The study was conducted in Addis Ababa, from February 15 to March 19. Addis Ababa is the capital city of Ethiopia. It has 11 sub-city and headquarters of Africa Union. The current metro population of Addis Ababa in 2022 is 5,228,000 which is located in the center of Ethiopia. There are around 16 public hospitals and 22 private hospitals in Addis Ababa. All hospitals provide comprehensive emergency obstetric care services in the city. Moreover, there are about 100 health centers that provide basic emergency obstetric care services in Addis Ababa.

The institutional-based cross-sectional study was employed

All pregnant women who were attending antenatal care (ANC) at public hospitals in Addis Ababa during the data collection period.

Those pregnant women who were attending ANC at selected public hospitals during the data collection period were included in the study.

All pregnant women who are on antenatal care follow-up in selected public hospitals were included in the study.

Pregnant mothers who are seriously ill during data collection

The sample size was determined by single population proportion with the following assumptions: Since this study was the first of its type in Addis Ababa P = 50% was taken, with a 5% margin of error, and 95% confidence interval and the final sample size was 422.

The sampling procedure started by identifying public health hospitals in Addis Ababa and four public hospitals was randomly selected by lottery methods and included in the study, these include Abebech Gobena Maternal and child health memorial hospitals, St’ Peter specialized hospital, Minliik II Hospital and Gandhi Memorial Hospital. The sample sizes for each school were allocated by using the proportional allocation formula.

Good knowledge: Participants who scored greater than or equal to the mean score8.

Poor knowledge: Participants who score less than the mean score8.

Good practice: Participants who scored greater than or equal to the mean score8.

Poor practice: Participants who scored less than the mean score8.

· Knowledge towards Covid19

· Practices towards Covid19 prevention measure

Socio-demographic (age, marital status, occupational status, religion), obstetric and reproductive factors (parity, gravidity, history of ANC)

Data were collected by four trained BSc midwives using an interviewer-administered structured questionnaire.

The collected data was tallied manually and entered into the Epi-data manager and exported to SPSS version 25 for statistical software. An exploratory analysis was carried out to determine the nature of data such as normality, the presence of outliers, and missing values. Before data analysis, we had got that the data was normally distributed based on the Shapiro-Wilks test. Then, the data were described using relative frequency and per cent for categorical variables. Continuous variables were also expressed by mean and standard deviation. Frequency and percentage were to describe the socio-demographic, reproductive and obstetric characteristics, knowledge and practice of the participants. The final report was expressed by tables and graphs. Finally, only those independent variables that maintain their association with outcome variables in multivariable logistic regressions (p-value <0.05) were used to construct the final models. The odds ratio with its p-value and confidence interval were reported in each logistic regression analysis. For measuring the strength of the association between the outcome and independent variables, Crude Odd Ratio (COR) and Adjusted Odd Ratio (AOR) along with a 95% Confidence interval (CI) were calculated.

The questionnaires were prepared in English and translated to Amharic languages and re-translated to English. The tools were pre-tested before the actual data collection among 20 pregnant women at Tirunesh Beijing Hospital, which was not included in the final study analysis. The training was given to 4 BSc midwifery data collectors and 1 supervisor for one-day duration. The data collection process was supervised by principal investigators and trained supervisors. The filled questionnaires were randomly checked for completeness and consistency.

Ethical clearance was obtained from the Kotebe Metropolitan University institutional research ethics board with the reference number አጠ5/38/12/6083. A formal letter from the research ethics board was submitted to the Addis Ababa health office and concerned bodies to obtain their cooperation. Then a letter of permission and support was written to each respective public hospital. Study subjects were informed about the study's purpose. At the time of data collection, written consent is taken from the participants to confirm whether they are willing to participate. Those unwilling to participate were allowed to do so. The confidentiality of responses was also ensured throughout the research process. The study was performed by the ethical standards of the Helsinki Declaration.

A total of 418 study subjects take part in this study making the response rate 99.05%. In this study, the maximum age of the participant was 44 and the minimum age was 20 with a mean and standard deviation of 29.89±5.438. The majority (48.3%) of the participant’s age group was 26-30 years followed by greater than or equal to 35 years which accounted for 20.6%. Most of the participant's educational status was completed primary school which accounted for 32%. From the study subjects about half of the participants were Orthodox in religion 51.2% and most of them have occupational status as a housewife (41.1%) Table 1.

In this study the majority (67.0%) of the subjects were multigravida and the rest 33.0 % were Primigravida. In more than three-quarters (80.6%) of the women the current pregnancy was wanted. Eighty-one (19.4%) of the study participants had previous pregnancy-related complications and sixty-two (14.8%) of the participants had current pregnancy-related complications. Ninety-Six (22.9%) of the study participants had a history of chronic disease Table 2.

About 417 (99.8%) pregnant women reported never hearing about COVID-19. Of those 195 who had ever heard about COVID-19 315(%) heard from mass media (TV 49.2%, Internet 31.75%, and Radio 19.04%), 120(24.5%) heard from a health care provider and 25(11.8%) heard from friends was the source of information. Around 49% of the respondents have good knowledge towards Covid-19 Figure 1, Table 3.

The pregnant women’s responses regarding preventive practice against COVID-19 infection was wearing a face mask regularly (89.1%), frequent hand washing and using sanitizers (87.1%), wearing a face mask when having flu-like symptoms (86.1%), avoiding hand contact having cough (73.9%) and keeping physical distancing (73.7%). The women participating in this study responding show that 54.8% of them were Reporting COVID-19 symptoms to health facilities but the remaining 45.2% did not report to the facility Table 4.

The prevalence of poor practice was found to be 45.7% whereas 54.3% had good practice towards preventive practice against COVID-19 infection. Hence, just 54.3% of study participants practiced COVID-19 infection prevention measures under scientific recommendation Figure 2.

In bivariate logistic analysis age, education, occupation, marriage, gravidity, parity, condition of pregnancy, previous complication of pregnancy outcomes, and number of ANC visits were significantly associated with knowledge towards covid-19 infection with a p-value of < 0.2. In multivariable logistic regression 26-30yrs (AOR=0.46, 95%CI: 0.014,0.12), 30-35yrs (AOR=0.093, 95%CI: 0.037-0.24),>36yrs (AOR=0.482, 95%CI: 0.234-0.99), no education (AOR=0.23, 95%CI: 0.099-0.52), Primary school (AOR=0.199, 95%CI: 0.104-0.4), Secondary school (AOR=0.282, 95%CI: 0.14-0.55), divorced (AOR=0.15, 95%CI: 0.065-0.34) and widowed (AOR=0.16, 95%CI: 0.024-1.03) were statistically significant Table 5.

In bivariate logistic analysis age, education, occupation, marriage, family size, previous complication of pregnancy outcomes, chronic health problems and number of ANC visits were significantly associated with preventive practice levels of covid-19 infection with a p-value of < 0.2. In multivariable logistic regression 30-35yrs (AOR=0.334, 95%: 0.115-0.97),>36yrs (AOR=0.28, 95%CI: 0.11-0.69), no education (AOR=0.06, 95%CI: 0.019-0.18), occupation (Civil servant) (AOR= 0.28, 95%CI: 0.122-0.66), marital status (divorced) (AOR=0.042, 95%CI: 0.01-0.18), family size (AOR=0.334, 95%CI: 0.169-0.66), previous complication of pregnancy outcomes (AOR=0.019, 95%CI: 0.01-0.061), chronic health problem (AOR=14.66, 95%CI: 0.457-39.4) and number of ANC visit (two) (AOR=5.704, 95%CI: 2.41-13.5) were statistically significant Table 6.

In this study, the proportion of pregnant women who had good knowledge towards COVID-19 prevention measures was 49% (95% CI 45%-54%). This finding is consistent with the study done in Debre Tabor Town8. But this finding is lower than studies conducted in Northern Ghana, 10 Wollega zones 11. And it is higher than the study conducted in South Africa 12 This difference is probably associated with variations in socio-demographic characteristics and study setting.

The preventive practice in the present study indicated that 54.3% (95% CI 49.3%-59.3%) of them were practicing well towards preventive measures for COVID-19. It is higher than the study conducted in Debre Tabor showed that 47.6% of participants have good practices towards COVID-198. This may be due to a variation of individual beliefs on preventive measures among pregnant mothers. But it is lower than the study conducted in Iran where 98.92% of study participants had good practice towards COVID-19. This might be due to our study participants having a high rate of not being educated compared with Iranian13 and probably associated with variations in socio-demographic characteristics, study setting

Participants whose age group was 26-30yrs, 30-35yrs and > 36yrs are 54, 90.7, and 51.8 times less likely to have good knowledge about COVID-19 when compared to participants whose ages were 20-25yrs. This is consistent However, a study done in Debre Tabor revealed that participants whose age group was found between 15 and 24 years were nearly five times more likely to have a sound knowledge about COVID-19 as compared to participants whose ages were greater than or equal to 35 years8. However, studies done in low‐resource Africans, aged older than 40 years are 81% less likely to have inadequate knowledge compared to those < 30 years 14. This could be due to the younger population being closer to social media, so they will have more information.

In this study, education is significantly associated with knowledge about COVID-19. Pregnant mothers with no education, in primary school and secondary school are 77%, 80.7%, and 71.8% less likely to have good knowledge about COVID-19 when compared to participants whose education level is college and above. Another study also revealed that those with no formal education were 6.3 times more likely to have inadequate knowledge than those with a tertiary education level 14. This could be due to those who are college and above are closer to updated information and they have a greater ability to interpret information that has been transmitted.

The finding of these studies revealed that marital status is significantly associated with knowledge about COVID-19. Pregnant mothers who are divorced, widowed and other (single) are 85%, 84%, and 66% less likely to have good knowledge about COVID-19 when compared to participants who are married. But this was not significant in the studies done in Debre Tabor town and Wollega zone 811. This indicates that being married will improve the likely hood of sharing and interpreting information.

The finding of these studies revealed that women aged 30-35 years and > 36yrs 66.6% and 72% less likely to engage in good COVID-19 preventive practices respectively than those aged 20-25yrs. This finding was consistent with a study done in Guraghe, those aged 20–24, 25–29 and 30–34 were 1.22, 1.32 and 2.57 times more likely to practice COVID-19 preventive measures9. Young age women are more likely to engage in good COVID-19 preventive practices 101114. This may be due to a younger population being more likely to adopt a new lifestyle 11.

In this study, those who are divorced are 95.8% less likely to have good COVID-19 preventive practices than those who are married. However, another study revealed that married pregnant women are 2.99 times more likely to have poor COVID-19 preventive practices14. Another study revealed that those who are single are more likely to engage in unsafe practices 15. This indicates that being married will improve the likely hood of sharing and interpreting information.

This study identified that educational status is significantly associated with good COVID-19 preventive practices. When compared to those who attend a college diploma and above, those who are not educated are 94% less likely to engage in good COVID-19 preventive practices. In another study, pregnant women with at least a primary education were also associated with good COVID-19 preventive practices 1014. The possible reason is illiterate women are not more exposed to health information, especially regarding COVID-19 and are therefore less likely to take positive measures to protect themselves against the disease.

According to these studies, occupation status is significantly associated with good COVID-19 preventive practices. Those who are civil servants are 72% less likely to engage in good COVID-19 preventive practices than those who engage in private business. A study done in Iran revealed that having health-care-related occupations was significantly associated with lower practice15. According to the study done in Nigeria, occupations requiring physical contact (like farmer, artisan and trading) were the factors associated with poor practice of preventive measures14. These might be due to the flexibility of their work in which those who have private business have the freedom to stop at any time they went and isolate themselves but civil servants are dependent on others so making their own decision regarding the preventive practices could be difficult.

This study identified that family size is significantly associated with COVID-19 preventive practice. Pregnant women who have a family size >4 are 66.6% less likely to have good COVID-19 preventive practices than those who have a family size <4. Once more, women having less than or equal to three children had a good level of practice. The possible explanation might be due as the number of children increases the family size may become crowded which makes it difficult to maintain the recommended distance8. Besides, an increased number of children may negatively affect the economy of the family thus less affordable some preventive measures like soap, alcohol-based hand sanitizers, and face masks that are used to prevent the spread of COVID-19 from person to person8. This might be due to; the crowdedness of the family size will make them focus on daily necessary things/accommodation instead of concerning about their health.

The previous complication of pregnancy outcomes is significantly associated with COVID-19 preventive practices. Pregnant women who don’t have any complications no their previous pregnancy outcomes were 98.1 times less likely to engage in good COVID-19 preventive practices compared to those who have previous complications of pregnancy outcomes. But this was not significant in the study done in Debre Tabor town8. Because it might be due to not knowing/experiencing the physical, mental and socioeconomic burden of pregnancy complications they will be careless on the COVID-19 preventive practice as well.

In this study, women who do not have a chronic disease were 14.6 times more likely to engage in good COVID-19 preventive practices compared to those who have a chronic disease. But another study revealed that women who had a chronic disease were more likely to engage in good COVID-19 preventive practices compared to those who have no chronic disease 910. It could be due to a loss of hope due to the effect of the disease on those who have chronic disease that they don’t engage in good preventive practice.

Those pregnant women who have ANC visits two times are 5.704 times more likely to engage in good COVID-19 preventive practices compared to those who have ANC visits four times. It could be due to an increased rate of 2 times ANC visits. But a significant association was not seen between the number of ANC visits according to the study done in Northern Ghana10.