Table 5.

The importance of each predictor of the apps use duration: decreasing order

Predictors	%IncMSE
Cigarettes/day measured at 1-month follow-up	135.43
Intention to quit smoking assessed after 1-month follow-up	125.52
Perceived helpfulness of the app measured after 1-month follow-up	56.59
Having quit smoking after 1-month follow-up	44.66
Current use of any smoking cessation app, as self-reported at 1-month follow-up	28.23
Age	26.98
Number of years smoking	22.02
Experiment group (treatment vs. control)	20.01
Use of nicotine medications after 1 month	15.68
Number of minutes before the first cigarette of the day	14.86
Use of nicotine medications at baseline	12.87
Use of heated tobacco product as self-reported after 1-month follow-up	12.67
Cigarettes/day measured at baseline	9.54
Sex	8.51
Depression screening test positive	7.92
Country (Switzerland vs. France)	6.76
Use of e-cigarettes at baseline	4.21
Use/smoke other tobacco products at baseline	3.51
Use of e-cigarettes after 1-month follow-up	1.57
Use of heated tobacco product at baseline	0.61

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