Learning Adolescent Psychosocial Interviewing Using Simulated Patients

The area of communication skills in adolescent medicine is emerging as a distinct and important part of the undergraduate curriculum. An appropriate level of confidence in dealing with the adolescent population is deemed a necessary educational requirement.1 Skills in psychosocial communication with adolescents differ from those required for younger patients and adults2,3,4; they include discussing confidentiality and adolescent risk-taking activities. Simulated patients can be used effectively in teaching and evaluating of communication skills.5,6 However, there is no report of using adolescent simulated patients to teach communication skills. The evidence available is inconclusive regarding the teaching time required to promote retention of communication skills, although a recent review7 suggests that one day's training or less is not effective. Long-term retention of these skills has been supported by only one paper,8 suggesting a need to follow students over time to ascertain the effect of communication skills training. Our study addressed two questions: (1) does feedback from a simulated adolescent patient and simulated mother lead to improvements in fourth-year medical students' psychosocial interviewing of adolescent patients? and (2) does this skill persist following the intervention? Method Final-year medical students (N = 68) from March 1998 through May 1999 were invited to participate, and 57 agreed. The 11 who were unavailable to participate were either interviewing for their postgraduate education, involved in presenting their own research, or unable to make the scheduled times for the simulations. Thirty five other class members were either randomly or self-selected to go to offsite locations for pediatrics, and therefore could not participate; however, this group acted as a non-randomized control arm to the study. A two-group (57 students in the intervention group and 35 in the control) prospective randomized double-bind study design was employed. The students were completing an eight-week core pediatrics rotation in a tertiary center, with seven to nine students per rotation. Study Question 1 Intervention. Four simulated cases were developed, each comprising both a medical component (epilepsy, diabetes, attention deficit disorder, or asthma) and risk-taking activities (smoking, drugs, boyfriend issues) in which the adolescent was scripted to be involved. Nine simulated mothers and ten female adolescents (mean age 13.6 years) were recruited using established procedures.9 Mother-and-daughter pairs were selected as this is the commonest adolescent presentation in medical practice. Young adolescents were chosen to provide a realistic presentation of this age group, which often presents a challenge to young doctors. The training for standardized feedback was achieved when all mothers reviewed a single taped scenario, scored this independently using a structured form, and then discussed the feedback they would provide the student in a group setting. The adolescents were guided by their partner mothers to give feedback, which the adolescents discussed in a focus group. At study entry, all students signed informed consent forms. They then interviewed a simulated mother-daughter pair. The students were randomly assigned to receive immediate feedback following the pretest interview from the simulated pair (F2), or to receive no feedback (F1). All students conducted a second interview four weeks later using a different case scenario. All students (F1 and F2) received feedback from the simulated pair following this post-test interview. Feedback was structured using a written modified Calgary-Cambridge guide10 and given verbally; both interview content and process were addressed. Measures. Three measures were taken: Questionnaire. At study entry, demographic data and students' self-ratings of prior experiences with adolescent medicine, confidence in dealing with adolescent patients, and anticipated future work with adolescents were collected. Pre-test. Students conducted a one-hour videotaped interview with a simulated adolescent and mother, using one of the four case scenarios, at the midpoint of their rotation. The videotaped interviews were scored by a psychologist who had been trained to reach an acceptable level of agreement with the principal investigator (KB) using the modified Calgary-Cambridge guide.10 Post-test. Four weeks later, each student conducted a second videotaped interview, using a different case scenario. Scoring was completed in the same manner as for the pre-test. Study Question 2 Intervention. The entire final-year class participated in a mandatory ten-station OSCE prior to graduation. This was two to 12 months after participation in the study (mean 6.6 months). One pediatrics station of this OSCE tested general pediatrics knowledge (students' performances in asking about medical aspects of the case) and adolescent psychosocial interviewing (students' performances in asking about psychosocial aspects, e.g., boyfriend, alcohol, drugs). The OSCE included 35 off-site students, those not involved in the adolescent interviewing study, i.e., those who had not been videotaped and had received no feedback (F0) and 45 of the 57 students who had completed their pediatrics rotation at the tertiary center and who had participated in the study (F1 and F2). Measures. The knowledge score and the psychosocial interviewing score on the pediatrics OSCE station were obtained from the checklists completed by the faculty examiner at the station. Data Analysis Study Question 1. A single psychologist, blinded to student group or time of interview, scored the tapes, using a modified Calgary-Cambridge Observation Guide.10 The psychologist evaluated eight aspects of the encounter: how the student initiated the session, collected information, gathered information, asked the parent for time alone with the patient, dealt with the adolescent alone, and acted before and during the examination and closure. Each section yielded a global score. Within seven of the sections there were between three and ten individual items. The section used to rate when the student was alone with the adolescent included 14 psychosocial elements (i.e., boyfriend issues, smoking, and drugs). The psychologist derived eight global ratings for each videotape. The global ratings for F1 and F2 students at pre- and post-test were compared using a paired t test. Regression analysis was conducted using student global ratings from the eight sections of the modified Calgary—Cambridge Observation Guide as the dependent (outcome) variable. The independent (predictor) variables were feedback, case type and simulator, gender, previous medical experience with adolescents, comfort level in relating to adolescents, future career plans, and the students' scores on the pre-test case. Study Question 2. The knowledge score and the psychosocial interviewing score on the pediatrics OSCE station were compared among the three groups (F0, F1, F2). Results Complete data were available for 52 of the 57 students (F2 = 31; F1 = 21) who completed both pre- and post-test interviews. Two tapes could not be rated, and three students did not complete the second interview. Study Question 1. The mean pre-test scores of the group receiving feedback after their first interview (F2) and those receiving no initial feedback (F1) were not statistically different (72.93, SD = 9.43 versus 72.77, SD = 8.08; p = 0.95). However, the group that received feedback immediately after their first interviews (F2) scored significantly higher on the post-test (82.81, (SD = 9.79) than did the F1 group (76.34, SD = 9.43); p = 0.02). No significant improvement was seen from pre-test to post-test for the group receiving no initial feedback (F1). However, the group receiving feedback (F2) improved significantly from pre-test to post-test (p = 0.02). Regression analysis revealed that receiving feedback was the only significant predictor (p =.021) of students' performances on the post-test case (R2 =.10 for the complete model). The other independent variables did not significantly predict post-test performance. Analyses also were conducted to determine whether or not the particular case scenario used had a significant influence on student performance. No statistically significant influence due to case difference emerged. Study Question 2. All students participating in the study received feedback either once (F1) or twice (F2). Both groups (n = 45) had significantly higher mean scores (p =.023) on the adolescent psychosocial inquiry on the final-year OSCE station (68.06, SD = 24.07) compared with the students (n = 35) who completed their core pediatrics rotation at the offsite placements (F0) (55.71, SD = 23.16). The groups did not differ significantly (p = 0.40) in their mean scores for the general knowledge aspects of this OSCE station (F1 and F2) (70.71, SD = 16.88) compared with F0 (67.53, SD = 16.69). After the OSCE the students were asked to comment on their clerkship experience. The simulated adolescent encounters were rated as one of the most positive learning experiences in the two years of clerkship. Discussion The main study finding is that the important communication skill of interviewing the adolescent patient can successfully be taught to undergraduate medical students. The teaching becomes faculty-independent when the simulated patients are scripted and trained in giving structured feedback. The training period, which was a one-hour interview (experimental), followed by 20 minutes of feedback, was much less than one day, which is the time reported in the literature as necessary for effective learning of these skills. This study poses questions for further research regarding optimal training time and the best method of reinforcement. For psychosocial interviewing with sensitive questioning, clerkship seems the optimal point of instruction; however, there is little evidence to inform where training in communication with adolescents should be placed in the medical curriculum. There are several limitations to this study. First, the sample was small, although representative of other randomized controlled trials in this field. Second, selection bias may have occurred, as the students who chose to complete their core pediatrics rotations at offsite placements were either randomly or self-selected. However, all students received the core pediatrics tutorials from the tertiary center by teleconference, along with detailed objectives. This ensured that all students received the same didactic curriculum. Third, although the study would have benefited from two independent raters, the increased cost was prohibitive. The psychologist rater was trained to use the modified Calgary—Cambridge Guide10 and underwent a mid-study validation of his scoring. Fourth, our sample was confined to mothers and daughters; whether the results would differ with mother—son simulator pairs is unclear. Fifth, although this study provides some indication that students' psychosocial communication skills can be improved and maintained over time, follow up was less than a year. Continued tracking of these doctors would be important to see whether this mastery is maintained into the residency years. Finally, application of these results must consider resources. At our medical school, standardized patients frequently supplement current teaching activities, and are part of the diagnostic assessment of student skills throughout the medical school curriculum. Expertise to train and administer such a program is quite involved from a logistic and monetary standpoint; although available at our medical school, this may not be the case everywhere. As this educational initiative relies on a realistic portrayal and structured feedback from the adolescent, time spent in recruitment and training of the standardized patients is important. Students overwhelmingly commented that feedback from a “real” adolescent was very helpful, as they had received little training in this area. Many of the students were very apprehensive on entry into the study, but were resoundingly positive after they had completed it. Because of the changing nature of the hospitalized patient population, standardized patients could be used to ensure that each student has exposure to common ambulatory problems. They could help ensure uniformity in teaching and learning of basic clinical skills. Interviewing an adolescent standardized patient who is involved in risk-taking activities provides the student an opportunity to practice psychosocial interviewing in a safe setting. The immediate feedback provided by the adolescent and mother is a powerful teaching tool. The student can then return to the clinical setting to apply these newly acquired skills. In conclusion, this randomized controlled trial has shown that final-year medical students can be taught adolescent interviewing skills and that these skills are retained for as long as a year. The teaching time required for such an intervention is short (90 minutes), and teaching can be independent of faculty once the simulators' training is completed. As the skill of talking to adolescents and their parents is an important part of physician training, we would recommend that medical schools consider this structured training for their curricula.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.