Antibiotic prescribing trends in primary care 2014-2022

Antimicrobial resistance (AMR) is a global healthcare challenge that governments and health systems are tackling primarily through antimicrobial stewardship (AMS). This should, improve antibiotic use, avoid inappropriate prescribing, reduce prescription numbers, aligning with national/international AMS targets. In primary care in the United Kingdom (UK) antibiotics are mainly prescribed for patients with urinary and respiratory symptoms (22.7% and 46% of all antibiotic prescriptions respectively). This study aimed to capture the time-series trends (2014-2022) for commonly prescribed antibiotics for respiratory and urinary tract infections in primary care in England. Trends for Amoxicillin, Amoxicillin sodium, Trimethoprim, Clarithromycin, Erythromycin, Erythromycin ethylsuccinate, Erythromycin stearate, Doxycycline hyclate, Doxycycline monohydrate and Phenoxymethylpenicillin (Penicillin V) were determined. In doing so providing evidence regarding meeting UK antibiotic prescribing rate objectives (a 15% reduction in human antibiotic use 2019-2024). Time series trend analysis of 62,949,272 antibiotic prescriptions from 6,370 General Practices in England extracted from the National Health Service (NHS) Business Services Authority web portal were explored. With additional investigation of prescribing rate trends by quintiles of the Index of Multiple Deprivation (IMD). Overall, there is a downwards trend in antibiotic prescribing for those explored. There is an association between IMD, geographical location, and higher antibiotic prescribing levels (prescribing hot spots). England has a well-documented North-South divide of health inequalities, this is reflected in antibiotic prescribing. The corona virus pandemic (COVID-19) impacted on AMS, with a rise in doxycycline and trimethoprim prescriptions notable in higher IMD areas. Since then, prescribing appears to have returned to pre-pandemic levels in all IMDs and continued to decline. AMS efforts are being adhered to in primary care in England. This study provides further evidence of the link between locality and poorer health outcomes (reflected in higher antibiotic prescribing). Further work is required to address antibiotic use in hot spot areas.

Strategies for de-implementation of low-value care-a scoping review

Background

The use of low-value care (LVC) is a persistent problem that calls for knowledge about strategies for de-implementation. However, studies are dispersed across many clinical fields, and there is no overview of strategies that can be used to support the de-implementation of LVC. The extent to which strategies used for implementation are also used in de-implementing LVC is unknown. The aim of this scoping review is to (1) identify strategies for the de-implementation of LVC described in the scientific literature and (2) compare de-implementation strategies to implementation strategies as specified in the Expert Recommendation for Implementing Change (ERIC) and strategies added by Perry et al.

Method

A scoping review was conducted according to recommendations outlined by Arksey and O’Malley. Four scientific databases were searched, relevant articles were snowball searched, and the journal Implementation Science was searched manually for peer-reviewed journal articles in English. Articles were included if they were empirical studies of strategies designed to reduce the use of LVC. Two reviewers conducted all abstract and full-text reviews, and conflicting decisions were discussed until consensus was reached. Data were charted using a piloted data-charting form. The strategies were first coded inductively and then mapped onto the ERIC compilation of implementation strategies.

Results

The scoping review identified a total of 71 unique de-implementation strategies described in the literature. Of these, 62 strategies could be mapped onto ERIC strategies, and four strategies onto one added category. Half (50%) of the 73 ERIC implementation strategies were used for de-implementation purposes. Five identified de-implementation strategies could not be mapped onto any of the existing strategies in ERIC.

Conclusions

Similar strategies are used for de-implementation and implementation. However, only a half of the implementation strategies included in the ERIC compilation were represented in the de-implementation studies, which may imply that some strategies are being underused or that they are not applicable for de-implementation purposes. The strategies assess and redesign workflow (a strategy previously suggested to be added to ERIC), accountability tool, and communication tool (unique new strategies for de-implementation) could complement the existing ERIC compilation when used for de-implementation purposes.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13012-022-01247-y.

Trends and Patterns of Antibiotics Use in China's Urban Tertiary Hospitals, 2016-19

Objectives: This study aimed to identify the trends in antibiotics utilization and patients costs, evaluating the effect of the policy and exploring factors associated with the irrational use of antibiotics. Methods: Based on the Cooperation Project Database of Hospital Prescriptions, data were collected from 89 tertiary hospitals in nine cities in China during 2016-2019. The study sample consisted of prescription records with antibiotics for 3,422,710 outpatient and emergency visits and 26, 118, 436 inpatient hospitalizations. Results: For outpatients, the proportion of treated with antibiotics declined from 14.72 to 13.92% significantly (p < 0.01). The proportion of antibiotic costs for outpatients decreased from 5.79 to 4.45% significantly (p < 0.01). For emergency patients, the proportion of treated with antibiotics increased from 39.31 to 43.45% significantly (p < 0.01). The proportion of antibiotic costs for emergency patients decreased from 36.44 to 34.69%, with no significant change (p = 0.87). For inpatients, the proportion of treated with antibiotics increased from 23.82 to 27.25% significantly (p < 0.01). The proportion of antibiotic costs for outpatients decreased from 18.09 to 17.19% with no statistical significance (p = 0.89). Other β-lactam antibacterials (1,663.03 ten thousand DDD) far exceeded other antibiotics categories. Stablely ranked first, followed by Macrolides, lincosamide and streptogramins (965.74 ten thousand DDD), Quinolone antibacterials (710.42 ten thousand DDD), and β-lactam antibacterials, penicillins (497.01 ten thousand DDD). Conclusions: The proportion of treated with antibiotics for outpatients and inpatients meet the WHO standards. The antibiotics use varied by different survey areas, clinical departments, patient gender, patient age and antibiotics categories. More efforts should focus on improving the appropriateness of antibiotics use at the individual level.

Impact of the childhood influenza vaccine programme on antibiotic prescribing rates in primary care in England

Vaccines are a key part of the global strategy to tackle antimicrobial resistance (AMR) since prevention of infection should reduce antibiotic use. England commenced national rollout of a live attenuated influenza vaccine (LAIV) programme for children aged 2-3 years together with a series of geographically discrete pilot areas for primary school age children in 2013 extending to older children in subsequent seasons. We investigated vaccine programme impact on community antibiotic prescribing rates. Antibiotic prescribing incidence rates for respiratory (RTI) and urinary tract infections (UTI; controls) were calculated at general practice (GP) level by age category (children<=10 years/adults) and season for LAIV pilot and non-pilot areas between 2013/14 and 2015/16. To estimate the LAIV (primary school age children only) intervention effect, a random effects model was fitted. A multivariable random-effects Poisson regression investigated the association of antibiotic prescribing rates in children with LAIV uptake (2-3-year-olds only) at GP practice level. RTI antibiotic prescribing rates for children <=10 years and adults showed clear seasonal trends and were lower in LAIV-pilot and non-pilot areas after the introduction of the LAIV programme in 2013. The reductions for RTI prescriptions (children) were similar (within 3%) in all areas, which coincided with the start the UK AMR strategy. Antibiotic prescribing was significantly (p < 0.0001) related to LAIV uptake in 2-3-year-olds with antibiotic prescribing reduced by 2.7% (95% CI: 2.1% to 3.4%) for every 10% increase in uptake. We found no evidence the LAIV programme for primary school age children resulted in reductions in RTI antibiotic prescribing, however we detected a significant inverse association between increased vaccine uptake in pre-school age children and antibiotic prescribing at GP level. The temporal association of reduced RTI and UTI antibiotic prescribing with the launch of the UK's AMR Strategy in 2013 highlights the importance of a multifaceted approach to tackle AMR.

Trends and patterns in antibiotic prescribing for adult outpatients with acute upper respiratory tract infection in Japan, 2008-2018

INTRODUCTION

This study aimed to characterize the patterns and trends in the use of major oral antibiotics prescribed for adult outpatients diagnosed with acute upper respiratory tract infections (AURTIs) in Japan between 2008 and 2018.

METHODS

We analyzed administrative claims data for adults in Japan, between April 2008 and September 2018. The trends in oral antibiotic prescription were illustrated using the prescribing rate, and tested using interrupted time series analysis. We also assessed the factors associated with antibiotic prescription for AURTIs.

RESULTS

Data on 7.54 million antibiotic prescriptions in 1,937,379 adults with AURTIs were analyzed; people ≥65-years old were scarcely included. The antibiotic prescribing rate declined from 49.9% in 2008 to 39.0% in 2018, and the rate of decrease accelerated after the national action plan on antimicrobial resistance was launched in April 2016. Acute nasopharyngitis was the most common indication. Cephalosporins (35.2%), macrolides (32.2%), fluoroquinolones (24.6%), and penicillins (5.88%) were the most commonly prescribed antibiotic classes. Sex, age, type and specialty of the facility, and season affected the likelihood of being prescribed antibiotics. The majority of antibiotics prescribed were broad-spectrum, but use of penicillins showed a higher rate of increase after 2016.

CONCLUSIONS

The percentage of antibiotic prescriptions for AURTIs decreased significantly after implementing the action plan, indicating that it was effective to some extent. However, since the use of broad-spectrum antibiotics was high, there is scope for improving the prescribing pattern, including the types of antimicrobial agents; this is considered to be a future issue.

Community-Based Antibiotic Prescribing Attributable to Respiratory Syncytial Virus and Other Common Respiratory Viruses in Young Children: A Population-Based Time-series Study of Scottish Children

BACKGROUND

Inappropriate antibiotic prescribing, such as for viral illness, remains common in primary care. The objective of this study was to estimate the proportion of community-prescribed antibiotics to children aged less than 5 years attributable to common respiratory viruses.

METHODS

We fitted time-series negative binomial models to predict weekly antibiotic prescribing rates from positive viral pathogen tests for the period 1 April 2009 through 27 December 2017 using comprehensive, population-based administrative data for all children (<5 years) living in Scotland. Multiple respiratory viral pathogens were considered, including respiratory syncytial virus (RSV), influenza, human metapneumovirus (HMPV), rhinovirus, and human parainfluenza (HPIV) types 1-4. We estimated the proportion of antibiotic prescriptions explained by virus circulation according to type of virus, by age group, presence of high-risk chronic conditions, and antibiotic class.

RESULTS

We included data on 6 066 492 antibiotic prescriptions among 452 877 children. The antibiotic-prescribing rate among all Scottish children (<5 years) was 609.7 per 1000 child-years. Our final model included RSV, influenza, HMPV, HPIV-1, and HPIV-3. An estimated 6.9% (95% confidence interval, 5.6-8.3%), 2.4% (1.7-3.1%), and 2.3% (.8-3.9%) of antibiotics were attributable to RSV, influenza, and HMPV, respectively. RSV was consistently associated with the highest proportion of prescribed antibiotics, particularly among children without chronic conditions and for amoxicillin and macrolide prescriptions.

CONCLUSIONS

Nearly 14% of antibiotics prescribed to children in this study were estimated to be attributable to common viruses for which antibiotics are not recommended. A future RSV vaccine could substantially reduce unnecessary antibiotic prescribing among children.

The drivers of antimicrobial use across institutions, stakeholders and economic settings: a paradigm shift is required for effective optimization

OBJECTIVES

Significant antimicrobial overuse persists worldwide, despite overwhelming evidence of antimicrobial resistance and knowledge that optimization of antimicrobial use will slow the development of resistance. It is critical to understand why this occurs. This study aims to consider the social influences on antimicrobial use within hospitals in Australia, via an in-depth, multisite analysis.

METHODS

We used a qualitative multisite design, involving 222 individual semi-structured interviews and thematic analysis. Participants (85 doctors, 79 nurses, 31 pharmacists and 27 hospital managers) were recruited from five hospitals in Australia, including four public hospitals (two metropolitan, one regional and one remote) and one private hospital.

RESULTS

Analysis of the interviews identified social relationships and institutional structures that may have a strong influence on antimicrobial use, which must be addressed concurrently. (i) Social relationships that exist across settings: these include the influence of personal risk, hierarchies, inter- and intraprofessional dynamics and sense of futility in making a difference long term in relation to antimicrobial resistance. (ii) Institutional structures that offer context-specific influences: these include patient population factors (including socioeconomic factors, geographical isolation and local infection patterns), proximity and resource issues.

CONCLUSIONS

The success of antimicrobial optimization rests on adequate awareness and incorporation of multilevel influences. Analysis of the problem has tended to emphasize individual 'behaviour improvement' in prescribing rather than incorporating the problem of overuse as inherently multidimensional and necessarily incorporating personal, interpersonal and institutional variables. A paradigm shift is urgently needed to incorporate these critical factors in antimicrobial optimization strategies.

An examination of trends in antibiotic prescribing in primary care and the association with area-level deprivation in England

Background

Internationally, there are growing concerns about antimicrobial resistance. This has resulted in increased scrutiny of antibiotic prescribing trends – particularly in primary care where the majority of prescribing occurs. In England, antibiotic prescribing targets are set nationally but little is known about the local context of antibiotic prescribing. This study aimed to examine trends in antibiotic prescribing (including broad-spectrum), and the association with area-level deprivation and region in England.

Methods

Antibiotic prescribing data by GP surgery in England were obtained from NHS Business Service Authority for the years 2014–2018. These data were matched with the Index of Multiple Deprivation (IMD) 2015 at the Lower Layer Super Output Area level Lower Layer Super Output Area (LSOA) level. Linear regression methods were employed to explore the relationship between antibiotic use and area-level deprivation as well as region, after controlling for a range of other confounding variables, including health need, rurality, and ethnicity.

Results

Over time, the amount of antibiotic prescribing significantly reduced from 1.11 items per STAR-PU to 0.96 items per STAR-PU – a reduction of 13.6%. The adjusted models found that, at LSOA level, the most deprived areas of England had the highest levels of antibiotic prescribing (0.03 items per STAR-PU higher). However, broad spectrum antibiotic prescribing exceeding 10% of all antibiotic prescribing within a GP practice was higher in more affluent areas. There were also significant regional differences – with the North East and the East of England having the highest levels of antibiotic prescribing (by 0.16 items per STAR-PU).

Conclusion

Although antibiotic prescribing has reduced over time, there remains significant variation in by area-level deprivation and region in England – with higher antibiotic prescribing in more deprived areas. Future prescribing targets should account for local factors to ensure the most deprived communities are not inappropriately penalised.

Variation in responsiveness to warranted behaviour change among NHS clinicians: novel implementation of change detection methods in longitudinal prescribing data

OBJECTIVES

To determine how clinicians vary in their response to new guidance on existing or new interventions, by measuring the timing and magnitude of change at healthcare institutions.

DESIGN

Automated change detection in longitudinal prescribing data.

SETTING

Prescribing data in English primary care.

PARTICIPANTS

English general practices.

MAIN OUTCOME MEASURES

In each practice the following were measured: the timing of the largest changes, steepness of the change slope (change in proportion per month), and magnitude of the change for two example time series (expiry of the Cerazette patent in 2012, leading to cheaper generic desogestrel alternatives becoming available; and a change in antibiotic prescribing guidelines after 2014, favouring nitrofurantoin over trimethoprim for uncomplicated urinary tract infection (UTI)).

RESULTS

Substantial heterogeneity was found between institutions in both timing and steepness of change. The range of time delay before a change was implemented was large (interquartile range 2-14 months (median 8) for Cerazette, and 5-29 months (18) for UTI). Substantial heterogeneity was also seen in slope following a detected change (interquartile range 2-28% absolute reduction per month (median 9%) for Cerazette, and 1-8% (2%) for UTI). When changes were implemented, the magnitude of change showed substantially less heterogeneity (interquartile range 44-85% (median 66%) for Cerazette and 28-47% (38%) for UTI).

CONCLUSIONS

Substantial variation was observed in the speed with which individual NHS general practices responded to warranted changes in clinical practice. Changes in prescribing behaviour were detected automatically and robustly. Detection of structural breaks using indicator saturation methods opens up new opportunities to improve patient care through audit and feedback by moving away from cross sectional analyses, and automatically identifying institutions that respond rapidly, or slowly, to warranted changes in clinical practice.

Government policy interventions to reduce human antimicrobial use: A systematic review and evidence map

BACKGROUND

Growing political attention to antimicrobial resistance (AMR) offers a rare opportunity for achieving meaningful action. Many governments have developed national AMR action plans, but most have not yet implemented policy interventions to reduce antimicrobial overuse. A systematic evidence map can support governments in making evidence-informed decisions about implementing programs to reduce AMR, by identifying, describing, and assessing the full range of evaluated government policy options to reduce antimicrobial use in humans.

METHODS AND FINDINGS

Seven databases were searched from inception to January 28, 2019, (MEDLINE, CINAHL, EMBASE, PAIS Index, Cochrane Central Register of Controlled Trials, Web of Science, and PubMed). We identified studies that (1) clearly described a government policy intervention aimed at reducing human antimicrobial use, and (2) applied a quantitative design to measure the impact. We found 69 unique evaluations of government policy interventions carried out across 4 of the 6 WHO regions. These evaluations included randomized controlled trials (n = 4), non-randomized controlled trials (n = 3), controlled before-and-after designs (n = 7), interrupted time series designs (n = 25), uncontrolled before-and-after designs (n = 18), descriptive designs (n = 10), and cohort designs (n = 2). From these we identified 17 unique policy options for governments to reduce the human use of antimicrobials. Many studies evaluated public awareness campaigns (n = 17) and antimicrobial guidelines (n = 13); however, others offered different policy options such as professional regulation, restricted reimbursement, pay for performance, and prescription requirements. Identifying these policies can inform the development of future policies and evaluations in different contexts and health systems. Limitations of our study include the possible omission of unpublished initiatives, and that policies not evaluated with respect to antimicrobial use have not been captured in this review.

CONCLUSIONS

To our knowledge this is the first study to provide policy makers with synthesized evidence on specific government policy interventions addressing AMR. In the future, governments should ensure that AMR policy interventions are evaluated using rigorous study designs and that study results are published.

PROTOCOL REGISTRATION

PROSPERO CRD42017067514.