Almost nine in 10 patients with alcohol problems were correctly identified by the screening test in a study by the University of Leicester.
Researchers say it could make it easier and quicker for GPs to spot patients who require help with a drinking problem but whom do not fall into at-risk groups.
One in five people develop a drinking problem at some point in their lifetime, but GPs struggle to screen patients because of limited consultation time.
The team analysed 17 studies on the effectiveness of simple questions to detect alcohol problems, including data from 5,646 patients seen in primary care.
This analysis, published in the BJGP, found two questions correctly identified 87.2% of cases, while accurately filtering out 79.8% of those without alcohol problems.
The two-question screening test for alcohol misuse
Researchers said patients identified through the two-question test should then take a longer, more accurate survey, such as the four-item CAGE or the 10-item AUDIT questionnaires. The combination of the short and a long-form tests identified 90.9% of all cases.
Two-question test 'works well'
Lead author Dr Alex Mitchell said: 'GPs don’t have the time to ask a long list of questions for every single patient. The shorter the survey, the more acceptable it is for GPs - but the greater the danger that it is inaccurate.
'Our work shows that asking just two questions to patients works modestly well, but they need to be followed up by a longer questionnaire of four questions in those who initially screen positive.
'At the moment, GPs ask patients about alcohol problems during about 3% of consultations. Routine screening using simple questions are an attempt to improve the situation.'