Daniel sent us this one — he recently came across someone unresponsive on a sidewalk and it got him thinking about first aid for intoxication. Not just the obvious stuff, but the syndromes people miss. Alcohol poisoning, drug poisoning, the really dangerous combinations like alcohol and benzodiazepines. His question is straightforward: if you encounter someone who's clearly intoxicated beyond just lightly drunk, what are the initial first aid steps, and what are the signs you need to escalate for emergency help immediately? There's a lot to unpack here.
I'm glad he asked, because most people's mental model of "too drunk" is someone throwing up in a bush at a party. The gap between that and a life-threatening CNS depression event is narrower than people realize. We're talking about a continuum where crossing the line can happen without anyone noticing until it's too late.
CNS depression being...
Central nervous system depression. Your brain is literally slowing down the signals that keep you breathing. Alcohol does it. Benzodiazepines do it. Opioids do it. Combine them, and you're not adding risks — you're multiplying them. The respiratory drive can just...
The person who looks peacefully asleep might actually be dying.
That's exactly the scenario emergency physicians lose sleep over. The "found down" patient who everyone assumed was just sleeping it off. There was a case in Boston a few years ago — college student, roommates put him in the recovery position, which was good, but they didn't realize his respiratory rate had dropped to six breaths per minute. By the time someone checked properly, he was in respiratory arrest.
Six breaths per minute. What's normal?
Twelve to twenty for an adult at rest. Below eight is a blaring alarm. Below six, you're watching someone's brain forget how to do the one thing it's supposed to never forget how to do.
Let's walk through this methodically. Someone's clearly intoxicated — could be alcohol, could be something else, could be both. You don't know them, you don't know what they took. What's step one?
Scene safety, always. Don't become patient number two. But assuming the scene is safe, the very first assessment is responsiveness. And I mean actually assessing it — not just shouting "hey buddy" from six feet away. You get close, you introduce yourself, you ask if they can hear you. If no response, you do a sternal rub.
Knuckles on the sternum, firm pressure. It's uncomfortable. Someone who's just deeply asleep will respond. Someone who's in a stupor or coma from CNS depression won't. That's your first decision point. If they don't respond to painful stimuli, you call emergency services immediately. Don't wait. Don't see if they come around. That's a nine-one-one call right there.
If they do respond?
Then you're gathering information. But here's where things get tricky — someone can be responsive and still be in serious danger. Responsiveness just means their brain is still talking to their body. It doesn't mean their respiratory drive is adequate, or that they're not going to deteriorate.
What are you looking for beyond just "are they awake"?
Three things that people consistently miss. First, respiratory rate. Actually watch their chest rise and fall and count for a full thirty seconds and multiply by two. Don't eyeball it for five seconds and guess. Second, skin signs — are they cold and clammy? Is there cyanosis, which is a bluish tint around the lips or nail beds? That means oxygen isn't getting where it needs to go. Third, pupillary response. If you have a small flashlight or even your phone light, check whether their pupils constrict. Pinpoint pupils that don't react — that's a classic opioid overdose sign. But dilated, sluggish pupils can also be a red flag with certain substances or with advancing CNS depression.
The average person is supposed to know all this?
The average person can learn the basics in about ten minutes, and those ten minutes might save someone's life. I'm not saying everyone needs to be a paramedic. I'm saying there are maybe five things to check, and if any of them are wrong, you escalate. That's the whole framework. Check breathing rate. Check skin color and temperature. Check responsiveness to stimuli. If any of those fail, call for help.
What about the airway specifically? You mentioned it but let's get concrete.
The airway is the thing that kills people fastest. Someone who's intoxicated can lose the muscle tone that keeps their airway open. Their tongue can fall back and obstruct their breathing. They can vomit and aspirate — that's when stomach contents go into the lungs. Aspiration pneumonia is a common cause of death in overdose cases, and it can happen hours after the initial event.
The recovery position is about protecting the airway.
And it's not just "roll them on their side." There's a specific technique. You want them on their side with their head tilted back slightly to open the airway, the top leg bent at the knee for stability, and the bottom arm extended out so they're not rolling onto their face. The goal is that if they vomit, gravity drains it out instead of it going into their lungs.
I've seen people put a backpack or something behind someone to keep them on their side. Is that right?
It's a reasonable improvisation, but it's not the primary technique. The bent knee is usually enough. The bigger issue is that people put someone in the recovery position and then walk away thinking the job is done. The recovery position is not a treatment. It's a holding pattern while you monitor them and wait for help. You still need to check their breathing every couple of minutes.
Because they could stop breathing even in the recovery position.
The position protects the airway. It doesn't protect the respiratory drive. If their brain stops sending the signal to breathe, no position in the world is going to save them without intervention.
Alright, let's talk about the specific combinations Daniel mentioned. Alcohol and benzodiazepines. What's actually happening there that makes it so dangerous?
This is where the pharmacology gets grimly elegant. Alcohol enhances the effect of GABA, which is the brain's primary inhibitory neurotransmitter. It's what slows everything down. Benzodiazepines — Valium, Xanax, Klonopin, Ativan — they work on a different binding site on the exact same GABA receptor complex. So you're hitting the same system from two different angles simultaneously. It's not one plus one equals two. It's more like one plus one equals eleven.
Both drugs are telling the brain to slow down.
The brain complies. The specific danger is synergistic respiratory depression. Either substance alone can suppress breathing at high enough doses. Together, they suppress breathing at much lower doses than either would require alone. A 2019 study in the Journal of Clinical Medicine found that the combination was present in roughly fourteen percent of all drug-related emergency department visits involving benzodiazepines. And those are just the ones that made it to the hospital.
Fourteen percent seems high for something people treat so casually.
It's staggeringly high when you consider that most people taking benzodiazepines are doing so under medical supervision. The problem is that "medical supervision" often means a prescription and a printed warning label that nobody reads. A lot of people don't realize that having a couple of drinks after taking their prescribed Xanax is categorically different from having a couple of drinks without it.
Then there's the recreational use angle.
Which is where the really unpredictable outcomes happen. Street benzodiazepines, counterfeit pills, mixing with alcohol deliberately to intensify the effect — that's Russian roulette. You don't know the dose, you don't know what else is in the pill, and you're combining it with a substance that amplifies its effect on the exact system that keeps you alive.
What about alcohol and opioids? That seems to dominate the headlines.
Same mechanism, different receptor system. Opioids work on mu-opioid receptors in the brainstem, which directly regulate respiration. Alcohol adds GABA-mediated depression on top of that. The combination produces a respiratory depression that's deeper than either alone. But there's a key difference in presentation. With pure opioid overdose, you get the classic triad: pinpoint pupils, respiratory depression, and unconsciousness. With alcohol in the mix, the pupils might not be pinpoint because alcohol can affect pupillary response differently. That makes it harder to recognize.
The thing people are taught to look for — the tiny pupils — might not be there.
Which is exactly why relying on a single sign is dangerous. You need the whole picture. Breathing rate, responsiveness, skin signs. If someone is breathing at eight breaths per minute and you can't rouse them, I don't care what their pupils look like. They need emergency help.
What about stimulants and alcohol? That's a different kind of dangerous, I'd imagine.
Different mechanism entirely. Cocaine and alcohol combine to form cocaethylene in the liver. Cocaethylene is more cardiotoxic than cocaine alone and has a longer half-life. It increases the risk of sudden cardiac death by something like eighteen to twenty-five times compared to cocaine alone. But the first aid picture is different. Someone on stimulants and alcohol might be agitated, hyperthermic, tachycardic — heart rate through the roof. The immediate danger is cardiac, not respiratory.
If someone's been drinking and they're agitated rather than sedated, you should think about stimulant involvement.
Or a head injury. And that's another thing that gets missed constantly. Intoxicated people fall. They hit their heads. The signs of a concussion or intracranial bleed — confusion, vomiting, altered consciousness — can look exactly like intoxication. If someone is more altered than you'd expect for the amount they reportedly drank, or if they're not improving over time, you need to consider that there might be a brain bleed happening.
That's a terrifying overlap.
It's why emergency departments have a low threshold for CT scans on intoxicated patients. The clinical exam is unreliable when someone's drunk. You can't do a proper neurological assessment on someone who's slurring their speech and stumbling regardless of whether their brain is bleeding.
We've got respiratory depression from alcohol and benzodiazepines or opioids, cardiac events from stimulant combinations, aspiration risk, head injuries masquerading as intoxication. What else are people missing?
Alcohol inhibits gluconeogenesis in the liver, which is the production of new glucose. If someone has been drinking heavily and not eating, their blood sugar can crash. Severe hypoglycemia can cause unconsciousness, seizures, and brain damage. The symptoms — confusion, sweating, altered mental status — overlap almost perfectly with intoxication. A blood sugar check takes five seconds with a glucometer. Most people don't carry one.
"they're just drunk" could actually be "their blood sugar is dangerously low.
Giving them more time to "sleep it off" while their brain is starving for glucose is a recipe for permanent neurological injury. This is especially true for people with diabetes, but it can happen to anyone. I've seen cases of young, healthy people with alcohol-induced hypoglycemia who were found unresponsive and assumed to just be drunk.
What about hypothermia? If someone's passed out outside.
Alcohol causes vasodilation — blood vessels near the skin dilate, which makes you feel warm but actually accelerates heat loss. An intoxicated person outside in cool weather can become hypothermic surprisingly fast, even at temperatures that wouldn't normally be dangerous. And hypothermia depresses the central nervous system, which compounds the alcohol's effects. It's a vicious cycle. Someone who's cold and drunk deteriorates faster than someone who's just cold or just drunk.
You find someone unresponsive outside. You've checked they're breathing. What do you do about the cold?
If you can't move them — and you shouldn't move someone unnecessarily if there's any chance of spinal injury — you insulate them from the ground. The ground pulls heat away much faster than air. Put a jacket, a blanket, anything under them. But don't actively rewarm them with hot water or heating pads. That can cause dangerous cardiac arrhythmias in a hypothermic person. Passive rewarming with blankets is the move until EMS arrives.
Let me pull us back to something practical. Someone's been drinking, they're conscious but clearly very intoxicated. They're at home, maybe a friend or family member. What does monitoring actually look like?
This is where most people fail. They check once, think the person looks okay, and go to bed. Monitoring means serial assessments. You check breathing rate, responsiveness, and skin signs every ten to fifteen minutes. You write it down if you have to. The trend matters. If their breathing rate was fourteen and now it's ten, that's a trend in the wrong direction even if ten is still technically within a concerning-but-not-emergency range.
If they're asleep?
You still check. Wake them up. If they can't be roused to the point of giving you a coherent response, that's a problem. The old advice about "let them sleep it off" has killed a lot of people. A sleeping drunk person should be rousable. Not easily — they might be grumpy and disoriented — but rousable. If you can't wake them, call for help.
What about the idea of giving them coffee or a cold shower?
Neither works and both can be harmful. Caffeine doesn't speed up alcohol metabolism. It might make a drowsy person slightly more alert, but it does nothing for their respiratory depression or their blood alcohol concentration. A cold shower can cause a dangerous drop in body temperature and, if the person is really intoxicated, a shock response that can trigger cardiac issues. It's one of those folk remedies that persists despite having no basis in physiology.
The coffee thing is so entrenched. I feel like every movie and TV show for decades has shown someone being given black coffee to sober up.
It's probably the single most dangerous piece of medical misinformation in popular culture, rivaled only by the idea that you should put something in someone's mouth during a seizure. Time is the only thing that metabolizes alcohol. The liver processes roughly one standard drink per hour, and nothing you eat or drink changes that rate. Coffee, cold showers, exercise — all they do is produce a wide-awake drunk person.
Which might actually be worse, because now they're mobile and still just as impaired.
A wide-awake drunk person can get into a car, fall down stairs, start a fight. The sedation is, in a perverse way, protective.
Let's talk about when to call for help. You've given us a lot of individual signs. Can you give me a clear, memorable set of criteria?
I'll give you five. If any one of these is present, you call emergency services. One: breathing rate below eight or above thirty. Two: unresponsive to painful stimuli. Four: cyanosis — blue lips or nail beds. Five: vomiting while not fully conscious or not able to protect their own airway. Any one of those, you make the call. Don't talk yourself out of it. Don't worry about embarrassing them or getting them in trouble. Dead is worse than embarrassed.
What about the gray zone? Breathing at ten, rousable but confused, pale but not blue.
That's the monitoring zone. But I want to emphasize something: if you're unsure, call. The threshold should be low. Emergency services would rather respond to a call that turns out to be non-critical than arrive to a respiratory arrest that could have been prevented. Paramedics are not going to be angry at you for calling about someone who's severely intoxicated and might be deteriorating.
There's probably a fear of legal consequences too. Especially if there are illicit substances involved.
This is where I need to be very clear. Most jurisdictions in the United States have Good Samaritan laws that provide legal protection for people who call for help in an overdose situation. The specifics vary by state, but the general principle is that the law wants you to call. The law prefers a live person over a prosecution. And even in places where the protections aren't perfect, the alternative is watching someone die.
That's a stark way to put it.
It's a stark situation. I've worked in emergency settings. I've seen the parents who found their kid too late because they were afraid of the police. I've seen the friends who drove someone to the hospital themselves instead of calling an ambulance because they didn't want the attention, and the person stopped breathing in the back seat. The regret is worse than any legal consequence.
What about naloxone? Should people carry it?
If you're around anyone who uses opioids — prescribed or otherwise — yes. Naloxone is an opioid receptor antagonist. It reverses opioid overdose by knocking opioids off the receptors, restoring respiratory drive within minutes. It's available over the counter in most states now, often as a nasal spray. It's easy to use, it's safe, and it doesn't cause harm if you give it to someone who turns out not to have opioids in their system. The worst that happens is you wasted a dose.
It doesn't work on alcohol or benzodiazepines.
It does not. Which is another reason the combination overdoses are so dangerous. If someone has alcohol, benzodiazepines, and opioids on board, naloxone might reverse the opioid component and restore breathing temporarily, but the alcohol and benzodiazepines are still there. The person can slip back into respiratory depression when the naloxone wears off, which happens in thirty to ninety minutes. This is why anyone who receives naloxone needs to go to the hospital, even if they seem fine afterward.
Naloxone is not a cure. It's a bridge.
It's a bridge to definitive care. And it's an incredible tool. The CDC reported that from 1996 to 2020, bystander-administered naloxone reversed nearly twenty-seven thousand overdoses in the United States. That's twenty-seven thousand people who would be dead without someone carrying a small nasal spray.
That's a remarkable number.
It's almost certainly an undercount, because not all reversals get reported. Naloxone is one of the few genuine miracles in emergency medicine. It takes someone who is blue and not breathing and within two or three minutes they're sitting up and talking. The first time you see it, you don't forget it.
Let's shift to the alcohol poisoning scenario specifically. No other drugs involved. What does that look like and what's different about managing it?
Pure alcohol poisoning is dose-dependent CNS depression. Blood alcohol concentration above about 0.3 percent is where you start seeing serious risk of coma and respiratory failure. 4 percent is potentially fatal. But those numbers are for people with some tolerance. A novice drinker can be in serious trouble at much lower levels. The presentation is progressive: confusion, stupor, coma, respiratory depression, hypothermia, and in severe cases, cardiac arrest.
The first aid is the same framework — airway, breathing, circulation, monitoring.
Same framework, but with one addition. Alcohol is directly toxic to the stomach lining. Severe alcohol intoxication often involves vomiting. The risk of aspiration is extremely high. If someone is vomiting and not fully conscious, they need to be on their side with their head positioned so vomit drains out. And I want to emphasize: never try to induce vomiting. That's another dangerous folk remedy. The alcohol is already absorbed. Making them vomit just increases the aspiration risk without removing meaningful amounts of alcohol from their system.
What about activated charcoal? That's in a lot of first aid kits.
Not for alcohol. Activated charcoal doesn't bind alcohol effectively. And giving anything by mouth to someone with a depressed level of consciousness is asking for aspiration. The only people who should be giving activated charcoal are medical professionals who have assessed that the airway is protected.
You mentioned earlier that people should check for medical ID bracelets or necklaces. Why is that relevant here?
Because someone who appears intoxicated might actually be having a diabetic emergency, a seizure, or a stroke. Medical ID jewelry can tell you about diabetes, epilepsy, medication allergies, or other conditions in seconds. It's not something most people think to check, but it can completely change your understanding of what's happening. If someone is confused and stumbling and you find a medical ID that says they have diabetes, your differential diagnosis just expanded dramatically.
The overlap between medical emergencies and intoxication is a theme here.
It's the central challenge of first aid in these situations. Intoxication mimics so many other conditions, and it also masks them. A drunk person with a head injury looks like a drunk person. A drunk person with hypoglycemia looks like a drunk person. A drunk person having a stroke looks like a drunk person. The safest assumption is that you don't know what's going on, and you treat the signs and symptoms you can observe rather than making assumptions about the cause.
Which brings us back to your five criteria. Those are agnostic to cause.
Breathing rate, responsiveness, seizure, cyanosis, vomiting without airway protection. Those criteria apply whether the person drank too much, took something, has a brain bleed, or all of the above. You don't need to diagnose the cause to recognize the emergency.
Let's talk about the recovery position in more detail. I feel like a lot of people have heard of it but might not actually know how to do it correctly.
The technique matters. Start with the person on their back. Kneel beside them. Take the arm closest to you and position it at a right angle to their body, palm up. Take the other arm and bring it across their chest, holding the back of their hand against their cheek on your side. With your other hand, grab the far knee and pull it up until the foot is flat on the ground. Then pull on that knee to roll them toward you. Once they're on their side, adjust the top leg so the hip and knee are bent at right angles. Tilt the head back slightly to open the airway. Check that the hand under the cheek is supporting the head and keeping the airway open. That's the standard technique. It's stable, it's effective, and it doesn't require any equipment.
You're checking breathing every few minutes after that.
Every two to three minutes. Put your hand near their mouth to feel for breath. Watch their chest. If they stop breathing, you start CPR and you call for help if you haven't already.
Is there anything else people should be doing while they wait for emergency services?
Gather information if you can, but don't delay care to do it. If there are empty bottles, pill containers, or other substances nearby, collect them to give to paramedics. Don't put yourself at risk handling unknown substances — some things can absorb through skin. But if there's a prescription bottle or a recognizable container, that information can save critical time in the emergency department. Knowing what someone took means the difference between immediate targeted treatment and a diagnostic workup that takes hours.
What about the person who refuses help? They're conscious, clearly intoxicated, but they're saying they're fine and they don't want an ambulance.
This is a really tough situation legally and ethically. An adult with decision-making capacity has the right to refuse treatment, even if that decision seems foolish. But intoxication impairs decision-making capacity. The question is whether it impairs it enough that they can't make an informed refusal. In general, if someone is oriented — they know who they are, where they are, and roughly what time it is — and they can articulate why they're refusing, you have to respect that. But if they're confused, disoriented, or unable to have a coherent conversation, they likely lack capacity, and you should call for help regardless of what they're saying.
That's a heavy call to put on a bystander.
And it's one of those situations where you have to trust your judgment. If your gut is telling you something is seriously wrong, listen to it. People who work in emergency medicine will tell you that the "something's not right" instinct is a legitimate clinical sign. It's not mystical. It's your brain pattern-matching against thousands of subtle cues that you're not consciously processing.
The uncanny valley of human behavior.
You see someone and your brain goes, "The signals don't add up." Don't ignore that.
Before we wrap up, I want to ask about something that's been in the background of this whole conversation. The alcohol and benzodiazepine combination specifically — how does that present differently from just being very drunk?
The key difference is the depth and quality of sedation. Someone who's very drunk is usually still somewhat responsive to stimulation. They might be sleeping deeply, but a sternal rub or a loud voice will get a reaction — maybe just a groan or a swat, but something. With benzodiazepine involvement, the sedation tends to be deeper and flatter. They're harder to rouse, and when you do rouse them, they slide back into unconsciousness faster. The respiratory depression is often more pronounced relative to the apparent level of intoxication. You might see someone who doesn't seem that drunk — they're not reeking of alcohol, they're not vomiting — but their breathing is dangerously slow. That's a red flag for polypharmacy.
Polypharmacy being multiple drugs.
Multiple central nervous system depressants specifically. And the thing about the alcohol-benzodiazepine combination is that it's incredibly common. Benzodiazepines are among the most prescribed medications in the United States. Something like thirty million people take them annually. A significant fraction of those people drink alcohol. Most of them are fine. But the margin of safety narrows dramatically as doses increase, and people don't always recognize when they're entering the danger zone.
The advice isn't "never mix them" — it's "understand that mixing them changes the risk profile dramatically.
Having one beer while on a low dose of a prescribed benzodiazepine is not the same as drinking heavily with a high dose or with short-acting benzodiazepines that hit fast. But the problem is that alcohol impairs judgment, so the person who planned to have one beer ends up having four, and the person who took their prescribed dose took a little extra because they were anxious, and suddenly you're in a situation nobody planned for.
The planning is being done by the part of the brain that's being impaired.
That's the cruel irony. The systems you need to make good decisions about intoxication are the same systems that intoxication disables. That's why bystanders matter so much. The person in trouble is often the least equipped to recognize they're in trouble.
To pull it all together for someone who wants a mental checklist — find someone intoxicated, what do you do?
Scene safety first. Check responsiveness — shout, then sternal rub. If unresponsive, call nine-one-one immediately. If responsive, check breathing rate for a full thirty seconds. Below eight, call nine-one-one. Check skin — cold, clammy, blue, call nine-one-one. Check airway — if they're vomiting or can't protect their airway, recovery position and call nine-one-one. If none of those red flags are present, monitor. Check every ten to fifteen minutes. If anything changes for the worse, escalate. Don't give coffee, don't give cold showers, don't try to induce vomiting. Don't leave them alone. And if you're unsure, call. Always err on the side of calling.
That's memorable. And it doesn't require medical training.
It requires being willing to pay attention and take action. That's it. Most bad outcomes in these situations happen not because help wasn't available, but because nobody recognized the need for it until it was too late.
The bystander effect, but with a medical twist.
Everyone assumes someone else will handle it, or that it's not that serious, or that they don't want to overreact. But in a CNS depression emergency, time is brain cells. Every minute of inadequate oxygenation is doing damage that might not be reversible.
I think we've covered the ground Daniel was asking about. Let's bring it home.
Now: Hilbert's daily fun fact.
Hilbert: The earliest known postal artifact from Hokkaido is a set of clay tablets from the third century used by the Ainu to communicate with the Ezo wolf population, requesting that wolves stop intercepting messengers traveling between settlements. The tablets were found in 1978 near what is now Asahikawa.
...right.
That's certainly a fact.
The Ezo wolf has been extinct for over a century, so I guess the postal request eventually worked.
Or the wolves just stopped reading their mail. Either way, that's our show. If you take one thing from this episode, let it be this: when in doubt, make the call. The worst-case scenario of calling for help is temporary embarrassment. The worst-case scenario of not calling is permanent.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop, and thanks to all of you for listening. If you found this useful, share it with someone who might need it. We'll be back next time.
Take care of each other out there.
