“Do you regularly have more than six drinks in one sitting? Or do you regret a drunken escapade that took place in the past year? Answering yes to both questions may be a sign that you have a drink problem," the Mail Online reports.
This comes following a systematic review, which is, essentially, a study of studies.
The review aimed to examine whether short and quick screening approaches (comprising just one or two questions) can successfully and accurately identify people with alcohol problems during a GP visit.
The prevalence of people attending an appointment who have some form of alcohol problem has been suggested to be as high as 30%.
From the seven papers identified, using a single screening question such as “How often do you have six or more drinks on one occasion?” or “As a result of your drinking or drug use, did anything happen in the last year that you wish didn’t happen?” was not very accurate.
However, asking two screening questions increased accuracy (sensitivity) to 87.2%, meaning that only around one person in seven would be missed.
Asking either one or two questions would not be recommended as a single approach, however, because they’re not accurate enough. Instead, they appear to serve well as an initial screening technique, if they are then followed by a standard screening questionnaire.
Where did the story come from?
The study was carried out by researchers from Leicester General Hospital, and received no sources of financial support.
The study was published in the peer-reviewed medical journal British Journal of General Practice.
The Mail Online’s reporting of the study is accurate, but didn’t make clear that the initial two-question screening was not proposed to be used by itself.
This initial screening would be followed up by one or more longer, validated alcohol screening questionnaires if an alcohol problem was initially suspected.
Describing this more measured approach may have been less newsworthy than warning people that their GP was going to make the decision based on just two questions.
What kind of research was this?
This was a systematic review that aimed to see whether asking one or two simple questions could be an accurate and acceptable method for general practice screening to find out if people have an alcohol problem.
Previous research, the study authors’ said, suggested that up to a third of people attending general practice may be drinking at a level harmful to their health (called at-risk drinking) or have an alcohol use disorder. Many researchers consider that GPs might be well placed to identify any drink-related problems, as they can offer help and support at an early stage.
This study aimed to review the global literature to see if there was any evidence to support the use of very short alcohol screening questions in general practice.
What did the research involve?
The authors searched three literature databases – MEDLINE, PubMed and Embase – up to January 2014, using various search terms, including different terms for alcohol use disorders and terms to identify screening questions. They looked at the identified studies and only included those that assessed one or two questions to identify alcohol problems. They appraised the quality of included studies, and extracted information including study setting, patient characteristics, sample size, questions used to identify alcohol problems and accuracy of the screening questions.
What were the basic results?
The researchers identified six publications investigating one-question screening, and two studies investigating two-question screening. All were diagnostic studies designed to investigate the accuracy of diagnosing alcohol problems. The sample size of individual studies ranged from 227 to 1333 participants.
Most studies used valid diagnostic criteria to identify alcohol use disorders, and the overall prevalence of disorders across studies was 21%.
A single-question approach helped identify 453 out of 800 individuals with an alcohol use problem, with a pooled sensitivity across studies of 56.6%. This means that 56.6% of people with an alcohol problem were correctly identified by the screening question as having a problem.
On the flip side, 43% of people with an alcohol problem would have incorrectly been given the all-clear (false negatives).
By comparison, the pooled specificity of the single question was 81.3%, meaning that 81.3% of people without an alcohol problem were correctly identified by the screening question as not having an alcohol problem suggested (18.7% false positive rate). However, the individual studies had quite variable sensitivity and specificity results.
The most accurate single questions appeared to be “How often do you have six or more drinks on one occasion?” and “As a result of your drinking or drug use, did anything happen in the last year that you wish didn’t happen?” Both were reported to have excellent performance for ruling out people with alcohol problems, with low false negative results.
For the two-question approach, the sensitivity was 87.2% (proportion accurately identified to have an alcohol problem), and the specificity was 79.8% (proportion accurately identified as not having an alcohol problem). The optimal combination of questions was “recurrent drinking in situations in which it is physically hazardous” combined with “drinking in larger amounts or over a longer period than intended”.
The currently used 10-item AUDIT alcohol use questionnaire was found to be the most accurate single method for identifying alcohol use disorders, followed by the 4-item CAGE questionnaire.
However, the difficulty with these methods is that they would take GPs much longer to use these as screening methods. They were followed in accuracy by asking two screening questions, followed by one question.
The most accurate method was considered to be a stepwise approach of asking two initial screening questions, followed by the AUDIT or CAGE questionnaires to confirm.
How did the researchers interpret the results?
The researchers conclude that “two brief questions can be used as an initial screen for alcohol problems, but only when combined with a second-step screen. A brief alcohol intervention should be considered in those individuals who answer positively on both steps”.
This systematic review has examined global literature to identify studies that have assessed using one or two screening questions in general practice to identify people with alcohol use problems. The pooled results across the seven publications found the prevalence of alcohol use problems to be 21%.
Using a single question such as “How often do you have six or more drinks on one occasion?” or “As a result of your drinking or drug use, did anything happen in the last year that you wish didn’t happen?” was not very accurate, having a sensitivity of just over half. This means that half the people with alcohol problems would be missed. However, asking two screening questions increased sensitivity to 87.2%, meaning that less than 13% would be missed.
The optimal two question categories were “recurrent drinking in situations in which it is physically hazardous”, combined with “drinking in larger amounts or over a longer period than intended”.
However, as the researchers highlight, asking either one or two questions would not be recommended as a single approach, because they’re not accurate enough. They would need to be followed by the longer 10-item AUDIT alcohol use questionnaire or the 4-item CAGE questionnaire in a stepwise approach.
Both of these questionnaires used alone are more accurate than either one or two questions used alone, but would take GPs much longer to use these as screening methods. However, asking an initial two questions followed by AUDIT or CAGE would be the best way to identify people with alcohol problems, hopefully targeting them towards interventions.
However, as the study says, there are limitations. Despite the systematic review design, only seven diagnostic studies were identified, and it was not possible to conduct subgroup analyses. For example, it can’t tell us whether accuracy differs by male or female sex, or how good the screening methods would be at identifying people with different types of alcohol problem (e.g. people with actual alcohol dependence, or just harmful or hazardous drinking habits).
The researchers say that in the UK, “experts have recommended routine alcohol screening focusing on new patient registrations, general health checks and special types of consultation”. However, as they say, there are many things to consider, including the acceptability of asking even a single question related to alcohol use, “as some questions may not be welcome in unselected primary care attendees”.
The researchers aptly put forward “a cautious recommendation for one or two verbal questions as a screening test for alcohol-use disorder in primary care, but only when paired with a longer screening tool to decide who warrants a brief alcohol intervention”.
Further research is needed to clarify the added value of this approach compared with clinical assessment without the use of screening questions.
Despite alcohol misuse being associated with psychological denial, most people with an alcohol problem know that they have a problem.
A good source of advice is your GP. Be honest with them about how much you drink.
If your body has become dependent on alcohol, stopping drinking overnight can cause severe withdrawal symptoms, and in some cases can even can be life-threatening, so get advice about cutting down gradually.
Your GP may refer you to a local community alcohol service. Ask about free local support groups, day-centre counselling and one-to-one counselling.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
Mail Online, 7 July 2014
Links to the science
British Journal of General Practice. Published online July 1 2014