This paper adds to the growing evidence base exploring antibiotic prescribing trends in England. It is the first though to consider antibiotic prescribing using a proxy indicator of health need, and it is also the first to explore the prescribing of broad-spectrum antibiotics (co-amoxiclav, cephalosporins, and quinolones) as a percentage of total antibiotic prescribing by area-level deprivation. We identified three key findings that will be of importance to healthcare policy discussions around antimicrobial resistance and prescribing targets: (1) there were significant inequalities in antibiotic prescribing by deprivation – with GP surgeries located in the most deprived areas having the highest levels of antibiotic prescribing (even when adjusting for a proxy indicator of health need); (2) there is also variation in the prescribing of broad spectrum antibiotics (co-amoxiclav, cephalosporins, and quinolines) with higher proportions of broad spectrum antibiotic prescribing occurring in more affluent areas; (3) there was also significant regional variation in antibiotic prescribing – with the highest levels of antibiotic prescribing observed in the East and North East of England.
Area deprivation has multiple influences on health so the pathways linking deprivation and inequalities in antibiotic prescribing patterns are complex. There is a large international literature on the relationship between health and place which suggests that geographical health inequalities exist as a result of both the characteristics of places (in terms of infrastructure and services, social factors and the physical environment) and of the people who live there (e.g. ethnicity individual level socio-economic status) . In terms of antibiotic prescribing, the most likely pathways linking deprivation to health system characteristics include healthcare access and quality, as well as the health needs of the population. So, the higher rates of antibiotic prescribing in the most deprived areas may reflect differences in GP prescribing behaviours – more deprived areas in England have lower GP provision per head of population than the least deprived and so are more reliant on locum doctors who may be more inclined to prescribe antibiotics [22, 23]. Additionally, there may be differences between symptom severity in patients living in different areas. Our proxy indicator of health need accounted for two key conditions (diabetes and COPD) but not the severity of these illnesses. Further, our adjustment of health need did not take into account other illnesses or conditions for which antibiotics might be prescribed (such as immunosuppression, or recurrent urinary tract infections). In terms of broad spectrum antibiotic prescribing being higher in more affluent areas, the reasons for this are not clear - although it has been previously shown that people accessing out-of-hours primary care services typically receive more broad-spectrum antibiotics compared to when they access in-hours primary care services . Living in an affluent area might, therefore, influence how primary care services are accessed, and, ultimately, the type of antibiotic prescribed; for example, people living in affluent areas may be able to navigate the healthcare system easier  and access emergency GP appointments, compared to people living in deprived areas. This should be explored in future studies.
Our broad findings are in keeping with previous research. As mentioned previously, the study by Curtis and colleagues, who described antibiotic prescribing trends across England for the years 1998 to 2017, showed that there was significant geographical variation in prescribing: at a Clinical Commissioning Group (CCG) level, the variation in overall antibiotic prescribing was two-fold, while for cephalosporin prescribing, the variation was seven-fold . The work also showed that higher prescribing trends were associated with a greater GP practice size, the proportion of patients greater than 65 years, or less than 18 years, ruralness and deprivation . Similarly, Covvey and colleagues evaluated antibiotic prescribing trends in Scotland, and concluded that higher rates of antibiotic prescribing were found in areas of higher deprivation, as measured according to the Scottish Index of Multiple Deprivation (SIMD) . This study also assessed antibiotic prescribing by antibiotic class stratified by SIMD quintile: the authors showed that the prescribing of quinolones, cephalosporins and other beta-lactams, was generally higher in more deprived areas. Although this study did not look at the proportions of broad-spectrum antibiotic prescribing as we did, this result seems to be in contrast to our findings, where we show the proportion of GP surgeries prescribing broad-spectrum antibiotics is higher in more affluent areas.
Another study by Mölter and colleagues, who analysed antibiotic prescribing by GP practice in England, identified spatial clusters of high and low spots of prescribing (so-called ‘hot’ and ‘cold’ spots) . The work showed that the distribution of antibiotic prescribing was heterogeneous, with the majority of the hot spots located in the North of England. Our results confirm this as we found that, when controlling for demographic, and health need variables, highest levels of prescribing were found in the North East, and East of England; lowest levels of antibiotic prescribing were consistently found in London. Overall, our study builds on previous research, and shows that even after using a proxy indicator to control for health need, there is evidence of significant inequalities for those in the bottom three deciles compared to those living in the least deprived decile.
Given the emphasis and strategic importance – at both a national and international level – of developing and implementing antibiotic stewardship polices, our findings are timely and potentially have important implications for policymakers. The national strategy of reducing the use of antibiotics appears to be working, given that our data shows a reduction in antibiotic items per STAR-PU each year. This is in line with other work that also shows a similar reduction in antibiotic prescribing . However, current prescribing targets only account for age and sex of the population served. For example, men aged 75 years and above are weighted as 1.0, while men aged between 35 and 44 years are weighted as 0.3; in contrast, women aged 75 years and above are weighted as 1.3, while women aged between 35 and 44 years are weighted as 0.6 . The weighting does not consider any measure of area-level deprivation or related-health need. Our results suggest that it would be prudent for any future antibiotic prescribing targets to acknowledge that GP surgeries located in the most deprived communities are likely to have a higher health need in terms of antibiotic use, and account for this in their targets. Local antimicrobial stewardship approaches should also be considered at an area level to account for specific pressures and needs. In addition, any future revision of the prescribing measure items per STAR-PU should also consider incorporating a measure of deprivation into their weighting. This finding was also echoed by Pouwels and colleagues , who suggest it would be advantageous to avoid the same prescribing targets for all GP practices, or it would be important to develop alternative approaches that encompass additional predictors of antibiotic prescribing. This is similar to the way in which NHS funding allocation policy incorporates deprivation .
Strengths and limitations
We believe our modelling results are robust; the residuals are normally distributed, and the VIF ratio is low (2.07) therefore homoscedasticity and multicollineararity are not considered problematic. However, we do acknowledge there are a number of limitations to our work. Firstly, we only assessed the amount of antibiotic prescribing according to items per STAR-PU; we did not consider the appropriateness of prescribing, nor did we consider the patient characteristics for whom the antibiotics were prescribed. It is possible, therefore, that the higher antibiotic prescribing observed in the most deprived areas were prescribed either unnecessarily or inappropriately. Indeed, Smith and colleagues showed that in an English primary care setting, most antibiotics are prescribed for conditions that only sometimes required antibiotics, which was dependent on patient specific indicators (e.g. co-morbidity) . Our study does not account for this. It would be prudent, therefore, for future work to assess the appropriateness of antibiotic prescribing for GPs located in deprived areas. Our data was obtained free of charge from the NHS Business Service Authority at the GP practice level; data at the patient – or individual prescriber level – was not available through this route. Collecting patient-level data, and making it freely available for research or system improvement purposes without expensive subscriptions would be advantageous as it would allow the assessment of prescribing appropriateness. A further limitation of our work is in relation to how we adjusted for health need: in our linear regression model, we only used COPD and diabetes prevalence as proxies for a health need measure. In addition to COPD and diabetes prevalence, there are other reasons that may contribute to increased susceptibility of developing a bacterial infection, including poor living conditions , reduced vaccination uptake , poor nutrition , and higher incidence of smoking . Frailty may also be associated with increased antibiotic prescribing, which could have been accounted for using the eFrailty index . These additional factors were not accounted for in our analysis. In addition, we also only analysed 4 years of data as prior to this, there were changes in the methodology of recording of prescribing data, making it challenging to investigate longer-term trends in antibiotic prescribing using our data sources. Furthermore, our study was ecological in design: thus, we acknowledge that relationships that apply at an area-level do not necessarily apply at an individual level – such an assumption would be committing the ecological fallacy. As such, there is no measure to link antibiotics prescribed at a GP level with patients who receive those antibiotics at an individual level.