When nutrition research became a nightmare

By Mauro Proença — Oct 30, 2026

Horror movies need monsters, but some of the most disturbing chapters in scientific history needed only a carefully controlled diet. From volunteers driven to the brink of starvation to prisoners deliberately made ill and institutionalized patients fed cavity-producing sweets, nutrition research has sometimes advanced at a terrible human cost. These experiments produced valuable knowledge, but they also raise an unsettling question: how much suffering can science justify in its pursuit of answers?

Image by ACSH using AI

Vampires, zombies, demons, and giant spiders populate popular culture and horror cinema. Who hasn't imagined being chased by Michael Myers after a horror movie or considered drinking coffee to avoid Freddy Krueger in their sleep after watching A Nightmare on Elm Street?

As terrifying as fictional horrors may be, scientific history contains genuine atrocities. During World War II, Japan's Unit 731 conducted brutal, often lethal human experiments as part of its biological warfare program in Manchuria. By disregarding participants' consent and safety, these experiments were crimes against humanity, not merely research that crossed an ethical line. But disturbing episodes in human experimentation are not confined to wartime atrocities. Nutrition research has its own troubling history, though the following studies fall into a different category. They pursued legitimate scientific questions, sometimes involved willing volunteers, and occasionally provided direct benefits to participants. Yet they also reveal how the pursuit of knowledge can lead researchers to accept, prolong, or deliberately inflict suffering. Their history raises a more nuanced question: when does the pursuit of scientific progress cease to justify the human cost?

When Hunger Became an Obsession

The Minnesota Starvation Experiment became a milestone in understanding the physical and psychological effects of semi-starvation and the challenges of rehabilitating malnourished populations after war. It also took a severe toll on participants. [1]

As World War II neared its end, scientists sought to understand prolonged starvation and restore the health of war-torn populations. Physiologist Ancel Keys, known for helping develop K-rations, designed the study at the University of Minnesota. Its 36 participants were conscientious objectors selected from more than 400 volunteers.

The experiment had three phases:

Three months on a regular diet to establish caloric needs and baseline weight.
Six months consuming roughly half their usual calories to lose 25% of initial body weight.
Three months of recovery testing different energy and nutrient intakes.

Their diet mimicked wartime European food supplies, consisting of potatoes, turnips, rutabagas, dark bread, and pasta, served in two meals daily Monday through Saturday and one larger meal on Sundays, alongside unlimited black coffee and water and daily cigarette rations.

Researchers aimed for weekly weight loss of about 1.1 kg and could further reduce intake if participants fell short. Volunteers also performed domestic and administrative tasks, expending about 3,009 kcal daily. Initially, participants lost weight rapidly and experienced slight declines in strength, while their moods remained generally good. Within weeks, however, one volunteer began cheating, buying sundaes and malted milks and stealing raw rutabagas. He developed severe psychological distress and briefly required psychiatric hospitalization.

Researchers introduced a buddy system to prevent volunteers from leaving the laboratory alone, but their condition continued to deteriorate as food became an obsession. They dreamed about it, licked their plates clean, lost interest in classes, cultural activities, dating, sex, and became increasingly irritable.

During the second phase of semi-starvation, symptoms intensified. Participants felt constantly cold, their body temperatures and heart rates dropped, and even sitting became uncomfortable. Resisting food-related urges became a struggle.

Initially allowed to chew gum, some men began chewing up to 40 packages a day. The second volunteer excused from the study consumed huge amounts of gum and admitted to eating scraps from trash cans, and, like the first, required a brief psychiatric hospitalization. Another participant broke the diet and developed urological complications, and one more was excluded because his weight loss did not match his recorded intake and energy expenditure.

During the rehabilitation phase, researchers randomly assigned the remaining participants to one of four energy-intake groups, each of which was further stratified by protein and vitamin levels. The lowest group initially received 2,200 kcal daily, only about 400 kcal more than during semi-starvation, and Keys eventually raised the levels when the men showed little improvement.

Recovery proved as difficult, if not harder than, semi-starvation. Despite the extra calories, improvements were modest, and weight returned slowly. The situation reached a critical point when one volunteer severed three fingers with an axe while chopping wood. He could not explain whether the act was intentional. As caloric intake rose, dizziness, apathy, and lethargy improved, but weakness, fatigue, loss of libido, and food obsession persisted. These effects began to subside consistently only after the experiment ended.

The volunteers knowingly accepted considerable personal risk in the hope of helping millions facing starvation after World War II. A modern Institutional Review Board (IRB) would likely reject the original protocol, not simply because volunteers were deliberately starved, but because the experiment continued despite substantial physical and psychological deterioration, with limited safeguards for stopping participation or ensuring timely recovery. The enduring question is not whether research may expose volunteers to harm, but how much harm society can ethically ask them to endure in pursuit of knowledge that may benefit others.

The Minnesota experiment was not an isolated case.

When Food Became Both the Cure and the Experiment

Pellagra results from a deficiency of vitamin B3 (niacin) or the essential amino acid tryptophan. It can cause the four “D’s”: dermatitis, diarrhea, dementia, and death. Spanish physician Don Gaspar Casal first described the disease in the 18th century after observing it among peasants. Field studies found that the disease was less common among wealthier people whose diets included more fresh foods and animal products. Three investigations strengthened the evidence.

In September 1914, two orphanages in Jackson, Mississippi, recorded 79 and 130 cases of pellagra, respectively. Both suffered from poor sanitation, overcrowding, and diets lacking animal-based foods. Beginning in mid-September, both institutions improved their diets by adding animal protein and legumes without deliberately changing sanitary conditions.

The results were striking. At one orphanage, none of the 67 monitored cases recurred, and no new cases appeared among 99 residents without a previous diagnosis. At the other, researchers recorded one recurrence among 105 children and no new cases among 69 residents without a history of pellagra.

At the Georgia State Sanitarium, researchers studied women with mental disorders. Among 72 women monitored from late 1914 to October 1915 on a diet similar to that used in the orphanages, none had a recurrence, including 18 with multiple previous episodes. In comparison wards, at least 15 of 32 women who remained on the standard diet had a recurrence of Pellagra.

Finally, a study conducted at a Mississippi State Penitentiary farm from April to October 1915 recruited white prisoners promised pardons upon completion and investigated whether a monotonous, cereal-based diet could induce pellagra in healthy men. Of the 12 inmates, aged 24 to 50, eleven completed the experiment. Researchers also monitored 35 prisoners without a history of pellagra as controls.

The groups differed in more than diet. Control quarters were dirty and pest-infested, while volunteers' quarters were regularly cleaned and their hygiene supervised, which should have protected them. Segregation under guard and lighter work further complicate interpretation.

Control prisoners received the standard prison diet, considered relatively adequate despite its high fat content and containing about twice the protein of the volunteers' diet. The volunteers ate a monotonous, cereal-based diet. One reported weakness and dizziness within two weeks; eventually, all 11 reported symptoms, mainly weakness and abdominal pain or discomfort. Everyone lost 6% to 22.6% of their body weight.

Six volunteers developed dermatitis nearly five months into the experiment. Specialists considered all six cases consistent with pellagra based on characteristic skin manifestations. Researchers concluded that the experiment, along with other evidence, showed that diet was a determining factor in the disease's causation and prevention.

Today’s IRB might approve the orphanage and medical ward studies; in both cases, no one was deprived of food, and there was a potential health benefit. However, informed consent is a concern, especially for orphaned children. By contrast, the prison study had two ethical red flags. Most critically, it induced a potentially fatal disease. Researchers intentionally created nutritional deficiencies in otherwise healthy prisoners, producing considerable weight loss and clinically recognizable pellagra. The prisoners had no prospect of direct medical benefit. Second, the promise of a pardon likely influenced their decision to participate in the study. [2]

The Bitter Truth About Sugar and Tooth Decay

Dental caries is a multifactorial disease in which acids produced by microorganisms in dental biofilm demineralize tooth tissues. Understanding this process took time. The frequency of consuming fermentable sugar is a key factor, alongside individual characteristics and oral health conditions.

One study that advanced this knowledge was conducted at Sweden’s Vipeholm Hospital, an institution for people with intellectual disabilities. Prior research had investigated the role of vitamins and minerals without demonstrating any clinical benefit, so investigators turned to study how different forms and frequencies of carbohydrate consumption affected tooth decay. Researchers compared sugar dissolved in liquids, incorporated into bread, and served as sticky toffees between meals. They later tested familiar sweets, including chocolate and caramels, in different quantities and at different times. They found that form, frequency, and timing of carbohydrate consumption altered dental caries.

Tooth decay increased primarily among participants who ate sticky sweets between meals. Those who consumed sugar in drinks or with meals generally saw little increase in cavities, although men who ate sugar-enriched bread developed more cavities during the second year, when the bread was served at every meal. Tooth decay rose sharply in those eating 24 toffees daily, prompting researchers to replace the toffees with an equal amount of fat earlier than planned. Tooth decay then returned to previous levels. Those eating caramels showed a similar pattern, but the changes occurred more gradually.

A modern IRB would not approve the study's design and methods. A historical analysis found that informed consent was inadequate or absent and that researchers selected vulnerable residents even though other participants were available.

In horror films, the monsters are generally easy to identify. In these experiments, the horror is less obvious: researchers pursuing answers to genuine health problems sometimes helped their subjects, sometimes knowingly harmed them, and sometimes relied on individuals whose freedom to refuse was limited. The lessons about starvation, pellagra, and tooth decay were real, but scientific value alone cannot justify the means used to obtain it. These experiments leave us with an unsettling question: not whether knowledge is worth pursuing, but who gets to decide what price others must pay for it.

[1] This section draws on “They Starved So That Others Be Better Fed: Remembering Ancel Keys and the Minnesota Experiment” and “The Minnesota Starvation Study,” from Twenty-Four Stories From Psychology.

[2] In a 2025 historical analysis, researcher Dana Landress argues that the participating prisoners were unusually wealthy, politically connected white men who used their influence to obtain participation and prearranged pardons. This complicates the conventional depiction of researchers exploiting powerless prisoners.

Tags: