Menu
Menu
Menu
Menu
Background Direct-to-beneficiary communication mobile programmes are among the few examples of digital health programmes to have scaled widely in low-resource settings. Yet, evidence on their impact at scale is limited. This study aims to assess whether exposure to mobile health information calls during pregnancy and postpartum improved infant feeding and family planning practices. Methods We conducted an individually randomised controlled trial in four districts of Madhya Pradesh, India. Study participants included Hindi speaking women 4-7 months pregnant (n=5095) with access to a mobile phone and their husbands (n=3842). Women were randomised to either an intervention group where they received up to 72 Kilkari messages or a control group where they received none. Intention-to-treat (ITT) and instrumental variable (IV) analyses are presented. Results An average of 65% of the 2695 women randomised to receive Kilkari listened to ≥50% of the cumulative content of calls answered. Kilkari was not observed to have a significant impact on the primary outcome of exclusive breast feeding (ITT, relative risk (RR): 1.04, 95% CI 0.88 to 1.23, p=0.64; IV, RR: 1.10, 95% CI 0.67 to 1.81, p=0.71). Across study arms, Kilkari was associated with a 3.7% higher use of modern reversible contraceptives (RR: 1.12, 95% CI 1.03 to 1.21, p=0.007), and a 2.0% lower proportion of men or women sterilised since the birth of the child (RR: 0.85, 95% CI 0.74 to 0.97, p=0.016). Higher reversible method use was driven by increases in condom use and greatest among those women exposed to Kilkari with any male child (9.9% increase), in the poorest socioeconomic strata (15.8% increase), and in disadvantaged castes (12.0% increase). Immunisation at 10 weeks was higher among the children of Kilkari listeners (2.8% higher; RR: 1.03, 95% CI 1.00 to 1.06, p=0.048). Significant differences were not observed for other maternal, newborn and child health outcomes assessed. Conclusion Study findings provide evidence to date on the effectiveness of the largest mobile health messaging programme in the world. Trial registration number Trial registration clinicaltrials.gov; ID 90075552, NCT03576157.
The study is an individually RCT with a parallel and unblinded design. At the time of randomisation in late 2018, women enrolled to the study (n=5095) were 12–34 weeks of gestation, more than 18 years of age, could speak and understand Hindi, and owned or had access to a mobile phone during the day when Kilkari calls were likely to come. Women were excluded who did not consent or who were mobile subscribers of Bharat Sanchar Nigam Limited (the state-owned telecommunications company) due to poor network coverage in the RCT districts in Madhya Pradesh. Given the shared nature of mobile phones, the husbands of women enrolled to the study were additionally interviewed as part of endline survey activities. Women in India have limited access to and use of mobile phones. Despite near universal household level phone ownership (92.8%), only 47.8% of Indian women have access to a mobile phone (41.6% in rural areas and 62.7% in urban).18 In the central Indian state of Madhya Pradesh (population 75 million) which is home to an estimated 20% of India’s total population, 19.1% of women rural areas and 49.5% in urban had access to a phone that they themselves could use in 2015.19 Beyond limitations in women’s phone access, Madhya Pradesh’s population health status falls below national level averages for most RMNCH indicators. In 2015, only 49.6% of women reported using any modern method for family planning and 12.1% reported having an unmet need for family planning.19 While over half (53.0%) of pregnant women attended ANC clinics in the first trimester, only 35.7% received the recommended four ANC visits.19 Despite high rates of institutional delivery (80.8%), only 54.9% of women and newborns received a postnatal health check within 2 days following birth.19 Among children, 34.4% were breastfed within 1 hour of birth and 58.2% were exclusively breastfed until 6 months of age.19 One in four children (25.8%) were wasted (weight-for-height) and 42.0% were stunted (height-for-age).19 Among children 12–23 months of age, 53.6% were fully immunised (Bacille Calmette-Guerin (BCG), measles and three doses each of polio and Diphtheria, Pertussis, and Tetanus (DPT)).19 Kilkari is comprised of 90 min of RMNCH content sent via 72 once weekly voice calls: 24 during pregnancy, 24 within the first 6 months postpartum and 24 from 7 to 12 months postpartum. Individual calls span an average of 77 s in duration and are framed as coming from ‘Dr Anita’. Across health content areas, 18% of cumulative call content is on family planning (benefits of family planning, modern reversible methods, sterilisation, pregnancy tests); 13% on child immunisations (diseases covered, doses); 13% on nutrition (malnutrition, growth monitoring, anaemia); 12% on infant feeding (quality of food, breast feeding, complementary feeding, anaemia); 10% on pregnancy care (ANC, institutional delivery, rest, nutrition, tetanus toxoid, emergency services); 7% on entitlements; 7% on diarrhoea; 7% on postnatal care (newborn danger signs, cord care, hypothermia); and the remainder on a range of topics including intrapartum care, water and sanitation (WASH) and early childhood development (figure 1). Summary of Kilkari content during pregnancy and up to 1 year post partum. WASH, water and sanitation. The mobile numbers provided by women randomised to receive Kilkari began receiving calls no later than 8 months (34 weeks) after conception, making them eligible for at least 57 of the 72 messages. The timing of RCT enrolment was intended to mirror the timing of enrolment to Kilkari elsewhere in India. Once subscribed, Kilkari used an algorithm to retry mobile numbers up to nine times—three times in each week—three times on the first day, and twice then twice each day for the next 3 days—to reach a subscriber. Respondents were given the option to unsubscribe from the service at any time. The primary outcome of this study was the reported practice of exclusive breast feeding for infants 0–6 months of age. The secondary outcome was the use of modern reversible contraceptive methods (including intrauterine contraceptive device (IUCD), injectables, oral contraceptive pills, emergency contraceptive pills, condoms) at 1 year post partum. The latter outcome was added after the trial registration but included in the published protocol. Three reasons underpin its inclusion. First, family planning messages constitute the largest overall message share (figure 1). Second, family planning calls occur throughout the extended postpartum window allowing for a large window of exposure to the intended messages. Third, the complexity of phone sharing practices in this population means that men may be the actual listeners of some of the Kilkari calls.20 Family planning represents includes behaviours which depend on both men and women both of whom may have been exposed to Kilkari calls. The outcome of immediate/early breast feeding noted in the trial registration was not emphasised as a primary or secondary outcome in our protocol. The study team felt that changes in these outcomes might be challenging to observe for two reasons. First, immediate/early breast feeding can be supply side dependent in the case of facility based and/or skilled attendant deliveries (which is why we reported it for normal deliveries and all deliveries) and thus, information provided to pregnant women is unlikely to move the indicator. Second, there are only two Kilkari messages (message 17: pregnancy month 8, week 1 and message 23: pregnancy month 9, week 3) which include content on early and immediate breast feeding. To assess the impact of Kilkari exposure on RMNCH outcomes, endline surveys were administered to enrolled women and their husbands after 12 months postpartum. The women’s endline survey included modules on RMNCH knowledge, practice, decision-making and demand for information and supply side services for the following health areas: pregnancy and delivery care, newborn care, child health, infant and young child feeding (IYCF), family planning, media consumption and reported demand for and receipt of health services from frontline health workers. The men’s survey sought to measure RMNCH knowledge; mobile phone ownership, use and literacy; and women’s access to phones. In addition to relying on respondent recall to answer survey questions, investigators also collected data from the participant’s Mother and Child Health card, which is issued to every pregnant woman at the local health facility and updated each visit and includes the child’s immunisation record. To detect a 5% difference in the reported practices of exclusive breast feeding and reversible modern contraceptive use assuming an alpha of 0.05, 80% power, the estimated sample size for a two-sample proportions test would be 3200 women. After adjusting for a design effect (variance inflation factor) of 1.25 due to clustering at the level of the Community Development Block (India’s subdistrict level administrative units), a 20% loss to follow-up from enrolment to endline, and a potential loss of 35% of women due to poor reporting of phone numbers and/or changes to original phone numbers provided, a total of 5100 women were targeted for enrolment in the study. We assumed that 40% of women in the reproductive age would have access to a mobile phone in 2018 at the start of the study, and that 12.5% of households would have a pregnant woman between 4 and 7 months, and thus anticipated needing to visit up to 750 000 households across the four study districts. Study participants were identified through a household listing survey carried out from July to October 2018 which included an intensive mapping of households and their residents and screened participants for eligibility. Women identified during the listing as meeting the eligibility criteria (Hindi speaking, 4–7 months pregnant, with access to a mobile phone during the day) were screened again by a separate baseline survey team. Those confirmed to meet eligibility criteria were administered a 1.5 hour baseline survey inclusive of modules on sociodemographic characteristics; phone ownership, use and digital competency; RMNCH knowledge and practices. Following the baseline survey, study participants who completed the face to face survey and consented to receive Kilkari messages were randomised to receive Kilkari calls (intervention) or not receive messages (control). Exposure to Kilkari spanned through the child’s first birthday. Study participants were interviewed at 12 months post partum as part of endline survey activities carried out from November 2020 to March 2021. No harms are anticipated for study participants enrolled to either study arm. The participants were randomised after stratification on a range of variables potentially associated with exposure (listening to messages) and likely to influence outcomes, including gestational age, parity, age of woman and ownership of phone. Stratification sought to ensure a balance of covariates between the intervention and control groups. The individual randomisation procedure was done on a block-by-block basis (block randomisation) after enrolment was completed for each block. This ensured similar number of intervention and control subjects for each community development block. Randomisation was performed by picking a random sample of participants for each Block using the sample command in Stata with the use of the above listed variables as stratifiers. The randomisation and allocation processes were carried out by the study investigators in late 2018 following the completion of baseline survey activities. The random sample picked was allocated to receive the intervention immediately after randomisation. The participants could not be blinded due to the nature of the intervention. The data collectors administering surveys were not aware of the allocation. To assess exposure to Kilkari content, call data records from the interactive voice response (IVR) system were linked to baseline and endline survey data. Listening patterns were assessed for each subscriber by call, for the duration of the their subscription to Kilkari. The content of the Kilkari calls was mapped to key health outcomes including knowledge and practice. Exposure was defined at a listening threshold of 50% or more of the cumulative duration of the calls mapped to the outcome. For example, 7 Kilkari calls (week 20, 24, 29, 34, 65, 66, 68) included content on reversible modern contraceptive methods, for a total of 10.5 min of audio content on this topic (online supplemental table 1). To be ‘exposed’ subscribers would have had to listen to 50% or more of the cumulative message total of these messages, for example, 5 min or more of the contents. bmjgh-2022-008838supp001.pdf Primary analyses of outcomes were done with modified intention-to-treat (ITT) analyses at the individual level so that outcomes were analysed regardless of the degree of listening to Kilkari. The term modified refers to the requirement of being able to determine the outcomes beyond 1 year of the postpartum period (ie, only those who were reached by the endline survey were included in the analysis). Relative risk (RR) for primary and secondary outcomes was calculated for the intervention group compared with the control group using log-binomial regression with general estimating equations to account for correlation within villages. To assess the impact of exposure on outcomes, compliance adjusted treatment effects were additionally generated using the instrumental variable (IV) methodology. The study arm based on randomisation was considered the instrument variable. As noted above, exposure was defined as listening to at least 50% of the cumulative duration of the calls mapped to the outcome. RR estimates were calculated using log binomial models for each outcome with adjustment for clustering at the village level. The IV analysis was carried out using the IVREG package in R.21 Frequencies and proportions are used to describe differences across intervention and control groups (ITT) as well as those exposed to 50% or more of Kilkari content. The Bill and Melinda Gates Foundation had no role in the study design; collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. All authors confirm that they had full access to all the data in the study and accept responsibility for the publication submitted. Patients were first engaged on identification in their households as part of a household listing carried out in mid/late 2018. Those meeting eligibility criteria were interviewed as part of the baseline survey, and ultimately randomised to the intervention and control arms. Prior to the administration of the baseline, a small number of patients were involved in the refinement of survey tools through qualitative interviews, including cognitive interviews, which were carried out to optimise survey questions, including the language and translation used. Finalised tools were administered to patients at baseline and endline, and for a subsample of the study population, additional interviews carried out over the phone and via qualitative interviews between the baseline and endline surveys. Unfortunately because of COVID-19 and associated travel restrictions, patients could not be involved in the dissemination of study findings. However, public dissemination of the results has occurred through a number of global level presentations.
N/A
