Bias in medical research and nutrition

We often assume that conclusions about food and medicine from research studies are reliable and valid. Historically, much of the research in nutrition was funded by the government, with important institutions such as NIH in the US and NIN in India. However, the decline in funding in the 21st century has led to private industries, especially food and pharma, filling the funding gap. Medical and nutritional research can be divided as follows:

(1) Basic research

Typically funded by the NIH, it focuses on broader public benefits such as laboratory studies and randomized trials. Projects such as the Human Genome and the Human Proteome are still funded by the NIH. However, the decrease in government funding has made this research less widespread.

(2) Applied research

Applied research uses science to prevent and treat disease. While some still receive NIH funding, most have been funded by the pharma, food and supplement industries. It can be divided into research supporting different foods, justifying supplements, or showing harm in certain foods or lifestyles.

(3) Public health research

This aims to improve public health at the population level, effectively funded by the government, although private industries are actively involved.

(4) Medical research

Focused on clinical trials, this research is mainly supported by pharmaceutical companies. Variation in how people respond to treatment (eg, due to gut microbiomes) complicates the results. For example, if a drug helps 30% of participants in the placebo group and 70% in the experimental group, it is considered effective, which leads to its prescription for everyone, even if 60% are not cured or not cured only because of the placebo effect. , they are still exposed to side effects without getting any benefit.

(5) Pharmaceutical research

This branch examines how chemicals interact with biological processes, playing an important role in drug development. Here are some important points to keep in mind:

5.1 The goal is not necessarily to cure disease, as doing so would conflict with the long-term profits of pharmaceutical companies.

5.2 Pharmaceutical companies often focus on controlling health indicators such as blood pressure, blood sugar, or cholesterol, which often require increasing doses over time.

5.3 These companies emphasize early detection, saying it saves lives. While this is true for certain conditions such as cancer, for many lifestyle diseases, it often leads to early initiation of medication. In addition, they insist on lowering the thresholds for starting treatment, as seen with blood pressure, blood sugar and cholesterol medications.

5.4 Pharmaceutical companies encourage strict control of these symptoms, often suggesting high or frequent doses. For example, increasing the dosage from two to three pills per day represents a 50% increase in business. This approach inspired the ACCORD study (2001-2009). However, the study backfired, as it showed an increase in the number of people dying in the experimental group, leading to early termination.

Bias in research

When we read research papers, we expect them to be honest and objective. However, various reasons can distort this view. Many studies have shown how research can be used to help a donor organization. Some of these options are summarized below.

(1) Financial bias

In the United States, government funding for research has declined dramatically over the past fifty years. In 2021, private sources invested $633 billion in research and development (R&D), while government funding reached only $76 billion. Private funding often comes with vested interests or conditions, and research shows that research funded by private sources rarely conflicts with the interests of the funder. This can happen for a number of reasons:

– Research can be done to avoid negative consequences. For example, using the Standard American Diet (SAD) as a control can make almost any type of diet seem healthy in comparison.

– In some cases, the research design can hide significant negative results. For example, one study asked college students to eat more white bread, and their health seemed to improve. However, concluding that white bread is healthy would be misleading, as the students had to reduce their intake of harmful foods such as eggs, sausages and bacon.

-Funders can cancel the research in the middle if they realize that the results will not help them. To prevent this, a law was passed requiring investigations to be registered with the government authorities before they start, to ensure that they cannot be canceled in the middle. However, this rule is not always strictly applied by executives who hope to find a comfortable consulting job when they retire.

– Before publishing any research, a general wording must be given. Funders often use this summary to minimize negative results. A notable example is when Senator Bob Dole changed the language of George McGovern’s dietary guidelines of 1977. Initially advocating a reduction in meat consumption, the guidelines were changed to encourage the consumption of lean meats. These two expressions have completely different meanings.

(2) Myth prejudice

Myths can distort the judgment of investigators. For example, it has long been recommended to eat more protein, despite evidence that excess protein damages the kidneys and liver. Similarly, the myth that plant protein is incomplete has been dispelled, yet it persisted for years.

The dairy industry promoted the idea that milk is the best source of calcium, despite evidence that it may contribute to osteoporosis.

(3) Expert pride bias

Respected experts can distort findings to protect prior positions. An example is Dr. Ancel Keys, whose *Seven Country Study* recommended a low-fat diet by excluding data from 15 countries that contradicted his view.

(4) Publication bias

Good results are more likely to be published,​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​ Negative results often face a reluctance to publish, leading to a distorted understanding of certain treatments.

For example, if three studies find that egg white therapy cures stage 2 breast cancer, all three studies may be published. However, if the following seven studies show that this is not the case, only one or two may be published. Future researchers may mistakenly conclude that the treatment is effective, leading to other studies based on limitations.

(5) Media bias

The media chooses sensational stories, which can lead to biased reporting. The food and pharma industries, key advertisers, drive publicity,​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​ For example, Dr. The high-protein, high-fat Atkins diet received more media coverage than Pritikin’s low-fat, high-carbohydrate diet, despite the latter. -rao was scientifically sound. Sensational books such as *Grain Brain* and *Wheat Belly* have also received a lot of coverage, driven by powerful Food Industry organizations that support such reports. Furthermore, since the media is heavily dependent on advertising revenue from the food and pharmaceutical industries, they are often reluctant to publish studies that criticize these industries.

(6) Government bias

Government agencies often face conflicts of interest, as officials seek to negotiate post-retirement jobs, and elected officials collect money from the industries they control.

The public relies on government agencies such as the Food and Drug Administration (FDA) to ensure that medical devices are safe.

There is an inherent conflict of interest in the drug and device approval process, as the FDA relies on clinical trials conducted by the same companies that make these products. Research findings are often ambiguous and open to debate. Sometimes the FDA has approved drugs and devices despite opposition from its staff. This has recently emerged with a device called MitraClip developed by Abbott Labs to repair leaky heart valves that was approved by the FDA in 2013. More than 17,000 reports have been filed complaining about this device.

(7) Other conflicts of interest

Political interests, personal interests, and other conflicts can complicate the investigation situation.

The end

It is important to approach research carefully. Understanding potential biases, especially the source of funding, helps reveal the full story of the research. Always check before accepting decisions.

EFFECT PRINCIPAL