Article Text


Alcohol related falls: an interesting pattern of injuries
  1. J J E Johnston,
  2. S J McGovern
  1. Accident and Emergency Department, The Ulster Hospital Dundonald, Northern Ireland
  1. Correspondence to:
 Mr J E Johnston
 Accident and Emergency Department, Newcastle General Hospital, Westgate Road, Newcastle upon Tyne NE4 6BE, UK;


Objective: To discover if there is a significant difference in the pattern and severity of injury sustained during falls in patients who have consumed alcohol and those who have not. To determine how pattern and severity of injury correlates with blood alcohol concentration.

Method: A prospective quasi-randomised controlled study between November 2001 and July 2002. All healthy adults between 16 and 60 years who had fallen from standing height were included. A systematic history and examination permitted calculation of injury severity scores as per abbreviated injury scale update 1998. Blood alcohol concentrations were obtained from intoxicated patients with consent.

Results: 351 healthy adult patients were included in the study, there were 238 in the no alcohol group, 113 had consumed alcohol and blood alcohol intake were obtained for 47. The alcohol group had a higher incidence of head injuries (46 (48%) versus 22 (9%)) with a lower incidence of limb injuries (39 (39%) versus 183 (76%)) than the no alcohol group. There was a significant difference in the pattern of injury between the alcohol and no alcohol groups (χ2, p<0.001) and there was a significant difference in the injury severity scores (p<0.001, Z = −2.5). In the alcohol group severity and pattern correlated with alcohol concentration at the time of injury. Patients with an alcohol concentration<2 g/l had mostly soft tissue limb injuries (58%), 2–2.5 mostly significant limb fractures (55%), and >2.5 mostly significant head injuries (90%).

Conclusions: Alcohol related falls are more often associated with severe craniofacial injury. The severity of both limb and head injury is greater and correlates directly with blood alcohol concentration.

Statistics from

Alcohol is the cause of many accidental falls presenting to the emergency department. There has been little research on this group, which represent a substantial proportion of the total attendance. Review of the literature demonstrated a Finnish study by Honkanen1 and an American study by Hingson2 investigating alcohol related falls although there has been more research on the effect of alcohol on general brain injury.

This study compares the pattern and severity of injury in those that have taken alcohol with those who have not. The study compares pattern and severity of injury with blood alcohol concentration within the alcohol group.


This study was designed to provide prospective data on falls in a healthy adult population presenting to The Ulster Hospital, a large district general hospital in Dundonald, Northern Ireland, with a total catchment population of 300 000, about 65 000 new attendances per annum in a mixed urban/rural environment.

The study period was for eight months from the 1 November 2001 to 30 June 2002. During this period all patients attending the department between the ages of 16 and 60 years with falls, simple slips or trips were identified at triage when JJEJ was on duty. Only patients who had fallen from a standing position to the ground were included in the study. Patients falling from a height, being pushed, not actually making contact with the ground with a body part higher than the ankle, and falls from horses, motorbikes, bicycles, sports, etc, were excluded from the study. All the patients were assessed by the same doctor (JJEJ), whose history and examination was standardised, to avoid intraobserver error.

Special regard was given to several details in the history, which included the following points.

  • Exact mechanism of injury

  • Exact time of the incident

  • Alcohol consumption amount and time

  • Degree of intoxication

  • Details of all injuries

Special regard in the examination was given to Glasgow coma scale, a precise anatomical description of injuries, results of investigations and treatment were recorded. The purpose of the study was explained and consent obtained from patients who appeared on examination to have ingested alcohol recently and a blood alcohol concentration was obtained if they were agreeable. The blood alcohol concentration at the time of injury was estimated using the time of injury, the time the blood sample was taken, and the blood alcohol concentration; 1.8 mg/l was added for each hour from the time of injury to the time the sample was taken.

Injury severity score (ISS) was calculated using the abbreviated injury scale 1998 update booklet.3 The injuries for each patient were coded individually; abbreviated injury scales were obtained and used to calculate the injury severity score.

The χ2 test was used to compare the distribution of injury in the alcohol and no alcohol groups, the Mann-Whitney U test was used to compare the ISS, with a p value of <0.05 considered to be significant.


During the study period 113 patients who had fallen had consumed alcohol (A), there were 238 patients in the no alcohol (NA) group.

In the (A) group head injuries were the commonest injuries, there were 46 (48%) patients of whom 33 (72%) had wounds, which required sutures or staples and one required internal fixation for a displaced maxillary fracture. The most severe trunk and limb injuries were also in this group with 11 (10%) of these patients requiring internal fixation for their injuries. The group included a patient with both a dislocated elbow with a comminuted radial fracture, a radial head fracture, a thoracic vertebral crush fracture, three patients with significantly displaced radial fractures, four patients with severely displaced ankle fractures, and two with displaced/comminuted finger fractures. In this group many of the patients came to the department the next day after having recovered from the effects of the alcohol. Four patients refused to give consent to phlebotomy because of needle phobia but we were able to record blood alcohol concentration for 47 patients who had evidence of recent alcohol consumption at the time of assessment.

There was a significant difference between the ISS of the (A) and (NA) groups with p<0.001 and Z score –2.5.

Figures 1 and 2 and tables 1 and 2 show:

Table 1

Correlation of blood alcohol concentration (g/l) and pattern of injury

Table 2

Pattern of injuries (number of injuries and as a percentage of total)

Figure 1

Correlation of blood alcohol concentration and injury severity score (ISS).

Figure 2

Injury type and blood alcohol concentration.

  • <2 g/l injury pattern is mostly soft tissue limb injury.

  • 2–2.5 mostly severe limb fractures requiring admission with internal fixation with an increasing number of more severe head injuries.

  • >2.5 g/l mostly severe head injuries.

Alcohol diffusely depresses central nervous system activity, effect varies from person to person depending on metabolism but a useful guide to levels is given below.

  • 300 mg/l impaired motor skills

  • 800 mg/l legal limit for driving

  • 5 g/l gross motor impairment

  • 2 g/l amnesia

  • 2 g/l coma

In the (NA) group the pattern of injuries was different with a total of 23 patients with head injury of which only eight had wounds, which required sutures or staples. The commonest injuries in this group were ankle/foot sprains 67, soft tissue wrist sprains 34, soft tissue knee injuries 29, musculoskeletal back injury 19, and musculoskeletal chest wall injuries 19. There were only 13 patients in this group who had sustained fractures or dislocations. The fractures were comparatively simple only two required internal fixations. The most common mechanism of injury was a slip on stairs or steps, accounting for 59 falls in the (NA) group and 18 falls in the (A) group.


This study shows different pattern and severity of injury in alcohol related falls. In alcohol related falls there was a greater incidence of craniofacial injury and a greater severity of injury. These differences can be accounted for by the inhibition of protective reflexes. In the intoxicated patient, the inability to put the outstretched hand to break the fall resulted in a lower incidence of limb injury and a greater force being transmitted to the head when it strikes the ground.

All the patients were seen by the same doctor to ensure assessments would be systematic and comparable although this has limited the size of the study. Patients were only recruited to the study when this doctor was on duty, the sample would have been larger if several doctors had been involved in the assessments. The sample was quasi-random as the assessing doctor was working shifts on a rolling rota. In this department nurse practitioners working mostly on daytime shifts see many patients with simple limb injuries as a result of falls so these are likely to be under-represented. The study depended on the diligence of the triage nurse to identify and direct the appropriate patients. The average person metabolises alcohol at a rate of 180 mg/l/h so 180 mg was added to the alcohol level for every hour of presentation after the fall.4–6 This is an estimate, as metabolism will vary within the population. We used serum alcohol concentrations to avoid sampling error but near patient testing with breath alcohol concentration may be easier and quicker with reliable results.

Injuries were not coded unless the appropriate information was available in the notes. The injury severity scores tended to underscore head injuries. Abbreviated injury scale guidelines state that there must be a clear history of loss of consciousness with evidence that it is directly attributable to brain injury rather than toxins, for example, alcohol. Our study emphasises that the injury is greater because of greater force involved as quantified by physical examination and radiological examination of injuries.

There has been research on the effect of alcohol on brain injury. Tate et al demonstrated that blood alcohol concentration was predictive of poorer function with reduced verbal memory over time and poorer visual and spatial functioning.7 Bombardier et al also showed neuropsychological impairments for up to one to two months after the traumatic brain injury, which directly correlated with blood alcohol concentration.8

Several animal studies have attempted to show that alcohol has long term neural protective or adverse effect for any given force applied to the brain. Studies by Kelly et al9 and Janis et al10 both on rats concluded that low to moderate alcohol levels at the time of injury provides a neural protective effect.

Kelly et al demonstrated that this effect was lost at high blood alcohol concentration, this may have been attributable to cardiovascular or respiratory depression. Janis et al demonstrated less cerebral damage in the intoxicated group.10

Other animal studies by Biros et al demonstrated no neural protective effect.11 A study by Zink et al demonstrated an adverse effect on both traumatic brain injury and haemorrhagic shock, cerebral tissue perfusion was shown to be worsened.12

Cooke et al used clinical parameters and neurological scores at presentation and at one hour in a group of intoxicated patients but were unable to predict which had intracranial injuries as evidenced by computed tomography.13

The literature is inconclusive on the neural protective or adverse effect of alcohol on the brain function and further research is required. Further research to find a correlation of blood alcohol with mechanism of injury as a predictor of intracranial injury, using magnetic resonance or computed tomography would be worthwhile. The intoxicated head injured patient has been classified at moderate risk of intracranial injury in current guidelines that indicate skull radiological investigation. As with all head injuries there has been a change of emphasis to investigate the underlying brain injury rather than the bony injury and the trend in United Kingdom with an increasing use of computed tomography rather than skull radiography is reflected in the new NICE guidelines.

In guidelines published last year by Servadei et al, the Neurotraumatology Committee of The World Federation of Neurosurgical Societies added patients with drug or alcohol consumption to the high risk group, independent of the clinical presentation and recommended computed tomographic investigation.

A Finnish study by Honkanen et al studying impact of alcohol intoxication in non-fatal trauma showed positive correlation between the severity of injury with the alcohol intoxication for car occupants and those injured in falls from stairs, but negatively in unspecific falls or falls at the same level.15 An earlier study by Honkanen1 studying a similar age group had shown that alcohol had been involved in 37% of accidents, falling was the commonest cause in this group. Some 47% in this group had head injuries, only 25% upper extremity injuries, tibia and ankle fractures were often associated with alcohol intake. Studies have shown that at blood alcohol concentrations over 1 g/l results in a significant swaying, decreased attention, visual acuity, and adaptation to brightness and glare.2

The commonest mechanism of injury was the simple slip or trip on stairs and steps. It has been shown that more than 10% of the 4000 deaths in the UK from home accidents resulted from falls on stairs.16 This could be reduced by addition of handrails to the side of the stairs. Of the national UK estimate (1999) of approximately 394 713 falls attending the emergency department in the 15–64 years age groups about 8% are attributable to stairs.

Emergency medicine physicians should be careful in the assessment of the intoxicated head injured patient. Subsequent injury is likely to be much greater than expected for a given mechanism and correlates with blood alcohol concentration; this is especially important as intoxicated head injured patients who are very difficult to assess. Guidelines both of investigation and subsequent management are changing, with more respect being given to this. Our study demonstrates that a blood alcohol of greater than 250 mg/% at the time of injury indicates a high force impact.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.