B McGowan1, C Bergin2, K Bennett3, M Barry1 1National Centre for Pharmacoeconomics, 2Department Of Genitourinary Medicine & Infectious Diseases and 3Department of Pharmacology and Therapeutics Trinity Centre, St James’s Hospital, James’s St, Dublin 8
Abstract The aim of the study was to identify outpatient antibiotic consumption between Jan 2000 and Dec 2005 through analysis of the HSE-Primary Care Reimbursement Services (PCRS) database as part of the Surveillance of Antimicrobial Resistance in Ireland (SARI) project. Total antibiotic consumption on the PCRS scheme between January 2000 and December 2005 expressed in Defined Daily Dose per 1000 PCRS inhabitants per day increased by 26%.The penicillin group represents the highest consumption accounting for approximately 50% of the total outpatient antibiotic use. Total DIDs for this group increased by 25% between 2000 and 2005. Co-amoxiclav and amoxicillin account for 80% of the total consumption of this group of anti-infectives. With the exception of aminoglycosides and sulfonamides which demonstrated a decrease in DID consumption of 47% and 8% respectively, all other groups of anti-infectives had an increase in DID consumption of greater than 25% during the study period. Antibiotic prescribing data is a valuable tool for assessing public health strategies aiming to optimise antibiotic prescribing.
IntroductionIn recent years there have been growing concerns worldwide about the level of unnecessary and inappropriate antibiotic prescribing and associated increased risk of antimicrobial resistance (AMR)1. International and national guidelines have been published looking at strategies to address these issues.2-8 In 1998 the European Antimicrobial Resistance Surveillance System (EARSS) was established with funding by the European Commission. It is an international network of national surveillance systems, which aims to collect comparable and reliable antimicrobial resistance data across Europe on invasive infections secondary to Staphylococcus aureus, Streptococcus pneumoniae, Escherichia coli, and Enterococcus faecium/faecalis for public health action. Ireland joined the EARSS surveillance programme in 1999. The Strategy for the Control of Antimicrobial Resistance in Ireland (SARI) was subsequently launched in June 2001 by the Department of Health and Children (DoHC). One of the specific terms of reference of SARI is the surveillance of antibiotic consumption. The aim of this study was to review antibiotic consumption on the HSE – Primary Care Reimbursement Services (PCRS) Scheme between January 2000 and December 2005 using its large prescription database.
MethodsThe HSE – PCRS scheme, formerly known as the General Medical Services (GMS) scheme covers approximately 30% of the population yet accounts for approximately 74% of volume and 70% of total expenditure on medicines prescribed under the Community Drugs Schemes9. The PCRS database was used to identify total antibiotic consumption between January 2000 and December 2005. All prescription data related to anti-infectives for systemic use (ATC classification J01) was analysed to identify trends in the prescribing of antibiotic therapy in the community setting. The database does not include non medical card holders such as persons covered under the Drug Payment Scheme (DPS scheme).Antibiotics are coded using the WHO Anatomic Therapeutic Chemical (ATC) classification and Defined Daily Dose (DDD) i.e. the average recommended adult daily dose is used as the unit of measurement of antibiotic consumption. There is no data available on clinical or microbiological diagnosis or indication for antibiotic usage within the database. The results are standardised for the PCRS eligible population for each year to account for changes in the PCRS population during the study period and in this case results are expressed in Defined Daily Doses per 1000 PCRS eligible population per day (DID). Both ATC classification and DDD are internationally agreed surveillance standards, allowing direct comparison of antibiotic consumption data between countries. Excel, JMP-In and SAS software were used to analyse the data.
ResultsTotal DID consumption of all anti-infectives for systemic use (ATC J01) increased by 26% over the study period. Consumption is divided into eight antibiotic groups according to the ATC classification: penicillins (J01C), cephalosporins (J01D), macrolides (J01F), quinolones (J01M), tetracyclines (J01A), sulfonamides (J01E), aminoglycosides (J01G) and other antibacterials(J01X). The penicillin group has had the highest consumption throughout the study period accounting for approximately 50% of the total outpatient antibiotic use. Total DIDs for this group increased by 25% between 2000 and 2005. Co-amoxiclav and amoxicillin account for 80% of the total consumption of this group of anti-infectives. The consumption of co-amoxiclav has increased by 80% within the study period, whereas total consumption of amoxicillin decreased by 9%.Table 1 illustrates the percentage increase or decrease in the DID consumption of the various antibiotic groups. Total DID consumption of macrolides has increased by 26% and 75% of this consumption is attributable to clarithromycin. Total consumption of the quinolone group of anti-infectives expressed in DIDs has almost doubled since 2000, 54% of which is attributed to the prescribing of ciprofloxacin. Ninety three per cent of the DID consumption of antibiotics which are listed under other antibacterials (J01X) is attributable to nitrofurantoin (81%) and colistin (12%). The total DID consumption of these anti-infective agents has increased three fold since 2000.
Total DID consumption of the aminoglycosides and sulfonamides decreased by 47% and 8% respectively within the study period.
For specific antibiotics there has been an increase in the total DIDs e.g. doxycycline 95%, clarithromycin 43%, flucloxacillin 42%, phenoxymethylpenecillin 22%, cefaclor 17% and minocycline 9%. There was a decrease of 14% and 34% in the consumption of erythromycin and oxytetracycline respectively between 2000 and 2005. The study identified marked seasonal variation in antibiotic consumption over the six-year period. Fluctuations were particularly noticeable in the penicillin, cephalosporin and macrolide groups where total DID consumption fell by 35% for the penicillins and cephalosporins and 40% for the macrolides between the months of January and August and increased again between August and December as illustrated in Figure 1.
There were similar patterns in prescribing of anti-infectives between the different healthboard regions as illustrated in Figure 2. When the results were standardised to the PCRS eligible population for each of the healthboard regions it was identified that the greatest DID consumption of anti-infective agents occurs predominantly in the Southern, Midlands and Midwestern healthboards while the lowest DID consumption of antiinfectives per 1000 PCRS eligible population occurs in the South Eastern healthboard region.
DiscussionTotal DID consumption of systemic antimicrobials (ATC class J01) increased by 25% between January 2000 and December 2005 on the PCRS scheme. Ferech et al10 reporting on the volume of outpatient antibiotic consumption in 25 European countries between 1997 and 2003 as part of the European Surveillance of Antimicrobial Consumption (ESAC) identified that outpatient consumption patterns demonstrated a year on year increase in Greece, Croatia, Ireland, Portugal, Luxembourg and Denmark. An initial increase followed by a reduction in antibiotic consumption was observed in Belgium, France, Czech Republic, Hungary, Poland, Slovenia and Sweden. An initial decrease followed by an increase was noted in the UK and Spain. Both Belgium and France have shown a sustained reduction in antibiotic consumption since 2001 which has been attributed to the nationwide campaigns to improve antibiotic use in these countries5,11. Total DID consumption of antibiotics varied by a factor of 3 between the country with the highest DID consumption which was Greece in 2003 and the country with the lowest DID consumption of anti-infectives which was the Netherlands. Both public and private prescriptions were analysed in the ESAC study. Ireland had the 11th highest rate of antibiotic consumption in this study where IMS data (wholesale data) was utilised for comparison.
The penicillins (ATC J01C) account for approximately 50% of all antimicrobial prescribing (ATC J01) on the PCRS scheme. Total DID consumption for this group of antibiotics has increased by 25% from 2000 to 2005, amoxicillin and co-amoxiclav accounting for 80% of the total consumption of penicillins in 2005. Similarly penicillins represented the most frequently prescribed antibiotic group in all countries in the study by Ferech et al12 ranging from 31% (Finland) to 63% (Denmark) of total DID consumption. They also found that penicillin use increased in most countries during the study period with the exception of France and the Czech Republic where consumption fell by more than 1 DID.
The consumption of co-amoxiclav on the PCRS scheme increased by 80% within the study period, whereas total consumption of amoxicillin decreased by 9%. Similarly Ferech et al12 also reported that there was a definite move from the prescribing of narrowspectrum penicillins to broad-spectrum penicillins across Europe specifically their combinations with B-lactamase inhibitors. This study identified marked seasonal variation in antibiotic consumption over the six-year period with fluctuations particularly noticeable in the penicillin, cephalosporin and macrolide groups. Similar results were identified in the ESAC study10 where the median increase of outpatient antibiotic consumption in winter months compared with summer months was 33% for 21 countries and ranged from 17% in Sweden to over 50% in Hungary and Slovakia. Seasonal variation in antibiotic consumption is most likely related to increased prescribing for respiratory tract infections in the winter months. It is important to highlight that while the PCRS data is restricted to medical card holders and does not include private prescriptions from the Drug Payment’s Scheme (DP Scheme) the same unit of measurement i.e. DIDs is used by the ESAC study which allows for broad comparisons to be made.
The majority of antibiotic prescribing (80%) occurs in primary care of which approximately 50% are prescribed for the treatment of respiratory tract infections3. A large proportion of such prescribing is as a result of viral pathogens rather than bacterial infections. It is estimated that up to 50% of antimicrobial prescribing in humans may be inappropriate13. The National Institute for Health and Clinical Excellence (NICE) published guidance on the prescribing for RTIs in July 200828 and there have also been numerous Cochrane reviews for these conditions6.These and other reports have concluded that the prescribing of antibiotics for the treatment of conditions such as sore throats, sinusitis, acute bronchitis and otitis media have minimal effect as many of these infections are self-limiting and the benefit must be weighed against the possible adverse reactions of the relevant antibiotics. Current antibiotic recommendations for the treatment of community acquired pneumonia by the British Thoracic Society and European Respiratory Society are: a beta-lactam plus/minus macrolide therapy26,27.
The appropriate use of antibiotics is essential to maximise health outcomes for patients and to reduce the development of further antibiotic resistance. Patient pressure and expectations often contribute to antibiotic prescribing even in situations where the value of such therapy is not entirely clear14. There is a strong link between the level of antibiotic consumption in the community and the level of antimicrobial resistance (AMR). This has been proven for beta-lactam antibiotic consumption and penicillin resistance in Streptococcus pneumoniae in a number of European countries1. Similar patterns have been identified in relation to macrolide consumption and resistance in group A streptococci15,25. Of increasing concern are the number of studies which highlight the association between quinolone use and clostridium difficile and MRSA29-31. The results of our analysis identified an increase in consumption of Quinolones in the PCRS scheme of 97% between 2000 and 2005. The World Health Organisation (WHO) has recognised AMR as a cause of prolonged morbidity, increased fatality and increasing the length of epidemics8,16 The incidence of antibacterial resistance among such espiratory pathogens as streptococcus pneumoniae and haemophilus influenzae has risen significantly over the past decades17. Goossens et al also identified that community antibiotic consumption correlated with geographic variation of resistance in Europe4.
Monitoring of antimicrobial utilisation has been recognised as a major component in antimicrobial resistance strategies, including the WHO Global Strategy of Containment of Antimicrobial Resistance and the Strategy for Antimicrobial Resistance in Ireland (SARI). Recommended strategies for the control of antimicrobial resistance include steps to control antibiotic utilisation18 combined with efforts to control the spread of infection19,20. The majority of European countries have placed an emphasis on two main strategies i.e. the appropriate and cautious use of antibiotic therapy both in the community and hospital setting with the guidance of appropriate antibiotic control policies and effective infection control practices in both the hospital and community settings14,21-24. Within the Irish context measures need to be put in place in the community to reduce antimicrobial consumption for e.g. the introduction of national prescribing guidelines. Tackling MRSA and Health Care Associated Infection (HCAI) is a priority for the HSE and as a result of the Say No to Infection campaign the HSE recently established a new HCAI Governance Group. It’s core five year objectives are: to reduce Health Care Associated Infections by 20%, to reduce MRSA infections by 30% and to reduce antibiotic consumption by 20% Such measures will inevitably improve patient well-being, reduce the incidence of antibiotic resistance and consequently reduce unnecessary expenditure in a budget constrained health service. Linking antibiotic consumption to resistance data across the different healthboard regions in Ireland will provide a system for monitoring variations in antimicrobial consumption and resistance over time and help assess the effectiveness of prevention programmes against AMR.
References