Healthy eating and drinking
Consumption of beef, chicken, tilapia, fruit, bottled or filtered water, milk, juice, and decaffeinated beverages were considered to be generally safe and healthy for the developing fetus. While recent risk factors associated with meat products have been described , the beneficial dietary supplementation of these choline-rich foods has been judged to outweigh the risks associated with hormone and antibiotic levels in certain meat and diary products. Likewise, high-sugar content in certain types of juices may be associated with the onset of gestational diabetes; however, the vitamins and nutrients in juices merit their inclusion in the healthy drinking category. The American Dietetic Association recommends, for pregnant women, a healthy diet in accordance with the Dietary Guidelines for Americans (2005) including a variety of daily grains, milk products, fruits, vegetables, and iron-rich meats [1, 35]. The majority of pregnant women reported eating healthy foods including tilapia, beef, chicken, or pork, fresh fruit, milk, and juice during their pregnancies. The frequencies at which women in our sample ate these foods did not strictly adhere to daily recommendations. Only a third of women, for example, reported eating fruit 7 or more times a week.
Unhealthy consumption habits
Consumption of tuna, salmon, canned goods, sugary desserts, fast foods, and drinking of tap water, caffeinated beverages, and alcoholic beverages during pregnancy have been deemed unhealthy due to the appearance of environmental toxins found to have harmful effects in the developing offspring.
The bioaccumulation of methylmercury (MeHg) in marine life, particularly tuna, presents a threat for developing fetuses whose mothers frequently eat this fish during their pregnancies, particularly because it is thought that mercury accumulates more readily in the fetal brain than in the maternal brain, interrupting patterns of cell fate, proliferation, migration, and neural outgrowth [36–38]. Because young children cannot metabolize mercury at the same rates as adults, exposure through either maternal or childhood consumption is of great concern. Even low levels of tuna consumption in children can readily result in blood mercury levels that exceed the heath limit . Epidemiological studies with cohorts from fish-eating populations have found that prenatal exposure to methylmercury has been associated with a myriad of developmental deficits involving attention, verbal learning, visuo-spatial and motor function, and delayed performance [25, 40]. The FDA reports that mean mercury levels of tuna range from 0.128 ppm (light canned tuna) to 0.689 ppm (big-eye tuna). However, mercury levels are highly variable and the FDA reported canned tuna to contain as much as 0.889 ppm for light canned tuna, and 1.816 ppm for big-eye tuna . Currently, the FDA and the Environmental Protection Agency (EPA) recommend that pregnant women eat no more than six ounces of tuna per week. However, tests performed by Consumer Reports on new samples of white tuna revealed that eating only 2.5 ounces of any of the new samples of white tuna would cause pregnant women to exceed the daily mercury levels that the EPA considers safe . Worryingly, more than half of the women (52%) in this study reported consuming tuna during their pregnancies, suggesting that pregnant woman are generally not aware of risk associated with tuna consumption.
Polychlorinated biphenyls (PCBs) are lipophilic compounds found in fatty tissues of marine life feeding in contaminated waters. Staggering levels of PCBs have been found in farmed salmon compared to their wild-type counterparts, and contaminated commercial salmon feed has lead to the bioaccumulation of these dangerous compounds . Prenatal exposure of PCBs has been linked to lower birth weights, smaller head circumferences, and abnormal reflex abilities in newborns, as well as to mental impairment in older children [26, 44–46]. The FDA limits limiting PCB residues in fish to 2 ppm . Although epidemiological studies seem to suggest PCB exposure is related to poor outcomes in neurodevelopment, the precise exposure patterns leading to these deficits have not been characterized . However, a recent meta-analysis of 12 European studies found an association between fetal growth and PCB exposure at low, clinically-relevant levels .
More than a quarter of the women (25.5%) surveyed reported consuming salmon, a commonly ingested source of PCBs. Although prenatal fish intake, including tuna and salmon, can be a good source of Docosahexaenoic acid (DHA), a fatty acid thought to be beneficial in development, fetuses may be at risk for adverse outcomes, and pregnant women should be advised to be selective about which fish they choose to consume, or seek supplementation with fish oil.
Bisphenol A (BPA) found in the lining of metal cans used for food represents a danger to developing fetuses whose mothers consumed a diet high in canned foods. BPA has received recent attention as a controversial ingredient in child sippy cups, baby bottles, and reusable water bottles, leading to a ban by the FDA on the use of the plastic additive in sippy cups and baby bottles [49, 50]. Perhaps less well-known are the dangers of BPA exposure in consuming canned foods. Leaching of BPA from the epoxy resin of metallic food cans has been demonstrated in many studies [49, 51, 52]. Prenatal exposure to BPA has been found to be associated with higher externalizing scores (hyperactivity and aggression) in two-year-old females and reproductive effects in rodent models [27, 28]. The majority of women surveyed (73.9%) reported consumption of canned foods during their pregnancies, with 11.9% reporting consumption at least 4 times a week, suggesting women may not be aware of the dangers of BPA exposure from canned foods. Although epidemiological studies concerning prenatal BPA exposure are lacking, animal studies warn of the detrimental effects of BPAs. Children of pregnant women maintaining a diet high in canned foods are at risk for adverse postnatal outcomes. This study has found that low income is inversely correlated with canned food consumption, suggesting that women of low SES in particular may be especially at risk.
Frequent consumption of sugary desserts during pregnancy may contribute to increased likely-hood of gestational diabetes mellitus (GDM), a condition of glucose intolerance that has been implicated in many pregnancy problems including macrosomia, large for gestational age (LGA) infants, and increased rates of cesarean delivery [53–57]. Hispanic women in particular are two and half times more likely than non-Hispanic whites to suffer from GDM . Increased sugar intake among pregnant adolescents has been linked to maternal gestational diabetes and LGA infants [57, 59, 60], and problems with gestational glucose control have been associated with neural tube defects . On the other hand, carbohydrate restriction has been shown to aid in maternal glycemic control, alleviating some of the adverse pregnancy outcomes seen in patients with GDM . More than a third of women (37%) ate sweet desserts more than one time per week, and 13.3% ate desserts more than 4 times per week. Children born of women maintaining a diet high in sweet desserts are at risk of macrosomia and postnatal obesity. Of course, women with GDM should be advised to closely monitor their sugar intake during pregnancy.
The past decades have seen an insurgence of reliance upon high-energy, low nutrient foods that correlate with rising rates of obesity. A prenatal diet high in fast food represents a possible danger to the developing fetus, as these foods usually contain high levels of fat and salt. Maternal diets high in fat have been associated with increased likelihoods of postnatal diet-induced obesity in offspring, and have the potential to influence epigenetic markers leading to altered postnatal gene expression and eating behavior [62, 63], Prenatal diets high in sodium levels have been linked to decreased gestational weight gain and an increased responsiveness to stress in adults [64, 65]. More than a quarter of women surveyed (25.7%) report eating fast foods at least once a week during their pregnancies. An alarmingly high percentage of surveyed women reported consuming fast foods more than four times per week (7.7%), and are at heightened risk for adverse fetal effects of high maternal salt and fat diets. Additionally, Hispanic populations are at risk for higher fat intakes from dairy foods . Based on the literature, prenatal advising should stress the importance of eating a healthy diet low in these energy-dense, nutrient low foods so as to lower future generations’ risk of obesity and stress conditions.
Drinking water has been found to have levels of many prenatal toxins including trihalomethanes, and certain drinking water disinfection by-products (dibromoacetic acids, or DBAs). The Drinking Water Quality Report for Downey, CA warns of 20 different pollutants in the city’s water, with eight chemicals existing at concentrations that exceed the health guidelines set by federal and state agencies: tri - and tetro-chloroethylene (DBAs), alpha particle activity, arsenic, radium 228, lead, radium 226, and combined radium . Many women in our sample (12.1%) reported drinking tap water during pregnancy, suggesting risk for exposure to many of these dangerous contaminants. Contaminants such as arsenic and DBA, and radium 226 have been found to result in central nervous system defects, oral cleft defects, and neural tube defects, and small for gestational age births, and risks for fetal death [67–69]. Women drinking tap water in areas where contamination is exceptionally high, as in Downey, CA, are at risk for adverse outcomes resulting from prenatal exposure to certain chemicals. Pregnant women should instead be encouraged to drink filtered or bottled water.
The spectrum of disabilities associated with prenatal alcohol exposure is termed FASD (Fetal Alcohol Spectrum Disorders). The prevalence of FASD has been difficult for researchers to ascertain. Some studies have found that anywhere from 0.5 - 2 cases per live births . One study found as high as 1 per 100 live births when the full range of FASD was taken into account . There is a large amount of data characterizing the severe complications in children who were exposed prenatally to alcohol. Children born with FASD often exhibit abnormal craniofacial features and have a litany of cognitive impairments including learning disabilities, decreased intelligence, decreased reaction time, slow sensory processing speed, language dysfunction, and behavioral disorders that are direct results of nervous system injury [72–75].
Self-reporting of maternal drinking
Information on the prevalence and pattern of maternal drinking has been notoriously difficult to ascertain. One possible diagnosis tool is the detection of biomarkers at birth. FAEEs (fatty acid ethyl esters) accumulate in the meconium after alcohol exposure in the first and second trimester . Such methods have been useful in detecting alcohol usage in at-risk subjects at birth, but are impractical for use in a comprehensive assessment of average alcohol usage during the entire gestational period. Thus, although biological methods of measuring alcohol usage during pregnancy have been developed, self-reports of maternal drinking remain the most effective methodology for maternal drinking prevalence assessments.
Ten women (5.8%) surveyed in Downey, CA reported drinking some time during their pregnancies. This number is slightly lower than national estimates of maternal drinking from the Centers for Disease Control and Prevention (CDC; 7.6% of pregnant women surveyed) . Differences in rates may be due to the demographic make up of our sample. Maternal drinking rates are highest in white populations of older individuals (35-44 years of age) . This study examined a population of relatively young (89.2% under 35) women who were predominantly Hispanic (87.4%). Methodological differences may also be at play. CDC data is obtained from the Behavioral Risk Factor Surveillance System (BRFSS), a telephone survey system that asks women to report alcohol usage within the last 30 days. Conversely, the current study surveys participants in late pregnancy or early post-pregnancy periods, when subjects might have difficulty remembering nutritional habits from early pregnancy. The chance of reporting inaccurate information from early pregnancy rises with the passing of time, and the stigma associated with maternal drinking might influence participants to deny usage if recall was suspect. However, despite the difficulties in obtaining accurate data through self-report, administration of a survey instrument to this population of women is the best way to obtain information regarding behaviors during pregnancy.
Methyl donors, prenatal vitamins, and alcohol consumption
Research has consistently shown that intake of folate, choline, and other methyl donors are integral to the healthy development of the fetus. Specifically, the metabolism of these nutrients provides methyl groups in one-carbon methylation pathways . Disruptions in one-carbon metabolism may result in decreased cognitive abilities  and serious birth defects . Folic acid is an important contributor of methyl groups for pregnant women. A recent study has suggested that intake of prenatal vitamins may reduce the risk of autism . Fortunately, intake of prenatal vitamins is fairly common: 83.4% of women report supplementing their diets with prenatal vitamins. Additionally, consumption of choline-rich foods was high: all women reported eating meat sometime during their pregnancies, and most reported eating meat at least once per week.
Because prenatal alcohol has been found to disrupt one-carbon metabolism, diets deficient in methyl donors may exacerbate the harmful effects of prenatal alcohol. Although women seem to be making an effort to curb their alcohol usage after recognizing their pregnancy, the number of women who report drinking during the first trimester is worrying (half of the ten women who reported drinking). Research has shown early pregnancy to be especially vulnerable to the teratogenic effects of alcohol, as deficiencies in methyl donor groups in the first month have been shown to result in neural tube birth defects in offspring .
Caffeine is a xanthine alkaloid that can be found in coffee, sodas, energy drinks, and tea. Studies labeling caffeine as a teratogen date back to the late 1960s , and in 1980, the FDA advised limiting intake of caffeine during pregnancy, noting the substance’s association with fetal mortality, birth defects, and decreased birth weights [83, 84]. The American Pregnancy Association recommends 150-300 mg as a safe daily dose of caffeine, although this is only based upon studies concerning risk of miscarriage . Published epidemiological studies noting the impact of prenatal caffeine on child behavioral and cognitive effects have been somewhat scarce and inconclusive [24, 86–88]. Animal studies, however, have found developmental delays, abnormal neuro-motor activity, and neurochemical disruptions, with some effects persisting until adulthood (for review, ). High levels of coffee consumption have been linked to fetal death after the second trimester [90, 91].
Caffeine has been consistently linked to abnormal motor activity and motor development [87, 88]. A majority of the subjects (80%) reported drinking caffeinated beverages during their pregnancies, with 14% of women reporting consumption of more than 4 caffeinated beverages per week. These numbers suggest that some children may be at risk for preventable persistent aberrations in neurochemistry and motor development.
Intake of over-the-counter and prescription medication
The number of women taking over-the-counter and prescription medications during pregnancy has increased within the past few decades . A small number of women report using aspirin (1.6%) and ibuprofen (3.1%) during pregnancy. Animal experiments have implicated aspirin and ibuprofen, both cyclooxygenase inhibitors, in a number of adverse fetal effects including physical malformations [93, 94] and postnatal cognitive deficits .
A small number of women (3.6%) reported taking prescription pain medications during pregnancy. Fetal effects as a result of prenatal opioid exposure are poorly understood, but seem to be related to poor developmental outcomes . Case control studies have made associations between prenatal use of opioid analgesics and congenital heart defects, the primary factor in birth-defect related infant mortalities .
Less than 10% (8.8%) of women reported using prescription anti-nausea medications. Use of anti-nausea medications have been found to be associated with acute non-lymphoblastic leukemia  although some FDA pregnancy category B drugs (considered safe in pregnancy) such as Zofran, are often prescribed for pregnancy related morning sickness. Our study did not differentiate between the pregnancy categories of anti-nausea medications used.
A paucity of conclusive research on the effects of prenatal exposure to prescription drugs may have led to the assumption that they are safe to prescribe. However, studies highlighting the possible dangers of prescription drugs, as well as some over-the-counter medications, suggest otherwise. Use of opioid analgesics and category C and above anti-nausea medications put pregnant women at risk for aversive offspring outcomes. In any case, women should be properly informed as to the possible dangers of prenatal exposure to these medications.
Changing consumption patterns after recognition of pregnancy
Reports of patterns of consumption, that is, information about the amount of a substance consumed and the period during pregnancy in which it was consumed, can provide a window into commonplace beliefs about what habits are healthy during gestation. For example, most women who consumed alcohol during pregnancy reported doing so only in the first trimester. Given that a significant number of women (48.6%) confirm their pregnancies halfway through the trimester, it is likely that these women who report drinking during the first few months are doing so before realizing they are pregnant. In contrast, the majority of women consuming caffeine during pregnancy report continuing their consumption of these beverages well into their second and third trimesters, suggesting that caffeine is not commonly regarded as harmful to the unborn fetus. The implications of this research are two fold: firstly, women of childbearing age hoping to conceive should be advised to eliminate all alcohol consumption, as effects of maternal drinking have dire consequences in the first trimester when the mother may not know she is pregnant. Additionally, every effort should be made during clinical prenatal care visits to inform pregnant women of the harmful effects of environmental toxins that can be readily transferred to the fetus as a result of uninformed, unhealthy consumption habits.
Several problems exist in attempt to extract such sensitive data (such as gestational consumption of alcohol) from participants, with underreporting and recall bias as major hurdles to accurate data collection. The uncomfortable nature of questions concerning maternal drinking may lead to extensive underreporting of alcohol consumption. Underreporting is suspected to be a major issue in the field . Nevertheless, self-reports of maternal drinking remain the most effective tool for sensitive assessments.
Unfortunately, our data was limited to women who agreed to participate in our survey, and we were unable to randomize samples. As survey distribution was offsite at a private medical group in Downey, CA, we do not have information on percentages of women who were not included in the survey. Small sample sizes for other ethnicities such as Whites, African Americans, and Asians made cross-ethnic studies unfeasible. Similarly, we were unable to conduct certain analyses on a low but significant population of women reporting alcohol use during pregnancy. The FBMIQ was designed to be a short 5-minute study assessing percentages of women who may be at risk for certain unsafe consumption habits and as such it does not represent a complete dietary intake assessment. The scope of the survey limited us to frequency and trimester information only in major food categories, thus we were unable to report frequencies of subcategory items (prenatal vitamins versus prescription pain medication, for example). Additionally,although many substances in foods and beverages are thought to be teratogenic at high levels, specific dosages and patterns of exposure leading to adverse outcomes are not yet known. Developmental deficits as a result of prenatal exposure to environmental toxins at clinically-relevant levels, along with reviews and meta-analyses of the current literature, are important areas of future research that will help inform future prenatal guidance protocols.