Highlights
-
Our study provides a detailed visualization of craniocerebral and vertebrospinal injuries.
-
Subarachnoid hemorrhage and contusions to the frontal and temporal lobes were the most common injuries acquired in fatal falls.
-
Age is the most critical risk factor in fatal fall injuries and is significantly associated with frontal contusions and subdural hematoma.
Introduction
Fatal falls are defined as a descent from a current position to a lower level, resulting in injuries that lead to death. These incidents can stem from various causes, including accidents, suicides and homicides, and affect individuals across all age groups. Reference Casali, Battistini, Blandino and Cattaneo1 In Canada, the number of fatal falls has been increasing at an alarming rate, with 6566 fatalities recorded in 2020 alone. 2 According to Statistics Canada, from 2000 to 2020, fatalities from falls among individuals aged 65 and older have increased significantly, while the rates for those under 65 have remained relatively stable (Figure 1). This trend may reflect underlying demographic changes, with the proportion of Canadians aged 65 and older increasing from 12.55% in 2000 to 17.90% in 2020. 3
Figure 1. Fatal falls in Canada from 2000 to 2020 (Statistics Canada). The number of fatal falls is increasing each year, primarily driven by the age group 65 years and older. The number for those under 65 years old remains relatively unchanged during these 20 years.
Several factors influence the types of injuries sustained by fall victims, including age, sex, pre-existing conditions, fall height and circumstances. It is essential for healthcare providers to be proactive in recognizing and informing individuals who are at risk of fall-related injuries to prevent additional harm or potential death.
Existing literature primarily focuses on fracture patterns, injury types, survival rates and recovery outcomes associated with falls. Reference Içer, Güloğlu, Orak and Ustündağ4–Reference Kent and Pearce5 However, fatal falls remain underexplored, particularly regarding their neuropathological implications. This retrospective study characterizes the craniocerebral and vertebrospinal injury patterns sustained by fatal falls within the Southwestern Ontario cohort. The relationships between these injuries and other intervening variables such as age, sex, comorbidities, fall height and fall surface are also investigated.
Materials and methods
Study design and data collection
This study was approved by the research ethics board of Western University (HSREB 120655). All autopsies were performed at the London Forensic Pathology Unit (FPU), the Department of Pathology, London Health Sciences Centre (LHSC). Autopsy reports between January 2000 and December 2020 were reviewed. Cases with a neuropathology consultation were included. Keywords “fall” or “descent from height” were used to search the autopsy warrant and the neuropathology consultation request form. There were 321 cases where keywords were mentioned within the reports. Sixty-three potential cases were initially identified after correlation of the final forensic autopsy reports, where the causes of death were attributed to fall-related injuries. External consult cases were excluded due to the lack of relevant clinical information. Forty-five cases were included for the final analysis. Craniocerebral injury was documented in all but one of the 45 cases due to severe autolysis. Variables affecting the impact of a fall were documented. Age, sex, scene information (location, height and landing surface), head/brain and spinal injury patterns, toxicology findings and comorbidities were obtained from the coroner’s warrants, neuropathology consultation reports and final autopsy reports. Falls from height and from stairs were not mutually exclusive; stair-related falls over 3 m were classified as both a high fall and a fall from stairs.
Neuropathology procedures
For all 45 cases included in this study, formal neuropathology consultations were performed. Whole brain or brain slices were retained and fixed in 10% neutral buffered formalin for 10 days before gross examination. Injuries were documented with photography. Representative sections were taken for microscopic examination using routine procedures. Reference Ding, Liu and Rebernig6,Reference Zhang, Suller-Marti and Ding7
Given that the case selection spans a 20-year period, the brain sampling protocols exhibited some variation over time. Nevertheless, a two-tiered framework was generally followed as a guiding principle. Tier 1 sampling comprised a set of core neuroanatomical regions collected in all cases, including representative cortical areas (frontal, temporal, parietal, occipital lobes and cingulate), subcortical structures (amygdala, hippocampus, basal ganglia, thalamus), the brainstem (midbrain, pons, medulla) and cerebellum (cerebellar vermis and hemisphere). Tier 2 sampling encompassed case-specific regions based on clinical context and gross pathology. This included additional sections through visible lesions (e.g., cortical contusions) or areas of clinical concern. In cases where diffuse axonal injury (DAI) was suspected, given the often subtle or absent gross findings, an expanded sampling protocol was employed. Targeted regions in such instances included the corpus callosum, internal capsule with adjacent deep gray matter, cerebellar white matter, caudal midbrain and rostral pons.
Additional neuropathological findings, such as neurodegenerative diseases and cerebral vascular diseases, were also documented where applicable. There is no established guideline for screening neurodegenerative diseases in this setting. A neurogenerative workup is typically initiated in decedents aged 65 years or older or when morphological features suggestive of such pathology are identified, such as amyloid plaques or Lewy bodies, which are often readily detectable on routine hematoxylin and eosin (H&E) stained sections.
Toxicology analyses were performed on all cases at the Ontario Centre of Forensic Sciences (www.ontario.ca/cfs).
Statistical analysis
Descriptive statistics were used to identify the most frequently observed neuropathological injuries and comorbidities. Contingency tables were created, and Fisher’s exact test was conducted to identify associations between the injury patterns and predictor variables (age, sex, toxicology, comorbidities, fall location and height of fall). The significance of the association was assessed using an alpha level of 0.05. To conduct logistic regression between these variables, toxicology and comorbidities were converted into binary (absent or present), as were the height of fall (>3 m or 0–3 m), age (older; ≥65 or younger; <65) and fall circumstances (stairs or no stairs). All statistical analyses were performed using the SPSS 29 (IBM Corp., Armonk, NY, USA) statistical software. Figures were generated using Microsoft Excel and BioRender.com.
Results
Patient demographics and fall circumstances
Of the 45 cases studied, 34 (75.6%) were male, and 11 (24.4%) were female. The mean age was 60.3 ± 18.1 years. Approximately half (48.8%) of our cases were under the age of 65. The mean time of survival was 50.8 hours, with two victims dying at the scene. Demographic information, comorbidities and information surrounding the circumstances of death are detailed in Table 1. The most common fall location was home (64.4%), followed by work (20.0%), which was most frequently from barns and construction sites. Cardiovascular disease was the most common pre-existing condition, followed by neurological disorders and metabolic disorders (e.g., diabetes) (Supplemental Table 1). Eight (18.2%) participants had a history of substance abuse, while 13 (29.5%) had a positive postmortem toxicology report (e.g., an elevated blood alcohol level). The majority of these patients were younger than 65 years old (Figure 2). There was a significant association (α = 0.05) between the presence of psychological disorders and younger age (Figure 2A and B).
Figure 2. Relationship among comorbidities, fall circumstances and age of death. (A) There are more substance use and psychological disorder histories in decedents who died under 65 years old, while more metabolic and cardiovascular disorders are observed in decedents over 65 years. (B) Younger decedents were more likely to have positive postmortem toxicology and to have fallen from greater heights in fatal fall cases. Falls from height were identified as most commonly from construction sites, home balconies and barns.
Table 1. Summary of patient demographics and fall circumstances
In terms of the circumstances of falls, stairs were the most common location. Falling from stairs was more common than all other falls combined (Table 1), accounting for 64.7% of all cases. There was no difference observed when we compared the individuals older than 65 years with their younger counterparts who fell from stairs. On the other hand, there was a significant association (α = 0.05) between a height of fall and age, with the younger group having 13.5 times higher odds of sustaining a fall from a greater height (>3 m). These high falls (long falls) were identified as most from construction sites, home balconies and barns. In addition, younger individuals were more commonly found to be intoxicated or under the influence at the time of fall, as shown by the positive postmortem toxicology report (OR = 9.63) (Figure 2).
Neuropathology findings
The neuropathological findings were summarized in Table 2. The mean number of craniocerebral or vertebrospinal injuries suffered was 7.69, with 16 being the greatest number of distinct injury types sustained by one individual. Craniocerebral injuries were grouped according to injury type and location, with multiple instances of the same injury type (such as multiple contusions) counted as one injury. Acute subarachnoid hemorrhages were the most documented injury among all cases (72.7%), followed by subdural and intraparenchymal hemorrhages (63.6%). Epidural hemorrhages were seen in 13.6% of the cases. In terms of cerebral contusions, the frontal and temporal lobes were the most common sites (47.7% and 45.5%, respectively), followed by the parietal and occipital lobes. Fractures to the skull base (36.4%) and frontal bone (27.3%) were the most common injuries to the cranium (Table 2 and Figure 3B and C). DAI, a type of severe diffuse traumatic brain injury often obtained in high-energy motor vehicle collisions (MVC), was also among the commonly documented injuries, accounting for 36.4% of all fatal fall cases. Secondary injuries, such as brain swelling and hypoxic-ischemic encephalopathy, were reported in 29.5% and 36.4% of the cases, respectively (Table 2). Vertebrospinal injuries were documented in less detail in the neuropathology reports. Injuries to the cervical spine were more frequent than in other spinal areas (17.8%) (Figure 3A). We then explored the association between case circumstances and injury patterns. Interestingly, our results showed that females are 4.83 times more likely than males to have an intraparenchymal hematoma. We also found low fall (also referred to as short fall for heights under 3 m) to be significantly associated with brain edema and the presence of subfalcine hernia (p = 0.006). Additionally, age was found to be significantly associated with subdural hematoma and frontal contusion, with individuals 65 years of age and older being 1.3 times more likely to suffer subdural hematomas and those under 65 more likely to suffer frontal contusions (OR = 4.0) (Table 3). No significant association was found between other variables and craniocerebral injuries. No associations were found in vertebrospinal injuries.
Figure 3. Neuropathology findings. (A) Distribution of the skull and spinal fractures. Skull base fracture is most common, followed by frontal, occipital and cervical spinal fractures. (B) Distribution of the intracranial injuries. Percentages of cases that had at least one intraparenchymal injury in each of the brain areas are indicated, and the specific injuries are illustrated as pie charts. Subarachnoid hemorrhage is most common, followed by temporal and frontal lobe contusions. (C) The left side injury is slightly more common than the right side.
Table 2. Summary of neuropathology findings. (*There were 45 cases included. Craniocerebral injury was not documented in one of the 45 cases due to severe autolysis.)
Table 3. Results of the Fisher’s exact test for craniocerebral injuries and select variables. Only results with significant association are reported
Discussion
In this study, we provide a comprehensive analysis of fatal falls in Southwestern Ontario over the last 20 years. Our findings reinforce existing evidence regarding the demographic and clinical factors associated with fatal falls while offering new insights into injury patterns and their underlying mechanisms.
There is a significant male predominance (75%), which has also been shown in studies by Caki et al. (82.5%), Kort et al. (86.6%) and Obeid et al. (77.8%). Reference Çakı, Karadayı and Çetin8–Reference Obeid, Bryk and Lee10 This gender disparity has been previously attributed to men’s greater engagement in high-risk activities or occupations. Reference Ding, Liu and Rebernig6 However, no definitive association can be established in our study due to the limited sample size of our cohort. Interestingly, females were significantly more likely to sustain intraparenchymal hematomas (OR = 4.83), a finding that warrants further investigation. Gender-specific anatomical or physiological differences, such as hormonal influences on vascular integrity, could play a role in this disparity. Future research should explore these potential mechanisms to better understand and address gender-related vulnerabilities in fall injuries.
The mean age of 60.3 years and nearly equal distribution of cases above and below 65 years indicate that falls impact both middle-aged and older populations, albeit with differing underlying risk factors. Older adults (≥65 years) were significantly more likely to have pre-existing cardiovascular or metabolic conditions, aligning with previous studies linking these comorbidities to increased fall risk, such as hypertension, Reference Deng, Li and Qin11,Reference Sibley, Voth, Munce, Straus and Jaglal12 arthritis Reference Mikos, Kucharska, Lulek, Kłosiński and Batko13,Reference Guo, Pei, Wei, Zhang, Yan and Han14 and diabetes. Reference Rashedi, Iranpour, Mohseni and Borhaninejad15 Gait disorders, a common consequence of such conditions, likely contribute to imbalance and fall susceptibility. Reference Montero-Odasso, Barbieri and Vitório16 Moreover, medication use, particularly diuretics and polypharmacy, represents a modifiable risk factor for falls and should be a focus of preventive efforts in this population. Reference Abu Bakar, Abdul Kadir, Idris and Mohd Nawi17
Another finding from our cohort is that individuals over the age of 65 are significantly more likely to fall from a low height of under 3 m. Most of these low falls resulted in minor trauma and occurred at home and from standing height, including scenarios such as in the washroom, getting out of bed or at the bottom of stairs. This population’s heightened susceptibility to fatality from ground-level falls is consistent with Spaniolas et al., who reported a threefold increase in mortality among individuals over 70 years following ground-level falls. Reference Spaniolas, Cheng, Gestring, Sangosanya, Stassen and Bankey18 These falls frequently resulted in subdural hematomas, likely due to age-related brain atrophy, which increases the risk of bridging vein rupture during minor trauma. Anticoagulant and antithrombotic medication use may further compound this risk, emphasizing the importance of careful medication management in older adults. Reference Sundblom, Sandberg and Ronne-Engstrom19–Reference Baechli, Nordmann, Bucher and Gratzl21–Reference Sundblom, Sandberg and Ronne-Engstrom19
Younger individuals, on the other hand, exhibited a higher incidence of falls from greater heights (>3 m) and a greater prevalence of intoxication or substance use. Nearly 30% of this group had positive toxicology reports, and they were 13.5 times more likely to fall from significant heights. These findings suggest that risk-taking behaviors and intoxication play a substantial role in the circumstances of fatal falls among younger adults. Although no direct association was found between toxicology and specific brain injuries, the link between substance use and fall risk warrants targeted preventive strategies.
Several studies have shown head injury to be the most common fatal injury in falls from a height. Reference Çakı, Karadayı and Çetin8,Reference Baechli, Nordmann, Bucher and Gratzl21–Reference Hörauf, Nau, Mühlenfeld, Verboket, Marzi and Störmann26 Subarachnoid hemorrhage is the most frequent brain injury in our cohort (72.7%), followed by cerebral contusions, specifically to the frontal and temporal lobes. These two injuries are common in cases of traumatic brain injuries and are often detected together. Reference Hörauf, Nau, Mühlenfeld, Verboket, Marzi and Störmann26 Alexis et al. reported 81.3% and 40.8% of autopsy cases suffered subarachnoid hemorrhage and cerebral contusions, respectively, Reference Alexis, Jagdish, Sukumar, Pandit, Palnivel and Antony27 and other studies showed similar findings. Reference Çakı, Karadayı and Çetin8,Reference Legaspi, Hickey, Pickup and Han25 Skull base fractures and frontal bone fractures were also frequently observed in our cohort. Thierauf et al. observed 291 cases of fatal falls from stairs, ground height and from a greater height and found 47% of the cases sustained skull base fractures, the majority of which had fallen from a height and down the stairs. Reference Thierauf, Preuss, Lignitz and Madea28 Similarly, Preuss et al. also noted that skull base fractures occurred in 75% of cases from stairs. Reference Preuss, Padosch, Dettmeyer, Driever, Lignitz and Madea29 DAI, a severe diffuse brain injury typically seen in high-energy trauma (e.g., MVC), is also prevalent in our cohort (36.4%), suggesting that falls, particularly from greater heights, can mimic the injury patterns of MVC.
Quantitative assessment of hemorrhage size, particularly epidural and intracerebral hemorrhages, may offer additional contextual information. However, such measurements were not consistently or systematically recorded in the autopsy reports spanning the 20-year study period. Consequently, hemorrhage dimensions were not extracted or categorized in this study, as retrospective estimation would be subject to considerable variability. Notably, existing literature suggests that the size or severity of intracranial hemorrhage does not reliably correlate with the height of a fall, particularly in pediatric populations. Reference Ibrahim, Wood, Margulies and Christian30 Likewise, alcohol intoxication has not been shown to significantly modify injury patterns. A retrospective study of fatal falls found no meaningful association between positive blood alcohol concentration and the nature or severity of head injuries. Reference Thierauf, Preuss, Lignitz and Madea28
A notable finding in this study is that a low fall of under 3 m was significantly associated with brain swelling and subfalcine herniation, highlighting the severe consequences of seemingly minor falls. A possible explanation is that in cases of low falls, the impact may not generate sufficient force to cause widespread brain injury but can instead lead to localized injuries such as subdural hematomas or contusions. These localized injuries will subsequently lead to secondary brain swelling and elevate intracranial pressure, particularly if the hematoma is unilateral. This is consistent with our cohort findings, where only two individuals died at the scene and the majority survived for a short period of time following the initial impact. A mean survival time of approximately 50.8 hours was observed. The extended survival interval may have allowed the progression of secondary brain swelling and herniation, providing a potential mechanism for delayed fatality after minor trauma. In contrast, falls from greater heights typically result in more extensive and diffuse brain injuries, such as DAI, and often lead to the patient dying at the scene. This emphasizes the importance of prompt medical evaluation for low falls.
In conclusion, this study characterizes the craniocerebral and vertebrospinal injury patterns and circumstances of fatal falls in Southwestern Ontario from 2000 to 2020. Subarachnoid hemorrhage and intraparenchymal hemorrhage were the most common injuries, with the frontal and temporal lobes and the skull base being the most frequently injured areas. Age was a significant predictor of injury type, with older adults more prone to subdural hematomas and younger individuals to frontal/temporal contusions. These findings underscore the importance of targeted prevention and early intervention strategies to mitigate the impact of falls across diverse populations. Future large-scale studies using statistical modeling are warranted to further explore these findings and develop evidence-based fall prevention measures.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/cjn.2025.10378.
Acknowledgments
The authors would like to thank the Office of the Chief Coroner for Ontario, the London FPU and the Pathology Translational Research Service of LHSC for assistance with case retrieval and autopsy report reviewing. This work was supported by the Pathology Research and Innovation Fund, Department of Pathology and Lab Medicine, LHSC (QZ) and the Undergraduate Summer Research Internship, Western University (RXC).
Author contributions
RXC: development of methodology, data analysis, original draft preparation; AEB: data collection, curation and analysis; WH: data analysis; QZ: project conceptualization, supervision, funding acquisition, manuscript review and editing.
Funding statement
Pathology Research and Innovation Fund, LHSC (QZ); Undergraduate Summer Research Internship, Western University (RXC).
Competing interests
The authors have no competing interests to disclose.
Open access
