Predicting and modifying substance use in childhood cancer survivors: application of a conceptual model.

PURPOSE/OBJECTIVES To identify factors that predict or modify substance use in childhood cancer survivors and to describe how a risk-counseling intervention reduced young survivors' substance use. DESIGN Secondary analysis of clinical trial data and primary analysis of medical record data. SETTING Outpatient clinic. SAMPLE 149 females and 118 males 12-18 years of age whose cancer had been in remission for at least two years were randomly assigned to intervention (n = 132) and standard care (n = 135) groups. METHODS Self-report questionnaires, abstracted medical record data, confirmatory factor analysis, and structural equation modeling. MAIN RESEARCH VARIABLES Smoking, alcohol consumption, knowledge, risk perceptions, motivation, and worry about cancer and treatment effects. FINDINGS Three factors directly predicted substance use at baseline: being in a higher grade in school (independent of age), feelings of being more susceptible to late effects of cancer therapy, and worrying more about cancer and its treatment. At follow-up a year later, grade in school and worry predicted increased substance use. In addition, a desire to change health behavior, influenced by the intervention and gender, predicted decreased substance use. The mechanism of influence of the intervention was evident: The intervention led to a need to change, which precipitated a desire to change and ultimately resulted in decreased substance use. CONCLUSIONS Young survivors' worries and concerns about their cancer and treatment-related late effects are a new intervention target. Motivation is sensitive to behavioral change interventions and positively affects risk reduction. IMPLICATIONS FOR NURSING Two new intervention strategies to address the impact of survivors' concerns about their cancer and its treatment are implied: (a) Replace substance use with new coping methods to reduce fear and anxiety, and (b) tailor motivation-based interventions to age and gender to communicate graphically and realistically to survivors the personal importance of behavioral change in modifying the risks of late effects.