Qualitative study of an educational intervention for GPs in the assessment and management of depression

Abstract

Background

Previous research has not shown any significant health gain for patients as a result of providing education about depression for GPs. Reasons for this, however, are unclear.

Aims

To explore relationships between process and outcome in the setting of a randomised controlled trial of a complex educational intervention designed to provide GPs with training in the assessment and management of depression.

Design of study

Qualitative study utilising semi-structured interviews.

Setting

General practice in the northwest of England.

Method

Semi-structured interviews with 30 GPs in Liverpool and Manchester who participated in a randomised controlled trial.

Results

Three major barriers to the effectiveness of the intervention were identified: the lack of the GP's belief that he/she could have an impact on the outcome of depression, the appropriateness of the training, and the organisational context in which doctors had to implement what they had learned.

Conclusion

Attitudes toward treating depression may need addressing at a much earlier point in medical education. If students are introduced to a biosocial model of depression at an early stage, they may feel more hopeful about their ability to intervene when faced with patients who exhibit significant degrees of functional disability in the context of apparently socially determined disorders. Postgraduate interventions should be tailored to the treatment of depression as a common chronic condition and be focused at the level of the organisation, not the individual practitioner.

INTRODUCTION

Recent research into the effectiveness of educational interventions for GPs regarding the detection and management of depression in primary care has produced disappointing results,^1–3 although the absence of simultaneous qualitative enquiry has made the reasons for the failure of these interventions unclear. We previously developed an educational intervention that we demonstrated to have an impact on acquisition of new skills.⁴ However, in a randomised controlled trial we were unable to demonstrate an impact on patient outcome.⁵ During the trial we conducted a qualitative evaluation in order to explore in detail how the educational intervention worked (or did not work), to identify potential barriers to changing professional behaviour by this method and to inform the design of future complex educational interventions.⁶

METHOD

Design of study

This was a qualitative study utilising semi-structured interviews. The design of the cluster randomised controlled trial and detailed descriptions of participants has been published in detail elsewhere.⁵

Following research ethics committee approval at all sites, all GP principals in Manchester, Liverpool, Sefton and Wirral were approached by letter. Recruited GPs (n = 38) were assessed on recognition of psychological disorders,⁷ attitudes to depression,⁸ and prescribing patterns (assessed by Prescribing Analysis and CosT data for the preceding 3 months). They also provided information in a brief semi-structured interview at baseline on previous experience of psychiatry and communication skills training. They were then randomised to receive training at baseline or the end of the study. We delivered 10 hours of training over 5 weeks; this consisted of brief factual presentations, role-play, and video feedback (Box 1).

Box 1. The educational intervention.

The Assessment and Management of Depression in Primary Care

A 10-hour course.

Aims of the course

To improve the assessment and management of depression in general practice by effective utilisation of pharmacological, physical, and social interventions that are realistic within the confines of the consultation.

Training methods

The key focus of the course is on acquisition of appropriate clinical skills. Each of the five 2-hour sessions consists of:

• ▸a brief presentation/lecture on each topic;
• ▸viewing of specially developed videotaped material;
• ▸the opportunity for each GP to role-play consultations as both the GP and a depressed patient in order to develop specific microskills;
• ▸these role-plays are videotaped and used for videofeedback in small groups; and
• ▸each GP receives written material to support each session.

Course timetable

• ▸Week one: assessing depression.
• ▸Week two: negotiating the treatment contract and drug treatment of depression.
• ▸Week three: problem-solving therapy and social interventions.
• ▸Week four: the question of suicide.
• ▸Week five: cognitive and behavioural skills.

GPs enrolled in the study were invited to take part in a longer face-to-face interview after the 3-month quantitative outcome data had been analysed.

Sample

After sending each GP a letter requesting a meeting with them to discuss the interim findings of the study, in-depth semi-structured interviews with 30 of the 38 GPs originally recruited were conducted. This number consisted of the 14 GPs still available in Manchester (one GP was on long-term sick leave and three had left the area) and 16 from a possible 20 in Liverpool: the remaining four declined to be interviewed further or were difficult to contact. At the time of the interview, 17 of the 30 had received the intervention; 13 were due to receive training at the end of the study. The split by city meant that of those in Manchester, nine had received the training (the intervention group) and five had not (the control group); in Liverpool there were eight doctors in each group.

Interviews

The interviews were all carried out at the GP's surgery with the exception of one, which took place at his home. The interview schedule (outlined in Box 2 and given in detail in Supplementary Box 1) contained a range of open-ended questions covering four areas: GPs' perceptions of what depression is, their feelings about treating depressed patients, their thoughts on issues surrounding training, and their responses to and explanations of the quantitative findings of the study at 3 months.

Box 2. Outline of the interview schedule.

• ▸
  What does depression mean to you in practice?

  1. Is depression treatable?
  2. What is it you are treating?
  3. Do you think there's a certain ‘type’ of person you ‘diagnose’ as depressed or who you treat for depression? What type? Why?
• ▸What is it like to treat depressed patients?
• ▸Do you find depression difficult to treat? Why?
• ▸Can training influence practice? How? If not, why not?
• ▸Why did you choose to get involved in this study/training? (intervention group only)
• ▸What do you think of the quantitative findings of this study?

The interviews took place when all 3-month quantitative data had been collected; this data was made available in pictorial form for comment as part of the interview. Due to the finding of differential effect by city in the interim findings, the Manchester GPs were shown a histogram of the Hamilton Rating Scale for Depression⁹ results at 3 months for Manchester patients only, and Liverpool GPs were shown the corresponding findings relating to Liverpool patients. In this way the Manchester GPs were commenting on a graph depicting the training resulting in a positive effect on clinical outcome at 3 months, whereas the Liverpool GPs were responding to a pictorial representation of negative findings. Neither group of doctors was aware of the conflicting findings in the other city.

How this fits in

Educational interventions for GPs on depression do not lead to improved outcome for patients. There is therapeutic pessimism about the chances of improving the lives of people who suffer from depression in primary care. The negative attitudes of doctors may need to be addressed much earlier in medical training and future interventions need to address the organisational barriers that prevent doctors from putting into practice what they have learned.

Each interview was audiotaped with the GP's permission and lasted on average 30–40 minutes. Following the interview, notes were made on the time and setting of the interview, together with the interviewer's comments on how well she knew the GP beforehand and her views on how they had related to each other. Her immediate reactions in terms of main themes, perceptions on how the interview went, and how she had felt at the time were noted. This is in accordance with the theoretical view that it helps in analysis of the data to place a personal interview in perspective in this way.¹⁰ The taped interviews were transcribed verbatim within 3 days of the interview.

Analysis

Interviews were analysed by utilising the grounded theory method.¹¹ The transcripts were coded and codes in each interview were compared with those in other interviews to create broader categories linking codes across interviews (the technique of ‘constant comparison’). Attribution theory¹² was used as a means for understanding the data in terms of the perspectives of the GPs and their interactions with their world. In accordance with grounded theory, the interview schedule was amended during the process of interviewing the GPs in response to the identification of new categories or themes produced during the concurrent coding of interview transcripts. Initial analysis was undertaken. Collaborative interpretation of data and verification of emergent categories in the data was undertaken to secure an appropriately reflexive mode of analysis.¹³

RESULTS

Several different themes emerged from the data. With the exception of knowledge of the training course, no clear differences were observed between themes in the interviews of the trained and untrained doctors. In this study we focused on those findings that help us to understand how the intervention did/did not work. These fell into three groups:

• whether GPs can have any impact on the outcome of depression;

• the irrelevance of mental health education to real primary care; and

• the difficulty of putting what is learned into practice.

Can GPs really have any impact on the outcome of depression?

Nine out of the 12 Manchester doctors were surprised by the positive results represented on the graph shown to them. Some believed improvement in the kind of depression they see to be primarily governed by factors outside their control:

‘… maybe only 25% of [recovery is] what we're actually doing with the patient, and it may be that they're getting better because of the other 75%.’ (GP 1, attended training.)

The GPs' accounts of depression reflected its social, rather than biological, nature in practice:

‘Depression … is an over-valued concept … a lot of people make depression an illness, it's not an illness, it sometimes is an illness but it isn't always an illness. I mean people can have bad moods … for lots of their lives … People can't get jobs or people didn't grow up the way they thought they should and … I don't like medicating context at all.’ (GP 16, control group.)

Many of the doctors (but not all), whether trained or not, perceived their role, therefore, to be limited in terms of how it might affect the outcome for many patients with depression. External social factors and the patients' personalities were considered to be as influential in their recovery, if not more so, than GPs own interventions:

‘We're just a very small part … a lot of it is what's going on in their lives.’ (GP3, attended training.)

Some, however, were very positive about the impact of their treatment of depression:

‘It's usually very satisfying because most of them do get better. It's nice to see the change.’ (GP 12, attended training.)

The irrelevance of mental health education to real primary care

Of the 30 GPs interviewed, the most common formal experience of previous training in psychiatry was, at the most, a senior house officer post of between 3 and 7 months (14 doctors). A further five had more extensive experience — one worked on a psychiatric unit as a clinical assistant, another had an interest in hypnotherapy, one had done a short psychiatry course as a GP trainee and two had been members of a Balint group. The remaining 11 had either received no training in psychiatry or none since that received as an undergraduate. Twenty-one out of the 30 (70%) had no previous training in any form of counselling and a third reported having received no postgraduate training in communication skills.

Not only had many of the GPs received little or no specifically relevant formal training since their time as a medical student, but the majority spoke of the irrelevancy of the undergraduate medical training they had received to general practice work. They commented in particular about the hospital-based nature of any practical experience they had in psychiatry, and how inappropriate this was for their practice as GPs.

‘The skills that you develop during the hospital phases of GP training are to large extent useless in general practice.’ (GP 28, attended training.)

‘On my rotation I did 6 months' psychiatry. It was completely and absolutely irrelevant. It was a consultant who was very organic and I had to do a weekly ECT clinic which I spent most of my time trying to get out of doing … and learned absolutely bugger all!’ (GP 15, attended training.)

How did this influence their views of the course?

Some doctors reported very real and positive differences in their management of depression as a result of attending the course. Many stressed the confidence-giving effect of the training as very important:

‘Training of any sort gives you confidence. Just taking time out away from seeing patients and thinking about an issue is helpful.’ (GP 4, control group.)

However, although the doctors generally considered that they had acquired new skills, this was not necessarily seen as the key element:

‘If you wanted my opinion I'd have said that having 10 hours of just thinking about depression was probably the single most important thing, and the various skill bits maybe less so.’ (GP 1, attended training.)

Additionally, just as their undergraduate medical training was perceived as irrelevant to their work in general practice, so some of the doctors considered that this intervention also failed to address the kind of depression they faced in their daily work and the changes necessary for its effective management:

‘I feel that a lot of the problems, people we see with depression in general practice, don't fit the criteria we were talking about on the course. That to talk about treating an episode of depression is not necessarily a realistic way to see things.’ (GP 7, attended training.)

Thus, the educational intervention was perceived as addressing management of ‘acute’ and discrete episodes of depression rather than the chronic or relapsing nature of depression associated with personality and social problems, which is increasingly recognised in primary care.¹⁴

The difficulty of putting what you learn into practice

All of the doctors considered that they would face some degree of difficulty in putting their new skills into practice in the busy setting of British primary care:

‘I mean most of us would want to do well but sometimes you just can't. I can imagine that often although we'll know what we're supposed to be doing we don't actually put it into practice.’ (GP 13, attended training.)

There was acknowledgement of an element of ‘old habits die hard’ in transferring new skills to everyday work. The fact that it is imperative to practise new skills immediately if there is to be any hope of a long-term effect, and that any effect is likely to take some time to become regular practice was reflected in the GPs' accounts. The main obstacle to changing their behaviour and so transferring their new skills and knowledge to their practice was, from the GPs' viewpoint, the structural constraint of time. Many reflected on how the time-limited nature of the consultation resulted in many good intentions regarding what is learned in training being curtailed, or how ‘time’ controls and dictates their day-to-day practice:

‘I need time and I'm not paid any extra money for doing this job. Good medicine in practice is one thing but resources available are a different thing.’ (GP 17, attended training.)

DISCUSSION

Summary of main findings

Three major barriers to the effectiveness of the intervention were identified:

• the lack of belief by the GPs that they could have an impact on the outcome of depression;

• the appropriateness of the training; and

• the organisational context in which doctors had to implement what they had learned.

Strengths and limitations of this study

The doctors who took part in this study were self-selected and were more likely have positive views about both depression and training in mental health skills than the general population of GPs. Therefore our findings of substantial barriers to the effectiveness of educational interventions for depression in this group are likely to be a replicated a fortiori among GPs in general.

GPs often give quite brief interviews and it can be problematic getting an in-depth perspective in a short period of time. Nevertheless, each interview lasted on average 30–40 minutes, and the majority of the interviews were relaxed and informal, allowing an exploration of the value and mechanisms behind the training. The specific findings cannot be considered to be generalisable to the wider population of doctors, but they may offer some insight into the range of difficulties involved in changing the behaviour of doctors in similar interventions.

Possible reasons why the training intervention did not work

How GPs ‘arrive’ at training interventions, in terms of their previous education and experience, will inform their interaction with the intervention and play a part in their evaluation of it. Medical training creates specific medical ‘dispositions’,^15,16 which remain with doctors and are influential throughout their professional lives. For many GPs their practice is considerably influenced by the training they received as medical students — training that may not only have taken place some time ago but, more importantly, is instrumental in creating a ‘biomedical’ disposition or perspective. Social and psychological matters may not be explicitly ignored, but they are still viewed as separate from the real object of medical practice.¹⁷

The power of educational and/or organisational interventions in primary care to improve the outcome of care for depression must be viewed in the light of doctors beliefs about the possibility of actually having the power to effect change in an illness that does not obey ‘biomedical’ rules and for which outcome is undoubtedly strongly influenced by social and psychological factors.¹⁸ Educational interventions during postgraduate training in general practice will struggle to challenge views about depression acquired earlier in medical training, and reinforced over many years in both lay and professional lives.¹⁹

This study reveals there may have been problems with the content of the training, in terms of its relevance to general practice. It might also have been difficult for doctors to utilise what they had learned in the context of busy everyday practice. However, another possible explanation that arises from this data is that the doctors' belief in the role that they can play in bringing about real change in patients' lives is limited. This may function as an attitudinal barrier, preventing the doctors from utilising the training effectively. We have already established with this group of GPs that their attitudes towards depression predicted their diagnostic response. That is to say that they were more likely to make a diagnosis of depression if they felt comfortable about treating it.⁷ We now wish to suggest two additional important dimensions to the issue of doctor's attitudes towards depression.

First, doctors' tendency to understand their patients as passive beings or dysfunctional machines²⁰ needs to be challenged in the context of a more active view of patients as people engaged in living their lives.²¹ Second, if the limitations of the role of doctors in influencing the outcome of depression are fully acknowledged, then GPs may paradoxically be encouraged to concentrate on the small, but nevertheless significant, contribution they can make to improve their patient's lives.

Implications for education and future research

Our findings support the view that postgraduate educational interventions alone seem unlikely to improve quality of care for depression. It may be more realistic to situate them as part of a broader range of interventions that address how care is organised for depression as a ‘chronic’ illness in primary care.¹ Commitment, both from doctors and policy makers, to addressing organisational barriers needs to be forthcoming if both doctors and, therefore, ultimately patients are to benefit from interventions to improve quality of care.

Educational interventions that do not address the ways that patients with symptoms of depression actually present in general practice, or review the ways that their care is organised, are unlikely to be effective in achieving change in patient outcomes. However, to achieve this we also need to challenge the potentially negative attitudes or ‘dispositions’ that doctors develop towards treating depression, perhaps as a result of exposure to ‘medical’ models of mental illness in early training that do not fit with their later experiences. This challenge should be made much earlier in their training than at the postgraduate level, the point at which most such interventions have been aimed to date. If students are introduced to a biosocial model of depression at an early stage, they may feel more hopeful about their ability to intervene later on, when faced with patients who exhibit significant degrees of functional disability in the context of apparently socially determined disorders.

Supplementary information

Additional information accompanies this article at http://www.rcgp.org.uk/journal/index.asp

Funding body

Department of Health ‘R & D initiative for Mental Health’ (NMH11C)

Ethics committee

North Manchester (ERC96/31), Central Manchester (CM/96/099), South Manchester (ERP95/099), South Sefton (PC/LC/EC.110.96), Liverpool (AAG/PM/96/136) and Wirral (57/96) Local Research Ethics Committees

Competing interests

Linda Gask has received funding from pharmaceutical companies for research into depression and speaking at educational meetings about depression in primary care. The other authors have no competing interests to declare