Complex intervention to optimise adolescent BMI pre-conception to address the double burden of malnutrition: A RCT in rural and urban South Africa

Throughout the world the numbers of people with non-communicable diseases (NCDs), like type 2 diabetes and high blood pressure, are increasing. The risk of developing diabetes is associated with being poorly nourished as a child then becoming obese later in life, being inactive, having a poor-quality diet, as well as influenced by genetic inheritance and poor intra-uterine growth. Low- and middle-income countries, like South Africa, have fast-growing numbers with NCDs yet health systems are struggling to cope with the ongoing burden of infectious disease (including HIV/AIDS and TB).
We have shown that in rural and urban South Africa a third or more young women are either overweight or obese by 20 years of age; while the problem of stunted growth is stubbornly persistent. Five to eight percent of 7 to 15 year olds already show early warning signs for the development of diabetes.

To combat an increasing prevalence of NCDs, it has been suggested that improving adolescent nutritional status may be a successful strategy, and that developing and evaluating pre-pregnancy interventions that promote nutrition and healthy behaviours in poor communities where childbearing tends to start at a younger age, may be part of the solution. Sub-Saharan Africa is the only region worldwide where the number of adolescents is predicted to grow, but it also faces a poor adolescent health profile. To-date, we are unclear about how best to intervene to improve the nutritional status of adolescents in low- or middle-income countries.

Our extensive formative and pilot trial research, has informed the development of an intervention targeting adolescent girls who reflect the double burden of malnutrition. The trial will enrol 1248 underweight or overweight/obese girls aged 14-19y from two-suitable settings in South Africa, one rural and one urban, using age- and sex-appropriate cut-offs. After baseline assessment and randomisation, trial participants will be reassessed after at least 18 months follow-up. If a participant becomes pregnant in this follow-up period, the reassessment will be carried out early in her pregnancy (<18weeks), and additional measures and antenatal clinic data will be collected during the pregnancy and from the infant at birth. The trial will include process and economic evaluations. This approach will allow us to assess the effects of our intervention - and whether there are grounds for a major expansion (scale-up).

The trial will employ and train community health workers (CHWs) to work with adolescents and their primary caregiver to address sub-optimal nutrition by: (i) promoting healthy behaviours and improving diets; (ii) providing a multi-micronutrient supplement to combat deficiencies, particularly anaemia; (iii) providing the primary caregiver with a conditional cash transfer to encourage and enable better household dietary diversity within the context of food insecurity and (iv) supporting adolescent girls who become pregnant to use antenatal care services earlier in their pregnancies and more frequently. CHWs will be trained in 'Healthy Conversation Skills', a set of skills to support behaviour change specifically developed and tested for use with socioeconomically disadvantaged women to improve their confidence so that they can achieve their health goals. CHWs will apply these skills in the rural and urban sites over the trial period, building relationships with adolescents and their families to encourage healthier adolescent lifestyles. The intervention is flexible and will focus on the most challenging areas for each participant. Health literacy seeks to increase adolescent access to health information, and develop their ability to use this information effectively. Adolescents who become pregnant will be further supported by CHWs to access and attend antenatal services, facilitate their caregiver's involvement, and reinforce optimal individual health during pregnancy.

South Africa faces a complex health burden with rising non-communicable disease against a background of prevalent infection. Malnutrition, undernutrition alongside overweight/obesity, is widely prevalent, imposing risks along the life course especially among adolescent girls.

We propose an intervention to optimise adolescent BMI pre-conception to address the double burden of malnutrition, using established research sites: rural Agincourt and Soweto.

We hypothesise that, by optimising change in BMI of adolescent girls at-risk, we can realise a triple return on investment: improved nutrition, reduced metabolic risk, and moderated peri-conceptional exposures to offset transgenerational risk for metabolic conditions.

We ask:
1. In high-risk nulliparous adolescent girls, can a community health worker-delivered intervention of nutrition (multi-micronutrient supplementation) and lifestyle change, with a conditional-cash transfer (i) achieve directionally-appropriate changes in BMI?, (ii) improve micronutrient status?, and (iii) influence individual behaviours?
2. Will these changes impact individual metabolic disease risk?
3. In those adolescent women who become pregnant, will reductions in variance of BMI from improved BMI status during the peri-conceptional period impact maternal glucose during pregnancy and new-born birth weight and adiposity?
Further: is the intervention package cost-effective; can findings contribute to strategies to tackle the adolescent double burden of malnutrition and reduce trans-generational risk for metabolic disease?

We will enrol 1248 underweight or overweight/obese girls 14-19y using age-/sex-appropriate cut-offs. After baseline assessment/randomisation, participants will be reassessed after 18-24 months follow-up. If a participant becomes pregnant, a reassessment will be conducted in pregnancy (<18weeks) and further measures collected during pregnancy and birth. We include both process and economic evaluations.

Trial will address a critical problem: South Africa's evolving burden of disease is complex due to (i) persisting underweight and micronutrient deficiencies; (ii) the highest prevalence of obesity in sub-Saharan Africa; and (iii) rising rates of non-communicable disease, such as type 2 diabetes (14% of black adult women live with T2D). Efforts to address the double burden of malnutrition in adolescent girls and young women may be an opportunity to address her future health and that of her offspring.

Impact of project on problem: Planned intervention during adolescence for girls, and in particular before they become pregnant in settings where childbearing is common at an early age, provides the best opportunity to impact effectively on an adolescent's future health and that of her offspring. The project team has engaged fully with target communities, adolescents, parents, educators and other key stakeholders, particularly the health sector, to ensure buy-in, collaboration, affordability and sustainability. We envisage that this intervention will reduce the risk of suboptimal nutrition in adolescents and contribute to lessening the present and future NCD burden in South Africa and similar middle- and low-income settings.

Trial will produce rigorous evidence: Including process and economic evaluations, which can inform potential later scale-up based within primary health care systems. Adolescents are a highly vulnerable and at-risk population for the double burden of malnutrition, with potential lasting effects for themselves and on the next generation. Early interventions to improve adolescent nutritional status and health are likely to have a significant impact and be more cost-effective than intervening later in life.

The intervention has great potential to impact on and strengthen the public health system (collaboration and co-production): Rigorous evaluation using hard endpoints will provide evidence for intervention scale-up within primary health care systems in South Africa, regionally and in other middle- and low-income countries undergoing complex health and nutrition transitions. In particular, we aim to work with and support the current national health system to adapt to these new and unfamiliar challenges. Thus the partnership with the South African Dept. of Health (DOH) was initiated early in the intervention development and needs assessment to ensure the project's relevance, with co-creation and input, and importantly to lay the basis for potential take-up of the evidence to shape policy and practice. The findings, experience and insight from the process evaluation will, we expect, prove instrumental in informing the DOH's Primary Health Care Re-engineering initiatives with CHWs and form an essential part of the pathway towards economic and societal impact. Notably, findings from this trial should translate into current adolescent health practice. Key local and national members of the DOH are part of the established Community Advisory Group and Scientific Committee.

In summary: impact of the trial will be through: (i) advancing our understanding on how to improve the nutritional status and health of adolescents living in poor rural and urban communities; (ii) understanding how a pre-pregnancy intervention during adolescence may leverage health benefits for adolescent women and their offspring if they become pregnant; (iii) shaping Dept. of Health policy and practice around CHWs; and (iv) enabling bi-directional knowledge transfer between UK and SA researchers and students involved in the study.