Dog-Related Falls Affecting the Lower Extremity Treated at United States Hospital Emergency Departments

By Mathias B. Forrester, BS

Background: Falls are a leading cause of non-fatal injuries in the United States (US). There are a variety of fall risk factors, among which are pets. Tens of thousands of fall injuries associated with dogs are treated at US hospital emergency departments (EDs) each year. The objective of this study is to describe dog-related falls affecting the lower extremity treated at US hospital EDs.

Methods: Dog-related falls affecting the lower extremity during 2000-2023 were identified using the National Electronic Injury Surveillance System (NEISS), a database of consumer product-related injuries treated at a representative sample of US hospital EDs. Cases reported to the NEISS can be used to calculate national injury estimates. The distribution of dog-related falls affecting the lower extremity was determined for patient demographics, injury circumstances, and management.

Results: Of an estimated 1,381,305 total dog-related falls treated at US hospital EDs during 2000-2023, 283,468 (20.5%) affected the lower extremity. The circumstances of the injuries were 50.5% tripped by or tangled with the dog; 13.8% pulled by the dog; 6.0% knocked down by, pushed by, or collided with the dog; and 29.7% other or unknown circumstances. The patients were 73.6% female, and 24.4% were aged 60 years or older. Strain or sprain was reported in 33.1% of the estimated falls, followed by fracture (28.8%), contusion or abrasion (19.1%), and laceration (5.5%). The patient was treated or evaluated and released in 89.5% of the estimated injuries.

Conclusion: Approximately half of dog-related falls affecting the lower extremity were due to being tripped by or tangled with the dogs. Almost three quarters of the patients were female, and almost one-quarter of patients were aged 60 years or older. The most frequently reported injuries were strain or sprain, fracture, and contusion or abrasion. Most patients were treated or evaluated at the ED and released. The information in this study may be useful for healthcare providers and others to identify those individuals most at risk for dog related falls affecting the lower extremity and creating plans to manage and prevent these injuries.

According to the American Veterinary Medical Association (AVMA), by year-end 2016, there were 77 million pet dogs in the United States (US), and 38.4% of US households owned a dog.¹ The 2021-2022 National Pet Owners Survey reported that 69.0 million US households owned a dog.²

Falls are a leading cause of non-fatal injuries in the US. There are a variety of fall risk factors, among which are pets. Tens of thousands of fall injuries associated with dogs are treated at US emergency departments (EDs) each year. The most common circumstances for dog-related falls are tripped over, pushed by, or pulled by the dog. The most common injuries in dog-related falls are fractures, contusions or abrasions, strains or sprains, and lacerations. The injury involves the leg or foot in roughly one-quarter of dog-related falls.³

The objective of this study is to describe dog-related falls affecting the lower extremity treated at US hospital EDs.

Methods

This descriptive study used data from the National Electronic Injury Surveillance System (NEISS), which is available at https://www.cpsc.gov/cgibin/NEISSQuery/home.aspx. The NEISS database has been described in detail in Lower Extremity Review previously.⁴ In brief, the NEISS, operated by the US Consumer Product Safety Commission (CPSC), is a database of consumer product-related injuries collected from a representative sample of approximately 100 US hospital EDs. National estimates are calculated from database records according to the sample weight assigned to each case based on the inverse probability of the hospital being selected for the NEISS sample.^5,6 Data are publicly available and de-identified; thus, the study is exempt from institutional review board approval. The previous study that examined dog-related fall injuries used data from a related database (National Electronic Injury Surveillance System – All Injury Program or NEISS-AIP).³ The NEISS-AIP has been described previously in Lower Extremity Review.⁷

Cases were dog-related falls affecting the lower extremity reported to the NEISS database during 2000-2023. The identification of cases was performed in stages. First, to identify as many dog-related injuries as possible, the NEISS database was searched for all records that included the following letter groups in the Narrative field (a text field that provides a brief summary of the circumstances and type of injury):

“dog,” or “pup,” or “airdale,” or “Akita,” or “Malamute,” or “shep,” or “hound,” or “beagle,” or “terrier,” or “collie,” or “doodle,” or “poodle,” or “boxer,” or “spaniel,” or “mastiff,” or “Corso,” or “chihuahua,” or “corgi,” or “dachshund,” or “dalmation,” or “dober,” or “labrador,” or “retriever,” or “dane,” or “Pyrenees,” or “husky,” or “ setter,” or “Maltese,” or “schnauzer,” or “dalmation,” or “Pomeranian,” or “pug,” or “rottw,” or “rotw,” or “Bernard,” or “shar p,” or “sharpie,” or “shih,” or “tzu,” or “weim,” or “whippet,” or “pitbull,” or “pit bull”

These letter groups were used because the dog might have been identified solely as a puppy or by breed. The letter groups were included in the names of some of the more common dog breeds and are not considered to be comprehensive.

Second, to identify as many falls as possible, this subset of records was searched for all records that included the following letter groups in the Narrative field:

“fall,” or “fell,” or “trip,” or “slip,” or “knoc,” or “nock,” or “down,” or “push,” or “pull,” or “foos,” or “T’D&F,” or “TD&F,” or “S’D&F,” or “SD&F,” or “bump,” or “ran into,” or “run into,” or “stumb,” or “jump,” or “land,” or “crash,” or “collid,” or “collis”

Third, the Narrative fields of the resulting second subset of records were individually reviewed to determine whether they appeared to represent dog-related falls. If so, the circumstance of the injury was assigned to one of the following categories:

1. Tripped by or tangled with the dog – includes if tripped or tangled by a leash, even if the leash was not attached to a dog
2. Pulled by the dog – includes if pulled by a leash, even if the leash was not attached to a dog
3. Knocked down by, pushed by, or collided with the dog – includes if the dog knocked into or pushed something the person was standing or sitting on or in (e.g., ladder, chair, stroller)
4. Other or unknown

Records were included if the Narrative indicated a trip or slip even if a fall was not explicitly stated. Records also were included if the person fell into or onto something other than the ground (e.g., wall, stairs, furniture). Records were excluded if, aside from a leash, the fall was due to a dog-related item (e.g., dog gate, dog cage, dog toy, dog urine).

Finally, those dog-related falls affecting the lower extremity were identified. The NEISS database contains two numeric fields for coding the affected body part (Body_Part and Body_Part_2) and two numeric fields for coding the type of injury or diagnosis (Diagnosis and Diagnosis_2). The Body_Part_2 and Diagnosis_2 fields were added in 2018, although they do not appear to have been used until 2019.⁶ For consistency over the entire study period, the Body_Part and Diagnosis fields alone was examined for the analysis. Only those records where the Body_Part field contained codes for a lower extremity (upper leg, knee, lower leg, ankle, foot, toe) were included in the study.

The variables examined were circumstances of the injury, treatment year and month, location where the incident occurred, patient age and sex, type of injury (diagnosis), affected body part, and disposition. Analyses were performed using Microsoft 365 Access and Excel (Microsoft Corporation, Redmond, Washington, US). National injury estimates were calculated by summing the values in the Weight numeric field in the publicly available NEISS database. The distribution of estimated injuries was calculated for each variable for all injuries and for all of the circumstance categories except for other or unknown because it was a heterogenous category. The CPSC considers an estimate unstable and potentially unreliable when the estimate is <1,200.⁵

Results

Of an estimated 1,381,305 total dog-related falls treated at US hospital EDs during 2000-2023, 283,468 (20.5%) affected the lower extremity. Table 1 shows the distribution of injuries by the circumstances of the injury. For both total injuries and lower extremity injuries, the most common circumstance was tripped by or tangled with the dog followed by pulled by the dog, and then knocked down by, pushed by, or collided with the dog. The proportion of total injuries comprised of lower extremity injuries was highest for being tripped by or tangled with the dog and lowest for being knocked down by, pushed by, or collided with the dog.

Table 2 provides the distribution of dog-related falls affecting the lower extremity by time period and location. The estimated number of injuries increased over the 4 6-year periods of the study for total injuries and injuries due to being pulled by the dog. The estimated number of injuries due to being tripped by or tangled with the dog increased during the first 3 6-year periods then remained relatively flat during the fourth 6-year period. The estimated number of injuries due to being knocked down by, pushed by, or collided with the dog declined from the first 6-year period to the second 6-year period then increased over the next 2 6-year periods. The estimated number of injuries was highest in April-September and lowest in October-March. For those injuries where the location of the injury incident was recorded, the highest estimated number of injuries occurred at home (or mobile home). This was true for all 3 major circumstances of the injury, although the proportion was lower for injuries due to being pulled by the dog. The proportion of injuries that occurred at other public property or street or highway was higher for injuries due to being pulled by the dog than for injuries due to being tripped by or tangled with the dog or being knocked down by, pushed by, or collided with the dog.

Table 3 presents the distribution of dog-related falls affecting the lower extremity by patient demographics. The estimated number of injuries increased with increasing age, with approximately one-quarter occurring in patients aged 60 years or older. The majority of patients were female. This pattern was observed for total lower extremity injuries and for the 3 major circumstances of the injury.

Table 4 shows the distribution of dog-related falls affecting the lower extremity by injury type and disposition. The highest estimated number of total lower extremity injuries involved the knee followed by the ankle, foot, lower leg, upper leg, and toe. This pattern was also generally found in injuries due to being tripped by or tangled with the dog and being pulled by the dog. However, for those lower extremity injuries due to being knocked down by, pushed by, or collided with the dog, the third most affected body part was the lower leg. The most frequently reported type of injury or diagnosis was strain or sprain followed by fracture, contusion or abrasion, and laceration. This pattern was observed for all 3 major circumstances of the injury. Most patients were treated or evaluated in the ED and released.

When the distribution of dog-related falls affecting the lower extremity was examined by most common injury type and 4 patient age groups (Table 5), patients aged 0-5 years had the highest proportion of total estimated injuries that were fractures. Among the 3 older age groups, the proportion of strains and sprains decreased and the proportion of fractures increased with increasing patient age. When the distribution of fractures alone was examined by the 4 patient age groups (Table 5), aside from patients aged 0-5 years, the proportion of total estimated fractures involving the lower leg decreased, involving the foot decreased, and involving the knee increased with increasing the patient age. Patients aged 60 years or older had the highest proportion of total estimated fractures that involved the upper leg, followed by patients aged 0-5 years.

Discussion

This study found that roughly half of estimated dog-related falls affecting the lower extremity were due to being tripped by or tangled with the dog, with 14% due to being pulled by the dog and 6% being knocked down by, pushed by, or collided with the dog. This pattern is similar to a previous study of dog-related falls, which found that the most common circumstance of the injury was fell or tripped over the dog (31.3%) followed by pushed or pulled by dog (21.2%).³

The estimated number of dog-related falls affecting the lower extremity increased over the 24-year period of the study. Other studies have reported increases in dog leash-related injuries, including those affecting the lower extremity, leash-dependent dog walking-related injuries, and fractures in elderly adults while walking leashed dogs.^4,8-10 The increasing trend in these dog-related injuries may partly be due to an increase in the number of pet dogs in the US. According to the AVMA, the rate of households owning dogs in the US increased from 36.1% in 2001 to 38.4% in 2016, and the estimated number of pet dogs increased from 72.1 million in 2006 to 76.8 million in 2016.¹

Dog-related falls affecting the lower extremity were seasonal. The highest estimated number of injuries was reported in July-September, and the lowest estimated number was reported in January-March. This is similar to a study of dog leash-related injuries that observed 27% of the injuries in June-August and 19% in December-February.⁹ The similarity may be due to many dog-related falls resulting from tripping over or tangling with the dog leash or being pulled by a leashed dog. People might be more likely to walk dogs in warmer weather.

The majority of dog-related falls affecting the lower extremity occurred at the patient’s home. Of particular note, the proportion of falls that occurred at home was lower and falls that occurred at other public property and street or highway was higher when being pulled by the dog than when being tripped by or tangled with the dog or knocked down by, pushed by, or collided with the dog. Falls due to being pulled by the dog may be more likely to occur when walking a leashed dog than falls due to the other 2 circumstances; injuries while walking a dog may be more likely to occur away from home.

Almost three-fourths of the patients involved in dog-related falls affecting the lower extremity were female. In addition, the estimated number of dog-related falls affecting the lower extremity increased with patient age, with almost one-fourth of patients aged 60 years or older. The prior study of dog- and cat-related falls reported 69% of the patients to be female and 24% aged 65 years or older,³ and other studies of dog-leash related injuries likewise found most patients to be female and 15-32% aged 60 years or older.^4,8-10 Females and older adults may be at particular risk of dog-related falls. Healthcare providers and others may want to take this into consideration when trying to prevent dog-related falls.

The most common injury type (diagnosis) in dog-related falls affecting the lower extremity was strain or sprain followed by fracture, contusion or abrasion, and laceration. Most such injuries might not be expected to require extensive hospital intervention. This is consistent with the observation that almost 90% of the patients were treated or evaluated and released from the hospital ED.

The pattern of injuries (diagnoses) varied with the patient’s age. Patients aged 0-5 years had the highest proportion of total estimated injuries that were fractures. Among older patients, the proportion of strains and sprains decreased and the proportion of fractures increased with increasing patient age. Moreover, the distribution of fractures by affected body part varied with the patient’s age. For patients aged 6 years or older, the proportion of total estimated fractures involving the lower leg decreased, involving the foot decreased, and involving the knee increased with increasing the patient age. Patients aged 60 years or older had the highest proportion of total estimated fractures that involved the upper leg, followed by patients aged 0-5 years. Healthcare providers might want to consider age differences in the pattern of injuries when planning education and prevention strategies for specific age groups.

There are general strategies for preventing dog-related falls. Healthcare providers and other organizations can promote public awareness about the circumstances or situations that can lead to falls, such as walking and chasing dogs. In addition, the public can be informed about the AVMA recommendations for obedience training for dogs to minimize hazardous behaviors such as pulling and pushing.³

This study has limitations. The source of the data was NEISS. NEISS primarily collects data on consumer product-related injuries treated at US hospital EDs. Thus, only those dog-related falls affecting the lower extremity considered to be related to consumer products would be included in the study, a subset of all dog-related falls affecting the lower extremity. However, the NEISS data are publicly available and represent all hospitals and all regions of the US. Also, the NEISS database has been used for previous dog-related injury investigations.^4,9-12 Moreover, the general pattern of dog-related falls affecting the lower extremity observed in the present study was similar to the pattern of dog- and cat-related falls found in a study that used data from the NEISS-AIP database, which is not limited to consumer product-related injuries.³

As another limitation, dog-related falls involving injuries were identified first by searching the record Narrative field for specific letter groups. If the Narrative field for a record of a dog-related fall injury did not include these letter groups, it would not be included in the study. In addition, the further selection of records to be included in the study and the sorting of them into the various circumstances of injury was performed by a single person and was based on the Narrative field, which contains a limited amount of information that at times was difficult to read and interpret. Errors in the selection and classification of records may have resulted in records being included or excluded erroneously or misclassified. Finally, only those dog-related falls affecting the lower extremity treated at hospital EDs were included in the study. The number of such injuries not seen at EDs is unknown.

In conclusion, the estimated number of dog-related falls affecting the lower extremity and treated at US hospital EDs has increased over the last 24 years. Approximately half of these injuries were due to being tripped by or tangled with the dogs. Almost three quarters of the patients were female, and almost one-quarter of patients were aged 60 years or older. The most frequently reported injuries were strain or sprain, fracture, and contusion or abrasion. Most patients were treated or evaluated at the ED and released. The information in this study may be useful for healthcare providers and others to identify those individuals most at risk for such injuries and creating plans to manage and prevent these injuries.

Mathias B. Forrester, BS, is an independent researcher in Austin, Texas. Now retired, he previously performed public health research for various university and government programs for 38 years.