Intended for healthcare professionals


One day or day one? Uptake of new prescribing guidance in general practice

BMJ 2019; 367 doi: (Published 02 October 2019) Cite this as: BMJ 2019;367:l5652

Linked research

Variation in NHS clinicians' response to new guidance

  1. Emma Wallace, senior lecturer in general practice1
  1. 1Department of General Practice, Royal College of Surgeons in Ireland, 123 Stephen’s Green, Dublin 2, Ireland
  1. Correspondence to: E Wallace emmawallace{at}

The pace of change is still highly variable, with implications for patient care and health expenditure

Internationally, the drive to improve healthcare quality and cost effectiveness is underpinned by evidence based medicine. However, the pace of change in healthcare is rapid and the implementation of new evidence into clinical practice is suboptimal.1 General practitioners offer comprehensive care at the first point of access for undifferentiated problems. Their role has changed substantially over time, with increasing demand for services coupled with the need to adopt medical and technological innovations swiftly.

In a linked article in The BMJ, Walker and colleagues (doi:10.1136/bmj.l5205) provide a large analysis of general practitioner uptake of prescribing recommendations.2 In healthcare, diffusion of innovation describes the process whereby warranted change is adopted into clinical practice. Typically, the rate of change varies over time with a small number of early adopters, a rapid middle adoption phase as momentum gathers, and a slow third phase as the innovation becomes embedded over time.3

The current study used OpenPrescribing (, an openly accessible anonymised database of all primary care prescribing by more than 8000 general practices in England. The authors examined the timing of response by general practitioners to new prescribing guidance for two drugs over a five year period. The first was a switch to generic oral contraceptive desogestrel from the branded form (Cerazette), and the second was to change the first line antibiotic choice for treatment or uncomplicated urinary tract infection (UTI) from trimethoprim to nitrofurantoin.

The study reported substantial heterogeneity between general practices in the length of delay before a noticeable change to prescribing, with an interquartile range of two to 14 months (median 8 months) for desogesterol, and five to 29 months (median 18 months) for UTI treatment. Although the pace of change—once started—also varied substantially, most practices showed a significantly favourable trend in changing prescribing practice for both treatments over time (78% for the desogestrel change and 82% for the change in UTI antibiotic).

Although the recommended change to oral contraceptive prescribing was clear cut and based on cost savings, the change in first line antibiotic for UTI required clinical judgment. Guidelines recommended a switch from trimethoprim to nitrofurantoin in late 2014,4 which was followed by a financial incentive introduced in 2017. However, trimethoprim was still recommended as a first line antibiotic choice for uncomplicated UTI by other UK clinical guidelines during this period, which could have affected prescribing decisions.5 Having a clear and consistent message across clinical guidelines is important to support the uptake of new practices.

Internationally, the impact of similar prescribing initiatives has been mixed. In Ireland, a “preferred drugs” initiative had limited impact on prescribing behaviour.6 The dissemination strategy for this initiative primarily involved writing to general practitioners without any financial incentive.6 A European study of generic risperidone substitution showed a steady reduction in risperidone prescribing overall, but had heterogeneity between countries in the uptake of generic prescribing.7

Previous research also shows substantial prescriber variation in relation to antibiotic prescribing and higher risk treatment.89 Unwarranted variation adversely affects both the quality of patient care and healthcare expenditure. A systematic review of reviews identified four themes underlying the gap between evidence and practice: external context, organisational factors, individual professional factors, and intervention characteristics.10 External contextual factors included existing policies, incentivisation structures, stakeholder buy-in, and available infrastructure. Organisational factors included the prevailing culture, resources, and ease of integration with existing processes. Individual professional factors included underlying values and competencies while intervention factors included evidence of benefit, ease of use, and adaptability to local conditions.10

Strategies to improve the implementation of complex interventions in general practice include audit and feedback, educational meetings, educational outreach, and reminders. Effectiveness varies, but all are associated with small to modest improvements in practice.11 Barriers to implementation include lack of access to the relevant evidence, concerns about its quality and applicability, patient related factors, and time constraints.12 Qualitative research indicates that doctors can struggle to unlearn old evidence, and clinicians report tensions between evidence and real world context.13

Research on the diffusion of innovation suggests the following steps to help maximise uptake of new recommendations.314 The first step is to identify and develop trusted opinion leaders to communicate, promote and support change. The second step is to identify innovators and invest in early adopters of change, moving from a culture of compliance to support. The third step is to make the activity of early adopters visible to encourage others.3 The OpenPrescribing platform, an opportunity to explore the prescribing patterns of similar practices, has great potential to inform and encourage change. This data driven approach can act as an enabler of evidence based medicine. Walker and colleagues have identified substantial variation between general practices in uptake of new prescribing guidance. Future research should now focus on why this variation is so large, and how open data can help to drive timelier uptake.


  • Research, doi: 10.1136/bmj.l5205
  • Competing interests: I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: None.

  • Provenance and peer review: Commissioned; not peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription