Karen Harris
Karen Harris is a professional writer and editor with more than two decades of experience in research-driven nonfiction writing. With a background in journalism and a Master's degree in English, she brings a meticulous approach to fact-finding, source evaluation, and clear communication of complex topics. Her work spans health, lifestyle, and consumer education, always with a focus on accuracy and readability.
Context
When breastfeeding is not possible, infant formula becomes the only alternative, and manufacturers increasingly add ingredients marketed with claims such as support for digestive health or cognitive development. Parents and clinicians often assume these claims rest on solid research. This systematic review asks a prior question: how trustworthy is the research itself?
The authors set out to assess the methodological quality, risk of bias, and conflicts of interest in randomized clinical trials of fortified, enriched, or supplemented infant formula in healthy infants. Rather than judging whether any formula works, the review evaluates whether the trials behind such products are reliable and independent.
Study Overview
Design: Systematic review following the PRISMA 2020 guidelines and registered in PROSPERO (CRD42023399640). Literature searches ran in MEDLINE (PubMed), EMBASE, and CINAHL in October 2022 and were updated in February 2023, with no language or date restrictions. Trial quality was assessed with the Cochrane Risk of Bias 2 (RoB 2) tool, and associations were tested with Fisher's exact test.
Included studies: 40 randomized controlled trials of fortified infant formula in healthy infants. The search returned 888 records; after screening and the addition of 7 papers from reference lists, 41 met the eligibility criteria, and 1 was removed as a duplicate publication, leaving 40. Papers were grouped by main outcome: microbiota (8), growth (8), miscellaneous (10), vitamins and minerals (3), neurodevelopment (3), bone function (3), fatty acids (3), and immune function (2). The trials were published between 1987 and 2023, were conducted from 1981 through 2020, and had initial sample sizes ranging from 60 to 1678 participants.
Comparators: Every included trial was required to have three arms: infants fed fortified (enriched or supplemented) formula, infants fed standard formula, and breastfed infants.
Outcomes assessed: Overall and domain-level risk of bias; the presence and correct disclosure of conflicts of interest; industry funding; and the direction of each trial's conclusions.
Funding: The review carries no funding statement. All four authors completed the ICMJE uniform disclosure form and declare no competing interests. PubMed indexes the paper as “Research Support, Non-U.S. Gov't.”
Evidence Certainty (GRADE): Not formally GRADE-rated. The review instead applied the Cochrane RoB 2 tool and rated all 40 included trials at high overall risk of bias.
Key Findings
Risk of bias
All 40 trials showed a high overall risk of bias. The “deviations from intended intervention (intention to treat)” and “missing outcome data” domains were especially problematic, and the authors describe the randomization process as the domain that most frequently displayed bias across the outcome groups. Only four trials used an intention-to-treat rather than a per-protocol analysis.
Industry ties
A potential conflict of interest with industry was identified in 33 of the 40 trials, and in 17 of those the conflict was not declared in the appropriate section. 35 trials were associated with industry through funding. 28 trials had at least one author with an industry affiliation, and in 14 trials people involved in analyzing samples or data had a declared relationship with industry.
Direction of conclusions
29 trials reported conclusions favorable to the fortified formula and only one reached an opposing conclusion. Table 1 of the paper prints a total of 28, which appears to be a typographical error: the per-subgroup values in that same table sum to 29, matching the abstract and the running text. 15 trials reported conclusions that were mutually contradictory or not fully in line with their own results.
No measured association
Using Fisher's exact test, the authors found no statistical association between the direction of a trial's conclusion and conflicts of interest, funding, financial ties, or donation of the formula. They attribute this partly to the small number of studies.
Reporting gaps
Only 15 trials reported the nutritional composition of the fortified formula and only 12 reported that of the standard comparator, which makes the true difference between the formulas hard to establish. Eleven trials did not clearly report their sample sizes. Several trials also included mixed-feeding arms, in which infants received both breast milk and formula, so the effect of the formula cannot be separated from that of breast milk.
Limitations
The authors note that they did not review the registered trial protocols in depth or flag their absence, and that they did not classify endpoints as surrogate versus patient-relevant because the outcomes were too heterogeneous for detailed analysis. The small number of included trials also limited the association testing. Beyond the authors' own list, risk-of-bias assessment with RoB 2 involves reviewer judgment (here carried out in pairs and settled by consensus), and the strict requirement for a three-arm design narrows the body of evidence the review could capture.
Neutral Interpretation
This review speaks to the quality and independence of the evidence base rather than to whether any specific fortified formula is beneficial. Its central finding is that the trial literature is methodologically weak, with every included trial at high risk of bias, and closely entangled with industry through funding, authorship, and data analysis. At the same time, the authors did not find a statistical link between industry involvement and favorable conclusions, so the review does not establish that funding caused the positive framing, and it names small sample size as a reason. The review's own authors declared no competing interests. Because the results rest on subjective risk-of-bias judgments and a stringent inclusion definition, they are best read as a call for more independent, better-reported trials rather than as a verdict on individual products.
Full Citation
García G, Pérez-Ríos M, Ruano-Ravina A, Candal-Pedreira C. Assessing conflict of interest reporting and quality of clinical trials on infant formula: a systematic review. J Clin Epidemiol. 2024;169:111313. doi:10.1016/j.jclinepi.2024.111313. PMID: 38432526.
Available at: https://pubmed.ncbi.nlm.nih.gov/38432526/
Disclosure
This summary is based on the published systematic review. It is provided for informational purposes only and does not constitute medical advice.