Impact of the Early Stages of the COVID-19 Pandemic on Coverage of Reproductive, Maternal, and Newborn Health Interventions in Ethiopia: A Natural Experiment

Background: The COVID-19 pandemic and response have the potential to disrupt access and use of reproductive, maternal, and newborn health (RMNH) services. Numerous initiatives aim to gauge the indirect impact of COVID-19 on RMNH. Methods: We assessed the impact of COVID-19 on RMNH coverage in the early stages of the pandemic using panel survey data from PMA-Ethiopia. Enrolled pregnant women were surveyed 6-weeks post-birth. We compared the odds of service receipt, coverage of RMNCH service indicators, and health outcomes within the cohort of women who gave birth prior to the pandemic and the COVID-19 affected cohort. We calculated impacts nationally and by urbanicity. Results: This dataset shows little disruption of RMNH services in Ethiopia in the initial months of the pandemic. There were no significant reductions in women seeking health services or the content of services they received for either preventative or curative interventions. In rural areas, a greater proportion of women in the COVID-19 affected cohort sought care for peripartum complications, ANC, PNC, and care for sick newborns. Significant reductions in coverage of BCG vaccination and chlorohexidine use in urban areas were observed in the COVID-19 affected cohort. An increased proportion of women in Addis Ababa reported postpartum family planning in the COVID-19 affected cohort. Despite the lack of evidence of reduced health services, the data suggest increased stillbirths in the COVID-19 affected cohort. Discussion: The government of Ethiopia’s response to control the COVID-19 pandemic and ensure continuity of essential health services appears to have successfully averted most negative impacts on maternal and neonatal care. This analysis cannot address the later effects of the pandemic and may not capture more acute or geographically isolated reductions in coverage. Continued efforts are needed to ensure that essential health services are maintained and even strengthened to prevent indirect loss of life. Data for this study come from the Performance Monitoring for Action (PMA) Ethiopia survey, a survey project comprised of an annual nationally representative cross-sectional survey, a panel survey following women from pregnancy through 1 year postpartum, and an annual Service Delivery Point (SDP) survey. The data for this analysis come from the panel survey. PMA Ethiopia is conducted in collaboration between Addis Ababa University and Johns Hopkins Bloomberg School of Public Health. PMA Ethiopia panel survey used multistage cluster sampling using probability proportional to size to select 217 enumeration areas (EAs) across six regions in Ethiopia, with region (Afar, Addis Ababa, Amhara, Oromia, Southern Nations Nationalities and Peoples, and Tigray) and residence (urban/rural) as strata. In Afar and Addis Ababa, only region was used for stratification. To identify women for the panel survey, a census was conducted among 36,614 households between October and November 2019. All women aged 15–49 were screened [32,792] and, if they reported being currently pregnant or having delivered within the past 6 weeks, were eligible for the panel study. In total, 2,889 women were identified as eligible and 2,855 enrolled to complete interviews at enrollment, 6 weeks, 6 months, and 1 year postpartum (Figure 1). Data used in this paper were reported at the 6-week interview, which had a follow-up rate of 93.3%. Study cohort diagram. *Women who were pregnant or 0–4 weeks postpartum at the time of the first panel interview received survey questions related to maternal care services they received up to the time of interview. Estimated or actual delivery dates of women were used to schedule a second interview, which was conducted when respondents were about 6 weeks postpartum. Some 6-weeks postpartum interviews were conducted before the COVID-19 pandemic; others were conducted during the COVID pandemic. PMA survey interview questions slightly for interviews conducted during the COVID pandemic. **Women who were 5–9 weeks postpartum at the time of the first panel interview received a combined set of survey questions that other women received during two separate interviews. All of these interviews were conducted before the COVID-19 pandemic. PMA Ethiopia paused data collection in early April due to the COVID-19 pandemic. At that time, questionnaires were modified to include a range of questions about COVID-19 knowledge and risk and the role of COVID-19 in care-seeking behaviors for MNH. When data collection resumed in June with enhanced safety protocols, including social distancing, COVID-19 symptom screening, and mandatory mask requirements, all women with outstanding surveys were interviewed using the updated questionnaires. As a sub-cohort of women had delivered prior to the onset of the COVID-19 pandemic and a sub-cohort delivered during the COVID-19 pandemic, a “natural experiment” within the PMA Ethiopia cohort was introduced, providing a unique opportunity to apply a pre-post cross-sectional study design to examine the early impact of COVID-19 on the coverage of peripartum care indicators. Women provided oral consent to participate at the initial household screening and prior to enrollment in the panel survey for all eligible women. All procedures were approved by both the Addis Ababa University [075/13/SPH] and Johns Hopkins Bloomberg School of Public Health [00009391] Institutional Review Boards. Additional information on the PMA Ethiopia survey can be found at Zimmerman (17). Restrictions to curb the spread of COVID-19 were introduced in Ethiopia between last March and early April, with some variation in date of introduction by regional states. In addition to structural disruptions, we assume this time also aligns with an increased public awareness of the potential threat of COVID-19. Translating this period of restriction into potential impact on health service access and use in the PMA cohort, we assume those women who gave birth in April or later could experience disruptions to late-ANC visits, care offered during childbirth, and services delivered in the first month after birth. If restrictions did impact service availability, we expect it would immediately affect labor and delivery care. Impact on ANC would be tempered due to repeat service visits throughout the pregnancy. For births that occurred in May 2020, disruption to antenatal service would translate to potential loss of the final pre-birth visit under a four-visit ANC schedule. Care delivered in the neonatal period could also have been impacted in births occurring as early as March 2020. In defining the appropriate COVID-19 affected and unaffected groups, we also considered the comparability of recall periods. Due to a pause in six-week post-birth follow-up interviews in April and May, births between February and April received follow-up interviews up to 25 weeks after birth (Supplementary Figure 1). This delay in follow-up could result in lower recall accuracy across indicators and significant bias in indicators with reference periods tied to the timing of interview administration (e.g., current breastfeeding or family planning use) or time between birth and interview (e.g., care-seeking for illness in newborn since birth). For our primary analysis, we defined our COVID-19 affected cohort as those born in May 2020 (average recall period: 9.4 weeks) or later and our COVID-19 unaffected cohort as births between August 2019 (start of post-birth data collection) and January 2020 (average recall period: 6.8 weeks). Births that occurred between February and April 2020 were excluded. We conducted a sensitivity analysis of indicators with a time-invariant reference period more loosely defining the unaffected cohort as August 2019 to February 2020 births (average recall period: 8.6 weeks) and the COVID-19 affected cohort as births in April 2020 or later (average recall period: 12.0 weeks). For indicators with unrestricted reference periods, therefore most susceptible to bias due to differences in recall period (i.e., vaccination, exclusive breastfeeding, care-seeking for infant illness, and postpartum family planning), we restricted the comparison of cohorts to only follow-up interviews that occurred more than five weeks and <10 weeks after birth (mean recall period COVID-19 unaffected cohort: 6.7 weeks; COVID-19 affected cohort: 7.9 weeks). We examined the effect of the COVID-19 pandemic and response on health interventions in the peripartum period. The PMA Ethiopia six-week postpartum questionnaire collected data on standard indicators of health practices and interventions during antenatal care, childbirth, and the neonatal period. Where an intervention could only be received through contact with the formal health system (e.g., blood transfusion) we report the indicator as the proportion of the population delivering at a facility that received the intervention. These indicators serve to assess changes in the content (and potentially quality) of service administered during the time period. Indicators of service contact (e.g., facility delivery) are calculated as a proportion of the total target population and demonstrate potential changes in both care-seeking behaviors and service access. Interventions or practices that can be accessed through multiple healthcare channels or do not require engagement with the healthcare system are similarly presented as coverage indicators among the total target population. To assess the effect of the COVID-19 pandemic and response on health practices, services, and outcomes, we compared these indicators in our COVID-19 affected cohort vs. our unaffected reference cohort. The primary analysis estimated the odds ratio of intervention receipt or practice (yes/no) for those in the COVID-19 affected cohort compared to the reference cohort using logistic regression adjusting for survey weights. We calculated the association at the national level, with and without adjusting for characteristics of the mother and birth. The adjusted regression assessed the cohort effect after accounting for variations in parity (first birth, 1-2 previous births, 3+ births), maternal education (none, attended primary, attended secondary or higher), maternal age, household wealth (relative quintile), urban vs. rural residence, and regional state. We also looked at associations between cohorts residing in Addis Ababa, other urban areas, and rural areas separately, with and without adjusting for covariates. We posited restrictions and COVID burden might have a greater impact in population centers that are more dependent on public transport, more vulnerable to economic shocks, and more susceptible to COVID-19 transmission. We also calculated the unadjusted coverage of each intervention or practice in both the COVID-19 affected and unaffected cohorts. We also compared the incidence of stillbirth and neonatal death in the two cohorts using Poisson regression. To account for potential left truncation of our data due to the absence of early stillbirths among women enrolled late in pregnancy, we restricted our stillbirth analysis to only those enrolled in either their first or second trimester of pregnancy.