Primary care

Morale among general practitioners: qualitative study exploring relations between partnership arrangements, personal style, and workload

Guro Huby, senior research fellow ^a, Marian Gerry, independent consultant ^c, Brian McKinstry, director of postgraduate medical educational research, SE Scotland ^d, Mike Porter, senior lecturer ^b, Johnstone Shaw, associate adviser in general practice, SE Scotland ^d, Robert Wrate, honorary senior lecturer ^e. 

^a Working Minds Research/Department of Community Health Sciences, Primary Care Research Group, University of Edinburgh, 20 West Richmond Street, Edinburgh EH8 9DX, ^b Department of Community Health Sciences, General Practice, University of Edinburgh, ^c Family Business Facilitation, St Ragnvald's Street, Kirkwall, Orkney KW15 1PR, ^d Lister Institute, 11 Hill Square, Edinburgh EH8 9DR, ^e Department of Psychiatry, University of Edinburgh, Young People's Unit, Royal Edinburgh Hospital, Edinburgh EH10 5HF

Correspondence to: G Huby guro.huby{at}ed.ac.uk

	Abstract

Top Abstract Introduction Methods Results Discussion References

Objectives: To explore general practitioners' experiences of wellbeing and distress at work, to identify their perceptions of the causes of and solutions to distress, and to draw out implications for improving morale in general practice.
Design: Three stage qualitative study consisting of one to one unstructured interviews, one to one guided interviews, and focus groups.
Setting: Fife, Lothian, and the Borders, South East Scotland.
Participants: 63 general practitioner principals.
Results: Morale of general practitioners was explained by the complex interrelations between factors. Three key factors were identified: workload, personal style, and practice arrangements. Workload was commonly identified as a cause of low morale, but partnership arrangements were also a key mediating variable between increasing workload and external changes in general practice on the one hand and individual responses to these changes on the other. Integrated interventions at personal, partnership, and practice levels were seen to make considerable contributions to improving morale. Effective partnerships helped individuals to manage workload, but increasing workload was also seen to take away time and opportunities for practices to manage change and to build supportive and effective working environments.
Conclusions: Solutions to the problem of low morale need integrated initiatives at individual, partnership, practice, and policy levels. Improving partnership arrangements is a key intervention, and rigorous action research is needed to evaluate different approaches.

	Introduction

Top Abstract Introduction Methods Results Discussion References

Morale among general practitioners is a current concern in the United Kingdom because of difficulties with recruiting and retaining the workforce needed to meet the targets of a primary care led NHS. ^{1 2} Work strain for British general practitioners increased after the introduction of the 1990 general practitioner contract,³ but satisfaction subsequently improved and stress from night visits fell.^4-6 Recent research into stress and malaise in general practitioners has examined individual experience of work and how organisational contexts shape this experience. ^{7 8} Firth-Cozens emphasised the importance of a well functioning team for reducing stress and improving performance,⁹ and Calnan et al showed that people respond differently to similar working conditions.¹⁰ More work needs to be done to link the experience of individual general practitioners with the practice context in which they work and with the wider political, economic, and social context of health service reform.

We report on the qualitative components of a multi-methods study of general practitioners' experiences of their work in South East Scotland. We aimed to explore general practitioners' experience of wellbeing and distress at work, to identify their perceptions of the causes of and solutions to distress, and to draw out implications for improving morale in general practice.

	Methods

Top Abstract Introduction Methods Results Discussion References

The qualitative study consisted of three phases of interviews and focus groups, each building on the previous phase, involving 63 general practice principals (box 1). To ensure rigour, the analysis included attention to ways in which the three phases of the research process influenced the substantive themes.¹¹

	Results

Top Abstract Introduction Methods Results Discussion References

The results are presented as an account of how the key themes emerged and were explored in the light of the relation between the research process and findings through the three phases.

Phase 1 interviewspartnership arrangements and personal style
The first phase consisted of open ended individual interviews. Respondents described a range of experiences of general practice, from deep distress to high levels of satisfaction. Contrary to expectations, ^{1 2} few accounts told of the pressures coming from increased workload and patient demand, although they varied in emphasis. Rather, in 13 out of 16 interviews, a strikingly similar story emerged of how experience of work was linked to partnership arrangements. This was a factor in accounts of distress and accounts of satisfaction. For example, one respondent had been off sick because of stress, which he attributed to the excessive demands of a full time post and inability to reduce his sessions:

"I went back to work about a year ago, but it took me about nine months to get back up to the nine sessions a week and my practice were pressurising me to get back up to nine sessions . . . because doing fewer sessions was never an option, partly because, you know, it is basically a male practice. There is one female partner who has a very poor financial deal in the practice . . . and I think if I were to drop a session or two it would make a farce of the whole inequitable system that is already in place . . . I was afraid that I would be dumped, I would be got rid of, and therefore I was saying this is just a temporary episode, I will be fine, and you couldn't be as open as you would want to" (respondent 3, phase 1).

A second example illustrates a more positive work experience, where the partnership had created structures that helped the respondent to contain her disposition towards greater patient involvement:

"We have had a look at the construction of surgeries, so I have more appointments kept back for the day so that I have got a fairer mix, because the biggest problem with time management is heartsink patients and because I tend to get more than my fair share of them, then a lot of my appointments are kept back for the same day so that I have got the variety and that helps with my time management, because a lot of the things that come in on the day are quick patients" (respondent 7, phase 1).

These two excerpts illustrate respondents' accounts of how partnership arrangements affected, and were affected by, the business of running a practice. How partnerships accommodated differences in working styles and speed of consultation were important, as were procedures for decision making within the partnership, seniority and sex of partners, and part time working. The three accounts that varied from this pattern included partnership arrangements as important, but the respondents described the role of partnerships in their lives differently from the majority. For example, one respondent said that the main problem of morale for him was a lack of career development, which, in turn, he saw as linked to his obligations to his partnership.

The second excerpt above also illustrates how partnership dynamics interacted with personal style. It was immediately clear from comparing the first interviews that people responded differently to similar issues within their partnerships, with very different outcomes. Partnership relations also affected the way respondents functioned at a personal level:

"Previously we couldn't work together in the [previous] practice; it just didn't work and it was very very stressful, and I didn't cope well with that at all . . . Yes, my self esteem, my confidence were at rock bottom and awful, non-existent . . . Now I wouldn't worry. People take me for who I am. If they don't like me, that's fine . . . I probably wouldn't like them. So that's fine" (respondent 8, phase 1).

Phase 2 interviewsworkload, partnership arrangements, and personal style
In the phase 2 interviews, we outlined the findings from phase 1 and explicitly explored partnership arrangements in a guided interview schedule. All respondents therefore told stories about how partnership arrangements shaped their working experience. However, when we asked explicitly how partnership arrangements affected morale, respondents tended to react by emphasising workload as a significant factor. Five of the 10 respondents were clear that this was more important than partnership arrangements. Workload factors included increased directives and paperwork, patient demand, and transfer of caseloads from secondary to primary care without an increase in resources:

"Pressure on us to reduce our ability to make clinical decisions. We are bombarded constantly with guidelines on everything. Restrictions on our prescribing choices. Doing tasks we see as useless, like, one we all wish we could be rid of is the over 75 annual health checks, which we see as totally useless and a waste of time . . . What else? Constant change. Higher patient expectations. Pressure to audit absolutely everything, which I agree is a good idea, but it is actually quite difficult to do it with no extra resources in terms of money to spend on staff to help" (respondent 4, phase 2).

Partnership arrangements nevertheless appeared as important, albeit implicitly so, in all accounts. A typical example was one respondent who emphatically criticised the emphasis on partnership arrangements in the interview, but only after giving a 20 minute unsolicited account of how difficulties in a previous partnership had led him to change practice.

It was also clear that personal style interacted with workload factors and partnership relations to create a particular work experience, but again relations were not made explicit. One respondent described his biggest problem as the open ended commitment he had to his patients and then went on to say:

"I don't know why I worry so much. If ever I got 20 patients needing a house call one day, the rest of the partners would say, look come on, we will take half of them for you, stop worrying about them. Fear of the unknown to some extent, what is going to come in and see you in the afternoon" (respondent 7, phase 2).

Workload was thus an important factor influencing morale, and it related to partnership dynamics and personal styles in intricate ways. The part of this whole that was brought into focus depended on the type of interview schedule. In the individual interviews, the interrelations between factors were not made explicit.

Phase 3focus groups

Making connections
The focus groups brought out these interrelations more clearly. Before the meetings, we sent participants a fictitious scenario about a practice that was outwardly successful, but which on closer inspection was somewhat dysfunctional in that partners did not communicate with each other and several partners had personal difficulties that had not been dealt with. We asked participants to discuss the problems the practice was facing and to suggest causes and possible solutions at personal, practice, and wider health service levels. As in the phase 2 interviews, several participants challenged what they saw as an implicit assumption that partnership arrangements were central in creating low morale and tabled workload issues as equally or more important.

Links between partnership and workloadtime and "space"
Out of this process links emerged between workload and partnership arrangements and their effect on morale that were absent from individual accounts. Practices that had equitable and inclusive partner and practice relationships managed workload better than practices in which people did not work well together:

"I think you are right, having practice meetings; the two common denominators in the two disastrous practices I was in, neither had any form of meeting at all, and I think the forum we have for the meeting on a Tuesday lunchtime, everybody does meet together and every so often we have all the staff in as well, which does allow people to air their gripes and glooms" (group 5).

"We have a problem in that we try to give continuous surgery first thing in the morning (to cope with increased patient demand). Someone is doing calls, then he is doing surgery when the others are out on calls, so there is something going on all the time and very often with the extras we see in our five minute, three minute slots, goes from 10 o'clock to 11 o'clock, which cuts out that time when we used to have our coffee break and we just don't get together in the same way. Also, I am the only full-timer in the practice, one is associate adviser, one does work for the local healthcare cooperative, two are part-time partners so they are not in all the time, there is one day when one partner is not in, this sort of thing. We don't see them as much as we used to" (group 3).

	Discussion

Top Abstract Introduction Methods Results Discussion References

Limitations of the study
A qualitative study with a small sample presents challenges of bias and also particular opportunities to check this. We controlled for bias in three ways. Firstly, we examined participants' reasons for taking part in the study, particularly in the focus groups as they were self selected. Most had personal experience of distress at work, but many had successfully dealt with or were actively dealing with this, and their experience provided appropriate data. Secondly, the 26 individual interviewees were purposely sampled, from 403 practitioners interested in taking part, to ensure variety by age, sex, marital status, size and location of practice, and morale. Thirdly, the analysis included a careful examination of the relation between the interview context and the findings, and this provided one of its main insights.

Respondents included only general practice principals. Further research should include study of practice teams and non-principals.¹⁰

Implications for future research
This paper has identified partnership dynamics, personal style, and workload related to changes in the NHS in Scotland as key factors in general practitioners' morale. The substantive findings from this qualitative study are similar to findings of quantitative studies on larger populations. ^{3 4 10} However, findings from large scale cross sectional studies do not directly map on to those of qualitative studies with smaller samples, and the two types of study can complement each other in interesting ways. In this study, rigorous attention to the way three different research contexts subtly, but appreciably, shaped the research outcomes indicates that the experience of morale in general practice is multifaceted, and people draw on different types of account to express this.

This has implications for further research. Different methods will access different aspects of work experience. The findings from any one study will depend on the questions asked and the context of questioning. Moreover, as general practice changes, the public and personal accounts describing the experience of work are also likely to change. As different research methods are applied in changing political and private contexts, other key factors will be identified. The crucial findings from this study are the complex interrelations between factors identified, the way these relations vary between individuals and contexts, and the way they are understood and managed.

Conclusions
Morale in general practice and primary care is complex. Concerns about workload were consistently expressed in three sets of qualitative data, obtained from stratified samples drawn from general practitioners in South East Scotland. However, throughout the research process, participants gave examples of innovative interventions at personal, partnership, and practice levels that were perceived as making important contributions to improved morale. Features of partnerships were identified that were key to coping with the increasing demands in primary care work while retaining a personal balance. These were respect for difference and flexibility to accommodate it, fairness in allocation of work and remuneration, and willingness to communicate at a personal level.

This research broadens the focus of attention from solutions at the level of the individual's coping strategy through to integrated solutions at partnership level and at the level of development of policy and resources. In practical terms, any measure to improve morale will have to enable people working in primary care to express and deal with the complexity of primary care work. In particular, rigorous action research is needed, in which several approaches to partnership and practice development are considered and developed.

	Acknowledgments

We thank all study participants for their time and contribution. This study is a development of the work of the late Pamela Baldwin of Working Minds Research.

Contributors: GH designed the study, did the one to one interviews, facilitated two of the focus groups and observed the other six, analysed the data, wrote the paper, contributed to planning dissemination of results, and is the guarantor for the paper. MG facilitated six of the focus groups, helped to refine and develop the coding frame, double coded all transcripts, and contributed to the paper and to planning dissemination of results. MD did the administration of the interviews and focus groups, transcribed the interviews, commented on the paper, and contributed to the dissemination of results. MP and RW helped with design of the study, observed one or two focus groups, read selected transcripts, and contributed to the paper and to planning dissemination of results. BMcK and JS helped with design of the study, observed one or two focus groups, read selected transcripts, contributed to the paper, and took a lead in organising meetings and events disseminating the results.

	Footnotes

Funding: The work was made possible by a bequest from the family of a local general practitioner.

Competing interests: None declared.

	References

Top Abstract Introduction Methods Results Discussion References

1.	Kmietowicz K. Quarter of GPs want to quit, BMA survey shows. BMJ 2001; 323: 887[Free Full Text].
2.	Scottish General Practitioners Committee. The reality behind the rhetoric: a survey of the views of GPs in Scotland on morale, service provision and priorities for improving primary care. Edinburgh: BMA, 2001.
3.	Sutherland VJ, Cooper CL. Job stress, satisfaction, and mental health among general practitioners before and after introduction of new contract. BMJ 1992; 304: 1545-1548.
4.	Sibbald B, Enzer I, Cooper C, Rout U, Sutherland V. GP job satisfaction in 1987, 1990 and 1998: lessons for the future? Fam Pract 2000; 17: 364-371[Abstract/Free Full Text].
5.	Heaney D, Gorman D, Porter M. Self recorded stress levels for general practitioners before and after forming an out-of-hours care centre. Br J Gen Pract 1998; 48: 1077-1078[Web of Science][Medline]
6.	Fletcher J, Pickard D, Rose J, Stewart-Brown S, Wilkinson E, Brogan C, et al. Do out-of hours co-operatives improve general practitioners' health? Br J Gen Pract 2000; 50: 815-816[Web of Science][Medline].
7.	Firth-Cozens J. Individual and organisational predictors of depression. Br J Gen Pract 1998; 48: 1647-1651[Web of Science][Medline].
8.	Howie JGR, Porter AMD. Stress and general practitioners. In: Firth-Cozens J, Payne RL, eds. Stress in health professionals, 2nd ed. Chichester: John Wiley, 1999.
9.	Firth-Cozens J. Hours, sleep, teamwork, and stress. BMJ 1998; 317: 1335-1336[Free Full Text].
10.	Calnan M, Wainwright D, Forsythe M, Wall B, Almond S. Mental health and stress in the workplace: the case of general practice in the UK. Soc Sci Med 2001; 52: 499-507.
11.	Malterud K. Qualitative research: standards, challenges and guidelines. Lancet 2001; 358: 483-489[CrossRef][Web of Science][Medline].

(Accepted 13 March 2002)