Article Text

Download PDFPDF

Original article
Management of patients with HIV-1 infection by pharmacist prescribers: an evaluation of practice
  1. Jane S Nicholls1,
  2. Mandeep Butt1,
  3. David Ogden2,
  4. Mel Snelling3,
  5. Rosy Weston4
  1. 1Medicines Safety Division, East and South East England Specialist Pharmacy Service, UK
  2. 2Pharmacy Department, St Georges Healthcare NHS Trust, London, UK
  3. 3Pharmacy Department, Oxford Radcliffe Hospitals NHS Trust, Oxford, UK
  4. 4Pharmacy Department, Imperial College Healthcare NHS Trust, London, UK
  1. Correspondence to Jane S Nicholls, Pharmacy Department, East and South East England Specialist Pharmacy Services, Northwick Park Hospital, Watford Road, Harrow, Middlesex HA1 3UJ, UK; Jane.nicholls{at}nhs.net

Abstract

Objectives This paper describes the self-evaluation and peer review of the practice of a group of self-selected pharmacist prescribers managing patients with HIV-1 infection in secondary care settings in the UK.

Methods Pharmacist prescribers who were responsible for managing an episode of care for adult HIV patients within defined clinical settings were invited to participate. The project lead worked with the volunteers to define, pilot and use a data collection tool for prospective data collection.

Results Four pharmacist prescribers agreed to participate and a tool was developed and used for prospective data collection over 6 months for a total of 95 patient consultations. The pharmacists were providing a number of different models of service delivery as either scheduled or unscheduled appointments, face-to-face or on the telephone. The primary purpose of the consultations ranged from initiation and optimisation of therapy to maintenance of supply of medicines and adherence advice. An attempt was made to look at patient outcomes in terms of viral load, and prescribing error rate was 1.2% of prescribed items.

Conclusions This evaluation suggests that pharmacist prescribers are able to safely and effectively manage episodes of care for patients with HIV-1 infection. Use of the data collection tool was found to be relatively simple and could be used for routine self-assessment or further study. Limitations include the small size and the self-assessment by practitioners. Further work should focus on an evaluation of the service against the BHIVA Standards 2013 and the patient experience.

  • patient outcomes
  • adherence
  • prescribing
  • non-medical prescribing

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

In the last 30 years, since HIV-1 infection was first recognised as causing significant morbidity and mortality, advances in antiviral treatments have allowed HIV disease to be managed as a chronic long-term condition. For many patients, this means taking as little as one pill a day, although adherence to therapy needs to be high to maintain long-term virological control without the development of resistant virus.1 Despite these advances in disease management, treatment requires regular monitoring for efficacy and safety with the support of a multidisciplinary team (MDT). Specialist pharmacists working in the field of HIV in the UK work effectively as MDT members often in clinic-based settings offering adherence support to patients and drug–drug interaction advice.2

In 2000, the Department of Health started a policy initiative to develop non-medical prescribing (NMP) in England with the aims of improving patient care, choice, access and patient safety through better use of health professionals’ skills and more flexible team work.3 Since 2006, pharmacists and nurses have been able to train to become independent prescribers, and more recently, prescribing of controlled drugs has been included.4 ,5 Evidence has shown that the benefits of NMP include: faster access to medicines, more flexible patient-oriented care, time savings and improved service efficiency. NMP has been found to be safe and acceptable to both patients and clinicians.6 Recent evaluations indicate that overall, nurse and pharmacist prescribing is safe and clinically appropriate.7 ,8 It is becoming a well-integrated and established means of managing a patient's condition; however, little or no specific evaluation of the services provided by pharmacist independent prescribers has been published.

Clearly, for a condition such as HIV-1 infection, where access to medicine is crucial, the pharmacist with prescribing skills can be even more effective as a member of the MDT. Not only those patients who are about to run out of medication but also stable patients in regular follow-up benefit from regular consultation with a pharmacist. Also initiating antiretroviral therapy may occur in the context of an adherence support consultation. The patient can be assessed and monitored, and have a medication review when necessary and medicines prescribed as required.

In HIV therapeutics, there are around 27 different licensed drugs and many more preparations, making prescribing accuracy challenging even for specialist clinicians. A study of 1306 prescribed items in a specialist HIV clinic in 2007 revealed an error rate of 8.7%, and 1.1% were classified as serious; undetected, they would have compromised treatment response or caused toxicity.9 For hospitalised HIV inpatients established on antiretroviral treatments, the risk is greater, with 22% having a prescribing error on admission in one study.10 This is likely to be because admitting teams cover a variety of specialties and are relatively unfamiliar with HIV infection management.

It is important that pharmacist prescribers are able to demonstrate to commissioners of health services that they are able to optimally manage patients with long-term conditions. In addition, many of these practitioners are working in relative isolation and they must be able to evaluate and reflect on their own practice. This paper describes the self-evaluation and peer review of the practice of a group of self-selected pharmacist prescribers managing patients with HIV-1 infection.

Objectives

  • To develop and pilot sets of clinical parameters that would enable pharmacist independent prescribers to examine their practice in relation to best practice as agreed collectively by the group.

  • To use these parameters to collect and analyse data in selected clinical settings.

Method

HIV pharmacists who were qualified as prescribers and who were responsible for managing an episode of care for adult patients within defined clinical settings were invited to participate. There was no existing template for collecting data during the patient consultations; therefore, the project lead worked collaboratively with the pharmacist prescribers to agree to a set of parameters with reference to a set of national standards that was being drafted at the time.1

The final parameters for data collection were agreed upon following a 2-week pilot period. They were amended again in collaboration with the pharmacist participants during a peer review meeting following the evaluation period. The anonymised data were submitted to and analysed by the project lead. Ethical approval was not required as the information was to be used for service evaluation and was not a part of a research project. To reduce selection bias, it was agreed that all patients seen for the period of data collection would be included in the analysis.

Results

Six pharmacist prescribers working with HIV patients were identified and four agreed to participate. All four worked in hospital settings. With the exception of viral load (VL) information, data were collected prospectively over a 6-month period, between October 2011 and March 2012. Box 1 gives an outline of the broad categories used to collect data during each consultation. The data collection tool including detailed categories is available on request.

Box 1

Broad categories used in the data collection tool

  • Anonymised patient identifier

  • Date of consultation

  • Type of episode/patient/location

  • Attended or DNA

  • Details of viral load/history/current measure

  • Actions arising from consultation: supply/tests/GP letter/interpretation of blood results/drug interaction/adverse drug reactions

  • Advice given to patient/types

  • Follow-up and monitoring performed

  • Details of prescribing/number of items/medicines/type

  • Details of prescribing errors made/classification/calculation of prescribing error rate

  • Details of referrals made

  • Time taken/min

Models of care

Data were collected for a total of 95 patient consultations. The pharmacists were providing a number of different models of service delivery as either scheduled or unscheduled appointments (see table 1). It was common for patients to consult the pharmacist for advice and supply of medicines without an appointment (44/95 consultations). They did not attend six of the scheduled appointments.

Table 1

Location and type of patient consultation

Prescribing

Table 2 shows the number of individual patients treated by each pharmacist and the number of consultations, as well as data showing the number of items prescribed. Eight patients were treated twice and one patient four times. However, patients were not always seen in person and some items were prescribed following a telephone consultation. Antiretrovirals (ARVs) were most commonly prescribed (159 items), although non-antiretroviral medications were also occasionally required by the patients (24 items). A prescription was not always needed, however, and no prescription was issued following 21 consultations. In almost 30% of consultations, pharmacists provided an urgent supply of medicines to patients who needed them (28/95).

Table 2

Number of patients seen and items prescribed by pharmacists

To give an idea of the nature of the consultations between the pharmacists and the patients, they were classified into complex, switch, starter, maintenance and other, according to the primary task required of the prescriber in addition to routine assessment and monitoring (table 3).

Table 3

Type of consultation

Prescribing errors

The prescriptions were clinically verified by another pharmacist within the hospital service at the point of dispensing so it was possible to capture data related to pharmacist prescribing errors. Data were not available for 13 of the items prescribed during four patient consultations. From the remaining total of 170 items prescribed, there were two errors detected on clinical verification giving a prescribing error rate of 1.2%.

On each occasion the prescribing pharmacist was consulted by the dispensary pharmacist and the prescription amended so the event was classified as a near miss, no harm occurred. Details of the potential adverse events are outlined below:

  • Wrong dose error: Correct drug prescribed but the dose was not adjusted upwards to take account of a potential drug–drug interaction

  • Length of supply not stipulated by prescriber

Advice and referral

Patients were usually offered advice as part of their care and during 29 (31%) consultations the advice was directly related to their medication as outlined in table 4. In the majority of cases this included advice on the drugs, how to take them, aims of treatment and the risks of developing viral resistance. Other examples of information given related to follow up and monitoring (38%) and what do if there were problems out of hours (16%).

Table 4

Type of advice given about medication (n=29)

Other functions of the prescribers included ordering tests including therapeutic drug monitoring (19%). In addition, a referral was made to a more senior clinician or to another service during 22 (23%) consultations. Some patients were referred to their General Practitioner (GP) for the investigation of non-HIV related symptoms and some were referred to other services, for example, dietician, mental health services.

Patient outcomes

When the participants met to review the data collected and the use of the data collection tool, it was agreed that the collection of VL data may be useful as a measure of successful therapy in the long term, and it was agreed to include this measure in the final data collection tool.

In practice, much VL data were not available retrospectively and could not be attributed solely to the care of the pharmacist prescriber. However, there was a subset of 12 patients for whom VL data were available and in whom ARV treatment was initiated at the start of the evaluation and their therapy subsequently managed in the first month by a pharmacist prescriber. The VL results for this subset showed 9 (75%) patients had an undetectable VL of <50 copies/mL at 6 months, 2 (17%) had VL <100 cop/mL and in one patient the follow-up was not available. In addition, the patient record indicated no new resistance in these 12 patients (100%) at 1-year follow-up, including two patients with virological failure (>400 cop/mL).

Discussion

This small evaluation suggests that pharmacist prescribers are able to safely and effectively manage episodes of care for patients with HIV-1 infection. The tool that developed was found to be relatively simple to use and could be used for service evaluation by others.

The four HIV pharmacist prescribing services have developed flexibly in response to an identified need in the host organisation, for example, urgent supply, telephone support and clinic appointments. Discussions with the pharmacists indicated that their scope of practice had changed and developed over time as the experience of the practitioners and their prescribing services had developed. The majority of the care was focussed on the initiation and optimisation of therapy with antiretrovirals.

Unscheduled consultations accounted for 46%, and nearly 70% of these consisted of reviews conducted over the telephone. In a patient group where medication is usually only available from a tertiary referral centre, where patients may not live locally and where adherence is crucial for disease control, safe and convenient access to medicines is vital and the pharmacist prescribing role forms an important part of the care of the multidisciplinary team.

Half of the patient episodes were for scheduled care provided in a clinical setting by one pharmacist. This model had the potential to allow the care provided to be examined over a longer period; however, most patients were only seen once in six months. A longer evaluation period might have allowed for a more detailed examination of patient outcomes and this was demonstrated when the participants decided to capture VL data retrospectively. The small subset of 12 patients for whom the pharmacist initiated ARV treatment showed promising data and there was no emergence of new viral resistance in the year since starting treatment. However, a larger evaluation is required to confirm this finding. Since the conclusion of this evaluation, detailed Standards have been published11 and measurable outcomes have been agreed. These do not specify the maintenance of undetectable VL but they do give a target for the emergence of new viral resistance in patients prescribed first line therapy as not more than 2% (Standard 4).

In all models of care, patients were being treated holistically by the multidisciplinary team and the function of the prescribing pharmacists within the team was emphasised by their ability to recognise the need for other specialist input. Referrals were made to other practitioners in 23% of consultations. Although it was not captured in this evaluation, the pharmacists recognised that patients were also frequently referred to them for expert adherence support. A consultation did not necessarily result in a prescription.

The clinical verification of the prescriptions suggested the majority were appropriate for the patient although a low number of errors were detected (1.2%). This finding reinforces the need for an independent check of prescriptions.12 The numbers of prescriptions was relatively small, making it difficult to draw firm conclusions; however, another study with medical prescribers reported an error rate of 8.7%9 suggesting an area for further study.

When interviewed about the mistake, the pharmacist said:

“This particular consultation was complicated in that I had no interpreter, and this may have distracted me from considering the interaction between ritonavir-boosted atazanavir and efavirenz. Clinically, it was a complex situation where I was prescribing an intensification regimen to a patient with rising viral load on Atripla therapy to prevent virological failure and resistance following Consultant advice. Fortunately, the patient's viral load quickly suppressed, boosted atazanavir was stopped and [the patient] now is maintained on Atripla alone.”

Advice and information were given on a wide range of topics related to medication and other aspects of treatment. In a third of consultations, the advice was recorded as specific to medication although this seems a low result since it seems likely that most appointments would have included a check related to medication and perhaps this aspect of care could be evaluated in more detail in future work. The importance of this type of adherence support is emphasised by the recent findings of Saberi's systematic review, which reported a positive association between HIV pharmacist activities and improvement in adherence and VL suppression.2 Drug regimens must be tailored to meet the particular requirements of patients and pharmacist prescribers possess the expertise to be able to balance patient preference, viral sensitivity, ARV pharmacology and patient comorbidities.

Although attempts were made to collect data from all patients who were seen in the evaluation period, the limitations include the self-selection of the pharmacists and the fact that we relied on self-assessment. The pharmacists involved reported benefits from the process of peer review and further work could be done to include more participants and the collection of data over a longer period, as well as collecting information on the patient experience. The new Standards published in 2013 could provide a basis for this work. In addition, an independent economic evaluation of the impact of the pharmacist service on patient care may help support the business case for new and existing services.

This is the first published evaluation of pharmacist prescribers working in this field in the UK. The BHIVA Standards 2013 state that, ‘Safe and effective prescribing of appropriate combinations of ARVs is crucial to maximise benefits, minimise adverse effects, avoid drug interactions and reduce the emergence of drug-resistant strains of HIV’. At a time, when there is pressure to measure the effectiveness of clinical services to patients, this evaluation contributes to evidence that pharmacists are a valuable part of the multidisciplinary team in the successful management of HIV infection.

Key messages

  • What is already known on this subject?

  • Literature exists on the general contribution of clinical pharmacists to the outcomes of patients being treated for HIV-1 infection but there is nothing that specifically examines the activities and the effectiveness of a group of pharmacist prescribers in this field.

  • There was no existing template for collecting data during the patient consultation in HIV outpatients.

  • Studies exist of prescribing errors made by physicians in HIV outpatient clinics and inpatient settings but there are no data on the corresponding error rate of pharmacist prescribers in these settings.

  • What this study adds

  • This self-evaluation of 95 outpatient consultations by four pharmacists suggests that pharmacist prescribers are able to safely and effectively manage episodes of care for patients with HIV-1 infection.

  • A data collection tool that allows pharmacist prescribers to examine the effectiveness of their practice in HIV outpatient settings has been developed and tested.

  • A total of 170 items prescribed by pharmacists were clinically verified by dispensing pharmacists acting independently and two errors were detected giving a prescribing error rate of 1.2%.

  • Cautious interpretation of a subset of patients where results were available at 1-year follow-up suggests favourable patient outcomes in terms of no new emergent resistance acquired, but further study is needed.

Acknowledgments

The authors would like to thank Sheena Castelino for participating in data collection and the local pharmacy teams for providing feedback on prescribing accuracy.

References

Footnotes

  • Collaborators Sheena Castelino.

  • Contributors The contributors to this paper have agreed to have their names listed as authors at the start of this paper, and in addition, a contributor who was not an author has agreed to be named at the end as she took part in the evaluation but did not contribute to the journal article. JSN is listed as the guarantor of the paper and accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish.

  • Competing interests None.

  • Ethics approval This was a service evaluation and did not require ethical approval. Patient data were collected without identifiers to allow for comparison of practitioner performance. Local Governance procedures were followed.

  • Provenance and peer review Not commissioned; externally peer reviewed.