Why Did Doctors Starve Diabetic Children Before Insulin Was Discovered?
The Allen starvation diet was not the only treatment available before insulin arrived in 1922, and the paper behind this article is a historical review of the published record rather than a clinical trial. Allan Mazur traces the therapy back to Frederick Allen’s experiments on dogs made diabetic by partial pancreatectomy, not to human evidence and not to the simple observation that carbohydrates raise blood glucose. Allen knew carbohydrates were the dietary component that most affects blood glucose, and so did many of his contemporaries, who prescribed an “animal diet” or another low carbohydrate variation to clear glycosuria. Critics of the Allen and Joslin regimen claimed as much success without near-total food privation, using diets low in carbohydrate but allowing more fat calories. Mazur’s conclusion is that clinical experience at the time gave little indication that severe undernutrition produced better outcomes than the low carbohydrate diets already in use.
Dr. Kumar’s Take
This is one of the more uncomfortable chapters in medicine, and reading Mazur’s account, the discomfort is not only that children were starved. It is that the justification offered for starving them was thinner than the profession believed. Allen and Joslin presented the regimen as a desperate but rigorous application of animal experimentation to human care, and textbooks have repeated that framing ever since. Mazur argues the justification was exaggerated: the public record contains only the briefest account of the relevant animal work, and a real alternative, low carbohydrate feeding with more fat, was sitting in plain view. Joslin himself looked back and wrote that they literally starved the child and adult with the faint hope that something new in treatment would appear, and that it was no fun to starve a child to let him live. I read that as an honest man’s verdict on his own practice, and as a reminder that a treatment can be sincere, prestigious, and still poorly grounded.
Key Findings
Frederick M. Allen and Elliott P. Joslin were among the most prominent diabetes specialists in America during the first third of the twentieth century. Joslin, the elder of the two, encouraged Allen’s early experimental work, and both men converged on the belief that severely calorie-restricted diets were the best therapy for diabetes. From 1915 until insulin was introduced in 1922, they promoted diets built on repeated fasting and prolonged undernourishment, not as a cure but for relief of symptoms and maximum extension of life.
Fasting usually does reduce glucose levels in diabetics. Prolonged calorie restriction introduced new hazards, most obviously death by starvation, which Allen and Joslin called “inanition.” It was already understood in Allen’s time that in patients of normal or lower weight, severe calorie restriction could lessen resistance to infection, and in children it might stunt growth. The regimen was a balancing act run by a specialist while the patient stayed under close supervision in a hospital or clinic, and patients tended to backslide once they went home. The permanent version of the diet was unpleasant, hard to maintain, and enervating, and many patients withdrew from it. Observers of the era asked whether such privation was too high a price for a modest extension of poor quality life.
Allen’s most famous patient was Elizabeth Hughes, daughter of Charles Evans Hughes, who was governor of New York, the Republican presidential candidate in 1916, and later chief justice of the U.S. Supreme Court. Elizabeth developed diabetes in 1919 at age 11, standing 4 feet 11 1/2 inches and weighing 75 pounds. Allen put her on a week of fasting followed by 500 calories daily with one fast day per week, which brought her weight down to 55 pounds. Once she was free of glycosuria, her diet was raised to 1,250 calories except on fast days, and her weight rose above 60 pounds. By the winter of 1921/22 she had deteriorated seriously and weighed 45 pounds. Her mother pleaded with Frederick Banting to include her as a trial patient for insulin. She regained weight, graduated from Barnard College, married, had three children, stayed active in civic affairs, and died of pneumonia in 1981 at age 73. Few of her friends knew she was diabetic.
Brief Summary
Allen and Joslin promoted fasting and prolonged undernutrition as the most advanced diabetes treatment of their day, and presented it as animal experimentation carried over to the bedside. Mazur’s review of the record finds that framing overstated. Allen’s book, published by Harvard University Press as Studies Concerning Glycosuria and Diabetes and running 1,179 pages, gives no prescription for calorie deprivation as a therapy for human diabetes and describes no beneficial effect of starvation in his animals. It contains conclusions that run against his later position, including “fat feeding is not to be feared in diabetics.”
Study Design
This is a review of the historical and published record, not a trial and not a study of patients. Mazur traces Allen’s career and the documentary trail behind the therapy. Allen graduated from the University of California at Berkeley and completed his M.D. there in 1907. He spent 1909 to 1912 as a teaching fellow at Harvard Medical School, conducting experiments on cats and dogs largely at his own expense, and produced diabetes in his animals by partial pancreatectomy, with severity depending on how much pancreas was removed. In 1913 he left Harvard for a post as nonresident assistant physician working on diabetes at the newly established hospital of the Rockefeller Institute in New York.
The earliest public statement of the treatment Mazur can locate was read at an American Medical Association session in June 1914. Allen reported that if enough of a dog’s pancreas is removed, glycosuria appears even on meat feeding; that in some of these dogs a few days of fasting makes the urine sugar-free; and that cautious protein and fat feeding, enough only to keep the animal thin, keeps such dogs free from diabetes. His longest experiment at that point was a single dog kept free of glycosuria for six months and still alive. He cautioned that if an attempt is made to increase the animal’s weight, glycosuria soon returns and must be checked by renewed fasting, adding that “such dogs, though very thin, are vigorous and lively.” Allen presented these results anecdotally, with no quantitative data and without reporting how many animals were used.
Results You Can Use
The starvation regimen reduced glucose and cleared glycosuria, and it was not a cure. At the time Elizabeth Hughes was treated, most child diabetics died from coma within months to a few years of diagnosis, and Mazur writes that it is difficult to judge how much Allen’s regimen prolonged her life. What changed her trajectory was insulin, not the diet. Meanwhile, practitioners using low carbohydrate diets with more fat calories claimed comparable success without near-total food privation, and clinical experience of the era gave little indication that severe undernutrition did better.
Why This Matters For Health And Performance
Calorie restriction is accepted today as beneficial for diabetics who are overweight. The failure in Allen’s era was applying near-starvation to patients of normal or lower weight, and to children, where the known costs included reduced resistance to infection and stunted growth. The broader lesson is about how a treatment gets adopted. Allen and Joslin were the most prominent specialists in the country, and their explanation, that severe undernutrition followed from animal experiments, was accepted and repeated long after the fact even though the published animal work supporting it was slight.
How to Apply These Findings in Daily Life
- Ask what evidence sits behind a treatment, not just who endorses it
- Notice when a rationale is offered retrospectively rather than published in advance
- Recognize that a therapy can lower a lab value and still harm the patient
- Understand why insulin transformed a disease that specialists could only manage by privation
- Remember that the alternatives available at the time are part of judging any treatment
- Advocate for accessible insulin and affordable medication
Limitations To Keep In Mind
This is one historian’s review of the surviving record, including an unpublished Allen memoir held in a private collection, so it depends on which documents survived and were located. Mazur is arguing about the justification Allen and Joslin gave, not running a comparison of outcomes, and the clinical claims of the era, on both sides, rest on anecdotal reports rather than controlled data. Allen’s own 1914 presentation gave no numbers and no count of animals, which limits what anyone can conclude from it now.
Related Studies
- The Internal Secretion of the Pancreas
- Insulin in America: A Right or a Privilege?
- 100 Years of Insulin: Why Is It So Expensive?
- Frederick Banting: Discoverer of Insulin
- Episode 24: The Discovery of Insulin
FAQs
How long could patients survive on the Allen Starvation Diet?
Mazur does not put a survival figure on the regimen. He notes that at the time, most child diabetics died from coma within months to a few years of diagnosis, and that it is difficult to judge how much Allen’s regimen prolonged the life of his most famous patient, Elizabeth Hughes.
Did any patients recover using starvation diets?
Allen and Joslin promoted the diets not as a cure but for relief of symptoms and maximum extension of life. Fasting cleared glycosuria and reduced glucose, and the disease remained. Elizabeth Hughes was freed of glycosuria under Allen’s care and still deteriorated to 45 pounds by the winter of 1921/22, before insulin reversed her decline.
How did families cope with watching their children starve?
Joslin’s own retrospective is the plainest account: Joslin wrote of literally starving the child and adult with the faint hope that something new in treatment would appear, and of how it was no fun to starve a child to let him live. Families faced a diet that was unpleasant, hard to maintain, and enervating, and many patients withdrew from it. Elizabeth Hughes’s mother went directly to Frederick Banting to get her daughter into the insulin trials.
Conclusion
The starvation era in diabetes care is usually told as a story of desperate physicians doing the only thing available. Mazur’s review complicates that. Low carbohydrate diets with more fat were already in use, critics claimed comparable results without near-total privation, and the animal work said to justify starving patients amounts, in the public record, to a brief anecdotal report. Insulin ended the argument in 1922 and turned a disease that killed children within months to a few years into one they could live with, which is why keeping insulin accessible still matters.
