A latent class analysis of DSM-IV and Fagerström (FTND) criteria for nicotine dependence

Abstract

Background: Nicotine dependence is associated with considerable morbidity and mortality. Two predominant classification systems, the Diagnostic and Statistical Manual (DSM-IV) and Fagerström Test for Nicotine Dependence (FTND), have been used to measure liability to nicotine dependence, yet few studies have attempted to simultaneously examine both sets of criteria.

Methods: Using a sample of 624 regular smoking individuals who are offspring of Vietnam Era Twin fathers ascertained for an offspring of twin study, we applied latent class analysis to the 7 DSM-IV and the 6 FTND criteria to classify individuals by their nicotine dependence symptom profiles. Post-hoc across-class comparisons were conducted using a variety of smoking-related variables and aspects of psychopathology. Whether a single class identified offspring at high genetic and environmental vulnerability was also investigated.

Results: The cross-diagnosis kappa was .30. A 4-class solution fit these data best. The classes included a low DSM-low FTND class and a high DSM-high FTND class; a moderate DSM-moderate FTND class, which was distinguished by moderate levels of smoking and intermediate levels of comorbid psychopathology; and a light smoking-moderate FTND class consisting primarily of lighter smokers with a more recent onset of regular smoking. High genetic and environmental vulnerability to nicotine dependence was noted in all classes with no statistically significant across-class differences.

Conclusions: In general, the DSM-IV and FTND criteria performed similarly to define a continuum of risk for nicotine dependence. The emerging class of light smokers should be further investigated to assess whether they transition to another class or remain as such.

Figure 1.

Endorsement probabilities of DSM-IV and FTND criteria in 624 regular smokers. HDHF = high DSM-high FTND; MDMF = moderate DSM-moderate FTND; LDLF = low DSM-low FTND; and LSMF = lighter smokers–moderate FTND (or low tolerance–moderate FTND). Note that for ordinal measures of time to first cigarette and cigarettes per day, the endorsement probability for the highest threshold is shown.