Different Approaches for Delivery of IPT in Pregnancy in Burkina Faso

Pregnant women are more susceptible to malaria than non-pregnant women, in particular during the first pregnancies. Malaria infection in pregnant women may cause anaemia, illness and abortion or stillbirth. It may also cause low birth weight through placental infection. In Africa the burden of malaria in pregnancy is highest in remote rural areas. IPT with sulphadoxine - pyrimethamine (SP) has been demonstrated to be effective in preventing pregnancy related malaria (anaemia, clinical cases and placenta infection), and to be beneficial to both babies and pregnant women, and increasing ANC attendance. In many Sub - Saharan African countries, IPT / SP is being adopted to replace chloroquine (CQ) as chemoprophylaxis was shown to be no longer efficacious. IPT/SP was adopted in 2005 by The Ministry of Health (MoH) of Burkina Faso to replace chemoprophylaxis with CQ. The new strategy is being implemented but no delivery approach was defined and presumably IPT/SP will only be delivered to pregnant women presenting at ANC visits. Although the expected compliance should be better than for the previous CQ weekly regime, it would be of importance to ensure a better coverage and higher compliance to make the new strategy effective. Therefore, in order to obtain a more efficient IPT / SP programme with a good level of compliance and coverage, several delivery approaches beside use of ANC alone will be explored. It is hypothesized that the IPT strategy using SP will be as inefficient as chemoprophylaxis with CQ if an effective delivery approach is not used.

General Objective: to assess different IPT/SP delivery approaches in rural Burkina Faso with the aim of improving coverage and compliance among pregnant women.

Specific objectives:

• To implement three approaches of IPT/SP delivery to pregnant women based on health centre distribution, and in addition community based distribution either by health services or by community members;
• To compare the effects of the approaches on compliance to and coverage of IPT / SP and their feasibility.

Material \& Methods

The research will be implemented in Burkina Faso. Malaria is endemic with seasonal transmission and is responsible for considerable ill health among pregnant women. The study site will be situated in the rural area of the Pissy health district which covers both the western peri-urban zone of Ouagadougou city and adjacent rural areas. There are a total of 25 health facilities in the district (both rural and urban) and a reference hospital at Ouagadougou. In 2004 the population of the district was approx. 542000 inhabitants. The estimated number of pregnancies was 27000. There is generally low attendance for antenatal care (ANC) and 15% of babies are born with LBW. In a study carried out in the Boromo rural district, only 51% of pregnant women had 2 ANC visits before delivery (Coulibaly, personal communication). HIV infection prevalence was found around 4% for the whole country (MoH, 2005) and there is no indication that it is different for the Pissy district catchment area.

The study participants will include pregnant women irrespective of gravidity residing in the study area and reachable by the approach that is applied in the zone.

It is a prospective comparative study of 3 different approaches of delivering IPT / SP in the catchment areas of rural health facilities. The approaches will be the following:

1. Passive health centre based delivery approach (PHC). IPT/SP will be delivered to all pregnant women presenting to the health centre either for ANC visit or for care seeking consultation. Drugs will be administered as a directly observed treatment (DOT). Some collation may be given to women as some of them refuse to take drugs because they have not eaten before. This approach, which is the most likely to be applied by the MOH, will serve as the control one.
2. Joint methods, with an advanced strategies delivery approach (JAS). In addition to passive delivery at health centres, the pregnant women will be reached during preventive activities the health staff carry out regularly in villages, such as immunization, health promotion, and even ANC visits. The women will be asked to present to the health staff for treatment and IPT/SP administered as DOT.
3. Community based distribution delivery approach (CBD). In addition to passive delivery at health centres, the pregnant women will be reached by traditional birth attendants (TBAs) or representatives of village women's associations (RWAs). The IPT/SP is then delivered also as DOT.

The two community - directed approaches will be further developed through workshops with health staff, communities, and resource persons to adapt them to the local situation. Each approach will be implemented in a zone constituted by the catchment area of a number of health centres to cover the required sample size. The zones will be selected in order to avoid contamination between them and randomly assigned to a delivery approach. The main outcomes to be measured are: a) the coverage of IPT (proportion of pregnant women who are administered the treatment), b) the compliance (number of administered doses per pregnant woman), c) the infection prevalence, d) the Hb rate, e) the most relevant difficulties and constraints to each approach, f) the acceptability to population and health staff and g) the performance of each approach to deliver IPT /SP. The sample size is calculated to measure a difference between groups of 10% in IPT administration coverage. Assuming we require 80% power (zβ = 0.84) of detecting a significant difference (P<0.05; zα/2 =1.96) if the intervention increases coverage by 10%, each group should be composed of n = 3841 pregnant women. Assuming a design effect of 1.5, n = 5762 women. The 3 geographical zones will be defined to each cover 5800 expected pregnancies.

Cross sectional surveys will be carried out at the beginning, during and at the end of the study period. The first will aim to ensure the study zones are comparable regarding the outcomes a) b) c) and d) and later on for comparison between zones. Furthermore, SP will be measured in urine using dipsticks for compliance measurement. An anthropological component will assess part of main outcomes e) and f) in health staff, communities, and pregnant women. Questionnaires, focus group discussions (FGD) and in-depth interviews (IDI) will be used for cross-sectional survey data collection. Data about pregnancy and IPT will be recorded from personal ANC book to be credible. The sample sizes will be calculated to be representative of the populations of different villages and zones.

Data on IPT / SP delivery will be also collected from records hold by health centre and TBAs / RWAs to measure main outcome e). Specific records different from the health centre's own records will be opened at each IPT/SP delivery site, including TBAs / RWAs, to register information on treated pregnant women. The information will be personal data (name, age, village, marital status, and profession), data on the current and previous pregnancies (rank of pregnancy, children, ANC visits, drugs given) and medical history (previous adverse reactions to drugs, fever, and malaria attacks), and the administered IPT/SP doses. As usual, the IPT / SP treatment received will be recorded in the women's personal ANC book as proof for further use, including the cross sectional surveys.

Quantitative data will be regularly entered with Epi Info software and analysed using Stata version 9. Qualitative data will be processed and reports produced. Proportions will be compared by ANOVA and means by student t test.

Training sessions will be organized for caregivers to harmonize IPT administration (as it is not yet in practice in Burkina Faso) and study implementation in each study zone. The PI will undertake monthly supervision in each zone. During the preparation phase, more information will be gathered to better understand the health district and centres organization and functioning system regarding ANC and malaria prevention.

Study period The study will be carried out over 24 months from June 2007.