It takes a village: A Pilot Study Encouraging Women to Use Maternal Care through Community Involvement

Last registered on November 08, 2022

Pre-Trial

Trial Information

General Information

Title
It takes a village: A Pilot Study Encouraging Women to Use Maternal Care through Community Involvement
RCT ID
AEARCTR-0010360
Initial registration date
November 04, 2022

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
November 08, 2022, 2:56 PM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Primary Investigator

Affiliation
University of California, Berkeley

Other Primary Investigator(s)

PI Affiliation
University of Ghana Business School

Additional Trial Information

Status
On going
Start date
2021-08-01
End date
2022-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
The World Health Organization (WHO) recommends a minimum of eight antenatal care (ANC) contacts, initiating ANC during first trimester, giving birth in facilities, and postnatal care (PNC) within 24 hours of birth to reduce maternal and perinatal mortality. These recommendations specifically refer to ANC “contacts” instead of “visits” to place the emphasis on continuous interactions between providers and pregnant women that does not necessarily imply a physical visit to health facility. In Ghana, although attendance of at least one ANC visit is nearly universal, attendance of the recommended number of visits remains below target, women typically start ANC late in pregnancy, and many still do not deliver in a health facility. Though it is the woman who bears the pregnancy, in Ghana and many other patriarchal societies, decisions to seek health services are often made by other stakeholders. For instance, it is often the mother-in-law who makes decisions about the merits of ANC and the husband who makes the financial decision. Many programs ignore this reality. We propose to test an intervention that focuses not only on encouraging pregnant women but also their support networks about the importance of seeking modern maternal care, taking into account deeply entrenched cultural beliefs.
External Link(s)

Registration Citation

Citation
Abuosi, Aaron and Aleksandra Jakubowski. 2022. "It takes a village: A Pilot Study Encouraging Women to Use Maternal Care through Community Involvement." AEA RCT Registry. November 08. https://doi.org/10.1257/rct.10360-1.0
Experimental Details

Interventions

Intervention(s)
The study involved two cross-randomized interventions: 1) community education at durbars, 2) enhanced ANC model during which monthly phone calls and one home visit were made to reinforce key messages, monitor pregnancy, and prepare birth plans together with husbands and mothers-in-law.
Durbars are village meetings where communities discuss important village matters. Since durbars take place on a usual basis, nurses coordinated with the local chiefs and elders to pre-arrange durbars to deliver the interventions. The chiefs and elders in turn employed their mechanisms of organizing durbars to invite the community members to participate. The key quality of durbars that we leveraged in our intervention is that these are ongoing meetings that happen on a regular basis and that involve village chiefs and other local authority figures who wield significant influence in the communities, and whose endorsement of the messages about the traditional beliefs about seeking ANC early and male involvement might be taken seriously. We sought the endorsement of our messages from the chiefs and elders prior to the durbars. Participation in the durbars was completely voluntary and was not monitored by research staff - community members came to these meetings when they wished, and they were free to leave whenever they deemed appropriate. We tested the impact of providing education about the importance of starting ANC in the first trimester by comparing whether the month of pregnancy when women first presented for ANC changed in the villages that held durbars vs. those that did not.
The durbars were held prior to any subject recruitment - we collected no identifiable or individual-level information about who attends the durbars. We obtained very general information about durbar attendance, but this information was tracked at the aggregate level using publicly observable information about the meetings (e.g. number of people, gender breakdown). This general attendance information helped us address the question of whether well attended & gender balanced durbars had larger impact.
After durbars were held, the study staff began enrolling women in the five study facilities. Women who agreed to participate were provided informed consent, were randomized to routine care or Enhanced ANC intervention which consisted of monthly phone calls and one home visit
Monthly calls: it was a follow-up on the progress of the pregnancy, encouraged women to come to the facility for in-person ANC check, and developed a birth preparedness plan. The innovation of this treatment is that instead of receiving pre-recorded messages (i.e. the CHPS program) women were connected with a midwife who can go over any possible complications and encouraged them to go in for a check-up. This activity was conducted by the nurses/midwives. Research staff were not involved in this activity.
Home visit: women randomized to this arm received one household visit from a community health officer (CHO) to develop a concrete birth preparedness plan with the pregnant woman and her household members. The health provider arranged a time to visit the household when either the woman’s husband or mother-in-law (ideally both) were at home, ensuring that the families developed a plan together. The health provider prompted the households to make specific choices for the plan, such as: identifying danger signs during pregnancy and labor, choosing the health facility where the woman can deliver, determining how long it takes to travel there, making specific plan on how they will travel there, establishing how much money was required for the birthing process, and planning how to set aside money for this expense. This activity was conducted by the nurses/midwives. Research staff were not involved in this activity.
Randomization to the durbar treatment was at the village-level; randomization to the phone call and home visit treatments was at the woman-level. The cross-randomization was to enable us to tease out the specific aspects of these intervention on outcomes of interest. Women who were randomized to the pure control arm received standard of care: encouraged to return for at least eight ANC contacts, to deliver in a health facility, and have a postnatal check each time she has contact with health providers.
Intervention Start Date
2021-08-12
Intervention End Date
2022-01-13

Primary Outcomes

Primary Outcomes (end points)
• Number of pregnancy months at first ANC visit: coded as the number of pregnancy months disclosed by women at 1st ANC visit (numeric). This number will be verified by calculating the length of time between First ANC visit and Expected Delivery Date. In case of discrepancies, we will use the number of months calculated based on EDD
• First ANC visit in first trimester - an indicator set to 1 if the woman is in her first trimester at first ANC visit, and 0 elsewhere (binary).
• Number of days between knowing about ANC and first ANC visit - number of days since the woman has found out about the pregnancy at her first ANC visit (numeric).
• Initiating ANC according to the woman’s preferences - an indicator set to 1 if woman started ANC no later than when she wanted to start (binary)
• Number of ANC visits: coded as the number of ANC visits woman makes during pregnancy (numeric)
• Quality of ANC: coded as the number of services women received during the ANC visits
• Knowing danger signs of pregnancy: Woman knows the main danger signs of pregnancy
• Birth plan made: woman developed a birth plan during pregnancy
• Birth plan used: woman used her birth plan for labor/delivery
• Support during pregnancy and birth:
o Woman was accompanied to ANC visit
o Husband assisted with making the birth plan
o Woman felt supported by husband during pregnancy
o Woman felt supported by husband during labor/delivery
o Husband accompanied to health facility for labor/delivery
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
The intervention’s objective was to provide more support to women during pregnancy and birth. An unintended consequence of this intervention may be that getting people in their support networks more involved in the pregnancy may ultimately take away some of the women’s empowerment and agency. We will test whether women score lower on women empowerment measures by intervention assignment.
We will ensure that the interventions were delivered as intended by testing whether women in control arm disclosed receiving the services.
We will also test who were the primary decision makers during pregnancy and childbirth, and the attitudes of study participants towards men and mothers-in-law being more involved in pregnancy, ANC, birth.
Finally, we will test whether pregnancy outcomes (baby weight, complications during birth, etc) differed by intervention.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Our study is designed to test whether education at durbars (or community meetings) to communicate the importance of timely and frequent prenatal care, combined with phone messages to reinforce the education, and household visits to develop birth preparedness plans with pregnant women, their partners, and mothers-in-law increases the use of maternal health services and improves health outcomes in northern Ghana. The general outline of our interventions and data collection procedures is as follows:
1. Random assignment of which communities receive ANC messages at durbars
2. Intervention 1: Education at durbars
3. Recruit and enroll pregnant women in study at 1st antenatal visit
4. Randomize individuals to receive phone calls and home visits
5. Conduct baseline survey (in person)
6. Intervention 2: Phone calls (monthly) and home visit (one time)
7. Conduct endline survey (in person unless women express preference for phone interview)
Experimental Design Details
Randomization Method
Randomization to the durbar treatment is at the village-level. Randomization was done in Stata 16 by randomly assigning the village to conduct a durbar or serve as control. The probability of a village being selected for a durbar is 50%.

Randomization to the phone call and home visit treatments is at the woman-level. We generated random numbers ahead of time that were assigned either to receive phone calls and home visits or to be controls (40% chance of being assigned to the intervention). These numbers were written down on envelopes, with the randomly generated number on the outside of the envelope and the treatment assignment included on the inside of the sealed envelope. After a woman agrees to be a part of the study and informed consent is obtained, a field officer will pick a random envelope and record the woman’s study ID (randomly-generated number) and treatment assignment. The information will later be verified in an electronic record system in which random numbers and treatment assignments will be tracked.
Randomization Unit
Durbar intervention was randomized at community level. Enhanced ANC intervention was randomized at individual level.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
We had 32 villages, half of which were randomized to intervention.
Sample size: planned number of observations
600 women.
Sample size (or number of clusters) by treatment arms
300 women in durbar and 300 in control communities; 240 in intervention and 360 in control for Enhanced ANC.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We set alpha at 0.05, power at 0.80, clusters at 32, and rho at 0.01. With these parameters, and the expected sample of 300-350 women per treatment arm (depending on attrition), we can expected to detect a difference of 11-12 percentage point difference in outcome.
IRB

Institutional Review Boards (IRBs)

IRB Name
Navrongo Health Research Centre Institutional Review Board (NHRCIRB)
IRB Approval Date
2021-07-16
IRB Approval Number
NHRCIRB420
IRB Name
The Committee for Protection of Human Subjects (CPHS) at UC Berkeley
IRB Approval Date
2021-06-08
IRB Approval Number
2021-04-14254
Analysis Plan

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials