Developing a short form of the PGSI

Abstract

AIMS: To provide the Gambling Commission and its stakeholders with advice on the feasibility and suitability of a new brief three-item measure for use in tracking the prevalence of problem gambling in the general population. METHODS: A new three-item instrument was developed on the basis of the Problem Gambling Severity Index (PGSI). One question assesses the behavioural dimension of the problem gambling construct and the other two questions adverse consequences. Data from a continuous omnibus telephone survey, and the British Gambling Prevalence Surveys 2007 and 2010 was used to assess internal reliability and item-total correlations of the three items as well as overall performance, including sensitivity, specificity, positive predictive value, negative predictive value, and overall classification accuracy. RESULTS: The results of this analysis confirm that the new measure is a valid brief assessment instrument for problem gambling. Using a cutoff of 3 or more provides a reasonable approximation of the 'true' rate of problem gambling as determined by both clinical assessment and the full PGSI. Using a cutoff of 1 provides an excellent approximation of the 'true' rate of at-risk, problem and pathological gambling as determined by clinical assessment. Although the PGSI Short-Form is quite suitable for population prevalence research, it is not suitable for use in clinical settings due to high rates of false negatives (i.e., true problem gamblers not being identified). CONCLUSION: The new brief assessment can be used in future Omnibus and other surveys as a relatively low-cost means to monitor problem gambling prevalence over time in Britain. However, to minimize false positives, the instrument should only be administered to participants who have gambled in the past four weeks and/or report gambling at least a dozen times in the past year.

Problem with this document? Please report it to us.