Dietary Diversity Practice and its Inuencing Factors Among Pregnant Women in Afar Region of Ethiopia: A Cross-sectional Study Design Supplemented by Qualitative Study.

Background: Women in low-income countries are frequently malnourished when they become pregnant, and the demands of pregnancy can exacerbate nutritional deficiencies, 20 particularly micronutrient deficiencies, with serious health effects on the fetus. Antenatal 21 nutritional supplements can help to improve birth outcomes and maternal health. As a 22 result, determining the magnitude of dietary diversity and its influencing factors among 23 pregnant women in the pastoral region of Afar, where no study has been conducted, is 24 an essential in order to establish an intervention program in the region. Method: A mixed study comprising 241 pregnant women and six focus group discussions 26 was conducted from October 1 to November 10, 2018. Participants in the quantitative 27 study were selected by a systematic sampling method, whereas those in the focus group 28 discussions were chosen by using purposive sampling method. The data was collected 29 using pretested questionnaires administered via face-to-face interviews. The relationship 30 between dietary diversity practice and its affecting factors was investigated using logistic 31 regression analysis. The strength of the association was determined by odds ratio with a 32 95 % confidence interval. Thematic framework was used to analyse the qualitative data. 33 Results: Seventy-three percent of pregnant women had poor dietary diversity. Dietary diversity was higher in younger pregnant women who were under the age of 20 years 35 (AOR=5.8; at 95% CI: 1.6-13.5) and aged between 21-25 years (AOR=3.9; at 95 percent 36 CI:1.1-12.2) than in older pregnant women with over the age of 30 years. Those 37 participants with a high average family income (above 4500 birr) had a good dietary 38 diversity when compared to those with family income less than 1500 birr (AOR=0.1:95% 39 CI;0.02-0.7) and between 1500-3000 birr (AOR=0.05:95% CI;0.01-0.2). Pregnant women who had one antenatal care visit practiced less dietary diversity than those who had four 41 or more (AOR=0.18: 95 percent CI; 0.04-0.8). Protein-rich foods (meat and eggs), 42 semisolid foods (porridge and cereal soup), milk, fruit (banana) and vegetable (cabbage) 43 were the most commonly avoided foods by pregnant women. These meals were 44 commonly avoided since they produced large babies and were attached to the fetus's 45 body. 46 Conclusion: The majority of study participants had a poor dietary diversity. Pregnant 47 women with a low family income and only one prenatal care visit were less likely than 48 those with a high family income and four or more antenatal care visits to practice dietary 49 diversity, respectively. Most pregnant women avoided high-protein diets, semi-solid 50 foods, milk, vegetable and fruit. Due to the presumptions of producing large fetus and 51 attached to the fetus's body, these foods were avoided. 52

Although few studies have been conducted in other parts of Ethiopia, the practice of 100 dietary diversity and its associated determinants among pregnant women was not studied 101 in the context of Afar pastoral region, where sociocultural and economic realities are 102 considerably different. In order to address this knowledge gap, a study was established 103 to assess the magnitude of dietary diversity practiced and its influencing factors among 104 pregnant women residing in Awash seven district, Afar region of Ethiopia.  141 A quantitative cross-sectional study complemented by qualitative study design was used 142 to determine dietary diversity practice and its influencing factors among pregnant women 143 from October 1 to November 10, 2018. The sample size for quantitative data was 144 calculated using a single proportion formula with the assumption of a small total 145 population (512). Thus, considering 95% confidence level with a 5% precision, and taking 146 57% inadequate dietary diversity from a prior study in Dire Dawa, Eastern Ethiopia (7),

147
The sample size was found to be 217. After adding for 15% non-response rate, the final 148 sample size was 250. Focus group discussions (FGDs) with groups of pregnant women 149 were conducted until the qualitative data was saturated.

151
All pregnant women who agreed to take part in the study and had lived in the study area 152 for at least one year prior to the study period were included. The study did not include 153 pregnant women who had chronic conditions like cancer or diabetes. This was due to the 154 fact that these illnesses are known to have an impact on a person's food consumption 155 and nutritional status. The respondents' health information was used to compile this data.

156
Pregnant women who had consumed special diets in the previous 24 hours owing to 157 holidays or celebrations were also excluded.

158
Sampling technique and procedures 159 The sampling interval was calculated by dividing the total number of pregnant women 160 who attended antenatal care in the health facility three months ago by the total sample 161 size. Then, using systematic sampling technique, the required study subjects were recruited. From the first two study subjects, the first study subject was selected by lottery, 163 and then every second study subject was selected until the required sample size was 164 achieved. But, pregnant women who had previously participated in the study did not re-165 interview. Purposive sampling was used to select focus group discussants (FGD) for the 166 qualitative study. The FGD participants, on the other hand, were not the same as those 167 who were sampled for quantitative data. were also consulted to amend the food list in the food groups, whether or not they 184 matched with local names of foods, and acceptable terminology was agreed upon (modification was carried out based on local language). The respondents were asked to 186 recall all foods (meals and snacks) consumed the previous day and night (24-hour recall).

187
After recalling all foods and beverages consumed, these food items were recorded. The 188 interviewers underlined the corresponding foods in the list under the appropriate food 189 categories and entered "1" in the column next to the food group if at least one food was 190 consumed. Once the recall was completed, look for food groups that were not consumed 191 by respondents. After ensuring that no meals from that food group were eaten, "0: was 192 filled in the right-hand column corresponding to the food group.  guiding questions used to interview focus group discussions was shown (Table 1).   and semi-solid foods (cereal soup and porridge), respectively, out of the total food 275 restricted participants. (Table 4). and milk production were taken by more than half of the participants (57.7%) (Table5).   1) 0.1 (0.03-0

301
In the qualitative study, a total of 38 pregnant women involving in six focus group 302 discussions, four of which were held with urban residents and two with rural inhabitants.

303
In urban residents, seven pregnant women participated in each of the two focus group 304 discussion and six study participants involved in the remaining each two focus group 305 discussion. In rural residents, each of the two focus group discussion had six participants.  income countries (2), but lower than a study in Kenya (4) that found cereal-based foods 397 were the most popular (99%). Furthermore, all food groups consumed by pregnant 398 women in the current study were lower than in a study done in north east Ethiopia (7), 399 with the exception of milk and milk products, which were common in the pastoral region Ethiopia's Oromia area (5)  involving in any process of this study.