Family Practice Vol. 20, No. 4, 363-369
© Oxford University Press 2003
Mental health |
Improving psychiatric interview skills of established GPs: evaluation of a group training course in Italy
Istituto di Psichiatria, Università di Bologna, Bologna, Italy,
a Western Psychiatric Institute and Clinic, Pittsburgh, PA, USA and
b National Primary Care Research and Development Centre, University of Manchester, UK.
Correspondence to Andrea Scardovi, MD, PhD, Istituto di Psichiatria, Università di Bologna, Viale Pepoli 5, 40123 Bologna, Italy; E-mail: andreascardovi{at}libero.it
Scardovi A, Rucci P, Gask L, Berardi D, Leggieri G, Berti Ceroni G and Ferrari G. Improving psychiatric interview skills of established GPs: evaluation of a group training course in Italy. Family Practice 2003; 20: 363–369.
Received 17 September 2002; Revised 25 February 2003; Accepted 28 March 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Background. A number of studies have shown that the communication style of GPs has a significant impact on their ability to recognize and manage psychiatric disorders. Italian training programmes do exist, but none have been evalutated for effectiveness.
Methods. Nine established GPs participated in a training programme consisting of twelve 3-h education sessions. Each session consisted of group discussion of videotaped patient interviews selected by GPs. Case discussion followed Lesser’s ‘problem-based approach’ criteria. Efficacy of training was assessed by measuring the pre–post change in accuracy in detecting psychiatric illness and changes in the interview skills of the GPs.
Results. Accuracy in detection of psychiatric illness increased significantly after training. Changes were seen in the interview style after training, such as use of open-ended questions and appropriate counselling in relation to problems presented by the patient. Physicians also improved their management skills by using a more negotiatory style and providing patients with supportive feedback. The way in which physicians gave advice and information to the patients improved significantly despite the fact that GPs were not instructed about how to give information to their patients.
Conclusions. Group training in problem-based interviewing utilizing video feedback is a robust, culturally transferrable model for improving the skills of established physicians. Our results suggest that training produces indirect effects that are the result of the teaching method rather than of explicit instructions. Further research is required to assess how to optimize the effect of educational interventions to ensure sustainability and maximal impact on measurable outcomes of care.
Keywords. Communication skills, primary care, training.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Management of common psychiatric disorders is one of the fundamental tasks of GPs. The average prevalence of fully fledged and subsyndromal ICD-10 disorders in primary care reported by the World Health Organization1 is 24 and 9%, respectively, and in the Italian2 replication of the study, the corresponding figures are 12.4 and 18%. Although frequently seen among primary care patients worldwide, common psychiatric disorders often require specific intervention strategies beyond the scope of routine general practice. GPs traditionally have sought to treat biological dysfunctions first and foremost. Indeed, because physical and mental disorders are interrelated, proper clinical management of the patients’ problems requires an approach that takes into account all factors, is goal oriented and does not expect instantaneous results.
In Italy, primary care is accessible and affordable; patients in need of help for psychological problems can access Community Mental Health Centres (CMHCs) directly, without being referred by their GPs. Nevertheless, the only connection between the CMHCs and primary care is through consultation, although the traditional consultation model has been proved to be ineffective.3 Shared management of mental disorders is an emerging trend that has been addressed recently by the Italian National Health Plan. As a result, the implementation of collaborative programmes between the CMHCs and primary care is now a priority. In order to improve the interface between psychiatry and primary care, a collaborative project involving psychiatrists from Community Mental Health Services, university researchers and GPs was implemented in Bologna. This project included a large epidemiological study2 and an outcome study comparing case review with usual care.4 One of the outcomes of this project was the identification of the need for physician-oriented specialized training in the management of psychiatric disorders seen in primary care. Designed to be more than a refresher course, a specialized training course was started, that focused training on providing GPs with the necessary skills to improve communication with their patients.
A number of studies have shown that the GP communication style has a significant impact on their ability to recognize and manage psychiatric disorders.5–11 Goldberg et al.6 showed that video feedback, when conducted in a controlled individual setting, does promote improvement in physician recognition abilities. No changes were seen, however, in interview management skills, and an excessive amount of time was required to conduct the training. Gask et al.7,8,12 showed that training programmes focused on video feedback conducted in a group setting and in combination with a reference model for management of psychological distress in general practice13,14 are effective in improving GP recognition and management skills.
Our training programme was based on methods set up by Gask et al.7,8 and had two main goals: (i) to test whether these training methods were replicable in Italy; and (ii) to evaluate the efficacy of training on two parameters: improved accuracy in GP detection of psychological distress in their patients and improvement in GP interview skills.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Training programme
A group of nine established GPs participated in a training programme organized by the Institute of Psychiatry at the University of Bologna. Their participation was voluntary and was part of the research programmes promoted by their professional organization. Physicians came from the greater Bologna area; their age range was between 40 and 55 years, and all had been in practice for at least 10 years. Two residents in psychiatry from the University acted as observers. The session co-ordinator was the first author (AS), an experienced psychiatrist who had attended a similar training course held by Dr L Gask in Yorkshire, UK.
Twelve 3-h education sessions were then scheduled over a 2-month period. An instructional video14 of the problem-based approach was shown at the beginning of the first session. Sessions consisted of group discussion of videotaped patient interviews selected by the physicians. Physicians were asked to record those cases of particular interest to them. Clinical problems observed repeatedly on the videotapes included patients with frank psychiatric problems, patients with somatic presentation of emotional distress and patients with complex emotional problems secondary to physical illness.
Case discussion followed criteria established in Lesser’s ‘problem-based approach’,13–16 an interview technique which facilitates GP understanding of the patient’s underlying problem(s) during consultation. During each session, the tape was stopped upon request to allow for discussion. Recognition of non-verbal cues was emphasized. GPs were encouraged to verbalize the impressions each patient gave them and to critique their colleague’s work in a constructive manner. As they watched each clinical situation, they were particularly encouraged to consider which emotions they were feeling. In accordance with what was first noted by Balint17 and later by Lesser,13 training focused on helping doctors to understand better what was actually happening during the interview, rather than immediately trying to figure out ‘what to do’.
Study design
A protocol was set up to evaluate training efficacy based on the model designed by Gask.7,8 We analysed two specific aspects.
GP accuracy in detection of psychiatric illness.. Prior to the start of the training sessions, each GP was asked to assess the severity of psychological distress (SPD) in a sample of 30 consecutive patients, using a 6-point scale (1 = no illness, 2 = possible, 3 = mild disorder, 4 = moderate, 5 = severe, 6 = very severe disorder).
Patients were asked to fill out the General Health Questionnaire18 (GHQ-12), a reliable screening instrument to assess the severity of psychological distress. The same assessments were repeated after training with a different set of 30 consecutive patients. The correlation between the GP’s assessment of SPD and the GHQ score was used as a measure of GP accuracy in detection of psychiatric disorder.
Changes in GP interview skills..
To identify changes that might have occurred in interview skills, each GP was asked to videotape 30 consecutive visits prior to the start of training. The same procedure was repeated for each GP after completion of training. Five pre-training and five post-training videotapes were chosen for each GP. The choice was based on consecutive videotapes reaching a pre-set cut-off value on two: a score of 
6 for the GHQ-12 and a score of 2 at the SPD scale filled out by the GPs.
The chosen interviews were then assessed independently by two raters using the last revision of the Interview Skills Scale.19 The scale allows one to rate a number of specific items, listed in Tables 1 and 2
, identified as representative of the doctors’ interviewing style. The items of this instrument are dichotomous or categorical and there is not an overall or a summary score.
|
|
To test the inter-rater reliability of this scale, 10 randomly selected pre- and post-training videotapes were examined independently by the two raters, who were blinded to the psychological condition of patients. The agreement between the two raters, measured by Cohen’s20 kappa, was 0.78, denoting a substantial level of agreement.21 All patients participating in the study provided written informed consent for the use of videotaped material. There were no refusals.
Statistical analyses
Spearman’s correlation coefficient (
) was used to measure the level of agreement between the patient’s assessment of psychiatric illness (GHQ-12) and that of the doctor (SPD scale). The coefficient was calculated for each GP both before and after training. Coefficients were then normalized by using Fisher Z-transformation,22 and analysis of variance (ANOVA) was carried out to assess the pre–post change in agreement between the patient’s and the doctor’s assessment of psychiatric illness. Changes occurring in interview skills, assessed with the Interview Skills Scale, were evaluated using the Wilcoxon test.23
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Accuracy in detection of psychiatric illness
A total of 251 patients before training and 264 after training filled out the GHQ-12 and were rated by the GP for the SPD.
The correlation coefficient between the GHQ-12 and the SPD scores was computed for each physician before and after training in order to estimate changes in accuracy of detection of psychiatric illness. Figure 1 shows the Spearman correlation coefficients for each GP, sorted by decreasing value at pre-training.
|
The mean Spearman correlation (
) between the SPD score and the GHQ score increased from 0.43 (pre-training) to 0.63 (post-training) (ANOVA F = 8.22, P = 0.021). Significant improvement in post-training can be seen in six of the GPs, who showed an increased correlation between the distress perceived by the patient and that detected by the doctor. Three of the GPs did not show any improvement.
Changes in physician interview skills
A total of 45 videotapes were rated before and 45 after training using the Interview Skills Scale. Pre- and post-training recognition, interview and management skills are detailed in Table 1
. Significant changes were seen in the interview style after training, together with a marked increase in the use of open-ended questions and appropriate advice in relation to problems presented by the patient. Physicians also improved their management skills by using a more negotiatory style and providing the patient with supportive feedback. The way in which physicians volunteered information and advised their patients changed dramatically after training (Table 2). Careful examination of the changes in communication style showed vast improvement in information provided ‘with an explicative link to both symptoms and diagnosis’ (12.2 at pre-training to 57.1 post-training, P < 0.001). Furthermore, there was a marked decrease in situations where doctors gave little or no information (10.2 at pre-training to 0 post-training, P < 0.001), or limited or partial information (see Table 2). Similar encouraging results can be seen regarding how doctors gave advice to patients (specific advice from 31.0 to 63.0; P = 0.05). The way in which physicians gave advice and information to the patients improved significantly despite the fact that no specific method for providing patients with information and advice was taught during the training. It is of note that the three GPs mentioned above who did not show any improvement in detection skills had a clear benefit from training. Indeed, their management style after training was less dispensory and more oriented to establish treatment goals with patients after providing them with clear information, advice and examples.
Finally, Table 3 indicates post-training changes in ‘utterances’. Open-ended questions increased, whereas the number of ‘closed’ questions remained unchanged. After training was completed, there was a recognizable increase in the physicians’ tendency to ‘probe’ by asking pertinent psychological and social questions, as compared with pre-training.
|
Data regarding medication prescribed have not been reported, as physicians stopped videotaping at the conclusion of the consultation, prior to the physical exam. As prescriptions are generally written once the medical exam is completed, these data are unavailable.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Interview skills training is a powerful tool for improving the ability of established physicians to recognize psychiatric disorders. We found that on average, the correlation between the ratings of psychological distress by patients and physicians increased from 0.43 to 0.63 (ANOVA F = 8.22, P = 0.021, in a group of nine doctors). Our findings are strikingly similar in order of magnitude to those of Gask et al.8 (correlation from 0.41 pre-training to 0.59 post-training; F = 11.9, P < 0.005, in a group of 14 doctors). While in the work conducted by Gask the improvement in detection of patient distress was greater for physicians with poor pre-training performance, in our study improvement was independent of that factor. Interview management skills also improved significantly, paralleling the results of Gask et al.7,8 even among those who did not improve in detection skills. Although a similar methodology was used in the two studies, the Italian and the English cultural and organizational contexts differ from one another. Thus, this cross-cultural validation is relevant because it provides sound evidence that training programmes should be promoted and instituted by virtue of their ability to improve performance of physicians at any level of expertise. This is especially important in Italy, where this type of programme is not the norm.3
Considering our findings in the light of the recommendations of a recent comprehensive review of the literature on educational interventions in primary care,24 we note that our programme may be effective because it ties learning to real clinical practice and utilizes interactive methods.
The methodology used seems to have an inherent validity. In this regard, when comparing how doctors gave information to patients before versus after training, dramatic improvement was seen in information provided, with an explicative link to both symptoms and diagnosis, while there was a marked decrease in situations where doctors gave either no information or partial information to patients (Table 2). This cannot be attributed to overt instruction during the training course; GPs were not instructed about how to give information to their patients. Indeed, these results suggest that training produces indirect effects that are both specific and significant, and that these effects are a result of the teaching method rather than of explicit instructions.
As early as 1957, Michael Balint17 highlighted the importance of what transpires during the doctor–patient consultation, but stressed that ". . . the events we are concerned with are highly subjective and personal, often hardly conscious, or even wholly beyond conscious control; also as often as not there exists no unequivocal way of describing them in words". Thirty years later, Gask and McGrath12 emphasized that it is not only possible, but essential, to define operationally what happens during the doctor–patient consultation. While it may not be possible to define unequivocally what transpires during doctor–patient communication, our study indicates, in line with the literature5–14,25–27 that, if specific training methods are used, GPs tend to increase their understanding of what goes on during a patient consultation. Therefore, we believe that effective GP training consists of assisting the physicians to understand what really happens during the interview and enabling them to incorporate this awareness into practice.
We identify four key methodological points in order to achieve successful training.
- As shown by Maguire et al.,28 Verby et al.29 and research cited previously, video feedback is an effective tool for improving GP interview skills. In our experience, video feedback promoted group discussion of cases by eliminating sophisticated terminology, and facilitating instead the use of a shared vocabulary. This is necessary for increasing physician awareness of his/her own behaviour in relation to patients, and reduces the need to refer to models and theories culled from experiences developed in other contexts.
- A case discussion model which respects the unique qualities of general practice and overcomes the initial difficulties of working in a group must be implemented. The problem-based approach13–16 enables physicians to discuss the emotional aspect inherent to the doctor–patient consultation, and assists the group in attaining sufficient shared awareness of this aspect of care. While it may not be possible to codify what occurs in each consultation, a well-formulated approach increases the likelihood of improving GP ability to understand the nuances of a patient consultation.
- The group setting constitutes another important methodological element. Not only a means to streamline the excessive time commitment required by an individual setting,6 a group setting is indispensable for achieving effective feedback. GPs tend to deem it necessary to apply diagnostic and therapeutic paradigms outside of the context in which the disorder is being treated. Through group experience, doctors gain technical awareness of the problems characterizing their clinical practice and learn how to deal with those problems in a way that is more tailored to the nature of their practice.
- The work of the co-ordinator consists of allowing the group to define the problems that occur in the videotaped interviews. To that end, and as Balint suggested,17 the training co-ordinator must keep the discussion focused on ‘understanding’, as the tendency of the group is to think first about knowing what to ‘do’. This aspect of the co-ordinator’s work is complex and thus requires appropriate training.30
We acknowledge that it has not been, so far, convincingly demonstrated that training programmes, on their own, produce change in measurable clinical outcomes in randomized controlled trials.24 Indeed, both the nine GPs who pioneered this training programme and all subsequent participants asked to repeat the course and to attend it on a regular basis as part of their continuing education programmes. Should this happen in the context of an organizational change, training programmes seem to have the potential eventually to induce significant effects on health outcome.
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Group training in problem-based interviewing utilizing video feedback is a robust, culturally transferrable model for improving the ability of established physicians to recognize psychiatric disorders. Our results suggest that training produces indirect effects that are both specific and significant, and that these effects are a result of the teaching method rather than of explicit instructions. Further research is required, however, to evaluate how to optimize the effect of educational interventions to ensure sustainability and maximal impact on measurable outcomes of care.
|
Acknowledgments |
|---|
We gratefully acknowledge the skilful collaboration of Drs G Carta, AR Scaramelli and M Sciulli in rating the videotapes.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
1 Üstün TB, Sartorius N. Mental Illness in General Health Care. Chichester (UK): Wiley and Sons; 1995.
2 Berardi D, Berti Ceroni G, Leggieri G, Rucci P, Ustun TB, Ferrari G. Mental, physical and functional status in primary care attenders. Int J Psychiatry Med 1999; 29: 133–148.[Web of Science][Medline]
3 Berardi D, Berti Ceroni G, Leggieri G et al. The collaborative project: Bologna–Psychiatry and General Practice. In Preventing Mental Illness: Mental Health Promotion in Primary Care. New York: Wiley; 1998: 262–266.
4 Berti Ceroni G, Rucci P, Berardi D, Berti Ceroni F, Katon W. Case review vs. usual care in primary care patients with depression: a pilot study. Gen Hosp Psychiatry 2002; 24: 71–80.[CrossRef][Web of Science][Medline]
5 Marks JN, Goldberg DP, Hillier VF. Determinants of the ability of general practitioners to detect psychiatric illness. Psychol Med 1979; 9: 337–353.[Web of Science][Medline]
6 Goldberg DP, Steele JJ, Smith C, Spivey L. Training family doctors to recognise psychiatric illness with increased accuracy. Lancet 1980; 6: 521–523.[CrossRef]
7 Gask L, Goldberg DP, McGrath G, Millar T. Improving the psychiatric skills of established general practitioners: evaluation of a group teaching. Med Educ 1987; 21: 362–368.[Web of Science][Medline]
8 Gask L, Goldberg DP, Lesser AL, Millar T. Improving the psychiatric skills of the general practice trainee: an evaluation of a group training course. Med Educ 1988; 22: 132–138.[Web of Science][Medline]
9 Millar T, Goldberg DP. Link between the ability to detect and manage emotional disorders: a study of general practitioner trainees. Br J Gen Pract 1991; 41: 357–359.[Web of Science][Medline]
10 Roter D, Hall J, Kern E, Barker R, Cole K, Roca R. Improving physicians’ interviewing skills and reducing patients’ emotional distress. Arch Intern Med 1995; 155: 1877–1884.
11 Giron M, Purificaciòn MA, Puerto-Barber J, Sànchez-Garcia E, Gòmez-Beneyto M. Clinical interview skills and identification of emotional disorders in primary care. Am J Psychiatry 1998; 155: 530–535.
12 Gask L, McGrath G. Psychotherapy and general practice. Br J Psychiatry 1989; 154: 445–453.
13 Lesser AL. Problem based interview in general practice: a model. Med Educ 1985; 19: 299–304.[Web of Science][Medline]
14 Lesser AL. Problem Based Interviewing in General Practice. An Instructional Videotape. Tavistock Publications; 1986.
15 Lesser AL. The psychiatrist and family medicine. A different training approach. Med Educ 1981; 15: 398–406.[Web of Science][Medline]
16 Gask L, Boardman AP, Standart S. Teaching communication skills: a problem-based approach. Postgrad Educ Gen Pract 1990; 2: 7–15.
17 Balint M. The Doctor, His Patient and the Illness. London: Pitman Medical Publishing Co. Ltd; 1957.
18 Goldberg DP. The Detection of Psychiatric Illness by Questionnaire. London: University Press; 1972.
19 Kaaya S, Goldberg D, Gask L. Teaching the skills of ‘reattribution’ a replicated study. Med Educ 1992; 26: 138–144.[Web of Science][Medline]
20 Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas 1960; 20: 37–46.[CrossRef][Web of Science]
21 Nunnally JC. Psychometric Theory, 2nd edn. New York: McGraw-Hill; 1978.
22 Fisher RA. On the probable error of a coefficient of correlation deduced from a small sample. Metron 1921; I: 3–32.
23 Siegel S. Non-parametric Statistics for the Behavioural Sciences. New York: McGraw-Hill; 1956.
24 Hodges B, Inch C, Silver I. Improving the psychiatric knowledge, skills and attitudes of primary care physicians 1950–2000. A review. Am J Psychiatry 2001; 158: 1579–1586.
25 Goldberg DP, Jenkins L, Millar T, Faragher EB. The ability of trainee general practitioners to identify psychological distress among their patients. Psychol Med 1993; 23: 185–193.[Web of Science][Medline]
26 Davenport S, Goldberg DP, Millar T. How psychiatric disorders are missed during medical consultations. Lancet 1987; ii: 439–441.[CrossRef]
27 Gask L, Morriss R. Training general practitioners in mental health skills. Epidemiol Psichiatr Soc 1999; 8: 79–84.[Medline]
28 Maguire P, Roe P, Goldberg DP, Jones S, Hyde C, O’Dowd T. The value of feedback in teaching interviewing skills to medical students. Psychol Med 1978; 8: 695–704.[Web of Science][Medline]
29 Verby JE, Holden P, Davies RH. Peer review of consultations in primary care: the use of audio-visual recordings. Br Med J 1979; 1: 1686–1688.
30 Gask L, Usherwood T, Standart S. Training teachers to teach communication skills: a problem-based approach. Postgrad Educ Gen Pract 1992; 3: 92–99.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||