General Practice

Extracontractual referrals: safety valve or administrative paperchase?

BMJ 1995; 310 doi: (Published 17 June 1995) Cite this as: BMJ 1995;310:1573
  1. Barbara Ghodse, service contracts managera
  1. a St George's Healthcare Trust, Grosvenor Wing, St George's Hospital, London SW17 0QT
  1. Correspondence to: Dr Ghodse.
  • Accepted 9 May 1995


Objective: To describe the extracontractual referrals of residents of a health authority during a six month period in 1994, identifying the number and cost of emergency and non-emergency referrals, including the number of cases costing more than pounds sterling20000 and those cases when payment was refused.

Design: Descriptive analysis of all extracontractual referrals submitted to the health authority between 1 April and 30 September 1994.

Setting: A health authority covering a poulation of 614000.

Results: Payment of pounds sterling2 583 693 was made to 263 different providers for 2400 episodes of care, of which 1469 were emergencies and 931 were elective or tertiary referrals. Authorisation was granted for an additional 1376 referrals for future treatment but was refused in 713 instances, mostly for technical reasons. Sixteen extracontractual referrals together accounted for over a fifth of total expenditure during the study period.

Conclusions: Handling large numbers of episodes of care on an individual cost per case basis imposes an enormous administrative burden on both purchasers and providers, diverting money away from patient care. Extracontractual referrals also expose health authorities to considerable financial risk and may undermine commissioning strategies. Measures are proposed to limit the number of episodes handled in this way.

Key messages

  • Key messages

  • Sixteen non-elective referrals, each costing over pounds sterling20 000, accounted for 21% of total expenditure

  • Mandatory payment for non-elective referrals can undermine a health authority's commissioning strategy and expose it to considerable financial risk

  • The high administrative costs of extra contractual referrals arise because a large number of episodes of care are handled individually

  • Measures should be introduced to limit the administrative burden of extracontractual referrals and other cost per case contracts to prevent further diversion of funds away from patient care


Hand in hand with the introduction of contracts for health services in 1991 came the extracontractual referral. Initially a cause of considerable anxiety,1 2 the management of extracontractual referrals has now become a complicated but largely routine administrative process carried out for reasons of confidentiality within designated “safe havens.” The number and complexity of the rules governing their management have multiplied steadily, and they have recently been “summarised” in a 100 page document.3 A separate paper, Establishing District of Residence, deals with bureaucratic niceties such as “hospital hoppers,” “holiday homes,” and “boarding school pupils from overseas.”4

Just as doctors in their consulting rooms take a history from their patients so administrators within the sanctity of safe havens systematically examine extracontractual referrals (most of which are rendered anonymous) to establish the health authority's liability for payment. In line with published guidance, checks carried out include:

  • Address and postcode, to ensure that the patient is a health authority resident

  • Whether the patient's general practitioner is a fundholder and, if so, his or her liability for payment

  • Whether the request is covered by a contract with that provider

  • Tariff, to ensure that the correct price is being quoted on the request or invoice

  • Timeliness of the request or invoice

  • The reason for the extracontractual referral, to identify whether treatment could be undertaken under a contract with another provider.

These checks, which are time consuming and tedious, often entail discussions with staff in the safe haven of the provider and with the patient's general practitioner. Unusual referrals (very expensive or uncommon treatments) require input from other health authority staff such as contracts managers and doctors in the department of public health medicine,5 who may need to speak to business managers or clinicians in the provider unit. Finally, whenever a request for an elective extracontractual referral is refused or referred again to another provider the patient and the referring practitioner must be informed of the reason for the decision.

The seclusion of extracontractual referrals within safe havens has tended to obscure this huge administrative burden. Other equally important issues, such as the financial risk to which health authorities are exposed by extracontractual referrals and the way in which they can undermine a health authority's commissioning strategy, have also received insufficient attention.

Many of the problems associated with extracontractual referrals also occur with cost per case contracts. As with extracontractual referrals the purchaser pays a fixed price for particular treatments as and when they arise, and again each episode of care requires separate administrative processing. The existence of a contract obviates the need for prior authorisation, however, which is a necessary stage in the processing of elective extracontractual referrals.

I describe the extracontractual referrals of the residents of one health authority during the period April to September 1994 inclusive. Some background information about the health authority is included so that the number and cost of its extracontractual referrals can be seen in the context of the total contract portfolio. I have discussed the implications of these findings as purchasing increasingly moves towards cost per case contracts, in which each episode of care is separately invoiced.


All extracontractual referrals received by the health authority between 1 April and 30 September 1994 have been included. They can be divided into two broad categories: emergency and non-emergency referrals. The latter includes elective referrals, for which prior authorisation must be requested, and tertiary referrals, when a patient is referred from one NHS consultant to another. Prior authorisation is not required and payment is mandatory for both emergency and tertiary extracontractual referrals. Provider units do not always identify tertiary referrals as such, however, and there had been no attempt to count them separately in this paper.

The number and cost of emergency and non-emergency extracontractual referrals were recorded, and the provider units from which they originated were identified. The number of extracontractual referrals for which payment was refused was also noted; the reason for refusal and the savings thus made were recorded. Finally, all extracontractual referrals costing more than pounds sterling20000 were separately identified to establish the nature and source of the referral.


The health authority studied has a population of 614000 and a recurrent cash limit of pounds sterling274 million. It has contracts with more than 50 providers, of which 30 are acute hospitals, including all the London teaching hospitals and special health authorities.


During the six month period (1 April to 30 September) pounds sterling2583693 was paid for 2400 extracontractual referrals: 1469 were emergencies costing pounds sterling929514 and 931 were non-emergencies costing pounds sterling1654179. Of the non-emergencies an unidentified number were elective referrals for which authorisation had been granted before the study period but the invoice was paid during this time. In addition, authorisation was granted for a further 1376 elective referrals (anticipated cost pounds sterling1117053) for which invoices had not been received, and 713 requests for authorisation or payment were refused. In total, therefore, 4489 extracontractual referrals (requests for authorisation or invoices, or both) were received during the study period, with a total financial commitment of pounds sterling3700746.

The 2400 referrals for which payment was made related to 263 different providers. Of these, 966 referrals were made to 26 providers with which the health authority has a contract. In particular, 556 were to one local, non-acute provider which excluded all highly specialist services, both inpatient and outpatient, from its contract.


Authorisation or payment was refused for 713 extracontractual referrals, 16% of the total received, with a total saving to the health authority of pounds sterling675604, 18% of the total financial commitment (table I). In 584 instances (82% of refusals) payment was refused for “technical” reasons: referral covered by a contract with that provider; retrospective request for authorisation; late or incorrect invoicing; patient was not a health authority resident; responsibility of a general practitioner fundholder, etc.


Refusals of extracontractual referrals

View this table:

In 113 cases, after discussion or correspondence, or both, with the patient and the general practitioner the patient was referred again for treatment under a contract. Of the 16 cases for whom treatment was refused, five were requests for treatment in private hospitals; eight did not meet agreed criteria for restricted treatments (sex change operations, assisted conception, cosmetic plastic surgery, dental implants); and in three cases the general practitioner did not support the patient's extracontractual referral for alternative medicine.


Sixteen extracontractual referrals, each costing more than pounds sterling20000 were paid for, of which 12 were emergency admissions and four were tertiary referrals (table II). Total expenditure on these 16 episodes of care was pounds sterling548527, 21% of total expenditure.


Extracontractual referrals costing more than pounds sterling20000

View this table:


I have shown the magnitude of the administrative task associated with extracontractual referrals, with nearly 4500 requests and invoices being received in a six month period by one health authority. Although the financial cost of their management was not calculated in this study, the audit commission estimated the average staff cost per extracontractual referral within the health authority alone as pounds sterling20-25 in 1991-2, commenting that “the average administration cost is significant, compared to the amount being invoiced for some of the cheaper extracontractual referrals such as single outpatient consultations.”6 Furthermore, it should be remembered that the handling charges incurred by providers are probably of at least a similar value; these become subsumed within their prices and thus become an indirect cost to purchasers.


Although most referrals in the present study originated from distant hospitals with which the health authority did not have a contract, 40% related to contracted hospitals. They occurred because of contract exclusions, which arose in two ways. Firstly, after specialty reviews or tendering, or both, the health authority identified preferred providers for certain specialties (dental specialties, otorhinolaryngology, ophthalmology) which were then excluded from contracts with other providers. Secondly, most providers now have a wide range of contract exclusions, usually for highly specialist treatments. These two factors, combined with the now almost universal practice of charging separately for each outpatient appointment, contributed greatly to the substantial workload related to extracontractual referrals and represented a major change from 1991, when contract exclusions were few and far between because of “steady state” and when outpatient appointments were rarely counted let alone charged for.

Individually, these factors are eminently reasonable: the ability to move contracts to preferred providers offering higher standards of care and better value for money is the raison d'etre of the internal market, and extracontractual referrals provide the essential safety mechanism to ensure that within this system the needs and wishes of individual patients and general practitioners are respected. The exclusion of highly specialist, high cost episodes of care also makes sense in that the residual contract reflects more accurately what is actually provided and is easier to manage. Finally, the separate identification of extracontractual referrals for outpatients is necessary to enable purchasers to control elective authorisation at the point of referral. The combined effect of these approaches, however, has been the administrative overload I have identified in this study.


One approach to reducing the health authority's administrative costs associated with extracontractual referrals would be to by pass the rigorous checking processes on the grounds that the handling charge in some cases equals or exceeds the cost of the referral. This would be an unattractive option, however, because, as this study has shown, considerable savings can be made in this way—important to purchasers operating within a strictly limited budget. Nevertheless, within the broader framework of the NHS such savings are more apparent than real. There is no such thing as “free” treatment and, when providers fail to obtain their proper reimbursement because of clerical inefficiency, the incurred costs become part of their overheads, to be recouped in prices at a later date.

The fact that so many inappropriate claims for payment were made illustrates the complexity of the whole process. Each year, 1 April brings boundary changes, mergers of health authorities, mergers of providers, changes in contracts, new contract exclusions, and new general practitioner fundholders, making it difficult for providers to know to whom a request or invoice should be submitted. Furthermore, some patients, happily unaware of the problems they cause, are inconveniently involved in programmes of care that straddle the end of the financial year.


The final point to emerge from this study is the financial risk to which the health authority is exposed as a result of exclusions in providers' contracts, with 0.7% (16 referrals) of extracontractual workload accounting for 21% of expenditure. All of these referrals were either tertiary or emergency admissions, for which payment is mandatory, so that no control could be exercised by the health authority over this large expenditure. This bears obvious parallels with the difficulties experienced in managing main contracts when there is an increased number of emergency admissions and, just as elective work may have to be delayed to control contracted levels of activity, so elective extracontractual referrals may have to be deferred to the next financial year. Furthermore, in the absence of mechanisms for agreeing funding for developments in service it is common for new and expensive treatments, formerly funded from research moneys, to be introduced to the market as extracontractual referrals. By their very nature they are always likely to be accessed as tertiary rather than as general practitioner referrals so that this provider led change in health service provision has the potential to undermine health authorities' commissioning strategies, forcing them to maintain sufficient financial reserves to fund expensive extracontractual referrals for a small number of patients rather than investing in services to meet the health care needs of the population. This is illustrated by the fact that within the six month period of the study the health authority received notification of five tertiary referrals for cochlear implants at a total cost of pounds sterling124000 (treatment has not yet taken place and therefore this projected expenditure was not included in the study data).


The implications of this study for future health care commissioning are considerable. In addition to anxieties about managing financial risk and about purchasing strategy, the most important concern relates to the administrative paperchase after each extracontractual referral. In just one health authority a conservative projection suggests that there will be at least 4800 completed referrals in one year. This number, multiplied by an equally conservative estimate of the total handling charge (including direct and indirect costs) of pounds sterling60, suggests that nearly pounds sterling0.3 million will be diverted from patient care to pay for the administrative processes necessary for the management of extracontractual referrals. These processes, including computer inputting as well as letters, telephone calls, and faxes between purchasers, providers, general practitioners, and patients are, as I have shown, an inevitable consequence of the present contractual system.

They occur because large numbers of episodes of care are being handled on an individual cost per case basis. Unfortunately it seems that the problem will get worse, at least for providers. More general practitioners are becoming fundholders and because of the smaller size of their contracts are more likely to manage them as cost per case contracts, with many of the problems associated with extracontractual referrals. Furthermore, even with block contracts for outpatient appointments general practitioners require much more detailed information on individual patients than health authorities do, thus placing increasing demands on the finance and information departments of providers, which do not have systems in place to deal with this bureaucracy.

A practical response would be, firstly, a limit to be imposed on the number of extracontractual referrals or cost per case episodes permitted between any provider and purchaser; where that number is exceeded in any year a contractual arrangement covering those referrals would be mandatory the following year. Secondly, agreed and guaranteed audit procedures should be in place in all providers to reassure purchasers (health authorities and fundholders) that episodes of care are correctly coded and allocated to contracts; queries on individual sets of data should be permitted only for treatments costing more than an agreed minimum.

The above measures offer a practical approach to controlling the burgeoning administration associated with extracontractual referrals and other cost per case contracts. They require prompt implementation to prevent the further diversion of scarce resources into pointless clerical tasks that have nothing to do with improving health care or the management of the NHS.

I thank the extracontractual referrals manager for generous help with data collection. This work was carried out while I was a commissioning manager for acute services in a health authority.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
View Abstract