Article Text


Clinical assessment compared with breathalyser readings in the ER: concordance of ICD-10 Y90 and Y91 codes
  1. R Touquet
  1. Correspondence to:
 Professor R Touquet
 A & E Department, St Mary’s Hospital, Praed Street, London, W2 1NY;

Statistics from

Clinical assessment of the amount of alcohol that has been consumed by a patient presenting to the emergency department is notoriously difficult. How should the caring professions respond to the “Epidemic of Binge Drinking in the United Kingdom (UK)”?1 Alcohol is a drug like any other, which depresses inhibitions initially, but is legal and used by 90% of society, including the medical profession.2

This paper, a work from twelve countries, defines the accuracy of clinical assessment (ICD-10 Y91) by comparison with alcohol breathalyser readings (ICD-10 Y90). It is well worked, multinational, with robust methodologies supported by the World Health Organization (WHO).

The paper shows that approximately 95% of patients, with blood alcohol concentrations (BAC—as measured by breathalyser) below 60 mg per 100 ml of blood, were correctly diagnosed as being not intoxicated. Approximately 85% of patients with BAC above 60 mg% were correctly assessed as being intoxicated. However, there was a low incidence of alcohol intake in this population of 4798 patients; approximately 80% had not been drinking alcohol within six hours of attending the emergency room.

The measurement instrument RAPS4—developed from optimal items across several screening instruments, including the well known CAGE questionnaire—does effectively detect dependency.

This paper, therefore, really confirms what we all suspect: that clinical assessment is difficult and for it to be of use clinicians need training.

So how does this paper affect our own practices in the UK? Training in the detection and management of alcohol misuse is judged as important,2 and this paper from Cherpitel et al emphasises this need. Emergency department staff, who may be inexperienced, need to be able to make pragmatic, prompt management decisions. The key question is “Is the patient a hazardous drinker who warrants referral for brief intervention with an alcohol health worker (AHW)”?3–5 as opposed to “Is the patient intoxicated at presentation to the accident and emergency department”?

The ICD-10 classifications do not answer this—hence their limitation.

If a patient is alert and orientated, then detecting hazardous drinking by questionnaire is effective. Use of breathalysers is perceived as judgemental causing resentment.

Patients, not alert and orientated, warrant blood alcohol levels to help clinical decision making,6–8 and to flag up possible alcohol misuse for later definition with questionnaire—for example, the Paddington Alcohol Test.

Cherpitel et al contribute to the literature by highlighting the need for accident and emergency department staff to be trained in assessing patients for possible alcohol misuse, and, further, that the WHO has no coding for the binge or dependent drinker. Such detection should trigger the offer of referral to the AHW, thereby making maximum use of the “teachable moment”.5

Alcohol abuse remains a major challenge in our everyday practice and the time has come to move from the reactive approach of dealing with the immediate problem to referring patients for proactive management of problem drinking. We have a long way to go before the recommendation from the Royal College of Physicians, that every acute trust must have their own AHW,9 is met.


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.