Suicide is a desperate conclusion to a psychological problem, one that may have been diagnosed prior to the attempt or may have yet to be found. Several attempts have been made to describe the characteristics of a patient who attempts suicide by jumping from a high place. Gore-Jones et al[11] and others have described the jumping patient as a 31-year-old single person with a diagnosis of a psychotic illness or borderline personality disorder, of which 60% will have had contact with mental health services[20]. Work from our own unit in the United Kingdom shows that in a series of 41 patients a Caucasian male aged 25-30 years is the most common patient to present with this mechanism, a picture also seen in the United States[20,21]. Other work has shown conflicting descriptors of patients. Chia and several European studies show that jumpers were usually female, young and single whereas many other reports show that jumping is more common in males[7,9,13,17,20,22-25].

Other demographics of the patient group have been identified. Whilst eighty percent of patients show features that the act was impulsive, three in every four patients have made previous suicide attempts[11]. Furthermore, Choi et al[16] showed that patients who jumped had a lower final grade of education, showing that severe injury following an attempt was associated with having attained a high school diploma rather than a university degree, perhaps indicating the social-economic situation in which the patient finds themselves at the time of their attempt.

Small et al[26] showed that inpatient suicide was a unique problem. The group showed that youth and social isolation in common with prolonged admission, a history of assaults and previous suicide attempts were particularly associated with jumping. Work in the inpatient population in the United Kingdom showed that jumping was likely to be the most common method of suicide for inpatients, as their access to alternative methods was likely to be restricted by virtue of sectioning or other controls in their environment[14].

The presence of a psychotic illness in patients who jump is agreed on in all series’ that examine the phenomenon. Psychosis may be diagnosed either before or after the attempt, but is typically made in association with the presentation. Estimates of pre-existing psychiatric illness, most commonly schizophrenia (which is more severe than that suffered by those who employ other methods of suicide) range from 10%-97%[3,4,11,19,21,22,24,26-37]. Nielssen et al[29] examined a cohort of survivors of deliberate jumps from height and noted that only 44% of the cohort had a diagnosis of psychotic illness, of which 44% had received no previous treatment, suggesting that they were in their first episode of psychosis. Further exploration of the psychological state of these patients showed that 20% had a robust diagnosis of delusional psychosis, suggesting that the first episode of psychosis is a significant risk factor for suicide attempt[29].

Although schizophrenia is the most often quoted psychotic diagnosis given to this cohort of patients, others diagnoses do feature. Stanford et al[35] found a diagnosis of personality disorder in 49%, depression in 25% and mood disorder in 18%. Kennedy et al[22] showed depression in 27% of patients, and Kontaxakis et al[24] showed affective psychosis-depressive type in several of their 46 patients. In contrast, our own cohort showed an absence of documented psychiatric disease either pre-existing or newly diagnosed during inpatient stay in 44% of patients; however depression features in 23% and psychosis in 13% of cases[21]. Organic illness is found within the cohort more than one might expect given the young age of patients. Wong et al[4] showed that 44% of patients had physical illness, while others list “serious somatic illness” as a comorbidity[24]. Substance abuse also features in patient backgrounds with Bostman quoting 15% of Finnish patients suffering alcoholism at the time of their jump[19,21,23,35].

The frequency of previous suicide attempt ranges from 23% to 75%[11,22,38]. Alongside this, the methods of suicide do seem to change following a failed attempt. Paraschakis found that most patients who failed to commit suicide with self-poisoning, self-inflicted wrist laceration or jumping switched method to jumping on their subsequent attempts[38]. Stanford reports a cohort of 55 patients with an 84% follow-up at a mean of 8 years, four had gone on to successful suicide[35].

Unfortunately, 55% of deliberate falls cause death, a figure influenced by the height of the jump and the physical and psychological comorbidities[16,17,39,40]. Dickinson et al[17] examined a cohort of 117 patients attended by the London’s Helicopter Air Ambulance Service who had fallen from height either accidentally or deliberately, and found that those who jumped rather than fell were more likely to die and that if head and chest injuries occurred, the height fall required for 50% of patients to die was found to be 11 m. If these injuries did not occur, the height fall required for 50% of patients to die increased to 22 m[17]. Further study by Türk et al[39] showed that suicidal falls occurred from a mean height of 23 m compared to 11 m for accidental falls, and that 79% of suicidal jumps were from a height greater than 16 m, whereas Copeland showed most suicidal falls were from 7 storeys (21 m) or higher[23].

Deliberate jumps occur from a range of structures, including residential buildings in 63%-83% of cases, bridges and hotels[4,14,25,32,36,40-44]. The prevention of suicide from bridges by the installation of barriers or other protections reduces the incidence of suicide from that site, an effect seen in both Europe and the United States[8,9,14,25,36,42,43,45].

The surface that the patient lands on influences both injuries and survival. Whilst jumping from buildings is likely to end with a solid surface, jumping from a bridge could lead to landing on water. Gill[46] showed that 77 suicides involving a jump into water showed few external signs of injury, suggesting drowning as the mode of death. Simonsen examined 10 cases of suicide with a water landing, and of the 10, drowning was the cause of death in six cases. Injuries, which were restricted to the thorax and spine, due to the fall caused death in four cases[47].

The spine is the most commonly injured body region following a deliberate fall from height, an association that is magnified if sacral fractures are included into this category[11,21,35,37,50-53]. Each group who have examined these injuries has identified a different spinal level as being most vulnerable. It seems that the most mobile spinal segments are particularly at risk particularly the thoracolumbar junction[28,35]. Wirbel et al[50] showed that of 36 patients with spinal injuries due to jumping, 33 had spinal injuries and Hahn et al[51] showed that 83% of jumpers had fractures of the thoracolumbar spine, usually at the thoracolumbar junction. Li et al[20] showed an incidence of 19% of neck injuries in their series, and Stanford et al[35] showed in his series of 55 patients with spinal cord injury as a result of jumping, 23 patients (42%) had a complete cord injury with C5 and L1 being the most commonly injured levels. Our own series shows that 15/41 patients sustained a thoracolumbar fracture (37%) and 5/41 patients a cervical spine injury[21].

It is reasonable to hypothesise that primary brain injury is responsible for a significant proportion of the 55% of early deaths due to suicide via jumping, supported by Richter et al[28] who showed that half of patients with a head injury fell or jumped from a single storey died[16]. Abel et al[41] showed that 30% of patients seen at their unit sustained craniofacial injuries, a figure in approximate agreement with Richter et al[28] (27%) and our own series (29%)[21]. However, Li et al[20] showing an incidence of head injury of 70% in their series of 124 lethal falls or jumps, suggesting the conclusion that the head injury is often fatal. Head injury is further explored by Dickinson et al[17] who showed that at a jump height of 11 m, 50% of those patients sustaining a head or chest injury die. Slightly contradictory to this, Türk et al[39] showed that patients jumping from heights between 11-25 m, head injuries were less common than when patients jumped from heights outside this range, and that 79% of suicides were from 16 m or higher.

The only reports of pelvic fractures caused by a deliberate jump have been made by Roy-Camille et al[53]. They describe an H-shaped transverse sacral fracture with vertical elements through the foramina. H-shaped fractures are now synonymous with the “Jumpers Fracture” are rare, with only 1.2% of pelvic fractures treated in a European trauma centre within a 9-year period being of this type[52]. They are however significant injuries, requiring surgery in all displaced fractures, especially with vertical shear fractures and associated L5 and S1 neurological damage[52,53].

Teh et al[48] found that their series of 57 jumpers showed a higher rate of pelvic injury in those who jumped from height when compared to fallers, and Richter et al[28] found a 30% incidence of pelvic fracture amongst 39 jumpers with the incidence of pelvic injury increased significantly once the height of the jump was above 7 m[48,51]. Within the cohort in our unit, pelvic fractures were found in 14/41 patients (34%), with jumpers fractures sustained in 4 of these, perhaps refuting a causal link between jumping and jumpers fractures[21,53]. Pelvic injuries due to jumping come with poorer long-term results when compared to other causes of pelvic fracture. It is evident that these patients have a lower health related quality of life score than other pelvic fracture patients, and that patients who are younger at the time of injury fare better when compared to more elderly[54].

The limbs are intuitively the most vulnerable to injury with any significant trauma and in jumping[50]. However, the degree to which they are injured and the pattern in which those injuries are found is not well described. Jumpers appear to sustain significantly more bilateral limb injuries than fallers which tend to be metaphyseal and epiphyseal[28,48,49]. Hahn et al[51] reported a limb fracture incidence to be 45% with calcaneal and ankle fracture to be the most common extremity injuries seen, with a respective incidence of 65% and 27%. Li et al[20] showed a 28% incidence of extremity fracture in jumpers, and Katz et al[37] showed that the lower limbs were the most common skeletal injury. These figures fall short of our own, where we found a 93% incidence of lower limb injury in our series of jump survivors[21]. No studies have yet published regarding the relative incidences of open and closed fractures in jumpers, although one would expect a tendency toward a higher incidence of open fractures in this group, due to the higher energy. Only two studies reporting on upper limb injuries. Our own work has shown an incidence of 8/41 upper limb injuries (most often open fractures), and Hahn et al[51] identified a 25% incidence of upper limb fracture in 39 jumpers[21].

Injuries to the thorax are associated with death in the jumper[17]. Chest injuries occur in up to 66% of jumpers, with abdominal injuries occurring much less frequently, estimated at being present in 6% of cases[20,28,41]. Dickinson details how the presence of a chest injury was associated with a higher risk of death, and that in the absence of chest (and head) injuries reduced the risk[17].

Abdominal injuries are rarely reported with one study reporting an incidence of 48% of abdominal injuries in a series of 139 fatal falls and jumps[20]. This figure varies from other published and raw data both of which give figure of 6% or less[21,28]. It may be that abdominal injuries, similar to cervical spine and head injuries, are often fatal in the very early stages (through massive haemorrhage or visceral trauma) and so could be overrepresented in the fatalities.