Impact of nutritional supplementation amongst PLHIV in Zambia

By Daphyne Williams

Daphyne Williams is currently Technical Advisor HIV and AIDS with Catholic Relief Services where she has worked for the past 2 years.

The author wishes to thank Cecilia Adalla, James Campbell, Elizabeth Jere, Linda Lovick and Shannon Senefeld for reviewing and contributing to this article. She would also like to acknowledge the work of CRS Zambia, SUCCESS staff and its partners in their continued efforts in HIV.

This article describes the impact of nutritional supplementation on people living with HIV/ AIDS, determined through a field study by CRS.

The Catholic Relief Services (CRS) Zambia Scaling-Up Community Care to Enhance Social Safety-nets (SUCCESS) project is based on a holistic approach to HIV care. It includes home-based care, community based counselling and testing, palliative care, the prevention of mother to child transmission of HIV, as well as targeted nutrition. From 2005 to 2006, CRS embarked on a targeted evaluation of the SUCCESS project's nutritional supplementation efforts¹.

At the time of the study and to date, there is little information on the impact of nutritional supplementation on the lives of people living with HIV (PLHIV) pre-antiretroviral treatment (ART) in high prevalence settings. This study built upon anecdotal project evidence that as a result of nutritional supplementation, PLHIV gained weight and regained some degree of previous activity levels. In many instances, it was observed that individuals previously bed bound were able to regain enough strength to re-enter family activity. This study hypothesised that chronically ill PLHIV, not on ART, who received nutritional support would experience multiple positive impacts as a result of the nutritional supplementation.

Method

A quasi-experimental design was used to examine the impact of nutritional supplementation on chronically ill PLHIV, not yet on ART, over a 6 month period. The design, vetted and approved by CRS and USAID, was approved by the Zambian Internal Ethics Review Board. The evaluation occurred in three dioceses in the implementation catchment areas of the Catholic diocese of Solwezi, Mongu and Monze. The study design had three arms:

• Home-based care (HBC) clients living with HIV who received high energy protein supplements (HEPS) and vegetable oil. (Solwezi)
• HBC clients living with HIV who received bulgur wheat or sorghum and common beans or peas. (Mongu)
• People living with HIV who received no food aid or nutritional support. (Monze)

The nutritional supplementation in Solwezi, in the Northwest province, was purchased locally using the President's Emergency Plan for AIDS Relief (PEPFAR) funds from USAID. In Mongu, the Western province, food was provided by USAID's Food for Peace programme through the Consortium for Southern Africa Food Security Emergency (C-SAFE), implemented by CRS.

The study sites were selected based on general similarities, including rural areas with mid-sized urban centres, topography, cultural background, levels of food insecurity, an existing long term project presence in Mongu and Solwezi and the expectation of expanding SUCCESS to Monze. However, the HIV prevalence in the areas varied somewhat at the time of the study. Solwezi had the second lowest HIV prevalence in the country, an estimated nine percent. Mongu's HIV prevalence was estimated at thirteen percent and Monze had the third highest prevalence in Zambia, eighteen percent. Participants for this study were solicited from current clients of the SUCCESS HBC programmes in Solwezi and Mongu. The HBC programme operating in the diocese of Monze was used as a control, as it did not have a food supplement programme. All of the HBC programmes offered similar services and activities. Initial eligibility for the HBC programmes was decided by clinical determination and referral. To be eligible to participate in the study, the HBC clients needed to fulfil all the following criteria:

• A registered HBC client (i.e. chronically ill) in Solwezi, Mongu or Monze
• Food insecure as determined by SUCCESS programme screening tool
• New to HBC nutritional supplementation component (cases)
• Not receiving food aid from another source (controls)
• 18 years old or older
• Not on ART, and
• Confirmed HIV status (by HBC volunteers).

HBC programme design

The HBC programmes followed very similar structures. In each programme, community members identified individuals as potential volunteers. After completion of HBC training, the volunteers visited HBC clients weekly or more, depending on level of illness. The HBC volunteers provided their clients with basic psychosocial and pastoral support as well as general health and prevention education. Individual client care plans were developed by nurses and followed by volunteers. Many of these care plans included directly observed treatment support (DOTS) for tuberculosis, clinical and social service referrals for clients and/or family members, care and support for orphans and vulnerable children, nutritional counselling and access to HIV voluntary and confidential counselling and testing.

Food distribution by CRS as part of HBC

The study utilised a written questionnaire developed in Lusaka, Zambia in close collaboration with representatives from partner implementers in Solwezi and Mongu. The questionnaire was composed of questions tested and used by CRS, C-SAFE, World Food Programme, and researchers from Johns Hopkins and other universities. The final survey instrument was written in English, but administered in the appropriate local language and back translated. The survey instrument targeted both the household and individual HBC client levels and focused on thirteen thematic sections including household demographics, coping strategies, household receipt and use of food aid, current medical treatment, anthropometric information, and quality of life measures.

Study participants

More than half of the HBC clients interviewed at baseline were from Monze, which ensured sufficient participants with similar food insecurity as in Solwezi and Mongu. Of the 931 participants identified in Monze, only 684 met the food insecurity criteria required for participation in the study, of which 168 were removed because they either received food aid or were on ART. At baseline, there were 516 participants from Monze, 413 from Mongu and 380 from Solwezi giving a total of 1,309 study participants.

At the end of the study, there were 911 HBC clients still participating. The reasons given for those who had not continued participation in the end line questionnaire included, among others, death, relocation and programme discharge. Of the remaining HBC participating clients, 578 client records were not included in the analysis because the clients started on ART or had unknown or HIV-negative test results. The final sample size consisted of 81 clients in Mongu, 124 in Solwezi and 123 in Monze forming a total of 328. Women who self-reported as pregnant or lactating, though included in the final overall analysis, were not included in the analysis of anthropometric measures.

Results

Food distribution by CRS as part of HBC

The participants in this study were similar in many ways. At least 60% were women and more than half were heads of households. The average household size ranged from 6.58 to 7.48. However, in Solwezi, clients were on average younger and more likely to be married than the clients in Mongu or Monze. In addition, the sources of income were quite diverse within and across dioceses. In terms of income, the most frequently reported earnings were less than the equivalent of 11 United States Dollars per month, across and within dioceses.

In the intervention sites, survey questions were included to determine the acceptability of the nutritional supplements as well as to verify use of the supplements. In Solwezi, where the monthly ration consisted of HEPS and oil, 93.9% of the clients surveyed reported eating HEPS in the 30 days prior to the survey. In Mongu, the ration of beans and bulgur wheat (or sorghum and peas) was reported to be eaten in the previous 30 days by 91.1% of respondents. However, the study found no significant change in the food consumption score, a proxy for the diversity and nutritional quality of the household diet. In Mongu and Solwezi, when combined as one intervention arm, there was a statistical change between baseline and end line food consumption score. However, analysis determined no correlation between food consumption scores and gender, age, selfreported health status or household size.

The study also measured coping strategies using the Coping Strategy Index (CSI), which measures the frequency and severity of household strategies to cope with food insecurity over the previous 30 days. The CSI is an inverse measure, meaning the higher the score, the more frequent and severe the coping strategies. For this study, there was a maximum CSI score of 177.5 and minimum of 35.5. At baseline, the mean score for Mongu was 87.11, and 81.22 and 79.32 for Solwezi and Monze, respectively. At end line, the mean scores were significantly lower in Solwezi (p<0.05) and Mongu (p<0.001). In Monze, the mean CSI score was significantly higher (p<0.001), which indicates more frequent and severe coping strategies used in Monze.

The study used the Eastern Cooperative Oncology Group (ECOG) scale, which measures performance status on a 5 point scale, ranging from '0, fully active, able to carry out all pre-disease activities without restriction'; to '4, completely disabled, cannot carry on any self-care, totally confined to bed or chair.' ECOG analysis examined the overall gain for all clients. The score was calculated by taking the difference between the end line and baseline ECOG scores. The majority of clients ranked between 2 and 3 on the ECOG scale and were, therefore, capable of self care and ambulatory, but were unable to work. Twice as many clients in Monze, the control arm, reported worsening performance during the study period. While both intervention sites reported increases in performance, the difference in the mean gain between the sites was not significant.

The anthropometric measures of Body Mass Index (BMI) and Middle Upper Arm Circumference (MUAC) were taken at baseline and end line. At baseline, the three study groups were not significantly different from each other in terms of mean BMI or MUAC. The mean end line BMI was not significantly different from baseline. However, there were slight increases in BMI in Solwezi and Mongu and a slight decrease in Monze from baseline to end line, although these changes were not statistically significant. With regard to MUAC values, there were significant improvements, with statistically significant changes in the combined intervention arms when compared to the control arm (p<0.001).

Conclusions

Home based care in Solwezi

This evaluation showed that even modest nutritional supplementation can improve the nutritional status of PLHIV in insecure households. Furthermore, nutrition support can improve mental and physical health, reduce the need for support and improve individual ability to carry out daily activities - essentially improving the quality of life for PLHIV not yet on ART. In general, MUAC measurements increased in clients who received nutritional supplements, while those who did not receive supplements saw decreasing MUAC measurements. Quality of life measurements saw improvements for the intervention clients. Additionally, the intervention sites saw significant decreases in AIDS related symptoms and need for caregiver support. The evaluation findings support the following recommendations:

• Increase funding and support for food and nutrition supplements to food insecure PLHIV.
• Further study the impact of nutritional supplements using larger sample sizes.
• Further consultation needed on an appropriate and feasible combination of assess ment tools for wide scale monitoring and evaluation.

Update

Since the study, Mongu has not participated in food distribution, but CRS continues to support the HBC programme. Monze did not become a programme site. The SUCCESS program plans to expand its nutritional assistance to 6 dioceses which cover nearly five Zambian provinces.

Currently, SUCCESS implements a therapeutic feeding programme targeting 11 hospices and selected parishes from Solwezi and Chipata Diocese HBC programmes. The new programmes were designed to comply with the U.S. government's release of policy guidance for the use of emergency plan funds to address food and nutrition needs. The policy guidance restricts food support to specifically identified target populations based on WHO criteria and guidelines. SUCCESS now procures ready-touse therapeutic food (RUTF), a high energy peanut-butter-like product, and HEPS for treatment of people in palliative care programmes with severe or moderate acute malnutrition. These foods are provided to clients according to clear entry and exit anthropometric criteria and are not designed as long term food support. SUCCESS no longer provides other food products through the palliative care programme, such as those used in the earlier study.

In 2008, SUCCESS embarked on a pilot project that utilizes the 'Food By Prescription' approach to target clients with clinical malnutrition. This approach has 'medicalised' food by distributing it to clients in small daily 'doses' dispensed in clinic settings. The pilot project integrates malnutrition assessment and RUTF and HEPS distribution into nine large HIV treatment clinics and their respective decentralised clinic sites. Clients collect the food as they collect their monthly HIV medication. The pilot project will examine how this model can be replicated in the Zambian context.

For more information on this article, contact Daphyne Williams, email: dwilliam@crs.org

For more information about this or other CRS evaluations, please visit the CRS website at www.crs.org or email HIVunit@crs.org

¹The findings from this study have been presented at the American Evaluation Association 2008 Conference and the 2008 International AIDS Society, Mexico City. A complete write up of the 'CRS SUCCESS Palliative Care Nutritional Supplementation Targeted Evaluation' can be found at http://crs.org/publications/list.cfm?sector=6.