VOL: 99, ISSUE: 04, PAGE NO: 36
Nicky Brooks, MPhil, BSc, DPSN, RGN, is research coordinator, De Montfort University;
Frank Durning, BSc, RGN, is NHS walk-in centre manager, division of primary care, Loughborough General Hospital, Charnwood and NW Leicestershire Primary Care Trust; Ian Bell, PhC, BA, DipHS, is principal pharmacist, Leicestershire Partnership NHS Trust; Joanne Charles, DipClinPharmacy, BPharm, is pharmaceutical advisor, Leicestershire Health Authority
Nicky Brooks, MPhil, BSc, DPSN, RGN, is research coordinator, De Montfort University;
Patient group directions (PGDs) are specific written instructions, which enable nurses to supply and administer drugs in response to the generalised direction of a doctor, rather than a patient-specific prescription (Box 1) (Department of Health, 2000). The increased use of PGDs reflects current policy in which nurses are extending their role within the management of medicines to provide more convenient and patient-centred care (DoH, 2000).
To date, there is limited evidence of nurses’ competence in the utilisation of PGDs. Nurse-led walk-in centres have been accused of not yet being on a ‘firm footing’ (Mountford and Rosen, 2001). The issues surrounding the supply of and resistance to antibiotics (DoH, 2000) and the implementation of eight antibiotic PGDs at a local NHS walk-in centre prompted the authors to conduct this study.
The NHS walk-in centre was established in July 2000 at Loughborough General Hospital. It provides a service from 7am until 10pm seven days a week and is open 365 days a year. Nurse advisers lead the service for the treatment of minor illnesses and injuries, and the centre employs a GP on a sessional basis.
At the time this study was undertaken there were eight PGDs in use at the centre, all of which were developed locally by a multidisciplinary team including a pharmacist and a microbiologist. These PGDs were for:
- Co-amoxiclav (for human and animal bites);
- Flucloxacillin - for soft tissue and skin infections;
- Erythromycin - for co-amoxiclav or flucloxacillin sensitivities);
- Chloramphenicol eye drops/ointment (for bacterial conjunctivitis);
- Trimethoprim (for urinary tract infections);
- Amoxicillin (for middle ear infections and tonsillitis).
GPs’ experiences of prescribing antibiotics to the general public serve as a useful comparison for nurses supplying antibiotics via PGDs. The prescribing of antibiotics, particularly by GPs, has been explored extensively over the past 20 years because of concerns of over-prescribing. Over-prescribing has been attributed to patients’ expectations (Macfarlane et al, 1997), GPs’ perceptions of patients’ expectations and the belief that a prescription can help preserve the doctor-patient relationship (Butler et al, 1998).
Primary care, where most infections are treated, accounts for about 80 per cent of antibiotic prescribing (Standing Medical Advisory Committee report, 1998). The cost of such prescribing is no small matter. For example, in 2000 prescribing costs amounted to approximately £143m according to the Prescription Pricing Authority (2002). Inevitably, therefore, the SMAC report directs many of its recommendations about good practice towards primary care, while recognising that it is the responsibility of everyone, health professionals and the public alike, to treat antibiotic agents as a ‘valuable’ resource.
The expansion of nurses’ role in the management of medicines continues through independent nurse prescribing, extended independent nurse prescribing (the formulary of which includes nine oral antibiotics), PGDs and supplementary prescribing. It is essential that nurses scrutinise their practice and learn from other prescribing professionals about judicious medicines’ management.
The trust’s local protocol guided the project ensuring that the necessary steps were taken in seeking approval for the project and in the handling and storage of confidential patient data. A case study design was chosen to focus on antibiotic provision in an NHS walk-in centre.
Two methods were used to establish the degree of compliance with the antibiotic PGDs: a clinical audit of all patients who were prescribed an antibiotic from October 2000-March 2001 and the assessment of staff’s knowledge of antibiotic PGDs. This article presents the findings from the clinical audit.
The stages of this audit were designed according to the criteria following on from the National Centre for Clinical Audit (1997). Thus criteria were taken from the individual antibiotic PGDs. In addition good practice criteria were agreed by the steering group and included in order to scrutinise the compliance and completeness of the clinical records. For example, good practice criteria agreed for the PGD for flucloxacillin for wound infection was to check for the presence of diabetes mellitus.
The clinical records held on a computer database were reviewed against the agreed criteria, made anonymous and analysed using the data package SPSS (version 10.0).
During data collection (October 2000-March 2001) 1,169 patients were supplied antibiotics (Box 2), representing 11.6 per cent of patients treated at the centre during that period. Of the 1,169 patients 847 (72 per cent) were assessed and supplied with antibiotics by the nurse. The remaining 322 patients were prescribed antibiotics by the GP. This may have been due to several factors: the patient’s infection was outside the remit of the eight PGDs, the patient may not have fulfilled all the criteria for eligibility for supply of an antibiotic by a PGD or the doctor may have been the health care professional who assessed the patient on presentation at the centre.
The utilisation of PGDs marks a more legitimate departure for nurses into wider medicines’ management. Previously, nurses supplied and administered medication under ‘group protocols’, which were never clearly defined in law (Baird and Morgan, 2001). This practice was repeatedly brought into question culminating with the Crown Review (DoH, 1999), which developed stringent criteria and called for a clarification of the law.
This study shows that nurses working at an NHS walk-in centre were able to supply antibiotics in a judicious manner according to locally agreed PGDs. More than 99 per cent of the drugs administered were the correct drug and dose according to the PGD (Table 1). Nurses have previously demonstrated consistency with following procedures and policies. Manias and Street (2000) showed qualitatively that nurses were able to adhere closely to policies and procedures, which gave them greater confidence in discussing treatment options, legitimising activities and demonstrating safe practice.
The judicious use of PGDs is also encouraging in terms of patient safety and record-keeping. Careful assessment and documentation are essential because staff at walk-in centres do not have access to any previous medical records of patients who ‘walk through the doors’. The completion of a PGD can be seen as an effective record-keeping tool ensuring that the relevant information is passed to the GP for completeness of medical records.
The judicious supply of antibiotics via PGDs could, however, be questioned as only 63 per cent of patients’ computerised records confirmed that the patient had no allergy or contraindication to the supplied antibiotics. Best practice should be to record patient-reported allergies and seek further clarification, despite the fact that this has sometimes been found to be an unreliable indicator of potentially serious reaction (Falconer and Gardener, 2000) and that patients’ self-perception of drug allergies are not always grounded in medical reality (Reeves et al, 1998). The consistency of recording of drug allergies is also an issue in hospital care (Selbst et al, 1999) and highlights the need for ongoing clinical auditing to ensure that health care professionals maintain consistent, comprehensive clinical records.
Less than one per cent of drugs supplied to patients at the walk-in centre were not given according to PGDs (Box 3). The failure to comply with PGDs constituted two doses not being recorded. This does not automatically mean that the patients did not receive the correct dose. Three patients who were given amoxicillin for otitis externa (infection of the outer ear) may also have had symptoms of middle ear infection. In the presence of outer and middle ear infection staff at the NHS walk-in centre supply amoxicillin. The patient who was given erythromycin as the first-line drug for tonsillitis may have had an allergy to amoxicillin that was not recorded. These examples highlight the importance of up-to-date record-keeping and the importance of recording the clinical encounter in sufficient detail to justify actions according to local policies and procedures.
The prescribing and supply of antibiotics and the ratification of PGDs is a minefield. This study has looked at compliance to locally agreed protocols in the form of local PGDs, which are in keeping with other national resources such as PRODIGY - the computer-based support system for GPs.
However, in the case of acute otitis media or otitis externa, analgesics are also suggested as first-line treatments, either as an alternative to or in addition to antibiotics. In the case of acute otitis media, many infections are viral and 80 per cent of uncomplicated cases are resolved spontaneously. A ‘wait-and-see approach’ for up to 72 hours for children who are not systematically unwell was found to be effective in a randomised controlled trial (Damoiseaux et al, 2000).
The retrospective clinical audit of the computer records presented difficulties. The software package was not sophisticated enough to be able to search for patients under key headings and this meant that every patient record had to be reviewed (more than 10,000). A new version of this software is being piloted with the intention of rolling this out nationally as soon as possible.
There were two systems of clinical record-keeping in operation for the supply and administration of antibiotic PGDs: paper and computer. As the results of the computer records showed that 49 per cent of the criteria had not been recorded (Tables 2-7), 20 paper records were randomly retrieved and matched with the computer records to determine whether staff were using both systems to form ‘a complete record’.
The results of this exercise demonstrated that when combining the computer and paper systems, 53 per cent of the records were complete, 32 per cent had one item missing, 5 per cent had two items missing, 5 per cent had three items missing and 5 per cent had five items missing. This shows that staff were using both record systems, and this practice has since been changed to ensure that record-keeping is as streamlined as possible.
Record-keeping in relation to antibiotics has also been an issue for hospital doctors (Seaton et al, 1999). The proportion of antibiotic indications and temperatures recorded in patient notes was assessed over four periods with feedback in between. Recording levels were 64 per cent at the baseline and rose to 85 per cent after the initial feedback. However, the programme showed that while audits and feedback improved the standards of record-keeping, sustaining the improvements was difficult. What this centre’s and Seaton et al’s (1999) studies highlight are the complexity and difficulty of maintaining complete records.
The UKCC (1998) advocated good record-keeping as an essential and integral part of nursing care. Nursing practice consistently identifies room for improvement in record-keeping. Reasons for not documenting care include: a lack of confidence and difficulty with written expression; time available for documentation; difficulty with the format of the documentation; viewing staffing levels and record-keeping as a burden separate from and in addition to nursing (Mason, 1999).
Incomplete documentation does not necessarily equate to poor standards of care (Anderson, 2000). However, in view of the relationship between increasing litigation and poor documentation (Byrne, 1999) record-keeping must be valued and invested in. Clinical audits can help practitioners reflect upon and improve their practice. In light of the recommendations of the Kennedy report (DoH, 2001) clinical audits are recommended as compulsory for all health care professionals who provide clinical care and active participation should be part of terms and conditions of employment.
The steering group assumed that auditing the computer records would provide sufficient detail. The random sample of paper records matched to the computer records identified that this was not the case. However, it still showed that there is room for improvement in the completeness of clinical records. However, this study demonstrates the importance of streamlining clinical records and with advances in electronic patient records may contribute to this being overcome in future.
Recommendations for further investigation include:
- This study only explored the supply and administration of antibiotics. Post-treatment outcomes such as the effect of the treatment and the degree of compliance should be studied in future;
- This study only investigated patients who received antibiotics. A future study should look at people who were refused antibiotics and determine whether this was appropriate and what other ‘treatments’ if any were used;
- Re-auditing the clinical records now that the PGDs have been in operation for a longer period and staff have received the necessary education and training may show a better picture in terms of compliance and completeness;
- In future, the capture of patient data should be facilitated by advances in information technology.
Evidence from this study goes some way to demonstrate that nurses are supplying and administering antibiotics judiciously and safely according to local policy. However, it should be taken into consideration that only 63 per cent of patients had allergy or contraindication checks recorded on their computerised records - this may well be a record-keeping issue.
A clinical audit has identified that there is room for improvement in the completeness of clinical records, and has proven to be an effective method in identifying areas of record-keeping that need further development. The findings of this study are transferable to other care environments where health care professionals seek to deliver and maintain high standards and a consistency of treatment for the patients they serve.