Around 1970, a fasting blood sugar below roughly 100 to 110 mg/dL on a whole-blood test was generally considered normal, though the exact number depended on which laboratory method was used and which set of guidelines a doctor followed. That range sounds familiar to modern ears, but the comparison is misleading. The tests, the sample type, and even the definition of “diabetes” were so different from today’s standards that a 1970 reading and a 2024 reading cannot be placed side by side without serious translation. The story of what counted as normal blood sugar half a century ago is really a story about how messy and unstandardized diabetes diagnosis used to be.
How Blood Sugar Was Measured in 1970
Modern glucose meters and hospital analyzers use enzymatic methods that react specifically with glucose molecules. In 1970, the dominant laboratory techniques were chemical colorimetric assays with names like Folin-Wu, Somogyi-Nelson, and ortho-toluidine. These methods worked by reducing a reagent and producing a color change that could be measured, but they differed in what they actually detected. The Folin-Wu method, for example, reacted not just with glucose but with other reducing substances in the blood, including uric acid and creatinine, which inflated the reading. Somogyi-Nelson was somewhat more specific because it used a protein-precipitation step to remove some of those interfering substances, but it still wasn’t measuring glucose alone.
The practical result was that two labs running the same blood sample through different methods could return meaningfully different numbers. A comparative evaluation of these older techniques found that both the Modified Folin-Wu and ortho-toluidine methods showed substantial imprecision and upward deviation from true glucose values, with average deviations reaching about 6% and nearly 11% respectively.1PubMed Central. METHODS FOR ESTIMATION OF BLOOD GLUCOSE : A COMPARATIVE EVALUATION A reading of 120 mg/dL on a Folin-Wu assay might correspond to something closer to 110 mg/dL or less on today’s enzymatic glucose oxidase methods. Doctors working in 1970 knew these methods weren’t perfect, but they were what was available, and “normal” ranges were calibrated to the specific assay each hospital used.
The Whole Blood Problem
Today, when you get a fasting glucose test, the laboratory almost always measures glucose in your plasma or serum, which is the liquid portion of your blood after cells are separated out. In 1970, many tests were performed on whole blood, meaning the red blood cells were still in the sample. This matters because red blood cells contain less water than plasma does, and glucose dissolves in water. The result is that a whole-blood glucose reading runs about 10% to 15% lower than a plasma glucose reading from the same person at the same moment.2JAMA. Serum Glucose Concentrations
That gap sounds small, but it’s clinically significant. A whole-blood fasting glucose of 100 mg/dL in 1970 might be equivalent to roughly 110 to 115 mg/dL in plasma terms. A 1966 JAMA paper argued that glucose should ideally be measured in serum or plasma because whole-blood values fluctuated with hematocrit, the proportion of red cells in the blood, which varies from person to person.2JAMA. Serum Glucose Concentrations A later study confirmed that a direct mathematical relationship exists between plasma and whole-blood glucose, with plasma values running consistently higher regardless of the preparation method used.3PubMed. The difference between the glucose concentrations in plasma and whole blood The transition from whole blood to plasma as the standard specimen happened gradually through the 1970s and 1980s, which means that “normal” values from 1970 and “normal” values from today are not measuring the same thing in the same way.
There Was No Single Standard for Diagnosis
If you ask what blood sugar level triggered a diabetes diagnosis in 1970, the honest answer is that it depended on which doctor you saw and which expert committee’s opinion that doctor followed. There was no universally agreed-upon diagnostic threshold. Multiple professional groups had published their own criteria, and those criteria disagreed on nearly every detail: the glucose load used in a tolerance test, the timing of blood draws, and the cutoff values that separated “normal” from “diabetic.”
By the end of the 1960s, the medical community acknowledged that relying on fasting blood sugar alone was identifying people too late in the progression of diabetes. This led to at least six different published recommendations for how to conduct an oral glucose tolerance test, with glucose loads varying from 50 to 100 grams and no consensus on whether the dose should be adjusted for body weight.4PubMed Central. The Oral Glucose Tolerance Test: 100 Years Later – Section: OGTT – History A patient tested with a 50-gram glucose drink and a patient tested with a 100-gram drink would naturally produce different blood sugar peaks, so a “normal” two-hour value in one protocol might look abnormal in another.
For fasting blood sugar specifically, a common rule of thumb in the late 1960s and early 1970s placed the normal range for whole blood below about 100 mg/dL, with values above 130 mg/dL on repeated testing suggesting diabetes. But this left a wide gray zone between 100 and 130 that different authorities handled differently. Some called it “borderline diabetes” or “chemical diabetes,” vague labels that would later be replaced by the more precise term “impaired glucose tolerance.”
What Happened After You Drank the Sugar
The oral glucose tolerance test was the gold standard for catching diabetes that fasting blood sugar missed, and the values considered “normal” during such a test in 1970 were surprisingly lenient by current standards. A Japanese study from that era attempted to define the true normal range statistically by excluding people with clearly abnormal results and computing the 98th percentile for each age group. The researchers found that the one-hour post-load blood sugar level considered normal was around 180 to 190 mg/dL, and the two-hour level was about 140 to 150 mg/dL, with both values increasing as people got older.5Journal of Chronic Diseases. Studies on normal blood glucose level—Statistical approach to interpretation of glucose tolerance test
By today’s standards, a one-hour glucose of 180 mg/dL after a standard meal or glucose load would be considered elevated, and a two-hour value of 150 mg/dL would place someone squarely in the “impaired glucose tolerance” category, which is now a recognized pre-diabetes diagnosis. The same study noted that the normal ranges for fasting blood sugar and the three-hour value were close to or slightly lower than the criteria that were already in use, meaning that the accepted diagnostic thresholds in 1970 were roughly appropriate for fasting values but arguably too generous for the post-load readings.5Journal of Chronic Diseases. Studies on normal blood glucose level—Statistical approach to interpretation of glucose tolerance test
The age effect matters here. Research from that period showed that blood sugar after a glucose load climbed steadily with age, and 1970-era criteria generally accepted this as normal aging rather than a sign of disease. A 65-year-old with a two-hour glucose of 160 mg/dL might have been told they were fine. Today, that same reading would prompt further evaluation and possibly a pre-diabetes label. The question of whether rising blood sugar with age is truly “normal” or simply common has been debated ever since, and the answer has shifted firmly toward treating it as a risk rather than an inevitability.
The 1979 Overhaul That Changed Everything
The disorganized state of diabetes diagnosis lasted through most of the 1970s until the National Diabetes Data Group published a landmark set of recommendations in 1979. This report created a unified classification system and, for the first time, established widely adopted diagnostic thresholds for fasting glucose and the oral glucose tolerance test.6PubMed. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance The NDDG criteria set the diabetes threshold at a fasting plasma glucose of 140 mg/dL or higher (on more than one occasion) or a two-hour oral glucose tolerance test value of 200 mg/dL or higher. Values between normal and diabetic were designated “impaired glucose tolerance,” replacing the murky older labels.
Shortly after, the World Health Organization issued its own recommendations based on both prospective and cross-sectional epidemiological data, creating categories for normal subjects, those with impaired glucose tolerance, and diabetics.7PubMed. When is diabetes?–a new look at diagnostic criteria for diabetes mellitus The WHO criteria were broadly similar to the NDDG criteria, and together these two documents formed the basis of diabetes diagnosis worldwide through the 1980s and into the 1990s.
For perspective on how far the thresholds have moved: the NDDG’s 1979 fasting cutoff of 140 mg/dL was lowered in 1997 by the American Diabetes Association to 126 mg/dL, and the concept of “pre-diabetes” was formalized with a fasting glucose of 100 to 125 mg/dL. Someone with a fasting glucose of 135 mg/dL in 1980 would not have met the diabetes threshold. The same reading today is solidly in the diabetic range. The biology didn’t change; the understanding of where risk begins did.
Why the Cutoffs Shifted Downward
The diagnostic thresholds weren’t lowered on a whim. Epidemiological studies across multiple populations showed that the risk of diabetes-specific complications, particularly damage to the small blood vessels in the retina, began at glucose levels well below the old 140 mg/dL fasting cutoff. Analysis of population data identified glycemic thresholds for retinopathy starting around a fasting plasma glucose of 6.4 to 6.8 mmol/L (roughly 115 to 122 mg/dL) and a two-hour post-load glucose of 9.8 to 10.6 mmol/L (about 176 to 191 mg/dL).8PubMed Central. Glycemic thresholds for diabetes-specific retinopathy: implications for diagnostic criteria for diabetes The evidence was clear that the old threshold was letting people develop retinal damage before they were ever told they had diabetes.
The Framingham Heart Study, which had been tracking cardiovascular outcomes in a large American cohort since 1948, provided further motivation. By the late 1970s, Framingham data showed that diabetic men had roughly twice the rate of cardiovascular disease compared to non-diabetic men, and diabetic women had three times the rate.9PubMed. Diabetes and cardiovascular risk factors: the Framingham study A related analysis confirmed a two- to threefold increased risk of clinical atherosclerotic disease across 20 years of surveillance.10PubMed. Diabetes and cardiovascular disease. The Framingham study But the Framingham researchers also found that even impaired glucose tolerance, that in-between zone below the diabetic cutoff, was associated with elevated cardiovascular risk factors and increased disease incidence.11PubMed. Diabetes and glucose tolerance as risk factors for cardiovascular disease: the Framingham study The implication was stark: people the 1970-era criteria classified as “normal” or merely “borderline” were already accumulating vascular damage.
Gestational Diabetes Had Its Own Moving Target
Pregnant women faced a separate set of criteria that were even more inconsistent. The first widely adopted standards for gestational diabetes came from O’Sullivan and Mahan in 1964, who tested 752 pregnant women with a three-hour, 100-gram oral glucose tolerance test using the Somogyi-Nelson method on whole blood. If a woman exceeded two or more of the threshold values, she was diagnosed with gestational diabetes.12PubMed Central. Trends in the Diagnosis of Gestational Diabetes Mellitus
These criteria had a fundamental limitation that went unrecognized for decades: the cutoff values were chosen based on the mother’s future risk of developing diabetes after pregnancy, not on outcomes for the baby. A woman’s blood sugar could be high enough to affect fetal growth and delivery complications but still fall below the O’Sullivan thresholds, because those thresholds were calibrated to a different endpoint entirely.12PubMed Central. Trends in the Diagnosis of Gestational Diabetes Mellitus It would take decades, and a large international study completed in the 2000s, before gestational diabetes criteria were recalibrated around fetal outcomes like birth weight and neonatal complications. In 1970, a pregnant woman with modestly elevated blood sugar might have been told her values were acceptable, when in fact her baby was being affected.
The original O’Sullivan criteria also had a technical shelf-life problem. They were developed using whole blood and the Somogyi-Nelson chemical method. As laboratories transitioned to plasma specimens and enzymatic glucose assays, the original numbers no longer applied directly, but many clinicians kept using them anyway, sometimes with ad hoc mathematical adjustments. This patchwork approach persisted well into the 1990s in some settings.
HbA1c Didn’t Exist as a Clinical Tool Yet
Today, the hemoglobin A1c test is one of the primary tools for diagnosing diabetes and monitoring long-term blood sugar control. It measures the percentage of hemoglobin proteins in your red blood cells that have glucose attached to them, giving an average picture of blood sugar over the preceding two to three months. In 1970, this test was not available to clinicians. Glycated hemoglobin was first used in routine clinical laboratories around 1977, and even then it took years before the assays were reliable enough for widespread diagnostic use.13PubMed Central. HbA1c standardisation: history, science and politics
The absence of HbA1c in 1970 is significant because it meant doctors had no way to assess what a patient’s blood sugar had been doing over time. A fasting glucose test captures a single snapshot, and it can be influenced by what the patient ate the day before, how well they slept, whether they were stressed, or whether they were fighting an infection. Two patients with the same fasting glucose of 95 mg/dL could have very different day-to-day glucose patterns: one might run steady and stable, while the other might spike to 250 mg/dL after every meal and simply happen to be at 95 the morning of the test. Without HbA1c, the second patient would look identical to the first.
HbA1c was eventually incorporated into diagnostic criteria in 2010, when the American Diabetes Association added an A1c of 6.5% or higher as a standalone diagnostic criterion for diabetes. The WHO adopted a similar threshold shortly after. Population studies confirmed that the glycemic threshold at which retinopathy risk rises corresponds to an A1c of roughly 6.3% to 6.7%.8PubMed Central. Glycemic thresholds for diabetes-specific retinopathy: implications for diagnostic criteria for diabetes This gave clinicians a tool that smooths out day-to-day variation and catches people whose average glucose is problematic even if their fasting value looks fine. Nothing remotely like it was available in 1970.
What This Means If You’re Comparing Old and New Results
People sometimes encounter blood sugar values from old medical records, family histories, or vintage health literature and try to compare them with today’s numbers. If you find a 1970-era result, several adjustments would be necessary to make it roughly comparable to a modern reading. If the test was run on whole blood, the equivalent plasma value would be about 10% to 15% higher. If the test used a non-specific chemical method like Folin-Wu, some of the measured “glucose” was actually other substances, so the true glucose was somewhat lower than the reported number. These two factors push in opposite directions and partially cancel each other, but not perfectly and not predictably, because the magnitude of each correction depends on the specific assay and the individual patient’s blood composition.
The takeaway is that an isolated number from 1970 cannot be reliably converted into a modern-equivalent value. What can be said is that the general boundary of normal fasting blood sugar hasn’t moved as dramatically as the diagnostic threshold for diabetes. A fasting level below 100 mg/dL was considered reassuring then, and it’s still considered normal today (now measured in plasma with enzymatic methods). The bigger shift has been in recognizing that the zone between “clearly normal” and “clearly diabetic” matters. In 1970, a fasting glucose of 120 mg/dL on a whole-blood test might have prompted a shrug and a note to recheck in a year. Today, an equivalent plasma value would trigger conversations about lifestyle changes, screening for complications, and possibly medication.
The Rise of Home Glucose Monitoring
In 1970, the only way to measure blood sugar was in a laboratory. Patients with known diabetes monitored their condition indirectly through urine glucose testing, which was crude and imprecise. Urine only shows glucose when blood levels exceed the kidney’s reabsorption threshold, typically around 180 mg/dL, meaning a patient’s blood sugar could be running dangerously high at 170 mg/dL and produce a completely negative urine test. Conversely, some patients spill glucose into their urine at lower thresholds than average, creating false alarms.
The first research on using enzyme-based electrochemical sensors to measure blood glucose was demonstrated around 1970, laying the groundwork for what would eventually become the portable glucose meter.14PubMed Central. Glucose Biosensors: An Overview of Use in Clinical Practice – Section: 3.2. Second-generation of Glucose Biosensors But commercially available home meters didn’t reach patients until the late 1970s and early 1980s, and even those early devices were expensive, bulky, and required careful technique. The transformation of diabetes management into something a patient could track in real time, multiple times a day, was still years away. In 1970, “knowing your blood sugar” meant getting a lab test done at a hospital or clinic, waiting for results, and learning a number that reflected one moment in time measured by one of several imperfect methods. The concept of a personal, continuous understanding of your glucose patterns simply didn’t exist.