Background It has been argued that as smoking prevalence declines in countries, the smokers that remain include higher proportions of those who are unwilling or unable to quit (a process known as ‘hardening’). Smokeless tobacco and e-cigarettes have been promoted as a strategy to deal with such smokers. If hardening is occurring, there would be a positive association between smoking prevalence and quitting, with less quitting at lower prevalence. There would also be a neutral or negative association between prevalence and the number of cigarettes smoked.

Methods We examined US state-level associations using the Tobacco Use Supplement (1992/1993–2010/2011) and Eurobarometer surveys for 31 European countries (2006–2009–2012) using regressions of quit attempts, quit ratios, and number of cigarettes smoked on smoking prevalence over time.

Results For each 1% drop in smoking prevalence, quit attempts increase by 0.55%±.07 (p<0.001) in the USA and remain stable in Europe (p=0.53), US quit ratios increase by 1.13%±0.06 (p<0.001), and consumption drops by 0.32 cig/day±0.02 (p<0.001) in the USA and 0.22 cig/day±0.05 (p<0.001) in Europe. These associations remain stable over time (p>0.24), with significantly lower consumption at any given prevalence level as time passed in the USA (−0.15 (cig/day)/year±0.06, p<0.05).

Conclusions Consistent with prior research using different data and methods, these population-level results reject the hypothesis of hardening as smoking prevalence drops, instead supporting softening of the smoking population as prevalence declines.