The impact of strengthened ward development committees on utilisation of reproductive maternal and child health services in Bayelsa State, Nigeria | BMC Public Health

Study design

This quasi-experimental study was used to assess the impact of strengthened Ward Development Committees (WDCs) on the utilisation of reproductive, maternal and child health services in Kolokuma/Opokuma LGA, Bayelsa State, Nigeria- a UNICEF priority LGA receiving WDC strengthening at the time.

Study setting

The study was conducted in Bayelsa state in the South-South geopolitical zone of Nigeria. The state is located at the core of the Niger Delta and has its capital as Yenagoa. It has a total area of 10,773 km² and a population density of 270.2 inhabitants per square kilometre [29]. The estimated total population for 2022 (from the 2006 population census) was 2,769,284 with women of childbearing age (WCBA) population of 609,843 [31]. The state is made up of eight LGAs and 105 geopolitical wards. About 70% of the eight LGAs are only accessible via waterways. The Kolokuma/Opokuma LGA has 11 wards and 94 communities (see Fig. 1). The LGA has a projected total population of 118,667 people with 26,108 Women of Childbearing Age (WCBA) and 23,735 children under the age of 5 years [32, 33]. The LGA is mostly landed with a lesser part being riverine. It is primarily an agricultural area.

Map of Nigeria and Bayelsa state showing the Local Government Areas and the focal LGA- Kolokuma/Opokuma

Study population

Two wards- Kaiama and Kaiama-Oloibiri were purposively selected for the intervention based on the states assessment and the poor RMNCH outcomes. They have about 18 communities with an estimated 4700 WCBA. The study population consisted of caregivers of children between the ages of 0 and 59 months in the intervention wards.

Inclusion criteria

Only caregivers who have lived in the selected wards for at least one year were eligible to participate in the study. Those who did not consent were excluded. Caregivers who had not lived with the child for at least 75% of the child’s lifetime as well as caregivers who were mentally impaired were excluded as they were deemed not to either have sufficient information about the child or give intelligibly coherent responses to the questions asked.

Sample size determination

The sample size for this study was 432 participants. This was determined using the Fleiss formula for sample size calculation in intervention studies given as [34]:

where:

N = minimum sample size

Zα/2 = 1.96 (from Z table) at type 1 error of 5%

Zβ = 0.84 (from Z table) at 80% power

p1—p2 = difference in the proportion of events (health service awareness). Health service awareness is assumed to be 50% at baseline. The intervention is expected to increase health service awareness to at least 60%.

P = pooled prevalence = [prevalence in study population (p1) + prevalence on control population (p2)] / 2

P = (0.50 + 0.60) / 2 = 0.55

N = 388

Adjustment for 10% Attrition Rate:

N = n/ (100-r %)

Where:

n = 388

r = anticipated non-response rate = 10%

N = 388/0.9

N » 432

Sampling technique

The two wards- Kaiama and Kaiama-Oloibiri were fairly equal in size and had almost the same population size therefore equal allocation was employed to ensure that respondents were equally spread across the wards. Though the estimated sample size was 432, only 300 participants met the eligibility criteria and participated in the study. In the second stage, households (HHs) were clustered within the communities with about 50 HHs per community. There were 20 – 25 HHs with eligible caregivers per community. All the eligible caregivers with children 0 – 59 months in the communities were studied.

Data was collected in three phases of the intervention spanning a period of 6 months- September 2021 to February 2022. The phases comprised pre-intervention, intervention, and post-intervention. During the course of the data collection, some participants were dropped and some were added. The interventions were as follows:

Pre-intervention

This involved community entry and advocacy followed by the identification of the WDC members in the intervention ward. They were briefed on their roles as WDCs, the aim of the study and their terms of reference within the study period. One (1) WDC was used in this study. A pre-tested interviewer-administered questionnaire to generate baseline data on the demand for maternal and child health care services was administered to the care givers.

Intervention in content

The intervention was centered on strengthening the existing WDC in the intervention wards: Kaiama and Kaiama-Oloibiri wards. The intervention lasted for a period of 6 months. This cuts across five major thrusts-

1. 1.

   Technical support for meetings: The state designated a minimum of three state supervisors and a maximum of 5 supervisors to attend and support the monthly WDCs meetings. The Terms of Reference (TOR) of the supervisors were to guide discussions around immunisation service delivery and routine delivery of Reproductive, maternal, newborn, child, and adolescent health + nutrition (RMNCH + N) services. The supervisors also guided the documentation of action points from the meetings and ensured tracking of the action points and progress outcomes from the meetings.

2. 2.

   Provision of Funding: The funding partner UNICEF through an already existing funding framework in the state provided extra funding support for transportation for the regular routine WDC meetings to make them functional in the ward.

3. 3.

   Provision of RMNCH + N Commodities: The state through its MOH procured and provided commodities (vaccines, zinc/ORS, deworming, vitamin A, family planning) to support the provision of RMNCH + N services in the identified PHCs in the wards to enhance demand creation and improve utilisation.

4. 4.

   Outreaches: HCWs were supported in microplanning and funds were made available to increase the number and frequency of outreach conducted in the wards, thus, there was a rise in the frequency of outreaches throughout the wards various communities. Additionally, the number of outreach locations grew as more settlements with distinctive topographies were visited to provide maternal and child health services including children under the age of five and increasing access to more RMNCH + N services inside the wards.

5. 5.

   Human Resource for Health: Health Care Workers (Community Health Extension Workers- CHEWS) were redeployed across facilities in the intervention ward to improve service provision in the facilities and to conduct outreaches. Supportive supervision was also strengthened to ensure continuous capacity-building support for HCWs during service delivery, and mentorship of women of childbearing age who accessed the services.They collaborated closely with the WDC to ensure alignment and effectiveness in their activities.

Mid- intervention

Three months (halfway) into the intervention, the same research instrument- the pretested interviewer-administered questionnaire that was used in getting baseline data earlier in the study was administered to collect midline data.

Post-intervention

This phase entailed data collection at the end of the study- after 6 months of intervention- with the same pre-tested interviewer-administered questionnaire administered at the pre- and mid-intervention stages to get endline data.

Data collection instrument

A structured pre-tested questionnaire was used to collect data at baseline, midline and endline of the intervention. The questionnaire was adapted from a similar study [27]. It was pre-tested with similar respondents in a ward in a different LGA (a ward in Yenagoa LGA).

Funds had already been disbursed, and the intervention was underway before data collection. Ten trained research assistants who were fluent in both English and Izon languages administered the questionnaires to the respondents. They used an open-data kit (ODK) tool after obtaining informed consent in each data collection phase. The baseline data was collected between 31st August 2021 and 15th September 2021, after the disbursement of UNICEF funds and the initiation of WDC strengthening activities. At this stage, the study aimed to assess whether the WDCs were actively fulfilling their roles. The midline data was collected three months after the baseline between 6 and 12th December 2021 to evaluate any changes following the intervention and stakeholder feedback while the end-line data was collected after the intervention between 11 to 14th April 2022.

Analytical approach

The data were extracted from the ODK collection tool into Microsoft Excel version 2019 and Statistical Package for Social Sciences (SPSS) version 25 was used for the analysis. The characteristics of all the respondents at baseline, midline and endline and the prevalence of the outcome variables were described using percentage. An assessment of the significant difference in the characteristics of the respondents at these periods was conducted with a test of association for categorical variables using χ2. A total of 300 respondents were enrolled at baseline, midline and endline. To assess whether there were significant changes in the outcomes across the three time points—baseline, midline, and endline, we used the Cochrane Q test and the Friedman test, depending on the nature of the variables. The Cochrane Q test was applied to binary outcome variables while for outcome variables with more than two ordinal data the Friedman test was employed. A series of logistic regression was used to examine the influence of socio-demographic factors on maternal health service utilisation, specifically focusing demographic factors on maternal health service utilisation, specifically focusing on family planning (FP) uptake, antenatal care (ANC) attendance, postnatal care (PNC), and place of delivery. For models with convergence issues or sparse categories (especially in ANC and PNC), a simplified model with key predictors (age group, education, and employment) was used to stabilise estimation. The logistics regression analyses were conducted in Python (statsmodels v0.14). The results were reported as odds ratios (ORs) with 95% confidence intervals (CIs) and corresponding p-values. Statistical significance was determined at p < 0.05.

Ethical considerations

Ethical approval was obtained from the Bayelsa State Health Research Ethics Committee, with the approval number BSHREC/Vol. i/21/09/01. Beyond this, each participant gave consent before the data collection exercise. Eligible participants were given a detailed written consent form containing information about the study. The consent forms explained the study process and their right to either decline or participate. Participants, who agreed to participate and gave their consent, signed a consent form.