Study Design and Setting
Institutional based cross-Sectional study was conducted in Debre Markos referral and Finote Selam hospitals from September 15, 2019 to October 15, 2019. Debre Markos referral hospital is situated at Debre Markos town found in East Gojjam zone in Amhara region. It serves for the total of 3.5 million populations. Debre Markos city is located in northwest Ethiopia (19). Finote Selam hospital is situated at Finote Selam Town in western Ethiopia. The hospital provides health service to more than 1.5 million populations in its catchments. The town is Located in the West Gojjam zone of the Amhara region (19).
Debre Markos referral and Finote Selam hospitals are found in the area of same socio- demographic characteristics. Around those hospitals teff is the major food which consumed by the population. In addition to teff, wheat, maize, potatoes, lentils, beans, peas, honey, dairy products, fruits and vegetables are the major food those are common in the areas. Those hospitals were selected randomly to get adequate sample size regarding with this study. Add information. Quantitative data was collected from diabetic patients who are attends services in Debre Markos referral and Finote Selam hospitals.
The ethical clearance was obtained from Debre Markos University, Health science college ethical committee. Official letter was written along with the ethical clearance and was submitted to Debre Markos referral and Finote Selam hospital office. This study was not had any risk consequence for study participants. Participants were also informed that participation was voluntary and participants who refused the consent had respect. Written consent was taken from the study participants after briefing them the objective of the study. There was no age of less than 18 years diabetes patients in the sample included. The confidentiality of participant’s related data was maintained by avoiding possible identifiers such as name of study participants only identification number was used as a reference. The interviewer administered questionnaires was kept safe throughout the whole process of the research work. During data collection time any diabetic client with medical problem findings such as any Diabetic complication, education was provided and contacted with health professionals found in those hospitals for support and treatment.
Sample Size Determination And Sampling Procedure
Sample size was calculated by a single population proportion formula:
n= [(Zα/2)² *P (1-P)] /d²
Since there was no previous study the sample size was calculated by assuming a 50% proportion (P) of utilization of low glycemic index foods, a confidence level of 95% (1.96), and 5% margin of error (d). Based on this assumption the
n= (1.96)²\(\frac{0.5\left(1-0.5\right)}{0.05²}\) = 384
After substitution the estimated sample size yields 384. By considering a 10% of non-response rate, 422 diabetic patients were planned to be included in the study.
Systematic random sampling was used to select 422 study participants from the study population. First 422 samples were proportionally allocated to Debre Markos referral hospital (241) and Finote Selam hospital (181). This data were taken from SMART care of those hospitals by filtering out DM patients specifically. Since there was no well-ordered or separated registration for DM patients, the sample was taken from those hospitals patient flow. Then K value was calculated. Based on k value lottery method was used to select the first respondents. Finally, the study participants were selected every K value was 2.4. Then the respondents were selected every 2 individuals until the required samples was full filed or interviewed.
Data Collection Procedure And Measurements
Data was collected by using interviewer administered structured questionnaire developed from different literatures such as (8, 20–22). The tool contains socio- demographic characteristics (6 items), food preparation and feeding technique (3 items), market access characteristics (4 items), health status related questions (4 items), life style related questions (6 items), and knowledge based questions (9 items). There are twenty food types in English version questionnaire and this questionnaire was translated to Amharic language. The respondents were asked to recall all foods and drinks they have taken in the three consecutive day’s meal history prior to the survey (17, 23) to assess utilization of low glycemic index foods of the respondents. By using different literatures the local food items were collected and summarized in twenty items. Six Professional nurses (two supervisors and four data collectors) were recruited to collect data. The data collectors use the codes “1’ for respondents who says yes and “0” for respondents who says no for low glycemic based questionnaire. The supervisors were follow the data collection technique and all the data collectors and supervisor were have regular meeting with the principal investigator at the end of each week of data collection period.
Operational Definition
Glycemic Index: is a scientific ranking of how the foods we eat affect our blood sugar levels in the 2 or 3 hours after eating. Foods are measured against pure glucose, which has a value of 100 on the index (1, 4, 5).
Low glycemic index: It indicates that the carbohydrates in a given food have a lower affect on postprandial blood glucose and possibly insulin responses. It has a GI value of < 55 gm (e.g. dairy products, fruits, vegetables…) (1, 4, 5, 24).
Utilized: Diabetic patients who are consumed above 75% from the total listed low glycemic index foods (25).
Diabetes Mellitus: is defined as a case identified for having the disease diabetes by the clinician.
Good Knowledge: respondents who answer correctly to knowledge related question and those who scored equal and above 75% of the total value.
Market access: Access of market measured in time and distance to get low glycemic food groups for study participants.
Long distance: If the distance between market and study participants is greater than 3 Km (26).
Adherence to Exercise: If the patient follow recommended level of exercise for more than 3 days in the last seven days (27).
Alcohol Consumption: If reported consumption of alcohol twelve month prior to the survey (28).
Statistical analysis
The collected data were coded and entered into Epi Data Version 3.1 and cleaned. The cleaned data set was exported to Statistical package for social sciences (SPSS) Version 25 software for analysis. Study participants utilization on low glycemic index food groups (utilized, none utilized), and socio-demographic characteristics and other independent variables those affect utilization of low glycemic index foods were presented using relevant descriptive statistics. Binary Logistic Regression was performed to see the association between the dependent variable and independent variables. The adequacy of the final model was checked by using Hosmer and Lemeshow goodness of fit test and which had value of significance was 0.740. An assumption of Binary Logistic Regression such as no multi-co linearity was checked and the Variance Inflation Factor (VIF) of all variables was less than 5 and all included in multivariable analysis. Both bivariable and multivariable analysis were conducted. Those variables that showed significant association at p less than 0.25 were included to multivariable analysis. Finally those variables that showed significant association at p less than 0.05 with 95% CI was declared statistically significant factors of utilization of low glycemic index foods.