What’s the Best Blood Sugar for Older Adults? (Know the Real Truth)

Added: 2026-05-11

[00:00:00]In 2008, a major diabetes trial was stopped early because the group achieving the tightest glucose control A1C below 6% was dying at a higher rate than the group with standard control.

[00:00:15]The trial was called Accord. The finding has been debated since the day it published. What hasn't been debated?

[00:00:23]serious hypoglycemic events requiring outside assistance were significantly more common in the intensive group. That finding has held across every subsequent analysis. What I want to sit with for a moment before getting to what accord means for targets, before getting to the numbers is the specific population that finding matters most for, not middle-aged adults with recently diagnosed type 2 diabetes. Older adults, people who have been managing diabetes for 15 or more years. These are people whose physiology responds to tight glucose control differently than the populations those original targets were built from. Because when you apply a chords hypoglycemia signal to that specific group, the clinical picture changes substantially. That's what this video is actually about. not accord as a historical controversy. Accord as a window into a specific physiological reality that most older adults managing diabetes have never had explained to them and that most quarterly appointments aren't designed to surface.

[00:01:41]Everything here is educational, not a replacement for a direct conversation with your own physician about your specific situation, medications, and targets. If you know your most recent A1C, drop it in the comments now. Just the number. And if you've had falls, add yes. Before the mechanism, I want to be specific about who I'm talking about because this isn't a universal statement about diabetes management. The patient I'm thinking about has been managing type 2 diabetes for more than a decade.

[00:02:17]Their A1C has been consistently well controlled, under seven, sometimes under 6.5. They take their medication as prescribed. their doctor is pleased with their numbers at every quarterly visit.

[00:02:32]On paper, this patient is a success story. What the paper doesn't show, whether this patient still feels their lows, whether the shaking and sweating and urgent hunger that used to signal a drop in blood sugar still arrive when blood sugar drops. Whether the alarm system that was supposed to catch a dangerous glucose event is still functioning. That's the clinical question that almost never gets asked at a standard diabetes appointment. Not because physicians aren't thorough, because the question requires knowing that losing those warning symptoms is a clinical event, not just getting used to medication. and that it changes what safe glucose management looks like for that specific patient. What that patient is actually living looks something like this. They mention fatigue at the quarterly visit and the doctor nods.

[00:03:35]Understandable at your age. They had a fall two months ago and described it as just lost my footing because they had no other explanation. The chart at that appointment read A1C 6.8. wellont controlled patient doing well.

[00:03:51]Everything the chart recorded was accurate. What it could not record was the silence. The mechanism behind hypoglycemia unawareness is one thing, not three. Understanding it fully changes how you read every other piece of information in this video. When blood sugar drops below safe threshold in a person with a functioning glucose alarm system, the nervous system fires adrenaline. Heart rate rises, sweating starts, hunger hits urgently. That response is calibrated. It fires at a specific glucose level because the body has learned that level represents danger. What glucose lows do repeated over time is re-calibrate that threshold. Every time blood sugar drops and the body manages through it without intervention, whether the person felt it or not, the nervous system adjusts. It records that the previous level was survivable. It shifts the trigger point slightly downward. One episode does almost nothing. 50 episodes over 5 years, shifts the threshold measurably.

[00:05:08]15 years of tight control with regular mild lows, many of them asymptomatic, many of them nocturnal, none of them recorded anywhere, can shift it far enough that the alarm stops firing at levels that are clinically dangerous. To be specific about what that progression looks like in glucose numbers, a functioning alarm typically fires somewhere around 70 milligrams per deciliter. In a patient with partial unawareness, that threshold may have shifted to 60. In a patient with severe unawareness, it can drop to 50 or even 45. The body has adapted to surviving at levels that should be triggering an emergency response. It simply no longer recognizes them. The end state is a patient whose blood sugar can drop into the 50s while they feel nothing. No shaking, no sweat, no hunger signal.

[00:06:11]They feel fine. The first indication that something went wrong is a fall or a confusion that clears an hour later or not waking the way they should. This is not a rare complication. Research suggests it affects roughly 40% of older adults who have lived with diabetes for more than 15 years. There's a question you can answer right now before any appointment, before any test that gives you a direct window into where your warning system stands. Think back to the last time your blood sugar ran low. Not a severe episode, a mild drop after a walk or a late meal. Did you feel it arriving? Did the shakiness come, the urgent hunger, the sweat? If yes and those signals still arrive reliably, your alarm system is functioning. That doesn't mean your target is correct, but you have a buffer that will catch a dangerous drop before it becomes a fall.

[00:07:16]If the answer is no, if lows no longer feel the way they used to, or if the only way you know blood sugar dropped is because the meter told you afterward, that is a clinical finding, not adaptation, not getting used to the medication, a measurable change in your physiology that belongs explicitly in the conversation about what target is safe for you specifically. Write that answer down. It is the most important piece of information your doctor doesn't have. The people in that 40% don't know they're in it. There is no standard test for it at a routine appointment. A clinician would have to ask directly.

[00:08:02]When your blood sugar runs low, do you still feel the shakiness and hunger you used to? Most don't ask. Most patients don't volunteer it because losing those symptoms doesn't feel like a clinical event. It feels like adjustment. Once you understand that threshold recalibration is happening silently in a significant portion of older adults on long-term diabetes medication, the question of what A1C to target stops being a guideline question and becomes a specific physiological one. How much buffer does this patient need against a low their body will no longer warn them about? That's the reasoning behind why the American Diabetes Association's current standards stratify targets for older adults rather than applying one universal number. For older adults who are generally healthy, living independently, cognitively intact, the target is under 7.5, not under seven. The half point isn't a loosened standard. It's a physiological buffer for a warning system that may no longer be fully reliable. A smaller meal, a longer walk, reduced appetite on a given day, any of those compress the margin. The 7.5 target keeps that margin intact. For older adults carrying multiple chronic conditions, functional limitations, a more complex medication picture under eight, the cardiovascular risk reduction in a 76 year old with three comorbidities is smaller than the hypoglycemia risk introduced by pushing for 6.9.

[00:09:58]The mechanism threshold recalibration is the reason that arithmetic doesn't favor the lower number for older adults who are frail, managing cognitive decline or carrying serious concurrent illness under 8.5. At this stage, a hypoglycemic fall carries consequences. Fracture, surgery, loss of independent living that are more immediate and more certain than anything a moderately elevated A1C produces across the same time frame. One thing worth naming before moving to the numbers themselves. For many older adults, the target currently being managed toward was set years ago at a different weight, a different level of kidney function, a different activity level, and has simply persisted without formal review, not adjusted in response to a reading, never reassessed from the ground up against today's physiology. On daily readings, the ADA recommends fasting blood sugar between 80 and 180 for older adults. That range is wide deliberately, but where you consistently sit within it matters. Running at 85 to 95 on an aggressive medication regimen eliminates the buffer the range was designed to provide. The range gives you room. the low edge of it does not. One additional assessment worth naming explicitly for patients with significant anemia or chronic kidney disease, whether the A1C they're being managed to is still giving accurate information.

[00:11:47]Altered red blood cell lifespan from either condition can make A1C read lower or higher than actual average glucose. A number that looks controlled may not be describing the glucose pattern we think it is. Before getting to the monitoring tool, one piece of clinical information that belongs in this conversation and almost never arrives there directly. Not all diabetes medications carry the same hypoglycemia risk and which one you're on changes how urgently the target conversation needs to happen. Metformin alone does not cause hypoglycemia. It works by reducing liver glucose output and improving insulin sensitivity. It lowers blood sugar but doesn't push it below safe levels independently. A patient on metformin only with an intact warning system has a meaningfully different risk profile than a patient on combination therapy. Sulphonilaras, glyazide, glimeride, giber stimulate the pancreas to release insulin regardless of what blood sugar is doing at that moment. They push glucose down whether a meal is coming or not. In older adults with threshold recalibration already in progress, sulfonil uras are the medication most consistently associated with undetected nocturnal lows. If you're on one, your target is not a passive number your doctor manages. It's an active safety variable that requires explicit discussion. Gibberide specifically is now considered inappropriate for older adults by the American Diabetes Association due to its long duration of action and the accumulation of active metabolites in the body over time. If you're on guride, that conversation about medication class is overdue. Insulin carries the highest hypoglycemia risk, and its interaction with a degraded warning system is the most immediately dangerous combination in older adult diabetes management.

[00:14:09]Doses set at one point in life aren't automatically correct at 75. Newer medication classes, SGLT2 inhibitors, GLP-1 receptor agonists, carry substantially lower hypoglycemia risk and are increasingly used in older adults for that reason. They're not appropriate for every patient, but if you're on a sulfonil ura or insulin and that conversation about medication class has never happened, it's overdue. Here's what two weeks of continuous glucose data showed in a patient pattern that repeats across older adults on combination therapy. A 71-year-old on metformin and glyipazide A1C of 7.1.

[00:14:56]Last office note reading satisfactory control. Glucose drop below 70 mgs per deciliter on 11 of 14 nights. Always between 1:00 a.m. and 4:00 a.m. Always asymptomatic. Always resolving before morning. Daytime readings were stable.

[00:15:18]The A1C accurately reflected the daytime average. 11 nocturnal lows in two weeks.

[00:15:26]None felt, none reported, none recorded anywhere in the medical chart. The A1C said 7.1. The A1C was telling the truth about daytime glucose. It had no mechanism for telling the truth about what was happening at 2 am. A continuous glucose monitor, a sensor worn on the upper arm or abdomen, reading blood glucose every few minutes around the clock, is what surfaces that data. No finger sticks on the phone screen. what this data looked like. A graph line running stable through the evening, then dropping sharply into the red zone between 1 and 3:00 a.m. 11 times across 14 nights before recovering before dawn.

[00:16:16]Nothing on that graph was visible in any quarterly lab result. The pattern visible only across 14 continuous days.

[00:16:26]glucose falling in the same 2hour window every night corresponding to when a medication from the evening dose reaches peak activity. That connection between medication timing and nocturnal lows in a patient with no remaining warning symptoms is the clinical finding that changes the management picture. Not the A1C, not the quarterly visit. What I'm looking for when I review this data, any reading below 70 is a low, regardless of whether it was felt. Below 54 is a serious low. If two weeks shows readings below 70 with no corresponding symptoms, no waking, no sweat, no hunger, that single finding is more actionable than anything else in the glucose picture. It confirms that the alarm is no longer functioning. Glucose variability matters independently of where the average lands. An A1C of 7.2 produced by readings swinging between 58 and 220 across the same week is a different clinical reality than an A1C of 7.2 produced by stable readings between 90 and 150. The number is identical. The physiological reality it summarizes is not on access. Medicare covers CGM for older adults with diabetes on insulin. If you're on insulin or a sulfanilure, ask your doctor directly whether you qualify. The Freestyle Libra is available over the counter at mostarmacies without coverage. 14 days of continuous data shows what years of quarterly testing was structurally unable to show. The appointment conversation that doesn't open anything useful goes like this. The patient mentions afternoon fatigue a fall last month. A general sense that something feels off. The physician checks the chart. A1C 6.9 blood pressure stable. Medication unchanged. The response, "You're doing well. Keep it up." The patient leaves with the same management plan and a faint sense that what they described didn't land anywhere. What broke down wasn't the physician's attention. It was that the symptoms arrived as separate complaints rather than as a connected clinical question. Fatigue, a fall, a vague sense of something being off. None of those individually generate a glucose reflex.

[00:19:16]Connected explicitly to blood sugar management and to the specific question of whether lows might be occurring without warning. They generate a completely different response. The question to bring in directly when my blood sugar runs low, do I still feel the symptoms I used to feel? If the answer is no or uncertain, ask whether hypoglycemia unawareness has ever been specifically assessed. For most patients on long-term diabetes management, it hasn't. Ask what medication you're on and whether it carries hypoglycemia risk. Ask when your current dose was last formally reviewed, not adjusted in response to a reading, but reassessed from the ground up against your current weight, kidney function, and activity level. If you're on a cell phona or insulin, ask whether alternatives with lower hypoglycemia profiles have been considered given current evidence on their use in older adults. Ask about your target directly. Is it still the right number? Here is where the clinical picture actually sits. And I want to be honest about it. What I don't know from here is where you sit in that picture.

[00:20:38]whether your warning system is intact, whether the target being managed toward was set when your physiology looked different than it does now, whether anyone has looked at what your glucose actually does at 2 a.m. Those are answerable questions, not automatically, but answerable if someone looks for the answers. Bring the selfassessment answer to the next appointment. Bring your medication name and ask directly about its hypoglycemia risk. If you're on insulin or a sulfonila, ask about CGM.

[00:21:16]If you have anemia or kidney disease, ask whether your A1C is still measuring what everyone thinks. Drop your current A1C in the comments if you know it. I want to see where people are actually being managed because the gap between what the evidence supports for older adults and what's happening in everyday appointments is wider than most people realize. And the comments make that visible. In the next video, I'm going to cover something that connects directly to what we discussed today. And it's a finding that surprised me when I first looked at the research carefully. There is a specific eating pattern that research shows can cut nocturnal glucose variability in older adults significantly.

[00:22:05]Not by restricting what you eat, but by changing when. That one is worth staying for. I'll see you in the next