The aim of the study is to validate an enlarged version of the Lie/Bet questionnaire for screening pathological gambling behaviour (Johnson, Hamer & Nora, 1998) in French. This version of the Lie/Bet has the following features: it was enlarged with one item derived from de CAGE screening for alcoholism (Ewing, 1984) and items were rated on a 4-points scale ranging from ‘All the time’ to ‘Never’ instead of being rated on a ‘Yes/No’ scale. The addition of the third item meets the need for a measure of social irritability about gambling. The use of 4-options responses scales is motivated by the necessity to measure severity of problem gambling and allow parametric epidemiological analysis. The study was conducted on three samples: one group of people concerned with gambling who participated to an online survey about gambling (N= 36), one group of out patients (N=33) and one group of respondents to the Swiss Health Survey 2007 (N=75). Analyses were conducted to measure internal consistency and convergent validity of the instrument. Results show good internal consistency of the enlarged Lie/Bet as well as good convergent validity.