[Instruments used to evaluate smoking habits: a systematic review]

Smoking is one of the most important causes of illness in the world, responsible for one out of every eight deaths. To know the instruments that characterize the use of tobacco is the first step to develop reliable and comparable research to tackle this challenge. The objective was to identify and compare instruments and domains used in population-based studies over the past five years aiming to determine smoking patterns. A systematic review was conducted on articles published from September 2002 to September 2007. The terms used were: (*Smok or tobacco) AND (Questionnaire or scale or score or instrument or assessment or form) AND (*cultural* or translat* or valid* or reproduc* or psychomet*). Out of 2,236 references, 186 articles were selected, of which only 91 informed the instruments used. From the 91 manuscripts, 49 distinct instruments were identified. The main topics were profile and prevalence (38%), dependence (24%) and motivation (10.8%). The researchers used standard instruments in 96% of studies of addiction. The majority of articles about profile and prevalence (79%) used self-developed questionnaires. The transparency and standardization of instruments and the preference for the use of validated questionnaires are crucial areas for quality and reproducibility of research on smoking.