What are the odds of a serious adverse drug reaction from prescribed medication?
Evidence quality 4.5/5
Eight-dimension review score against the quality rubric . Each dimension scored 1–5.
- D1 Source grounding
- 5/5
- D2 Source authority
- 5/5
- D3 Arithmetic
- 4/5
- D4 Uncertainty
- 4/5
- D5 Scope
- 5/5
- D6 Prose
- 4/5
- D7 Perception honesty
- 4/5
- D8 Caveat completeness
- 5/5
Lifetime probability · lifetime, US adult
1 in 58
1.7% lifetime chance
Most people underestimate this.
range 1 in 87 to 1 in 29
● your factors — click this risk ▾ to reveal
≈ As likely as
Perceived
There is no standing survey that isolates "fear of a serious adverse drug reaction from a prescribed medication." Prescription drug risk sits in a cultural blind spot: patients skim the side-effect leaflet, notice that most entries end in "rare," and file the category under "things that happen to other people." The same readers who will happily interrogate a vaccine's one-in-a-million myocarditis signal tend to accept an SSRI, an NSAID, or a beta-lactam antibiotic with almost no probabilistic framing at all. The gap between that intuition and the aggregate hospitalisation and mortality numbers is one of the larger mismatches on this site.
Rough estimate: most US adults would guess well under 1 in 1,000 for a lifetime fatal ADR
Source: editorial intuition, not polled
Actual
~50,000-150,000 fatal adverse drug reactions per year, US (midpoint ~75,000)
US adults, all ages pooled
Show derivation
Two different headline numbers live in this entry and it is worth separating them. (1) Lifetime probability of at least one *hospitalisation-level* serious ADR: CDC reports roughly 500,000 ADE-related hospitalisations per year in the US. Against ~258 million US adults that is a per-adult-year hazard of about 1.94 per 1,000. Compounded over 59 years of remaining adult life: 1 − (1 − 0.00194)^59 ≈ 0.108, i.e. roughly 1 in 9 adults experience a serious ADR requiring hospitalisation at some point. (2) Lifetime probability of a *fatal* ADR: Lazarou et al. (JAMA 1998) estimated ~106,000 fatal ADRs in US hospitals in 1994; more recent analyses have argued that figure is on the high end of the plausible envelope, with defensible modern point estimates in the 50,000-150,000/year range. Using a midpoint of ~75,000 fatal ADRs per year against 258 million US adults gives a per-adult-year hazard of ~2.91 × 10^-4 and a compounded lifetime figure of 1 − (1 − 0.000291)^59 ≈ 0.0171, or roughly 1 in 58. The headline normalized value reports the fatal-ADR lifetime number so it can be compared directly with other mortality entries on the site; the hospitalisation-level figure is discussed in the body text.
Caveats: The population-level number here papers over heterogeneity that matters. Age is …
The population-level number here papers over heterogeneity that matters. Age is the dominant axis: adults 65+ account for more than 600,000 of the ~1.5M annual ADE ED visits, so a healthy 30-year-old on a single medication faces a risk well below the headline and an 80-year-old on eight medications faces a risk several multiples above it. Drug class matters almost as much — CDC identifies anticoagulants (~21% of ADE ED visits), diabetes agents including insulin (~14%), and antibiotics (~13%) as the three biggest categories, with opioids and antiplatelets close behind. The Lazarou 106,000 fatal-ADR/year figure has been contested downward in the literature since publication, which is why Likelier's uncertainty band spans roughly 50,000-150,000 US fatal ADRs per year rather than pinning to the 1994 point estimate. Finally this entry covers only adverse reactions to medications taken as prescribed: overdose (accidental or intentional) is tracked under <a href="/fears/drug-overdose">drug-overdose</a>, and drug-induced anaphylaxis — roughly 59% of all fatal anaphylaxis per Jerschow et al. — is the iatrogenic slice of <a href="/fears/anaphylaxis-fatal">anaphylaxis-fatal</a>.
Risks at similar odds
Other risks with roughly the same likelihood — useful for calibration.
Medication misuse
What are the odds of being harmed by not reading drug labels or mixing medications?
Benzo dependence
What are the odds of developing benzodiazepine dependence after a standard prescription?
Bipolar disorder
What are the odds of developing bipolar disorder at some point in your lifetime?
Recently viewed on this device
Stored locally — clear anytime.
Pick challenger
The Lazarou meta-analysis is the reference point everyone starts from. Pooling 39 prospective studies of US hospitalised patients, Lazarou, Pomeranz, and Corey reported a 6.7% incidence of serious adverse drug reactions and a 0.32% incidence of fatal ADRs per admission, extrapolating to roughly 2.2 million serious and 106,000 fatal ADRs among US inpatients in 1994 — a number that, if accurate, placed adverse drug reactions between the fourth and sixth leading cause of death in the country. Pirmohamed’s independent UK analysis of 18,820 prospective admissions six years later found almost the same admission-level rate (6.5% of admissions ADR-related) with a more conservative 0.15% case-fatality, and CDC’s ongoing surveillance puts current US ADE volume at roughly 1.5 million emergency department visits and 500,000 hospitalisations per year. Put together, the lifetime probability of a hospitalisation- level ADR for a typical US adult is something like 1 in 9; the lifetime probability of a fatal one lands around 1 in 58, with an uncertainty band spanning roughly 1 in 90 to 1 in 30 depending on which modern re-estimate of the Lazarou figure you believe.
The comparative frame is the part that is genuinely underrated. Prescription medications taken as directed cause a per-exposure rate of serious adverse events that runs roughly 100 to 1,000 times higher than the per-dose rate for modern vaccines, where serious adverse events sit in the one-in-100,000-to-one-in-a-million band. The cultural attention runs in exactly the opposite direction: a myocarditis signal at 1 in 10,000 triggers years of front-page debate, while a 6.5% admission-level ADR rate from ordinary prescribing has been common knowledge in clinical pharmacology for thirty years without meaningfully reshaping public intuition. Part of that is framing — an ADR attributed to “the patient’s underlying condition” is invisible in a way a vaccine reaction is not — and part is that prescription drug risk is distributed across thousands of individually-small events rather than clustered into newsworthy episodes.
The headline does not apply evenly. The fatal-ADR burden concentrates sharply in older adults on multiple medications: CDC’s own surveillance shows that adults 65+ account for more than 600,000 of the roughly 1.5 million annual ADE ED visits, and the three drug classes responsible for the majority of those visits — anticoagulants (~21%), diabetes agents including insulin (~14%), and antibiotics (~13%) — are also the classes whose risk profile scales steeply with age, renal function, and polypharmacy. A young-otherwise-healthy adult on a single well-tolerated agent faces a risk a small fraction of the headline number; a frail 80-year-old on eight concurrent medications faces a risk several multiples of it. The population-average 1-in-58 figure is the right answer to “how large is the aggregate burden of prescription medication mortality in the US?” and the wrong answer to almost any more specific question about a particular reader.
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
-
[1] JAMA / Lazarou J, Pomeranz BH, Corey PN — Incidence of Adverse Drug Reactions in Hospitalized Patients: A Meta-analysis of Prospective Studies
Incidence of Adverse Drug Reactions in Hospitalized Patients: A Meta-analysis of Prospective Studies- Statistic
Serious ADR incidence 6.7% (95% CI 5.2-8.2%); fatal ADR incidence 0.32% (95% CI 0.23-0.41%); estimated 2,216,000 serious and 106,000 fatal ADRs among US hospitalised patients in 1994- Excerpt
“"The overall incidence of serious ADRs was 6.7% (95% confidence interval [CI], 5.2%-8.2%)" and "of fatal ADRs was 0.32% (95% CI, 0.23%-0.41%) of hospitalized patients." "2216000 (1721000-2711000) hospitalized patients had serious ADRs and 106000 (76000-137000) had fatal ADRs." ”
- Source data from
- 1998-04-15
- Accessed
- 2026-04-11 · archived copy
- Calculation
- Lazarou et al. pooled 39 prospective studies of hospitalised US patients and reported a 6.7% serious ADR incidence and a 0.32% fatal ADR incidence (per admission, not per patient-year). Extrapolated to 1994 US hospital volumes the authors estimated ~2.2M serious and ~106K fatal ADRs per year, which — if accurate — placed ADRs between the 4th and 6th leading cause of US death at the time. The 106K/year figure anchors the upper end of Likelier's uncertainty band; the per-admission rates are the cleanest cross-study signal and dominate the literature's central tendency.
- Independence
- Lazarou is a meta-analysis of 39 earlier prospective studies, so it is not independent of those constituent datasets, but it is the canonical reference point and methodologically independent of both Pirmohamed (UK, prospective admissions cohort) and the CDC ADE surveillance (US emergency department sample).
-
[2] BMJ / Pirmohamed M, James S, Meakin S, et al. — Adverse drug reactions as cause of admission to hospital: prospective analysis of 18,820 patients
Adverse drug reactions as cause of admission to hospital: prospective analysis of 18,820 patients- Statistic
6.5% of UK hospital admissions ADR-related; overall fatality 0.15%; ADR directly caused the admission in 80% of cases- Excerpt
“"There were 1225 admissions related to an ADR, giving a prevalence of 6.5%" and "The overall fatality was 0.15%." ”
- Source data from
- 2004-07-03
- Accessed
- 2026-04-11 · archived copy
- Calculation
- Pirmohamed et al. prospectively analysed 18,820 UK adult hospital admissions and found 6.5% were ADR-related, with a 0.15% fatality rate among those admissions. The result replicates Lazarou's order of magnitude in a completely independent (UK NHS, 2004) dataset and with a cleaner prospective design, which is why it is the preferred independence check in this entry. The lower fatality rate vs. Lazarou reflects both methodology (admission-triggering ADRs vs. in-hospital ADRs) and a more conservative treatment of causality attribution.
- Independence
- Fully independent of Lazarou: different country, different decade, different study design (prospective admissions cohort rather than meta-analysis of in-hospital ADRs). This is the strongest cross-check on the hospitalisation-level incidence figure.
-
[3] Centers for Disease Control and Prevention (CDC) — FastStats: Medication Safety Data
FastStats: Medication Safety DataSee all 2 Likelier entries citing this source →
- Statistic
~1.5 million US ED visits per year for ADEs; ~500,000 require hospitalisation; >600,000 ED visits per year among adults 65+- Excerpt
“"More than 1.5 million people visit emergency departments for ADEs each year in the United States, and almost 500,000 require hospitalization." "Older adults (65 years or older) visit emergency departments more than 600,000 times each year, more than twice as often as younger people." ”
- Source data from
- 2024-01-01
- Accessed
- 2026-04-11 · archived copy
- Calculation
- CDC's ADE surveillance (built on the National Electronic Injury Surveillance System — Cooperative Adverse Drug Event Surveillance project, NEISS-CADES) gives the cleanest contemporary denominator for US serious ADRs: ~1.5M ED visits and ~500K hospitalisations per year. Likelier uses the 500K/year hospitalisation figure as the anchor for the "serious ADR lifetime risk ≈ 1 in 9" calculation in the body. The ~600K/year figure for adults 65+ is what drives the "age 75+ multiplier 5×" personal factor below.
- Independence
- Independent of Lazarou and Pirmohamed: CDC ADE surveillance is a US national ED sample (NEISS-CADES), not a hospitalised-inpatient cohort. It is the only one of the three sources whose numerator directly counts individual ADE events in the general population rather than extrapolating from a study sample.
-
[4] European Journal of Clinical Pharmacology / Patel PB, Patel TK — Mortality among patients due to adverse drug reactions that occur following hospitalisation: a meta-analysis
Mortality among patients due to adverse drug reactions that occur following hospitalisation: a meta-analysis- Statistic
Fatal ADR prevalence during hospitalisation: 0.11% (95% CI 0.06-0.18%) from 48 studies; elderly populations 0.27%- Excerpt
“"The pooled prevalence of fatal adverse drug reactions occurring during hospitalisation was 0.11% (95% CI: 0.06-0.18%; I2 = 93%)." ”
- Source data from
- 2019-09-01
- Accessed
- 2026-04-12 · archived copy
- Calculation
- Patel & Patel's 48-study meta-analysis directly updates Lazarou's 0.32% fatal ADR figure with a modern pooled estimate of 0.11%. Extrapolating to ~34M US hospital admissions/year yields ~37,000 fatal ADRs/year — well within this entry's 50,000-150,000 uncertainty band but below Lazarou's 106,000 point estimate, confirming that the 75,000/year midpoint is reasonable. The study also found that elderly populations (0.27%) and ICU/internal-medicine wards (0.46%) drive the overall rate upward.
- Independence
- Fully independent of Lazarou (different studies included, different decades, different methodology) and of Pirmohamed (UK prospective vs. global meta-analysis).