JMIR Publications

ABSTRACT

Background:

Unintended pregnancy is associated with poorer health outcomes for both women and their children. Fulfilling unmet need for contraception is essential in avoiding unintended pregnancies. Although the most effective methods of contraception are available in Bolivia, unmet need among women aged 15-19 is estimated to be 38% (2008) and the adolescent fertility rate is 71 per 1,000 women (2016). Mobile phones are a popular mode by which to deliver health behaviour support. The London School of Hygiene & Tropical Medicine and the Centro de Investigación, Educación y Servicios in Bolivia developed a contraceptive behavioural intervention for young Bolivian women delivered by mobile phone. The intervention development was guided by behavioural science and consists of short instant messages sent through an app over four months.

Objective:

The objective of this trial was to evaluate the effect of the intervention on young Bolivian women’s use of and attitudes towards the effective contraceptive methods available in Bolivia (oral contraceptive pills, intrauterine devices, injectables, implants and the patch).

Methods:

This was a parallel group, individually randomised superiority trial with a 1:1 allocation ratio. Women were eligible if they were aged 16-24, owned a personal Android mobile phone, lived in La Paz or El Alto, reported an unmet need for contraception (i.e. are sexually active, not on effective contraception and want to avoid a pregnancy) and could read Spanish. The target sample size was 1310 participants. Participants allocated to the intervention arm had access to an app with standard family planning information plus intervention messages. Participants allocated to the control arm had access to the same app plus control messages. Co-primary outcomes were use of effective contraception and acceptability of at least one method of effective contraception at four months. Secondary outcomes were use of effective contraception during the study, acceptability of the individual methods, service uptake, unintended pregnancy and abortion. Process outcomes included knowledge, perceived norms, personal agency and intention. Outcomes were analyzed using logistic and linear regression. We also asked participants if they experienced physical violence since joining the study.

Results:

640 participants were enrolled and 67% (n= 429) contributed follow-up data for the co-primary outcome, use of effective contraception. There was no evidence that use differed between the groups (33% in the control arm vs 37% in the intervention arm, adjusted OR 1.19, 95% CI 0.80-1.77, p = 0.40). There was a borderline significant effect regarding the acceptability outcome (63% in the control arm vs 72% in the intervention arm; adjusted OR 1.49, 95% CI 0.98-2.28, p = 0.06). There were no statistically significant differences in any of the secondary or process outcomes. Intervention dose received was low. In the control group, 3% reported experiencing physical violence compared to 2% in the intervention group (Fisher’s exact test p=0.75).

Conclusions:

This trial was unable to provide definitive conclusions regarding the effect of the intervention on use and acceptability of effective contraception due to under recruitment. While we cannot strongly recommend implementation, the results suggest that it would be safe and may increase acceptability of effective contraception if the intervention messages were offered alongside the download of the app. Clinical Trial: ClinicalTrials.gov Identifier NCT02905526