Intended for healthcare professionals

General Practice

Purchasing for all: an alternative to fundholding

BMJ 1994; 308 doi: (Published 05 February 1994) Cite this as: BMJ 1994;308:391
  1. J P Graffy,
  2. J Williams
  1. Academic Department of General Practice and Primary Care, Medical Colleges of St Bartholomew's and the Royal London Hospitals, London Hackney General Practitioners' Forum, London E3 2AN
  1. Correspondence to: Dr Jonathan P Graffy, Statham Grove Surgery, Statham Grove, London N16 9DP.
  • Accepted 16 December 1993


Objective: To evaluate general practitioner participation in a district health authority's purchasing work.

Design: Questionnaire study of 131 Hackney general practitioners and 33 senior health service managers; review of the minutes of 28 meetings of the Hackney General Practitioners' Forum and the contract between City and Hackney Health Authority and the St Bartholomew's NHS Trust.

Setting: Hackney General Practitioners' Forum.

Main outcome measures: General practitioners' and managers' perceptions of how representative and effective the general practitioners' forum is; proportion of new quality targets and service developments contributed by general practitioners; main issues discussed by the forum and impact on district health authority policy.

Results: 99 (76%) general practitioners and 27 (82%) managers responded. Both groups perceived the forum as representative. 92% (24/26) of the managers thought the forum was effective but only 74% (70/95) of general practitioners did so, largely because some doubted that the forum was listened to. 75% (103/138) if quality targets and 55% (16/29) of service developments planned in the 1993-4 contract were contributed by general practitioners. They also lobbied successfully for more resources for urology and community mental health services.

Conclusions: Input into commissioning via a general practitioners' forum can be both representative and effective. General practitioners need to work closely to achieve a consensus and those involved need administrative support. The relation between general practice and public health medicine needs to be strengthened.


  • Implications

  • General practitioners offer health authorities grassroots feedback; in return they can demand a say in the priorities set for secondary care

  • General practitioners must develop a consensus if they are to be listened to

  • Commissioning partnerships may be highly effective but need adequate resources and trust

  • Schemes including all general practitioners should be eligible for the financial support fundholding now receives


General practitioners have always had opinions on the quality of the hospital care their patients receive. Around 5% of consultations lead to a referral,1 and patients often want to discuss the options available or to report their experiences after seeing a specialist. Until the NHS and Community Care Act 1990 split the NHS into purchasers and providers2 general practitioners who wanted to question the quality of the secondary care their patients received, or the priorities set by local hospitals, had to negotiate largely on the hospital's terms. Since the reforms, however, the commissioning process has offered an alternative way to influence secondary care.

Though fundholders may have more influence on the hospital care covered by their budgets, their small size and the fact that the district health authority still purchases most of their patients' care means they too can have an interest in the decisions the district health authority makes. To purchase effectively, district health authorities need to know what is happening in the health service locally, and as a result many have come to see general practitioners as a source of information and advice. It is this recognition of a common interest between general practitioners, demanding better secondary care for their patients, and health authorities, seeking grassroots feedback, which has led to a range of schemes to involve general practitioners in health care commissioning.*RF 3-6* The way this has developed in different areas has been influenced by the extent of fundholding, the ability of general practitioners to coordinate their approach, and the commitment of the district health authority.7 Some district health authorities have developed a genuine partnership with general practice whereas others have been more reluctant to share decisions.

In City and Hackney most general practitioner participation in commissioning has been through a “general practitioners' forum.” This paper evaluates the approach the forum has taken.

Hackney general practitioners' forum

The general practitioners' forum was established in 1985 with support from the department of general practice at St Bartholomew's Medical College and has become a focus for discussion about health services locally. All City and Hackney general practitioners are invited to the monthly meetings, which rotate from surgery to surgery to encourage more to attend. When the NHS reforms were introduced the forum adopted a twin track approach, trying to win influence through the district health authority but considering a fundholding consortium if this was not achieved. In the event only two practices chose to become fundholders.

The first links with the district health authority entailed setting up advisory groups, which helped draft the first contract with the local provider but then lapsed.8 By 1992 it was apparent that a more coordinated approach was needed if general practitioners were to participate effectively, so the family health services authority agreed to fund an administrator and (together with the district health authority) seven sessional purchasing adviser posts. These general practitioner advisers have met monthly as a core group, but each has established his or her own links with particular specialties. These links include attending joint planning meetings, work on clinical guidelines, audits, and educational activities, all of which have helped the advisers establish a clearer picture of the current services and how they could be improved.


The minutes of 28 forum meetings from December 1989 to June 1993 were analysed to provide attendance rates and details of the main issues discussed. The minutes of two further meetings were not available. The issues reported were grouped into categories and used to design a questionnaire, which was then piloted and sent to all general practitioners in City and Hackney. The doctors were asked how important they considered these issues were as well as how representative and effective they thought the forum was. Two reminders were sent. Details of doctors' qualifications were taken from the medical list of the family health services authority.9

Additionally, 33 senior NHS managers who either headed departments at the family health services authority, district health authority, or local provider or chaired joint planning groups were asked their views on how representative and effective the forum was. The SPSS/PC+ package10 was used to analyse the results, with X(super2) tests to assess the response rate and attendance at meetings and non-parametric tests for the questionnaire results.

Finally, the 1993-4 contract between City and Hackney Health Authority and the St Bartholomew's NHS Trust11 was analysed to see how many of the new items were contributed by or had significant input from general practice. The attribution of these was confirmed, item by item, with the general practitioners concerned and the health authority contracts manager who wrote the document.


Ninety nine (76%) of the 131 general practitioners and 27 (82%) of the 33 health service managers responded. General practitioners who failed to respond were significantly less likely to have attended any forum meetings (X(super2) = 19.9, df=1, P<0.001). There were no significant differences in practice size, year of registration, and qualifications between those who did and did not respond.


The minutes of the 28 forum meetings showed an average attendance of 21 general practitioners. Forty three (33%) doctors had never attended a meeting, 67 (51%) had attended two or more meetings, and 35 (27%) had attended five or more. However, 78 (60%) worked in practices which were represented at five or more meetings.

Doctors who had attended twice or more were significantly more likely to be members of the Royal College of General Practitioners and to be approved for child health surveillance or minor surgery (X(super2) values 4.36, 7.51, 4.60 respectively; df=1, P<0.05 in each case). Practice size, year of registration, and inclusion in the obstetric list were unrelated to attendance.

How representative is forum seen to be?

When asked whether they agreed with the statement “The GP forum represents my views well” 90% (85/94) of general practitioners said they definitely or probably did (table I). Three of the nine who disagreed said they perceived the forum as something of a clique, and two said that as fundholders they thought the forum did not represent them. Eighty nine per cent (24/27) of the health service managers agreed with the related statement “The GP forum is representative of local GPs,” suggesting the forum is recognised as a representative body. However, several suggested the forum tended to represent the more developed practices, perhaps because they had met doctors from those practices more often at meetings. Most managers said they probably agreed the forum was representative whereas more of the general practitioners definitely agreed, a difference which was significant when the rankings given by each group were compared (Mann-Whitney U test, z = 2.1109, P<0.05).

TABLE I - Perceptions of Hackney General Practitioners' forum. Figures are numbers (percentages) of general practitioners and NHS managers completing questionnaire

View this table:

How effective is forum seen to be?

Ninety two per cent (24/26) of the NHS managers but only 74% (70/95) of the general practitioners agreed that the forum was “effective at influencing the health service locally,” a difference in perception which was significant (Mann-Whitney U test, z = −3.35, P<0.001) (table I). Seven of the 25 general practitioners who doubted the forum's effectiveness commented that they had not seen it achieve results; five thought the forum lacked power because its role was advisory; and five thought the government, district health authority, and providers did not listen to representations. The NHS managers were more positive, but this should be interpreted with caution, as they may have been reluctant to admit if they were not open to general practitioner influence.

A significant majority of general practitioners considered the forum to be more representative than effective (Wilcoxon test, z = −3.89, P<0.001) but, interestingly, most of the managers scored the forum's effectiveness higher (Wilcoxon test, z = −2.01, P < 0.05).

General practitioners' concerns and strategy

The minutes of the 28 forum meetings reflect the concerns of general practitioners during a period of great change. The NHS reforms, general practitioner contract, and more recently the Tomlinson report have been important issues, but concern to avoid a two tier NHS and work with the purchasing authority to improve secondary care have also been recurring themes.

The establishment of a core group and appointment of an administrator in April 1992 marked an important change. Before that, general practitioners concerned in joint planning or purchasing contributed mainly as individuals, but since then the core group members have worked to establish a common approach. The strategy adopted, jointly with the health authority, was to specify individual quality targets to be met by each provider directorate (box). It was agreed that a general practitioner and senior clinician from the trust would attend each directorate's contract meeting to ensure the targets set were both realistic and fully implemented. To evaluate this approach the quality targets and planned developments in services in the 1993-4 contract11 were counted to determine who had contributed them (table II). This showed that 75% (103/138) of the quality targets and 55% (16/29) of the service developments planned for individual directorates had come largely from general practitioners. Most of these were for general medicine, care of the elderly, obstetrics and gynaecology, or paediatrics, specialties which general practitioners have frequent clinical contact with. They also illustrate how the general practitioner advisers brought ideas from joint planning or the collaborative guidelines project12 and codified them in the contract document.

TABLE II - Origin of new requirements in 1993-4 contract

View this table:

The general practitioner advisers also contributed to quality specifications13 for communications with general practice, prescribing, dealing with non-attenders, and in vitro fertilisation. However, the specifications for health promotion, advocacy, and implementing the Health of the Nation and patient's charter were largely the work of the public health department. An audit of hospital communications was conducted jointly with the local provider and “service alert” forms are used to inform purchasers about individual problems, but a more systematic approach to contract monitoring is planned by selected practices.

Urology services locally had been a particular concern, so two practices audited the care their patients had received. This showed that patients waited too long for admission and pointed to the need for a bladder cancer register to improve follow up. After discussions with the urology consultants the purchasing authority specified time limits for admission for cystoscopy and prostate surgery but agreed to increase the contract from 900 to 1100 episodes to enable the urologists to meet these targets.

One consistent priority has been support for community and mental health services. In 1990 the forum successfully opposed plans for hospital departments to manage the community health services and in 1992 helped to ensure the community, mental health, and care of the elderly budgets were shielded against cuts. The general practitioner adviser on mental health has encouraged the district health authority to develop more community based services, with the result that mental health received most of the district health authority's 1993-4 growth money.

The Tomlinson report14 with its recommendation to close St Bartholomew's and the Queen Elizabeth Children's Hospitals has been a major concern and the forum was able to produce detailed responses to most of the changes proposed. Two meetings were held with the minister of health, Brian Mawhinney, to express concern about the level of inpatient care needed in a deprived area and the ability of the Homerton Hospital to cope if the proposed closures were implemented.

When asked which of the forum's activities they saw as most important, general practitioners gave the results shown in table II. The scores, on a seven point scale from “not important” to “very important,” confirmed that most doctors thought the issues the forum had addressed were important. “Better hospital services,” “resources to develop general practice,” and “informing general practitioners about what's going on locally” were seen as the three highest priorities (table III).

TABLE III - General practitioners' priorities for forum

View this table:


The supporters of fundholding have claimed it “offers an opportunity to make the health service responsive to general practitioners, acting on behalf of patients,”15 but many fear it will lead to a two tier health service, waste a great deal of money on administration, and allow rationing decisions to undermine the doctor-patient relationship.16 Do “commissioning partnerships” between general practitioners and health authorities offer the benefits but not the risks of fundholding?

This study has shown that most general practitioners in Hackney feel well represented by the general practitioners' forum and that managers respect it as a representative body. However, it points to the need to draw in more local doctors and to improve links with the two fundholding practices.

Though most of the managers were impressed by the forum's effectiveness, a quarter of the general practitioners had doubts about the extent of their influence. In part, this reflects the way that the health service agenda in London has been dominated by the Tomlinson report and the provider changes which have followed. As a result, developing purchasing and ensuring quality of care have at times seemed to have a lower priority. This is best illustrated by the dilemma about whether to impose financial penalties on a local hospital for failing to meet targets, when the real issue is whether the hospital will survive at all.

Fundholding at present concentrates mainly on a limited range of elective work, but the general practitioners' forum has been able to influence a wider range of services. For example, it is questionable whether mental health services would have been given such a high priority had most local practices opted for fundholding. The isolation of fundholding practices may also limit their ability to achieve strategic changes in the pattern of service delivery, whereas by working with the district health authority general practitioners may be able to achieve this.

This study also highlights the need to review the relation between general practice and public health medicine. As Hannay has pointed out,17 we are both concerned in health promotion; we are both concerned in purchasing. Though nationally the two branches of the profession may have lost touch, locally our links have been productive. The public health department has concentrated more on health and health promotion, as well as researching the efficacy of treatments like tonsillectomy. General practitioners have brought their generalist's knowledge of the health service and incorporated this into the detail of contracts. They can monitor performance, by auditing the care their patients receive, and at times be a voice for others who cannot talk freely in today's NHS, an approach illustrated by the forum's public campaign in response to the Tomlinson report.

If general practice is to have an increasing role in purchasing, schemes like this will need political commitment from the Department of Health. Those involved will need training, administrative help, and funding for time spent on purchasing. General practitioners need to work closely to develop a consensus if they want to be listened to and, similarly, health authorities must be prepared to share important decisions and offer representation at every level if a commissioning partnership is to be based on trust.

This model has developed in an area where the general practitioners' forum already provided a voice for general practice. It is seen as both effective and representative, and this review documents the influence the forum has had on services locally. In other areas general practitioners may take different approaches, but the opportunity to help commission more responsive services is there for all.

We thank the general practitioners and health service managers who completed the questionnaires; Chris McManus for statistical advice; Clare Goodhart, Gabby Tobias, and Gene Feder for commenting on the manuscript; and Nick Evans for checking the attribution of changes to the contract with the Barts NHS Trust.

Quality targets proposed by general practitioners

* All women who miscarry should be offered counselling

* 80% of people over 75 admitted as emergencies should be seen by a geriatric consultant or senior registrar within 48 hours

* Patients needing cystoscopy should have this done within six weeks after general practitioner referral

* A random audit of 100 diabetic patients should be done and the results made available to the district health authority

* A shared care system should be developed for patients receiving second line drugs for rheumatoid arthritis


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
View Abstract