Intended for healthcare professionals

General Practice

Organisation of primary care services outside normal working hours

BMJ 1994; 309 doi: (Published 17 December 1994) Cite this as: BMJ 1994;309:1621
  1. Lesley Hallam,
  2. David Cragg
  1. Centre for Primary Care Research, Department of General Practice, University of Manchester, Rusholme Health Centre, Manchester M14 5NP, research fellow, clinical lecturer.
  1. Correspondence to: Ms Hallam.
  • Accepted 9 September 1994


Objective: To determine the use and organisation of out of hours services in primary care.

Design: Telephone survey.

Setting: Family health services authorities in England and Wales.

Main outcome measures: Rate of use of out of hours care, methods of provision, and role of authorities.

Results: 12-25% of authorities were unable to answer one or more key questions in the survey because of insufficient information. The mean number of night visits made per unrestricted principal per 1000 patients per year was 35.3. 13 of the 19 authorities with averages above 40 covered large towns or cities. 81 authorities had at least one commercial deputising service. In 46 metropolitan districts and one other district over 75% of general practitioners had consent to use a deputising service, although not all did so. Information on cooperation between practices was limited. 22 cooperatives were recognised by the authorities, nine were not officially recognised, and a further 13 were nearing institution. Only two cooperatives were in areas with extensive use of deputising services.

Conclusion: Methods of providing out of hours care are changing, and without good information systems family health services authorities will not be able to monitor the effect on quality and cost effectiveness of care.

Key messages

  • Key messages

  • Dissatisfaction among general practitioners and patients is leading to important changes in the provision of out of hours primary care

  • This study showed that, with the exception of London, demand for out of hours care was highest in urban areas

  • The mean proportion of visits provided by deputising services in each family services authority was about a third; 13 authorities could not obtain this information

  • Most authorities had limited information on cooperation between practices

  • The information systems of family health services authorities need to be improved to enable them to monitor the quality and cost effectiveness of new developments in providing out of hours service


Fundamental changes in the delivery of primary medical care outside normal surgery hours are being considered in Great Britain.1 2 3 A recent review highlighted the rising demand from patients and growing disaffection among providers.4 It also pointed out that, although innovative services already exist,5 6 much of our knowledge about demand and provision is fragmentary or anecdotal. We conducted a survey of family health services authorities to determine current organisational patterns in out of hours services in general practice and the demand for care between 10 pm and 8 am.


During July and August 1993 we conducted telephone interviews with officers responsible for out of hours arrangements in family health services authorities in England and Wales. The interview schedule was based on a comprehensive literature review,4 discussions with several authorities and the Department of Health, and three pilot interviews. We asked about the number and composition of responsible practices, the extent and nature of information held on out of hours care, the methods used by general practitioners to provide cover, and the cost of night visits. Detailed questions were asked about cooperation between practices; the availability and use of deputising services; formal and informal links between authorities and deputising services; innovations in the delivery of care and the authority's role in these. We sent a summary of the schedule to the authorities before the interview to enable informants to collate data. Any missing information not held by the informant but believed to be available elsewhere in the authority was sought by letter.


Information was obtained from 97 of the 98 family health services authorities in England and Wales, with varying degrees of comprehensiveness. Authorities differed in the amount and type of information held and the manner in which it was collated and stored, reflecting differences in service provision and the perceived value of data. Seven authorities (two London boroughs, two metropolitan districts, and three shire counties) could answer few questions at interview and did not respond to follow up letters. Between 12% and 25% of authorities were unable to answer one or more key questions on provision and use. This was often because of inability to aggregate data from practice profiles or disaggregate financial data. Data were often distributed throughout different departments, leading to collation difficulties, particularly where reorganisation was taking place.


The mean annual number of night visits made per unrestricted principal per 1000 patients throughout England and Wales was 35.3. Table I shows the wide variation among authorities in the average number of night visits made per 1000 patients. Thirteen of the 19 authorities with averages above 40 were in the large conurbations of Greater Manchester, West Midlands, and South Yorkshire. Their mean number of night visits was 47 per 1000 patients with a range of 41 to 59. Of the remaining six authorities with rates above 40, three were in industrial areas in the north and north Yorkshire regions. Among the 10 Greater London authorities who provided information, the average number of night visits per 1000 patients was 25 (range 11 to 34). Rural and semirural authorities had much lower rates than those covering predominantly urban and suburban areas, with the exception of Greater London.


Mean number of night visits per general practitioner/ 1000 patients/year

View this table:

Information was not available in 14 authorities. Although the total cost of fees was known, the number of claims could often not be calculated because of lack of discrimination between lower rate fees of pounds sterling 15, primarily applicable to visits by deputising services, and higher rate fees of pounds sterling 45 applicable to general practitioners providing their own cover.


Among the 85 authorities able to provide sufficient data the average proportion of night visits attracting lower rate fees throughout England and Wales was around 34% (table II). Eighty one authorities had at least one commercial deputising service operating in their area. In 38 cases only part of the area was covered. However, in 47 authorities, 46 of which were in metropolitan districts or London boroughs, over 75% of general practitioners had consent to use a service.

Table II

TABLE II—Percentage of night visits attracting lower rate fees

View this table:

Consent was not an accurate predictor of use. For instance, in one authority where 80% had consent, only 21% of night visit fees were paid at the lower rate.


It was difficult to determine the extent of cooperation between practices in providing out of hours cover. Many authorities require that practices inform them only of rotas exceeding 10 general practitioners since doctors in such rotas will normally qualify for lower rate night visit fees unless they are visiting their own patients. Others felt unable to devote resources to aggregating information held on individual practices.

No common vocabulary or definitions existed for describing collaboration between practices. We therefore defined two terms. Informal arrangements between individual practices, mainly between 10 or fewer general practitioners, are referred to as joint rotas. The term cooperative is applied to more formally organised, non-profit making services with many subscriber members. We found, however, some examples of collaborative working that did not readily fit either category.

Table III shows the extent of cooperation between practitioners from more than one practice in the 73 authorities able to provide detailed information. Cooperation was generally low, and there were no apparent associations between types of area and extent of cooperation.


Extent of cooperation on out of hours visiting between general practitioners from different practices

View this table:

Data not available for 25 authorities.

In August 1993, 22 cooperatives existed that were recognised by their authorities. Nine other groupings seemed to merit the label cooperative: four were not formally constituted and had not sought recognition and five had been refused recognition on grounds of size or unacceptable financial arrangements. Plans for 13 more cooperatives were at an advanced stage. In addition, 34 large groupings of general practitioners and 12 small groups or individuals were investigating cooperative arrangements. Three authorities were promoting area wide developments. Nearly 2000 general practitioners were reported to be members of recognised or unrecognised cooperatives at the time of the survey. Sizes varied from 15 to over 200. Only two cooperatives were in areas with extensive deputising service cover, but cooperatives were being planned in nine other such areas.

In areas where cooperatives would be viable but none were present, finances were considered the most serious obstacle to their formation. Very few authorities had been able to provide substantial financial help with setting up costs because of regulations on the use of cash limited General Medical Services funds, and a shortage of development funds that were not cash limited. Many authorities believed that the growth of cooperatives would largely depend on changes in regulations and funding. Three attempts to form cooperatives had reportedly failed, at least partially on financial grounds. Some Greater London authorities were planning to use the London initiative zone funding for projects. Elsewhere support took the form of small grants and loans, accommodation in unused district health authority premises, extensions of the cost-rent scheme to cooperative bases, and seconded staff to help in development.

Once a cooperative was operational, it was generally expected to be self financing. Though two family health services authorities provided support for cooperative based “walk in” centres, only one authority top sliced its General Medical Services budget to provide general support. The interpretation of night visit fee regulations differed between authorities. Though some gave all payments to cooperative members at the lower rate, others supported a system of networking with members grouped into rotas of 10 general practitioners. Visits to patients registered within the groups of 10 attracted higher rate fees. Cross cover and locum arrangements were allowed within the network in a few cases.

Family health services authorities with functioning cooperatives saw few disadvantages to them. Individual authorities pointed to the possible dilution of doctor-patient relationships, and the possibility of encouraging unnecessary call outs and dissatisfaction among patients with the characteristics of the visiting doctor. Concern was also expressed about the impact of cooperatives on local general practitioners who did not wish to participate, particularly if cooperatives displaced existing deputising services, attracted new patients, or consumed development budgets.

Generally, however, there was widespread support for cooperatives among authorities who had them. Many cooperatives worked to standards agreed with and monitored by their authorities. Higher standards were also believed to result from peer pressure, doctors' vested interests, and authorities' greater influence over the activities of their registered practitioners. The cost of out of hours visits was lower where cooperative cover replaced personal cover. General practitioners were believed to benefit from fewer periods on call, access to a wider range of equipment, better communication, and increased security in cooperatives that employed drivers. Their local knowledge increased their ability to mobilise other resources to assist patients. Less tangibly, they were not responsible for the activities of unknown doctors and their involvement in a cooperative lessened professional isolation.


Seventy six authorities with 20 113 responsible general practitioners had precise information on practices that provided all their own out of hours cover without recourse to joint practice rotas, cooperatives, or deputising services. A total of 5406 (27%) general practitioners worked in such practices. Table IV shows the variations between authorities in the extent of own cover. In areas where 75% or more of general practitioners had consent to use a deputising service, less than one third provided all their own cover. Of the five authorities with extensive own cover, three were sparsely populated rural areas and two had rejected deputising services.


Proportion of general practitioners in practices providing all their own out of hours cover

View this table:

Eighty four authorities knew which of their single handed practitioners were providing all their own out of hours cover. They identified 160 in this position (less than 6% of all singlehanded general practitioners). Such unsupported general practitioners were equally rare in rural and urban areas. Of the 16 most rural authorities (those with a population density below 160 people per km2) 14 with data knew of just 26 singlehanded practitioners without regular relief. Though the relief available was sometimes limited—for instance to shared weekend cover only—most singlehanded general practitioners were thought to have adequate out of hours support. The cost of deputising service cover for a small practice and the difficulty and cost of obtaining locum cover for more extended periods were seen as more problematic.


No clear pattern emerged for the development of centralised out of hours services. Such services were run or being planned by deputising services, cooperatives, individual practices, and sessionally employed general practitioners and were sited in hospital accident and emergency departments, central clinics, consulting suites within deputising service and cooperative offices, and general practitioners' surgeries. Hours of operation varied from occasional periods of high demand, through regular weekday evenings or weekend days, to all periods outside normal surgery hours. Examples of almost every combination of providers, locations, and times were found. It was thus impossible to quantify the full range of provision.

At the time of the survey family health services authorities knew of 12 commercial deputising service branches offering evening or weekend book in services to attenders. Twelve cooperatives were reported to be offering a similar service. Two of these services were hospital based, one of which was restricted to Sundays only. Though only two authorities were able to provide survey based figures of attendance rates (10% of all patients making telephone contact with the service in one case and 30% in the other) other authorities offered the subjective view that relatively few patients chose to attend. One evening book in surgery had reportedly been discontinued because of lack of attenders.

Walk in or book in emergency services offered by individual practices were mainly regular Sunday morning surgeries. These were widespread in only one area, though three other areas were aware of such surgeries in some practices. Some Sunday services had been withdrawn because patients could not be persuaded to attend.

Eight schemes linked to accident and emergency departments, 12 linked to cooperatives, and five where neither siting nor staffing had yet been agreed were being considered. In addition, schemes to use community and satellite hospitals for primary care emergency centres, to set up a nurse practitioner staffed minor injuries centre, and to provide 24 hour access to two health centres that would involve working shifts were also reported. Attempts to establish centre based services in Humberside, Trent, and Mersey regions had been abandoned because of problems with funding, siting, or general practitioner cooperation.


The roles and responsibilities of family health services have changed greatly since 1990.7 8 This survey suggests that, with respect to out of hours care, not all have successfully implemented a shift from administration to management. Though practice profiles enable authorities to inform and advise patients, without the ability to aggregate practice based data they cannot effectively monitor the level, quality, and cost effectiveness of services. This is concerning, particularly at a time when innovative and largely unevaluated methods of service delivery are being adopted.

Wide variations exist between authorities in the demand for care outside surgery hours, with average night visiting rates ranging from 11.4 to 58.8 per 1000 patients. Rates of use were generally higher in urban areas, though use in London was much lower than in other cities. This may be due to the increased use of accident and emergency departments by London residents.9 Deputising services were also more heavily used in urban areas, and further studies controlling for potentially confounding sociodemographic and geographical factors are required to determine whether this is linked to the increased use of out of hours care. Education campaigns designed to reduce inappropriate consulting might be more effective if targeted at particular areas or groups of patients.

Deputies were reported to have carried out 46% of visits between 11 pm and 7 am in 1989.10 In this study the mean proportion of night visits attracting lower rate fees was 34% with the overall proportion among all responding authorities being 32%. Though some visits by members of cooperatives and large rotas (more than 10) attract lower rate fees, most such fees will be for deputising service visits.

Expansion of cooperative cover will be considerable if financial support is available. Some confusion exists, however, about what constitutes a cooperative. Differences in size, accommodation, organisation, financing, business status, and employment of staff affect the way some authorities treat these groups. Their uncertain status is reflected in variations in the interpretation of rules and regulations governing financing. Though some authorities monitor performance of cooperatives, others do not. The growth of cooperatives is unevenly spread throughout the country. Few metropolitan districts where deputising service coverage is extensive had cooperatives. The impact of the formation of cooperatives on the viability of deputising services for the other general practitioners needs to be considered.

In future, a growing number of out of hours services will require some patients to attend an emergency centre. This is a central feature of the package of measures currently being negotiated.3 Though uniformity in service organisation and delivery is not necessarily desirable, there is a danger that the uncoordinated growth of such schemes will lead to different standards of service throughout the country. The effect of walk in or book in services on demand and use cannot be predicted, and few attempts have been made to evaluate and monitor them. Access to an integrated, comprehensive database will play an important part in ensuring that high quality, cost effective services are delivered, irrespective of the method of delivery.

This survey was funded by the Department of Health. The views of the authors do not necessarily represent those of the funding body.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.