Primary Care

General practitioners' perceptions of sharing workload in group practices: qualitative study

BMJ 2004; 329 doi: (Published 12 August 2004) Cite this as: BMJ 2004;329:381
  1. Ruben Branson, general practitioner associate1,
  2. David Armstrong (david.armstrong{at}, reader in medical sociology1
  1. 1 Department of General Practice, King's College London, London SE11 6SP
  1. Correspondence to: D Armstrong
  • Accepted 14 June 2004


Objective To explore general practitioners' beliefs and experiences of distribution of workload and teamwork between doctors in general practice.

Design Qualitative semistructured interview study.

Setting South London.

Participants 18 general practitioners from 11 practices.

Main outcome measures Perceptions and experiences of distribution of workload and teamwork between doctors.

Results Equitable distribution of workload was a common concern among general practitioners in group practices. Several ways of addressing the problem were identified, including relying on trust, creating systems of working based on explicit rules such as points' systems, and improving communication. Improvement of communication was hampered by the taboo nature of the problem.

Conclusion Resentment about perceived inequalities in workload places a further burden on general practices. The issue of working together warrants further support.


Despite its centrality to the everyday work of general practitioners, the distribution of workload within a practice has not been the subject of major study. Research on practice organisation has tended to focus on interprofessional teamwork,14 with mainly anecdotal contributions on the relationship between doctors.5 Yet working with other general practitioners has become much the standard in the NHS with its falling numbers of single-handed practices and increases in the size of partnerships, such that nearly a third of general practitioners now work in groups of six or more people.6 This group size ignores the increasing number of non-principal staff such as general practitioner assistants, associates, locums, registrars, and general practitioners on the retainer or flexible career scheme who contribute to the size of working groups.

Working together presents many benefits for general practitioners. These include economies of scale in premises and staffing, shared cover for out of hours' work and holidays, and prevention of professional isolation. There may, however, be costs in sharing workload. Recent studies examining morale in general practitioners have highlighted the importance of partnership relationships in dealing with the increased workload in primary care. One study, for example, noted that fairness in the allocation of work, remuneration, and personal communication between partners were important in coping with increased workload and were related to morale.7 In a study of dissatisfaction among new general practitioners in an inner city area, the stress of working in a partnership was a major reason for expressing regret at joining a practice.8 We carried out a qualitative study of general practitioners' perceptions and experiences of distribution of workload and teamwork in group practices.


Our study took place in inner city practices in south London within the health authority area of Lambeth, Southwick, and Lewisham. This is a densely populated area with a high cultural and ethnic mix of patients.

We purposively sampled the general practitioners. They were chosen to give a mix of practice sizes, general and personal medical service contracts, number of principals and non-principals, sociodemographic characteristics, and structure of the practice regarding distribution of workload. We tried to interview two general practitioners from each practice to include different perspectives on the same work environment.

General practitioners were contacted by telephone and invited to take part in the study. The study was explained, and an explanatory letter was emailed or posted before the interview. The interviews were conducted in locales convenient for the participants. The semistructured interview schedule (administered by RB) was used to introduce different issues to the respondents who were then encouraged to clarify their thoughts by way of several follow up questions (box).

Respondents were allowed to withdraw consent before, during, or after the interviews. Interviews lasted between 35 and 65 minutes and were audiotaped.

The interviews were transcribed and analysed using ATLAS/ti. The analysis proceeded throughout the study to allow emergent themes to be fed back into the data collection. These themes and the research question then formed the basis of the coding strategy. Regular review and discussion of evolving themes contributed to the data synthesis and interpretation.

Follow up questions in interview schedule

What do you think of teamwork between doctors in your practice?

How is workload distributed in your practice? Cover visits, paperwork, etc

How has this developed/where did this come from?

What do you think about communication in your practice?

What do you feel works well?

What do you feel does not work so well?

What are your experiences in other practices?

Are there any issues that you think are important?


Overall, 20 general practitioners were contacted and agreed to be interviewed, but two cancelled the appointment. Eighteen interviews were therefore included in the analysis. The general practitioners came from 11 practices, which ranged in size from three full time partners to 15 doctors (nine full time partners and four part time partners supported by two salaried assistants).

The workload problem

All the general practitioners recognised the problem of inequalities in the distribution of workload either in their current practice or in a previous practice. The most common basis for inequality, mentioned by all but two respondents, was the difference in consultation style between fast consulters, who tended to stick to time, and slow consulters, who tended to overrun clinics. Patients were usually allocated in equal numbers to each general practitioner, so fast consulters finished their surgeries early and found themselves doing other tasks such as seeing patients who arrived without appointments, doing visits, and organising repeat prescriptions. This led to resentment:

But you can't help thinking why am I dealing with all the phone calls just because I have finished my surgery and Dr X is running way over time and not picking up that extra slack. (Practice 4, general practitioner A)

Slow consulters, however, thought that the speed of faster consulters was often a false economy. They saw themselves as attending to more complex psychological and social problems, sorting out more presenting problems, and providing a high quality service. They therefore believed that their workload was increased and so were their levels of stress compared with that of faster consulters:

You notice when your colleagues are on holiday and some of their appointments are BP checks and repeat prescriptions for hypertension. Now that is not going to take a long time but it will fill up your slots so no wonder you see your work quickly if that's what you're day is full of. So I don't think that speed is necessarily quality. (Practice 1, general practitioner A)

Variation in consulting style was largely explained as simply the way different doctors practised medicine, and the perceived inequalities that might result were therefore unintentional. But there was also the view that colleagues sometimes deliberately manipulated their workload:

Some of the GPs are minimalist, some of the principals and they are out of the door at 6 o'clock every night and they are off for their half day at noon on the dot. I don't know how they do it… I think that perhaps he is not sorting the patient perhaps it's being landed on us. (Practice 3, general practitioner A)

Although concern with the distribution of workload seemed to be an ongoing feature of working in a partnership, all the general practitioners (except one who was about to leave the practice) somehow managed their situation despite the divisive forces engendered by resentment. Several solutions were described that general practitioners thought underpinned stable and good working relationships. None of these strategies was perfect, and practice equilibrium was constantly threatened by new conflicts.

The need for flexibility

Some general practitioners, especially those in small practices, described the importance of flexibility and trust. These doctors stressed the reciprocity inherent in flexible systems and the need to “get along”:

The key to that system working is showing that if you are flexible with your partners they will be flexible with you and it is just a basis underpinning of any partnership that it's a give and take thing. (Practice 5, general practitioner B)

The doctor's approachability was seen as an important factor in maintaining a flexible system, yet at the same time it could exacerbate tensions if all doctors were not viewed as equally approachable. Receptionists, for example, who needed to find a doctor for a task, were reported as asking the most approachable, which meant that those with a more difficult manner escaped the extra work.

Structuring workload

The alternative to an informal system of workload distribution that relied on flexibility and trust was a more explicit division of labour in which strict rules governed the distribution of tasks. In some practices this was achieved by keeping lists for each doctor, which implied areas of responsibility for specific patients. In other practices the division of labour had become so formalised that points were awarded for specific duties and tied into financial remuneration. The rules were clear and there were less grey areas in distributing workload. Respondents from these practices reported that they were able to predict duties and workload in advance:

I think the more traditional and the less structural the organisation the more of the chances that there is an uneven style of work with the amount of paperwork. If you introduce a rigorous system that takes that away and that means more of an even distribution of everything. (Practice 4, general practitioner A)

Points' systems seemed particularly important for larger heterogeneous practices in which there was a variety of doctors, particularly practices with non-principals and part-time partners. In one practice, for example, although six of 10 doctors were partners only two were full time.

Practices that used structured systems for allocating tasks often found it difficult to quantify all workload owing to the nature of clinical work and the existence of areas of “hidden work” such as consultations by telephone. Yet for many respondents the question of who determined the allocation of points to particular tasks was much more important. Indeed, lack of involvement in developing a “fair” system was a common source of resentment about workload. Salaried employee doctors in particular seemed rarely involved in developing points' systems and were also unlikely to be able to benefit financially from working harder:

You see we have the points system but I am not involved in deciding the points because I am only a salaried doctor. I seem to do more on calls than the partners do and also spend more time seeing patients but where is the reward? (Practice 3, general practitioner B)

So we share on-call equally amongst the partners and the PMS doctors do slightly less. They are not on the point system they are on contract. This is where problems can arise because the PMS doctors think they are doing too much and sometimes we think they are actually doing less than we do for the amount of pay they get so I think that is where some of the cracks appear. In fact the salaried doctors have fewer points which are reflected in their salary so in fact they are on an equal pay but sometimes they find it difficult to understand. Practice 3, general practitioner A)

There were also reports of practices that had moved from structured systems to more informal ones, in one case when the mix of partners changed:

I think myself and the new doctor that arrived here basically have built up that level of trust… I think there were such big issues about trust in another partner (since left) and basically when that trust had all gone… I think there was no real hope of the practice working. (Practice 11, general practitioner A)

Improving communication

Despite the range of concerns about inequality in workload, most general practitioners felt it was difficult to discuss these issues with colleagues:

I'm sure it's a reason practices break-up and because it's a taboo subject that you can't talk about as you are afraid of raising it and you're afraid to rock the practice and break it up so you go on working along with your feelings bottled up inside. (Practice 2, general practitioner A)

Only one general practitioner described how she had discussed the subject with a colleague. This happened in a small informal practice and the doctor concerned stated that they were both close friends and able to discuss these issues. Once this colleague had left the practice, however, she felt unable to communicate these issues to the same degree with her other colleagues.

For some general practitioners, regular practice meetings provided a useful mechanism for discussing problems and sharing personal grievances. One respondent, however, described good informal communication within her practice but was critical of practice meetings. She told of her experience in another practice where there was a lot of formal communication but to little effect:

It was dreadful. That is one of the reasons I left. We had weekly practice meetings which tended to go on for hours which did not achieve anything and then other meetings were called in the week. We seemed to have more meetings than surgeries towards the end. (Practice 8, general practitioner A)

One of the powerful ways in which communication served to reduce resentment about workload was to enable direct comparison of workload over previous weeks. These usually took the form of computerised summaries of activities such as the number of patients seen, visits done, and on-call duties. A common observation was that despite initially having feelings of working harder than colleagues these perceptions were not confirmed by the data:

I think in general people think they are being dumped on. I'm doing more than I should. If you can see it in black and white in actual fact you are doing the same as everyone else. You only remember the bad weeks. You remember the week you are doing twice as many duty doctors, two prescriptions that week rather than one, you forget that it's been weeks and weeks since you've done a duty prescription so it varies. (Practice 5, general practitioner B)


Perceived inequalities in workload can be a major cause of resentment whenever general practitioners work together. Such resentment is an unintended consequence of partnership working which was first introduced into the NHS with the good intentions of sharing the burden of patient demand, achieving economies of scale for such areas as attached staff and premises, and preventing. professional isolation. Yet while many practices offer supportive environments for general practitioners, working together can also create yet another source of stress for busy health professionals.

The view that workload was not distributed equitably seemed closely linked to the size of the practice or, perhaps more accurately, its heterogeneity. When general practitioners had common working styles and small practices, there could be a system of mutual trust and flexibility. Increasing practice size and different types of general practitioners meant this style of organisation could not work as effectively, and the friction was expressed as resentment over workload distribution. This problem was usually addressed by a more explicit allocation of tasks so that workload could be seen to be fair. At its most extreme this involved introducing a points' system that was tied to remuneration, but such a formal system of allocating work was only seen to be fair if it had emerged through consensus: general practitioners outside the agreement could feel even more disadvantaged as the inequality became enshrined in the points themselves.

What is already known on this topic

Most general practitioners now work in group practices

Little is known about their perceptions of shared workload

What this study adds

Teamwork produces tensions, particularly over the perceived distribution of workload

Small practices mainly rely on trust for managing these potential areas of conflict

Larger or more heterogeneous practices often introduce rules for allocating workload

Perceived inequity in workload is a major source of resentment and stress

Better communication, both formal and informal, might seem the solution to many of these resentments, but the very topic of workload often seemed too emotive to discuss.9 When it was discussed, however, especially alongside data on actual workload, respondents observed that resentments were often dissipated by data that showed that workload was not as misallocated as assumed.

Although problems with the distribution of workload were commonly reported these were not a constant feature of teamwork. Practices seemed to go through phases of contentment and resentment with workload distribution. This seemed to be an aspect of practice growth (increasing size highlighting the need for different ways of working) and the natural course of practice dynamics, with accumulating resentments being resolved by showing the perceptions of inequality having no basis in fact, or introducing “fairer” rules on work allocation. The steady trickle of practice break-ups is, perhaps, evidence of the final cost of not getting work relationships right.

The need for explicit rules to manage interaction might be described in terms of trust (several respondents mentioned this concept). When colleagues could be trusted to carry out a fair share of the workload there was no need for close monitoring of behaviour. But when that trust began to break down, it was transferred to the points' system, which ensured equity of workload, except for those general practitioners not involved in establishing the system who therefore had little trust in it.

A crisis of trust is not unique to partnerships in general practice. Indeed, the decline of trust across modern society and across healthcare organisations has been the subject of much comment.10 11 Yet trust remains an important component of successful teamwork whether in relationships with colleagues or in the belief that organisations are fair; it has been argued that the existence of trust is a prerequisite for good quality care.12


  • This study derives from an MSc in general practice carried out at Guy's, King's, and St Thomas's School of Medicine. We thank the general practitioners who took part.

  • Contributors RB and DA devised the study, carried out the analysis, and wrote the paper. RB collected the data. DA will act as guarantor for the paper.

  • Competing interests None declared.

  • Ethical approval King's College Hospital research ethics committee.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
View Abstract