For most of human history, a diagnosis of what we now call Type 1 diabetes was a death sentence, typically carried out within weeks to months. Children wasted away. Adults grew skeletal. The only treatments available amounted to variations on dietary restriction, and while these could buy some time, they could not replace what the body had lost: the ability to produce insulin. Type 2 diabetes, by contrast, progressed more slowly, and many people with milder forms lived for years with the condition, sometimes without ever knowing they had it. The distinction between these two diseases was not clearly understood until the twentieth century, which makes the pre-insulin story messier and more tragic than a single narrative can capture.
Recognizing the Disease Without Understanding It
Physicians noticed diabetes thousands of years before anyone understood what caused it. Ancient Egyptian medical texts from around 1500 BCE describe a condition involving frequent urination and weight loss. By the first century CE, Greek and Roman doctors had recorded its clinical features and fatal course with surprising accuracy. Galen, writing in the second century, attributed the disease to a weakness in the kidneys, reasoning that the organs had lost the ability to retain fluids properly.1PubMed Central. A history of diabetes mellitus or how a disease of the kidneys evolved into a kidney disease For centuries, that was the dominant theory: diabetes was a urinary disorder, not a metabolic one.
Ancient Indian physicians independently noticed something Galen had not emphasized: the urine of affected patients tasted sweet. This observation resurfaced across cultures and centuries, noted by the Persian polymath Avicenna around 1000 CE and by European anatomists much later.1PubMed Central. A history of diabetes mellitus or how a disease of the kidneys evolved into a kidney disease In the seventeenth century, the English physician Thomas Willis formalized this observation by adding “mellitus,” from the Latin for honey, to the disease’s name.2Europe PMC. Milestones in the history of diabetes mellitus: The main contributors That sweet urine was glucose spilling over because the body could not use it, but no one had the tools to explain that mechanism yet. For patients, the practical consequence was simple: doctors could identify the disease, but they had almost nothing to offer.
The Dietary Approach and Why It Sometimes Worked
With no understanding of insulin or blood sugar regulation, pre-modern physicians gravitated toward the one lever they could pull: food. If sugary urine meant something was wrong with digestion, maybe controlling what went in would control what came out. Various dietary regimes were tried across centuries, ranging from all-meat diets to milk-based fasting to reduced-carbohydrate plans. Some of these accidentally worked for people with Type 2 diabetes, because cutting carbohydrates does reduce the demand on whatever insulin the body still produces. For Type 1 patients, whose bodies produce essentially no insulin at all, the benefit was marginal at best.
By the early 1900s, dietary intervention had become more systematic. In 1903, the German physician Carl von Noorden introduced what became known as the oatmeal cure: a regimen built around oatmeal, butter, and vegetable protein.3JAMA. THE OATMEAL DIET IN THE TREATMENT OF DIABETES MELLITUS The logic was that oatmeal, being a slowly digested complex carbohydrate, might be tolerated better than refined sugars or starches. This approach had some real utility for people with milder diabetes. Their blood sugar could be partially managed by controlling the type and amount of carbohydrate they consumed, and oatmeal days became a recognized clinical tool that persisted well into the twentieth century.4PubMed. Short-Term Dietary Oatmeal Interventions in Adults With Type 2 Diabetes: A Forgotten Tool
But for severe cases, gentler dietary strategies were not enough. That desperation drove a more extreme approach.
Starvation Diets and the Allen Plan
In the decade before insulin’s discovery, two prominent American physicians, Frederick Allen and Elliott Joslin, became the leading voices for an approach that was exactly what it sounds like: near-total starvation. Allen’s method, developed around 1910 and refined over the following years, involved fasting patients until their urine tested free of sugar, then reintroducing food in tiny increments. Some patients were kept on as few as 400 to 500 calories a day for extended periods.5PubMed Central. Why were “starvation diets” promoted for diabetes in the pre-insulin period?
The rationale was grim but internally consistent. If the body could not handle sugar, and if eating any food eventually produced sugar in the blood, then the only way to keep blood sugar manageable was to eat almost nothing. Allen and Joslin believed this was a genuine life-extension strategy: better to live thin and weak than to die quickly from uncontrolled diabetes. And for some patients, it did extend life, possibly by months or in rare cases a year or two beyond what an unrestricted diet would have allowed.
The human cost, though, was devastating. Patients on the Allen plan were emaciated, constantly hungry, and physically fragile. Photographs from the era show children who look like famine victims. Critics at the time called the approach “starvation dieting,” and the label was not unfair: some patients did starve to death on the treatment itself.5PubMed Central. Why were “starvation diets” promoted for diabetes in the pre-insulin period? The question families faced was whether death by diabetic coma or death by wasting was more humane. Many chose to follow the plan because it was the only option that offered any hope at all. Joslin kept meticulous records, and his data showed that patients who adhered strictly to the diet survived longer on average than those who did not, but “longer” often meant months, not decades.
What the starvation era exposed, more than anything, was how different the two forms of diabetes were. Patients with the milder form (what we now call Type 2) could sometimes maintain a functional, if miserable, existence on restricted calories. Their bodies still made some insulin, and reducing the metabolic load could keep things in check. Patients with the severe form (Type 1) were fighting a losing battle regardless. Their bodies made no insulin, and no amount of caloric restriction could substitute for a hormone they simply did not produce.
Herbal Remedies and Early Drug Experiments
Diet was not the only strategy people tried. For centuries, folk medicine traditions in Europe and Asia included plant-based treatments for “sugar sickness.” One plant in particular, Galega officinalis, commonly known as goat’s rue, had been used in European herbal medicine for centuries. It turned out to contain guanidine, a compound that was shown in 1918 to lower blood glucose.6PubMed. Metformin: historical overview This was a genuinely important finding, and chemists quickly began synthesizing guanidine derivatives to use as medications. Some of these drugs were tested in diabetic patients during the 1920s and 1930s, but most were abandoned because of serious toxicity.
One of those guanidine derivatives was metformin, now the most widely prescribed diabetes drug in the world. It was synthesized early in this wave of research but was not developed for clinical use until decades later. The irony is worth noting: the chemical ancestor of a modern first-line treatment existed before insulin did, but the science was not yet sophisticated enough to separate the helpful compounds from the toxic ones.6PubMed. Metformin: historical overview
Other pre-insulin remedies were less scientifically grounded. Opium was prescribed by some physicians to manage symptoms, particularly the intense thirst and frequent urination. Various mineral waters, tonics, and patent medicines circulated as supposed cures. None of them addressed the underlying metabolic problem. At best they provided comfort; at worst they delayed what little effective treatment existed.
Connecting the Pancreas to the Disease
The conceptual breakthrough that eventually made insulin possible came in 1889, when two researchers in Strasbourg, Oskar Minkowski and Josef von Mering, surgically removed the entire pancreas from a dog. The experiment was not actually designed to study diabetes. But afterward, a lab assistant noticed that the dog was urinating constantly and could not be housebroken no matter what. Minkowski followed a hunch, tested the urine for sugar, and found glucose levels above ten percent.7PubMed Central. I-European research, the cradle of the discovery of the antidiabetic hormone: the pioneer roles and the relevance of Oskar Minkowski and Eugène Gley The dog had developed severe diabetes within days of losing its pancreas. Minkowski confirmed the result by repeating the surgery in several more animals.
This established that the pancreas was central to sugar metabolism, but it did not explain how. Over the next decade, pathologists began examining the pancreases of patients who had died of diabetes and found damage concentrated in specific clusters of cells, which had been identified by the anatomist Paul Langerhans. These cell clusters, now called the islets of Langerhans, appeared to degenerate in diabetic patients while the rest of the pancreas remained relatively intact.8PubMed Central. ON THE RELATION OF CHRONIC INTERSTITIAL PANCREATITIS TO THE ISLANDS OF LANGERHANS AND TO DIABETES MELUTUS The pattern suggested that something produced by these specific cells was missing in diabetes.
The race was then on to isolate whatever substance those cells produced. Researchers knew the pancreas was the key organ, they suspected the islets were the key tissue, and they had good reason to believe that some kind of internal secretion was involved. But extracting it proved extremely difficult because the pancreas also produces digestive enzymes that destroy proteins, and the mystery substance was itself a protein.
The Experiments That Almost Got There
Before Banting and Best made their famous breakthrough in 1921, other researchers came tantalizingly close. In 1915, Israel Kleiner, working in New York, published results showing that pancreatic extracts could lower blood glucose in normal dogs and in dogs whose pancreases had been removed. By 1919, he had demonstrated that intravenous administration of a pancreatic emulsion achieved a significant drop in blood sugar in depancreatized dogs without causing toxic side effects.9PubMed Central. The Nobel Prize of Physiology or Medicine, 1923: controversies on the discovery of the antidiabetic hormone
Kleiner’s work was promising, but he did not have the resources or institutional support to push the research to a clinical stage. Several other groups in Europe had also produced partially purified pancreatic extracts that showed glucose-lowering effects in animals. The frustrating reality is that the basic principle was demonstrated multiple times in laboratories before anyone managed to create a preparation pure enough and safe enough to inject into a human patient. The gap between “this works in a dog” and “we can treat a dying child” turned out to be enormous, involving problems of purification, dosing, and manufacturing that required the collaboration of biochemists, clinicians, and eventually the pharmaceutical industry.
When Banting and Best did succeed in 1921, with the biochemist James Collip producing a purified extract, and when Leonard Thompson became the first human patient successfully treated in January 1922, the era that historians have called the “frustration era” finally ended.10PubMed Central. The Discovery of Insulin: An Important Milestone in the History of Medicine Children who had been wasting away on starvation diets gained weight within weeks. The transformation was so dramatic that early photographs of insulin-treated patients, before and after, remain some of the most striking images in the history of medicine.
Why Type 2 Patients Had Better Odds
One reason the pre-insulin story is more nuanced than “everyone died” is that the two main forms of diabetes behave so differently. What we now call Type 2 diabetes typically develops in adulthood and involves the body’s cells becoming resistant to insulin rather than losing the ability to produce it entirely. Many people with Type 2 diabetes in the pre-insulin era would have experienced gradual symptoms: increased thirst, frequent urination, slow wound healing, blurred vision. Some would have managed these symptoms for years or even decades through dietary changes, physical labor, and plain luck.
In an era without blood tests or widespread medical screening, many people with mild Type 2 diabetes probably lived and died without ever receiving a formal diagnosis. Their condition might have been attributed to aging, obesity, or general ill health. The complications that eventually killed them, such as heart disease, kidney failure, or infections that would not heal, were common enough in the general population that no one necessarily connected them to a single underlying metabolic disease.
Type 1 diabetes told a completely different story. It often struck children or young adults suddenly and progressed rapidly. Without insulin, the body cannot use glucose for energy, so it breaks down fat and muscle instead, producing toxic byproducts called ketones. This process, now known as diabetic ketoacidosis, leads to coma and death within days to weeks if untreated. Even with the best available pre-insulin care, including Allen’s starvation regimen, most Type 1 patients survived only one to three years after diagnosis.
What Families Actually Faced
The medical literature can make this era sound clinical and distant, but the lived experience was horrifying in ways that statistics do not capture. A family whose child developed Type 1 diabetes in 1915 faced an immediate and ongoing crisis. The child would be desperately thirsty, losing weight rapidly, and producing enormous volumes of urine. A doctor, if consulted, could confirm the diagnosis by tasting or testing the urine for sugar. The treatment plan, if the family was fortunate enough to see a specialist like Allen or Joslin, involved putting the child on a diet so restrictive that the child would essentially be starving.
Parents had to enforce these diets against children who were begging for food. Locks on pantries and iceboxes were common in households with diabetic children. Some families hired nurses specifically to prevent the child from sneaking food. The psychological toll on both the child and the parents was enormous, and the knowledge that even perfect adherence would only delay the inevitable made it worse. When insulin arrived, families who had been living under this regime for months or years experienced something close to a miracle. But for those who died in the years just before insulin became available, the tragedy was compounded by how close they came to surviving.
The Geography and Class Divide
Access to even the limited pre-insulin treatments varied enormously by wealth and location. Allen’s program was based at the Physiatric Institute in New Jersey, a private facility that catered to patients who could afford extended residential care. Joslin practiced in Boston and saw a mix of patients, but his most intensive dietary management required close medical supervision that was simply unavailable to most people. The oatmeal diets, starvation regimens, and careful monitoring described in the medical literature of the era represent the best-case scenario: what happened to patients who had access to the most advanced physicians of their time.
For a farmer in rural Appalachia, a laborer in industrial Manchester, or a villager in colonial India, the diagnosis played out with far less medical sophistication. Local doctors might recognize the symptoms but had little to offer beyond general dietary advice. Patent medicines and folk remedies filled the gap. In many cases, people simply wasted away and died without anyone fully understanding what had killed them. The pre-insulin era was not a single story; it was many stories, shaped by class, geography, and the random chance of which physician a patient happened to see.
How Goat’s Rue Led to Modern Medication
The connection between the pre-insulin herbal tradition and modern pharmacology is more direct than most people realize. Goat’s rue had been used in European folk medicine for symptoms we now associate with diabetes long before anyone isolated the active compounds. When researchers identified guanidine as the glucose-lowering agent in 1918, they opened a research path that would eventually produce three families of synthetic drugs: the biguanides, the diguanides, and the monoguanides. Several of these were tried clinically in the 1920s and 1930s, but insulin’s dramatic success overshadowed them, and their toxic side effects gave doctors little reason to keep experimenting.6PubMed. Metformin: historical overview
Metformin, which belongs to the biguanide family, was shelved for decades before being revived in the 1950s by the French physician Jean Sterne, who recognized that it was far less toxic than its chemical relatives. It took until the 1990s for metformin to reach the American market, almost eighty years after the guanidine connection was first established. Today it is recommended as first-line therapy for Type 2 diabetes worldwide, and researchers are investigating its potential benefits for aging, cancer prevention, and cardiovascular disease. The pre-insulin herbalists who prescribed goat’s rue were, in a sense, stumbling onto a genuine pharmacological principle. They just lacked the tools to refine it.
What Changed Overnight in 1922
The arrival of insulin did not solve diabetes, but it transformed the disease from an acute death sentence into a chronic condition that could be managed. The first human treated with purified insulin, Leonard Thompson, was a fourteen-year-old boy who weighed just 65 pounds and was near death on Allen’s starvation diet. Within weeks of starting insulin injections, he gained weight, regained energy, and his urine cleared of sugar. He lived another thirteen years before dying of pneumonia, likely complicated by the long-term vascular damage that diabetes causes even with treatment.
Thompson’s story captures both the triumph and the limitation of insulin. It saved lives that would otherwise have ended quickly, but it introduced a new set of challenges: the need for daily injections, the risk of hypoglycemia from too much insulin, the financial burden of an ongoing pharmaceutical requirement, and the long-term complications that accumulate over decades of living with diabetes. Parents of children with Type 1 diabetes today still navigate a demanding care regimen that affects career decisions, family finances, and daily routines in profound ways.11PubMed Central. Caregiver burden among parents of children with type 1 diabetes: A qualitative scoping review The disease is manageable now, but it is never easy. Before 1922, it was not manageable at all.