Trends over time in prescribing by English primary care nurses: a secondary analysis of a national prescription database

Access to safe and affordable medicines is one element of all countries’ strategies for ensuring public health [1]. Some countries have legislated for nurses to have authority to prescribe medicines, as part of national strategies for improving safety and efficiency in access to prescribed medicines, particularly in primary care settings and more countries are considering doing so [2]. The introduction of prescribing by nurses in most countries has been in response to either issues of shortages of physicians in rural and remote areas [3‐8] or in response to perceived inefficiencies in primary health care delivery [9, 10]. There is variation between countries as to the classes of medicines that nurses are authorised to prescribe and the qualifications of nurses with that authority. For example all registered nurses in South Africa can prescribe, but only for some classes of medicines, while in Australia only those with nurse practitioner qualifications can prescribe from a state- or territory-approved medicine list within their scope of practice [1].

The extent to which a health care innovation becomes integrated into usual practice depends on the interaction between features of the innovation, the adopter(s), and the context [11]. Government commissioned evaluations from Northern European countries at single points in time report the innovation of nurse prescribing as well received by patients, clinically appropriate and safe [12, 13]. At a theoretical level the context for nurse prescribing is one of the shifting and contested boundaries of jurisdiction between professional groups in health care [14]. Prescribing by nurses’ remains a disputed innovation [15, 16] including by some medical professionals [17‐20]. At the same time, although nursing professional organisations in many countries have actively sought prescriptive authority [21‐24], some nurses have expressed concerns regarding appropriate training, support and remuneration for this role [25‐28] and ambivalence to a more medically orientated role [29‐31]. The contextual influences are demonstrated by the variation in rates of self-reported levels of use of prescribing authority by nurses. Surveys of nurse practitioners in the United States of America report over 90% regularly using their authority to prescribe [32, 33], in comparison to a survey of Australian nurse practitioners with endorsement to prescribe, which found that 41% did not prescribe [34]. The adoption of health care innovation also changes over time. Studies of nurse prescribing to date have drawn prescribing evidence at single points in time or from aggregated data [12, 13, 35, 36], often in relation to programmes targeted at specific patient populations [5, 37] or particular types of medicines such as analgesics or antidepressants [38, 39].

This study investigated the extent of the use of prescribing authority by primary care nurses over five years in the context of English government health policies which supported greater flexibility in health professional roles to improve access and efficiency in primary care [40, 41]. An incremental history of legislation and implementation in England [42‐47] culminated in 2006 with registered nurses with additional independent prescribing qualifications being authorised to prescribe any licensed medicine, including some Controlled Drugs, for any medical condition within their clinical competence and scope of practice (see Table 1).

This incremental history has also resulted in two types of qualification for prescribing by nurses [48]:

A previous systematic review [51] and subsequent updating found no United Kingdom (UK) studies that reported from objective prescribing data by nurses in primary care (although as the review noted some studies may have included data from primary care but it was not possible to separate from that in hospital settings). This study addressed the following research questions:

The study design was a secondary data analysis of the administrative records of National Health Service (NHS) primary care prescriptions and contextual primary care environment and workforce data in England available in the public domain.

Between 2006-2010 the number of nurses registered by their employers to prescribe with ePACT rose by 18% from 30,753 to 36,281 (Figure 1). The greatest increase was those with independent nurse prescribing qualifications (INPs) from 5014 to 12975 while the Community Practitioner Nurse Prescribers (CPNPs), who use a limited formulary, decreased by 2437.

While the numbers of nurses actively prescribing, as recorded on ePACT by 2 or more prescriptions in a year, rose in the time period by 18% from 13,391 to 15,841, this remained at 43% of all those nurses registered with ePACT to prescribe. A greater percentage of those with independent nurse prescribing qualifications registered with ePACT were actively prescribing in the time period (from 65% to 72% annually) than CPNPs, who decreased as active prescribers between 2006 and 2010 (Figure 2).

Nurses prescribed 1.2% of all items on ePACT during the time period rising from 1.1% (8,760,634/773,090,199) in 2006-7 to 1.5% (13,573,943/907,152,654) in 2009-10.

This study has used NHS prescription administrative databases to examine trends in prescribing practices of 30,753 (2006) rising to 36,281 (2010) primary care nurses in England. To our knowledge this study presents data on the prescribing practices of the largest number of primary care nurses to date.

We found that NHS employers authorised greater numbers of primary care nurses to prescribe over the time period. The percentage actually prescribing as recorded by the prescription administrative system remained static over time (43%). We have objectively quantified the percentages of nurses that are using their independent nurse prescriber (INP) qualification and demonstrated a decline in use of community nurse prescribing (CNP) qualifications, not reported previously. We found higher percentages of nurses in primary care not using their prescribing qualifications than reported by generic surveys, including primary care, in the USA [32, 33] and the UK [12, 29] but lower than Australia [34] where prescribing rights have been more recently introduced. The finding of regional variation, with areas with no prescribing by practice employed nurses, has been noted before [58] but the evidence here suggests this feature has not changed over time.

These findings suggest that this innovation has yet to be fully adopted as usual or normal practice for primary care nurses with prescribing qualifications [11]. Some UK studies have suggested nurses’ lack of employer support [30, 59] and there is a need for change champions in local health care systems [60], however these were nurses who had employer support as evidenced by the employer registering them as prescribers with the ePACT database. It raises questions as to why such numbers of nurses have gained the prescribing qualification for it then not to be used in practice. One hypothesis, derived from the literature [30, 59] and requiring testing, is that nurses who have access to a clinical mentor or supervisor are more likely to prescribe than those who no or little access. A second hypothesis that requires further investigation is that the ambivalence to this role is more wide spread amongst primary care nurses than previously reported and they are choosing not to prescribe and as such acting as ‘street-level bureaucrats’ [61] i.e. as front line staff making policy through their implementation decisions. Primary care nurses acting as street-level bureaucrats in the face of policy implementation has been noted before in the UK [62] and other countries [63].

While the number of nurses and the volume of prescribing by nurses increased over the five year period, prescribing in primary care remains an activity mainly undertaken by doctors in England. This has been noted previously [64] and in other countries where nurses prescribe in primary care [36]. We found that the largest volume of items prescribed by nurses in primary care (both INPs and CNPs) were those items used in common nursing care activities practice i.e. wound dressings, incontinence and stoma devices. Beyond these, the medicine categories where there had been the greatest increase in both volume and percentage of prescribing compared to GPs, were those that could be bracketed as health promotion e.g. contraception, smoking cessation [65]. While practice nurses have become involved in the chronic disease management processes for primary care patients [66, 67] it is not evident from this study that as a group they undertake significant prescribing with these patients. Surveys of practice nurses in the UK over the last two decades show that health promotion and family planning activities are ranked as the most frequent [68, 69]. One hypothesis, that requires further investigation, is that nurse prescribing is most acceptable to both nurses and others when it improves their efficiency in delivering primarily nursing interventions, treatments or health promotion within their scope of practice. Such investigation could include the prescribing practices of nurses with specific responsibilities for patients with long term conditions such as community matrons in the English setting [70].

We found tentative evidence that there may be higher rates of nurse prescribing in areas with lower ratios of GPs to patient populations (Spearman’s rho = -0.16) and higher levels of deprivation (Spearman’s rho = 0.19). We suggest this requires further investigation over time and in the face of predicted shortages of GPs in deprived and rural areas, where the UK and other countries have difficulty in attracting and retaining family doctors [71, 72].

Our findings of the first five years following the introduction of independent nurse prescribing, involving all classes of medicines, suggests that while the English policy objectives were for increased flexibility professional roles [40, 41], this has only been at the margins of medical practice. However, it may have significantly improved access and efficiency in health care for some groups of patients and released medical time. Prescribing by nurses in primary care for specific patient groups has the potential to release general practitioner time. Economic modelling, from one UK study, for patients with infections and those with hypertension, suggested the involvement of independent nurse prescribers was less expensive compared to a GP only prescribing model [12]. Further investigation is required over longer periods and specifically examining questions of efficiency, improved access and cost effectiveness for different patient populations.

The study has a number of limitations. As an analysis of a data set established for financial reimbursement, it cannot investigate at the patient or prescriber level. Hence, we were not able to address other aspects of nurse prescribing activities such as ceasing medications. Nor is it able to address questions such as clinical safety or health economics. The data are limited to five years of prescriptions in NHS England and other mechanisms which circumvent the ePACT database may be in place such as patient group directions [73] (known elsewhere by terms such as ‘standing orders’ [74]) with bulk purchase e.g. vaccines, masking a greater level of nurse activity in prescribing. However, despite these limitations this study provides empirical data and insights not available elsewhere as to the types of prescribing undertaken by primary care nurses, over time and from a national perspective. As such it offers some valuable information to nurses and policy makers both in the UK and elsewhere and in addition sets a research agenda for future study.

The percentage of prescriptions written by nurses in primary care in England is very small in comparison to general practitioners and there has been little change in that over five years. The adoption of any innovation in health care systems is influenced by a range of contextual and individual factors. Our findings suggest that nurse prescribing has been most frequently used in situations where it is seen to have relative advantage by all stakeholders, in particular when it supports efficiency in nursing practice, most commonly of wound and incontinence management but also the health promotion activities of nurses in general practice. It is in these areas that there appears to be flexibility in the prescribing role between nurses and general practitioners.