Epi info version 7

A cross-sectional study was employed from May 1^st to August 15^th, 2019 in Addis Ababa, the capital city of Ethiopia. The total population of the city was 2,738,248 consisting of 1,304,518 men and 1,433,730 women. A total of 45 hospitals (11 governmental, 31 private, and 3 NGO) are found in the city. According to the Addis Ababa health administration office, an estimated 30,000 HCP are engaged in clinical and other related works in Addis Ababa.
The sample size was calculated by Epi info version 7.1 considering the following parameters; P: 74.4% of WTP for SHI (16 ), d = margin of error is 5%, 95% CI = Za/2 = 1.96%, 10% nonresponse rate, design effect: 1.5, and the final sample size became 480. Multistage sampling was used to select study participants. First, 15 hospitals randomly selected (5 government, 9 private, and 1 NGO) from the 45 hospitals found in Addis Ababa. Second, the sample was proportionally allocated for the selected hospitals and the actual study participants were selected using the lottery method.
Data was collected using an interview questionnaire which was prepared by reviewing similar WTP studies and modified to fit the local context (8 , 11 , 16 –22 (link)). It was pretested among 10% of the sample size of the study participants, which were not included in the actual study. The data were collected by five public health officers and supervised by two assistant professors. Respondents were asked about their maximum WTP for SHI when they first expressed their willingness to join. Subsequently, respondents were invited to choose a lottery ticket from a stack of unmarked envelopes. Each respondent was randomly assigned to one of three initial values; 3%, 4% of monthly salary, and 5% of monthly salary. A maximum of three trials were performed with each respondent if the respondent was not satisfied with the results of the earlier bids. If the answer was “yes,” the interviewer increased the bid by 1% until the respondent says “no” and vice versa. Finally, those who chose 3% and above are considered as WTP yes (16 , 17 (link), 23 ).
The data were entered into Epi info version 7.1 and exported to SPSS version 23 for data processing and analysis. Descriptive data were presented in frequency with percent and mean with standard deviation. Logistic regression analysis was carried out and all explanatory variables that were significantly associated with the outcome variable in the bivariate analyses (P < 0.05) were entered into multivariate logistic regression model. Crude and adjusted odds ratios with their 95% confidence interval (CI) were determined, and statistically significant association was asserted based on P value less than 0.05. Model fitting test was performed using the likelihood ratio test, and multicollinearity was checked using the variance inflation factor.

The data were checked for completeness and coded manually. EPI-INFO version 7 and SPSS version 20 were used for data entry and analysis, respectively. Descriptive statistics, such as frequencies and percentages were computed to describe the study population in relation to relevant variables. Bivariate and Multivariable logistic regression analyses were carried out to see the presence of association between dependent and the independent variables. Variables with p-values of < 0.2 in the Bivariate analysis were further fitted to multivariable logistic regression analysis. Adjusted odds ratios with 95% confidence intervals were computed and variables with p- values of < 0.05 in the multivariable analysis were considered as statistically significant.

Data were cleaned, code and entered into Epi-info version 7.1 then exported to SPSS version 23 for analysis. Descriptive analysis was carried out to see the distribution of independent variables. Binary logistic regression was used to examine associations between the dependent variable and each independent variable. Based on the bivariate analysis those factors whose crude associations to the timing of antenatal care booking at p < 0.2 was entered into the multivariate analysis to get adjusted odds ratio.
The strength of association was determined by using a crude odds ratio in the bivariate analysis and adjusted odds ratio in multivariate analysis. P-values and 95% confidence interval was used to determine the level of significance of the association. P < 0.05 considered as statistically significant. Hosmer and Lemeshow Test were used for checking the model fitness of logistic regressions.