Examining the integration of Food by Prescription into HIV care and treatment in Zambia

By Kate A. Greenaway, Elizabeth C. Jere, Milika E. Zimba, Cassim Masi and Beatrice Mazinza Kawana

Kate Greenaway is Senior Technical Advisor, HIV Unit, Catholic Relief Services, Baltimore, MD

Elizabeth Jere is Senior Technical Advisor, STEPS OVC, Catholic Relief Services, Lusaka, Zambia

Milika Zimba is Programme Manager, Children's AIDS Fund, Lusaka, Zambia

Cassim Masi is Executive Director, National Food and Nutrition Commission, Lusaka, Zambia

Beatrice Mazinza Kawana is Deputy Executive Director, National Food and Nutrition Commission, Lusaka, Zambia

This research was funded by a USAID Cooperative agreement (award number 690-A-00-06-00093-00). The authors would like to acknowledge the support of the Ministry of Health (Zambia), the National Food and Nutrition Commission (Zambia) and the participating HIV care sites. We are also grateful to the interviewees, focus group participants and research assistants for their participation in this study. Finally, we acknowledge the Food and Nutrition Technical Assistance (FANTA-2) team for their guidance and advice throughout the design and implementation process.

There is increasing evidence that antiretroviral therapy (ART) outcomes and nutrition interventions are closely linked. Studies from sub- Saharan Africa have established that low Body Mass Index (BMI) at ART initiation is a significant predictor of early mortality and that malnutrition plays a substantial role in disease progression^1,2,3. In late-stage HIV infection, unintended weight loss is common: up to 25 percent of clients experience dramatic, life-threatening weight loss.

Clients collect their prescribed HEPS at Wusikile Mine Hospital

FBP is a treatment approach that targets moderately and severely malnourished individuals with ‘medicalised’ doses of specific nutrition supplements. While empirical evidence about causal relationships between nutrition support, weight gain and improved treatment outcomes among ART clients is lacking, there is evidence that weight gain at three months on ART is strongly associated with survival⁴ and that nutrition supplements have a positive effect on ART adherence⁵.

Zambia has a generalised HIV epidemic, where more than 900,000 Zambians are living with HIV (PLHIV), with 280,000 on ART⁶. Research conducted in 2007 revealed startlingly high rates of malnutrition among adult PLHIV starting ART: 33.5% had a BMI < 18.5 kg/m2, and 9% had a BMI less <16.0 kg/m2 (⁷).

Catholic Relief Services FBP Pilot in Zambia

To address malnutrition in people living with HIV (PLHIV), Catholic Relief Services (CRS) Zambia worked in partnership with Zambia’s National Food and Nutrition Commission (NFNC) to pilot a Food by Prescription (FBP) programme as an adjunct to a USAID-funded palliative care grant. The evaluation was undertaken to understand the practical implications of FBP implementation and to gather information on client outcomes.

The Zambia FBP model prescribes and dispenses specialised nutrition commodities in response to clinical malnutrition (Figure 1). Small daily ‘doses’, packaged in individual sachets, are intended to reduce intra-household sharing, institutionalise the concept that these foods are ‘medicines’, ease calculation of recommended daily allowance (RDA) and aid monitoring at the household level.

The model requires that nutrition assessment, education, counselling and support (including food dispensing) be synchronised with HIV care and treatment services. The pilot tested the model in three types of settings: clinical facilities (eight), hospices (ten) and home based care (two). Procedures varied by setting to accommodate pre-existing systems and emphasising integration rather than establishment of parallel systems. Staff training, using the (draft) national FBP guidelines, was provided in the 22 sites.

Clients were admitted to the FBP component of HIV care according to anthropometric criteria. BMI was most often used. Mid-upper arm circumference (MUAC) was used to assess pregnant and lactating women, as well as clients whose heights could not be taken. Children were assessed using weight-for-height z scores (WHZ). As dictated by the national protocol, adults with severe acute malnutrition (SAM) were prescribed both Ready to Use Therapeutic Food (RUTF) and High Energy Protein Supplement (HEPS) in sufficient quantity to meet 100% RDA. Those with moderate acute malnutrition (MAM) received HEPS to meet 50% RDA. Clients were re-evaluated regularly and discharged when anthropometric assessment indicated nutrition rehabilitation. At the time of the evaluation, the pilot had reached 5,360 clients.

Methodology of pilot evaluation

To evaluate the pilot programme, purposive sampling was used to select 11 evaluation sites: six ART clinics, four hospices and one HBC site representing locality (urban or rural), supporting organisation and size of programme.

From the 1,671 clients enrolled at the 11 selected sites, the evaluation team planned for a purposive sample of ten discharged clients per site (total of 110). Difficulties in communication and logistics resulted in identification of 91 clients (84 adults and 7 children). The guardian most conversant with the child’s health/illness situation was interviewed.

Quantitative data was collected and analysed from client records, monthly reports and field visit reports. Variables of interest were age, sex, weight, anthropometry on admission and discharge, length of stay on treatment and reason for discharge. Data were edited and entered into Excel. Data were exported to Statistical Application Systems (SAS) for further cleaning and analysis. Analyses involved descriptive and inferential statistical analyses including frequencies and distributions of all variables.

Structured individual interviews were then held with 91 clients, ten administrators and 38 health care workers. Focus group discussions (FGDs) of five to ten participants, drew further information from clients (four groups) and service providers (five groups) using openended questions and scenario methods, led by a moderator and recorded by a note-taker. Qualitative data were analysed manually by the evaluation team, who read through the interviews to identify emerging themes.

Results

Integration - Clinical facilities: The majority of ART clinics achieved integration of FBP and ART services as intended in Figure 1. Facilities with the weakest community outreach programmes had the highest number of defaulters.

Some facilities, having achieved competence with FBP implementation, extended services to satellite sites. These were significantly more difficult to manage but reduced congestion at the hospitals and greatly increased FBP enrolment.

Integration - HBC: HBC programmes provided a decentralised, ‘one-stop’ service model. Clients were assessed, counselled, prescribed and dispensed rations by a trained caregiver at the parish office. Home-based follow-up was provided by assigned HBC providers whose role was to support both FBP and ART adherence. HBC service providers and clients reported that integration was seamless.

Integration - Hospice: Service models varied considerably among hospices, with hybrids of centralised and decentralised models tried. Success with integration varied, possibly due to reliance on lay counsellors and volunteer caregivers (in contrast to technical staff employed by ART clinics) as well as less rigorous recordkeeping (in many hospices), and less management oversight in some cases.

Service provision: Anthropometric assessment was often done selectively (on clients who appeared malnourished) rather than as a routine aspect of the standard of care. While weighing and recording weight is a standardised practice, BMI is rarely calculated and MUAC is seldom used. FGD respondents at all sites noted the need for additional training and supervision to ensure adherence to admission/ discharge criteria and to improve skills and consistency in nutrition assessment.

Respondents consistently reported that active supervision positively influenced staff commitment to providing nutrition assessment and education, and improved accuracy of record-keeping. Many staff requested incentives for providing FBP services. The pilot, however, was neither designed nor budgeted to accommodate incentive requests.

Only 11% of clients reported that they were linked to livelihood activities, illuminating the enormity of the gap in this aspect of FBP programming.

Food storage and dispensing: Lack of FBP commodity storage space was cited as a significant challenge. CRS generally disbursed commodities every two months to accommodate storage limitations. Some alternate storage locations, such as kitchens, did not meet storage standards for temperature and humidity. There was no consensus among service providers regarding the ideal location for food dispensing but agreement that each setting should evaluate its options with emphasis on creating the most seamless, efficient pathway for clients.

Overall, supply chain management was a significant challenge, with three primary difficulties noted:

1. Month-to-month new enrolment numbers varied considerably
2. Length of client enrolment varied
3. Short shelf-life of selected commodities reduced prepositioning options.

Monitoring and Evaluation: The project sought to align with ‘the 3 Ones’⁸ by contributing to a single national reporting system, but was obliged to create a parallel approach because the existing system, SmartCare, did not allow for the capture of comprehensive nutrition data (e.g. BMI and WHZ). Lack of nutrition training, combined with the lack of tools and systems for data collection, have resulted in a nationwide gap in the detection, tracking and treatment of malnutrition among PLHIV, especially adults.

Client weight gain and BMI: All sites showed an increase in client BMI between admission and discharge. Among adult clients, the average BMI on admission was 17.6 kg/m2 and the average BMI on discharge was 20.5 kg/m2. The overall average increase in BMI pre-FBP to post-FBP was 2.9 kg/m2. Most clients required three to six months of nutrition rehabilitation to qualify for discharge.

Of the 22% of clients already discharged from the programme at the time of the evaluation, 997 (84%) met discharge criteria, 127 (11%) died from various causes, 45 (4%) were unknown or lost to follow-up and 18 (1%) were removed from treatment because of medical complications.

Client health status: Clients were asked to rate their pre- and post-intervention health status using the Eastern Cooperative Oncology Group (ECOG) performance scale⁹. The percentage of clients who were ‘fully active’ went from 5% pre-FBP to 51%, post-FBP. Only 1% of clients remained ‘completely disabled’ post-FBP, compared to 17% pre-FBP.

Limitations

Site records and quantitative datasets had numerous missing anthropometric data which limited the scope of analysis. Geographic distance, communications challenges, delayed project start-up and time constraints resulted in a disproportionate number of enrolled clients (thus fewer-than-planned rehabilitated and discharged) represented in the evaluation sample.

The short six-month project implementation period was sufficient to measure integration activity but necessitated pooling of clients across several sites in order to obtain a sufficient sample, which may have masked site-specific patterns.

It should be noted that weight gain, BMI and ECOG performance cannot be attributed exclusively to a FBP intervention. It is understood that nutritional status and activity level are likely to improve with ART only, or with some other combination of ART and nutrition.

Conclusions and recommendations

Integration of FBP into existing HIV care and treatment was successfully adapted to facility, home-based care (HBC) and hospice service delivery settings. Integration did not interrupt existing service delivery and can be accomplished using available human and material resources. The ‘medicalisation’ concept was appreciated and understood by both clients and service providers and the selected rations were successful in treating malnutrition. Weight gain and body mass index (BMI) improved while percentage of discharges cured (i.e. nutritionally rehabilitated) exceeded standards. In addition, activity levels and perception of wellness improved dramatically.

The keys to success were on-going support for application of nutrition concepts and careful record-keeping, and the identification of site coordinators who brought both technical nutrition knowledge and a high level of commitment to the pilot project. However, on-going training is required in nutrition, record-keeping and reporting. Future implementers would benefit from formal incorporation of new (FBP) tasks through either scopes of work for key staff, or the full integration of FBP responsibilities into standard job descriptions. These additional tasks may have implications for remuneration. Furthermore, integration of FBP commodities into the medical stores procurement and distribution system would reduce duplication of effort and promote national ownership.

The national ART M&E systems must be expanded to capture nutrition data. To foster timely discharge, linkages to wrap-around food security and livelihood programmes should be designed from the early stages of project conceptualisation.

Children and pregnant/lactating women were under-represented, suggesting that Maternal & Child Health (MCH) and Prevention of Mother to Child Transmission (PMTCT) programmes should be more intentionally included in scale-up plans. With regard to the use of MUAC, it was suggested that it be used for screening only, applying an increased cut-off to trigger referral of potential clients for assessment by a clinician.

For more information, contact Kate Greenaway, email: kate.greenaway@crs.org

¹Johannessen A, Naman E, Ngowi BJ, Sandvik L, Matee MI, Aglen HE, Gundersen SG, Bruun JN: Predictors of mortality in HIV infected patients starting antiretroviral therapy in a rural hospital in Tanzania. BMC Infect Dis 2008, 8(52).

²Stringer JS, Zulu I, Levy J, Stringer EM, Mwango A, Chi BH, Mtonga V, Reid S, Cantrell RA, Bulterys M et al: Rapid scale-up of antiretroviral therapy at primary care sites in Zambia: feasibility and early outcomes. Jama 2006, 296(7):782-793

³Zachariah R, Fitzgerald M, Massaquoi M, Pasulani O, Arnould L, Makambe S, Harries AD: Risk factors for high early mortality in patients on antiretroviral treatment in a rural district of Malawi. AIDS 2006, 20(18): 2355-2360.

⁴Madec Y, Szumilin E, Genevier C, Ferradini L, Balkan S, Pujades M and Fontanet A: Weight gain at 3 months of antiretroviral therapy is strongly associated with survival: evidence from two developing countries. AIDS 2009, 23(7): 853-861.

⁵Cantrell RA, Sinkala M, Megazinni K, Lawson-Marriott S, Washington S, Chi BH, Tambatamba-Chapula B, Levy J, Stringer EM, Mulenga L, Stringer JS: A pilot study of food supplementation to improve adherence to antiretroviral therapy among food-insecure adults in Lusaka, Zambia. J Acquir Immune Defic Syndr 2008, 49(2):190-5.

⁶Ministry of Health and National AIDS Commission: Zambia Country Report: Monitoring the Declaration of Commitment on HIV and AIDS and the [sic] Universal Access Biennial Report, Lusaka; 2010.

⁷Koethe JR, Lukusa A, Giganti MJ, Chi BH, Nyirenda CK, Limbada MI, Banda Y, Stringer JS: Association between weight gain and clinical outcomes among malnourished adults initiating antiretroviral therapy in Lusaka, Zambia. J Acquir Immune Defic Syndr 2010, 53(4):507-513.

⁸The ‘3 Ones’ is a set of three key elements that underpin a coordinated national response: One agreed HIV/AIDS Action Framework that provides the basis for coordinating the work of all partners. One National AIDS Coordinating Authority, with a broad-based multi-sectoral mandate; and One agreed country-level Monitoring and Evaluation System. (UNAIDS, 2004)

⁹Oken, MM, Creech, RH, Tormey, DC, Horton, J, Davis, TE, McFadden, ET, Carbone, PP: Toxicity And Response Criteria Of The Eastern Cooperative Oncology Group. Am J Clin Oncol 1982, 5(6):649-655.