Introduction Health systems are often weakened by public health emergencies that make it harder to access health services. We aimed to assess maternal, newborn and child health (MNCH) service utilisation during the first 6 months of the COVID-19 pandemic compared with prior to the pandemic. Methods We conducted a mixed study design in eight health facilities that are part of the Birhan field site in Amhara, Ethiopia and compared the trend of service utilisation in the first 6 months of COVID-19 with the corresponding time and data points of the preceding year. Result New family planning visits (43.2 to 28.5/month, p=0.014) and sick under 5 child visits (225.0 to 139.8/month, p=0.007) declined over the first 6 months of the pandemic compared with the same period in the preceding year. Antenatal (208.9 to 181.7/month, p=0.433) and postnatal care (26.6 to 19.8/month, p=0.155) visits, facility delivery rates (90.7 to 84.2/month, p=0.776), and family planning visits (313.3 to 273.4/month, p=0.415) declined, although this did not reach statistical significance. Routine immunisation visits (37.0 to 36.8/month, p=0.982) for children were maintained. Interviews with healthcare providers and clients highlighted several barriers to service utilisation during COVID-19, including fear of disease transmission, economic hardship, and transport service disruptions and restrictions. Enablers of service utilisation included communities’ decreased fear of COVID-19 and awareness-raising activities. Conclusion We observed a decline in essential MNCH services particularly in sick children and new family planning visits. To improve the resiliency of fragile health systems, resources are needed to continuously monitor service utilisation and clients’ evolving concerns during public health emergencies.
We conducted the study in eight health facilities in the Birhan North Shewa Zone, Amhara Region, Ethiopia. The field site was established in June 2018. The Birhan field site is a community-based continuous follow-up study of health and demographic conditions that provides up-to-date information on the catchment population and establishes a population frame to nest studies. We selected all catchment health facilities for this study, including five health centres, two primary hospitals (one public and one private) and one referral hospital. These facilities provide essential MNCH services for both the rural majority population and urban population within the field site catchment and non-catchment areas. The health centres provide antenatal care (ANC), postnatal care (PNC), delivery, abortion, routine immunisation (RI), integrated management of neonatal and childhood illness (IMNCI), and family planning (FP). Each health centre also has a minimum of five service extension health posts, mainly for FP and RI in each kebele (the lowest administration unit in Ethiopia), and each health post sends monthly activity reports to health centres. Two public hospitals (one primary and one referral) and one private general hospital also provide the aforementioned essential MNCH services, except for RI, which is given mainly in health centres and catchment health posts. Mixed phenomenological qualitative and facility-based cross-sectional study designs were employed. For the quantitative part of the study, a facility-based cross-sectional survey was conducted to assess the impact of COVID-19 on essential MNCH service provision or utilisation and provider-side barriers to service provision and utilisation in the Birhan field site catchment health facilities. We interviewed 91 MNCH HCPs (doctors, nurses, midwives and clinical officers available at the time of data collection) with uniformly structured questionnaires about their perception of client flow and possible barriers for respective sections. Twelve out of 91 HCPs were working in two MNCH departments and were interviewed twice. In addition to this, we extracted retrospective, healthcare utilisation time-series data from each facility using monthly facility reports and medical registers. Retrospective facilities’ service statistics were collected over an 18-month period from March 2019 to August 2020 using Computer Assisted Field Editing. We extracted data from the uniformly structured questionnaires, entered it into the Open Data Kit (ODK), and collected and uploaded the data to the ODK aggregate. The monthly facility reports and medical registers data were collected separately. The health centres’ monthly reports include services given in the health posts that are extension sites for the health centre, but the facility registers are exclusively for services given in the health centres. In addition to the cross-sectional study, we implemented a phenomenological qualitative design using in-depth interviews to assess client and provider-side barriers and enablers to service provision/utilisation in the Birhan field site catchment health facilities. We sampled and conducted in-depth interviews until we reached theoretical saturation. For this section of the study, we interviewed 10 facility or department heads, and 9 mothers (delivered at home or facility and had ANC or missed ANC follow-up). An interview guide with open-ended questions was translated from English to Amharic and was used to elicit the qualitative information from informants. We conducted in-person interviews with facility or department heads, women who visited facilities during COVID-19, and women who delivered at the facilities and phone interviews with women who missed an ANC follow-up or delivered at home. With the permission of the respondents, we recorded all interviews and transcribed all records into English for further analysis. To ensure the safety of the data collectors and participants, data collectors wore masks and practised physical distancing during training and data collection from 2 to 20 November 2020. The extracted data were exported to Stata V.17.0 for analysis and the average MNCH service utilisation was calculated each month to quantify the changes pre-COVID-19 (March–August 2019) and during the COVID-19 (March–August 2020) pandemic. To control for potential seasonal fluctuations in service utilisation, March–August 2019 and March–August 2020 were considered pre-COVID-19 and COVID-19 periods, respectively. Across all health facilities, we had 48 paired months of observations (6 months for each of 8 facilities) for all essential MNCH variables except for RI, which was only administered at the five health centres (and corresponding extension health posts), resulting in 30 paired months. Errors were found in some cases where medical records were misplaced and data for some months were missing or partially filled. To avoid the effect of missing and partially filled values, analogous months’ data from the same facility were excluded from the data analysis. Finally, we compared visits for each MNCH service in the pre-COVID-19 and COVID-19 period using a two-tailed independent sample t-test. We repeated the analysis for the initial 2 months (March–April 2020) of the pandemic and the analogous period (March–April 2019) to examine changes in service utilisation at the onset of the COVID-19 pandemic. We used a significance level of α=0.05 for all statistical tests. In addition to the quantitative metrics listed above, English language transcript data were entered in Dedoose software for qualitative data analysis. After familiarisation with the data, the content of the data was coded line by line for thematic analysis following a framework theory approach to describe and interpret health providers’ and communities’ perceptions of barriers and enablers to MNCH service provision. The framework approach involves using some pre-assigned themes to initially categorise data while also adjusting and iterating the coding scheme to accommodate newly emergent themes, subthemes, and categories through inductive interpretation.9 Coded data were examined for potential relationships and themes were also assessed across relevant participant demographic categories to understand different user perspectives. Findings were described under pre-assigned and newly emerged themes. As in-person meetings were restricted by local authorities during protocol development and data collection due to COVID-19 pandemic, we were not permitted to involve clients or the public in study design or reporting and dissemination plans of our research.