Background (brief):

1. Burden: Maternal and neonatal mortality continue to be a serious public health problem in Bangladesh. Facility-based healthcare services reduce the probability of preventable causes of maternal and neonatal death. Community clinics (CCs) are located closer to the community than other healthcare facilities are. However, only 2.8% of all CCs are ready to provide antenatal care (ANC) and 2.4% of these facilities are ready to provide delivery services. Despite the wide acceptance of midwifery care for ANC and delivery services, they are posted only up to the Upazila Health Complexes and at selected Union Health and Family Welfare Centers.
2. Knowledge gap: There is a dearth of evidence on whether the utilisation of ANC, normal vaginal delivery, and postnatal care (PNC) services can be increased if midwives are assigned to CCs to provide these services.
3. Relevance: Midwives were trained to provide antenatal and postnatal care, conduct normal deliveries, and manage CC-specific pregnancy-related complications. Midwives are posted up to union-level facilities, and the Government of Bangladesh has a strategic plan to enable selective community clinics for normal delivery services in addition to antenatal and postnatal care.

Hypothesis (if any): N/A

Objectives: This study aims to assess the feasibility, adoption, acceptability, appropriateness, fidelity, coverage, and possible barriers and enablers of introducing Midwifery-embodied Community Clinic (MCC) model in selected community clinics of Bangladesh.

Methods: We will implement a Midwifery-embodied Community Clinic (MCC) model in two CCs of the Baliakandi sub-district of Rajbari district. Trained midwives will provide ANC, select low-risk pregnant women, conduct their normal deliveries, and provide PNC services in CCs, while the CHCP will provide essential newborn care during NVD. High-risk mothers will be referred to the upper level of healthcare facilities for managment. In addition, counselling services will be provided to mothers by the Health Assistants and Family Welfare Assistants during their domiciliary visits. Transport services using the ambulances of the Upazila Health Complex will be integrated into the model to transport patients to the CCs and upper-level healthcare facilities in case of emergencies.

Outcome measures/variables: The outcome measures will be feasibility, adoption, acceptability, appropriateness, fidelity, coverage, and possible barriers and enablers of the MCC model.

Study design and setting:

This is a feasibility study in which a Midwifery-embodied Community Clinic (MCC) model will be introduced in two Community Clinics (CCs) located in the Baliakandi sub-district of the Rajbari District of the Dhaka division of Bangladesh. Rajbari district lies between 22º40' and 23º50' North latitudes and between 89º19' and 90º40' East longitudes (19). The total area of the district is 1092.28 sq. km and the Baliakandi sub-district is 229 sq. km (19). It has 22 Community Clinics in 7 unions covering 258 villages (20, 21). The total population of the Baliakandi sub-district is 207,086, of which 103,411 are females (19). The proportion of home deliveries in the Baliakandi sub-district is around 21% (22). The model will be implemented in the Delua and Sonapur Community Clinics of Narua and Nawabpur unions of Baliakandi sub-district which are located approximately 12km away from the Baliakandi Upazila Health Complex (Nearest CEmONC facility).

Steps of the study

1. Pre-implementation period 1.1 Stakeholder engagement We will engage potential stakeholders from the beginning of the study. These stakeholders will include personnel from the government, development partners, and technical bodies.

   1.2 Baseline data collection Before the implementation of the MCC model, we will conduct household survey and health facility assessment. We will adopt both qualitative and quantitative approaches to collect data.

   1.2.1 Household survey We will enrol all women with a history of childbirth in the last 12 months of the survey period and will agree to participate in the household survey. We will conduct key informant interviews (KII) with the the program managers, technical experts, and local level government personnel. We will also conduct in-depth interviews with the service providers and care receivers.

   1.2.2 Health Facility Assessment We will conduct a health facility assessment at the baseline of the study to identify gaps in the necessary medicines, equipment, and logistics so that the necessary resource allocation can be ensured and the facilities become ready to provide ANC, normal delivery, and PNC services.

   1.3 Capacity development of the service providers After ensuring key stakeholder engagement from the beginning of the study, we plan to implement the MCC model. The OGSB will train the study midwives on ANC, NVD, and PNC and the CHCPs on Essential Newborn Care (ENC). Healthcare providers will receive specific hands-on training following national guidelines suitable for CCs (24). HAs and FWAs will be briefed on the availability of maternal healthcare services and essential newborn care in community clinics.

2. Implementation period 2.1 Facility-level interventions We will prepare CCs and ensure the supply and maintenance of the necessary logistics, equipment, and medicine during the implementation period, so that these facilities can provide antenatal and postnatal care and conduct normal vaginal deliveries. We will train midwives, and CHCPs on ANC, NVD, ENC, PNC, and CC-specific complication management. The HAs and FWAs will be briefed on their roles and responsibilities regarding the availability of midwifery-embodied maternal and newborn healthcare services in the community clinics, and helping the mothers and their families on the BPCR plan.

The Maternal Health Standard Operating Procedure (SOP) recommends tier-specific management of mothers during the antepartum, intrapartum, and postpartum periods, while the Comprehensive Newborn Care Package (CNCP) recommends tier-specific newborn care and complication management. Healthcare providers will follow recommendations specific to community clinics to provide ANC, delivery, ENC, and PNC services. Midwives will identify high-risk and low-risk mothers, refer high-risk women to the Upazila Health Complex for further management and provide scheduled antenatal and postnatal care to low-risk women.

The ANC and PNC services will be provided during routine working hours. The delivery service will be available on a 24/7 on-call basis. Midwives will conduct deliveries of low-risk mothers who want to have their deliveries in CCs. Otherwise, they will be encouraged to choose from any of the higher levels of healthcare facilities equipped with the resources to manage obstetric cases properly. During normal vaginal delivery, midwives will conduct the delivery, and closely observe the mothers and their newborns for at least six hours postpartum. If any complications develop, they will provide initial management, stabilize the mothers and refer them to higher-level healthcare facilities. The CHCPs will be responsible for providing essential newborn care to newborns following delivery, and if any complications develop, they will provide appropriate resuscitation to the newborn and referral.

Service data (ANC, NVD, ENC, and PNC) will be recorded in government-approved service registers (ANC, EmONC, and PNC registers). This information will be collated monthly in the monthly reporting form.

Community-level interventions The HAs and FWAs will inform pregnant women and their families about the availability of the ANC, NVD, ENC, and PNC services at their nearest CCs and strengthen birth preparedness and complication management (BPCR) plans. Thus, they will have a BPCR plan that will help them make prompt decisions about where to go, from whom to seek care, and how to reach the target facilities. This combined counselling of midwives during ANC, and HAs and FWAs during domiciliary services will address the first delay, the delay in decision-making.

2.3 Integration of transport system The ambulance service available in the Baliakandi Upazila Health Complex will be utilised upon receiving authorisation from the UH&FPO. Local transport will also be integrated and a list will be prepared which will be given to pregnant women. Pregnant women will call the drivers if they develop any sign of labour or any danger sign. The available drivers will transport the pregnant woman to their preferred healthcare facility for delivery and the complicated cases (mother or newborn) to the referral facility following CC-specific stabilisation. The place of delivery will depend on the preferences of the pregnant women and their family members. This mechanism will reduce the second delay, the delay of transport.

At present, the Baliakandi Upazila Health Complex has one functional ambulance for patient services. We will make a list of key contact person(s) and transport drivers, including the ambulance drivers of Baliakandi Upazila Health Complex (UHCs), nearby health facilities, and local transport drivers, and populate them among community people through the patients seeking services at the CCs and through the community health workers and data collectors while conducting household surveys. Through this study, we will invest in strengthening care coordination between CCs and high-level facilities as a part of the Birth Preparedness and Complication Readiness (BPCR) plan.

2.4 Local government-level interventions A coordination committee will be formed with the UH&FPO, the focal person of the CG-CSG members, CHCPS, midwives, HAs, FWAs, drivers of the ambulance, and owners of the local transport. icddr,b will encourage the organisation of monthly coordination committee meetings where the service registers will be reviewed, service delays will be accounted for, barriers and challenges of the implementation of different activities of the model will be identified and addressed, and experiences will be discussed. This monthly discussion will create accountability among midwives, CHCPs, HAs, FWAs, and transport drivers to the focal person of the CG-CSG members at the community level and the UHFPO at the facility level for task achievement and timely response. The pre-assignment of services to specific service providers and regular coordination committee meetings will address the third delay; the delay in the initiation of the service in the healthcare facility.

2.5 Midline data collection During the midline data collection period, we will conduct semi-structured observations of the activities of midwives and unstructured observation of the CHCPs, HAs, and FWAs. We will also conduct in-depth interviews to collect the opinion of care receivers and service providers. We will identify gaps and operational challenges in introducing the model in the community clinics and take the necessary steps to solve the problems.

Post-implementation period 3.1 Endline survey We will conduct an end-line survey using a mixed-method approach to assess the feasibility of introducing the MCC model into the routine health system of Bangladesh. In-depth Interviews (IDI) and key informant interviews (KII) will be conducted with respondents to understand the adoption, acceptability, appropriateness, fidelity, coverage, barriers and enablers of the MCC model.

3.2 Dissemination of the findings We plan to organize a dissemination workshop with policymakers and relevant stakeholders to communicate the study findings and translate them into policies and practices. This will be an in-person workshop in which participants will be allowed to discuss the findings. In addition to this dissemination, we plan to develop scientific manuscript(s) and publish them in peer-reviewed journal(s) to translate this knowledge into other similar settings. We will also develop a policy brief.

Data collection Health facility assessment We will assess the availability, readiness, and functionality of community clinics at the baseline of the study to identify the gaps and ensure necessary resource allocation for ANC, PNC, and NVD service provision.

Household survey Household surveys will be conducted to assess the coverage of maternal health and essential newborn healthcare services among the primary stakeholders in the catchment areas of the selected community clinics. Information on care-seeking behaviour, antepartum, intrapartum, and postpartum service utilisation of the consenting respondents will be obtained to get baseline and end-line estimates. Relevant data collection instruments previously used in similar evaluations and the Bangladesh Demographic and Health Surveys (BDHS) will be further adapted for the planned data collection. Trained data collectors will collect data.

Service register data extraction We will extract data from the service registers used to capture ANC, NVD, ENC, PNC, and routine work of the healthcare providers of the community clinics to evaluate the fidelity of the model.

Qualitative interviews We will conduct in-depth interviews with the service receivers, and care providers, and key-informant interviews with the policymakers, technical experts, UH&FPO, and focal persons of the Community Group and Community Support Group members to understand the feasibility, adoption, appropriateness, barriers and enablers of the implementation of the model. Data collectors trained on qualitative research will conduct the interviews.

Data quality checks, storage and access During data collection, the data records will be checked by members of the management team. All individual identifiers will be removed from the data to be anonymised. The data will be stored on a Secure Server, which is automatically backed up to an off-site location after collection. The data stored on this secure server will be encrypted so that they can only be accessed by those with the correct encryption key, which will only be available to members of the immediate research team working on the analysis of the data. Fully anonymised data will be shared in line with the open-access requirements of peer-reviewed journals.

Sample size:

Household survey We will follow the convenient sampling technique to recruit the study participants. We will include all recently delivered women in the catchment areas of the selected community clinics who have a history of birth outcome in the past 12 months of the survey period. A sample size of 60 to 120 is required to adjust the precision of the sample size of this pilot/ feasibility study (25). The annual birth rate in Bangladesh is 20/1000 live births in rural areas and each CC coverage is around 6000-8000 population (13). In addition, according to recent estimates from the Baliakandi Health and Demographic Surveillance data, 400-450 women, on average, became pregnant per month in Baliakandi. Distributing this across seven unions, around 57-65 women got pregnant per union per month in Baliakandi. We anticipate that we will be able to enrol 70 participants.

Qualitative interviews We will conduct in-depth interviews with 8±2 pregnant women about their ANC experiences, 8±2 recently delivered women about their delivery experiences, and 8±2 women about their PNC experiences who will receive the respective services from the selected community clinics. We will also interview 8±2 key informant, and 8±2 service providers. We will employ an emergent sampling strategy. An emergent sampling strategy will ensure that we can respond to the unexpected and provide the flexibility to sample additional participants where necessary to fulfil the research objectives.

Data analysis Quantitative analysis All data will be assigned an appropriate identifier to maintain their anonymity. Quantitative data will be stored in Microsoft Excel, and STATA 15.0 will be used for performing the analysis. We will use descriptive statistics as appropriate (frequency, mean, standard deviation, etc.). We will perform multivariable logistic analysis to assess the changes in their service utilization (ANC, normal vaginal delivery, and PNC services) among the study participants of the catchment population of the community clinics.

Qualitative analysis A thematic analysis approach will be adopted to analyze the qualitative data. All the interviews and field notes will be recorded and transcribed verbatim. Transcription will be performed by the respective interviewers immediately after the interviews. Upon preparing the transcriptions of interviews, all of the data will be coded by the researchers based on the a-priori codes (e.g. feasibility, adoption, acceptability, appropriateness, fidelity, and barriers and enablers). To establish inter-coder reliability, a minimal set of interviews will be coded by different researchers to ensure that the codes and coding process are well understood, and that everyone is on the same page in terms of understanding the concepts. Upon establishing inter-coder reliability, a master code list will be generated, based on which the rest of the transcripts will be coded. The researcher will also search for inductive codes that can explain these themes. For all inductive codes, formal discussions with all coders and co-investigators will be organised to ensure uniformity in the analysis. The next step in the analysis will be to generate summaries of codes based on similarities and contrasts and analyse them using a thematic framework. The theoretical orientation would be phenomenological, in which the phenomena of interest would be the topics being documented. Data will be analysed using NVivo software.

Inclusion criteria the study participants are as follows:

1. Pregnant women or have a history of childbirth within one year of the survey period
2. The participants must be living in the catchment areas of the Delua and Sonapur Community Clinics of Narua and Nawabpur unions of the Baliakandi sub-district of Rajbari district
3. Healthcare providers of the selected community clinics (midwife, CHCP, HA, and FWA)
4. UH&FPO, and focal person of the community group and community support group
5. Policymakers from Maternal Health Programme, MNC&AH, DGHS and Maternal Health, DGFP
6. Technical experts of the Obstetrical and Gynaecological Society of Bangladesh (OGSB)

Exclusion criteria of study participants are as follows:

1. Respondents who will not give consent to participate in the study