Post-ban self-reports on economic impact of smoke-free bars and restaurants are biased by pre-ban attitudes. A longitudinal study among employees.

INTRODUCTION: Objective sales data have indicated that the Norwegian indoor smoke-free regime implemented in June 2004 did not affect the hospitality business negatively. This paper investigates whether self-reports on the economic impact of the smoking ban from employees in the hospitality sector gave similar results, and whether post-ban self-reports on the economic impact of the ban were influenced by pre-ban attitudes towards smoke-free policies.
METHODS: A random sample of 516 employees in bars and restaurants stated their attitudes towards smoke-free policies shortly before the ban became effective. One year later, the same respondents gave their perceptions of changes in patronage for their workplace. Data were collected using self-administered questionnaires and telephone interviews.
RESULTS: 56.4% of the hospitality workers stated that the ban on smoking had led to a large reduction (28.2%) or some reduction (28.2%) in the number of patrons, a result inconsistent with analysis of sales data. After adjusting for demographic and smoking-related variables, a negative pre-ban attitude significantly increased the odds for reporting a negative economic impact post ban (odds ratio 2.48, confidence interval 1.48-4.14).
CONCLUSIONS: Subjective reports of the economic impact from an indoor ban on smoking are influenced by attitudes towards smoke-free policies and should not be considered as a valid measure of the economic effect.