Can we predict who will quit? Does it matter?

Vangeli et al (Addiction 106: 2110-2121, 2011) identified eight large, population based studies of smoking cessation in the real world. The surveys included both US and other countries. Six used data acquired between 2000 and 2010. These studies offer the most generalizable test of predictors of making a quit attempt or predicting success once one tries to quit. I was very surprised with their outcomes. Before this study I had thought older smokers make fewer attempts but are more successful, that women make more attempts but are less successful, and that higher education/income, smoking restrictions and motivation predicted both making a quit attempt and success. But looks like demographics are not good predictors, that dependence is good negative predictor both of attempts and success, and that motivation predicts attempts but not success. One other surprise to me was that high dependence predicts fewer attempts to quit. So to me, this suggests maybe we should try more aggressive motivational interventions in this group. Another surprise was that motivational and self-efficacy measures did not predict abstinence success because several meta-analyses of motivational interviewing interventions indicate they do increase successful abstinence. No sure how to integrate these paradoxical findings. It’s actually important to figure this out because most tobacco control interventions focus on motivating smokers and assume that such motivation will increase both an attempt and success. With this model, treatment is not that necessary. If instead, we believe motivation is not that important for quitting and instead overcoming dependence is more important, then that suggests treatment is important. Finally, I cannot help but comment that many in tobacco control will say that cessation is mostly a motivational problem, not a dependence problem. But then will turn around and say the tobacco companies are evil for producing dependence and trapping smokers into smoking. The former is contrary to the later. Predictors of Quit Attempt and of Duration of Abstinence

	Probability a quit attempt occurred	Successful abstinence after a quit attempt
Age	No	No
Sex	No	No
Smoking restrictions home	Yes	No
Smoking restrictions work	No	No
Cigs/Day	Yes (inverse)	Yes (inverse)
Dependence	Yes (inverse)	Yes (inverse)
Past quit attempt	Yes	Yes (inverse)
Longest prior abstinence	Yes	Yes
Motivation to quit	Yes	No
Intention to quit	Yes	No
Confidence to quit	No	No
Enjoyment of smoking	Yes	Not tested