A community-based, cross-sectional study design was used from March to April 2018 in the Shire town Northwest zone of Tigray, Ethiopia, located 1087 kilometers from Addis Ababa, the capital city of Ethiopia and 309 kilometers from Mekelle, the capital city of Tigray. The study employed a single population proportion sample size determination formula taking the magnitude of health-seeking behavior 72.7% from previous studies conducted in Bahir dar 15, 5% marginal error (d) with 95% confidence. The final sample size with the design effect of 1.5 and 10% non-response rate was 504.
Multi-stage sampling technique was used for interviewing of 504 study participants. Three Kebels were selected by simple random sampling technique and to get the sample size from every three Kebels proportional allocation to sample size was done. Systematic sampling technique (K=16) was used to choose the household for an interview and the starting mother was selected using a lottery method. Data was collected using a pre-tested interviewer-administered structured questionnaire by three BSc degree holder nurses. The tools were adapted from different kinds of literature 13, 15-17 then modified and contextualized to fit the local situation and research objective.Modern Healthcare seeking behavior: Activities that are taken by mothers who wanted care for their children who have a health problem in the formal health care institutions (government health facilities, private hospitals/clinics) 2 Caregiver: The responsible individuals who taking care of a child either the relatives of the child or the mother13 Training and orientation were given for data collectors. The questionnaire was initially prepared in English and then translated into Tigrigna version (local language) by different experts of both languages.
The questionnaire was pre-tested on five percent of the sample size in Aksum town. The collected data were reviewed and checked for completeness and consistency by the supervisor and principal investigator on a daily base. The data were coded, entered, cleaned edited using EPIDATA version 3.
1, and then exported to SPSS Version 22.0 for analysis. Binary logistic regression analysis was used to look at the statistical association between the outcome variable and every single independent variable. Variables which showed statistical significance during bivariate analysis at ? 20% (p-value ?0. 20) were entered into multivariate logistic regression to isolate an independent effect of the predictors by using the backward elimination method.
The Hosmer-Lemenshow test was used to check the appropriateness of the model for analysis. Results were presented using tables and texts. Adjusted odds ratios (AOR) with 95% CI, were estimated to assess the strength of associations and statistical significance was declared at a p-value