Table 2.
Relationship Between Measures of Dependence and Observed Smoking Behaviors Among Intermittent Smokers (ITS)
| Variable | Mean CPD | Max CPD | Max run | Proportion of days smoked | ||||
| β | p | β | p | β | p | β | p | |
| FTND | ||||||||
| Linear | 0.45 | <.0001 | 0.39 | <.0001 | −0.26 | <.0003 | 0.43 | <.0001 |
| Quadratic | 0.24 | <.02 | 0.17 | NS | 0.04 | NS | −0.14 | NS |
| TTFC | ||||||||
| Linear | −0.26 | <.0004 | −0.21 | <.003 | 0.33 | <.0001 | −0.51 | <.0001 |
| Quadratic | 0.29 | <.007 | 0.26 | <.02 | −0.12 | NS | 0.19 | <.05 |
| NDSS-T | ||||||||
| Linear | 0.44 | <.0001 | 0.37 | <.0001 | −0.34 | <.0001 | 0.46 | <.0001 |
| Quadratic | 0.12 | NS | 0.10 | NS | 0.18 | <.04 | −0.24 | <.003 |
| WISDM primary | ||||||||
| Linear | 0.53 | <.0001 | 0.51 | <.0001 | −0.41 | <.0001 | 0.53 | <.0001 |
| Quadratic | −0.07 | NS | 0.00 | NS | 0.22 | <.02 | −0.25 | <.004 |
| WISDM secondary | ||||||||
| Linear | 0.26 | <.0003 | 0.28 | <.0001 | −0.25 | <.0004 | 0.33 | <.0001 |
| Quadratic | −0.01 | NS | 0.00 | NS | 0.12 | NS | −0.13 | NS |
| HONC continuous | ||||||||
| Linear | 0.30 | <.0003 | 0.27 | <.0008 | −0.13 | NS | 0.31 | <.0002 |
| Quadratic | 0.01 | NS | −0.02 | NS | 0.08 | NS | −0.11 | NS |
| HONC- dichotomous | ||||||||
| Lineara | 0.15 | NS | 0.17 | <.04 | −0.16 | <.05 | 0.24 | <.003 |
Note. All models were weighted by race and controlled for history of daily smoking. CPD = cigarettes per day; β = standardized regression coefficient; FTND = Fagerström Test for Nicotine Dependence, modified to exclude CPD item when used in CPD models; TTFC = time to first cigarette (hr); NDSS = Nicotine Dependence Syndrome Scale; WISDM = Wisconsin Inventory of Smoking Dependence Motives; HONC = Hooked on Nicotine Checklist; NS = nonsignificant.
Only linear (not quadratic) relations are possible with a dichotomy.