Abstract
Previous studies have suggested that women have low dietary intake of fruits and vegetables. This study’s objective was to test the effectiveness of a novel nutrition intervention (education about prenatal flavor learning) on increasing fruit and vegetable intake in a group of primarily Latina women at an urban prenatal clinic. Methods: The Harvard Service Food Frequency Questionnaire (HSFFQ) was administered to 2 groups at the same clinic at 2 time points for each group. The first group was a nonintervention, comparison group. The second (intervention) group received specific information about how a pregnant woman’s food choices can influence subsequent solid food preferences of her infant, with encouragement given to increase fruit and vegetable choices. The HSFFQ was administered pre- and post-intervention for this group. Results: Combined fruit and vegetable intake declined from the administration of Q#1 to Q#2 in both the comparison (n = 28) and intervention (n = 31) groups. The decline was primarily the result of a decrease in vegetable intake, but it was not statistically significant. In the comparison group, only 23.3-36.6% of women were eating adequate daily servings of vegetables, and in the intervention group 32.3%-38.7%. In both the comparison and intervention groups, over 74% of the women were eating adequate daily servings of fruit at both time points. Conclusions: In this Latina population of pregnant women, there was no difference in fruit and vegetable intake after receiving education about prenatal flavor learning. These findings suggest that education alone may not be sufficient to change health behaviors.
Introduction
Several studies have noted that food sources of many nutrients are deficient in maternal diets.1-5 Low-income women seem to be particularly vulnerable to deficient dietary intake, especially with regard to fruits and vegetables.1,2,6,7 Although some of this low intake may be attributed to lack of access,8,9 prior research suggests that food choice and nutrition education are also significant factors.10-13 What is not widely known is that what a mother eats during her pregnancy also affects her infant’s nutritional preferences after birth, a concept identified as “prenatal flavor learning.”14,15 In a classic study, Mennella et al. obtained amniotic fluid from women who had undergone routine amniocentesis in the second trimester of pregnancy after ingesting either garlic capsules or placebo. 14 Blinded, independent raters found that maternal ingestion of garlic capsules resulted in a detectable odor of garlic from amniotic fluid (odor is a significant component of flavor). In another study, 46 pregnant women were randomly assigned to consume either 300 mL of carrot juice or water for 4 days per week for 3 consecutive weeks in their last trimester of pregnancy. 15 After the infants’ birth, at 4 weeks after mothers began to supplement their infant’s diets with solid food (cereals), infants were videotaped while eating cereal mixed with water on one test day and cereal mixed with carrot juice on another test day. The videotapes of the feedings were reviewed by independent raters who were blinded to the experimental conditions of the study. The raters were looking for negative facial expressions by the infants in response to each spoonful of cereal offered. Negative facial expressions were chosen for rating, as the authors explain that these are easier to discriminate than positive expressions. Some examples given of these negative facial expressions were nose wrinkling, brow lowering, and head turning. The results demonstrated that those infants who had had prenatal exposure to the carrot flavor exhibited significantly fewer (p = .01) negative facial expressions to the carrot-flavored cereal relative to the plain cereal when compared with infants who had had no prenatal exposure to the carrot flavor. Cell biologists have only recently identified the specific neuroanatomic changes that occur in olfactory bulb glomeruli with in utero exposure to odorants from metabolites of the mother’s diet, and they have also noted subsequent preferences for the odorants by mouse pups. 16
By adding the concept of prenatal flavor learning to nutrition education interventions, we can empower pregnant women to influence their children’s future health and add another potential strategy in the prevention of childhood obesity. Esposito and colleagues note that in utero environmental exposures can have a subsequent effect on childhood obesity.
17
Although there are many factors thought to be contributing to obesity, prior research suggests that a diet high in fruits and vegetables can be of significance in combating obesity, and increasing the percentage of Americans eating adequate fruits and vegetables had been a goal of
Briefel and colleagues found that on a given day, 25%-30% of infants and toddlers consume no fruit, and 20%-25% consume no vegetables. 20 For the Latino community in particular, low intake of fruits and vegetables would be of particular concern. The prevalence of overweight Latino youth has risen dramatically in the past few decades, and Latinos are also known to be at high risk for developing metabolic syndrome and type 2 diabetes.21-23
The primary aim of this study was to determine if education about prenatal flavor learning would increase fruit and vegetable consumption in pregnant women. It was hypothesized that the intervention would result in increased consumption of fruits and vegetables.
Materials and Methods
Subjects
A convenience sample of pregnant women was recruited at an inner-city prenatal clinic. Women were excluded from the study if they did not speak or read English or Spanish. This study received Institutional Review Board approval, and all subjects gave their written informed consent prior to active participation in the study.
Design
Participants from the intervention group completed a baseline (pretest) of the Harvard Service Food Frequency Questionnaire (HSFFQ) after the first prenatal visit. The HSFFQ has been validated for use in populations of low-income pregnant women.24-26 The HSFFQ asks about dietary intake over the past 4 weeks and includes targeted questions about fruit and vegetable intake. It is a 4-page questionnaire that can be self-administered and requires only basic literacy skills. Both English and Spanish forms were used for the study, and the one selected was based on each participant’s language preference.
The intervention consisted of 2 components: (1) general prenatal nutrition education emphasizing the importance of dietary intake of fruits and vegetables, and (2) educational information about how what a woman eats during pregnancy affects her baby’s food preferences after birth. The general prenatal nutrition information was provided through the use of March of Dimes pamphlets 27 (available in English and Spanish), which were reviewed with research participants. In addition to the usual nutrition information provided by the March of Dimes pamphlets, the novel nutrition education intervention included the use of a clinic poster and individual pamphlets (“My Child Loves Healthy Food,” also available in English and Spanish). These educational materials were specifically designed to provide information about prenatal flavor learning and how a mother’s food choices during pregnancy affect her baby’s food choices after he or she is born. Pamphlets were reviewed via 20 minutes of face-to-face counseling by study investigators with individual participants in a private room adjacent to the clinic. The poster was reviewed in a group setting with subsequent placement in the clinic waiting room. Participants then completed the HSFFQ at a minimum of 4 weeks post-intervention. For the intervention group, the pretest questionnaire (Q#1) was administered at a mean of 15.7 (± 3.8) weeks gestation and the posttest questionnaire (Q#2) was administered at a mean of 28.5 (±3) weeks gestation. For the comparison group, Q#1 was administered at a mean of 15 (±4) weeks gestation, and Q#2 at 29.5 (±4) weeks gestation. The comparison group received the usual care of the clinic without receiving information about prenatal flavor learning.
Data and Statistical Analysis
Statistical analyses on aggregate data were performed using the SPSS (Chicago, IL) and Sigma Plot (Systat Software, Inc., San Jose, CA) computer programs. Comparison of means and differences for daily portions of fruits and vegetables were done with 2-tailed, Student
Results
A total of 65 women completed this study. Results were excluded from data analysis if women were found to be carrying twins (n = 3), or if we were unable to obtain complete antenatal data (as a result of leaving the country or transfer to another clinical facility). Demographic data are presented in Table 1.
Population Characteristics
For infant birth weight data, n = 27 for intervention group and n = 25 for comparison group.
For the comparison group, there was a decline in mean combined fruit and vegetable intake from Q#1 (8.4 servings/day) to Q#2 (6.5 servings/day). At the time of Q#1, only 36.6% of women were eating the recommended minimum daily number of vegetable servings. At the time of Q#2, only 23.3% of women were eating the recommended minimum daily number of vegetable servings. However, over 80% of the women were eating the recommended minimum number of fruit servings daily at the time points for both Q#1 and Q#2.
For the intervention group, there was a smaller decline than the comparison group in the mean combined fruit and vegetable intake from Q#1 (8.6 servings/day) to Q#2 (8.3 servings/day); however, the difference was not statistically significant (
Discussion
In this study, combined fruit and vegetable intake declined during pregnancy, although most of this decline was accounted for by a decline in vegetable servings. There were no significant differences as a result of the intervention. Differences may have been significant if the study had been more adequately powered. A meta-analysis of nutrition education intervention research among pregnant women had noted significant self-reported differences in similar dietary measurements when total sample sizes ranged from n = 43 to n = 173, a wide range that made effect size difficult to determine for power analysis. 10
The results of this study seem to suggest that nutrition education alone is not sufficient to change behavior. Cultural preferences may be a very significant factor in fruit and vegetable consumption. 28 Fruit intake remained stable and at an adequate level in both groups. As one participant in our study stated, “We’re from the islands; we eat a lot of fruit.” The women in this study should be commended for eating adequate amounts of fruit. However, for those women who are overweight, obese, or concerned about excessive weight gain, increasing the intake of non-starchy vegetables could help to decrease total caloric intake, assisting them with weight management goals. The mean body mass index for both groups in this study population was in the “overweight”category; however, as demonstrated in Table 1, there were wide standard deviations for body mass index.
Education alone about prenatal flavor learning did not succeed in increasing fruit and vegetable consumption. Further study is needed that would incorporate the concept of prenatal flavor learning with other factors known to influence eating behaviors (such as availability,cost, and culture) to test intervention strategies designed to promote healthful eating for pregnant women.
Footnotes
Acknowledgements
We thank the women who participated in this study and the expert staff of the prenatal clinic, Patricia Chico, Carol Baumeister, and Mariluz Rivera. In addition, we wish to thank Villanova students Carla Cardona (pamphlet design), Lyndsay Escajeda (poster design) Lisa Wetzel-Effinger, Heather Maroszan, and Abdullah al Mahrouqi for their contributions to this project.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
This work was supported by a VERITAS grant from Villanova University’s Office of Research and Sponsored Projects.