Weiler correctly highlights some of the limitations of primary care prescribing data:1 Research databases such as the Clinical Practice Research Datalink (CPRD, formerly the GPRD) record information on issued prescriptions, prescribing analyses and cost (PACT) data records prescriptions that are dispensed. Neither of these methods will record the proportion of prescriptions that are actually taken by the patient and we are unaware of any research studies that have specifically addressed this important question. GPs can record delayed prescriptions using the Vision software system, but it is unclear how frequently this Read Code is used.
However, estimates suggest 80–90% of antimicrobials are prescribed in the community so even if we are overestimating antibiotic use, it is clear that the majority of antimicrobials are prescribed and dispensed in primary care. Research studies based on both issued and dispensed prescriptions using different populations all deliver the same message: there is major heterogeneity in prescribing, and this offers scope to reduce antibiotic use.2–4
There is a clear need for better surveillance data on antimicrobial prescribing and resistance in hospitals and in primary care and this is the rationale behind the English Surveillance Programme for Antimicrobial Utilization and Resistance (ESPAUR).5 Optimising antibiotic use is a fundamental part of the response to antimicrobial resistance. This will only be achieved by changing patient and clinician behaviour in the community and in hospitals.
- © British Journal of General Practice 2015
