Return to Home Page
Primary Care Pharmacy March 2000 Vol 1 No 2 p52-54

Practice prescribing systems: an obstacle to progress?

By Gill Speak PhD, MRPharmS

Introducing a prescribing policy can improve prescribing management

Primary care groups (PCGs) are approaching the end of their first year, with the more progressive ones already moving toward trust status. In a short space of time, there has been a shift in primary care toward corporate working at a local level, with some devolution of commissioning from health authorities to the PCGs. The next phase will entail a regrouping of PCGs into larger units - primary care trusts - modelled on secondary care. In order to make the transition from PCG to primary care trust, current systems will have to undergo radical changes. Organisation at practice level is the key to containing prescribing costs. For many pharmacists now working with PCGs, there has not been time to assess practice prescribing systems on a large scale. However, it is clear from recent published studies that many primary care pharmacists are aware that practice prescribing systems threaten the PCG's viability. What exactly does the term prescribing systems embrace and what is the threat to prescribing budgets? This article will seek to address some of the more prevalent weaknesses of prescribing systems and ways in which a pharmacist might help in resolving them.
In a study conducted in Northern Ireland some years ago, practices were assessed on the basis of 26 prescribing indicators.1 These indicators measured good management, not clinical quality, in repeat prescribing. Pres-cribing management is often overlooked in primary care. This is not to say that practice managers lack training or experience but, in common with the pace of change in both workplace and home today, increasing complexity demands specialist input. The nature of that input is determined by the purchaser-provider chain that drives clinical quality in prescribing. In other words, the end-provider in the prescribing process is the pharmacist, who is, therefore, most likely to recognise when systems fail.
Unless all practice business is recorded on a computer, data will be unreliable for audit and review purposes and, more importantly, work will be being duplicated. This wastes time and can result in corners being cut in vital areas to keep on track. One way of encouraging better prescribing management is to ensure that the practice has a prescribing policy, one of the criteria used by the NHS Executive to define high quality medical services.2

Prescribing policies

The purpose of a written prescribing policy is to guarantee a standard of working that is not subject to external pressure or to individual priority setting. In order to meet the NHS criteria for high quality medical services, a prescribing policy should include:

  • review of patient care at appropriate intervals
  • Compilation of an up-to-date list of repeat medication
  • Recording of current and recent drug therapy (including doses)
  • Use of a formulary
  • Increased generic prescribing

In practice, the prescribing policy will need to provide detailed guidance on each step of the prescribing process if the above criteria are to be satisfied.

Turnover time and prescription collection

A prescribing policy must acknowledge problems caused by, for instance, public holidays, lack of resources and doctor availability and it must provide a realistic means of delivering the service offered to patients. So, the repeat prescription turnover time should be explicitly stated within practices. It is important that the turnover time set by the practice is realistic and is sufficiently long to allow for checking, signing, filing and review of prescriptions before collection by the patient. A procedure should be put in place to assist staff in dealing with patients who have returned sooner than expected to collect their prescription. Without such guidance, good management quickly cedes to crisis management, queue jumping and patient-led systems.
In addition, a prescribing protocol needs to anticipate problems and stipulate arrangements for receipt and collection of prescriptions at weekends or during public holidays. Requests for repeat prescriptions could be stamped with the date and time as the prescription box is emptied. This would ensure that priorities are not changed to suit some patients at the expense of others. This information could then be used to audit overall efficiency, as well as minimising disruption in the practice and inconvenience to the patient. It might be feasible to allocate a time slot for collection of repeat prescriptions by homes, pharmacies, suppliers, etc, outside surgery hours, when reception is less busy. These prescriptions could be signed for in a bound book to reduce disruption and possible duplication of prescriptions if those requested cannot be found.

Lost Prescriptions

A policy might contain a flow chart for dealing with lost prescriptions (see Figure 1). This would ensure a smooth, efficient and standardised approach, which would minimise the time spent by receptionists away from their posts and would define the point at which a duplicate script is issued.

Figure 1
Figure 1 Lost Scripts

Repeat prescriptions

The task of producing repeat prescriptions could benefit from a similar review. Many practices still accept telephoned requests throughout the working day. This blocks more urgent calls, inconveniencing both doctors and patients. It also encourages a "one-stop" approach to printing the prescription, with no written record of the request. Items that are notoriously difficult for receptionists, such as appliances and dressings, tend to be added as requested, because this is easier than trying to match up the patient's description with the computer listing. Nor does a "one-stop" system allow scope for checking whether the quantity requested by the patient is reasonable. Understandably, this can be a costly burden on the practice and can affect the practice workload.
A practice policy should, therefore, specify that the important task of producing repeat prescriptions be undertaken by named receptionists between agreed times and in a place away from interruptions. Requests not received on repeat prescription slips should be discouraged, to facilitate the task of matching the request to the patient's record. Some practices operate a two-tier system, promising a faster turnover for requests received on the practice repeat prescription slip. Use of repeat prescription slips also distinguishes between repeat and acute prescriptions. Externally produced documentation rarely, if ever, acknowledges this distinction. When Zermansky reported that 66 per cent of repeat prescriptions studied showed no evidence of authorisation by a doctor, he highlighted, among other things, the need for a definition of a "repeat prescription".3 The fact that a patient has had the medication previously does not authorise them to request it again. Thus, a prescribing policy should include such a definition, to discourage a modern tendency to indulge in "mail order shopping" for medication. The accepted definition of repeat prescriptions came from the National Audit Office in 1994, which described them as "a prescription issued without a consultation".4

Prescribing interval and review

Prescribing intervals can become rather flexible, according to patient demand. There is a tendency for patients and practice staff to imagine that prescribing at longer intervals will reduce the practice workload. However, in practice, the facility for monitoring compliance and therapeutic response is lost. Zermansky found no evidence in 72 per cent of repeat prescriptions studied that a review had taken place during the previous 15 months.3 Longer prescribing intervals also lead to increased costs, because of difficulties in estimating appropriate quantities of test strips, inhalers, vials and other original packs. A prescribing policy should define the number of days' treatment normally issued on a repeat prescription and, where possible, list the exceptions to this rule. Such a list might include hormone replacement therapy products, the contraceptive pill and drugs taken "as required" rather than regularly.
A prescribing policy should also address the concept of review, so that patients understand the need to see their doctor at specified intervals. As Zermansky pointed out, computer software does not often support practice staff in identifying early or late requests, nor does it stop the user from reauthorising the prescription of drugs. A policy would need to state those review periods for given drugs, to avoid a gradual "rounding up" to the longest period.

Nursing and residential homes

Any prescribing policy must address monthly requests for prescriptions from nursing and residential homes. Community pharmacies reportedly need about a week from receipt of the prescriptions to delivery to the home, so requests are sent to the practice about halfway into the monthly cycle. There are concerns around anticipated needs for so-called "prn" items and original packs. That apart, the practice needs to be geared to process these requests, not on receipt, but to a calendar. This facilitates recognition of requests coming in too soon and it assists the homes to work to a plan.
Some homes still send in requests for individual patients as they run out of their medication. Reasons given for this lack of uniformity may be lack of storage space to house one month's medication for all residents at one time, or it may be that the practice is unwilling to issue one-off prescriptions to bring new medication into line with the others. Automated prescribing, in this case, encourages inflexibility. A calendar plan agreed with the home and pharmacy should ensure that prescriptions are not requested erratically. This avoids undue wastage of staff time or medication, both of which are costly commodities, and it also facilitates clinical review.

Discharge medication

Another area of prescription requests that gives cause for concern is discharge medication given to patients on leaving hospital. Zermansky found that hospital initiated drugs were "particularly prone to be added without clinical appraisal". In 56 per cent of repeat prescriptions analysed, there was no instruction to continue the drug long-term.3 McGavock et al found that only two out of 57 practices "always checked hospital prescriptions for errors and to see whether there were any therapeutically important differences between a patient's previous maintenance therapy and what had then been recommended."1 This results, at best, in long lists of obsolete drugs on the practice computer, increasing the potential for error and wastage at the point of issue. At worst, it leads to duplication of therapy, with a future drain on health care resources to treat unplanned clinical outcomes. A prescribing policy should stipulate who should deal with each stage of the recording and management of discharge notes and the conditions under which they should be archived.

Patient demand

One of the biggest problems in general practice prescribing systems is patient demand. Receptionists need to know how to handle this professionally, without unnecessary disruption of the work schedule. The growing problem of requests for acute prescriptions (ie, requests for medication to treat what is perceived to be both an urgent need and a self-limiting condition) is largely unstudied. The practice policy should specify appropriate responses to requests to treat, for example, coughs, colds, head lice, infantile teething and febrile conditions. Several management options are available: the practice nurse or health visitor; the pharmacist; and finally, the doctor. But a list of medicines that the practice will not prescribe could also be displayed on reception and listed in a prescribing policy. A standardised method of documenting these requests is advisable, to help with longer term resource management and clinical audit. A request form ensures that the doctor receives adequate information for patient management. From the receptionist's viewpoint, if the request is date-stamped there can be less likelihood of patients returning at a busy time demanding immediate satisfaction.

Conclusion

All of the solutions to the commonly encountered issues discussed above should be included in a prescribing policy. They are designed to help manage time and resources effectively. If fully utilised, a computer system should help to implement these provisions. However, as stated at the beginning, much depends on how the computer is used. Slavish dependence on computerisation can create new problems. Batch printing of repeat prescriptions, may be efficient but it reduces the likelihood of the doctor thinking critically about each prescription.5
Furthermore, without a policy demarcating lines of responsibility and promising remedial action for failure to observe that policy, there is little control over who adds or edits prescribing information. Attempts at formulary implementation, generic prescribing, or cost-effective prescribing usually fail here.
The number of prescribing policies in circulation suggests that there can never be one formula to suit every situation. However, putting together a template might be a suitable starting point for discussion. Both doctors and practice managers should discuss the clinical and managerial issues around prescribing but it is strongly recommended that receptionists' views are considered, too. Practice and theory rarely coincide and "shop-floor workers" know how and why the short cuts are taken, which ultimately cost the practice dear. Communication is the key.

Dr Speak is a prescribing facilitator in Lancashire

References

1. McGavock H, Wilson-Davis K, Connolly JP. Repeat prescribing management - a cause for concern? Br J Gen Pract 1999;49:343.
2. Allocation of DDRB £60m to GPs. HSC 1999/107. London: Department of Health, 1999.
3. Zermansky AG. Who controls repeats? Br J Gen Pract 1996;46:643.
4. A Prescription for Improvement: towards more rational prescribing in general practice. London:Audit Commission, 1994.
5. Harris CM, Dajda R. The scale of repeat prescribing. Br J Gen Pract 1996;46:649.