Commercial Conflicts in Personalized Health Products

What research says about sponsorship and results

A Cochrane review of 75 papers compared industry-sponsored drug and device studies with studies sponsored by others. Industry-sponsored studies more often had favorable efficacy results (risk ratio 1.27, 95% confidence interval 1.17 to 1.37). [1] That review covered drugs and medical devices, so it does not measure personalized nutrition directly.

In nutrition, an older analysis of 206 articles about soft drinks, juice and milk found that industry funding may bias conclusions in favor of sponsors' products. [2] Neither study proves bias in any single trial. They explain why disclosure is worth reading.

Evidence label: sponsorship and favorable results are established as an association in drug and device research; for personalized nutrition specifically, insufficient evidence, because no review of this field was found in the sources read.

What disclosures look like in this field

The table shows ties reported in the papers' own statements. It is not a judgment of the findings.

Disclosures vary in how much influence they reveal. A funder on a scientific advisory board that contributes to design and analysis is different from a funder with no role. [3] [7]

The same issue appears in guidelines and reviews

Opinion and guidance papers can carry ties too. One proposed framework for judging gene-based diet advice noted that some authors had worked for or founded genetic-testing services. [8] The same paper noted that this field is unregulated, with no defined standards beyond some commercially adopted codes of practice. [8] A review of 104 companies' nutrition gene tests found the market unregulated, and only about half disclosed the genes used. [9]

Oversight is limited

In the United States, the FDA generally does not review direct-to-consumer tests for general wellness or low-risk purposes before they are offered, and the evidence behind them varies. [10] The FDA also does not approve dietary supplements or their claims before marketing. [11] So a product being on sale does not tell you it was checked.

How to read a disclosure

• Check the role. Employee, founder, consultant and paid speaker are different levels of tie.
• Check who analyzed the data. Independent analysis is stronger than analysis by the seller.
• Look for independent repeats. A result confirmed by unrelated teams is more reliable. A large trial that tested a gene-based approach and found no added benefit is an example of a result that does not favor a product's pitch. [12]
• Look for pre-registered outcomes. See registered trials versus published results.
• Do not dismiss everything. Company-funded trials can be well run. They deserve the same careful reading as any other.

Related guides

For general principles, see funding and conflicts of interest in health research and financial conflicts in emerging health products. For what the trials found, see long-term outcomes in personalized nutrition trials.