Main objectives: To evaluate the impact of weekly vitamin A supplementation (VAS) to women of reproductive age (15-45 years) on maternal mortality in rural Ghana, and to compare this with the impact on overall mortality.

Hypotheses:

1. Weekly supplementation with vitamin A (7000 µg retinol equivalent [RE]) to reproductive age women will reduce maternal deaths by 33%.
2. This impact will be achieved by reductions in both pregnancy-related and non-pregnancy-related deaths.
3. There will be a reduction in non-maternal deaths, similar in size to that in maternal non-pregnancy related deaths.

Outcome measures: Maternal mortality rate, and overall mortality rate. Deaths will be identified through monthly demographic surveillance, and classified as maternal (pregnancy-related, non-pregnancy-related) or non-maternal using verbal autopsies.

Pregnancy accounts for nearly 600,000 deaths of women each year; maternal health problems are the largest contributors to the disease burden of adult women. Conventional primary health care approaches, which included Traditional Birth Attendant training and antenatal screening, had little impact on the maternal mortality ratio. Instead, the Safe Motherhood paradigm now aims to ensure emergency obstetric care (EMOC) at the district hospital level for the 10-15% of women who develop potentially life threatening complications, and is moving towards recommending that professionals attend all deliveries.

While the latter configurations of care have been shown to reduce maternal mortality, they require considerable political will, attention to health systems, and expansion of access to supervised delivery and EMOC. For the poorest countries, such capacity is some years down the line. Low-tech interventions which effectively reduce maternal mortality, and which can be delivered at the community level would be a welcome addition to the armamentarium of public health measures for preventing maternal mortality. Should vitamin A supplementation prove to be effective in reducing maternal mortality, or indeed all-cause female mortality, it would provide such a tool. Moreover, as there is considerable policy and programmatic interest in VAS for children, it is likely that such interest can be broadened to encompass supplementation for women. Furthermore, it is increasingly recognised that poverty not only increases the risk of ill health, but that ill health in turn plays a major role in creating and perpetuating poverty. A community-based intervention such as Vitamin A is likely to address the needs of the very poorest women, as these are the individuals least likely to have access to emergency obstetric care and professional birth attendants.

This will be a cluster-randomised double-blind placebo-controlled trial. All women between the ages of 15 and 45 years will be randomised, according to their cluster of residence to receive weekly capsules of either 7000 RE of vitamin A in peanut oil or identical looking placebo capsules containing peanut oil only. Thus, supplements will be delivered to women both in antenatal and inter-pregnancy periods.

The trial will be conducted by the Kintampo Health Research Centre (KHRC) in four contiguous districts - Kintanpo, Techiman, Wenchi and Nkoranza -- in the Brong Ahafo region of Ghana. The districts fall within the forest-savannah transitional ecological zone, and vitamin A rich food sources are less available than in the forest regions to the south. Data from previous studies by KHRC and from a national prevalence survey, both indicate a VAD problem of public health significance in the area -- 26% of breastmilk samples have retinol concentrations lower than 30µg/dl, exceeding the WHO cut-off of 25% for defining areas with a severe problem (WHO, 1996). VAS has been found to substantially reduce childhood morbidity and mortality in similar areas, thus it is suitable for testing the potential benefits of VAS to women.

All women aged 15-45 years who are permanent residents in the study areas will be eligible for recruitment into the trial. They will be identified from existing databases. Permanent residence is defined as having been resident in the area for the three months preceding the start of recruitment, with intention to remain in the study area for the following 12 months. There will be no exclusions to participation, except for women who have nightblindness or other signs of VAD. These, and any women who develop VAD in the course of the study will be treated according to current IVACG recommendations (IVACG, 1997). They will continue to be followed, but will be given vitamin A and considered separately in the analysis. Continuous recruitment will be done for women who migrate into the study area, or those who become eligible by age as the study progresses. Allocation to treatment will be determined by the cluster of residence.

Inclusion Criteria:

• All women of reproductive age (15 to 45 years) who are permanent residents in any of the 4 districts in rural Ghana (Kintampo, Wenchi, Techiman, and Nkoranza)

Exclusion Criteria:

• There will be no exclusions to participation, except for women who are unable to give their informed consent to join the study