How to Know If Someone Put Something in Your Drink

Drink spiking is difficult to detect in the moment, but there are real warning signs: unexpected changes in your drink’s taste or appearance, and a sudden onset of symptoms like confusion, dizziness, or extreme drowsiness that feel disproportionate to how much you’ve consumed. The challenge is that the most commonly used substances are chosen precisely because they’re hard to notice, and the window for confirming what happened through testing is alarmingly short. What follows is what the research actually tells us about spotting a spiked drink, what gets used, how reliable the available detection tools are, and what steps matter most if you suspect something has gone wrong.

Changes in Taste and Appearance

One of the few lab studies to directly test whether people could detect drugs dissolved in drinks found that all nine sedating drugs tested altered the appearance or taste of the beverages they were added to, at least to some degree.¹ Two drugs in particular, zopiclone (sold as Imovane) and carisoprodol (Somadril), produced a taste that was noticeable in every beverage tested. The others were subtler, but none were completely undetectable. That sounds reassuring, but there’s a catch: in a loud bar with a flavored cocktail or sugary mixer, subtle bitterness or a slight cloudiness is easy to miss. Strongly flavored or dark-colored drinks mask both taste and visual changes far more effectively than water or light-colored beverages.

If your drink tastes saltier, more bitter, or chemically “off” compared to what you ordered, that’s worth paying attention to. A change in color, unexpected fizzing, or a layer of residue at the bottom of the glass are also red flags. But the honest reality is that relying on your senses alone is not a dependable safety strategy. Many spiking agents dissolve clearly and have minimal flavor at the doses used.

Physical Symptoms That Don’t Match What You Drank

The most reliable real-time indicator isn’t in the glass; it’s in your body. If you feel dramatically more intoxicated than your alcohol intake can explain, something may be wrong. Benzodiazepines, one of the drug classes associated with drink spiking, can cause confusion, impaired thinking, memory loss, drowsiness, poor coordination, and dizziness.² GHB produces a similar constellation: sudden sleepiness, nausea, loss of muscle control, and a feeling of being “out of it” that comes on fast.

Key symptoms to watch for, in yourself or in someone you’re with:

Sudden drowsiness: Feeling like you could fall asleep standing up after just one or two drinks.
Confusion or disorientation: Losing track of where you are, who you’re with, or what you were doing moments earlier.
Loss of coordination: Stumbling, slurring, or having trouble holding objects, far beyond what your alcohol consumption would predict.
Memory gaps: Realizing you can’t recall the last 20 minutes or more.
Nausea or vomiting: Particularly when it comes on abruptly and early in the night.
Visual disturbances: Blurred or tunnel vision not typical for your level of drinking.

The critical distinction is the mismatch. One beer shouldn’t leave you unable to walk. Two glasses of wine shouldn’t cause a total blackout. If the effect dramatically exceeds the cause, treat it as suspicious. Tell a trusted friend immediately, move to a safe location, and consider going to an emergency department. Time matters enormously for detection, as we’ll get to shortly.

What Actually Gets Put in Drinks

The popular image of drink spiking involves exotic “date rape drugs,” but the research paints a more complicated picture. In a study of alleged spiking cases presenting to a London emergency department, alcohol was by far the most prominent substance: about 90% of patients tested positive for ethanol, and the majority had blood alcohol levels associated with significant intoxication.³ Illicit drugs were found in about 15% of cases, and drugs that couldn’t be explained by the patient’s own use were found in roughly 10%. Unexplained sedative exposure specifically was confirmed in only about 3% of cases.

An Australian study of 97 suspected spiking cases found a similar pattern: researchers identified only nine plausible spiking cases and did not find a single confirmed instance of a sedative drug being secretly placed in a drink at a pub or nightclub.⁴ Illicit drugs were present in about 28% of the group, and the average number of standard drinks consumed was nearly eight. In another emergency department study, drug testing on samples from 42 patients found drugs of misuse in about a fifth of cases.⁵

This doesn’t mean spiking never happens. It clearly does. But it does mean that extra alcohol added to drinks (doubling shots without someone’s knowledge, for example) is a far more common mechanism than a covert sedative, and that voluntary alcohol and drug consumption sometimes produce symptoms people later attribute to spiking. One survey-based study found that almost a quarter of respondents reported having experienced drink spiking, but toxicological and police data suggest the perceived likelihood of spiking is considerably overestimated.⁶ None of this should discourage someone who suspects spiking from seeking help. The point is that spiking with extra alcohol is still spiking, and the appropriate response is the same regardless of the substance involved.

Why GHB Is Especially Hard to Catch

Among the drugs that do get used in drink spiking, GHB stands out for how quickly it vanishes from the body. A clinical trial measuring GHB elimination found a median half-life of just 1.2 hours. At three hours after ingestion, 80% of participants still had detectable blood levels above the standard cutoff, but by four hours that dropped to a third, and by six hours GHB was undetectable in every single participant.⁷ That six-hour window is practically invisible when you factor in the time it takes for someone to realize something is wrong, get to a hospital, and have blood drawn.

The detection challenge gets worse: GHB is naturally present in the human body at low levels. Forensic toxicologists have to distinguish between what’s supposed to be there and what was added from outside, and those concentration ranges overlap in routine testing.⁸ Advanced techniques exist that can tell the difference by analyzing the carbon isotopic signature of the GHB molecule, since synthetic GHB and the body’s own GHB come from different chemical origins.⁹ But this technique is specialized and not available in most emergency departments. The practical implication for you is that if you suspect GHB exposure, getting to a hospital and requesting blood and urine collection as quickly as possible is genuinely urgent. Every hour that passes makes detection less likely.

Do Drink-Testing Products Actually Work?

You may have seen test strips, coasters, or nail polishes marketed as tools to detect drugs in your drink. The research on these is not encouraging. A lab evaluation of drug-detection coasters designed to screen for GHB and ketamine found that while the coasters correctly identified drug-free drinks 100% of the time, their ability to detect spiked drinks was dismal: 0% sensitivity for ketamine and somewhere between 31% and 69% for GHB, depending on how you interpret ambiguous results.¹⁰ The researchers concluded the coasters were inadequate as screening tools.

The picture for test strips is somewhat more nuanced. Older, first-generation GHB test strips were largely insensitive and produced colors that were hard to read in dark or colored beverages like beer, whiskey, and red wine. Newer second-generation strips performed substantially better in lab conditions, detecting GHB at concentrations low enough to be clinically relevant.¹¹ This is an improvement, but worth putting in context: these were lab evaluations, not field tests in noisy, dimly lit venues. And they only detect GHB, not the broader range of substances someone might use.

Experimental technologies continue to be developed, including cellulose-based smart straws that change color in response to pH shifts, which can signal certain types of adulteration.¹² These are still at the research stage rather than consumer-ready, and pH-based detection can’t distinguish a spiking agent from a naturally acidic or basic ingredient in a cocktail. The bottom line on personal detection products is that the technology is improving but is nowhere near reliable enough to serve as your primary defense. A negative test result from any of these products should not make you feel safe; a positive result, however, should be taken seriously.

What to Do If You Suspect Your Drink Was Spiked

The steps that matter most are the steps you take fast. If you feel suddenly and unexpectedly intoxicated, or you notice your drink looks or tastes wrong:

Stop drinking it: Put it down, and don’t let anyone pressure you into finishing it. If possible, keep the drink or a sample of it. Research on benzodiazepines dissolved in beverages has shown that the drugs remain detectable in the liquid over time, giving forensic labs something to work with if the drink is preserved.¹³
Tell someone you trust: A friend, a bartender, a bouncer. Do not leave the venue alone with someone you don’t know well.
Get to a safe location: If symptoms are progressing, call emergency services or go to the nearest emergency department.
Request testing early: At the hospital, ask for both blood and urine samples to be collected and preserved for toxicology. The detection window for some substances is only a few hours.
Report it: File a police report even if you’re uncertain. Early reports help establish a timeline, and samples collected at that point can be tested later.

Preserving the drink itself is a point researchers have specifically recommended. Because drugs like GHB vanish from the body so quickly, the drink may be the only piece of physical evidence available. Store it sealed and refrigerated if you can, but getting it to police or hospital staff is more important than handling it perfectly.

Looking Out for Others

Drink spiking is often something bystanders are better positioned to spot than the person targeted. Research on barroom aggression found that perpetrators tend to seek out people who appear rapidly intoxicated, and that heavily intoxicated individuals are less able to detect threatening cues, resist, or call for help.¹⁴ In one observational study cited in that research, fewer than 1% of sexually aggressive incidents in bars ended with the aggressor being removed from the venue. The implication is clear: formal safeguards at venues are often absent, and the people most at risk are the least equipped to protect themselves in the moment.

If someone in your group or even a stranger at a bar looks dramatically more impaired than the situation warrants, checking on them is one of the most effective interventions available. Ask if they’re okay, offer to help them contact someone they trust, and don’t leave them alone with someone who seems to be steering them somewhere. Bar staff, when trained in bystander intervention, can also play a critical role, but many venues don’t invest in that training.

The Psychological Aftermath

Whether or not toxicology ultimately confirms the presence of a drug, the experience of believing your drink was spiked carries real psychological weight. A content analysis of people describing drink-spiking experiences on Reddit found that many reported long-term mental health effects and felt that emergency services responded inadequately when they sought help.¹⁵ The uncertainty itself is part of the harm: not knowing what happened to you, not being able to prove it, and facing skepticism from institutions that should be helping.

In the Australian study mentioned earlier, 35% of patients still believed they had been spiked even after their toxicology results came back without evidence of a covert drug.⁴ This is sometimes cited as evidence that people are wrong about their experiences, but it’s more complicated than that. Negative toxicology doesn’t mean nothing happened. Testing may have been too late. The substance may have been additional alcohol rather than a detectable drug. The detection thresholds used by the lab may have missed a compound at low concentration. And the subjective experience of suddenly losing control in a social setting is frightening regardless of what a blood test says afterward.

If you’ve gone through this, the lack of a confirmed positive test doesn’t invalidate what you felt. Counseling or support from organizations that specialize in sexual violence and substance-facilitated assault can be helpful even without a forensic finding to point to.

Why the Evidence Is Thinner Than You’d Expect

One thing that becomes apparent when you look at the research on drink spiking is how little of it there is relative to the scope of the problem people describe. The studies tend to be small, conducted at single emergency departments, and limited to patients who self-present saying they think they were spiked. People who were spiked but didn’t recognize it, didn’t seek help, or attributed their symptoms to having had too much to drink never enter the data at all. Meanwhile, the forensic challenges with rapidly eliminated drugs like GHB mean that even genuine cases often produce negative lab results.

The gap between self-reported experience and confirmed toxicological findings is wide, and both sides of the debate oversimplify it. Dismissing all spiking claims because forensic confirmation rates are low ignores the massive detection problem. But treating every suspicion as confirmed spiking without considering voluntary overconsumption doesn’t serve people well either, because the most common cause of unexpectedly severe intoxication remains alcohol itself, sometimes consumed voluntarily in amounts people underestimate and sometimes added without consent. The practical takeaway isn’t to doubt yourself, but to understand that the fastest possible medical attention and sample collection gives you the best chance of an answer, and that looking after your own drink and staying aware of how it tastes and how you feel remain the most reliable real-time defenses available.