Assessing the level of adherence and its determinants is important in appraising the overall effectiveness of trials. The present study aimed to evaluate the extent of adherence and its determinants in a pragmatic randomized controlled trial of Fe prophylaxis during pregnancy in Maputo, Mozambique.

A pragmatic randomized controlled trial.

Two health centres (1o de Maio and Machava) in Maputo, Mozambique.

Pregnant women (≥12 weeks’ gestation, ≥18 years old, non-high-risk pregnancy; n 4326) attending prenatal care consultations at two health centres were randomized to receive routine Fe (n 2184; 60 mg ferrous sulfate plus 400 μg folic acid daily throughout pregnancy) or selective Fe (n 2142; screening and treatment for anaemia and daily intake of 1 mg folic acid).

The level of adherence was 79 % for having two or more visits, 53 % for adequate prenatal care and 67 % for complete intake of Fe/folic acid tablets during the trial. The correlation between the adherence measures ranged between 0·151 and 0·739. Adherence did not differ by trial arm, but there were centre differences in adequate prenatal visits and intake of tablets. Older women (>20 years) and those with a history of abortion were more likely to achieve greater adherence, whereas an increased number of previous births decreased the likelihood of adherence. HIV positivity decreased the likelihood of adherence in one trial centre and increased it in the other.

The variation in adherence by trial centre, women’s characteristics and outcome measures suggests that adherence in trials fully depends on participants’ behaviour and can be increased by paying attention to contextual factors.