliver standard facility-based care for patients who are clinically unstable on ART) leads to a lower or equal (noninferior) risk of virological failure compared with the standard of care (standard facility-based care for all ART patients).
Grimsrud A, 20147 Community Health Centre (CHC) Gugulethu, Cape Town, South Africa DR 2006 Comparative cohort study Compare a nurse-managed, decentralized model of care for stable ART patients with a doctor-managed ART clinic, for patients receiving ART in primary care in Cape Town, South Africa
Grimsrud A, 201540 Hannan Crusaid Treatment Centre (HCTC), CHC Gugulethu, Cape Town South Africa CAC 2012 Descriptive study Describe the implementation, early outcomes, and lessons learned from the community adherence clubs (CAC)s given the limited evidence base for community-based models
Grimsrud A, 201641 Community Health Centre (CHC) Gugulethu, Cape Town South Africa CAC 2012 Comparative cohort study Describes outcome loss to follow-up (LTFU) and viral rebound over the first 18 months of CAC implementation in Cape Town, South Africa, and compares patient outcomes under the CAC model of care to those of patients managed in facility-based primary care
Hanrahan CF, 201842 Witkoppen Health and Welfare Centre Johannesburg, South Africa AC 2014 A pragmatic, open-label, parallel randomized controlled trial Compare the effectiveness of community- versus clinic-based adherence clubs concerning loss from club-based care and viral suppression
Long L, 201143 Themba Lethu Clinic/Crosby Clinic, Johannesburg, South Africa DR 2008 Quasiexperimental Evaluate the implications of this down-referral strategy for treatment outcomes and costs
Luque-Fernandez MA, 201344 Ubuntu clinic, Khayelitsha, Cape Town, South Africa AC 2007 Retrospective cohort evaluation Evaluate the effectiveness of adherence clubs compared with traditional clinic-based care in maintaining or improving long-term retention-in-care and virologic suppression
Mantell JE, 201945 3 clinics in 2 rural districts in Mashonaland Central and Mashonaland West Provinces, Zimbabwe CARG 2014 An exploratory qualitative study Identify facilitators and barriers to CARG participation by HIV-positive men, with inputs from recipients of HIV care, community members, HCWs, donors, and policymakers
Mudavanhu M, 201946 Witkoppen Health and Welfare Centre Johannesburg, South Africa AC 2014 A mixed-methods study Explore patient acceptability and attitude toward the community and clinic-based adherence clubs
Mukumbang FC, 201847 Western Cape District Hospitals (WCDOH), South Africa AC 2011 Realist evaluation (Case study) Test the hypothesis (the initial program theory) of the adherence club to validate, reject, or modify the initial program theory. To obtain a refined program theory of the adherence club intervention based on the operation of the intervention in the identified primary health care facility
Mukumbang FC, 2019_SAJHIVM48 1 Provincial PHC in Western Cape province, South Africa AC 2014 Retrospective cohort analysis and an explanatory qualitative approach Test a theory on how and why the adherence club intervention works and in what health system context(s) in a primary health care facility in the Western Cape Province
Mukumbang FC, 2019_Plos149 1 PHC in Mitchell's Plain, Cape Town, South Africa AC 2012 Retrospective cohort analysis and an explanatory qualitative approach Unravel the mechanisms explicating how, why, for whom, and in what circumstance the adherence club program works at a community health center in Cape Town
Pasipamire L, 201850 (CAG) 16 Primary care centers in the Shiselweni region, Swaziland CAG 2015 Program evaluation (Retrospective analysis Compare retention in care model and retention on ART among 3 care models, ie, CAG, Outreach, and Treatment clubs and to determine factors associated with all-cause attrition.
Pasipamire Outreach 1 Primary and 1 secondary care facility OR 2015
Pasipamire AC 1 large health centre AC 2015
Pellecchia U, 201751 Mikolongwe Health Centre and Khonjeni Health center, Thyolo, Malawi CAG 2012 Qualitative study Report the findings of a qualitative study to assess the perceived benefits and limitations of CAGS from a patient and a health care worker (HCW) perspective.
Prust ML, 2018,5 CAG 30 heterogeneous sites in Malawi—8 CAGs CAG 2012 Qualitative study Describe the qualitative component of the process evaluation that explored patients and provider perspectives on the key benefits and challenges associated with models of differentiated care for stable patients
Prust FTR 30 heterogeneous sites in Malawi—4FTR FTR 2012
Prust MMS 30 heterogeneous sites in Malawi MMS 2012
Rasschaert F, 201452 20 clinics in Tete province, Mozambique CAG 2008 Qualitative study Assess the relevance, the dynamic, and the impact of CAG
Rasschaert F, 201453 MSF Project Tete province, Mozambique CAG 2008 Qualitative evaluation Highlights the components, which might facilitate and/or jeopardize the sustainability of the CAG model, and formulates recommendations to guarantee its long-term sustainability
Selke HM, 201029 Mosoriot rural health centre, Kosirai, Kenya HD 2006 Community randomized clinical trial Evaluate the clinical outcomes of patients enrolled in an innovative HIV care delivery system which used PLWAs as community care coordinators (CCCs), aided by an electronic decision support tool, to deliver medications and provide follow-up care to patients on ART in the community
Sharp J, 201954 Ubuntu clinic, Khayelitsha, Cape Town, South Africa AC 2012 A descriptive retrospective cohort study Describe the outcomes of patients referred directly to ACs after viral suppression after specific adherence support
Tsondai PR, 20174 Cape town health district, South Africa AC and CAC 2007 A retrospective observational cohort study Describe and explore possible predictors of LTFU and viral rebound for a representative sample of patients receiving their ART within ACs in Cape Town, South Africa
Vandendyck M, 201555 Health Centre (HC) Nazareth clinic, Roma District, Lesotho CAG 2012 Mixed methods Study how CAG dynamic was perceived by different stakeholders, and study retention among patients in conventional care and CAG members in HC Nazareth.
Venables E., 201956 Ubuntu ART clinic, Khayelitsha and Gugulethu CHC, Western Cape Province AC 2016 A qualitative study 1. Explore perceptions of ACs among former and current AC members, as well as those who had never joined a club, in 2 settings in Cape Town, South Africa, including the perceived advantages and disadvantages of the differentiated model mechanisms. 2. Explore the experiences of patients referred out of ACs back to routine clinical care
Vogt F, 201757 Kabinda Referral Hospital, Kinshasa, DRC CDDP 2010 Cohort study Assess outcomes and risk factors for attrition after decentralization in this project
Wringe A, 201858 District Hospital and 10 health centers in Chiradzulu, Malawi SMA 2008 a retrospective cohort analysis Describe long-term retention in care, and risk factors for attrition from care among clinically stable ART patients accessing SMCC over the period from 2008 to 2015. To estimate the number of clinic appointments “saved” as a result of SMCC
AC, facility-based adherence clubs; CAC, community-based adherence clubs; CAG, community ART group; CARG, community ART refill group; CDDP, community drug distribution point; DMD, decentralized medication delivery; DR, down referral (DR) from hospital to PHC; FTR, fast track refills; MMS, multimonth scripting; OR, outreach; SMA, 6-monthly appointment; SMCC, 6-monthly clinical consultation; LTFU, lost-to-follow-up.

Risk of Bias in Included Studies

Over 90% of quantitative studies had a high to moderate risk of bias (see File 3, Supplemental Digital Content, The risk of bias in qualitative studies was considered moderate. All qualitative studies did not state the philosophical perspective from which the studies were conducted or the theoretical perspective of the researcher within the research.

Characteristics of Included Interventions

Of the 39 interventions (Table 2), adherence club (41%) and community ART group (20%) were the most commonly reported intervention types.

TABLE 2. - DSD Intervention Type Description
Intervention Type DSD model6 Country Number of Studies in Review n,% (Citation) DSD Intervention Description
Adherence clubs (AC)—clinic and community-based Health care worker managed groups South Africa, Swaziland 16, 414,9,40,41,44,47,50 Groups of 25–30 stable adult ART patients led by a health care worker or peer who meets every 2–3 months within or out of the facility for group counseling and ART refill
Community ART groups (CAG) Client-managed groups Mozambique, Lesotho, Swaziland, Malawi 8, 215,35,36,50–53,55 Self-formed groups of 2–6 stable adult ART patients living within the same geographical area who meet within the community monthly for group counseling and drug distribution. One CAG member visits the clinic monthly to collect ART refill for the group and consultation on a rotational basis so that each member visits the clinic at least once every 6-mo
Community ART refill groups (CARG) Client-managed groups Zimbabwe 2, 532,45 Self-formed groups of 4–12 stable adult ART patients who live and meet within the same community for group counseling. A group member is appointed for a clinical visit every 3-months for drug-refills, whereas the whole group visits the clinic for annual consultation.
Community drug distribution points/decentralized medication delivery (CDDP/DMD) Community-based individual model South Africa, DR Congo 2, 538,57 Peer-led centers within the community where stable adult patients come for 3-monthly ART refills with 1-yearly clinical consultation at the facility
Down-referral (DR) Facility-based individual model South Africa 3, 87,34,43 Referral of stable adult ART patients from secondary health facilities to primary health centers for the continuation of care, one of the first models tried.
Home delivery (HD) Community-based individual model Tanzania, Kenya 2, 529,39 Monthly delivery of ART by CHW to stable adult patients at home or any other location within the community
Multimonth scripting (MMS) Facility-based individual model Malawi 1, 35 Three-monthly clinical consultation with drug refill at the health facility
Outreach (OR) Community-based individual model Swaziland 1, 350 ART drug refill integrated into existing outreach programs held in the community
Six monthly appointment/fast track refill (SMA/FTR) Facility-based individual model Malawi, Guinea 4, 105,30,31,58 Six-monthly clinical consultation with 3-monthly ART refill by CHW

The intervention types adherence club and down referral were conducted mostly in urban settings except for one adherence club intervention.50 Of the 2 community drug distribution point interventions, one was conducted in a large urban center, whereas the other was a multicenter study spanning both rural and urban settings.57 Community ART groups, community ART refill groups, and outreach interventions were predominantly in rural settings. Primary care providers in over 85% of interventions were peers and/or lay health care workers (Lay-HCW). Interventions were mostly funded externally, apart from locally funded adherence clubs in South Africa, a situation reflecting the funding status of HIV programs in general.

Outcome measures reported included 1-year retention (in 20/29 studies, of which 8 were comparative), 1-year viral suppression and 1-year loss-to-follow-up (in 10 studies each), and 1-year mortality (in 8 studies). Three studies reported costs, of which only 1 article reported both provider and client-incurred costs.9,38,43

Overall Sustainability Scores per Intervention Type and Study

Across the 9 DSD intervention types, sustainability scores ranged from 67.5% for home delivery in Kenya to 95.8% for adherence clubs in South Africa. DSD interventions implemented in South Africa, for example, adherence clubs and decentralized medication delivery, scored the highest (94.2%–95.8%), followed by the community ART groups implemented in Mozambique (93.3%) and those in Lesotho (90.8%).4,9,35,44,51,52,55 Conversely, interventions with minimal engagement of key stakeholders, for example, home delivery, 6-month appointments, and down referral, scored the lowest with 67.5%, 71.8%, and 73.3%, respectively. The scores per study are shown in Table 2, Supplemental Digital Content,

Sustainability Scores—Top-Scoring Constructs

The top 12 constructs scored between 95% and 100%. Evidence for sustainability was reported across all interventions for 6 constructs, namely, value system, no opposition, problem awareness, project type, expertise, and client-related outcomes (see details in Fig. 2A).

Scoring of sustainability constructs including (A) the highest and (B) the lowest scoring constructs of 40 constructs assessed. The 3 numbers displayed for each bar in the 2 figures from left to right represents (1) percentage of all DSD interventions evaluated reporting little/no evidence, (2) percentage reporting moderate, and (3) percentage reporting sufficient evidence that the construct was achieved. *Response refers to the score as described in methods: (1) little evidence, (2) moderate evidence, and (3) sufficient evidence.

Sustainability Scores—Least Scoring Constructs

The least 12 scoring constructs scored between 15% and 53%. Sustainability constructs were least reported in the following areas (Fig. 2B): community participation and the involvement of program champions (15% of studies); community awareness of interventions and clients involvement in intervention design, planning, and processes—23% of studies7,36,52; a shared goal developed with all stakeholders, resources, and adequate funding necessary to continue interventions (30% of studies); other constructs with minimal evidence included ownership of interventions, infrastructure, satisfaction among staff, power, and readiness to continue delivering the intervention with little or no external support.4,9,40

Constructs Not Described Across Studies

Over 50% of studies provided some form of evidence for all constructs. Of the remainder, 14 studies did not describe between 1 and 3 constructs. The involvement of champions was the construct most frequently not described (13%, 33.3%). Community awareness followed a similar trend. Other constructs not commonly described include readiness (30%), funding (25%), community participation (13%), roles/responsibilities (13%) job description (10%), and shared goals (10%).

Comparing Sustainability Domains Across DSD Intervention Types

In Figure 3, we present the median and interquartile ranges for domain-specific scores per intervention type. In Table 3, we describe the domain scores in detail. The figure highlights the relative performance of the 6 sustainability domains across the 9 unique DSD intervention types. Median scores are highlighted as 6 separate boxplots with different colors representing the 6 domains per DSD intervention type.

Plot of median scores (with IQR???) obtained per domain for the 9 unique DSD intervention types in the review. The DSD intervention types include AC, ie, adherence clubs both facility and community based; CAG, ie, community ART groups; CARG, ie, community ART refill groups; CDDP, ie, community drug distribution points; DR, ie, down-referral; HD, ie, home delivery; MMS, ie, multimonth scripting; OR, ie, out-reach; and SMA, ie, 6-monthly appointment. Plotted scores were derived by obtaining the average of total construct scores making up each of the 6 domains as indicated by the colors, ie, (orange) intervention design and delivery, (brown) external environment, (green) organizational setting, (teal blue) people involved, (blue) intervention processes, and (violet) resources as described in methods. AC, facility-based adherence clubs.
TABLE 3. - Sustainability Scoring Across all DSD Intervention Types by Domains
Domain Overall Scores Least Scoring DSD Interventions and Constructs within Domain Challenging Top Scoring DSD Interventions and Constructs within Domain Supporting
Intervention design and delivery Range—2.3 to 3.
Four of 9 constructs in this domain scored maximally across all interventions.
Least scoring—SMA, MMS HD, and CAG.
Dependence on donor-driven systems, eg, for monitoring and reliable drug supply makes sustainability doubtful.
High scorers include AC, CAG, DR, and OR.
All intervention types had clear designs, good outcomes, and displayed intended benefits
There were adequate expertise, capacity building, and use of improvement methods.
External environment Range from 2.0 to 3.
Two of 4 constructs in this domain were top scorers
SMA and DR scored least closely followed by MMS, HD, and CDDP
There was generally minimal involvement of the larger community in design and implementation.
CAG and OR were top scorers likely because
the urgency for DSD was acknowledged, and there was strong political support and motivation to explore DSD options.
Many interventions have already been scaled up regionally and countrywide.
Organizational setting Range from 2.2 to 3.
Two of 6 constructs in this domain had maximum scores with another top scoring
HD, DR, and CARG performed the least.
The readiness to sustain the interventions with minimal external support was however doubtful in most countries except South Africa.
AC scored highest. Similarly, CAG, CDDP, SMA, MMS, and OR all score ≥2.7 benchmark
High acceptability, management support, and flexibility to adapt and expand existing resources to reach more patients characterize DSD
DSD objectives align with health system priorities and strategic plans
Observable benefits provide visibility that patients are a priority.
No opposition to DSD was documented.
Intervention process Range from 2.3 to 3.
Two of 7 constructs in this domain were top scorers
AC, CAG, HD, and DR scored the least.
Structures for coordinating and monitoring are still mostly donor-funded
Lack of clarity in roles with the existing system especially in CAG
A shared goal across stakeholders including PLHIV with clear responsibilities for sustainability was reported minimally.
CDDP scored highest followed by AC and CARG
Clients are motivated to participate as models create safe outlets for ART refill even where HIV stigma is high.
The simplicity of interventions and reduced workload promote the buy-in of facility staff.
Many countries have updated guidelines and developed other job aides to promote DSD
Resources Range from 1.2 to 3.0.
Four of 5 constructs in this domain scored least
Least performing across all models in this review.
HD, DR, CAG, and CARG were all low scoring
Funding, infrastructure, resources, and staff required were mostly provided by an external donor.
AC in south Africa scored highest
Interventions were embedded and implemented within the routine HIV service delivery
Peers engaged as human resource in care provision process reducing the need for highly skilled staff
Reduced client burden in clinics free limited staff to do more work
People Range from 1.5 to 3.0.
Five of 9 constructs in this domain were among the least scoring
DR, HD, and CARG were the least scoring
Patients are mostly involved at the level of implementation and rarely in design and planning.
Power to adapt and will to own limited
Limited report of the use of champions beyond the immediate community of PLHIV across intervention types
CAG scored optimally followed by AC likely because of adequate engagement of key stakeholders in implementation in Mozambique and South Africa
Good collaboration, acceptability, and stakeholder engagement
Client satisfaction as DSD promote active participation with HCW as allies
AC, facility-based adherence clubs; CAC, community-based adherence clubs; CAG, community ART group; CARG, community ART refill group; CDDP—community drug distribution point; DMD, decentralized medication delivery; DR, down referral (DR) from hospital to PHC; FTR, fast track refills; HD, home delivery; MMS, multimonth scripting; OR, outreach; SMA, 6-monthly appointment.

Facilitators of and Challenges to Sustaining DSD Interventions

Across studies, DSD interventions were mostly acceptable to clients and staff alike. Among clients, preference for DSD was explained as representing a convenient option, reduced time spent accessing care, reduced frequency of clinic visits, reduced transport cost, increased peer support, reduced absenteeism from work, material support among members, and improved self-management.5,9,31,35,37,43–47 For staff, the reduced workload was the most commonly cited reason for preference in addition to more effective use of HCW time attending to sicker patients, reduced number of clients lost-to-follow-up, and decongestion of clinic.5,59,60

Fear of stigma because of unintentional disclosure by participating in the intervention was a recurrent challenge across studies. Other challenges were mostly health system–related including protocol violations because of pressure from clients to be enrolled, staff shortage, long viral load result turn-around-time, poor documentation and data quality, inadequate understanding of intervention by staff, insufficient supervision, ARV and cotrimoxazole stock-out, differences in implementation across facilities, inadequate awareness, and low community involvement, inefficient drug supply chain, restrictive policies limiting the roles of CHW, unconducive venue for intervention, and lack robust monitoring systems.5,29–31,34,36,39,40,50,51,55 A few client-related challenges were reported, for example, group conflicts and low male participation.

Sensitivity Analysis

The definitions of stable on ART varied across included studies (see File 4, Supplemental Digital Content, Seven interventions used the base definition, whereas 4 (10%) did not give a specific definition (Base−) and 28 (72%) provided a more detailed definition (Base+). We did not observe any trend in the likelihood of sustainability when comparing across the 3 categories (Fig. 4).

Sensitivity analysis showing sustainability scores vs studies applying varying detail in criteria used for stable patient definition. Base = stable patient definition given by specifying CD4 count or VL, and months on ART specified. Base+ = base stable definition plus other criteria specified, for example, opportunistic infection, weight, adherence, and residence. Base− = stable patient not defined or no CD4/VL specified. Benchmark = score above which DSD is likely to be sustainable. The teal blue, green, and red dots represent the sustainability scores per DSD intervention. Multiple dots per intervention types represents the scores for included studies. The purple dots represent the benchmark score above which we assumed that interventions are likely to be sustained.


Our review shows that DSD interventions targeting HIV-positive individuals established on ART in sub-Saharan Africa may be sustainable but may require additional support in aspects such as resources and stakeholder involvement to enhance sustainability. Indeed, we found that DSD interventions were potentially sustainable for the domains design and delivery, organizational setting, external environment, and intervention process. The domains of resources and people involved, however, received lower sustainability scores.

The comprehensive definition by Moore et al, and the framework by Lennox et al, captures the complexity of the sustainability concept and enabled our multidimensional examination of sustainability.12,13 Although distilled primarily with evidence from high-income countries, the constructs and domains proposed by the framework remain relevant in low-income and middle-income settings. First, it is recognized that similar processes drive sustainability across settings,61 and second, the domains align with key areas highlighted in the sustainability discourse about donor-funded interventions in Africa.62–65 The framework allowed us to expand sustainability assessment from a sole resource perspective to a more comprehensive view of program continuation.66 Financing considerations no doubt matter when discussing other sustainability domains in SSA where HIV programs rely heavily on external funding. However, securing funding has been shown not necessarily translate to systemic efficiency or deal with other drivers of sustainability such as social norms and practices.66 Domains are not mutually exclusive, although they are presented separately. We proceed to discuss our findings across domains adopting a broad view.

Intervention Design and Delivery

This was the highest-scoring domain. DSD interventions show effective client-related outcomes, adequate problem awareness and expertise, sufficient evidence, and appropriate design types, all of which are factors that enhance sustainability.67,68 Other constructs in this domain, that is, capacity building, improvement methods, and project duration, were reported as accomplished by most studies, whereas the existence of structured monitoring systems was instrumental to highlighting achievements toward the 90-90-90 targets.46,58,69

External Environment

This domain relates to laws and policies impacting vulnerability to HIV (eg, among sex workers, lesbian, gay, bisexual, and transgender and access to HIV-related services (ie, demand and supply))70 and scored moderately. Unfavorable legislation targeting vulnerable groups has been documented in several countries such as Eswatini, Tanzania, Kenya, and Malawi.71–73 Reassuringly, nearly all studies score high for political support with reports of policy updates to incorporate DSD and enabling requirements into national HIV strategic plans, for example, Mozambique, DRC, South Africa, Eswatini, and Zimbabwe.4,52,57 The limited community awareness (besides the PLHIV and HCW directly involved) is likely linked to the sociocultural context within which interventions are implemented and the behavior and perspectives of stakeholders in which stigma cannot be overlooked.5,45,46 Stigma is a complex social construct that remains an issue.74–77 Pantelic et al78 recommended prioritizing a combination of interventions at different levels to tackle stigma. Although participants acknowledge the benefit of DSD, fear of unintentional disclosure was a recurrent reason for nonparticipation.51,55 Poverty is another relevant structural issue which not only increases vulnerability to HIV but also limits access to services.79,80 Gender-based violence, gender inequality, and cultural beliefs that condone oppressive male dominance disproportionately affect women.81 Despite these, included studies across countries report the spread of DSD interventions to multiple sites within districts and regions.5,32,37,43,50,58

Organizational Setting

The deliberate inclusion of DSD interventions in the policies and guidelines of primary care systems was evident across studies. In our review, the organizational setting domain provided most constructs that enabled the evaluation of DSD sustainability, alongside domains such as intervention process and resources. Most health system values and culture align with the DSD strategy which may explain management support and having no opposition reported in studies. Also, adaptations that optimize intervention's fit within the environment were frequently reported. CHW-driven programs such as DSD are poorly integrated into the formal health system in many African countries.82–84 Apart from South Africa, studies show that governments have been unable to facilitate this integration.50,53 DSD models require health systems with readiness and capacity adapted to community services. Paper-based data management systems are common across Africa and do not support the level of tracking required in DSD.29,43,52,55,85 Investment in a robust electronic health information system is desirable for ensuring the retention of clients who receive ART out-of-facility. Conversely, the large quantities of ART dispensed at once to PLHIV in DSD demand an efficient logistics system to prevent stock-out. Six-month appointments, fast-track refills, multimonth scripting, and adherence club interventions although successful, were mostly funded externally which poses a risk for sustainability.5,30,31


The people domain scored poorly across interventions. The extent of participation, ownership, collaboration, and power exercised by stakeholders, in DSD interventions were found to be generally less than optimal. Apart from adherence clubs in South Africa and community ART groups in Mozambique which reported evidence of stakeholder engagement and good client involvement, PLHIV participation appeared passive. Although there were reports involving networks of PLHIV, community awareness and involvement of the larger community where these interventions were implemented was rarely reported. Even within facilities, an adaptation of the adherence clubs which was integrated with other chronic diseases faced similar challenges.47 Despite evidence that program champions help reinforce positive behavior, there was little report of their engagement in our review.86 A right-based approach to health, as promoted by WHO, is desirable and involves meaningful stakeholder participation to guarantee the values and preferences of beneficiaries are incorporated in the design, planning, implementation, monitoring, and evaluation of any intervention.87


This domain performed moderately and involves processes necessary for the continued delivery of interventions. DSD interventions show minimal complexity to implement. They reveal incentives including perceived reduction of workload which may explain the wide acceptance and belief in the interventions. Simplicity and belief in the value an intervention add to sustainability.88 Ambiguity in roles and responsibilities and not updating job descriptions to reflect current roles were challenges especially the community ART groups.5,51 Having a shared vision among stakeholders is advocated for sustainable DSD but poorly described in studies.62,89,90


This was the lowest scoring domain. Funding underlies most other aspects of sustainability, for example, staff, infrastructure, all of which were currently supported by donor funds. A system sensitive to changes in epidemiological trends (especially within subgroups) will inform targeted interventions and facilitate sustainability. DSD interventions report encouraging retention rates and close monitoring but realizing the full potential of DSD requires funds,4 stigma reduction,47,51 and establishment of new management structures.9,44 The call for shared responsibility by the UN in agreement with the African Union is a step in the right direction and has facilitated an increase in domestic investment in HIV programs.91

Recommendations for Sustainable DSD Interventions

A clear vision for institutionalizing DSD, innovative monitoring as PLHIV remotely access various DSD services, and capacitating the health system with basic human and material resources will be required to facilitate DSD sustainability. In addition, materializing universal health coverage, leveraging the influence of opinion leaders, and tapping into local partnerships will all be crucial to sustaining DSD. (Table 4).

TABLE 4. - Recommendations to Support the Sustainability of DSD Interventions
Domain Recommendation
Design Innovate and monitor—an investment in electronic health information systems will facilitate adequate DSD monitoring and tracking of all PLHIV as they maintain minimal contact with the health system. Portable devices, eg, mobile phones, PDAs, etc., can be programmed for real-time data collection with appropriately validated decision support algorithms for service delivery to absolve LCHW from making clinical decisions requiring more trained personnel.29,92
External environment Communicate and collaborate—the relevance of DSD should be conveyed to opinion leaders within the community where DSD is conducted to raise awareness and tap into appropriate local capital and partnerships beyond the health sector crucial for sustainability.89,90
Organization Integrate and capacitate—the need to formalize roles already played by LHCW and to recruit and train them in adequate numbers is advocated to secure current achievements34,51 and support initiative continuity.93–95 Capacitate health system with an efficient and reliable supply chain management system to minimize stock-outs96 and conduct routine surveillance of drug resistance as an integral part of HIV programs.1,97
People Engage and empower—engaging influential people as champions at different levels to reinforce messages will promote the acceptance of HIV as a normal disease and support a positive attitude toward HIV necessary to curb stigma within the larger society.98–100
Process Assign and define—having a clearly articulated vision shared among different stakeholders involved in DSD and clarifying roles and responsibilities will streamline activities and stimulate suitable adaptations necessary to institutionalize DSD.62
Resources Horizontalize and diversify—although global financing mechanisms such as the global funds, PEPFAR will remain relevant, it will be crucial to explore horizontalizing such vertical funds into general risk pools, operationalize “shared responsibility” as articulated by the AU, and scale up universal health coverage (UHC).101,102


DSD entails an assortment of interventions in the literature, and it is possible our search missed some relevant articles. However, the wide variety of search terms used aimed to describe many known DSD terminologies likely minimized the articles missed. The community ART groups and adherence clubs being the most implemented may have biased our findings. Because we could only evaluate items that were included in published reports, other sustainability constructs could have been fulfilled, but not reported in the publication. Nonreport of constructs implies that the sustainability scores calculated may have been underrated as a result. This likely had minimal effect on our findings because nonreport followed a random pattern across studies. This review did not assign weights to the individual sustainability domains, which may have influenced the conclusions.66 Limited evidence suggests that domains rank differently in importance in the sustainability of community-based programs.86,103 The nonuniform domains used in different studies however by make extrapolation challenging. Similarly, we used cut-offs which we assumed set the standard high to estimate which construct, domain, or overall score was indicative of sustainability. We conducted a sensitivity analysis to investigate trends in sustainability with variations in definitions of individuals established on ART among DSD interventions. Most studies included were observational in design, and therefore, we can draw no firm conclusions on causality as a result. Our evaluation, we believe nevertheless provides useful constructs and domains to consider for DSD sustainability.

By using an existing framework, this review complements existing sustainability research and moves the discourse from theory to practice. Future sustainability research will benefit from leveraging this framework to build consensus on if a minimum set of sustainability constructs can be developed, validating constructs by weighting according to relevance and significance and recommending benchmarks.104 Such a standard toolkit could provide the basis for measuring and comparing the sustainability of interventions across settings. There has also been debate in the literature about the extent to which complex phenomena can be described and understood by lists of constructs or factors alone.105–107 Therefore, future work should also explore the dynamic nature of sustainability constructs and the interaction between them.


We reviewed DSD interventions to identify, score, and rank the constructs and domains reported in included studies which can be used to estimate the likelihood of sustainability. The community ART groups and adherence clubs were found to be most likely to be sustainable. With the right investment, DSD models were generally observed to be potentially sustainable. This work provides insight into how specific constructs and domains support or hinder the sustainability DSD of different DSD types. Our results provide a resource that policymakers can use to inform decisions about which DSD intervention to implement based on their potential sustainability.


The authors thank Rene Spijker of the Amsterdam UMC library for his contribution in conducting both the initial and final search for articles included in this review.

Although this research was funded by the Erasmus Mundus Joint Doctorate (EMJD) Fellowship program of the European Union (EU), the views expressed in this article are those of the author(s) and not necessarily those of the EMJD or EU.

This research was supported by the NIHR Applied Research Collaboration Northwest, London. The views expressed in this publication are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.


1. World Health Organization. Guidelines Guideline on When to Start Antiretroviral Therapy and on Pre-exposure Prophylaxis for HIV. Geneva, Switzerland: World Health Organization; 2015.
2. International AIDS Society I. Differentiated Service Delivery Home Page. Available at: Accessed February 16, 2021.
3. Decroo T, Rasschaert F, Telfer B, et al. Community-based antiretroviral therapy programs can overcome barriers to retention of patients and decongest health services in sub- saharan africa: a systematic review. Int Health. 2013;5:169–179.
4. Tsondai PR, Wilkinson LS, Grimsrud A, et al. High rates of retention and viral suppression in the scale-up of antiretroviral therapy adherence clubs in Cape Town, South Africa. J Int AIDS Soc. 2017;20(Suppl 4):51–57.
5. Prust ML, Banda CK, Nyirenda R, et al. Multi-month prescriptions, fast-track refills, and community ART groups: results from a process evaluation in Malawi on using differentiated models of care to achieve national HIV treatment goals. J Int AIDS Soc. 2017;20(suppl 4):21650.
6. Differentiated IAS. Care for HIV: a decision framework for antiretroviral therapy delivery. Differ (International AIDS Soc). 2016:1–56.
7. Grimsrud A, Kaplan R, Bekker LG, et al. Outcomes of a nurse-managed service for stable HIV-positive patients in a large South African public sector antiretroviral therapy programme. Trop Med Int Heal. 2014;19:1029–1039.
8. Decroo T, Panunzi I, Das Dores C, et al. Lessons learned during down referral of antiretroviral treatment in tete, Mozambique. J Int AIDS Soc. 2009;12:6.
9. Bango F, Ashmore J, Wilkinson L, et al. Adherence clubs for long-term provision of antiretroviral therapy: cost-effectiveness and access analysis from Khayelitsha, South Africa. Trop Med Int Heal. 2016;21:1115–1123.
10. Waldrop G, Doherty M, Vitoria M, et al. Stable patients and patients with advanced disease: consensus definitions to support sustained scale up of antiretroviral therapy. Trop Med Int Heal. 2016;21:1124–1130.
11. Okere NE, Urlings L, Naniche D, et al. Evaluating the sustainability of differentiated service delivery interventions for stable ART clients in sub-Saharan Africa: a systematic review protocol. BMJ Open. 2020;10:e033156.
12. Moore JE, Mascarenhas A, Bain J, et al. Developing a comprehensive definition of sustainability. Implement Sci. 2017;12:110.
13. Lennox L, Maher L, Reed J. Navigating the sustainability landscape: a systematic review of sustainability approaches in healthcare. Implement Sci. 2018;13:27.
14. Koh GA, Askell‐Williams H. Sustainable school‐improvement in complex adaptive systems: a scoping review. Rev Educ. 2020;9:281–314.
15. Walugembe DR, Sibbald S, Le Ber MJ, et al. Sustainability of public health interventions: where are the gaps? Heal Res Pol Syst. 2019;17:1–7.
16. Karsan Y, Anderson C, Boyd MJ, et al. Exploring barriers to the sustainability of an electronic administration system in long-term care facilities: a case study approach. Res Soc Adm Pharm. 2020. doi:10.1016/j.sapharm.2020.08.003.
17. Nadalin Penno L, Davies B, Graham ID, et al. Identifying relevant concepts and factors for the sustainability of evidence-based practices within acute care contexts: a systematic review and theory analysis of selected sustainability frameworks. Implement Sci. 2019;14:1–16.
18. Byrne AL, Harvey C, Chamberlain D, et al. Evaluation of a nursing and midwifery exchange between rural and metropolitan hospitals: a mixed methods study. PLoS One. 2020;15:1–22.
19. Ford JH, Gilson A. Influence of participation in a quality improvement collaborative on staff perceptions of organizational sustainability. BMC Health Serv Res. 2021;21:1–14.
20. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Heal. 1998;52:377–384.
21. Joanna Briggs Institute. Critical Appraisal Tools: Qualitative Checklist. Vol 6; 2005. doi:10.1002/9780470988343.app1/pdf.
22. Calhoun A, Mainor A, Moreland-Russell S, et al. Using the program sustainability assessment tool to assess and plan for sustainability. Prev Chronic Dis. 2014;11:130185.
23. Lockwood C, Munn Z, Porritt K. Qualitative research synthesis. Int J Evid Based Healthc. 2015;13:179–187.
24. Luke DA, Calhoun A, Robichaux CB, et al. The Program Sustainability Assessment Tool: a new instrument for public health programs. Prev Chronic Dis. 2014;11:130184.
25. Nurjono M, Shrestha P, Ang IYH, et al. Implementation fidelity of a strategy to integrate service delivery: learnings from a transitional care program for individuals with complex needs in Singapore. BMC Health Serv Res. 2019;19:177.
26. Schaap R, Bessems K, Otten R, et al. Measuring implementation fidelity of school-based obesity prevention programmes: a systematic review. Int J Behav Nutr Phys Act. 2018;15:75.
27. The R Foundation. R: The R Project for Statistical Computing. 2018. Available at: Accessed February 16, 2021.
28. Wickham H. Ggplot2: Elegant Graphics for Data Analysis. 2nd ed. Cham, Switzerland: Springer; 2016. Available at:
29. Selke HM, Kimaiyo S, Sidle JE, et al. Task-shifting of antiretroviral delivery from health care workers to persons living with HIV/AIDS : clinical outcomes of a community-based program in Kenya. J Acquir Immune Defic Syndr. 2010;55:483–490.
30. Bekolo CE, Diallo A, Philips M, et al. Six-monthly appointment spacing for clinical visits as a model for retention in HIV Care in Conakry-Guinea: a cohort study. BMC Infect Dis. 2017;17:766.
31. Bemelmans M, Baert S, Goemaere E, et al. Community-supported models of care for people on HIV treatment in sub-Saharan Africa. Trop Med Int Heal. 2014;19:968–977.
32. Bochner AF, Meacham E, Mhungu N, et al. The rollout of Community ART Refill Groups in Zimbabwe: a qualitative evaluation. J Int AIDS Soc. 2019;22:e25393.
33. Bock P, Gunst C, Maschilla L, et al. Retention in care and factors critical for effectively implementing antiretroviral adherence clubs in a rural district in South Africa. J Int AIDS Soc. 2019;22:e25396.
34. Brennan AT, Long L, Maskew M, et al. Outcomes of stable HIV-positive patients down-referred from a doctor-managed antiretroviral therapy clinic to a nurse-managed primary health clinic for monitoring and treatment. AIDS. 2011;25:2027–2036.
35. Decroo T, Telfer B, Biot M, et al. Distribution of antiretroviral treatment through self-forming groups of patients in Tete Province, Mozambique. J Acquir Immune Defic Syndr. 2011;56:e39–e44.
36. Decroo T, Koole O, Remartinez D, et al. Four-year retention and risk factors for attrition among members of community ART groups in Tete, Mozambique. Trop Med Int Heal. 2014;19:514–521.
37. De Jager GA, Crowley T, Esterhuizen TM. Patient satisfaction and treatment adherence of stable human immunodeficiency virus-positive patients in antiretroviral adherence clubs and clinics. Afr J Prim Heal Care Fam Med. 2018;10:e1–e8.
38. Fox MP, Pascoe S, Huber AN, et al. Adherence clubs and decentralized medication delivery to support patient retention and sustained viral suppression in care: results from a cluster-randomized evaluation of differentiated ART delivery models in South Africa. PLoS Med. 2019:16;e1002874.
39. Geldsetzer P, Francis JM, Sando D, et al. Community delivery of antiretroviral drugs: a non-inferiority cluster-randomized pragmatic trial in Dar es Salaam, Tanzania. PLoS Med. 2018;15:e1002659.
40. Grimsrud A, Sharp J, Kalombo C, et al. Implementation of community-based adherence clubs for stable antiretroviral therapy patients in Cape Town, South Africa. J Int AIDS Soc. 2015;18:19984.
41. Grimsrud A, Lesosky M, Kalombo C, et al. Community-Based adherence clubs for the management of stable Antiretroviral therapy patients in Cape Town, South Africa: a cohort study. J Acquir Immune Defic Syndr. 2016;71:e16–e23.
42. Hanrahan CF, Schwartz SR, Mudavanhu M, et al. The impact of community-versus clinic-based adherence clubs on loss from care and viral suppression for antiretroviral therapy patients: findings from a pragmatic randomized controlled trial in South Africa. PLoS Med. 2019;16:1–15.
43. Long L, Brennan A, Fox MP, et al. Treatment outcomes and cost-effectiveness of shifting management of stable ART patients to nurses in South Africa: an observational cohort. PLoS Med. 2011;8:e1001055.
44. Luque-Fernandez MA, Van Cutsem G, Goemaere E, et al. Effectiveness of patient Adherence groups as a model of care for stable patients on antiretroviral therapy in Khayelitsha, Cape Town, South Africa. PLoS One 2013;8:e56088.
45. Mantell JE, Masvawure TB, Mapingure M, et al. Engaging men in HIV programmes: a qualitative study of male engagement in community-based antiretroviral refill groups in Zimbabwe. J Int AIDS Soc. 2019;22:1–11.
46. Mudavanhu M, West NS, Schwartz SR, et al. Perceptions of community and clinic-based adherence clubs for patients stable on antiretroviral treatment: a mixed methods study. AIDS Behav. 2019;24:1197–1206.
47. Mukumbang FC, Marchal B, Van Belle S, et al. “Patients are not following the [adherence] club rules anymore”: a realist case study of the antiretroviral treatment adherence club, South Africa. Qual Health Res. 2018;28:1839–1857.
48. Mukumbang FC, van Wyk B, Van Belle S, et al. “At this [adherence] club, we are a family now”: a realist theory-testing case study of the antiretroviral treatment adherence club, South Africa. South Afr J HIV Med. 2019;20:1–14.
49. Mukumbang FC, Van Wyk B, Van Belle S, et al. Unravelling how and why the Antiretroviral Adherence Club Intervention works (or not) in a public health facility: a realist explanatory theory-building case study. PLoS One. 2019;14:e0210565.
50. Pasipamire L, Nesbitt RC, Ndlovu S, et al. Retention on ART and predictors of disengagement from care in several alternative community-centred ART refill models in rural Swaziland. J Int AIDS Soc. 2018;21:e25183.
51. Pellecchia U, Baert S, Nundwe S, et al. “We are part of a family”. Benefits and limitations of community ART groups (CAGs) in Thyolo, Malawi: a qualitative study. J Int AIDS Soc. 2017;20:21374.
52. Rasschaert F, Telfer B, Lessitala F, et al. A qualitative assessment of a community antiretroviral therapy group model in Tete, Mozambique. PLoS One. 2014;9:e91544.
53. Rasschaert F, Decroo T, Remartinez D, et al. Sustainability of a community-based anti-retroviral care delivery model—a qualitative research study in Tete, Mozambique. J Int AIDS Soc. 2014;17:18910.
54. Sharp J, Wilkinson L, Cox V, et al. Outcomes of patients enrolled in an antiretroviral adherence club with recent viral suppression after experiencing elevated viral loads. South Afr J HIV Med. 2019;20:1–7.
55. Vandendyck M, Motsamai M, Mubanga M, et al. HIV/AIDS Research and Treatment Community-Based ART Resulted in Excellent Retention and Can Leverage Community Empowerment in Rural Lesotho. A Mixed Method Study Citation. Hyderabad Telangana State, India: Openventio Publishers; 2015.
56. Venables E, Towriss C, Rini Z, et al. Patient experiences of ART adherence clubs in Khayelitsha and Gugulethu, Cape Town, South Africa: a qualitative study. PLoS One. 2019;14:1–18.
57. Vogt F, Kalenga L, Lukela J, et al. Brief report: decentralizing ART supply for stable HIV patients to community-based distribution centers: program outcomes from an urban context in Kinshasa, DRC. J Acquir Immune Defic Syndr JAIDS. 2017;74:326–331.
58. Wringe A, Cawley C, Szumilin E, et al. Retention in care among clinically stable antiretroviral therapy patients following a six-monthly clinical consultation schedule: findings from a cohort study in rural Malawi. J Int AIDS Soc. 2018;21:1–9.
59. Dudhia R, Kagee A. Experiences of participating in an antiretroviral treatment adherence club. Psychol Heal Med. 2015;20:488–494.
60. Bateman C. MSF again paves the way with ART. South Afr Med J Suid-Afrikaanse Tydskr Vir Geneeskd. 2013;103:71–73.
61. Scheirer MA, Dearing JW. An agenda for research on the sustainability of public health programs. Am J Public Health. 2011;101:2059–2067.
62. Dharmayat KI, Tran T, Hardy V, et al. Sustainability of “mhealth” interventions in sub-Saharan Africa: a stakeholder analysis of an electronic community case management project in Malawi. Malawi Med J. 2019;31:177–183.
63. Piot P, Abdool Karim SS, Hecht R, et al. Defeating AIDS—advancing global health. Lancet. 2015;386:171–218.
64. Brundage SC, Bellamy WM, Lamptey P, et al. Terra Nova: How to Achieve a Successful PEPFAR Transition in South Africa. Washington, DC: Center for Strategic and international studies; 2011.
65. Mogeni T, Tegang S-P, Kumar Pothapregada S, et al. 2013 Sustainability Review of Global Fund Supported HIV, Tuberculosis and Malaria Programmes Final Report Commissioned by the Technical Evaluation Reference Group (TERG) of the the Following People from the Global Fund Contributed to This Report. Geneva, Switzerland: The Global Fund; 2013.
66. Oberth G, Whiteside A. What does sustainability mean in the HIV and AIDS response? Afr J AIDS Res. 2016;15:35–43.
67. Oliveira SRDA, Medina MG, Figueiró AC, et al. Fatores estratégicos para a sustentabilidade de uma intervenção na saúde em nível municipal no Brasil. Cad Saude Publica. 2017;33. doi:10.1590/0102-311X00063516.
68. Herrera-Sánchez IM, León-Pérez JM, León-Rubio JM. Steps to ensure a successful implementation of occupational health and safety interventions at an organizational level. Front Psychol. 2017;8:2135.
69. Wilkinson L, Harley B, Sharp J, et al. Expansion of the Adherence Club model for stable antiretroviral therapy patients in the Cape Metro, South Africa 2011-2015. Trop Med Int Heal. 2016;21:743–749.
70. Ferguson L, Nicholson A, Henry I, et al. Assessing changes in HIV-related legal and policy environments: lessons learned from a multi-country evaluation. PLoS One. 2018;13:e0192765.
71. Zahn R, Grosso A, Scheibe A, et al. Human rights violations among men who have sex with men in southern africa: comparisons between legal contexts. PLoS One. 2016;11:e0147156.
72. NERCHA. The Kingdom of Swaziland SWAZILAND GLOBAL AIDS RESPONSE PROGRESS REPORTING 2014. Mbabane, Swaziland: National Emergency Response Council on HIV/AIDS (NERCHA); 2014.
73. Briana D. Kenya Homosexuality: High Court Considers Case to Make it Legal—CNN. CNN. 2018. Available at: Accessed April 19, 2019.
74. Virlogeux V, Zoulim F, Pugliese P, et al. Modeling HIV-HCV coinfection epidemiology in the direct-acting antiviral era: the road to elimination. BMC Med. 2017;15:217.
75. Agnarson AM, Masanja H, Ekstrom AM, et al. Challenges to ART scale-up in a rural district in Tanzania: stigma and distrust among Tanzanian health care workers, people living with HIV and community members. Trop Med Int Heal. 2010;15:1000–1007.
76. Rueda S, Mitra S, Chen S, et al. Examining the associations between HIV-related stigma and health outcomes in people living with HIV/AIDS: a series of meta-analyses. BMJ Open. 2016;6:e011453.
77. UNAIDS. UNAIDS Warns That HIV-Related Stigma and Discrimination is Preventing People From Accessing HIV Services—UNAIDS. 2017. UNAIDS. Available at: Accessed April 20, 2019.
78. Pantelic M, Steinert JI, Park J, et al. “Management of a spoiled identity”: systematic review of interventions to address self-stigma among people living with and affected by HIV. BMJ Glob Heal. 2019;4:e001285.
79. Oramasionwu CU, Daniels KR, Labreche MJ, et al. The environmental and social influences of HIV/AIDS in Sub-Saharan Africa: a focus on rural communities. Int J Environ Res Public Health. 2011;8:2967–2979.
80. Brodish PH. An association between neighbourhood wealth inequality and HIV prevalence in sub-Saharan Africa. J Biosoc Sci. 2015;47:311–328.
81. Kata F. Gender-Based Violence Increases Risk of HIV/AIDS for Women in Sub-saharan Africa—Population Reference Bureau. Population Reference Bureau. 2011. Available at: Accessed April 20, 2019.
82. Mottiar S, Lodge T. The role of community health workers in supporting South Africa's HIV/AIDS treatment programme. Afr J AIDS Res. 2018;17:54–61.
83. Long A, Mbabali I, Hutton HE, et al. Design and implementation of a community health worker HIV treatment and prevention intervention in an HIV hot spot fishing community in rakai, Uganda. J Int Assoc Provid AIDS Care. 2017;16:499–505.
84. De Neve JW, Garrison-Desany H, Andrews KG, et al. Harmonization of community health worker programs for HIV: a four-country qualitative study in Southern Africa. Tsai AC. PLOS Med. 2017;14:e1002374.
85. Mate KS, Bennett B, Mphatswe W, et al. Challenges for routine health system data management in a large public programme to prevent mother-to-child HIV transmission in South Africa. Castro A. PLoS One. 2009;4:e5483.
86. Ceptureanu SI, Ceptureanu EG, Luchian CE, et al. Community based programs sustainability. a multidimensional analysis of sustainability factors. Sustain. 2018;10:1–15.
87. WHO. Human Rights and Health. WHO. Available at: Published 2017. Accessed April 22, 2019.
88. Herlitz L, MacIntyre H, Osborn T, et al. The sustainability of public health interventions in schools: a systematic review. Implement Sci. 2020;15:217.
89. Morton KL, Atkin AJ, Corder K, et al. Engaging stakeholders and target groups in prioritising a public health intervention: the Creating Active School Environments (CASE) online Delphi study. BMJ Open. 2017;7:13340.
90. Griffiths J, Maggs H, George E. Stakeholder involvement: background paper prepared for the WHO/WEF joint event on preventing noncommunicable diseases in the workplace (dalian/China, september 2007). Who. 2008;44:0–40.
91. UNAIDS (Joint United Nations Programme on HIV/AIDS). Panel 1 :Shared Responsibility—a New Global Compact for HIV: Key Issues. New York, NY; 2011.
92. 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;22:173–174.
93. Virani T, Lemieux-Charles L, Davis D, et al. Sustaining change: once evidence-based practices are transferred, what then? Healthc Q. 2009;12:89–96.
94. Curry SJ, Mermelstein RJ, Sporer AK. Sustainability of community-based youth smoking cessation programs: results from a 3-year follow-up. Health Promot Pract. 2016;17:845–852.
95. Milne S, Greenaway S, Conway K, et al. What next? Sustaining a successful small-scale alcohol consumption harm minimization project. Subst Use Misuse. 2007;42:1933–1944.
96. EVOLUTION of the VISIBILITY and ANALYTICS NETWORK IN SOUTH AFRICA ROB BOTHA-PwC Public Sector Chief of Party Global Health Supply Chain-Technical Assistance.
97. World Health Organisation. Global HIV Drug Resistance Laboratory Network; 2014. Available at:
98. FHI. Engaging public health champions to garner support for innovations. Natl Collab Cent Methods Tools. 2011. doi:10.1002/14651858.CD000125.pub3.
99. About—Champions for an AIDS-free Generation in Africa. Available at: Accessed April 29, 2019.
100. PEPFAR and BONEPWA+ Launch the Treat All Champions Campaign | U.S. Embassy in Botswana. Available at: Accessed April 29, 2019.
101. Atun R, Silva S, Knaul FM. Innovative financing instruments for global health 2002-15: a systematic analysis. Lancet Glob Heal. 2017;5:e720–e726.
102. World Health Organisation (WHO). HIV, Universal Health Coverage and the Post-2015 Development Agenda; 2015.
103. Savaya R, Elsworth G, Rogers P. Projected sustainability of innovative social programs. Eval Rev. 2009;33:189–205.
104. Villeneuve C, Tremblay D, Riffon O, et al. A systemic tool and process for sustainability assessment. Sustain. 2017;9:1–29.
105. Pettigrew AM, Woodman RW, Cameron KS. Studying organizational change and development: challenges for future research. Acad Manag J. 2001;44:697–713.
106. Reed JE, Kaplan HC, Ismail SA. A new typology for understanding context: qualitative exploration of the model for understanding success in quality (MUSIQ). BMC Health Serv Res. 2018;18:584.
107. Kaplan HC, Provost LP, Froehle CM, et al. The model for understanding success in quality (MUSIQ): building a theory of context in healthcare quality improvement. BMJ Qual Saf. 2012;21:13–20.

sustainability; differentiated service delivery; HIV/AIDS; sub-Saharan Africa; patient-centered care

Supplemental Digital Content

Copyright © 2021 The Author(s). Published by Wolters Kluwer Health, Inc.