Study design and area
A Community-based, cross-sectional study design was carried out from January 18–30, 2017 in Dessie and Combolcha towns. Dessie is the capital city of South Wollo administrative Zone, and is one of the three metropolitan towns in the Amhara Regional State. It is located in the Northern part of Ethiopia about 401 Kms from the capital city, Addis Ababa to the North East, and about 480 Kms away from the capital city of the Amhara Regional State, Bahir-Dar, to the east. According to the 2013–2014 years, the total population in the town is 195,661 (male 94,285, female 101,376).
Combolcha town is also 374 kms far away from Addis Ababa and it is one of the industrial zones in the nation. Administratively Combolcha is structured in 11 Administrative Units (called Kebeles in Amharic). The town has relatively hot climatic condtion as compared to Dessie. According to 2015/16, the total households are 30, 631, in which 23, 102 (75.4%) households are living in urban areas and 7, 529 households are residing in rural areas. Housewives or household heads that had been living for at least 6 months and above in the randomly selected Kebeles were included in the study. Severely ill individuals and those who cannot prepare foods in their home were excluded from the study.
Sample size was determined using single population proportion formula by considering the prevalence of insufficient iodine intake at household level in Tigray a nearby region in Ethiopia as 63% [20], 95% confidence level based on standard normal distribution, 5% as degree of precision and 1.5 as design effect. Assuming 5% for non-respondent, the sample size was 524 households.
Multistage sampling technique was used to select target respondents. Dessie city administration has ten administrative units (Kebeles) and Combolcha town has five administrative units (Kebeles). Then three Kebeles from Dessie town and one Kebele from Combolcha town were selected randomly. The sample was allocated proportionally to the size of households in the two areas. The allocated sample for Dessie town was reallocated to the three Kebeles based on the total households resided. The total households in the selected Kebeles of the two Towns were divided to the allocated sample sizes to get the sampling interval and the study subjects were systematically drawn.
Data measurement and quality control
All required data were assembled using a questionnaire organized from EDHS, UNICEF procedure protocol and from other studies [11, 20, 21] via Open Data Kit (ODK) tool. The questionnaire was compiled from validated pre-established sources and was modified contextually as per the objectives of the study [2, 10, 19,20,21,22]. Six data collectors who had Bachelor of Science (BSc) in medical laboratory technicians with previous data collection experiences, were recruited. Additionally, two Master of Public Health (MPH) professionals with previous experiences in handling survey field work were selected as supervisors.
Before the commencement of actual survey, pre-test was conducted on 5% of the sample in areas found other than the study site. Detailed investigation was carried out to shuffle the tool in order to address the desired study objective. A 3 days intensive training was given for data collectors and supervisors on how to record, compile and complete the questionnaire. The RTK was done under close supervision and if there is any suspicion the test was repeated.
Determination of iodine content from dietary salt by using RTK, socio-demographic, economic variables, sources of iodized salt, knowledge about iodine and its effect, common food sources, container/shelf of iodized salt, timing of addition during cooking and other associated factors were obtained by interview using smart phones.
One medium sized tea spoon of iodized salt was taken from each households. The semi-quantitative estimation of the salt iodine by using a RTK is based on the reaction between starch and iodine to form starch-iodine complex. This test solution contains an acidic buffer and a reducing agent, which convert potassium iodate (KIO3) to elemental iodine (I2) [11].
The white cup found from the test kit was filled with salt and spread the salt surface flat first. Secondly, two drops of the test solution on the surface of the salt were added and the colour of the test sample is compared with the standard colour chart (< 15 ppm or > 15 ppm) [13, 16], within 1 min and iodine content was determined. If the colour of test sample did not appear after 1 min, three to five drops of the recheck solution were added on a fresh sample, followed by two drops of test solution were added on the same spot and then the colour was compared with the colour chart to record the content (recorded as “adequate” for blue colour, as “Low” for gray colour and as “nil/zero” for white colour).
Data management and analysis
The data collected by ODK software were uploaded to cloud server of the user account. Once the data had been collected and stored, it was downloaded in the form of comma delimited file type. Then, it was finally transported to STATA version 12 for analysis. The data were explored for missing data, distribution of outcome variable, and test of parallel lines and model fitness information.
The association between the adequacy of iodized salt and related factors was performed through ordinal logistic regression. The main effect was specified for the location of the proportional odds model, test of parallel lines checked (p-value = 0.34), Goodness of fit statistics was checked (P-value = 0.67) and the Link function was Logit during the analysis. Candidate variables with p-value < 0.2 moved to the final model to identify predictor variables and finally, a p-value of < 0.05 (two-tailed) was used to declare statistically significant variables.
Socio-economic status of the households was measured using assets by Principal Component analysis. After check the assumption of sample adequacy, communality and presences of complex struction, the score was ranked from the first (poorest) to fifth (richest) quintile (constituting 20% of distribution in each category).