Capturing habitualness of drinking and smoking behavior in humans.

backgroundRecent findings suggest that overreliance on habit may be common in individuals diagnosed with addiction. To advance our understanding of habit in clinical samples and from behavioral measures, this study examines the interrelations between self-reported habit index for smoking and drinking as well as behavioral measures of intraindividual variability in smoking and drinking.

methodsTreatment-seeking heavy drinking smokers (N = 416) completed the Self-Report Habit Index (SRHI) adapted for both smoking and drinking. "Behavioral habitualness" was computed from the degree of intraindividual variability in patterns of smoking and drinking over the past month. Using the 28-day Timeline-Follow Back (TLFB) interview, we derived two measures of intraindividual variability: interclass correlation (ICC) and autocorrelation [AR(7) coefficients].

resultsSelf-report measures of habit were robustly associated with clinical severity of drinking and smoking with higher habit scores indicating greater severity of drinking and smoking, respectively. ICC and AR(7) coefficients, the behavioral measure of "patterness" and putative habit, were not associated with SRHI scores. While ICC for smoking was associated with higher nicotine dependence scores, this pattern was not found for drinking ICC and alcohol problem severity.

conclusionsThese results support the construct validity of the self-report measures of habit for smoking and drinking, as well an initial evaluation of behavioral measure of smoking "patterness" as a potential proxy for habit smoking. Because habit represents a complex phenotype with limited clinical translation, additional studies capturing a wider range of substance use severity and coupled with brain-based validation methods are warranted.