DS had to report smoking daily, an average of 5–30 cigarettes per day (CPD), while ITS just had to report smoking 4–27 days per month. The study sample is described in greater detail in Shiffman et al. (2012c). DS (n = 218) were 37.2% African American (AA), 43.6% female, and averaged 40.7 (SD = 11.3) years of age. ITS (n = 252) were 31.0% AA, 50.4% female, and averaged 36.0 (SD = 12.3)
years of age. DS reported having smoked on average 25.2 (SD = 10.9) years, averaged 15.0 (SD = 5.9) CPD, and had an average Fagerstrom Test for Nicotine LGK-974 datasheet Dependence (FTND; Heatherton et al., 1991) score of 5.1 (SD = 1.9). ITS smoked for an average of 18.0 (SD = 12.1) years, averaged 4.5 (SD = 2.9) CPD on days in which they smoked, and had a mean FTND score of 1.4 (SD = 1.6), with nearly half of ITS (44.4%) having FTND scores of 0. Among ITS, CITS (n = 152) averaged 38.2 (SD = 12.4) years of age and NITS’ (n = 80) mean age was 33.3 (SD = 11.6) years. [Classification data were missing for 20 ITS, who thus do not participate in these analyses.] CITS were 37.5% AA and 56.6% female. Among NITS, 21.3% were AA and 41.3% were female. CITS reported having smoked, on average, for 20.2 (SD = 12.1) years, an average of 5.1 (SD = 3.1) CPD on smoking days, and had an average FTND score of 1.7 (SD = 1.8). NITS reported smoking an average of 13.4 (SD = 10.7) years,
3.3 (SD = 1.8) CPD on smoking Megestrol Acetate days, and scored a mean of 0.9 (SD = 1.2) Fludarabine cost on the FTND. Only participants (n = 471) who had non-missing
scores for all 13 WISDM scores were included: 41 participants dropped from the study before completing the WISDM and 3 skipped multiple items. One additional participant who responded “7” to all WISDM items, yielding a highly implausible profile with no variation (SD = 0), and undefined standardized scale scores, was dropped. Finally, a subset of participants completed WISDM assessments for which 2 items (one contributing to the Taste-Sensory scale; one contributing to the Weight Control scale) were systematically missing. For these individuals, scores for the Taste-Sensory (n = 115) and Weight Control (n = 117) scales were imputed using highly predictive multivariate regression (R2s = .98) from other subscale items. Subjects completed the 68-item WISDM, which was scored to yield 13 subscales. Piper et al. (2004) reported high reliability for all subscales, and a consistent factor structure in daily and non-daily smokers, suggesting it is suitable for use in ITS. Subjects were interviewed about their smoking history, to determine whether they had ever smoked daily for 6 months or more (CITS) or not (NITS); see Shiffman et al. (2012c) for a more complete description. Profile scatter was indexed by the within-profile standard deviation across the 13 scales of the WISDM.