Falls from height are the fourth leading cause of unintentional injury in the United States.1 Variable patterns of injury may be observed owing to the height of the fall, the nature of the impact surface, and body position.2 In addition, differences in injury pattern between intentional and accidental fallers have also been reported.3,4 However, the height of the United States–Mexico border fence is relatively uniform and the impact surface is somewhat homogeneous, so these individuals intend to land safely on his or her feet. Injured individuals are often transferred to trauma centers in the United States for evaluation and management and are largely or completely unable to assume the cost of their hospitalizations.
Each year, an untold number of unauthorized immigrants attempt to cross the United States–Mexico border. This border extends approximately 1980 miles from San Diego, California, to Brownsville, Texas.5 The US Border Patrol Tucson Sector is responsible for patrolling the 262 miles of international border between the Arizona–New Mexico state line and the Yuma county line. Of that 262 miles, 210 miles has some form of a barrier in place with 71 miles consisting of pedestrian style fencing (Vivien McLoughlin, email communication, January 2017). Unauthorized individuals frequently attempt to scale these fences to avoid detection by US Border Patrol agents and often do so without the use any safety apparatus. A recent report suggests that the incidence of jumping and falling injuries at the border fence are increasing, posing a public health dilemma.6 A thorough understanding of these patterns of injury may help emergency department personnel more accurately diagnose and treat this population.
The purpose of this study was to assess patterns of injury, sex disparity, and length of stay in individuals presenting with musculoskeletal injuries sustained in a fall from the international border fence. To the authors' knowledge, no previous studies addressing patterns of musculoskeletal injury in this population exist.
Materials and Methods
After obtaining institutional review board approval, The University of Arizona Medical Center Trauma Registry was queried to identify all patients who sustained falls from the United States– Mexico border fence between February 1, 2004, and February 1, 2010. Unauthorized border crossers who presented after falls while trekking through harsh terrain or were admitted for dehydration or other exposure-related diagnoses were excluded. Injury radiographs and medical records were reviewed. The following data were extracted and analyzed: age, sex, fracture location, type, laterality, presence of open fracture, length of stay, and number of operative procedures.
Fractures were classified by the involved bone or pattern of fracture to include the following: metatarsal, tarsal, calcaneus, talus, rotational ankle injuries, pilon, diaphyseal tibia and/or fibula, plateau, patella, distal femur, diaphyseal femur, proximal femur, acetabulum, pelvic ring, clavicle, scapula, proximal humerus, diaphyseal humerus, distal humerus, proximal radius and/or ulna, diaphyseal radius and/or ulna, distal radius and/or ulna, carpus, metacarpal, phalangeal, and rib. Long bones were divided into thirds such that proximal and distal groupings included fractures of the both epiphysis and metaphysis. Dislocations were recorded by involved joint. Vertebral fractures were classified into thoracolumbar spine (T10-L5), thoracic spine (T1–T9), and cervical spine (C1–C7).
Average number of fractures between sexes was compared using Student's t test. Patients were further divided by sex and into 5 age groups: 10 to 20, 21 to 30, 31 to 40, 41 to 50, and 51 to 60 years. The average number of fractures and frequency of open fracture in each age group were then compared with the total population and between groups with a Student's t test (adjusted for multiple hypotheses). The association between fractures types was assessed using a Monte Carlo simulation, where the number of fractures per patient was controlled, whereas the type of fracture sustained was randomized. The significance of the association was assessed with a 2-tailed test that was adjusted for the multiple hypotheses tested.
Admission was defined as a hospital stay requiring a physician admission or observation order. Where applicable, operative reports were reviewed to determine the number of procedures performed to treat each individual requiring formal, operative intervention. Nonbillable procedures and reductions performed by resident orthopedic surgeons in the emergency department were excluded. All statistical analyses were performed using the Anaconda Python distribution (Continuum Analytics, Austin, Texas).
During the study period, 174 individuals who had sustained orthopedic injuries as a result of falls from the United States–Mexico border fence were identified. The study population contained 93 (53%) women and 81 (47%) men with an average age of 31.5 years (range, 11–56 years). The age group distributions were 14 (8%) patients in the 10 to 20 years group, 65 (37.4%) in the 20 to 30 years group, 57 (32.8%) in the 30 to 40 years group, 31 (17.8%) in the 40 to 50 years group, and 7 (4%) in the 50 to 60 years group.
The average±standard error (SE) number of fractures observed in the study population at-large was 1.58±0.07 fractures. Men sustained an average±SE of 1.77±0.12 fractures per individual compared with 1.43±0.07 for women; this difference was statistically significant (P=.015). However, there were no statistically significant differences in average number of fractures between age groups. Overall, fractures of the thoracolumbar spine occurred most commonly, followed by rotational ankle and tibial pilon fractures. In total, 50 open fractures were observed, occurring in 44 patients (25.2% of individuals; 18.4% of all fractures). There were no significant associations between sex and frequency of open fracture. The Table lists the frequency of fracture and percentage of open fracture by site.
Frequency of Fracture and Percentage of Open Fracture by Type
Several associated fracture types were observed, including fractures of the thoracolumbar spine and calcaneus (z=15.82) and thoracolumbar spine and tibial pilon (z=21.08). Fractures of the pelvic ring were associated with numerous fracture types, including the thoracolumbar spine (z=13.3), humerus (z=20.5), and radius (z=13.3). In addition, fractures of the diaphyseal tibia and fibula were associated with fractures of the calcaneus (z=6.81).
Of the patients presenting to the trauma center, 144 (82.8%) required admission for management of their injuries. On average, length of stay for admitted patients was 3.5 days (range, 1–15 days) and underwent 0.75 (range, 0–3) surgical procedures. There was no mortality observed in this series.
The severity of an injury in a fall can depend on the height, nature of the impact surface, and body orientation.7 Reproducible patterns of injury may be observed between individuals who sustain vertical deceleration injuries from similar height and circumstance. Previous authors have described patterns of injuries that occur in those who jump intentionally with suicidal intent and those who fall accidentally.3,4 In the current series, the authors examine several vertical deceleration injuries in individuals who intend to land safely on their feet and continue their journey. To the authors' knowledge, no such series exists to date.
Overall, the injury patterns the authors identified in falls from the international border fence are consistent with patterns found in prior studies of vertical deceleration injuries. In 1948, Ciccone and Richman8 reported the anatomical distribution of 2709 fractures due to accidents related to parachute jumps. Although fractures of spine occurred much less frequently in their series, the proportion of lower extremity injuries is relatively similar. In addition, fractures of the calcaneus, which are typically thought to occur as a result of axial loading on the heels when landing on one's feet, occurred in 16% of individuals in this study group. This is similar to rates cited in the literature.3,4,8 In addition, concurrent fractures of the calcaneus and spine are reported to occur in as many 10% to 75% of patients who fall from height.4,9
The current authors also found a statistically significant association between fractures of the thoracolumbar spine and calcaneus. Surprisingly, they also observed a significant association between fractures of the thoracolumbar spine and tibial pilon fractures, which supports the hypothesis of loading through the appendicular skeleton with subsequent transmission to the spinal column. Fractures of the pelvis are also known to occur during high-energy, vertical deceleration events.9,10 This may explain the association of pelvic fractures with other fracture types observed in this series. The observed association between fractures of the diaphyseal tibia and calcaneus was somewhat surprising and, to the authors' knowledge, has not been described elsewhere.
The average number of fractures sustained by male fallers in the current series was significantly higher than that observed in female fallers. This gender disparity is somewhat unexpected but may represent an important difference in border crossing behavior between genders. Sapkota et al5 have suggested that young, Mexican men are more likely to use dangerous routes and crossing points when attempting to gain unauthorized access to the United States. The current authors believe this variation may be explained by attempts to cross in urban centers where the border fence is highest.
Unauthorized immigrant populations entering the United States receive considerable attention in the lay press. The topic of border security and financial costs of unauthorized immigration are frequently debated among policy experts and the efficacy of federal anti-immigration measures remains questionable.11 This study illustrates the humanitarian and financial crisis faced by trauma centers in the southwestern United States charged with caring for unauthorized immigrants who fall from the international border fence. More than 80% of the individuals represented in this study required hospital admission. The true cost of these admissions is difficult to quantify; however, the overwhelming majority of these individuals are largely or completely unable to pay for their stays and hospitals have no means of recovering the associated expenses once repatriation has occurred.
The current study has several limitations. First, its retrospective nature is associated with inherent inaccuracies of data collection. The authors also assume that individuals crossing the border fence are doing so with intent to land safely on the opposite side. However, human smugglers in the authors' state are known to push, kick, or intentionally drop fence crossers to flee approaching law enforcement agents. Unfortunately, there is no method for separating these individuals from the remainder of the study group. In addition, unauthorized border crossers are transferred to several medical facilities in the authors' area, only the most severely injured present to our level I trauma center. Finally, the current sample size of 172 patients may be underpowered to detect significant differences in injury patterns.
Falls from the United States–Mexico border fence are an important, and often underappreciated, cause of musculoskeletal injury in the southwestern United States. Patterns of injury observed in this population are similar to those reported for other populations of vertical deceleration injury. A significant proportion of these individuals require hospital admission but are unable to pay for services rendered, leaving hospitals and physicians uncompensated. Prevention strategies targeting young, Mexican men and women that discourage crossing the international border fence are needed to decrease incidence of border fence related injury and decrease strain on health care systems.
- Rockett IR, Regier MD, Kapusta ND, et al. Leading causes of unintentional and in tentional injury mortality: United States, 2000–2009. Am J Public Health. 2012; 102(11):e84–e92. doi:10.2105/AJPH.2012.300960 [CrossRef]
- Warner KG, Demling RH. The pathophysiology of free-fall injury. Ann Emerg Med. 1986; 15(9):1088–1093. doi:10.1016/S0196-0644(86)80134-2 [CrossRef]
- Richter D, Hahn MP, Ostermann PA, Ekkernkamp A, Muhr G. Vertical deceleration injuries: a comparative study of the injury patterns of 101 patients after accidental and intentional high falls. Injury. 1996; 27(9):655–659. doi:10.1016/S0020-1383(96)00083-6 [CrossRef]
- Teh J, Firth M, Sharma A, Wilson A, Reznek R, Chan O. Jumpers and fallers: a comparison of the distribution of skeletal injury. Clin Radiol. 2003; 58(6):482–486. doi:10.1016/S0009-9260(03)00064-3 [CrossRef]
- Sapkota S, Kohl HW III, Gilchrist J, et al. Unauthorized border crossings and migrant deaths: Arizona, New Mexico, and El Paso, Texas, 2002–2003. Am J Public Health. 2006; 96(7):1282–1287. doi:10.2105/AJPH.2005.075168 [CrossRef]
- Kelada A, Hill LL, Lindsay S, Slymen D, Fortlage D, Coimbra R. The U.S.-Mexico border: a time-trend analysis of border-crossing injuries. Am J Prev Med. 2010; 38(5):548–550. doi:10.1016/j.amepre.2010.01.028 [CrossRef]
- Steedman DJ. Severity of free-fall injury. Injury. 1989; 20(5):259–261. doi:10.1016/0020-1383(89)90162-9 [CrossRef]
- Ciccone B, Richman RM. The mechanism of injury and the distribution of three thousand fractures and dislocations caused by parachute jumping. J Bone Joint Surg Am. 1948; 30(1):77–97. doi:10.2106/00004623-194830010-00010 [CrossRef]
- Lowenstein SR, Yaron M, Carrero R, Devereux D, Jacobs LM. Vertical trauma: injuries to patients who fall and land on their feet. Ann Emerg Med. 1989; 18(2):161–165. doi:10.1016/S0196-0644(89)80107-6 [CrossRef]
- Gupta SM, Chandra J, Dogra TD. Blunt force of lesions related to the heights of a fall. Am J Forensic Med Pathol. 1982; 3(1):35–43. doi:10.1097/00000433-198203000-00008 [CrossRef]
- Amuedo-Dorantes C, Puttitanun T, Martinez-Donate AP. How do tougher immigration measures affect unauthorized immigrants?Demography. 2013; 50(3):1067–1091. doi:10.1007/s13524-013-0200-x [CrossRef]
Frequency of Fracture and Percentage of Open Fracture by Type
|Site||No. of Occurrences||No. of Open Fractures||Open Fractures by Site|