What are the odds of a child being killed or seriously injured by falling from a window or balcony?
Evidence quality 4.38/5
Eight-dimension review score against the quality rubric . Each dimension scored 1–5.
- D1 Source grounding
- 4/5
- D2 Source authority
- 5/5
- D3 Arithmetic
- 5/5
- D4 Uncertainty
- 4/5
- D5 Scope
- 5/5
- D6 Prose
- 4/5
- D7 Perception honesty
- 3/5
- D8 Caveat completeness
- 5/5
Lifetime probability · lifetime, subgroup
1 in 855
0.1% lifetime chance
Most people underestimate this.
range 1 in 1,429 to 1 in 500
● your factors — click this risk ▾ to reveal
≈ As likely as
Perceived
No published US perception survey isolates window or balcony falls as a standalone parental fear, so this entry uses editorial intuition. Parents of toddlers tend to think about windows the way they think about stair gates and outlet covers — a once-and-done childproofing item that fades from attention after the first year. The American Academy of Pediatrics has issued a dedicated policy statement on falls from heights since 2001 precisely because the underlying intuition runs the wrong way: peak risk is concentrated in ages 1 to 4, not infancy, and the mechanism is climbing onto sofas, beds, or chairs adjacent to an opened window — a developmental window most parents associate with greater autonomy and lower fragility. Balcony rail spacing and the absence of window guards in most US private housing are not common dinner-table topics. The fear is genuinely underrated relative to how preventable the events are.
Rough estimate: Most parents likely treat windows as a low-priority childproofing item past the first year, when ages 1-4 are actually the peak-risk window.
Source: editorial intuition, not polled
Actual
~5,180 US children aged 0-17 treated in EDs annually for window-fall injuries (Harris 2011, NEISS 1990-2008)
US children aged 0-17
Show derivation
Harris VA, Rochette LM, Smith GA (Pediatrics 2011, NEISS 1990-2008) estimate 98,415 US ED visits for window-fall injuries among children 0-17 across the 19-year study period — an average of 5,180/year. The mean age was 5.1 years and children 0-4 accounted for approximately 65% of injuries. Splitting against the ~73 million US children under 18: - 0-4 (≈20M children): 65% × 5,180 ≈ 3,370 cases/year → ~16.8 per 100,000/year - 5-9 (≈20M children): ~25% × 5,180 ≈ 1,295 cases/year → ~6.5 per 100,000/year Cumulative through age 10 (independent-trial compounding): - 0-4 window: 1 − (1 − 0.000168)^5 ≈ 0.000840 - 5-9 window: 1 − (1 − 0.000065)^5 ≈ 0.000325 - Combined: 1 − (1 − 0.000840)(1 − 0.000325) ≈ 0.00116 Rounded: ~0.00117, or ~1 in 860 per US child by age 10. Adding balcony falls (not captured in Harris's window-specific NEISS extraction) pushes the figure moderately higher; international series and the AAP policy statement bracket windows + balconies together as the same prevention category, so the headline is best read as a window-anchored lower bound for the combined mechanism. Fatal window falls are an order of magnitude rarer: SafeKids / UC Davis (2024) estimates roughly 8 fatal window falls per year in US children under 5, suggesting an annual under-5 fatality rate around 0.4 per 100,000 and a cumulative fatal probability through age 10 on the order of 1 in 380,000 — two-and-a-half orders of magnitude below the serious-injury headline. Scope is subgroup_lifetime: this is the probability that a given US child experiences at least one qualifying ED visit during the first decade of life, not a US adult's remaining lifetime probability.
Caveats: The headline is the per-child cumulative probability of an ED visit for a window…
The headline is the per-child cumulative probability of an ED visit for a window-fall injury through age 10, derived from Harris et al. (Pediatrics 2011) NEISS data spanning 1990-2008. Three structural caveats apply. First, NEISS captures windows specifically as the product code; balcony falls are not separately tracked at national scale in the US, so the headline is a window-anchored lower bound for the combined window-plus-balcony mechanism that the question asks about. Second, ~25% of these ED visits result in hospitalization and ~26% involve a head or brain injury — the headline therefore mixes "serious enough to warrant ED imaging" with "admitted to the hospital" and "brain injury", which span a wide severity range; readers wanting a fatal-only number should subtract roughly two orders of magnitude (cumulative fatal window-fall probability through age 10 is on the order of 1 in 380,000 per child). Third, the Harris dataset ends in 2008; subsequent NEISS extractions (Academic Pediatrics 2020) find the annual incidence in 0-4-year-olds has declined modestly since then, consistent with steady but slow adoption of window guards and stops in US private housing. Risk is heavily concentrated in age 1-4, in dwellings of two or more stories, and in homes without operable window guards or 4-inch opening limiters — the NYC "Children Can't Fly" experience shows that the residual risk in guard-equipped dwellings drops by roughly 96%, putting the achievable floor well below the population-average headline. Excludes intentional falls, falls from playground or sports equipment (separate mechanism), and falls from cribs or changing tables (covered in infant-fall-from-furniture).
Risks at similar odds
Other risks with roughly the same likelihood — useful for calibration.
Baby walker injury
What are the odds an infant in a baby walker is treated in the emergency department for a walker-related injury?
Untreated childhood scoliosis
What are the odds of serious disability from untreated childhood scoliosis?
Child rear bike seat
What are the odds of serious injury to a child riding in a rear-mounted bike seat?
Toddler stair fall
What are the odds of a toddler suffering serious injury from falling down stairs?
Child fall head injury
What are the odds of a child suffering a serious head or eye injury from a fall?
Recently viewed on this device
Stored locally — clear anytime.
Pick challenger
Harris, Rochette, and Smith’s 2011 Pediatrics analysis of the Consumer Product Safety Commission’s National Electronic Injury Surveillance System estimated 98,415 US children aged 0-17 treated in emergency departments for window-fall injuries between 1990 and 2008 — an average of 5,180 per year. The mean age was 5.1 years and boys accounted for 58.1% of cases. Splitting that flow against the ~73 million US children under 18, and re-weighting against the study’s finding that children aged 0-4 made up 65% of injuries, gives an annual rate of roughly 16.8 per 100,000 for under-5s and ~6.5 per 100,000 for ages 5-9. Compounded as independent trials across the first decade of life, the cumulative per-child probability of a window-fall ED visit through age 10 sits around 1 in 860. One in four of those ED visits ends in hospital admission, and the under-5 age band carries 3.22 times the head-injury rate of older children. Fatal window falls are a smaller subset — SafeKids and UC Davis Health cite roughly 8 fatal window falls per year in US children under 5, which works out to a cumulative fatal probability through age 10 on the order of 1 in 380,000, two-and-a-half orders of magnitude below the serious-injury headline.
The mechanism is well-characterized and remarkably uniform across US case series. Vish, Powell, Wiltsek, and Sheehan’s 2005 Injury Prevention analysis of 90 Chicago pediatric trauma-center cases found a median age of 2 years and that 98% of falls were from the third floor or lower — the dominant mental image of a window fall as a high-rise event is not what the US epidemiology shows. The recurring mechanism is a toddler climbing onto a sofa, bed, or chest placed adjacent to an opened window, leaning against the insect screen (which is not designed to bear human weight), and falling through. The American Academy of Pediatrics’ 2001 policy statement Falls From Heights: Windows, Roofs, and Balconies recommends limiting accessible window openings to four inches or less, opening double-hung windows from the top sash only, installing operable window guards (openable for fire egress) on second-story and higher windows, and ensuring balcony, deck, porch, and fire-escape railings have vertical spacing of no more than four inches. None of these interventions are universal requirements in US private housing outside specific municipal mandates.
The strongest natural experiment in the US literature is the New York City “Children Can’t Fly” program. Charlotte Spiegel and Francis Lindaman’s 1977 American Journal of Public Health paper documented the program’s launch in 1972 and a 50% reduction in reported window falls in the Bronx between 1973 and 1975. The 1976 NYC Health Code amendment requiring landlord-provided window guards in any apartment housing children aged 10 or younger has since been credited with a roughly 96% reduction in pediatric window-fall hospitalizations relative to the pre-program baseline — a magnitude of effect that is rare in injury prevention and that has held across five decades. The implication for calibration is straightforward: the headline 1-in-860 figure is the population average across US housing, in which window guards are not the norm. In dwellings where guards or 4-inch opening limiters are installed and maintained, the residual risk falls by roughly a factor of 25, putting a guard-equipped child’s ten-year window-fall probability well below 1 in 20,000. The fear is genuinely underrated relative to the gap between what is achievable and what is the default, and the prevention menu is unusually concrete: a fixture, a measurement, and a maintenance check.
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] Pediatrics (American Academy of Pediatrics) — Harris VA, Rochette LM, Smith GA — Pediatric injuries attributable to falls from windows in the United States in 1990-2008
Pediatric injuries attributable to falls from windows in the United States in 1990-2008- Statistic
98,415 US children 0-17 treated in EDs for window-fall injuries 1990-2008 (avg 5,180/year); mean age 5.1 years; 58.1% boys; 25.4% hospitalized; ages 0-4 had 3.22× higher head-injury rate and 1.65× higher hospitalization/mortality rate than 5-17; hard landing surfaces 2.05× more head injury and 2.23× more hospitalization/death than cushioned- Excerpt
“"An estimated 98 415 children (95% confidence interval [CI]: 82 416-114 419) were treated in US hospital EDs for window fall-related injuries during the 19-year study period, averaging 5180 patients per year. The mean age of children was 5.1 years, and boys accounted for 58.1% of cases. One-fourth (25.4%) of the patients required admission to the hospital." ”
- Source data from
- 2011-09-01
- Accessed
- 2026-05-24 · archived copy
- Calculation
- Primary US population denominator. 5,180 cases/year against ~73 million children 0-17 → ~7.1 per 100,000/year averaged across the full pediatric age range. Re-weighting against the 65% concentration in ages 0-4 gives ~16.8 per 100,000/year for under-5 and ~6.5 per 100,000/year for 5-9, which combined as independent trials over the 0-10 window yields the ~0.00117 cumulative-probability headline. The Harris paper is the first nationally representative US dataset on pediatric window falls and remains the canonical denominator for any per-child US window-fall probability estimate.
- Independence
- NEISS-based national sample, Consumer Product Safety Commission surveillance pipeline. Independent of the trauma-registry and single-center sources cited below.
-
[2] Pediatrics (American Academy of Pediatrics) — Committee on Injury and Poison Prevention — American Academy of Pediatrics: Falls from heights: windows, roofs, and balconies
American Academy of Pediatrics: Falls from heights: windows, roofs, and balconies- Statistic
Approximately 140 deaths from falls occur annually in US children younger than 15; ~3 million children require ED care for fall-related injuries each year; AAP recommends limiting window opening to ≤4 inches, opening double-hung windows from the top only, installing operable window guards on second-and-higher-story windows, and balcony rail spacing ≤4 inches- Excerpt
“"Approximately 140 deaths from falls occur annually in children younger than 15 years, and 3 million children require emergency department care for fall-related injuries [...] preventive strategies [include] the installation of window guards and balcony railings." ”
- Source data from
- 2001-05-01
- Accessed
- 2026-05-24 · archived copy
- Calculation
- Authoritative US framing for the broader fall-death denominator (140/year across all fall mechanisms, all children under 15) and the prevention menu (4-inch window opening limit, top-sash opening, operable guards on 2nd-story and higher, balcony rail spacing ≤4 inches). Used here as the pediatric policy anchor: window and balcony falls share a prevention framework even though the surveillance data sets are typically separated by product/mechanism. The 140-deaths-per-year figure is the all-falls pediatric total — window-fall-specific fatality is a small subset (~8 fatal per year in under-5s per SafeKids 2024).
- Independence
- AAP policy statement synthesizing the field; methodologically distinct from NEISS surveillance and from the NYC public-health intervention evaluations. Provides the prevention-recommendations basis for the personal_factor_multipliers below.
-
[3] American Journal of Public Health (1977; reprinted Injury Prevention 1995) — Spiegel CN, Lindaman FC — Children can't fly: a program to prevent childhood morbidity and mortality from window falls
Children can't fly: a program to prevent childhood morbidity and mortality from window falls- Statistic
NYC 'Children Can't Fly' program began 1972; reported window falls in the Bronx declined 50% from 1973 to 1975; NYC Board of Health amended Health Code in 1976 to mandate landlord-provided window guards in apartments housing children aged 10 and younger; subsequent NYC data document a ~96% reduction in pediatric window-fall hospitalizations relative to pre-program baseline- Excerpt
“"Reported falls declined 50 percent from 1973 to 1975 [...] In 1976 the Board of Health amended the Health Code to require that landlords provide window guards in apartments where children ten years old and younger reside." ”
- Source data from
- 1977-12-01
- Accessed
- 2026-05-24 · archived copy
- Calculation
- Foundational US public-health evaluation showing the multiplier effect of window-guard mandates. The within-three-years 50% Bronx reduction is the directly attributable figure in the original paper; the broader 96% reduction headline frequently cited (e.g. by NYC DOHMH, Nationwide Children's Hospital press releases, and SafeKids) refers to the decade-after-mandate hospitalization reduction in NYC pediatric window falls and informs the window-guard multiplier in the personal-factor table (0.04× under mandated-guard conditions versus the unmandated baseline).
- Independence
- Pre-NEISS-era public-health intervention evaluation from the NYC DOHMH and Bronx Lebanon Hospital Center; independent of the modern NEISS and AAP synthesis sources above. The 50-year track record of the NYC window-guard mandate is the strongest natural-experiment evidence for the effectiveness of operable window guards in this dataset.
-
[4] Injury Prevention (BMJ) — Vish NL, Powell EC, Wiltsek D, Sheehan KM — Pediatric window falls: not just a problem for children in high rises
Pediatric window falls: not just a problem for children in high rises- Statistic
90 Chicago pediatric trauma-center cases 1995-2002; median age 2 years; 98% of falls were from third floor or lower; three deaths; head trauma and extremity fractures most common- Excerpt
“"The authors reviewed 90 cases; 55 were male. The median age was 2 years. [...] Ninety eight percent of falls were reported to be from the third floor or lower. [...] The most common injuries were head trauma and extremity fractures." ”
- Source data from
- 2005-10-01
- Accessed
- 2026-05-24 · archived copy
- Calculation
- Used to justify the building_type personal-factor multiplier and to anchor the "modest-height fall" framing in the prose: the prevailing mental image of a window fall as a high-rise event is wrong in the US epidemiology. 98% of Chicago cases were from the third floor or lower, with a 2-year median age that mirrors the under-5 concentration in Harris 2011. Three deaths in 90 trauma-center cases gives a single-center case fatality of ~3.3% conditional on reaching a pediatric trauma center, which is roughly consistent with the order-of-magnitude gap between the ED-visit headline and the cumulative fatal estimate in the assumptions field.
- Independence
- Single-center Chicago trauma-registry sample; corroborates the Harris NEISS-based national age distribution at the case-mix level and adds the building-height distribution that NEISS does not capture.