BACKGROUND: Patient behavior accounts for half or more of the
variance in health, disease, mortality and treatment outcome and costs.
Counseling using motivational interviewing (MI) effectively improves the
substance use and medical compliance behavior of patients. Medical
training should include substantial focus on this key issue of health
promotion. The objective of the study is to test the efficacy of teaching
MI to medical students.
METHODS: Thirteen fourth-year medical students volunteered to
participate. Seven days before and after an 8-hour interactive MI
training workshop, each student performed a video-recorded interview
with two standardized patients: a 60 year-old alcohol dependent female
consulting a primary care physician for the first time about fatigue and
depression symptoms; and a 50 year-old male cigarette smoker
hospitalized for myocardial infarction. All 52 videos (13 students×2
interviews before and after training) were independently coded by
two blinded clinicians using the Motivational Interviewing Training
Integrity (MITI, 3.0). MITI scores consist of global spirit (Evocation,
Collaboration, Autonomy/Support), global Empathy and Direction,
and behavior count summary scores (% Open questions, Reflection
to question ratio, % Complex reflections, % MI-adherent behaviors).
A "beginning proficiency" threshold (BPT) is defined for each of these
9 scores. The proportion of students reaching BPT before and after
training was compared using McNemar exact tests. Inter-rater
reliability was evaluated by comparing double coding, and test-retest
analyses were conducted on a sub-sample of 10 consecutive interviews
by each coder. Weighted Kappas were used for global rating
scales and intra-class correlations (ICC) were computed for behavior
count summary scores.
RESULTS: The percent of counselors reaching BPT before and after MI
training increased significantly for Evocation (15% to 65%, p<.001),
Collaboration (27% to 77%, p=.001), Autonomy/Support (15% to 54%,
p=.006), and % Open questions (4% to 38%, p=.004). Proportions
increased, but were not statistically significant for Empathy (38% to
58%, p=.18), Reflection to question ratio (0% to 15%, p=.12), %
Complex reflection (35% to 54%, p=.23), and % MI-adherent behaviors
(8% to 15%, p=.69). There was virtually no change for the Direction
scale (92% to 88%, p=1.00). The reliability analyses produced mixed
results. Weighted kappas for inter-rater reliability ranged from .14 for
Direction to .51 for Collaboration, and from .27 for Direction to .80 for
Empathy for test-retest. ICCs ranged from .20 for Complex reflections to
.89 for Open questions (inter-rater), and from .67 for Complex reflections
to .99 for Reflection to question ratio (test-retest).
CONCLUSION: This pilot study indicates that a single 8-hour training
in motivational interviewing for voluntary fourth-year medical students
results in significant improvement of some MI skills. A larger sample of
randomly selected medical students observed over longer periods
should be studied to test if MI training generalizes to medical students.
Inter-rater reliability and test-retest findings indicate a need for caution
when interpreting the present results, as well as for more intensive
training to help appropriately capture more dimensions of the process
in future studies.