Reducing antibiotic prescribing rates in young children in an outpatient primary care setting—a systemwide quality improvement initiative

Abstract

Objectives

To improve antimicrobial stewardship (AMS) and reduce unnecessary antibiotic prescriptions in young children in a British primary care setting.

Methods

Forty-nine general practices in the South of England each hosted a 1 h in-house workshop, facilitated by trained local pharmacy professionals. This type of educational outreach approach using TARGET (Target Antibiotics Responsibly, Guidance, Education and Tools) antibiotic materials has previously been shown to reduce antibiotic dispensing in a UK primary care setting. The workshop included a review of antibiotic prescribing data, a presentation on paediatric AMS showcasing locally agreed paediatric prescribing guidelines and safety-netting resources from the Healthier Together website, and formulation of a local action plan. The primary outcome measure was total oral antibiotic prescriptions (‘items’) dispensed per 1000 patients aged under 5 years for the year after the workshop, compared with the previous year’s dispensing.

Results

The median prescribing rate for children under 5 years of age changed from a baseline of 48.9 per 1000 patients prior to the intervention to a new median monthly prescribing rate of 39.0 per 1000 patients following the intervention. There was no increase in paediatric presentations to primary care following the intervention.

Conclusions

This low-cost intervention has the potential to reduce primary care antibiotic prescribing in children and we did not detect an increase in GP attendance rates after this intervention in our study. It could easily be rolled out nationwide.

Introduction

Antibiotics are a major driver of antimicrobial resistance (AMR). In the UK, around 80% of antibiotic prescribing occurs in primary care,¹ so any national strategy to reduce AMR should target antibiotic prescribing in this setting. Nationally, despite antibiotic consumption falling between 2014 and 2019,² further reductions are required to attain the low levels achieved in other northern European countries, such as the Netherlands.³ Whilst attention has focused on adults, young children are high recipients of antibiotics⁴ and data suggest an increasing burden of antimicrobial-resistant infections, especially in young infants.⁵ Furthermore, studies show substantial overprescribing by GPs, particularly in uncomplicated respiratory tract infections (RTIs).^6,7 RTIs make up more than 50% of presentations to GPs and of these, approximately 60% are prescribed antibiotics,^8,9 although there is good evidence to suggest that the majority of children gain little or no benefit from them.¹⁰

Streptococcus pneumoniae, Neisseria meningitidis and Haemophilus influenza type B are the most common infectious agents responsible for bacterial meningitis and sepsis in children.¹¹ The introduction of conjugate vaccines against these agents has resulted in a marked reduction in their prevalence in children, explaining the significant reduction in the rate of serious infections over recent years.^11–13 Yet despite its self-limiting nature, febrile illness in childhood accounts for a large GP workload, with one in four consultations resulting in antibiotics.¹²

Key antibiotic stewardship resources available to help facilitate appropriate antimicrobial prescribing include the UK Health Security Agency and Royal College of General Practitioners TARGET (Treat Antibiotics Responsibly, Guidance, Education and Tools) antibiotics toolkit¹⁴ and the Healthier Together programme.¹⁵

TARGET includes a range of resources to help influence prescribers’ and patients’ personal attitudes, social norms and perceived barriers to optimal antibiotic prescribing. Resources have been developed with input from both prescribers and patients and include audits, training materials, patient information leaflets and antibiotic prescribing self-assessment tools, amongst others.

Healthier Together has developed national safety-netting resources for parents in conjunction with the Royal College of Paediatrics and Child Health. Safety-netting resources provide clear information about monitoring of symptoms and actions to take if they occur. This is a programme co-designed with parents and healthcare professionals from primary and secondary care, aimed at improving the health of 0–18-year-olds. The focus is on healthcare professional and public education, promoting shared decision-making. The Healthier Together website contains clear safety-netting (When should I worry?) information on common childhood infections and includes the option of sending this directly to parents using a short messaging service functionality. Evidence supports this approach in reducing antibiotic prescribing and reducing subsequent primary care presentations without reducing satisfaction with care.¹⁶

The driver diagram in Figure 1 summarizes the quality improvement approach taken to reduce antibiotic prescribing in primary care. Many of the proposed improvement opportunities were already embedded in the TARGET toolkit.

Figure 1.

Driver diagram for quality improvement in antibiotic prescribing.

The TARGET and Healthier Together programmes also incorporate the recommendations from the National Institute for Health and Care Excellence to improve antimicrobial stewardship (AMS).¹⁷

Education of prescribers

The use of educational outreach (also known as ‘academic detailing’) is an established evidence-based method of making long-term changes to prescribing behaviours.^18,19 Antibiotic prescribing, specifically, can be improved through multifaceted interventions including multilevel education, after addressing local barriers to change.²⁰ Educational outreach and TARGET’s effectiveness in reducing antibiotic dispensing in British general practice has previously been described in a randomized control trial.²¹ This study found that within usual service provision, TARGET antibiotic workshops can help improve antibiotic use, and therefore should be considered as part of any national AMS initiatives.

This manuscript outlines findings from a quality improvement initiative to improve AMS and reduce unnecessary antibiotic prescriptions in young children in a British primary care setting.

Methods

The work described in this article uses the SQUIRE guidelines framework²² to report the experience of improving antibiotic prescribing in the former West Hampshire Clinical Commissioning Group (WHCCG).

Context

In 2018, WHCCG provided care to a population of 550 000 people (including 27 500 children under 5 years old) through its 49 constituent GP practices.

CCGs were clinically led statutory NHS bodies responsible for the planning and commissioning of healthcare services (including primary medical care from general practices) for their local area, prior to being replaced by integrated care systems in July 2022.²³

Intervention

As part of a local Medicines Optimisation prescribing incentive scheme, individual general medical practices were invited to host an educational outreach meeting on paediatric AMS. Trained local pharmacy professionals (clinical pharmacists and pharmacy technicians known to GP staff) delivered a 1 h face-to-face workshop, adapted for paediatrics using resources produced by the Healthier Together programme.

All 49 WHCCG practices accepted, and the meetings took place over a 6 month period (October 2018–March 2019). The combined TARGET/Healthier Together workshop included a review of antibiotic prescribing data and a presentation on paediatric AMS showcasing locally agreed paediatric primary care prescribing guidelines, as well as safety-netting resources, and formulation of a local action plan. The workshop was delivered to both clinical and non-clinical members of the primary healthcare team.

Study of the intervention

Data were obtained from the NHS Business Services Authority Information Services team online business intelligence tool, ePACT2.²⁴ This allows registered users across England and Wales to view, analyse and present age-specific prescribing data, and monitor prescribing trends. Data on antibiotic use are collected electronically from pharmacies when each antibiotic prescription is dispensed. Each prescription dispensed is referred to as an ‘item’.

Primary care appointment consultation data were provided by the Hampshire and Isle of Wight Paediatric Urgent Care dashboard developed by the South, Central and West Commissioning Support Unit.

Measures

Primary outcome measure

Data on total oral antibiotic items per 1000 children under 5 years of age were collected for WHCCG, as well as for the whole of England, between January 2015 and January 2020.

Secondary outcome measures

GP attendance rates for children aged under 5 years over the study period (2015–20) were used to identify whether an unintended consequence of the intervention was increased demand for primary care services.

Data analysis

A run chart was used to compare the trend in antibiotic prescribing rate with the baseline median trend to identify sustained reduction in monthly antibiotic prescribing rate over the study period for WHCCG and the rest of England as a comparison. Due to the strong seasonal nature of antibiotic prescribing, a run of 12 or more consecutive points above or below the baseline trend (rather than the usual 9) was used to indicate a signal of a change in prescribing rate.

A run chart was also used to see trends in GP consultation rate by children aged 0–4 years in WHCCG, per quarter from 2015–20.

Ethics

National Research Ethics Committee approval was not required, as the study did not recruit NHS patients, through the NHS. This decision is in accordance with the NHS ‘defining research’ guidelines.²⁵ Routine ePACT2 data, available to registered primary care organization users was used in the analysis, therefore no ethical consent was required.

Results

Data were analysed from 48 out of the 49 practices that participated in the interactive workshop. This was due to a merger of GP practices resulting in the prescribing data for one practice no longer being available.

Antibiotic prescribing

In WHCCG the initial median monthly prescribing rate was 48.9 per 1000 patients (Figure 2). A trend of 12 consecutive points below this median began in April 2018, with a new median monthly prescribing rate of 39.0 per 1000 patients. In England there was no period of sustained change from the initial monthly prescribing rate of 58.7 per 1000 patients.

Figure 2.

Monthly antibiotic prescribing rates per 1000 children under 5 years old in England and WHCCG, January 2015 to January 2020. Blue—monthly prescribing rates per 1000 children under 5 years old. Solid orange—median prescribing rate defined using 24 months of data initially and 12 months when rebased. Dashed orange—period of continued trend. Pink—period of 12 consecutive months below median. The start of the deviation in prescribing rates did not coincide with the start of the quality improvement initiative in October 2018. This is due to seasonal variation in prescribing antimicrobials, causing prescribing to reduce at roughly the same time of year as would be expected over the summer months. Following the intervention, however, antibiotic prescribing rates did not return to the higher levels of previous winter months, indicating a different pattern in WHCCG compared with England.

GP attendances

The run chart of GP consultation rate identified no signal of a change in consultation rate between January 2015 and January 2020 (Figure 3).

Figure 3.

Trend in GP consultation rate by children aged 0–4 years in WHCCG, per quarter 2015–20. Vertical dashed line represents the implementation date of the initiative.

Discussion

Summary

This study suggests that a multifaceted intervention including teaching on AMS, review of antibiotic prescribing data, showcasing of prescribing guidelines, safety-netting resources and formulation of a local action plan can reduce antimicrobial prescribing in children in a UK primary care setting. The intervention was not associated with unintended consequences such as an increase in primary care presentations.

Interpretation

Studies support improvement in reducing inappropriate use of antimicrobials in this age group without an increase in harm from lack of prescribing.^4,10,26 We measured rates of GP presentations as a surrogate marker for adverse infection outcomes, as the primary healthcare provider for children with worsening infection would be the GP practice. We found no increase in GP presentations associated with the intervention.

Strengths and limitations

The major strength of this intervention was its ease of implementation and acceptability. The intervention was rolled out across all 49 practices in the CCG, reaching more than 300 GPs in their place of work. None of the practices declined to participate. The 1 h meeting was relatively easy to accommodate into a busy practice schedule. It was well received by the clinicians and non-clinical staff in the GP practices, who were keen to arrange a regular future annual AMS meeting (as recommended by the TARGET materials). Practice managers welcomed the resultant Practice Action Plans as evidence of progress towards attaining stewardship targets.

Limitations include the metric used to quantify antibiotic prescribing. Although ‘days of therapy’ is often used as the conventional metric of antimicrobial consumption, we used the metric of ‘items dispensed per 1000 practice patients aged under 5 years’ because this metric is routinely collected by NHS England. This metric provides data on whether antibiotics were prescribed, but does not provide data on the duration of antibiotic courses, with some evidence suggesting that course lengths exceed those recommended nationally.²⁷

In addition, the dataset collected included all prescribed oral antibiotics, as opposed to antibiotics specifically used to treat RTIs. However, this is unlikely to have impacted on the findings of our study, as the vast majority of oral antibiotics prescribed to young children are for RTIs.²⁸

It was beyond the scope of this quality improvement initiative to determine whether these findings generalize beyond WHCCG, which would require a more robust study design, including, for example, a cluster randomized clinical trial.

Furthermore, data were not collected on confounding factors, which meant we were unable to account for differences in the target population between WHCCG and the other CCGs. Although we appreciate that various factors can impact antimicrobial prescribing,²⁹ the approach we adopted was to compare prescribing rates in WHCCG with national prescribing rates, assuming that apart from the intervention, other factors would have remained relatively constant.

Conclusions

Our findings demonstrate that a low-cost intervention can have an impact on antibiotic prescribing in a target intervention group. The success of the intervention relied on a clear strategy and good leadership, provision of prescribing data to frontline clinicians to facilitate peer benchmarking and the availability of high quality safety-netting resources to support shared decision-making between clinicians and parents. This intervention has the potential to be rolled out across the country.

In order to fulfil the requirements of the NHS Long Term Plan³⁰ and HM Government’s UK 5-year national action plan³¹ for tackling AMR, further progress is required to continue to reduce the number of unwarranted antibiotic prescriptions dispensed in primary care. We encourage senior health professionals to take on the role of AMS Champions, to promote AMS within their teams, recognizing the influence that senior prescribers can have on prescribing practices of colleagues. AMS Champions, along with the resident pharmacy professionals working in general practices, provide a dedicated workforce with access to in-house data and in-depth knowledge of prescribing patterns and behaviours within general practice. Such an intervention not only has an impact on rates of antibiotic prescribing but is also associated with reduced rates of presentation to primary care.

Funding

We received no funding to conduct this study. It was carried out as part of routine health improvement practice of the WHCCG Medicines Optimisation team (a team of pharmacists and pharmacy technicians employed by the CCG and working in primary care to optimize the use of medicines). The educational outreach meetings in GP practices were incentivized as part of the CCG Medicines Optimisation Prescribing Incentive Scheme. Research paper publication costs were met by the United Kingdom Health Security Agency.

Transparency declarations

There are no conflicts of interest.