Linked evidence from seven dormant randomized clinical trials suggested that kids given nutritionally modified baby formula in infancy did not achieve higher academic performance compared to kids who were given standard formula, possibly debunking the claim that modifying formula milk promotes cognitive development.
While a number of studies have demonstrated that breastfeeding is the best option for infant nutrition, infant formulas remain a widely used alternative around the globe. Thus, optimizing the composition of formula products to enhance nutrition and reduce disparities in health outcomes between breastfed and formula fed infants has become an important public health issue; in particular, manufacturers often modify formula in an attempt to impart long-term cognitive outcomes on par with those seen in breastfed babies, Maximiliane L. Verfürden, PhD, of UCL Great Ormond Street Institute of Child Health in London, and colleagues wrote in The BMJ.
Verfürden and colleagues conducted their study to assess three types of modified infant formula that are “widely available and have been proposed to promote cognitive development: formula enriched with nutrients, formula supplemented with long chain polyunsaturated fatty acids (LCPUFAs), and follow-on formula fortified with iron.”
The study authors linked a series of “dormant” randomized controlled trials of infant formulas conducted from 1993-2001—two of nutrient enriched formula, two of LCPUFA supplemented formula, one of iron-fortified follow-on formula, one of formula supplemented with nucleotides, and one of formula supplemented with sn-2 palmitate—to national student records in England in order to establish whether or not modified formulas led to different academic performance outcomes (measured by differences in scores on General Certificate of Secondary Education [GCSE] exams) compared to standard formula. The seven trials assessed were referred to as dormant because the primary outcome for each study had already been collected and further active follow-up had been discontinued; “however, participant identifiers were retained enabling reactivation of the trials through linkage,” they explained.
They found that “differences in academic performance between modified and standard formulas were consistent with differences measured in the original trials and in the external literature; that is, no benefit of the infant formula modifications on cognitive outcomes.”
“We found no evidence of improved cognitive ability, measured by differences in scores for GCSE mathematics at age 16 years, for nutrient enriched versus standard formula in infants born small for gestational age at term, or in preterm infants after discharge from hospital, in LCPUFA supplemented versus unsupplemented formula in term or preterm infants, or in high versus low iron follow-on formula in term infants,” they wrote. “Upper 95% confidence limits excluded a moderate benefit of more than 0.27 standard deviation for any of the modified formulas (equivalent to about half a GCSE grade).”
In addition to suggesting that modified formulas provide no cognitive improvements compared to standard baby formulas, Verfürden and colleagues also argued that their analysis demonstrates the feasibility and effectiveness of following up on dormant clinical trials through data linkage without participants needing to reconsent.
“This approach, with linkage to school data and potentially to health databases, can be used to maximize the use of dormant trials to look at the remaining gaps in evidence about modification of formula feeds at critical stages of development,” they wrote. “This study sets a precedent for other trials and cohorts to use linkage to administrative data to answer important questions about long term outcomes in children and young people.”
For this analysis, the study authors linked trial data from the seven included baby formula trials to routinely collected data from the National Pupil Database, which contains data for all children attending state schools in England and all children who took national curriculum tests such as the GCSE. Each trial randomized infants to either modified or standard formula—schools were not informed about trial participation or group allocation.
The prespecified primary outcome was the standardized difference within trials in mathematics GCSE score at age 16 years; the study authors chose math as opposed to English for the primary outcome “because examination results for mathematics are considered to be graded less subjectively.” Prespecified secondary outcomes were mean difference in standard deviation scores for GCSE English at age 16 and for mathematics and English at age 11; odds ratio of receiving five or more GCSEs at grade C or above in the intervention group compared to controls; and the odds ratio of being allocated special educational needs support at any age.
The study randomized 1,763 formula trial participants, of whom 1,607 (91.2%) were linked to a pupil record in the National Pupil Database.
“No benefit was found for performance in mathematics examinations at age 16 years for any modified formula: nutrient enriched in preterm infants after discharge from hospital, standard deviation score 0.02 (95% confidence interval -0.22 to 0.27), and nutrient enriched in small for gestational age term infants -0.11 (-0.33 to 0.12); LCPUFA supplemented in preterm infants -0.19 (-0.46 to 0.08) and in term infants -0.14 (-0.36 to 0.08); iron follow-on formula in term infants -0.12 (-0.31 to 0.07); and sn-2 palmitate supplemented formula in term infants -0.09 (-0.37 to 0.19),” they found. “Participants from the nucleotide trial were too young to have sat their General Certificate of Secondary Education (GCSE) examinations at the time of linkage to school data. Secondary outcomes did not differ for nutrient enriched, high iron, sn-2 palmitate, or nucleotide supplemented formulas, but at 11 years, preterm and term participants randomized to LCPUFA supplemented formula scored lower in English and mathematics.”
Why the infants randomized to LCPUFA supplemented formula experienced a dip in secondary academic performance measures “is unclear,” Verfürden and colleagues noted. “The content of docosahexaenoic acid in human milk is variable and greatly influenced by maternal diet. This natural variability in breast milk makes the optimal dose of docosahexaenoic acid in infant formula uncertain. Furthermore, LCPUFAs derived from sources other than human breast milk, and in isolation from other components present in human breast milk, likely have different biological properties compared with LCPUFAs naturally occurring in human breast milk. Given the potential associations between the source of LCPUFAs and adverse cognitive outcomes, long term follow-up of trials testing infant formulas from other sources of LCPUFAs is recommended, including use of innovative trial designs.”
This point is particularly important, they added, given a recent mandate to add docosahexaenoic acid to all infant and follow-on formulas in the European Union. Given the findings of this study, Verfürden and colleagues suggested that this legislation should be reconsidered, as the mandate “might have the potential for considerable harm and could also inhibit future research by limiting equipoise.”
In an editorial accompanying the study, Charlotte Wright, MD, and Ada L. Garcia, PhD, both of the University of Glasgow in Glasgow, U.K., called the linked research by Verfürden et al “impressive.” What’s more, they noted, while the current study suggests modified baby formulas do not provide a benefit for child cognitive development, there is some evidence suggesting that they can potentially cause harm.
For example, some manufacturers of baby formula advertise the addition of iron to follow-on milks, “which could be interpreted as an advantage of formula milk over breast milk,” as breast milk has little iron content, they wrote. However, while Verfürden and colleagues found no cognitive differences between kids given iron-supplemented formula compared to standard formula, a 2019 study suggested that children given baby formula supplemented with iron had reduced cognition at age 16, and an earlier analysis suggested iron supplementation may have adverse effects on child growth. Given these findings, “it is time to consider whether current regulations governing the composition of formula milks need review worldwide.”
Wright and Garcia also noted that the reliability of large formula trials is questionable; in a large systematic review of formula milk trials published last month in The BMJ, Boyle et al found that most such trials lacked transparency and contained selective reporting, 80% of trials were at high risk of bias, and only 14% of trials were conducted independently of formula manufacturers.
“Giving babies infant formula instead of breast milk has been shown convincingly and repeatedly to place babies at risk of harm,” they added. “Recently published evidence suggests a need to better regulate research into infant formulas and to ensure that this evidence is used to remove unnecessary and potentially harmful nutrients from formula milk, and to prevent misleading promotional claims.”
Study limitations included that analyses of academic performance were only powered to detect differences between groups of greater than 0.32-0.43 standard deviation; the composition of formulas and neonatal care have changed since the time in which the formula trials were initially conducted; and the cognitive benefits of the randomized formula modifications may have been “diluted by unmeasured compensatory interventions in the control group over time, masking a true benefit of the intervention,” the authors noted.
-
Linked evidence from seven dormant randomized clinical trials suggested that kids given nutritionally modified baby formula in infancy did not achieve higher academic performance compared to kids who were given standard formula.
-
The study authors argued that this study sets a precedent for trials and cohorts to use linkage to administrative data to answer questions about long-term outcomes in children and adolescents.
John McKenna, Associate Editor, BreakingMED™
The study authors reported support from the Economic and Social Research Council UCL, Bloomsbury and East London Doctoral Training Partnership and a Great Ormond Street Hospital Charity Research starter grant for the submitted work. Verfürden declared support from the Economic and Social Research Council UCL Bloomsbury and East London Doctoral Training Partnership, as well as the Great Ormond Street Hospital Charity.
Wright is a member of the U.K. Scientific Advisory Committee on Nutrition.
Cat ID: 138
Topic ID: 85,138,730,191,41,138,43,192,94,925
