Search

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.addbeh.2006.03.030 Byline: Ji-Hyun Lee, Thaddeus A. Herzog, Cathy D. Meade, Monica S. Webb, Thomas H. Brandon Keywords: Statistical analysis; Binary longitudinal data; GEE; Tobacco relapse; Relapse-prevention-intervention Abstract: Longitudinal study designs in addictive behaviors research are common as researchers have focused increasingly on how various explanatory variables affect responses over time. In particular, such designs are used in intervention studies that have multiple follow-up points. These designs typically involve repeated measurement of participants' responses, and thus correlation within each participant is expected. Correct inferences can only be obtained by taking into account this within-participant correlation between repeated measurements, which can complicate the analysis of longitudinal data. In recent years, generalized estimating equations (GEE) has become a standard method for analyzing non-normal longitudinal data, yet it often is not utilized by addiction researchers. The goal of this article is to provide an overview of the GEE approach for analyzing correlated binary data for behavioral researchers, using data from an intervention study on the prevention of relapse to tobacco smoking. Author Affiliation: H. Lee Moffitt Cancer Center and Research Institute, The University of South Florida, Tampa, Florida, USA

Relapse prevention remains a major challenge to smoking cessation efforts. T. H. Brandon, B. N. Collins, L. M. Juliano, and A. B. Lazev (2000) found that a series of 8 empirically based relapse-prevention booklets mailed to ex-smokers over 1 year significantly reduced relapse. This study dismantled 2 components of that intervention: the amount of content (number of booklets) and the frequency of contact. Content and contact were crossed in a 2 x 2 factorial design. The criteria of at least 1 week of abstinence at baseline was met by 431 participants, 75%-85% of whom returned 12-, 18-, and 24-month follow-up questionnaires. Eight booklets produced consistently higher point-prevalence abstinence rates than did a single booklet, but frequency of contact did not affect outcome. Moreover, the high-content interventions were highly cost-effective.

Individuals with schizophrenia have a much higher prevalence of tobacco smoking, a lower cessation rate, and a higher incidence of tobacco-related diseases than the general population. The initial challenge has been to motivate these individuals to quit smoking. This study tested whether motivational interviewing is effective in motivating smokers with schizophrenia or schizoaffective disorder to seek tobacco dependence treatment. Participants (N = 78) were randomly assigned to receive a 1-session motivational interviewing (MI) intervention, standard psychoeducational counseling, or advice only. As hypothesized, a greater proportion of participants receiving the MI intervention contacted a tobacco dependence treatment provider (32%, 11%, and 0%, respectively) and attended the 1st session of counseling (28%, 9%, and 0%) by the 1-month follow-up as compared with those receiving comparison interventions.

Practicing psychologists are well positioned to provide at least minimal interventions for tobacco dependence among their clients. Because smoking covaries with psychopathology, a substantial proportion of psychologists' clients are likely to be smokers. Psychologists have expertise in motivating behavior change, and they have greater contact and stronger relationships with their patients than do most other health providers. Despite these advantages, the current study found that psychologists were less likely to intervene for tobacco use than for other client behaviors, such as alcohol or illicit drug abuse. Psychologists do not regularly ask their clients whether they smoke, advise them to quit, assess their willingness to quit, assist them with quitting, or arrange follow-up. Recommendations for additional training are provided.

Key Words tobacco, addiction, drug use, nicotine, dependence * Abstract Cigarette smoking is a leading cause of mortality and morbidity and a particularly common and intractable addictive disorder. Research shows […]

R. Eisenberger's (1992) learned industriousness theory states that individuals display differing degrees of persistence depending on their history of reinforcement for effortful behavior. These differences may influence the development, maintenance, and cessation of addictive behaviors. In cross-sectional studies, E. P. Quinn, T. H. Brandon, and A. L. Copeland (1996) found that cigarette smokers were less persistent than nonsmokers, and R. A. Brown, C. W. Lejuez, C. W. Kahler, and D. R. Strong (2002) found that smokers who had previously abstained for 3 months were more persistent than those who had never quit. The present study extended these findings by using a prospective design. A pretreatment measure of task persistence (mirror tracing) completed by 144 smokers predicted sustained abstinence throughout 12 months of follow-up. Moreover, persistence predicted outcome independent of other significant predictors: gender, nicotine dependence, negative affect, and self-efficacy.

Pavlovian conditioning models have influenced the development of cue exposure treatments for drug abuse. However, poor maintenance of extinction performance (renewal) after treatment is a common problem. A treatment-analogue experiment tested the role of context in renewal, as well as a potential strategy for reducing renewal. Seventy-eight social drinkers completed extinction trials to reduce saliva and urge reactivity to alcohol cues and were randomly assigned to a renewal test in either the same context as extinction, a different context, or the different context containing a cue from the extinction context (E-cue). As predicted, the different context produced greater renewal than the same context and renewal was attenuated when the E-cue was present. These results offer preliminary evidence for the context dependence of extinction to alcohol cues and for the use of an extinction cue to improve the generalizability of exposure therapies.

The balanced placebo design (BPD) was used to evaluate the independent effects of nicotine dose and smoking-related expectancies on self-reported anxiety, urge to smoke, and withdrawal symptoms. After anxious mood was induced, participants smoked either a de-nicotinized cigarette or one with standard nicotine content. Nicotine dose was crossed with instructions that the cigarette was either de-nicotinized or standard. Nicotine cigarettes produced greater anxiety reduction than de-nicotinized cigarettes. Nicotine instructions attenuated anxiety only among those who held relevant expectancies. Nicotine dose and instructional set interacted such that either nicotine cigarettes or instructions that the cigarettes contained nicotine were sufficient to reduce urge to smoke. Implications of these findings and methodological issues regarding use of the BPD with cigarettes are discussed.

Behavioral interventions often are administered in groups, yet the effects of group membership rarely have been evaluated. The current research examined 33 groups of clients (M = 5.5 clients per group, SD = 2.5) volunteering for a group smoking cessation intervention. The intervention consisted of 6 group therapy sessions over an 11-day period. Attendance at the sessions and smoking behavior during the 11-day period were the dependent variables. Hierarchical linear modeling (HLM) revealed a statistically significant (p < .05) degree of interdependence among group members (intraclass correlation = .44 and .32 for attendance and smoking, respectively). Groups were characterized disproportionately by high attendance and infrequent smoking or low attendance and relatively frequent smoking, with fewer moderate groups than would be expected by chance. Group membership effects dissipated within a month following treatment. These findings suggest the need for assessing such effects and for future research examining their causes.

Research into the causal role of outcome expectancies in smoking motivation and behaviour is presented. It is suggested that health risk expectancies could be very strong motivators for behaviour change.

The Smoking Consequences Questionnaire (SCQ) was originally designed to measure the outcome expectancies of college students for cigarette smoking. A new version is developed that is intended for adult smokers. Two different forms of SCQ-Adult are introduced, with one consisting of subjective expected utility items and the other consisting of probability items. Analysis of these measures indicates that the specificity of expectancies increases along with smoking experience.

Smokers (N= 126) were randomly assigned to 6-session smoking cessation treatments consisting of 1 of 2 counseling strategies (skills training or support) and I of 2 nicotine exposure strategies (nicotine gum or rapid smoking). Counseling and nicotine strategies were completely crossed; all four combinations resulted in equivalent 1-year abstinence rates. Skills training produced higher initial cessation and more coping responses posttreatment than did support. Rapid smoking, but not nicotine gum, produced tachycardia to the taste of cigarettes posttreatment, consistent with cigarette aversion. The treatments were differentially effective among subpopulations of smokers: Subjects high in pretreatment negative affect responded best to support counseling; those low in pretreatment negative affect responded best to skills training. Self-reports of pretreatment craving predicted response to the nicotine exposure treatments., The comparative effectiveness and possible routes of impact of two intervention techniques in smoking cessation treatments are analyzed. Counseling treatments, either supportive counseling or skills training, and nicotine exposure treatments (rapid smoking and nicotine chewing gum) are applied to a sample population of smokers. Results indicate that all four combinations have led to equivalent one-year abstinence rates, but that the treatments were differentially effective.