Why Medications Affect People Differently: The Real Science Behind Drug Side Effects

Why Medications Affect People Differently: The Real Science Behind Drug Side Effects
By Frankie Torok 2 January 2026 14 Comments

Ever taken the same pill as someone else and had a totally different experience? One person feels fine, another ends up in the ER. It’s not luck. It’s biology. Medications don’t work the same for everyone - and the reasons why are deeper than most doctors even talk about.

It’s Not Just About the Drug

Most people think side effects are just bad luck. You take a pill, you get dizzy, nauseous, or worse. But that’s not random. It’s your body’s unique reaction to the medicine. Your genes, your age, what else you’re taking, even your gut bacteria - all of it plays a role. The World Health Organization calls these unwanted reactions adverse drug reactions (ADRs). And they’re not rare. In Europe, 3.6% of hospital admissions are caused by them. In the UK, ADRs cost the NHS £770 million every year. That’s not just a statistic - it’s people getting sick from medicines meant to help them.

Your Genes Are the Main Player

Your DNA decides how fast or slow your body breaks down drugs. The key players are enzymes called cytochrome P450 - especially CYP2D6, CYP2C9, and CYP2C19. These are like little factories in your liver. Some people have versions of these enzymes that work super fast. Others have broken versions that barely work at all.

For example, about 5-10% of white Europeans are poor metabolizers of CYP2D6. That means drugs like codeine, antidepressants, or beta-blockers stick around too long. They build up. Side effects spike. On the flip side, 1-2% of Europeans - and up to 29% of Ethiopians - are ultra-rapid metabolizers. Their bodies clear drugs too fast. The medicine doesn’t work. They think it’s ineffective, so they take more. And then - boom - overdose.

This isn’t theory. It’s documented. A 68-year-old woman in a JAMA case study kept having dangerous bleeding from warfarin, even though her dose hadn’t changed. Turns out, she had two broken copies of CYP2C9. Her body couldn’t process the drug. Once her doctors knew, they cut her dose by 60%. Her INR stabilized. She went home.

Age Changes Everything

As you get older, your body changes - and so do your reactions to drugs. Older adults have more body fat and less muscle. Fat-soluble drugs like benzodiazepines or antidepressants get stored in fat tissue. They linger. That’s why seniors are 3-5 times more likely to have falls from sedatives. Their kidneys and liver don’t filter drugs as well. Even if the dose is “normal,” it’s too much for them.

And it’s not just age. If you’re sick with an infection or inflammation, your liver enzymes slow down. That means drugs you’ve taken for years suddenly become stronger. A simple cold can turn a safe dose into a dangerous one.

Elderly woman surrounded by shattering genetic code as warfarin dosage adjusts digitally.

Drug Interactions Are Silent Killers

Taking two pills together can be like mixing chemicals. Amiodarone - a heart rhythm drug - blocks the enzyme that breaks down warfarin. That can double or even triple warfarin levels in your blood. Result? Uncontrolled bleeding. NSAIDs like ibuprofen are fine for most people. But if you have a certain gene variant and take steroids at the same time? Your risk of a stomach bleed jumps from 1-2% to 15-30%.

Polypharmacy - taking five or more medications - is the biggest risk factor for ADRs in older adults. One study found these patients have ADR rates 300% higher than younger people. It’s not the drugs alone. It’s the combinations. And most doctors don’t check for them.

Pharmacogenomics: The Future Is Here (But Not Everywhere)

There’s a solution: pharmacogenomics. That’s just a fancy word for using your genes to pick the right drug and dose. The FDA has included genetic info on over 300 drug labels. For 44 of them, they give specific dosing advice based on genetics.

Take clopidogrel. It’s a common blood thinner after a heart attack. But if you’re a poor metabolizer of CYP2C19 - 2-15% of people - the drug doesn’t work. You’re still at risk of a clot. Testing for this variant could save lives. Yet, it’s still not routine in most hospitals.

Warfarin is another win. Genetic testing for CYP2C9 and VKORC1 explains 30-50% of why people need different doses. In trials, patients who got genotype-guided dosing reached safe levels faster and had 31% fewer major bleeds.

In pediatric cancer, St. Jude’s Hospital cut severe side effects from mercaptopurine by more than half using genetic testing. That’s not a small win. That’s life-changing.

Holographic gene map alerts pharmacist about ineffective clopidogrel due to genetic variant.

Why Isn’t Everyone Getting Tested?

Because it’s still not easy. Only 18% of U.S. insurers cover pharmacogenomic tests. Most doctors haven’t been trained to read the reports. A 2023 survey found 68% of physicians feel unprepared to use genetic data. Hospitals lack the systems to automatically flag risky gene-drug combinations in electronic records.

The cost used to be $2,000 per test. Now it’s down to $250. That’s cheaper than a month’s supply of many prescriptions. But the biggest barrier? Doctors don’t ask. They don’t know what to ask. And patients don’t know to ask for it.

What Can You Do?

You don’t need a genetic test to protect yourself. Start here:

  • Know your meds. Keep a list - including over-the-counter drugs and supplements.
  • Ask your pharmacist: “Could this interact with anything else I’m taking?”
  • If you’ve had a bad reaction to a drug before, tell your doctor. Don’t just say “it made me sick.” Say: “I got dizzy after taking X. I stopped it.”
  • If you’re on long-term meds like warfarin, clopidogrel, or antidepressants, ask: “Is there a genetic test that could help me get the right dose?”
  • If you’re over 65, ask if your dose is too high for your age.

The Bigger Picture

The future isn’t just about one gene. It’s about hundreds. New tools called polygenic risk scores are starting to combine dozens - even hundreds - of genetic signals to predict how you’ll respond to a drug. Early results show they’re 40-60% more accurate than single-gene tests.

The U.S. Medicare system just started covering pharmacogenomic testing for 17 high-risk drugs in January 2024. The EU now requires genetic data in all new clinical trials. Academic hospitals are building programs. The market is growing fast - from $5 billion in 2022 to an expected $24 billion by 2029.

But technology alone won’t fix this. We need doctors who understand genetics. We need systems that warn before a bad mix happens. We need patients who speak up.

Medications aren’t one-size-fits-all. They never were. The sooner we accept that, the fewer people will get hurt by the very things meant to heal them.

Why do some people get side effects from a drug while others don’t?

It’s mostly due to genetics. Your genes control how your body absorbs, breaks down, and responds to drugs. For example, variations in CYP2D6 or CYP2C19 enzymes can make you a slow or fast metabolizer, causing drugs to build up or wear off too quickly. Age, other medications, liver or kidney function, and even infections can also change how your body handles medicine.

Is pharmacogenomic testing worth it?

For certain drugs, yes. If you’re taking warfarin, clopidogrel, certain antidepressants, or chemotherapy drugs like mercaptopurine, genetic testing can prevent serious side effects or help find the right dose faster. Studies show it reduces hospitalizations and bleeding events. For most people on simple medications, it’s not yet essential - but it’s becoming more accessible and affordable, with tests now under $250.

Can I get genetic testing for drug reactions through the NHS?

Currently, routine pharmacogenomic testing isn’t standard in the NHS. It’s mostly used in specialized settings like cancer care or cardiology. But that’s changing. With Medicare in the U.S. covering testing for 17 drugs in 2024, pressure is growing for similar coverage in the UK. If you’re on a high-risk medication and have had a bad reaction, ask your doctor about referral to a pharmacogenomics clinic.

What drugs are most likely to cause genetic-related side effects?

Warfarin (blood thinner), clopidogrel (antiplatelet), certain antidepressants like SSRIs and TCAs, statins (cholesterol drugs), codeine and other opioids, and chemotherapy agents like 6-mercaptopurine. The FDA has labeled over 300 drugs with pharmacogenomic information. If you’re on long-term treatment for heart disease, depression, or cancer, ask if your drug is on the list.

Does having a family history of bad drug reactions mean I’ll have them too?

It’s a strong clue. If your parent had a severe reaction to a drug like codeine or warfarin, you may carry the same gene variant. It doesn’t guarantee you’ll react the same way - other factors like age and other meds matter - but it’s a red flag. Tell your doctor. It could save you from a dangerous reaction.

Are over-the-counter drugs safe if they’re not prescription?

Not necessarily. NSAIDs like ibuprofen and naproxen cause stomach bleeds in 1-2% of users each year - but that risk jumps dramatically if you have certain genes or take steroids. Even common supplements like St. John’s Wort can interfere with antidepressants or birth control. Just because it’s sold without a prescription doesn’t mean it’s risk-free.

14 Comments
Haley Parizo January 2 2026

So let me get this straight-we’re still letting doctors guess what drug does what to your body? That’s not medicine, that’s Russian roulette with a prescription pad. I’ve seen people die because their ‘normal’ dose was a lethal cocktail for their genes. We’re not talking about rare edge cases here-we’re talking systemic negligence. If your liver can’t process codeine, why the hell are you even getting it? This isn’t science. It’s corporate laziness dressed up as healthcare.

Ian Detrick January 3 2026

Man, I wish I’d known this before I got hospitalized from a ‘safe’ dose of an SSRI. My doc said ‘it works for most people’-turns out I’m one of those weird CYP2D6 ultra-rapid metabolizers. Took me three years to find someone who even knew what that meant. Now I get tested before anything new. It’s not expensive. It’s not complicated. It’s just not taught. And that’s the real tragedy.

Sarah Little January 4 2026

Pharmacogenomic profiling is not a luxury-it’s a biomolecular imperative. The CYP450 polymorphisms, particularly in CYP2C19 and CYP2D6, are clinically actionable variants with Level A evidence per the CPIC guidelines. The failure to integrate pharmacogenomics into EHRs represents a catastrophic failure of translational medicine. We are operating in a pre-genomic era while the data is already here. This isn’t innovation-it’s institutional malpractice.

innocent massawe January 6 2026

Wow. This is real. In Nigeria, we just take what the pharmacy gives. No testing. No questions. My uncle took warfarin and bled for days. They said ‘maybe bad luck.’ Now I know it was his genes. This info should be everywhere-on posters, on radios, on phones. Not just for rich countries.

🙏

Philip Leth January 7 2026

Bro, I got prescribed gabapentin for anxiety. Felt like I was underwater. Told my doc, he said ‘give it time.’ Two weeks later I was hallucinating. Turns out I’m a slow CYP2D6 metabolizer. Now I take CBD instead. No test needed. Just listen to your body. The system’s broken. Don’t trust the script. Trust your vibe.

Angela Goree January 7 2026

STOP. JUST STOP. We’re not talking about ‘maybe’ or ‘some people.’ We’re talking about PEOPLE DYING because doctors are too lazy to read a damn genetic report. And now you want to make it optional? No. No. NO. This isn’t ‘personal preference’-it’s a public health emergency. The FDA has labeled over 300 drugs. The EU mandates it. But here? We’re still playing ‘guess the gene’ like it’s 1987. Shame. Shame. Shame.

Joy F January 8 2026

Let’s be real: this isn’t about science-it’s about capitalism. Pharma doesn’t want you to know your genes dictate your reaction because then they’d have to personalize everything. And personalization? That’s not scalable. That’s not profitable. They’d rather you get sick, get hospitalized, get another prescription, get another bill. This is a $770 million problem… that’s just the tip of the iceberg. And they’re not fixing it. They’re monetizing it.

They’ll sell you a $250 test… then charge you $2,000 to interpret it. And they’ll make you wait six months. Meanwhile, your liver is failing. This isn’t medicine. It’s a performance art of exploitation.

Palesa Makuru January 9 2026

It’s amusing how Americans treat genetics like some mystical oracle. You think your CYP2D6 variant makes you special? In South Africa, we don’t have access to tests, so we rely on clinical observation-something your over-medicated culture has forgotten. Your ‘personalized medicine’ is just a privilege for the insured. Meanwhile, my cousin died on ibuprofen because her doctor didn’t know she was homozygous for HLA-B*58:01. No test. No second chance. Just another statistic. You’re not saving lives-you’re commodifying them.

Hank Pannell January 9 2026

There’s something beautiful here-our biology is uniquely coded, and we’re just beginning to decode it. I’ve been on antidepressants for a decade. Three different ones. All failed. Then I got tested. Turns out I’m a CYP2C19 poor metabolizer. Switched to sertraline. No more brain fog. No more nausea. Just… life. It’s not magic. It’s biology. And if we can make this routine, we’re not just preventing side effects-we’re restoring dignity to the patient experience. Why isn’t this standard? Because we still treat people like data points, not individuals.

Lori Jackson January 11 2026

Oh, so now we’re blaming genetics because doctors can’t be bothered to do their jobs? Let me guess-you also think your gluten intolerance is ‘in your DNA’ and not just because you eat 12 bagels a day. This is the new pseudoscience: blaming your genes for bad lifestyle choices. Warfarin? Maybe you’re not taking it right. Antidepressants? Maybe you’re not sleeping. Stop looking for genetic scapegoats. Take responsibility. Your body isn’t a lab experiment-it’s your responsibility.

Wren Hamley January 12 2026

Imagine your body as a jazz band. Some folks got a fast drummer (ultra-rapid metabolizers), others got a sax that’s out of tune (poor metabolizers). The song’s the same-warfarin, clopidogrel, you name it-but the rhythm? Totally different. And if the bandleader (your doctor) doesn’t listen to the instruments, the whole thing crashes. We need conductors who know the score. Not just guys with a clipboard and a 10-minute consult.

erica yabut January 13 2026

Let’s be honest: most people don’t even know what ‘CYP2D6’ means. And yet you want them to demand genetic testing? This isn’t empowerment-it’s elitist nonsense. The average person can’t even spell ‘pharmacogenomics.’ You’re not helping. You’re alienating. The real solution? Better doctor training. Not more tests. More empathy. Less jargon. Less ‘you’re special because of your SNPs.’ Just… care.

Tru Vista January 14 2026

LOL. You think this is new? My grandma took aspirin in 1952 and got a bleed. They didn’t have DNA tests. She still lived to 98. Maybe you’re just overmedicated? Stop looking for complex answers when the solution is: take less drugs. Duh.

Vincent Sunio January 15 2026

While the article presents a compelling narrative, it fundamentally misrepresents the evidentiary threshold for clinical implementation of pharmacogenomic testing. The majority of the cited studies are observational, with small sample sizes and selection bias. The FDA’s labeling of 300 drugs includes many with only Level C evidence. To suggest that routine testing is standard of care is not merely premature-it is a misapplication of medical ethics. Until randomized controlled trials demonstrate reduced mortality across diverse populations, this remains speculative. One must not confuse accessibility with efficacy.

Say something