Only 1 Vitamin D death around the world in a decade

Accidentally took 600,000 IU of Vitamin D for 21 days
(Prescription was for Growth hormone, NOT Vitamin D)

Note: 12,600,000 IU. A bottle containing 100 capsules of 50,000 IU = 5 million IU

Deep Research by Claude AI - June 2026

Fatal vitamin D toxicity is vanishingly rare. Across a decade of global medical literature, only a handful of deaths are attributable to vitamin D excess, and in the largest US poison-surveillance dataset (NPDS 2000–2014, 25,397 single-substance vitamin D exposures), there were "no fatalities, but five (0.02%) major effect outcomes" (Spiller et al., Hum Exp Toxicol 2016).
The clearest confirmed fatal case (2016, India) involved a massive prescribing error — 600,000 IU/day for 21 days in a previously healthy 10-year-old. The most publicized adult case (David Mitchener, UK, 2023) was an 89-year-old with significant cardiac/renal disease in whom vitamin D toxicity was one of several listed causes — contributory rather than sole.
Trial-level mortality signals concern dosing regimen, not toxicity deaths: large RCTs of bolus/high-dose vitamin D show increased falls/fractures and a borderline cancer-mortality signal (D-Health), but no excess of toxicity deaths. Routine daily supplementation shows no all-cause mortality harm.

Key Findings

Toxicity requires extreme, sustained intake: typically tens of thousands to hundreds of thousands of IU/day for weeks-to-months, producing serum 25(OH)D well above 150 ng/mL (>375 nmol/L), suppressed PTH, and hypercalcemia.
The mechanism of death is hypercalcemia and its sequelae: acute kidney injury, nephrocalcinosis, cardiac arrhythmia/conduction block, vascular and soft-tissue calcification, encephalopathy, and pancreatitis.
Predisposing conditions dramatically lower the toxic threshold: granulomatous disease (sarcoidosis, TB), CYP24A1 loss-of-function mutations, and chronic kidney disease.
Most "incidents" are manufacturing/compounding/dosing errors and mislabeled supplements — the dominant modern cause — but the overwhelming majority of even these high-exposure cases survive with treatment.
Poison-control surveillance shows thousands of vitamin D exposures annually with essentially no deaths, providing strong denominator context.

Details

1. Published fatal case reports (2015–2025)

Tier A — death with vitamin D as the primary/triggering cause:

Narsaria, Sankar & Lodha, "Fatal Outcome of Accidental Vitamin D Overdose," Indian Journal of Pediatrics, 2016. https://doi.org/10.1007/s12098-016-2109-z A 10-year-old boy being treated for short stature was erroneously prescribed 600,000 IU of vitamin D daily for 21 days (intended to be growth hormone). He developed abdominal pain and vomiting, then refractory hypercalcemia (calcium persistently >14 mg/dL despite hemodialysis), 25(OH)D >160 ng/mL, suppressed PTH (5.34 pg/mL), hypertensive encephalopathy with posterior reversible encephalopathy syndrome (PRES) on neuroimaging, pulmonary edema, and hypercalcemia-induced acute pancreatitis. He died of his illness complicated by nosocomial sepsis and shock. The authors note that "death due to vitamin D toxicity has not been reported" — i.e., they presented this as a sentinel fatal case. This is the cleanest confirmed fatality in the window: a previously healthy child, gross overdose, no competing cause.

Tier B — death with vitamin D toxicity listed as contributory among multiple causes:

David Mitchener (UK), HM Assistant Coroner for Surrey Jonathan Stevens, Regulation 28 Prevention of Future Deaths report, dated 19 January 2024; inquest concluded 28 December 2023. An 89-year-old man was admitted to East Surrey Hospital on 10 May 2023 with hypercalcaemia and died there on 20 May 2023 despite treatment. Per the report, ante-mortem tests "revealed vitamin D levels at 380 – the maximum level recordable by the laboratory" (units not stated in the report; presumably nmol/L). Post-mortem cause of death: 1(a) congestive cardiac failure and acute-on-chronic kidney failure; 1(b) hypercalcaemia; 1(c) vitamin D toxicity; 2. ischaemic heart disease. He had taken vitamin supplements purchased from NaturPlus UK for at least the preceding 9 months, with "no warnings on or in the packaging." Inquest conclusion: death by misadventure. Causation note: vitamin D toxicity appears on the death certificate but within a multifactorial cause-of-death in an elderly man with significant cardiac and renal disease — properly classified as contributory, not sole.

Context — a pre-window case-series death (denominator framing): A case series of adults from Kashmir (an endemically deficient region) documented 10 vitamin D-toxicity patients (doses 3.6–210 million IU cumulative over 1–4 months); nine recovered and "one patient expired due to multiorgan failure." Published 2011, before the 2015–2025 window, but it illustrates that even in hospitalized severe-toxicity series, fatality is the exception (1/10).

Notable severe but non-fatal cases (the survivable norm): Numerous 2015–2025 reports describe severe hypercalcemia, AKI, complete heart block, and need for peritoneal/hemo-dialysis, with 25(OH)D levels up to 746 ng/mL and ~2016 ng/mL — virtually all resolving with hydration, calcitonin, bisphosphonates/denosumab, and steroids. This pattern is the rule: vitamin D toxicity is usually reversible.

2. Notable incidents, errors, outbreaks, and regulatory actions

Manufacturing/compounding errors are the leading modern mechanism. A Turkish series (Kara et al.) reported 7 children (aged 0.7–4.2 years) with toxicity from a fish-oil supplement mismanufactured with excessive vitamin D; initial median calcium 16.5 mg/dL (range 13.4–18.8) and median 25(OH)D 620 ng/mL (range 340–962). All survived.
Compounding error (Brazil): a 53-year-old with diabetes, hypertension, and CKD received a dose ~2,000-fold higher than prescribed; severe hypercalcemia and worsening renal failure; survived.
Infant dosing/concentration errors: multiple reports of infants given ~40,000–50,000 IU/day (up to ~50× the upper limit) from dropper/concentration errors, causing severe hypercalcemia and nephrocalcinosis; survived with treatment. A separate report documents 25× overdosing (2,000 IU/drop product) over seven months causing nephrocalcinosis in a 22-month-old.
Foodborne outbreak (2025, India): a household cluster of 3 adults developed severe hypercalcemia and AKI from cooking oil contaminated with vitamin D3 at 1,286 µg/mL (~3.34 × 10⁶ nmol/L), confirmed by LC-MS/MS; one required dialysis; all recovered. Neighbors using the same oil were also affected.
US recall (FDA, Feb 2024): Nordic Naturals voluntarily recalled one lot (~3,800 units; lot 234909) of Baby's Vitamin D3 Liquid (labeled 400 IU/0.25 mL) due to a manufacturing error producing a "super potent dose." No adverse events were reported; FDA later terminated the recall.
UK regulatory response: the Mitchener Regulation 28 report was sent to the Food Standards Agency and the Department of Health & Social Care, urging warning labels on supplement packaging ("there is a risk that future deaths could occur unless action is taken"). NHS England separately documented a prescribing-frequency error pattern — a review of the National Reporting and Learning System found 42 reports over two years of patients receiving high-strength vitamin D more frequently than intended, some hospitalized for hypercalcaemia — prompting British National Formulary and Specialist Pharmacy Service guidance updates.
Animal/veterinary (context only): in 2018–2019 the FDA recalled dozens of dry dog-food products for toxic vitamin D levels, with some pet deaths. Not human deaths, but it demonstrates the same manufacturing-error pathway.

3. Poison-control / surveillance data (strong denominator)

US NPDS, 2000–2014 (Spiller et al., Human & Experimental Toxicology 2016; 35(5):457–61): 25,397 single-substance vitamin D exposures; exposures rose ~1600% between 2005 and 2011 (annual mean rising to 4,535). The verbatim outcome: "There were no fatalities, but five (0.02%) major effect outcomes." Cases were limited to vitamin D alone (multivitamins excluded); 98% ingestion; majority young children.
US NPDS, 2021 (39th Annual Report, Gummin et al., Clinical Toxicology 2022; 60(12):1381–1643): ~11,718 single-substance vitamin D exposures (the figure cited in StatPearls and traceable to this report), more than half in children under 5, overwhelmingly unintentional. Across the broader "vitamins" category there were ~50,120 single exposures with 7 major outcomes and no deaths. Vitamin D exposures exceeded combined vitamins B, A, C, and E — reflecting how widely it is used.
US NPDS, 2022 (40th Annual Report, Gummin et al., Clinical Toxicology 2023; 61(10):717–939): of ~2.06 million human exposures, vitamins do not appear among the substance categories associated with the largest number of fatalities; no vitamin D death is recorded. (The exact 2022 vitamin D single-substance count sits in Appendix B Table 22B and was not separately machine-extractable, but the fatality tables contain no vitamin entry.)
Cross-year pattern: independent reviews of the NPDS Table 22B death column across multiple recent annual reports consistently find no vitamin D deaths in any year 2015–2024.
Limitations: NPDS is passive and self-reported, lacks a sales denominator, captures exposures (not adjudicated poisonings), and supplement fatality abstracts are frequently missing — so the system likely undercounts. But the near-total absence of deaths against tens of thousands of exposures per year is a strong rarity signal.

4. Epidemiological / mortality-signal evidence (RCTs and meta-analyses)

This is a distinct question from toxicity death — it concerns whether supplementation regimens shift mortality.

VITAL (Manson et al., NEJM 2019; 380:33–44): 25,871 US adults, 2,000 IU/day vitamin D3 for a median 5.3 years. No effect on all-cause mortality (HR 0.99, 95% CI 0.87–1.12) or major CVD. For cancer death (341 deaths) the HR was 0.83 (95% CI 0.67–1.02) — non-significant — strengthening to HR 0.79 (0.63–0.99) excluding year 1 and HR 0.75 (0.59–0.96) excluding the first 2 years. No mortality harm.
D-Health (Neale et al., Lancet Diabetes & Endocrinology 2022; 10:120–128): 21,315 Australians aged ≥60 (10,662 vitamin D / 10,653 placebo), 60,000 IU/month for 5 years. No reduction in all-cause mortality (per-protocol OR 1.18, 95% CI 1.00–1.40, p=0.06). In analyses excluding the first 2 years, the vitamin D arm showed "a numerically higher hazard of cancer mortality than those in the placebo group (HR 1.24 [95% CI 1.01–1.54]; p=0.05)." The authors invoked the precautionary principle for this monthly bolus regimen in vitamin D-replete people. This is a hypothesis-generating signal, not a demonstration of toxicity deaths.
Bolus-dosing harm signal (falls/fractures): Sanders et al. (JAMA 2010; 303(18):1815–1822), annual single 500,000 IU in older women, found 171 fractures vs 135 (26% more) and ~15% more falls, with "an incidence RR of falling in the vitamin D group vs the placebo group of 1.31 in the first 3 months after dosing and 1.13 during the following 9 months." The ViDA trial (monthly 100,000 IU) found no benefit on falls/fractures. The mechanism appears independent of hypercalcemia. These are morbidity signals tied to large intermittent doses, not fatal toxicity.

What this does NOT show: None of these trials show people dying of vitamin D toxicity/hypercalcemia. The D-Health cancer-mortality finding is borderline (p=0.05), derived from a post-hoc/exploratory analysis, not replicated in VITAL, and concerns a specific monthly bolus regimen — it is a within-trial signal, not established causation, and emphatically not "vitamin D toxicity deaths."

5. Context, denominator, thresholds, and predisposition

How rare: Tens of millions supplement daily; US poison centers log roughly 10,000+ vitamin D exposures per year with essentially zero deaths. A Mayo Clinic population study (Olmsted County, 2002–2011; 20,308 measurements) found only one case of true acute toxicity (25(OH)D 364 ng/mL in someone taking 50,000 IU/day for >3 months) and <1% of people exceeded 100 ng/mL. Confirmed fatal cases in the world literature over the past decade number in the single digits.
Dose/duration to reach fatal toxicity: chronic intake on the order of 10,000–60,000+ IU/day for months is generally needed to produce toxicity (Mayo/StatPearls cite ~60,000 IU/day for several months); fatal cases involved far higher (e.g., 600,000 IU/day). Symptomatic toxicity generally shows 25(OH)D >150 ng/mL and often 300–1,000+ ng/mL.
Predisposing factors: (a) granulomatous disease (sarcoidosis, TB) — extrarenal 1α-hydroxylase converts 25(OH)D to active 1,25(OH)₂D, so even modest doses cause hypercalcemia; up to ~10% of sarcoidosis patients are hypercalcemic, and case reports document hypercalcemia/AKI after ordinary cholecalciferol doses; (b) CYP24A1 loss-of-function mutations — impaired vitamin D catabolism; can present in adulthood with hypercalcemia/nephrocalcinosis after even modest supplementation; (c) chronic kidney disease; (d) concurrent calcium supplementation.

Recommendations

Editorial stance for the resource: State plainly that fatal vitamin D toxicity is extraordinarily rare and almost always requires gross overdose (manufacturing/compounding/prescribing error) or a predisposing condition. Do not imply that ordinary supplemental doses (≤4,000 IU/day) cause death — the evidence does not support this.
Tier the evidence explicitly: (A) one clean confirmed fatal case — pediatric, 600,000 IU/day prescribing error; (B) one high-profile contributory case — Mitchener, multifactorial, elderly; (C) surveillance showing ~zero deaths against tens of thousands of exposures annually; (D) trial mortality signals that concern dosing regimen, not toxicity.
Flag the trial signal honestly but proportionately: present the D-Health cancer-mortality finding (HR 1.24, p=0.05, exploratory) as a borderline, regimen-specific signal not replicated in VITAL — and separate it cleanly from "toxicity deaths."
Thresholds that would change the message: if future NPDS annual reports begin recording vitamin-D-attributed deaths, or if a large RCT confirms a significant all-cause or cancer mortality increase from a defined regimen, the "safe in normal use" framing should be revisited. Until then, the data support routine daily dosing as safe and large intermittent bolus megadosing as the regimen of concern.
For at-risk groups (sarcoidosis/granulomatous disease, known CYP24A1 mutation, CKD): recommend caution, lower thresholds, and monitoring of serum calcium and 25(OH)D before and after supplementation.

Caveats

Confirmed vs. contributory: the literature rarely adjudicates causation rigorously; "death attributed to vitamin D" often means it was one of several listed causes (e.g., Mitchener). True sole-cause fatalities are rarer still — essentially the pediatric overdose case in this window.
Surveillance undercounts: NPDS is passive and self-reported, lacks a denominator, and supplement fatality abstracts are often missing; absence of recorded deaths is strong but not absolute proof of zero deaths.
Publication bias: case reports favor dramatic/severe presentations; the survivable majority and asymptomatic high-25(OH)D cases are under-reported, which can paradoxically make toxicity look more dangerous than it is.
Unit confusion is a recurring real-world hazard (ng/mL vs nmol/L; IU vs mcg; drops per mL), itself a driver of dosing-error toxicity — and a reason the Mitchener "380" value lacks a stated unit.
Trial signals are regimen-specific: harm signals attach to large intermittent boluses (annual/monthly megadoses), not to daily physiologic dosing; do not generalize across regimens.