1 Background

An imbalance in maternal nutrition during pregnancy has been proven to be associated with low birth weight. Consequently, it is considered a leading factor for neonatal or child mortality and morbidity [1,2,3]. Furthermore, abnormal birth weight might increase the risks of different disorders in adulthood [4]. Deficits in physical or functional growth and intellectual development of offspring or occurrence of behavioral problems were reported to be significantly associated with low birth weight [5, 6]. The World Health Report 2000 stated that 18 million infants with low birth weight were born annually, constituting 14% of total births [7]. This statistic is equal to 11.56% in Iran, emphasizing the necessity for developing corrective strategies by policymakers to reduce contributing pressures imposed on the healthcare system and community as a whole [8].

Literature has found several factors contributing to improper birth weight [9,10,11]. Mother’s BMI, dietary intake during pregnancy, maternal age, physical activity, harmful behaviors such as cigarette smoking or drug use, hypertension, anemia, urinary tract infections, and malnutrition were among the most cited [12,13,14]. Among different factors, study findings agreed on the critical role of adequate maternal nutrition during pregnancy in giving birth to healthy babies with standard anthropometric measures [15]. Evidence suggests that pregnant women have an increasing need to obtain various macro and micro-nutrient-rich foods, particularly those containing iron, calcium, protein, folic acid, and sufficient fruits and vegetables. Proper dietary balance during pregnancy is necessary for adequate fetus growth and provision of adequate plasma volume needed for more significant circulatory needs of maternal organs and placenta [16, 17]. Accordingly, dietary diversity score (DDS) is an indicator of dietary quality status that evaluates the diversity of consumed foods in a given period [18]. This measure is a tool to measure the dietary diversity of pregnant women [19, 20]. Furthermore, as diet is a modifiable factor, it investigates the association between maternal dietary diversity and birth outcomes. A study by Madzorera et al. [21] revealed an inverse association between diet quality and low birth weight [21]. Similarly, a prospective cohort study in rural Ethiopia found a significant correlation between high DDS during pregnancy and a lower risk of low birth weight in offspring [22]. These findings propose that DDS can play a vital role in birth outcomes. However, to our knowledge, there are a limited number of relevant studies, particularly among women in developing countries, who do not fully observe healthy diet patterns [23]. There are also some contingencies between different study findings across many countries that an ideal birth weight is related to healthier diets rich in fruits, vegetables, seafood, and beans [24, 25]. Thus, the association between dietary patterns among mothers and birth weight is still indecisive.

Due to these inconsistencies in dietary patterns of populations living in different geographical areas with diverse cultures or background characteristics, we conducted a research study to investigate the relationship between maternal dietary diversity and neonatal weight among pregnant women referred to health centers in Qazvin, Iran, in 2019.

2 Methods

2.1 Study design and participants

This is a cross-sectional study of 1076 pregnant women referred to healthcare centers and facilities affiliated with the Qazvin University of Medical Sciences from April to June 2019. Permission to conduct the study was approved by the Research Ethics Committee at Qazvin University of Medical Sciences (IR.QUMS.REC.1394.128). We selected 15 facilities among 28 healthcare centers providing prenatal care for a covered population in Qazvin. Then, all pregnant women referred to these centers who were at 28–36 weeks of gestation agreed to participate in the research and were enrolled.

2.2 Eligibility criteria

Women eligible to be enrolled in the study were residents of Qazvin City in an age group of 18–45 years old, had a singleton and low-risk pregnancy, with no history of infertility, chronic physical or mental illness, hypertension, anemia, urinary tract infections, malnutrition, drug addiction or smoking behavior. Those with multiple pregnancies and clinical diagnosis of heart disease, diabetic mellitus, kidney disease, thyroid disease, autoimmune disorder, preeclampsia, placental abruption, and premature rupture of membranes and pregnant women who had hepatitis B or HIV infection were excluded from the study. After excluding women who were not eligible for participation based on exclusion criteria, 1320 qualified ones remained. Of these, 1240 mothers were available for follow-up, and 1076 provided dietary information for the study analysis.

2.3 Data collection

Birth weights and other neonatal anthropometric profiles for 1076 infants were measured and recorded within 24 h of birth. Women's weight and height before pregnancy were extracted from their medical records to calculate pre-pregnancy BMI. Their relative height and weight were measured when referring to the medical center to measure maternal BMI. Data on demographic characteristics and socio-economic factors of pregnant women, including age, number of pregnancies, education, income, and occupation, were obtained through a general information questionnaire. Some questions also belonged to women's health status (such as blood pressure and disease history), which were collected and added to the study results.

We used a food frequency questionnaire (FFQ) to retrospectively assess maternal dietary intake and frequency of eight food categories consumption (including beverages, bread and cereals, fat, vegetables, fruits, meat/fish/liver/chicken, dairy products, and sweets). We conducted the dietary intake assessment during the past 4 weeks and asked all pregnant women to complete the questionnaire during their third trimester of pregnancy [26]. The FFQ contained food type and intake frequency, containing 61 food items [27]. The responses were given into six grades (“never,” “1–3 times per 4 weeks,” “1–3 times per week,” “4–6 times per week,” “1–2 times per day” and “more than two times per day”). The detailed questionnaire's content has been reported and validated in the literature [28,29,30]. Exploratory factor analysis was performed to reduce the 61 dietary food items from the FFQ into manageable items. The rotated factor loadings greater than 0.3 were used as a threshold value for a variable to belong to a latent group [31]. Table 3 depicts the factor loadings of different food items using factor analysis.

To determine maternal DDS, we used a 24-h dietary diversity score to ask the pregnant women to mention the foods they had consumed in the previous 24 h. Therefore, the food items on the FFQ were aggregated into ten groups, as suggested by the Food and Agriculture Organization (FAO) [32]. If a woman had consumed at least one of the food items in each food group, the score of 1 and otherwise the score of 0 was given. Consequently, women were categorized into two groups “low” (DDS < 4) or “high” (DDS ≥ 4) dietary diversity groups [33].

2.4 Data analysis

Research data were analyzed in SPSS software version 20. We used factor analysis with a varimax rotation to determine dietary patterns. Nineteen foods with factor loadings of 0.3 or more were considered to have a significant role in describing the diet pattern of study participants.Means and standard deviations were used to describe continuous variables. The correlations between maternal and newborn infants’ anthropometric measurements were tested by the Pearson test. Furthermore, to compare dietary diversity scores in women with different socio-demographic and health characteristics, a T-test and ANOVA test were used. In all comparisons, differences were considered statistically significant at P < 0.05.

Then, we considered two categories of low (DDS < 4) and high (DDS ≥ 4) dietary diversity and compared birth weight means in the mentioned groups by using an independent Student’s t-test. Finally, the association between DDS and birth weight was investigated after adjusting for baseline differences, including maternal weight gain during pregnancy, hemoglobin level, father’s occupation, mother’s weight, and mother’s BMI. Therefore, a log-binomial model was run, and the adjusted RRs (ARRs) and 95% CI values were reported.

3 Results

3.1 Anthropometric characteristics

Overall, 1076 pregnant women and born infants participated in the research, whose anthropometric characteristics are shown in Table 1.

Table 1 Anthropometric Parameters for Newborn Infants and Mothers
Full size table

3.2 Correlation between maternal and infants’ anthropometric measurements

Findings related to associations between maternal and neonatal anthropometric measurements are depicted in Table 2. Data confirm that maternal weight and height had a significant positive relationship with infants’ BMI. In contrast, no significant correlation was found between other neonatal and mothers’ anthropometric measurements.

Table 2 Correlation between maternal and newborn infants’ anthropometric measurements
Full size table

3.3 Dietary patterns

Data analysis revealed two different dietary patterns (healthy and non-healthy), explaining mothers' dietary habits in the research. The loading factors related to the list of foods are shown in Table 3. Of the 61 foods included in the questionnaire, 40 were descriptive of mothers' diet, including fried potato, sausages, canned fish, oil, salad dressing, salty snacks, cookies, sugar, salt, coffee, and coca were classified in the non-health category.

Table 3 Loading factors for foods using varimax rotation
Full size table

3.4 Women dietary diversity, maternal characteristics and infants’ birth weight

As findings reveal, dietary diversity is significantly related to maternal weight gain during pregnancy, hemoglobin level, father’s occupation, mother’s weight, and mother’s BMI (p < 0.05). Mothers with higher dietary diversity scores were those with normal BMI and normal levels of hemoglobin who had 8–10.9 kg maternal weight gain and were headed by an employed husband (Table 4).

Table 4 Association between dietary diversity, and maternal characteristics
Full size table

3.5 Birth weight of neonates in different dietary diversity groups

The newborns’ mean birth weight of mothers with inadequate diet intake was 2.17 kg, with an SD of 0.17, while those from mothers with adequate diet were 3.34 kg, with an SD of 0.22 (p < 0.05) (Table 5).

Table 5 Birth weight outcomes based on mother’s dietary diversity score
Full size table

3.6 Association between dietary pattern and birth weight

Regression analysis revealed a statistically significant association between dietary diversity score and birth weight. The risk of low birth weight in newborns of mothers with inadequate dietary pattern was 2.6 times higher (ARR = 95% CI 2.6, 1.4, 5.12) compared to the newborns from mothers who had adequate diets (Table 6).

Table 6 Regression analysis of birth weight and dietary diversity score in study participants
Full size table

4 Discussion

In developing countries, dietary diversity has always been considered an essential determinant of diet quality and adequacy among pregnant women. Thus, most dietary guidelines recommended to increase the consumption of various nutrients during pregnancy to improve mothers’ health status and, consequently, infants’ birth outcomes [33, 34]. In this regard, we defined two dietary patterns (including healthy and non-healthy diets) and compared infants' birth weights and some maternal characteristics in the patterns. Out of 61 foods included in the FFQ questionnaire, 40 were descriptive of mothers’ diet, including fried potato, sausages, canned fish, oil, salad dressing, salty snacks, cookies, sugar, salt, coffee, and coca were classified in the non-healthy category. We also identified that maternal weight gain, hemoglobin level, father's occupation, and mother's BMI influenced mothers' dietary patterns. Besides, women headed by employed husbands were more likely to choose a healthy diet, while illiterate mothers with higher BMI levels had significantly higher loading for a non-healthy diet. These findings align with related literature, which found significant associations between dietary diversity and some maternal characteristics, including weight gain during pregnancy and mothers’ hemoglobin levels [19, 35]. Most of the studies agree that diet diversity was significantly associated with sufficient consumption of macronutrients and micronutrients and, as a result, higher levels of energy and nutrient adequacy [36,37,38]. Our findings added that after adjustment for maternal characteristics, mothers in the unhealthy diet group were at higher risk of giving birth to low-birth-weight neonates. We explain that pregnant women who consume diverse diets might receive more essential nutrients crucial for the healthy growth of their newborns [39]. Previous studies also confirmed our findings and highlighted the significant role of diverse dietary foods during pregnancy on the normal birth weight of neonates [40,41,42]. In contrast to our study, some literature did not find a similar association. The small sample size or ignoring the effect of potential confounding factors might have affected the results and led to some variances [43,44,45].

Another important finding of our paper was the significant relationship between maternal factors, including maternal weight gain, hemoglobin level, father’s employment, mother’s weight, and BMI and maternal dietary diversity score. Similarly, Mayén et al. revealed that the high socioeconomic status of pregnant mothers was associated with higher dietary diversity scores [46]. In line with these findings, Kiboi et al. reported that some socioeconomic factors, such as education level, employment status, and household income, are among the key factors in the dietary diversity of pregnant women [47]. Rammohan et al. highlighted that low maternal education and economic status were significantly related to poor dietary diversity [44]. Similar findings were in line with the results and reported that women living in affluent households had better nutritional status than poor ones [48, 49]. According to different study findings, dietary diversity in pregnant women could prevent low birth weight in neonates by influencing gestational weight gain [50]. Mothers who consume diverse and healthy foods are more likely to gain higher gestational weight and deliver babies with higher weight [51, 52]. In addition, dietary diversity reduces the incidence of nutritional deficiencies in mothers, sequentially resulting in improved fetal growth [53, 54].

After adjusting possible covariates, our study revealed that the risk of low birth weight was 2.6 times higher in mothers with insufficient dietary diversity intake. Our result is consistent with previous literature [55,56,57]. The prevalence of low birth weight in African countries was the highest owing to their considerably disadvantaged nutritional status [58]. Another study conducted in India affirmed the findings and declared that pregnant women with a low intake of fruit and vegetables were more likely to give birth to low-weight babies [57]. Accordingly, Kjøllesdal and Holmboe-Ottesen [56] mentioned that a maternal diet containing oil, high-fat meat, and sugar was associated with low birth weight [56]. Generally, the literature confirmed that unhealthy and poor-quality diet is significantly associated with adverse physical conditions of infants at birth and even during their growth [55,56,57,58]. Some studies disagreed with the results and reported that more consumption of fruits, vegetables, and beans during pregnancy did not affect birth weight [42]. Similarly, another study reported no association between maternal macronutrient intake during pregnancy and infant birth weight [42]. These variances might be due to differences in study designs, sample size, analysis methods, socioeconomic status, and cultural disparities.

4.1 Rigor of study

Although diet quality has a vital role in improving maternal characteristics, there have been few studies on the dietary patterns of pregnant women in different regions of Iran. Thus, conducting such kinds of studies highlights rational relationships between dietary patterns and some of the maternal factors that ultimately affect a society's birth outcomes.

The current study has several limitations. The retrospective approach of the research in collecting data regarding pregnancy dietary intakes might have led to reminding biases unavoidably. To control the limitation, researchers emphasized the importance of study objectives while asking the participants about their dietary intakes and highlighted the necessity for credible data for analysis and subsequent decision-making. Focusing on a limited period for collecting dietary data from pregnant women could also influence the precision of study results. Thus, conducting similar studies with randomized control trial designs would be beneficial to obtain the most relevant and reliable evidence of causal relationships.

5 Conclusion

Our findings revealed that diversity in dietary intake among pregnant women plays a significant role in the healthy growth of neonates and their birth outcomes. Insufficient intake of dietary diversity during pregnancy can potentially increase the likelihood of low birth weight in newborns. Thus, we recommend that health system authorities prioritize the planning for healthy nutrition of pregnant mothers in different geographical regions, specifically remote, rural, and deprived areas. Likewise, nutritional recommendations should be integrated into prenatal caregiving to provide pregnant women with information, knowledge, and attitudes.