Abstract
Background:
Two randomized trials found women with low blood docosahexaenoic acid (DHA; an omega 3 fatty acid) had fewer early preterm births (<34 weeks gestation) if they were assigned to high dose DHA supplementation, however, there is currently no capacity for clinicians who care for pregnancies to obtain a blood assessment of DHA. Determining a way to identify women with low DHA intake whose risk could be lowered by high dose DHA supplementation is desired.
Objective:
To determine if assessing DHA intake can identify pregnancies that benefit from high dose DHA supplementation.
Study Design:
This secondary analysis used birth data from 1310 pregnant women who completed a 7-question food frequency questionnaire (DHA-FFQ) at 16.8 ± 2.5 weeks gestation that is validated to assess DHA status. They were then randomly assigned to a standard (200 mg/day) or high dose (800 or 1000 mg/day) DHA supplement for the remainder of pregnancy. Bayesian logistic regressions were fitted for early preterm birth and preterm birth as a function of DHA intake and assigned DHA dose.
Results:
Participants who consumed less than 150 mg/day DHA prior to 20 weeks’ gestation (n=810/1310, 58.1%) had a lower Bayesian posterior probability (pp) of early preterm birth if they were assigned to high dose DHA supplementation (1.4% vs 3.9%, pp=0.99). The effect on preterm birth (<37 weeks) was also significant (11.3% vs 14.8%, pp=0.97).
Conclusion:
The DHA-FFQ can identify pregnancies that will benefit most from high dose DHA supplementation and reduce the risk of preterm birth. The DHA-FFQ is low burden to providers and patients and could be easily implemented in obstetrical practice.
Keywords: Docosahexaenoic acid, diet, prenatal supplements, pregnancy, preterm birth
Introduction
Each year approximately 15 million infants in the world are born preterm. Preterm birth (PTB) is the primary cause of infant mortality worldwide; and infants who survive have a higher risk of child disability and high societal cost compared to infants born at term (1). Available treatments for reduction of PTB are extremely rare, but a recent Cochrane Review (2) concluded there is strong evidence omega-3 fatty acids, especially docosahexaenoic acid (DHA), reduce early PTB (EPTB, <34 weeks gestation) and PTB (<37 weeks gestation) by 42% and 11%, respectively. Most of the trials that informed the results provided at least 500 mg/day of docosahexaenoic acid (DHA) (2).
At the time the Cochrane Review was published, randomized trials to determine if DHA during pregnancy could reduce EPTB were ongoing in Australia and the United States (US). In contrast to trials included in the Cochrane Review, both were conducted after prenatal DHA supplements were marketed in those countries. Both trials found that randomization to high dose DHA, 800 mg/day vs placebo (3) or 1000 mg/day vs 200 mg/day (4), reduced EPTB in participants with low baseline blood DHA. Participants with high DHA levels in their blood prior to 20 weeks’ gestation, the majority of whom were taking a prenatal supplement with DHA, had a low rate of EPTB and did not benefit from more than 200 mg/day of DHA (4), the amount recommended for pregnancy from seafood or a supplement (5–7). While the evidence is therefore strong that high dose DHA lowers the risk of EPTB in women with low DHA status early in pregnancy, the need to obtain a blood measure of DHA status to identify them is a significant barrier to clinical implementation.
A DHA Food Frequency Questionnaire (DHA-FFQ) with only 7 questions about food and supplement intake is a valid indicator of blood DHA (8) and can be reliably completed by women at their first prenatal visit (9). Collecting DHA intake would be a lower burden to patients and providers than a blood measure of DHA. In this study, we asked if DHA intake at baseline alone could identify pregnancies for which high dose DHA supplementation lowered risk of EPTB and PTB. We used the results from two randomized clinical trials of DHA supplementation during pregnancy in which participants completed the DHA-FFQ before randomization to 200 mg/day or high dose DHA, i.e., 800 (10) or 1000 (4) mg/day. Both trials were funded by the Eunice Kennedy Shriver National Institute of Health and Human Development and conducted in the US between 2016 and 2021 (ClinicalTrials.gov: NCT02626299 and NCT02709239) (4,10).
Materials and Methods
Participants
Participants were enrolled in one of two randomized clinical trials that assigned them to either 200 mg/day DHA or a higher dose of 1000 mg/day (Assessment of DHA on Reducing Early Preterm Birth or ADORE) (4) or 800 mg/day (Prenatal Autonomic Neurodevelopmental Assessment or PANDA) (10) from 16.8 ± 2.5 weeks gestation until delivery. There were 1400 participants enrolled in the two trials, however, 85 were lost to follow-up and 5 did not have a valid assessment of DHA intake, leaving an n=1310. The primary outcomes of both trials are published, and each report includes a CONSORT diagram (4,10). Both trials were registered with ClinicalTrials.gov (NCT02626299; NCT02709239), approved by the University of Kansas Medical Center Institutional Review Board (STUDY00003455 and STUDY00003792), and conducted between 2016 and 2021. The participants in both trials were similar in age (mean of 30.2 years in ADORE and 30.3 in PANDA). However, Black race was slightly different in the two trials (22% in ADORE and 13% in PANDA). Combining the two studies was appropriate because both studies were operated out of the same DHA lab at KUMC with the same protocols such as clinical trials management system, DHA-FFQ procedure, and RBC analysis. Participants provided written consent for their respective study, were 18 years of age or older with a singleton pregnancy and were enrolled before 20 weeks’ gestation. Gestational age at baseline was determined in both studies as previously reported (11). The medical record was used to determine birth dates.
DHA food frequency questionnaire (DHA-FFQ)
The DHA-FFQ consists of 7 questions and was developed to assess intake of DHA from foods and supplements (12). It has good validity in adults (12,13) and pregnant women (8). In both randomized trials, the DHA-FFQ was administered at the baseline visit in the language preferred by each participant (English or Spanish). The 7 questions estimate average daily DHA intake from food and supplements in the previous 2 months. Because the questions only ask about intake of the major contributors to DHA intake (seafood, eggs, poultry, liver and supplements), the assessment time is greatly reduced compared to FFQs that assess total intake. Moreover, we found that only 71% of participants reliably completed the National Cancer Institute’s Diet History Questionnaire-II (847/1191) whereas we obtained DHA intake for 96.8% of participants with the DHA-FFQ (1355/1400). The DHA-FFQ is published in English (12). We created a version in Spanish (8) using appropriate cultural and linguistic adaptions (14,15).
Statistical Analysis
Participants were analyzed by assigned DHA dose. We utilized a mixture of three normal distributions to model gestational age at birth as a continuous time-to-event value (16). We utilized the continuous data to dichotomize EPTB/PTB and to model EPTB/PTB in subgroups with low (<150 mg/day) and high (≥150 mg/day) DHA intake based on results of the DHA-FFQ. The subgroups for intake were determined by an analysis that showed women who were consuming more than 150 mg/day of DHA did not benefit from more than 200 mg/day with a low rate of EPTB. We used this same intake as the cut point in analyzing for the effect of high dose DHA on PTB. Because this was a secondary analysis, no power analysis was conducted.
Two Bayesian logistic regressions were fitted as a function of baseline DHA intake from the DHA-FFQ and dose assignment (high dose [800/1000 mg] or low dose [200 mg]). The first Bayesian logistic regression treated the outcome as EPTB. We used sqrt(DHAFFQ) for computational convenience as DHAFFQ gave some numerical instability in the Bayesian model. We explored whether DHAFFQ/100 would help with numerical stability and there was no appreciable difference, so we proceeded with sqrt(DHAFFQ) as input to the Bayesian logistic regression. The rate of EPTB logistic regression allowed the impact of the square root of DHA-FFQ to be different depending on the dose assigned. This is written as logit(EPTB)=B1+Gamma1*sqrt(DHA-FFQ) for the high dose group and logit(EPTB)=B2+Gamma2*sqrt(DHA-FFQ) for the low dose group. The B1 and B2 are the intercepts for low and high doses and Gamma1 and Gamma2 are the respective slopes for sqrt (DHA-FFQ). The intercepts had normal priors that were fairly noninformative and had negligible influence on the final inference each [mean −3.5 and standard deviation 1.5, corresponding to a prior median EPTB rate of 3.2% and 95% equal tailed credible interval of 0.2% to 40%. The slopes were non-informative and very close to flat centered around 0]. We replicated this Bayesian logistic regression model for PTB.
For the mixture models we created tables and calculated the posterior median of EPTB/PTB and equal-tailed 95% credible intervals by assigned dose for the total sample and by subgroups with intakes <150 mg/d) and ≥150 mg/day. For the total sample and by subgroups we calculated the posterior probability (pp) that high dose (800/1000 mg/day) has lower EPTB or PTB than low dose (200 mg/day). For each of the logistic regressions we calculated the posterior median of EPTB and PTB and equal-tailed 95% credible intervals as a function of DHA-FFQ with separate curves for each dose assigned. We utilized OpenBUGS version 3.2.3 rev 1012 for all Bayesian analyses. All analyses were fitted using 10,000 burn-in draws of Markov chain Monte Carlo, followed by 40,000 draws for inference.
Results
Population characteristics
There were no demographic differences between groups assigned to the low DHA dose (200 mg/day) or the high DHA dose (800 mg/day or 1000 mg/day) in either trial (4,10). The demographic characteristics of participants in the combined trials and the subgroups who consumed <150 mg/day or ≥150 mg/day are shown in Table 1. Most (58%) consumed <150 mg/day of DHA at baseline from diet and supplements. As can be seen in Table 1, women consuming <150 mg) at baseline compared to those consuming ≥150 mg, were more likely to be from racial/ethnic minorities (non-Hispanic/Latina Black and Hispanic/Latina), to have fewer years of formal education, to not take a DHA supplement, to have lower annual household income (<$50,000) and to have lower dietary DHA intake. They were also less likely to achieve a maternal postpartum red blood cell phospholipid DHA >5.5%, the preplanned study criteria for adherence to capsule intake in ADORE (17).
Table 1.
Population Characteristics
| Baseline Characteristic | ALL (n=1395) |
DHA Intake <150mg (n=810 [58.1%]) |
DHA Intake >150mg (n=585 [41.9%]) |
|---|---|---|---|
| Study / Site | |||
| ADORE (site 1)* | 486 (34.8%) | 294 (36.3%) | 192 (32.8%) |
| ADORE (site 2)† | 359 (25.7%) | 187 (23.1%) | 172 (29.4%) |
| ADORE (site 3)‡ | 251 (18.0%) | 191 (23.6%) | 60 (10.3%) |
| PANDA | 299 (21.4%) | 138 (17.0%) | 161 (27.5%) |
| Language of preference | |||
| English | 1237 (88.7%) | 690 (85.2%) | 547 (93.5%) |
| Spanish | 158 (11.3%) | 120 (14.8%) | 38 (6.5%) |
| Birth Data Available? | |||
| Yes, birth data available | 1310 (93.9%) | 754 (93.1%) | 556 (95.0%) |
| Ethnicity & Race § | |||
| Non-Hispanic or Latino | 1112 (79.7%) | 616 (76.1%) | 496 (84.8%) |
| AIAN | 6 (0.54%) | 4 (0.65%) | 2 (0.40%) |
| Asian | 38 (3.42%) | 21 (3.41%) | 17 (3.43%) |
| Bi- or Multi- Racial‖ | 31 (2.79%) | 16 (2.60%) | 15 (3.02%) |
| Black | 277 (24.91%) | 204 (33.12%) | 73 (14.72%) |
| NHPI | 2 (0.18%) | 2 (0.32%) | 0 (0.00%) |
| White | 758 (68.17%) | 369 (59.90%) | 389 (78.43%) |
| Hispanic or Latino | 280 (20.1%) | 192 (23.7%) | 88 (15.0%) |
| Asian | 1 (0.36%) | 1 (0.52%) | 0 (0.00%) |
| Bi- or Multi- Racial‖ | 3 (1.07%) | 1 (0.52%) | 2 (2.27%) |
| Black | 12 (4.29%) | 9 (4.69%) | 3 (3.41%) |
| Other Race | 150 (53.57%) | 106 (55.21%) | 44 (50.00%) |
| White | 114 (40.71%) | 75 (39.06%) | 39 (44.32%) |
| Unknown | 3 (0.2%) | 2 (0.3%) | 1 (0.2%) |
| Black | 1 (33.33%) | 1 (50.00%) | 0 (0.00%) |
| Unknown | 1 (33.33%) | 1 (50.00%) | 0 (0.00%) |
| White | 1 (33.33%) | 0 (0.00%) | 1 (100%) |
| Black (yes/no) ¶ | |||
| No, not Black | 1083 (77.6%) | 583 (72.0%) | 500 (85.5%) |
| Yes, Black | 311 (22.3%) | 226 (27.9%) | 85 (14.5%) |
| Unknown | 1 (0.1%) | 1 (0.1%) | 0 (0.00%) |
| Adherence to Study Treatment # | |||
| n/a | 213 (15.3%) | 146 (18.0%) | 67 (11.5%) |
| Not compliant | 141 (10.1%) | 114 (14.1%) | 27 (4.6%) |
| Compliant | 1041 (74.6%) | 550 (67.9%) | 491 (83.9%) |
| Taking DHA-containing Supplement at Enrollment (yes/no) | |||
| No Supplement | 670 (48.0%) | 593 (73.2%) | 77 (13.2%) |
| Yes Supplement | 725 (52.0%) | 217 (26.8%) | 508 (86.8%) |
| Annual Household Income | |||
| Less than $10,000 | 170 (12.19%) | 143 (17.65%) | 27 (4.62%) |
| $10,000 – $14,999 | 80 (5.73%) | 67 (8.27%) | 13 (2.22%) |
| $15,000 – $24,999 | 151 (10.82%) | 111 (13.7%) | 40 (6.84%) |
| $25,000 – $49,999 | 228 (16.34%) | 154 (19.01%) | 74 (12.65%) |
| $50,000 – $99,999 | 307 (22.01%) | 157 (19.38%) | 150 (25.64%) |
| $100,000 – $149,999 | 269 (19.28%) | 102 (12.59%) | 167 (28.55%) |
| $150,000 – $199,999 | 88 (6.31%) | 28 (3.46%) | 60 (10.26%) |
| $200,000 or more | 68 (4.87%) | 22 (2.72%) | 46 (7.86%) |
| Unknown | 34 (2.44%) | 26 (3.21%) | 8 (1.37%) |
| Baseline RBC DHA (% total fatty acids) | 6.5±1.8 (6.3) | 5.8±1.5 (5.6) | 7.4±1.7 (7.4) |
| PP RBC DHA (% total fatty acids) | 9.1±3.1 (8.9) | 8.5±3.1 (8.1) | 10±2.8 (10) |
| Dietary DHA Intake (mg/d) | 88.5±79.8 (70) | 60.9±36.3 (56) | 126.7±104.3 (111) |
| Supplemental DHA Intake (mg/d) | 72.2±104.6 (11) | 12±24.1 (0) | 155.9±115.3 (190) |
| TOTAL DHA Intake (diet & supplement) (mg/d) | 160.6±134.1 (123) | 72.9±38.7 (71.5) | 282.1±124 (249) |
| Maternal Education (years) | 14.7±3.1 (15) | 13.8±3 (13) | 15.9±2.7 (16) |
| Paternal Education (years) | 14.3±3.1 (14) | 13.5±3 (13) | 15.3±2.9 (16) |
| Age at Enrollment (years) | 30.2±5.5 (30.2) | 29.2±5.6 (29.1) | 31.5±5 (31.3) |
| Parity (each) | 1.1±1.4 (1) | 1.3±1.4 (1) | 0.9±1.3 (1) |
| GA at Enrollment (weeks) | 16.8±2.5 (17.3) | 16.7±2.6 (17.3) | 16.9±2.3 (17.3) |
| GA at Birth (weeks) | 38.7±1.9 (39.1) | 38.6±2.1 (39) | 38.9±1.5 (39.1) |
DHA, docosahexaenoic acid; ADORE, Assessment of DHA on Reducing Early Preterm Birth; PANDA, Prenatal Autonomic Neurodevelopmental Assessment; AIAN, American Indian or Alaskan Native; NHPI, Native Hawaiian or other Pacific Islander; RBC, red blood cell; n/a, not applicable; PP, postpartum; GA, gestational age; NHL, Non-Hispanic or Latino.
Data are n (%) or mean±SD (median).
ADORE-Kansas City.
ADORE-Columbus.
ADORE-Cincinnati.
Ethnicity and Race were reported by the participant. Ethnicity choices included: Hispanic, Non-Hispanic, and Unknown. Race categories included: AIAN, Asian, Black or African American, NHPI, Other, specify, Two or more, specify, Unknown, and White. Those who selected race as Other, specify or Two or more, specify were categorized by the investigators based on the additional details provided. There was one race and three ethnicities reported by the participant as unknown. There were no other missing data points for this variable in this data set.
DHA Intake <150mg: NHL, Asian, White (n=4); NHL, Black, Native American (n=1); NHL, Black, Native American, White (n=4); NHL, Black, White (n=7); Hispanic, Native American, White (n=1). DHA Intake ≥150mg: NHL, Asian, Black (n=1); NHL, Asian, Black, White (n=1); Hispanic, Asian, White (n=1); NHL, Asian, White (n=5); NHL, Black, White (n=7); Hispanic, Native American, White (n=1); NHL, Native American, White (n=1).
Categorized by the investigators based on the participants’ reported race. Includes those from the Bi- or Multi- Racial category who reported their race as partially Black or African American.
Defined by postpartum RBC DHA. n/a = no postpartum RBC sample; Not adherent = postpartum RBC DHA <5.5%; Adherent = postpartum RBC DHA >5.5%.
Early preterm and preterm birth
Table 2 shows the observed proportion of EPTB (<34 weeks) by assignment to 200 mg/day or 800/1000 mg/day, the posterior median percent of EPTB and 95% Bayesian credible intervals, and the Bayesian posterior probability (pp) that 800/1000 mg/day is superior to 200 mg/day. Table 3 shows the same results for PTB (<37 weeks). In both tables, results are shown overall and by baseline DHA intake. Bayesian pp representing the probability the rates of EPTB and PTB rate are lower in participants assigned to the higher doses compared to 200 mg/day. Probability is measured using a scale of probability from zero to 1. A probability of 1 means that the event is certain to happen.
Table 2:
DHA intake estimated by the DHA-FFQ1 and early preterm (< 34 week) birth: Overall and by baseline DHA intake
| Overall Proportion of Births (%) | Bayesian posterior median % (95% CI)2 | BPP3 (1000/800 better than 200 mg) |
|||
|---|---|---|---|---|---|
| 200 mg N = 632 |
1000/800 mg N = 678 |
200 mg N = 632 |
1000/800 mg N = 678 |
||
| Overall | |||||
| Early preterm birth < 34 wk | 14/632 (2.2) | 10/678 (1.5) | 2.6 (1.4–4.0) | 1.6 (0.7–2.8) | 0.89 |
| Baseline DHA intake 1 | |||||
| Low DHA (<150 mg) | 13/357 (3.6) | 6/397 (1.5) | 3.9 (2.3–6.0) | 1.4 (0.0–2.9) | 0.99 |
| High DHA (≥150 mg) | 1/275 (0.4) | 4/281 (1.4) | 0.8 (0.0–2.4) | 2.2 (0.8–4.2) | 0.10 |
Docosahexaenoic acid food frequency questionnaire;
confidence interval;
Bayesian posterior probability
Table 3.
DHA intake estimated by the DHA-FFQ1 and preterm (< 37 week) birth: Overall and by baseline DHA intake
| Overall Proportion of Births (%) | Bayesian posterior median % (95% CI)2 |
BPP3 (1000 and 800 better than 200 mg) |
|||
|---|---|---|---|---|---|
| 200 mg N = 632 |
1000/800 mg N = 678 |
200 mg N = 632 |
1000/800 mg N = 678 |
||
| Overall | |||||
| Preterm birth < 37 wk | 66/632 (10.4) | 54/678(8.0) | 12.6 (10.6–14.8) | 10.0 (8.3–12.0) | 0.97 |
| Baseline DHA-FFQ 1 | |||||
| Low DHA (<150 mg) | 43/357 (12.0) | 34/397(8.6) | 14.8 (12.0–18.0) | 11.3 (9.0–13.8) | 0.97 |
| High DHA (≥150 mg) | 23/275 (8.4) | 20/281(7.1) | 10.1 (7.7–12.9) | 8.7 (6.2–11.7) | 0.77 |
Docosahexaenoic acid food frequency questionnaire;
confidence interval;
Bayesian posterior probability
Overall, there was a high probability that a high dose was superior in reducing EPTB (pp=0.89), but the high dose was clearly superior for participants with an average daily DHA intake <150 mg at baseline (pp=0.99). The Bayesian calculated posterior median percent and 95% credible intervals were 3.9% (CI 2.3–6.0%) and 1.4% (CI 0–2.9%) for the low and high doses of DHA, respectively (Table 2). The observed rates were similar to the Bayesian calculation. On the other hand, participants consuming >150 mg/day of DHA at baseline had a very low rate of EPTB when assigned to 200 mg (0.8%, CI 0.0–2.4%).
There was a high probability (pp=0.97) that the total sample (i.e., independent of baseline DHA intake) benefited from assignment to 800 mg/day or 1000 mg/day DHA compared to 200 mg/day with a lower rate of PTB (12.6% vs 10%, pp=0.97) (Table 3). Participants with a baseline intake <150 mg/day of DHA had a 23.6% reduction in PTB, from 14.8% to 11.3%, pp=0.97, whereas there was a smaller 13.9% reduction for those with a baseline intake ≥150 mg/day, from 10.1% to 8.7%, pp=0.77 (Table 3).
Discussion
Principle findings
DHA intake was estimated using a validated DHA questionnaire (DHA-FFQ) (8) that can be reliably completed without assistance (9) and that could easily be administered in a clinical setting. The DHA-FFQ predicted participants whose risk of EPTB and PTB was reduced by consuming a DHA supplement of 800 or 1000 mg/day compared to 200 mg/day. Participants who started the study with an average daily DHA intake of <150 mg had a 64% lower rate of EPTB and a 24% lower rate of PTB if they were assigned to 800 or 1000 mg/day compared to 200 mg/day DHA. Therefore, our study supports providing pregnant women with low intake with between 800 and 1000 mg/day of DHA during pregnancy. In contrast, participants who were consuming ≥150 mg/day at baseline do not appear to need a high dose of DHA to reduce EPTB. In contrast to EPTB, there is convincing evidence that high dose DHA reduces PTB rates among all women, even those with higher DHA intake at baseline (pp=0.77). This raises the question of whether universal supplementation with a high dose of DHA might be appropriate for all. The major disadvantage of this approach may be the cost of the supplement. However, it is also important to consider other possible benefits of a higher dose of DHA, such as observed for the women who participated in PANDA and ADORE (19,20), and to carefully evaluate the safety data from randomized clinical trials that provided high dose DHA during pregnancy in attempting to answer this question.
Low levels of education and income and perceived stress are associated with lower DHA intake in pregnant women and those of childbearing age (21–26), and lower socioeconomic status is associated with higher rates of PTB (27). The association with EPTB is evident here for participants assigned to 200 mg/day DHA who started with an average daily intake <150 mg/day. Compared to those who started with an intake ≥150 mg/day, they had a nearly 5-fold risk of EPTB (3.9% vs 0.8%, respectively). Arguably the most important finding of this study is that the disparity in risk of EPTB associated with lower socioeconomic status disappeared when these same participants were assigned to high dose DHA supplementation. In contrast, although their risk of PTB was significantly reduced by assignment to high dose supplementation (from 14.8% to 11.3%, pp=0.97), evidence of disparity related to lower socioeconomic status remained as women with higher baseline DHA intake had dose-equivalent PTB rates of 10.1% and 8.7%.
The mechanism by which DHA reduces EPTB and PTB is unknown. Most theories center around the known properties of DHA metabolites to reduce or resolve inflammation (28). Results from one of these trials (4) are evidence that 1000 mg/day compared to 200 mg/day of DHA modified sRAGE and IL-6, immunoregulatory factors involved in the initiation of parturition (29). Although it is outside of the scope of this project, future research is necessary to elucidate the mechanisms underlying the significant effect of DHA on EPTB and PTB.
Clinical implications
A modified version of the DHA-FFQ administered in an electronic data capture system (REDCap) (30,31) prior to the first prenatal visit had a sensitivity of 92.1% for identifying DHA intake <150 mg/day compared to the interview method used in these trials (9). It is now standard of care for women in our prenatal clinics at the University of Kansas Health System to complete the survey in REDCap prior to their first prenatal visit. The result triggers a message for the patient and their provider regarding DHA supplementation that depends upon their average daily DHA intake. The link is an example of the online survey, which could be easily implemented by other providers of prenatal care: https://redcap.kumc.edu/surveys/?s=XLP7DJDWF4
Strengths and limitations
A strength of the study is that women were randomly assigned and provided the doses of DHA that we evaluated. Another strength is that there were no demographic differences between groups assigned to the low DHA dose (200 mg/day) or the high DHA dose (800 mg/day or 1000 mg/day) in either trial. Because of this, we are confident in our finding that those with low DHA intake in early pregnancy benefited from a higher DHA dose with a lower EPTB rate despite having different maternal characteristics from high baseline consumers. Another strength is that most participants (58%) in these trials were consuming <150 mg/day of DHA at baseline. This was fortuitous as had the trials been conducted in a population with high intakes of seafood or use of prenatal DHA supplements, it is unlikely we would have found an effect of DHA dose on EPTB and PTB.
A limitation is that although the data were collected prospectively, this is a secondary analysis. Trials designed to test the effects of DHA dose on low DHA consumers in pregnancy are needed. The DHA-FFQ offers a method to identify those participants.
Conclusion
A simple, easy to administer 7-question survey (DHA-FFQ) identified pregnant women with low DHA intake in two randomized clinical trial and found that their risk of EPTB and PTB was reduced by 64% and 24%, respectively, if they were assigned to a high dose DHA supplement (800 or 1000 mg/day) compared to 200 mg/day. The DHA-FFQ also identified women who did not benefit from high dose DHA supplementation in respect to EPTB and PTB (i.e. women consuming ≥ 150 mg/day). Our results do not question the advice for some women to supplement with 200 mg/day of DHA during pregnancy, however, they offer strong evidence that 200 mg is not sufficient for women who have a low intake of DHA in early pregnancy, where low intake is defined as <150 mg/day. In conclusion, the DHA-FFQ identifies women who could benefit from high dose DHA supplementation at least as effectively as a blood measure of DHA (4) but with far fewer barriers for clinical implementation.
Figure 1 illustrates the posterior median EPTB (<34 weeks) rate and 95% credible interval by dose and DHA intake (DHA-FFQ) including all participants. The reduction in EPTB in participants assigned to the higher dose can be determined from the figure and range from 20% to 80%, with the greatest reduction in participants who had the lowest baseline DHA intake.
Figure 1.
Superiority of 800 or 1000 mg DHA compared to 200 mg DHA to reduce early preterm birth (EPTB <34 weeks gestation) among participants who reported low DHA intake from food and supplements at baseline (results from the 7-question food frequency questionnaire). The solid lines represent the posterior median rates of EPTB and the dotted lines represent the 95% credible intervals. See results for early preterm birth by DHA intake <150 mg and ≥150 mg day in Table 2.
Figure 2 illustrates the posterior median PTB (<37 weeks) rate and 95% credible interval by dose and DHA intake (DHA-FFQ) including all participants. Among participants who consumed less than 100 mg/day of DHA at baseline, assignment to the higher dose resulted in a 30% reduction in PTB. Those who consumed between 100 and 150 mg/day at baseline had a 20% reduction.
Figure 2.
Superiority of 800 mg or 1000 mg DHA compared to 200 mg DHA to reduce preterm birth (PTB <37 weeks gestation) among participants who report low DHA intake from food and supplements at baseline (results from the 7-question food frequency questionnaire). The solid lines represent the posterior median rates of PTB and the dotted lines represent the 95% credible intervals. See results for preterm birth by DHA intake <150 mg and ≥150 mg day in Table 3.
Acknowledgement
We thank the participants who participated in these trials during pregnancy and a very large group of support staff who were responsible for recruiting participants and carrying out the large number of duties needed for the conduct of the randomized clinical trials.
Funding statement
This work was supported by grants from The Eunice Kennedy Shriver National Institute of Health Child Health and Human Development (NICHD) (R01HD083292, R01HD086001) and the National Institutes of Health Office of Dietary Supplements (R01HD083292-03S1). The NICHD had no role in study design, data collection, data analysis, data interpretation, or writing of this article. Life’s DHA™-S oil, DSM Nutritional Products LLC, Switzerland donated the investigational capsules for both trials but had no role in study design, data collection, data analysis, data interpretation, or writing of this article.
Abbreviations
- ADORE
Assessment of DHA On Reducing Early Preterm Birth
- EPA
eicosapentaenoic acid
- EPTB
early preterm birth
- DHA
docosahexaenoic acid
- FFQ
food frequency questionnaire
- NICHD
Eunice Kennedy Shriver National Institute of Child Health and Human Development
- PANDA
Prenatal Autonomic Neurodevelopmental Assessment
- PTB
preterm birth
- US
United States
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of Interest
SEC has received honorariums for presentations about DHA in infancy and pregnancy. KMG was the PI of R01HD086001. SEC, BJG and CJV were PIs of R01HD083292, CJV was an employee of RB Nutrition, which produces infant formulas and supplements with DHA at the time the study was conducted, however, RB was not involved in the study execution or analysis. She conducted this study through her role as an Adjunct Professor at The University of Cincinnati. The other authors have no competing interests.
References
- 1.March of Dimes, The Partnership for Maternal, Newborn & Child Health, Save the Children, World Health Organization. Born too soon: the global action report on preterm births. Geneva: World Health Organization, 2021. (https://www.who.int/pmnch/media/news/2012/201204_borntoosoon-report.pdf). [Google Scholar]
- 2.Middleton P, Gomersall JC, Gould JF, Shepherd E, Olsen SF, Makrides M. Omega-3 fatty acid addition during pregnancy. Cochrane Database Syst Rev 2018;11(11): Cd003402. 10.1002/14651858.CD012409.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Simmonds LA, Sullivan TR, Skubisz M, Middleton PF, Best KP, Yelland LN, et al. Omega-3 fatty acid supplementation in pregnancy – baseline omega-3 status and early preterm birth: exploratory analysis of a randomised controlled trial. BJOG Int J Obstet Gynaecol 2020;127(8):975–81. doi: 10.1111/1471-0528.16168. [DOI] [PubMed] [Google Scholar]
- 4.Carlson SE, Gajewski BJ, Valentine CJ, Kerling EH, Weiner CP, Cackovic M, Buhimschi CS, Rogers LK, Sands SA, Brown AR, Mudaranthakam DP, Crawford SA, DeFranco EA. Higher dose docosahexaenoic acid supplementation during pregnancy and early preterm birth: A randomised, double-blind, adaptive-design superiority trial. EClinMedicine 2021; 36:100905. https://doi.org/10.1016;j.eclinm.2021.100905. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Food and Agriculture Organization of the United Nations. Fats and fatty acids in human nutrition. Report of an expert consultation. FAO Food and Nutrition paper 91. Rome. 2010. Available at http://www.fao.org/docrep/013/i1953e/i1953e00.pdf. [PubMed] [Google Scholar]
- 6.Koletzko B, Cetin I, Brenna JT for the Perinatal Lipid Intake Working Group. Consensus statement. Dietary fat intakes for pregnant and lactating women. B J Nutr 2007;98:873–7. [DOI] [PubMed] [Google Scholar]
- 7.March of Dimes. Omega-3 fatty acids during pregnancy. March of Dimes Web Site, 2009. Available at http://www.marchofdimes.com/pnhec/159_55030.asp.
- 8.Crawford SA, Christifano DN, Kerling EH, Gajewski BJ, Valentine CJ, Gustafson KM, Mathis NB, Camargo JT, Gibbs HD, Sullivan DK, Sands SA, Carlson SE. Validation of an abbreviated food frequency questionnaire for estimating DHA intake of pregnant women in the United States. Prostaglandins Leukot Essent Fatty Acids 2022; 177:102398. doi:10.1016;j.plefa2022.102398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Christifano DN, Crawford SA, Lee G, Gajewski BJ, Carlson SE. Utility of a 7-question online screener for DHA intake. Prostaglandins Leukot Essent Fatty Acids 2022; 177:102399. doi: 10.1016/j.plefa.2022.102399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gustafson KM, Christifano DN, Hoyer D, Schmidt A, Carlson SE, Colombo J, Gajewski BJ, Mathis NB, Sands SA, Chollet-Hinton L, Brown AR, Mudaranthakam DP. Prenatal docosahexaenoic acid effect on maternal-infant DHA equilibrium and fetal neurodevelopment: A randomized clinical trial. Pediatr Res 2021. Sept 22;1–10. doi: 10.1038/s41390-021-01742-2. Online ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.American college of obstetricians and gynecologists committee opinion: committee on obstetric practice. Number March 2017;688. [Google Scholar]
- 12.Kuratko C Food-frequency questionnaire for assessing long-chain omega-3 fatty-acid intake: Re: Assessing long-chain omega-3 polyunsaturated fatty acids: a tailored food-frequency questionnaire is better, Nutrition (Burbank, Los Angeles County, Calif.) 29 (2013) 807–808. 10.1016/j.nut.2012.10.013. [DOI] [PubMed] [Google Scholar]
- 13.Yurko-Mauro K, McCarthy D, Rom D, et al. , Beneficial effects of docosahexaenoic acid on cognition in age-related cognitive decline, Alzheimers Dement. 6 (2010) 456–464. 10.1016/j.jalz.2010.01.013. [DOI] [PubMed] [Google Scholar]
- 14.Cheung H, Mazerolle L, Possingham HP, Tam K, Biggs D. (2020). A methodological guide for translating study instruments in cross-cultural research: Adapting the ‘connectedness to nature’scale into Chinese. Methods in Ecology and Evolution, 11(11), 1379–1387. [Google Scholar]
- 15.Gibbs HK, Camargo JMTB, Owens S, Gajewski B, Cupertino AP. Measuring nutrition literacy in Spanish-speaking Latinos: An exploratory validation study. J Immigr Minor Health 2018;20 (60:1508–1515. doi: 10.1007/s10903-017-0678-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gajewski BJ, Reese CS, Colombo J, Carlson SE. Commensurate priors on a finite mixture model for incorporating repository data in clinical trials. Stat Biopharm Res 2016;8(2):151–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Carlson SE, Gajewski BJ, Valentine CJ, Rogers LK, Weinger CP, DeFranco EA, Buhimschi CS. Assessment of DHA on reducing early preterm birth: the ADORE randomized controlled trial protocol. BMC Pregnancy and Childbirth 2017;17:62. doi: 10.1186/s12884-017-1244-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Zhang Z, Fulgoni VL, Kris-Etherton PM, Mitmesser SH. Dietary Intakes of EPA and DHA Omega-3 Fatty Acids among US Childbearing-Age and Pregnant Women: An Analysis of NHANES 2001–2014. Nutrients. 2018;10(4):416. Published 2018 Mar 28. doi: 10.3390/nu10040416 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Christifano DN, Chollet-Hinton L, Mathis NB, Gajewski BJ, Carlson SE, Colombo J, Gustafson KM. DHA supplementation during pregnancy enhances maternal vagally-mediated cardiac autohomic control in humans. J Nutr 2022; Aug 11:nxac178. Doi: 10.1093/jn/nxac178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Christifano DN, Gustafson KM, Carlson SE, Sultanna N, Brown A, Sands SA, Colombo J, Gajewski BJ. Maternal docosahexaenoic acid exposure needed to achieve maternal-newborn EQ. Nutrients 2022. Aug 12:14(16)3300. Doi: 10.3390/nu14163300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.DeNavas-Walt C, Proctor BD. Income and Poverty in the United States: 2014. U.S. Government Printing Office; Washington, DC: 2015. U.S. Census Bureau, Current Population Reports, P60–252. [Google Scholar]
- 22.Rendall MS, Weden MM, Brown J. (2021). Family and household sources of poverty for Black, Hispanic, and White newborns. Journal of Marriage and Family, 1–17. 10.1111/jomf.12781 [DOI] [Google Scholar]
- 23.Coleman-Jensen A, Rabbitt MP, Gregory CA, Singh A. Household food security in the United States in 2019. Economic Research Service, US Department of Agriculture 2020. Available at https://www.ers.usda.gov/publications/pub-details/?pubid=99281. [Google Scholar]
- 24.Hinkle SN, Zhang C, Grantz KL, Sciscione A, Wing DA, Grobman WA, Newman RB, D’Alton ME, Skupski D, Nageotte MP, Ranzini AC, Owen J, Chein EK, Craigo S, Samrawit S, Yisahak SF, Liu A, Albert PS, Louis GMB, Grewal J. Nutrition during Pregnancy: Findings from the National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies–Singleton Cohort, Curr Dev Nutr 2020; 5(1):nzaa182. doi: 10.1093/cdn/nzaa182.eCollection 2021 Jan. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Fowles E, Gabrielson M. First trimester predictors of diet and infant birth weight in low-income pregnant women. J Community Health Nurs. 2005;22:117–30. [DOI] [PubMed] [Google Scholar]
- 26.Keenan K, Hipwell AE, Bortner J, Hoffmann A, McAloon R. Association between fatty acid supplementation and prenatal stress in African Americans: A randomized controlled trial. Obstetrics Gynecology; 2014;124:1080–1087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. https://www.commongroundhealth.org/insights/library/preterm-birth-rates-and-socioeconomic-status.
- 28.Best KP, Gibson RA, Yelland LN, Leemaqz S, Gomeresall J, Liu G, Makrides M. Effect of omega-3 lcpufa supplementation on maternal fatty acid and oxylipin concentrations during pregnancy. Prostaglandins Leukot Essent Fatty Acids 2020; 162:102181. doi: 10.1016/j.plefa.2020.102181. [DOI] [PubMed] [Google Scholar]
- 29.Valentine CJ, Khan AQ, Brown AR, Sands SA, Defranco E, Gajewski BJ, Carlson SE, Reber KM, Rogers LK. Higher-dose DHA supplementation modulates immune responses in pregnancy and is associated with decreased preterm birth. Nutrients 2021;13:4248. doi: 10.3390/nu13124248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) – A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009; 42:377–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Harris PA, Taylor R, Minor BL, Elliott V, Fernandez M, O’Neal L, McLeod L, Delacqua G, Delacqua F, Kirby J, Duda SN. REDCap consortium, the REDCap consortium: Building an international community of software partners. J Biomed Infomr 2019. May 9. doi: 10.1016/j.jbl.2019.103208. [DOI] [PMC free article] [PubMed] [Google Scholar]