Table 3.

Predictors of the number of times the app was opened, ranked in decreasing order of importance

Predictors	%IncMSE
Intention to quit smoking assessed after 1 month	44.31
Cigarettes/day measured at 1-month follow-up	39.96
Perceived helpfulness of the app measured after 1 month	27.09
Experiment group (treatment vs. control)	25.89
Having quit smoking after 1-month follow-up	14.97
Current use of any smoking cessation app, as self-reported at 1-month follow-up	9.98
Number of years smoking	9.72
Age	8.31
Cigarettes/day measured at baseline	5.94
Number of minutes before the first cigarette of the day	3.63
Use of nicotine medications after 1 month	3.12
Depression	3.08
Use of heated tobacco product as self-reported after 1 month	2.99
Use of e-cigarettes at baseline	2.44
Use of e-cigarettes as self-reported after 1 month	2.23
Country (Switzerland vs. France)	0.77
Use of nicotine medications at baseline	0.61
Use/smoke other tobacco products at baseline	0.57
Use of heated tobacco product as self-reported after 1-month follow-up	0.49
Sex	0.03

%IncMSE, percent increase in mean squared error, a statistical measure indicating the level on the predictor variable importance in the regression machine learning algorithm.