Factors associated with dietary diversity and physical activity of pregnant women in Bangladesh: A cross-sectional study at an antenatal care setting

Introduction: Both dietary diversity and physical activity during pregnancy are very important since they are known to affect pregnancy and birth outcomes. However, little is documented on dietary diversity and physical activity among pregnant women in Bangladesh. Accordingly, this study was designed to assess the dietary diversity and physical activity level, as well as their associated factors, among pregnant women at an antenatal care setting in Bangladesh. Methods: An institution-based cross-sectional study was conducted with 96 randomly selected pregnant women who sought antenatal care at a hospital in Khulna City, Bangladesh in July 2020. Dietary diversity and physical activity were assessed using the 24-h dietary recall method of consuming seven food groups, and standardized Pregnancy Physical Activity Questionnaire (PPAQ), respectively. Simple and multivariable logistic regressions were conducted to determine the factors associated with dietary diversity and physical activity among pregnant women. Results: Around 70% of participants had high dietary diversity and 58.3% were physi-callyactive.Beingjobholders,gettingcounselingondiet,andbeingfromhigherincome andfood-securedfamilieswereidentifiedaspotentialdeterminantsofthehighdietary diversity of participants. Having higher secondary education, being job holders, and being from families with a size of below five were more likely to be physically active compared to others. Conclusion: In light of these findings, there is a need to support existing policies and devise new policies targeting these variables. Improving awareness on the impor-tance of dietary diversity during pregnancy, incorporating counseling services, ensur-ing household food security especially during this pandemic situation, and providing women with diet counseling are recommended to increase pregnant women’s dietary diversity and physical activity level.


Dietary diversity during pregnancy
Nutritional requirements are higher during pregnancy than other periods of life in order to meet the metabolic and physiological needs of the mother and the growing fetus. 1,2 Proper nutrition during pregnancy is a crucial determinant of the growth and development of infants as well as the long-term health outcomes of mothers. 3,4 However, pregnant women are more vulnerable to suffer from nutritional deficiency due to factors such as physiological vulnerability, inadequate education, social or cultural traditions, and household workloads. [3][4][5] Malnutrition during pregnancy can permanently affect the physiological development of the fetus, increase the risk of intrauterine growth restriction, low birth weight, preterm delivery, and maternal morbidity and mortality. 6 Thus, nutritional improvement during pregnancy is necessary to decrease the global burden of maternal and child malnutrition, morbidity, and mortality. 7 Previous scientific studies suggest that the adequacy of micronutrient intake through diet can be measured by the dietary diversity (DD) scores of pregnant women. 8 Measuring DD score (DDS) is relatively simple, easy to apply, and can reflect the micronutrient intake of pregnant women. 9 When a pregnant woman's diet is not from diverse food groups, it will be deficient in essential nutrients and as a result, the fetus will not be provided the nutrition required for its healthy growth. 10 Nutritional deficiency and poor nutritional status of pregnant women are widespread problems in lower-and lower-middle-income countries. 11 The prevalence of undernutrition among females in Bangladesh is higher, and dietary intake from diversified food groups is lower, compared to other developing countries. 12 In Bangladesh, the leading causes of lower DD include low family wealth index, lack of education, and inability to afford food due to periodic unemployment. 13 A longitudinal study on maternal DD conducted in rural Bangladesh reported that the DDS declined with comparatively worse household food security, an association that can be best explained by the wealth index of the family. 14 It has been reported that having a financial crisis is linked with lower intake of various micronutrients among pregnant women in Bangladesh. 15

Physical activity during pregnancy
Adequate physical activity (PA) during pregnancy is considered to be safe and beneficial for both the fetus and pregnant woman and may improve pregnancy outcomes. 16,17 PA can reduce the risk of preeclampsia, 17 pelvic pain and back pain, 18 gestational weight gain during pregnancy, 19 anxiety and depressive symptoms, 20

Objectives
The objectives of this study were (a) to assess the DD and its associated factors, and (b) to investigate the level of self-reported PA of pregnant women and factors that affect PA at an antenatal care setting in Khulna City, Bangladesh.

Study design and setting
An institution-based cross-sectional study was conducted in a Mother

Study participants
The study participants were pregnant women who visited the Mother

DD of pregnant women
The DDS for pregnant women was assessed using 24-h recall, a method that was also used by another study among Bangladeshi women. 30 Participants were asked about consumption status (consumed or not con- used to classify each activity by intensity. 33

Household food security
The household food security of the participants was measured using the same 5-question scale that was used in a previous study among pregnant women in Bangladesh. 34 Participants were asked regarding their past 12 months' food intake based on the following five terms: (a) eating a full-stomach meal, (b) skipping an entire meal, (c) eating less food due to lack of food, (d) eating wheat instead of rice, and (e) borrowing food from neighbors or relatives. A qualitative score of 0 to 3 was attached to each question, with 0 being the most food-secure response.
By summing up the score of each question, participants' household food security status was assessed. A household was considered food secure if the score was 0 and food insecure if the score was greater than or equal to 1.

Statistical analysis
Data were checked, cleaned, and analyzed using SPSS version 23.0 and STATA version 16.0. The DDSs were dichotomized as category 1 for those who had high DD and category 2 for those who had low DD. Similarly, the PA variable was also dichotomized as category 1 for those who were physically active and category 2 for those who were physically inactive. Descriptive analyses were done to determine means, frequen- by the odds ratio (OR) with 95% confidence Interval (CI). Variables with a p-value < 0.05 were considered to be statistically significant.

Ethical approval
The research protocol was reviewed and approved by the Research

Socio-demographic characteristics, DD, and PA status
Among the study participants, the mean age was 32.   were housewives were physically inactive, however, most of the pregnant women engaged in a job (80%) were physically active in the study (Table 3).

Factors associated with DD of pregnant women
Our study found that a major proportion of pregnant women had sufficient DD intake. In the present study, the prevalence of high DD was higher than that of previous studies conducted in Ethiopia (43.6%), 35 Kenya (20%), 36 and Ghana (46%). 37 This difference might be due to their study methodology, mainly variation for the food group involved and its food category, in that our study contains seven food groups with two categories, and other studies contained different numbers of food groups and different numbers of categories. Additionally, different geographical locations and seasonal variability might be other reasons for the difference, as in an impoverished rural area in Ghana, where recurrent drought and flood occurred. 37 Beyond this, variation in different demographic and socio-economic characteristics, as well as reporting bias (due to self-reporting), might be the possible reasons for this incompatibility. 38 The simple logistic regression model of the present study demonstrates that the level of maternal education was significantly positively associated with DD, with a higher level of education associated with TA B L E 3 Chi-square analysis showing the distribution of physical activity level based on different socio-demographic characteristics of study participants -p values less than 0.2 were bolded.
higher DD. This finding is in accordance with studies conducted in Kenya, 39 Tanzania, 40 and Ghana. 41 A previous study also reported that women with higher education tend to intake a diet containing variety of food groups. 42 A possible explanation for this is that women with higher education may be more likely to understand educational messages about nutrition delivered by the media and know the importance of a diverse diet. Additionally, educated women have better employment opportunities and higher income, which can further improve their household food security status and thus increase consumption of diversified food. 43 According to this study, pregnant women's DD was associated with the wealth status of their households. The odds of high DD increase with the monthly income of the families. This finding supports the studies conducted among pregnant women in Bangladesh, Ghana, and Kenya. 37,39,44 This reason could be that households with higher wealth status might have access to buy a variety of foods and thus, the dietary intake of pregnant women in those households will come from diversified food groups.
We also found that pregnant women from food-secure households were more likely to have higher DD scores, compared to those from food-insecure households. A previous study from Malaysia revealed similar findings. 45 Another Bangladeshi study also reported a positive association between food security and DD of pregnant women. 44 Food security increases the accessibility of adequate and diversified food due to the economic status of respondents. Previous studies also reported that individuals who had food insecurity and low socioeconomic status are more prone to low diversity diets. 46

Factors associated with PA of pregnant women
More than half (58.3%) of the pregnant women in this study physically active. This is lower than the study findings in the United States 49 and Ireland, 50 which were 77.5% and 78.5%, respectively. It is higher than study findings in Nigeria 51 and Brazil, 52 which were 49.0% and 30.8%, respectively. The discrepancies in the level of PA between other studies and ours may be due to differences in sample characteristics, and tools used for measurements. 32 Women with no formal education or below primary education were more likely to be inactive in the present study. Similar findings were also observed in the earlier studies from Australia, 53 Brazil (Nascimento et al., 2015 54 ), and Canada. 55,56 Another study also argued that pregnant women with higher education tended to be more active during pregnancy than women with lower education. 49  Employed women were more likely to be physically active during pregnancy than housewife women. This finding is in line with that of earlier studies in Iran 57 and the United States. 58 The observation that half of the pregnant women of this study were housewives, accounted for lower participation in the occupational activity. In addition, most of the pregnant women in Bangladesh usually take maternity leave that may also make them spend lower time at the occupational activity. In addition, most pregnant women feel more comfortable and safer doing household activities than engaging in occupational or sports activities during pregnancy. 32

Limitations
Since this is a cross-sectional study, a causal relationship cannot be established. In addition, the figure of the DD practice, as well as the PAs and their intensity, might be influenced due to biases associated with self-report measures, such as recall bias and social desirability bias.
Furthermore, this study did not address the availability of food in markets and the extent of partners' support, which may also influence their dietary practices and PAs.

CONCLUSION
Mother's educational attainment/level and occupation, household

ACKNOWLEDGMENTS
The authors would like to thank the Director, Dr. Rougon Ara Begum, for her kind cooperation to conduct the study in the healthcare center.
The authors would also like to thank the participants who participated voluntarily in the study. We are also thankful to Anna Utsinger (A native English speaker from the United States) for reviewing the manuscript in order to remove any grammatical error. This research received no specific grant from any funding agency in the public, commercial or notfor-profit sectors.

CONFLICT OF INTEREST
The authors declare no conflict of interest.