AIDS CARE, 2018 https://doi.org/10.1080/09540121.2018.1441970

Review of differentiated approaches to antiretroviral therapy distribution Nicole Davisa, Natasha Kanagata, Melissa Sharera,b, Sabrina Eagana, Jennifer Pearsona and Ugochukwu “Ugo” Amanyeiwec

aJSI Research & Training Institute, Inc, Boston, MA, Washington, DC; bDepartment of Public Health, St. Ambrose University, Iowa, USA; cUnited States Agency for International Development

ABSTRACT In response to global trends of maximizing the number of patients receiving antiretroviral therapy (ART), this review summarizes literature describing differentiated models of ART distribution at facility and community levels in order to highlight promising strategies and identify evidence gaps. Databases and gray literature were searched, yielding thirteen final articles on differentiated ART distribution models supporting stable adult patients. Of these, seven articles focused on distribution at facility level and six at community level. Findings suggest that differentiated models of ART distribution contribute to higher retention, lower attrition, and less loss to follow-up (LTFU). These models also reduced patient wait time, travel costs, and time lost from work for drug pick-up. Facility- and community-level ART distribution models have the potential to extend treatment availability, enable improved access and adherence among people living with HIV (PLHIV), and facilitate retention in treatment and care. Gaps remain in understanding the desirability of these models for PLHIV, and the need for more information the negative and positive impacts of stigma, and identifying models to reach traditionally marginalized groups such as key populations and youth. Replicating differentiated care so efforts can reach more PLHIV will be critical to scaling these approaches across varying contexts.

ARTICLE HISTORY Received 3 July 2017 Accepted 6 February 2018

KEYWORDS Antiretroviral therapy; decentralization; differentiation; community ART programs; differentiated care; ART distribution; HIV treatment; task shifting

Introduction

In order to maximize the number of patients receiving ART, low- and middle-income countries (LMICs) are beginning to implement treatment delivery models that are “decentralized” (where services are shifted from higher- to lower-level facilities) or “differentiated” (where services are adapted to the needs of individuals on ART) (WHO, 2013). There is no one-size-fits-all approach to delivering ART, and this review categorizes models into two broad categories: facility- and commu-nity-level. This review deepens the understanding of the current evidence, identifies gaps, and highlights promising strategies for potential replicability and scale-up by presenting successes and challenges associ-ated with adopting a client-centered approach at each level of service delivery across varying contexts.

Methods

PubMed, Google Scholar and gray literature, including project documents and conference abstracts, published in English from 2000–2015 were searched using a combi-nation of key terms (Table 1). Results were further refined to include only gray and peer reviewed literature published

since 2006 when new funding increased HIV treatment availability. The study population of focus was limited to stable adults (18 years and older) on ART, who are the main target of most differentiated models currently being implemented. Articles focused on pregnant or post-partum women, adolescents, patients with comorbidities (such as tuberculosis), or key populations were excluded as they have specific needs beyond the scope of this review. Articles were categorized by ART distribution approach: distribution at either facility or community level.

Searches identified 416 published articles and 27 gray literature documents by three reviewers. Abstracts were reviewed independently via two lead authors and then ranked by a subject matter expert. Each reviewer used inclusion and exclusion criteria to identify articles detail-ing models at facility or community level, resulting in 138 articles receiving a full review by the two lead authors (see Figure 1).

Results

Thirteen articles were included in this review: seven focused on approaches at the facility level and six focused on community-level approaches (Figure 2). The study

CONTACT Sabrina Eagan sabrina_eagan@jsi.com; Melissa Sharer melissa_sharer@jsi.com 44 Farnsworth Street, Boston, MA 02210, USA © 2018 Informa UK Limited, trading as Taylor & Francis Group

2 N. DAVIS ET AL.

Table 1. Search terms. Search term base Addition terms for search combinations

Antiretroviral AND [Adherence OR Attrition OR Clubs OR therapy Decentralization OR Delivery OR Differentiated

OR Dispensing OR Enablers OR Innovative OR Scale up OR Task Shifting OR Technology]

HIV treatment AND [Adherence OR Attrition OR Clubs OR Decentralization OR Delivery OR Differentiated OR Dispensing OR Enablers OR Innovative OR Scale up OR Task Shifting OR Technology]

ARV AND [Adherence OR Attrition OR Clubs OR Decentralization OR Delivery OR Differentiated OR Dispensing OR Enablers OR Innovative OR Scale up OR Task Shifting OR Technology]

design, approach type, and effects reported in each article are presented in Table 2.

Facility-level models

At the facility level, differentiated approaches to ART distribution include shifting ART responsibility to lower-level facilities and cadres, strengthening the

pharmacuetical system and tailoring pharmacy services using technology.

Benefits A study in Malawi demonstrated that involving non-physician clinicians in ART initiation could double ART enrollment (Bemelmans et al., 2010). Furthermore, a computerized dispensing system shifts ART dispensing to primary health facilities and community distribution points for stable patients in South Africa, with patients reporting significantly reduced waiting times (Mabirizi, Embrey, Saleeb, & Aboagye-Nyame, 2015). Similarly, South Africa’s Chronic Dispensing Unit (CDU) was introduced as an outsourced, centralized service dispen-sing prepackaged medications to stable clients via facili-ties and community sites including mobile clinics, clubs, and private pharmacies. Findings show ART patients saved money and time by accessing their medication at local facilities (Magadzire, Marchal, & Ward, 2015).

Figure 1. PRISMA diagram of literature review process.

3 AIDS CARE

Figure 2. Distribution models at facility and community levels.

In Ethiopia and Kenya, strengthening pharmaceutical systems led to improved efficiency of ART distribution. Models in South Africa and Malawi showed stable clients enrolled in “fast track” pick-up lanes saved time by col-lecting prepackaged ART (Mabirizi et al., 2015). Auto-mated pharmacy dispensing units in South Africa allowed clients to avoid queues for ART refills. Although a formal evaluation has not been published, reports from one clinic indicated improvements in patient wait times, increased time for counseling, and adherence (Turquoise PR Communications, 2014).

In Namibia, pharmacists using the mobile Electronic Dispensing Tool (EDT), which enabled health workers to monitor the adherence of each patient through pill count and time of prescription refill, have reported reduced bur-den of dispensing ART and increased pharmacy efficiency and availability to provide care (Sumbi et al., 2013).

Challenges Reported challenges with facility-level differentiation of ART distribution included increased patient loads,

reducing patient attention and decreased quality of care, potentially leading to LTFU and treatment failure (Fatti, Grimwood, & Bock, 2010).

Community-level models

Differentiated models of community-level ART distri-bution include ART adherence support groups, the use of “expert patients” for peer support involving com-munity health workers (CHWs) in treatment, and community drug distribution points (Bemelmans et al., 2014).

Benefits Community-based ART distribution may significantly reduce the frequency at which clinically stable clients are required to attend a health center, which decreases clinicians’ workload, resulting in time savings for the cli-ent and for facility staff and allowing health center resources to be applied to improved management of complex clinical cases (Bemelmans et al., 2014).

4 N. DAVIS ET AL.

Table 2. Description of reviewed documents and associated reported outcomes. Study and site Type of differentiated approach Study design Limitations to study design Effects reported

Facility-Level Interventions (7) Bedelu et al., 2007 South Africa Task Shifting Decentralization of Case Study / Program . Limited generalizability . Improved geographic

care description access for patients . Reduced loss to follow

up . Reduced workload for

health facility staff . Improved retention

Bemelmans et al., 2010 Malawi Task Shifting Decentralization of Case Study/ Program . Limited generalizability . Improved geographic care description access for patients

. Improved enrollment into ART treatment

. Better efficiency of ART services

. Improved retention

Fatti et al., 2010 South Africa Decentralization of Care Retrospective cohort . Potential for poor data . Improved geographic study quality or limited data access for patients

availability . Lowered time and/or costs for patients attending clinics

. Improved retention

. Improved virological suppression

Mabirizi et al., 2015 Ethiopia & Pharmaceutical System Program Report . Descriptive program � Increased availability of Kenya Strengthening reports ART � Better efficiency of

ART services

Magadzire et al., 2015 South Automated Pharmacy Dispensing Case Study . Limited availability of . Improved geographic Africa Community ART distribution data access for patients

points . Limited generalizability . Reduced client waiting times

. Improved patient autonomy, agency and resilience

Sumbi et al., 2013 Namibia Task Shifting Pharmaceutical Program Report . Limited description of . Reduced client waiting System Strengthening analysis times

. Limited generalizability . Reduced workload for health facility staff

Turquoise PR & Automated Pharmacy Dispensing Program Report . Limited description of . Reduced client waiting Communications, 2014. South analysis times Africa . Limited generalizability . Better efficiency of ART

services . More time for staff

Community-Level Interventions (6) interaction with patients

Bemelmans et al., 2014 DRC, Adherence clubs Community ART Case Study Series . Program reporting . Lowered time and/or Malawi, Mozambique, South distribution points . Limited generalizability costs for patients Africa Appointment Spacing attending clinics

. Improved social support for patients

. Improved patient autonomy, agency and resilience

. Reduced workload for health facility staff

. Improved retention

Decroo et al., 2011 Adherence clubs Case Study . Program reporting . Lowered time and/or Mozambique . Limited generalizability costs for patients

attending clinics . Improved social support

for patients . Reduced workload for

health facility staff . Improved retention

Dudhia & Kagee, 2015 South Adherence clubs Case Study . Qualitative data . More time for staff Africa . Small sample size interaction with patients

. Limited generalizability

(Continued )

5 AIDS CARE

Table 2. Continued. Study and site Type of differentiated approach Study design Limitations to study design Effects reported

.

.

Lowered time and/or costs for patients attending clinics Improved social support for patients

Grimsrud et al., 2015 South Africa

Adherence clubs within 500 ft of facility

Case Study/ Program description

.

.

No comparison group Limited generalizability

.

.

Improved geographic access for patients Reduced workload for

.

.

health facility staff Lowered time and/or costs for patients attending clinics Improved virological suppression

“MSF,” 2012. DRC, Leotho, Adherence clubs Community ART distribution points Appointment Spacing

Case Study (cross case study?)/ program description

. Limited generalizability .

.

Lowered time and/or costs for patients attending clinics Reduced workload for

.

.

health facility staff Improved social support for patients Improved retention

Rasschaert et al., 2014 Mozambique

Adherence clubs Program evaluation . Limited generalizability .

.

.

Reduced workload for health facility staff Improved social support for patients Improved patient autonomy, agency and resilience

ART distribution at community levels was found to reduce patient costs related to transport, time and work lost due to frequent clinic visits (Bedelu, Ford, Hil-derbrand, & Reuter, 2007; Bemelmans et al., 2010; MSF, 2012). One study indicated that a patient’s departure from the “sick” role to that of a participant or leader of a community group may improve both self-efficacy and self-management skills. This shift in perception may contribute to their sense of empowerment, and increase their agency and ability to self-manage their care (Dudhia & Kagee, 2015).

Models like adherence clubs or community-distri-bution-points include a component of peer support/edu-cation that can increase access and reduce time/travel needed to collect ART (MSF, 2012).

Studies in South Africa and Mozambique found that stable patients receiving ART via community groups had higher rates of viral suppression and higher CD4+ counts compared to those who collected ART at health facilities (Bemelmans et al., 2014). Decroo et al. (2011) found patients enrolled in community-based delivery methods or using community distribution points have higher rates of retention, lower rates of attrition, and less LTFU.

Additionally, community-based models exhibited flexibility – including across different regions/countries

or populations (Grimsrud, Sharp, Kalombo, Bekker, & Myer, 2015). These models’ adaptability makes them desirable for implementation and scale-up given their potential to adapt to ever-changing needs and contexts (Rasschaert et al., 2014).

Challenges Reported challenges with community based ART distri-bution included: low enrolment in community adher-ence clubs (MSF, 2012); low uptake of community services due to stigma (Bemelmans et al., 2014) and poor coordination and management (Rasschaert et al., 2014).

Discussion

Evidence shows decentralization of ART distribution mechanisms to the facility level can lead to better out-comes and increased accessibility, which is reinforced by similar findings in South Africa, Kenya, and Malawi (Chan et al., 2010; Fatti et al., 2010; Massaquoi et al., 2009; Reidy et al., 2014). Similarly, community-level dis-tribution mechanisms reduced patient waiting times (Mabirizi et al., 2015) as well as transport costs and time lost from work (Bedelu et al., 2007; Bemelmans et al., 2010; IAS, 2016; Decroo et al., 2013; MSF, 2012).

6 N. DAVIS ET AL.

Furthermore, models such as adherence clubs offered the additional benefit of peer support (MSF, 2012; Reda & Biadgilign, 2012). Results revealed the importance of CHWs’ role in facilitating access to ART for rural and underserved populations (Arem et al., 2011; Gusdal et al., 2011; Mukherjee & Eustache, 2007; Torpey et al., 2008).

Our findings have shown the benefits and challenges associated with various ART delivery models and high-lighted the importance of understanding which individ-uals are most suited to community- or facility-based decentralized ART delivery mechanisms. The review reinforces that community ART distribution can comp-lement health facility/provider-driven service delivery and enhance patients’ ability to manage HIV. Findings further reinforce that efforts driven, owned by, and embedded in the communities themselves are sustain-able (Decroo et al., 2009).

There is a need for more in-depth information on posi-tive and negative effects related to stigma and ART distri-bution. Future efforts should focus on determining if the models explored in this review are translatable to individ-uals who are not clinically stable and struggle with reten-tion and adherence. Finally, more attention is needed on effective coordination and structured supervision as ART distribution moves from facility to community level. It remains critical that PLHIV must still be linked to the health care system and able to access more advanced care when needed in order to ensure that these approaches continue to represent an effective, con-venient, and supportive alternative for ART distribution.

Limitations

Widespread implementation of ART distribution models is relatively new, and the breadth of results likely will continue to emerge. Further, the risk of bias is a chal-lenge to assess as most of the studies did not employ ran-domization, were observational, and have small sample sizes. Finally, due to the heterogeneity of outcome measures, it was not feasible to conduct meta-analysis.

Conclusions

The evidence shows differentiated models of care have the potential to extend the reach of treatment, and reduce barriers to accessing ART, and this review broad-ens the knowledge base and understanding of the models that do exist. Additionally, emerging evidence describes the potential of health and pharmaceutical systems strengthening and community-based mechanisms to dis-tribute ART which may lead to more innovation. Much of the literature available focused on reporting results of

the specific interventions carried out; however, there were few details on the specific enablers and barriers that are key to replication and scale-up. Additionally, all models focused on stable populations and did not provide in-depth information any positive or negative impacts on self-perceived stigma. It is therefore impor-tant to consider the expansion and variety of differen-tiated models to serve the different contextual needs of those living with HIV (Brinkhof et al., 2008) as no one model will suit all.

Authors’ contributions

Natasha Kanagat, Nicole Davis, Melissa Sharer, and Sab-rina Eagan conceived the study design; Nicole Davis and Natasha Kanagat carried out the review; Sabrina Eagan and Melissa Sharer double-screened most included papers; Nicole Davis, Natasha Kanagat, and Melissa Sharer wrote the first draft; and Sabrina Eagan commen-ted on all subsequent drafts. Jennifer Pearson critically read and revised the manuscript. Ugochukwu “Ugo” Amanyeiwe critically read and provided comments on the manuscript. All authors reviewed and approved the final version.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was made possible by the generous support of the American people through the U.S. President’s Emergency Plan for AIDS Relief (PEPFAR) with the United States Agency for International Development (USAID) under the terms of Cooperative Agreement Strengthening High Impact Interven-tions for an AIDS-free Generation (AIDSFree), number AID-OAA-A-14-00046. The contents are the responsibility of AIDSFree and do not necessarily reflect the views of USAID, PEPFAR, or the U.S. Government.

References

Arem, H., Nakyanjo, N., Kagaayi, J., Mulamba, J., Nakigozi, G., Serwadda, D., … Chang, L. W. (2011). Peer health workers and AIDS care in Rakai, Uganda: A mixed methods oper-ations research evaluation of a cluster-randomized trial. AIDS Patient Care and STDs, 25(12), 719–724. doi:10. 1089/apc.2010.0349

Bedelu, M., Ford, N., Hilderbrand, K., & Reuter, H. (2007). Implementing antiretroviral therapy in rural communities: the Lusikisiki model of decentralized HIV/AIDS care. The Journal of Infectious Diseases, 196(Suppl. 3), S464–S468.

Bemelmans, M., Baert, S., Goemaere, E., Wilkinson, L., Vandendyck, M., van Cutsem, G., … Ford, N. (2014). Community-supported models of care for people on HIV

7 AIDS CARE

treatment in sub-Saharan Africa. Tropical Medicine & International Health, 19(8), 968–977. doi:10.1111/tmi. 12332

Bemelmans, M., Van Den Akker, T., Ford, N., Philips, M., Zachariah, R., Harries, A., … Massaquoi, M. (2010). Providing universal access to antiretroviral therapy in Thyolo, Malawi through task shifting and decentralization of HIV/AIDS care. Tropical Medicine & International Health, 15(12), 1413–1420.

Brinkhof, M. W., Dabis, F., Myer, L., Bangsberg, D. R., Boulle, A., Nash, D., & Anglaret, X. (2008). Eearly loss of HIV-infected patients on potent antiretroviral therapy pro-grammes in lower-income countries. Bulletin of the World Health Organization, 86(7), 559–567. doi:10.2471/BLT.07. 044248

Chan, A. K., Mateyu, G., Jahn, A., Schouten, E., Arora, P., Mlotha, W., … van Lettow, M. (2010). Outcome assessment of decentralization of antiretroviral therapy provision in a rural district of Malawi using an integrated primary care model. Tropical Medicine & International Health, 15(s1), 90–97.

Decroo, T., Panunzi, I., das Dores, C., Maldonado, F., Biot, M., Ford, N., & Chu, K. (2009). Lessons learned during down referral of antiretroviral treatment in Tete, Mozambique. Journal of the International AIDS Society, 12, 6. doi:10. 1186/1758-2652-12-6

Decroo, T., Rasschaert, F., Telfer, B., Remartinez, D., Laga, M., & Ford, N. (2013). Community-based antiretroviral therapy programs can overcome barriers to retention of patients and decongest health services in sub-saharan Africa: A systema-tic review. International Health, 5(3), 169–179. doi:10.1093/ inthealth/iht016

Decroo, T., Telfer, B., Biot, M., Maïkéré, J., Dezembro, S., Cumba, L. I., … Ford, N. (2011). Distribution of antiretro-viral treatment through self-forming groups of patients in Tete province, Mozambique. JAIDS Journal of Acquired Immune Deficiency Syndromes, 56(2), e39–e44.

Dudhia, R., & Kagee, A. (2015). Experiences of participating in an antiretroviral treatment adherence club. Psychology, Health & Medicine, 20(4), 488–494. doi:10.1080/13548506. 2014.953962

Fatti, G., Grimwood, A., & Bock, P. (2010). Better antiretro-viral therapy outcomes at primary healthcare facilities: An evaluation of three tiers of ART services in four South African provinces. PLoS One, 5(9), e12888.

Grimsrud, A., Sharp, J., Kalombo, C., Bekker, L.-G., & Myer, L. (2015). Implementation of community-based adherence clubs for stable antiretroviral therapy patients in Cape Town, South Africa. Journal of the International AIDS Society, 18, 19984.

Gusdal, A. K., Obua, C., Andualem, T., Wahlstrom, R., Chalker, J., & Fochsen, G., on behalf of the INRUD-IAA project. (2011). Peer counselors’ role in supporting patients’ adherence to ART in Ethiopia and Uganda. AIDS Care, 23 (6), 657–662. doi:10.1080/09540121.2010.53253

International AIDS Society (IAS). (2016). Differentiated care for HIV: A decision framework for antiretroviral therapy delivery. Durban, South Africa. Retrieved from http:// www.differentiatedcare.org/Portals/0/adam/Content/ yS6M-GKB5EWs_uTBHk1C1Q/File/Decision% 20Framework%20REPRINT%20web.pdf

Mabirizi, D., Embrey, M., Saleeb, S., & Aboagye-Nyame, F. (2014). Pharmaceutical system strengthening interventions to improve access to antiretroviral therapy, Arlington, VA: Management Sciences for Health. Retrived from: http:// apps.who.int/medicinedocs/en/d/Js21842en/.

Magadzire, B. P., Marchal, B., & Ward, K. (2015). Improving access to medicines through centralised dis-pensing in the public sector: A case study of the chronic dispensing unit in the Western Cape province, South Africa. BMC Health Services Research, 15, 8. doi:10. 1186/s12913-015-1164-x

Massaquoi, M., Zachariah, R., Manzi, M., Pasulani, O., Misindi, D., Mwagomba, B., … Harries, A. D. (2009). Patient retention and attrition on antiretroviral treatment at district level in rural Malawi. Transactions of the Royal Society of Tropical Medicine and Hygiene, 103(6), 594– 600. doi:10.1016/j.trstmh.2009.02.012

Médecins Sans Frontières (MSF) International. (2012). Closer to home: Delivering antiretroviral therapy in the commu-nity. Retrieved from http://www.msfaccess.org/content/ closer-home-delivering-antiretroviral-therapy-community

Mukherjee, J. S., & Eustache, F. E. (2007). Community health workers as a cornerstone for integrating HIV and primary healthcare. AIDS Care, 19(Suppl. 1), 73–82. doi:10.1080/ 09540120601114485

Rasschaert, F., Decroo, T., Remartinez, D., Telfer, B., Lessitala, F., Biot, M., … Van Damme, W. (2014). Adapting a com-munity-based ART delivery model to the patients’ needs: A mixed methods research in Tete, Mozambique. BMC Public Health, 14, 1. doi:10.1186/1471-2458-14-364

Reda, A. A., & Biadgilign, S. (2012). Determinants of adher-ence to antiretroviral therapy among HIV-infected patients in Africa. AIDS Research and Treatment, 1, e000125.

Reidy, W. J., Sheriff, M., Wang, C., Hawken, M., Koech, E., Elul, B., … Abrams, E. J. (2014). Decentralization of HIV care and treatment services in Central province, Kenya. Journal of Acquired Immune Deficiency Syndromes (1999), 67(1), e34–e40. doi:10.1097/QAI.0000000000000264

Sumbi, V., Indongo, L., Mabirizi, D., Lates, J., Mwinga, S., Kangudie, M., & Blom, A. (2013). Implementing a nationwide electronic pharmacy dispensing system in low-resource settings: The electronic dispensing tool in Namibia’s ART Program Republic of Namibia Ministry of Health and Social Services. Retrieved from www. siapsprogram.org

Torpey, K. E., Kabaso, M. E., Mutale, L. N., Kamanga, M. K., Mwango, A. J., Simpungwe, J., … Mukadi, Y. D. (2008). Adherence support workers: A way to address human resource constraints in antiretroviral treatment programs in the public health setting in Zambia. PLoS ONE, 3(5), e2204. doi:10.1371/journal.pone.0002204

Turquoise PR Communications. (2014). Automated pharmacy at SA’s busiest HIV clinic reduces wait times from 4 hours to 18 minutes. Retrieved from http://www.health24.com/ Medical/HIV-AIDS/News/Automated-pharmacy-at-SAs-busiest-HIV-clinic-reduces-wait-times-from-4-hours-to-18-minutes-20131129

World Health Organization (WHO). (2013). Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: Recommendations for a public health approach. Geneva: Author.