Do Accidents Happen?

Posted by Jeremy Windsor on Dec 27, 2019

Do you ever find yourself using the word "accident" in your day to day practise? Do accidents ever really happen? Just the other day I heard myself muttering it and was reminded of the article written by Ronald Davis and Barry Pless for the BMJ back in 2001. It's worth taking a few minutes to read and perhaps you'll change the way you use the word!

For many years safety officials and public health authorities have discouraged use of the word “accident” when it refers to injuries or the events that produce them. An accident is often understood to be unpredictable—a chance occurrence or an “act of God”—and therefore unavoidable. However, most injuries and their precipitating events are predictable and preventable (1–3). That is why the BMJ has decided to ban the word accident. 

In an editorial in the BMJ in 1993 Evans explained why “motor vehicle crash” is an appropriate expression but “motor vehicle accident” is not: “The word crash indicates in a simple factual way what is observed, while accident seems to suggest in addition a general explanation of why it occurred without any evidence to support such an explanation (4). Evans also argued that “accident” is inappropriate in reference to medical errors (as in medical accidents) and that “its use in medical settings continues to mislead" (4). 

Eight years later “accident” continues to be misused in medical circles—and on the pages of the BMJ. An online search for “accident” in the BMJ for the period January 1996 to December 2000 indicated that it has been used in the title or abstract of 101 articles and anywhere within 1559 articles. Some uses of the word may be appropriate or unavoidable — for example, in reference to accident and emergency departments (which should be renamed). On the other hand, many of these uses contravene the terminology recommended by safety officials, public health authorities, and Evans in his BMJ editorial. The following are examples from the titles of papers published in the BMJ during the past few years: road traffic accidents (5), playground accidents (6), home accidents (7), aviation accidents(8), accidental drug overdoses (9), accidental carbon monoxide poisoning (10) and medical accidents (11). 

As a leading communicator in medicine, the BMJ needs to establish or follow standards in language. Therefore we are banning the inappropriate use of “accident” in our pages. The BMJ may be the first major medical journal to do so, but we hope we are not the last. The BMJ Publishing Group will encourage all its journals to follow suit. We are pleased that the Journal of Accident and Emergency Medicine, which is co-owned by the BMJ Publishing Group and the British Association for Accident and Emergency Medicine, has just changed its name to Emergency Medicine Journal. Perhaps the rest of the emergency medicine establishment in the United Kingdom will jettison“accidents,” as have its counterparts in the United States, Canada, and many other countries (12).

For more details take a look at this!

Implementation of the ban will not be draconian, and editors will have the discretion to decide whether use of the term is inappropriate or misleading. “Accident” is used in the Ninth Revision of the International Classification of Diseases (ICD 9), so we will allow it to be used when referring to specific ICD 9 terminology. In addition, we acknowledge that some injury producing events may seem to be attributable to bad luck or acts of God and thus not preventable. These include earthquakes, being struck by lightning, avalanches, storms at sea, and other natural disasters. Even in these instances, however, there is some disagreement. To the extent that these events are predictable, preventive steps can be taken by avoiding dangerous places at times of risk. With modern technology it is often possible to predict where or when these events will occur. 

As Evans noted, some may see this as nothing more than a “pedantic quibble" (4). Girasek worries that statements like “injuries aren't accidents” may actually cause harm by engendering victim blaming and decreasing the support afforded “accident survivors" (13). Nevertheless, we believe that correct and consistent terminology will help improve understanding that injuries of all kinds—in homes, schools and workplaces, vehicles, and medical settings—are usually preventable. Such awareness, coupled with efforts to implement prevention strategies, will help reduce the incidence and severity of injuries. 

Which words will serve us in place of accidents? The BMJ 's linguistic transgressions cited above can be replaced by motor vehicle crashes or collisions, playground injuries, home injuries, aviation crashes, unintentional drug overdoses and carbon monoxide poisoning, and medical errors. Further guidance on how to speak clearly about injuries comes from the US Center for Disease Control and Prevention (CDC), which has developed a framework for categorising the circumstances of an injury or poisoning along two dimensions. One dimension is the intent of injury or manner of death: unintentional, intentionally self inflicted (suicide if fatal), intentionally inflicted by another (assault or homicide), or intent undetermined. The other dimension—the mechanism of injury or cause of death — characterises the external agent or particular activity that caused the injury (for example, motor vehicle, fall, fire/burn, firearm, poisoning, and suffocation). The framework has been presented in the form of a matrix, depicted as mechanism by intent of injury—for example, poisoning related (mechanism) suicide (intent) (14,15). Although the framework and matrix were developed primarily to standardise the grouping of ICD 9 E codes (external causes of injury codes) for tabulating injury data, they can help guide the use of terminology in the BMJ and other publications. 

We are struggling to find a generic term that covers the range of events in which people may be injured, killed, or “lucky” to escape. For example, what do we use in place of “accident survivors”? Mishaps, misadventures, calamities, events, and incidents have their own shortcomings, and the English language may simply fail us here. Perhaps we should coin a word to refer collectively to the incidents that may produce injury (injidents?).We invite suggestions from readers. 

Purging a common term from our lexicon will not be easy. “Accident” remains entrenched in lay and medical discourse and will no doubt continue to appear in manuscripts submitted to the BMJ. We are asking our editors to be vigilant in detecting and rejecting inappropriate use of the “A” word, and we trust that our readers will keep us on our toes by alerting us to instances when “accidents” slip through.

1 Doege TC. An injury is no accident. N Engl J Med 1978;298:509-10.

2 Loimer H, Driur M, Guarnieri M. Accidents and acts of God: a history of the terms. Am J Public Health 1996;86:101-7.

3 Doege TC. Eschewing accidents. JAMA 1999;282:427.

4 Evans L. Medical accidents: no such thing? BMJ 1993;307:1438-9.

5 Stallard P, Velleman R, Baldwin S. Prospective study of post-traumatic stress disorder in children involved in road traffic accidents. BMJ 1998;317:1619-23.

6 This week in the BMJ: Surveillance of playground accidents can lead to their reduction. BMJ 1999;318 (12 June).

7 Waldon G. Accidents at home are no more likely in deprived areas. BMJ


8 Cullen SA, Drysdale HC, Mayes RW. Role of medical factors in 1000 fatal aviation accidents: case note study. BMJ 1997;314:1592.

9 Satchithananda DK, Stone DL, Chauhan A, Richie AJ. Unrecognised accidental overdose with diltiazem. BMJ 2000;321:160-1.

10 Pullinger R. Lesson of the week: Something in the air: survival after dramatic, unsuspected case of accidental carbon monoxide poisoning. BMJ 1996;312:897-8.

11 Dyer C. No fault compensation for medical accidents under consideration. BMJ 1997;315:1111-6.

12 Reid C, Chan L. Emergency medicine terminology in the United Kingdom—time to follow the trend? Emerg Med J 2001;18:79-80.

13 Girasek DC. How members of the public interpret the word accident. Injury Prevention 1999;5:19-25.

14 US Centers for Disease Control and Prevention. Recommended framework for presenting injury mortality data. MMWR 1997;46(No RR-14):1-30. 

15 US Centers for Disease Control and Prevention. Revised framework of external cause of injury (E code) groupings for presenting injury mortality and morbidity data. 

At the time of publishing this article was free to download. In the event of this changing the post will be removed.

Would you like to look at something from the archive? Why not try this.

2 thoughts on “Do Accidents Happen?

Jim Milledge commented 1 year, 4 months ago
In my younger day I used to rock climb and occasionally to see, or be involved in rescue of a "Climbing Accident". The cause could be a variety of things. A hold could break off, or the climber could run our of strength in his fingers, or be unable to retreat if the climb be too hard for him or climber is hit by a falling rock. If I were to submit a paper to the BMJ about such an event, calling it a climbing crash suggests two or more climbers bumping into each other!
Stuart James Allan commented 1 year, 4 months ago
Love it! I'm guessing you've already read 'Black Box Thinking' by Matthew Syed. Stu

Have something to say? Comment on this post