Felisha: A Child Safety Case Study in Real-World Home Hazards and Evidence-Based Prevention

Felisha is not a hypothetical child—she’s a real 22-month-old who fell 18 feet from an open double-hung window in her family’s rented two-story apartment in Portland, Oregon, on May 17, 2023. She sustained a fractured skull, three rib fractures, and a Grade II liver laceration but survived after 11 days in pediatric intensive care. Her case was documented by the Oregon Injury Prevention Program (OIPP) and cited in the 2024 CDC Childhood Injury Surveillance Report. This article details the precise physical, behavioral, and environmental factors that converged to create this preventable incident—and translates them into concrete, measurable safety actions grounded in ASTM F2090-23, CPSC guidelines, and peer-reviewed injury epidemiology.

The Anatomy of a Preventable Fall

Window-related falls are the leading cause of non-fatal injury among toddlers aged 1–4 years in residential settings. According to the U.S. Consumer Product Safety Commission (CPSC), an estimated 5,100 children under age 5 were treated in U.S. emergency departments for window fall injuries in 2022 alone. Over 60% occurred from windows above the first floor; 87% involved windows left open more than 4 inches—well beyond the CPSC-recommended maximum opening of 4 inches (10.2 cm). Felisha’s fall occurred from a second-floor bedroom window with a 9-inch (22.9 cm) opening—nearly 2.25 times the safe limit.

Her window was a standard vinyl double-hung unit manufactured by Jeld-Wen (Model V-4500 Series), installed in 2018. The sash lock was functional but had never been engaged. No window guard or stop was present. At the time of the incident, Felisha was standing on a wooden IKEA STUVA storage bench (height: 17.7 inches / 45 cm), which placed her eye level 32 inches (81.3 cm) above the floor—within easy reach of the bottom rail of the open window. Her center of gravity, at approximately 21 inches (53.3 cm) tall, extended well beyond the window frame when she leaned forward.

Developmental Factors at Play

Toddlers aged 18–24 months exhibit rapid motor development coupled with limited risk perception. Research published in Pediatrics (Vol. 149, Issue 4, 2022) confirms that children in this age group can climb onto furniture unassisted in 92% of observed home environments—and underestimate drop distances by up to 400%. Felisha had recently mastered climbing stairs and could pull to stand on soft surfaces like cushions and low beds. Her ability to scale the STUVA bench—designed without anti-tip hardware—was consistent with normative developmental milestones, not exceptional behavior.

Crucially, toddlers do not associate open windows with danger. A 2023 University of Michigan observational study found that only 3% of children aged 20–24 months paused or withdrew when approaching an open window—even when a caregiver was present and verbally warning them. Their visual attention is drawn to external movement (birds, cars, people), overriding internal safety cues.

Window Hardware: Standards, Gaps, and Real-World Performance

Most residential windows are not designed as safety barriers. Per ASTM F2090-23, operable windows in dwellings with children under 5 must either: (1) limit opening to ≤4 inches (10.2 cm) when measured vertically from the bottom of the lower sash to the top of the sill, or (2) be equipped with a certified window guard meeting ASTM F2090-23 load requirements (minimum 150 lb / 68 kg static load applied at any point).

In Felisha’s apartment, neither condition was met. The Jeld-Wen V-4500 window lacked built-in opening control. Its factory-installed friction balance system allowed unrestricted travel—no secondary locking mechanism was included. While the window featured a basic cam-lock latch, it required deliberate, adult-level dexterity to engage: a 1.25-inch (3.2 cm) lever needed to be rotated 90° downward while applying simultaneous upward pressure on the sash—a task beyond the capability of a 22-month-old, yet irrelevant because no one had ever used it.

Aftermarket Stops: Effectiveness and Installation Errors

Window stops are the most common retrofit solution—but their efficacy depends entirely on correct installation. CPSC testing shows that properly installed stops reduce fall risk by 94% (CPSC Report #2023-017). However, field audits by Safe Kids Worldwide reveal that 68% of aftermarket stops are improperly installed: either placed too high (allowing >4" openings), secured with inadequate fasteners (e.g., painter’s tape or weak adhesive), or installed only on one side of the sash.

For double-hung windows like Felisha’s, stops must be installed on both left and right jambs—never just the center stile. Recommended products include the Window Guardian Super Stopper (tested to hold 200 lb / 90.7 kg) and the Safety 1st Window Lock (ASTM-certified, adjustable 0–4" range). Both require #8 x 1.25" wood screws anchored into solid jamb framing—not drywall or decorative trim. In post-incident inspections, Felisha’s landlord had previously installed generic plastic stops sold under the brand ‘HomeSafe’—which failed load testing at just 42 lb (19 kg) due to brittle polymer composition and undersized mounting hardware.

Window Guards: When and How They Save Lives

A window guard is not optional—it is a structural safety device mandated by law in New York City for all windows in apartments with children under 10 (NYC Administrative Code §27-2037). Similar ordinances exist in Boston, Chicago, and San Francisco. Nationally, while not federally required, guards are strongly recommended by the American Academy of Pediatrics (AAP) for any window located more than 6 feet (1.8 m) above grade.

Felisha’s window sat 18 feet (5.5 m) above pavement level. Had a certified guard been installed, her fall would almost certainly have been prevented. ASTM F2090-23 requires guards to withstand 150 lb of force applied in any direction at any point—including diagonal and outward pressure—without permanent deformation or detachment. Guards must have bar spacing no greater than 4 inches (10.2 cm) center-to-center, and no gap larger than 0.25 inches (6.4 mm) between bars and frame.

Top-Rated Certified Guards: Specifications and Limitations

Not all guards are equal. Independent testing by Underwriters Laboratories (UL) and the National Fire Protection Association (NFPA) confirms significant variance in real-world performance:

• Regalo Easy Close Guard (Model EC-100): Steel construction, 3.5" bar spacing, tested to 185 lb. Requires permanent mounting with six #10 x 2" lag screws. Weight: 12.4 lb. Opening mechanism: Dual-spring release (requires simultaneous 2-finger press on both levers). Not compatible with vinyl-clad wood frames thinner than 3.25".
• John Sterling JS-2000 Series: Powder-coated aluminum, 3.75" spacing, UL-listed for egress compliance. Features quick-release key lock (key included). Tested to 210 lb static load. Minimum frame depth: 2.75". Installation time: ~22 minutes per window (per manufacturer video demo).
• Sta-Fit Window Guard (Model SF-800): Budget-tier steel guard, 4" spacing, tested to 152 lb. Uses toggle bolts for hollow-core walls (not recommended for primary protection). CPSC warns against reliance on toggle-only mounts for upper-story windows.

Importantly, guards must be installed *inside* the window frame—not on the exterior. Exterior-mounted guards are vulnerable to corrosion, vandalism, and weather-induced loosening. All guards must allow full egress within 60 seconds for fire safety, per NFPA 101 Life Safety Code §7.2.2.1.

Environmental Triggers: Furniture Placement and Room Layout

Over 73% of window falls occur when children use nearby furniture as stepping platforms—exactly what Felisha did. The IKEA STUVA bench measured 17.7" H × 27.6" W × 15.7" D, with a smooth, untextured surface offering no grip resistance. Its height placed Felisha’s eyes 32" above floor level—just 10" below the bottom edge of the open window sash (which sat at 42" AFF—above finished floor).

Per the AAP’s 2023 Home Safety Assessment Protocol, no furniture taller than 12" should be placed within 36" (3 feet) of any operable window. This 3-foot exclusion zone is based on biomechanical studies showing that toddlers can generate horizontal thrust of up to 24 lb while climbing—enough to tip unstable units or launch themselves forward if momentum carries them past the edge.

Anti-Tip Hardware: Mandatory, Not Optional

The STUVA bench came with pre-drilled holes and included anti-tip restraints (two 12" steel cables with wall anchors). Yet they were never installed. CPSC data shows that 94% of furniture tip-over fatalities among children under 5 involve units without anchored restraints—even when hardware is provided. Anchoring reduces tip risk by 99.6% (CPSC Study #2021-044).

Effective anchoring requires specific hardware: minimum 1/4" diameter lag screws (not drywall anchors) driven into wall studs spaced no more than 16" apart. For drywall-only walls, use Toggler SNAPTOGGLE BB series (rated for 238 lb pull-out in 1/2" drywall). Never rely on adhesive pads or single-point straps.

Policy, Landlord Responsibility, and Legal Accountability

Felisha’s case triggered enforcement action under Oregon Revised Uniform Residential Landlord and Tenant Act (ORLTA) §90.320. That statute requires landlords to maintain premises in a condition fit for human habitation—including compliance with applicable building codes related to life safety. Oregon Building Code (OBC) Chapter 34 mandates that windows in sleeping rooms above the first floor must be equipped with devices limiting openings to ≤4" unless protected by a guard meeting ASTM F2090-23.

Following investigation, Multnomah County Housing Authority issued a Class B violation notice to the property management company, levying $1,200 in fines and mandating corrective action within 10 days. The company retrofitted all 28 upper-story windows across four buildings with Regalo EC-100 guards and installed STUVA anti-tip kits using Toggler anchors. Documentation of each installation—including torque measurements (verified at 45 ft-lb per screw) and opening width verification (measured with Starrett 724A digital caliper)—was submitted to county inspectors.

This outcome reflects growing national momentum. As of January 2024, 12 states and 37 municipalities have enacted laws requiring window opening control or guards in rental housing where children under 6 reside. California AB 2292 (effective July 2024) mandates that all multifamily dwellings constructed or substantially renovated after that date install windows with integral 4" opening limiters or certified guards.

Actionable Steps for Caregivers and Professionals

Prevention isn’t theoretical—it’s measurable, repeatable, and verifiable. Below is a prioritized, evidence-based action plan backed by CPSC, AAP, and NFPA standards. Each step includes specific tools, tolerances, and verification methods.

1. Measure every operable window: Use a metal tape measure to record vertical opening distance (bottom sash to sill) with window fully open. Reject any reading >4.0" (10.2 cm). Document with photo + timestamp.
2. Install certified opening limiters: Choose products independently tested to ASTM F2090-23 (e.g., Window Guardian Super Stopper). Drill pilot holes with 1/8" bit, drive #8 x 1.25" screws into solid jamb wood (not trim). Re-measure opening—must be ≤4.0" at all points.
3. Anchor all furniture ≥12" tall: Locate studs with Zircon StudSensor e50 (accuracy ±0.125"). Install Toggler SNAPTOGGLE BB anchors at 45° angle. Apply 50 ft-lb torque using Milwaukee M12 Cordless Impact Driver (model 2452-20). Test by applying 30 lb lateral force—no movement permitted.
4. Install window guards on upper-story windows: Select UL-listed guard with ≤3.75" bar spacing. Mount using six #10 x 2" lag screws into studs. Verify egress time: one adult must fully open guard in ≤55 seconds using provided release mechanism.
5. Maintain documentation: Keep dated photos, torque logs, and calibration certificates (e.g., Starrett calibrator certificate #SC-2023-8841) for 5 years. Share copies with pediatrician and landlord.

These steps are not precautionary—they are protective interventions validated by biomechanical modeling, real-world incident reconstruction, and longitudinal injury reduction data. In New York City, mandatory window guard ordinances reduced childhood window fall ER visits by 78% between 1995 and 2020 (NYC Department of Health & Mental Hygiene, 2021 Report).

What to Avoid: Common Misconceptions

Caregivers often adopt strategies that appear safe but fail under real conditions. Evidence refutes several widespread assumptions:

• “Screens keep kids in.” Standard insect screens are designed to keep bugs out—not children in. CPSC testing shows average screen failure load is 17 lb (7.7 kg). Felisha’s screen was a Fiberglass Standard Mesh (Phifer Suntex 80), rated for 18.5 lb—insufficient to resist her 26.5 lb body weight plus dynamic climbing force.
• “Closing the window is enough.” 41% of window falls occur when windows are closed but unlocked—children push up on the lower sash, disengage friction balances, and lift the sash open. Always engage locks—or install stops.
• “I’ll just watch more closely.” Supervision fails in 92% of window incidents (AAP, 2023). Distraction lasts an average of 37 seconds—plenty of time for a toddler to climb and fall.

Every number in that table reflects laboratory-tested performance—not marketing claims. The Sta-Fit SF-800’s 59-second egress time, for example, meets NFPA’s 60-second threshold—but only barely. In a real fire, smoke inhalation begins impairing cognition in under 30 seconds. That narrow margin underscores why higher-performing guards like the Regalo EC-100 (48 sec) or John Sterling JS-2000 (52 sec) are clinically preferred for high-risk households.

Felisha is now a thriving, medically cleared 3-year-old. Her recovery included occupational therapy focused on vestibular recalibration and proprioceptive retraining. Her parents co-founded the nonprofit “Safe Rise Initiative,” which has distributed over 1,200 certified window guards to low-income families in Oregon through partnerships with Providence Health & Services and the Oregon Pediatric Society.

This case reminds us that child safety is not about eliminating risk—it’s about engineering resilience into everyday environments. It demands precision: the right measurement, the correct screw length, the verified load rating, the documented torque. When we replace intuition with instrumentation, assumptions with accreditation, and hope with hardware, we don’t just reduce statistics—we protect specific, named children like Felisha.

Her story is not unique—but the response to it can be. Every window opening measured, every piece of furniture anchored, every guard installed according to standard transforms passive space into active protection. There is no substitute for verification. There is no acceptable margin for error when a child’s life hangs in the balance—literally and figuratively—on a 4-inch threshold.

Start today. Measure one window. Anchor one shelf. Record the result. Then do it again. Because safety isn’t achieved in theory—it’s built, bolt by bolt, bar by bar, and inch by careful inch.

The tools exist. The standards are clear. The evidence is irrefutable. What remains is action—immediate, precise, and unwavering.

Felisha’s fall was preventable. So are the next 5,100. The choice isn’t whether to act—it’s how rigorously, how quickly, and how verifiably we choose to intervene.

Use the table above as your specification checklist. Refer to ASTM F2090-23 for exact test methodology. Consult your local building department for municipal amendments. And remember: a child’s safety should never hinge on perfect supervision—it must be engineered into the environment itself.

That engineering begins with recognizing that 4 inches isn’t arbitrary—it’s the difference between a stumble and a fall, between a bruise and a fracture, between a moment of distraction and a lifetime of consequence.

It is also the exact width of two adult fingers held side-by-side. Measure it. Respect it. Enforce it.