2
demonstrated, many physicians fail to counsel patients. The most common reasons cited for lack of
counseling include inadequate training and time pressures. Our intervention will target medical
students in the early stages of training. The proposed intervention will provide a foundation for
medical learners in stage specific counseling and will aid physicians in primary practice to help
their patients stop smoking. The rationale for this program is that providing education early and
allowing students to use these skills with patients in the community can help: 1) future physicians
with confidence in smoking cessation counseling, 2) physicians in the community who may not
have adequate time to counsel patients, and 3) patients whose health may be at risk from smoking.
IV. Research Methods
Study Design: Randomized cross-over trial consisting of two smoking cessation counseling
interventions: 1) counseling intervention including patient education, written material and follow-
up by students who have been trained in stage specific tobacco cessation techniques, and 2)
counseling intervention that includes patient education, written material and follow-up by students
who have been trained in non-smoking cessation techniques (exercise counseling).
Setting: Community practice sites in internal medicine, family medicine and pediatrics
throughout Connecticut where medial students attend weekly continuity sessions with physician
preceptors.
Study Subjects: 80 first-year medical students and 308-350 patients aged 16 years or older in
the students’ community practice sites who are seeing the students’ physician preceptor for any
reason and meet criteria of smoking one or more cigarette daily in the previous week.
Randomization: Students will be randomized by the day they attend their Principles of Clinical
Medicine Course and trained in stage specific tobacco cessation counseling or exercise counseling.
After 6 months, students will receive training in the other behavioral counseling technique.
Main Outcome Measures: patients’ quit rate, stage of change, desire to quit, motivation to quit,
confidence in quitting (self-efficacy), and nicotine dependence at 1, 3, 6, and 12 months.
Process Measures: patient satisfaction with the quality of care in general, and satisfaction with
the quality of care related to smoking cessation counseling.
Analyses: patient level analyses of main outcome and process measures comparing patients
who received counseling from students trained in smoking cessation counseling and patients who
received counseling from students trained in exercise counseling adjusting for potential
confounding factors. We will use logistic regression for dichotomous outcomes and linear
regression for continuous outcomes. We will use generalized estimating equations (GEE) and
random effects modeling to allow us to adjust for time-dependent covariates