eMJA     The Medical Journal of Australia

Home | Issues | eMJA shop | My account | Classifieds | Contact | More... | Topics | Search   

Abstract

MJA 1997; 167: 124-127  

Introduction

To update these data, DUSC convened a working group in 1995, comprising representatives of DUSC, the Australian Pharmaceutical Manufacturers Association and the Therapeutics Resource and Educational Network for Doctors (TREND) project of the Royal Australian College of General Practitioners.

This group reviewed Australian and international data on antibiotic sales and dispensing to determine patterns of antibiotic use in Australia between 1990 and 1995, and to compare these with patterns in similar developed countries. To understand trends in drug use, the group also analysed prescriber surveys of the indications for antibiotic use.  

Methods

Prescription dispensing data: These were obtained from the database maintained by the DUSC secretariat which monitors dispensing of prescription medicines through community pharmacies in Australia. These data include dispensing to some private hospitals, but not public hospitals. The database contains information on all subsidised prescriptions processed by the Health Insurance Commission, together with an estimate of non-subsidised prescriptions from an ongoing survey of a sample of about 250 community pharmacies.² Measurement units are either prescription volumes or number of defined daily doses (DDDs)/1000 population/day. The DDD is based on the assumed average daily dose of the drug when used for its main indication by adults. It is the unit approved by the World Health Organization for drug use studies and allows for comparisons independent of differences in price, preparation and quantity per prescription.³

Sales data: Data on sales of oral antibiotics to retail and hospital markets in a number of countries were obtained from Intercontinental Medical Statistics (IMS), Melbourne. This is a commercial market research organisation which collects sales data and surveys general practitioner (GP) prescribing in 33 countries.⁴ Data were retrieved as unit sales by form and strength and drug class and then converted to DDDs/1000 population/day.

Prescriber survey: Diagnoses for which patients were prescribed anti biotics and patient ages were obtained from the Australian Medical Index (AMI), the survey of GP prescribing conducted by IMS. This uses a sample of 420 GPs stratified in line with the total Australian GP population by metropolitan/country location, age, year of graduation, and practice size (patient numbers). These GPs record all patient encounters over seven consecutive days in each quarterly survey period. The data are then projected to obtain quarterly estimates of prescribing patterns by diagnosis and by patient age and sex for the Australian GP population. At the end of a four-quarter cycle, GPs are invited to continue participating. About half agree, and the remainder are replaced by new recruits. As GPs start in various quarters of the year, there is continual turnover; GPs are not usually kept in the survey for more than two years.

TREND project: Use of antibiotics for management of upper respiratory tract infection (URTI)/pharyngitis and influenza was examined using the database of 33 203 doctor-patient encounters recorded in August and September 1994 by the TREND project. This project was part of the development of a new practice assessment in therapeutics option within the RACGP Quality Assurance and Continuing Education Program. Participating GPs recorded prospectively their drug and non-drug management, including lifestyle advice, investigations and referrals, for 110 consecutive patient encounters of all types (including surgery and telephone consultations, home, hospital and nursing home visits).

The GP sample was drawn from two urban and two rural areas in NSW which were selected randomly from all NSW Divisions of General Practice, after matching within urban (16 Divisions) and rural (11 Divisions) strata. Matching variables were the index of relative socioeconomic disadvantage (from the 1991 Census⁵ ) and GP-to-population ratios (from the Medicare claims database as at the end of June 1993) (Gordon Calcino, Acting Director, Technical Support Section of the General Practice Branch, Department of Health and Family Services, personal communication). The rural areas included both major and outlying towns, but not remote rural areas. All GPs from the four areas were eligible for the survey, based on a list provided by the General Practice Branch from the Medicare claims database. Of the 783 GPs approached, 303 (38.7%) completed the practice assessment in 1994. Participants had a similar sex distribution to the 1994 NSW GP population, but a somewhat younger average age.⁶  

Results

International comparisons of antibiotic usage

Sales of oral antibiotics to hospitals in 1994 and the percentage this represented of the total market are shown in Box 1 (above). The percentage was similar in Australia, Canada and the UK (7%, 9% and 11%, respectively), lowest in the US (5%) and highest in West Germany (38%).

Figure 2 shows the percentage split of the oral antibiotic market by drug class in 1994. Australia had the highest percentage use of tetracyclines (25.5% of total oral antibiotics), but the lowest use of fluoroquinolones (2.2%) and mid range use of penicillins -- both narrow spectrum (7.1%) and broad spectrum (35%). The overall profile of antibiotic use in Australia was similar to that in the UK.  

Types of antibiotics used in Australia and indications

Converting prescription volume to DDDs/1000 population/day altered the relative ranking of some drugs. For example, drugs in the tetracycline group, which have increased maximum quantities available for treatment of severe acne, rose in rank (1990). In contrast, cefaclor, which has a high DDD (1.5g) by Australian standards, fell from third to seventh position (1995).

Box 2 also shows antibiotic prescribing profiles for various indications in 1995. For sinusitis, the most prescribed antibiotics were doxycycline (20.7%), amoxycillin-clavulanate (18.1%) and cefaclor (15.1%). For bronchitis, the most prescribed antibiotic was amoxycillin (18.1%), followed closely by roxithromycin (16.5%) and cefaclor (15.2%). In urinary tract infections, trimethoprim-sulfamethoxazole (28.5%) was most commonly prescribed, followed by cephalexin (18.9%) and amoxycillin-clavulanate (17.2%).

Antibiotic prescribing for upper respiratory tract infection (URTI)/pharyngitis and influenza was examined in the TREND sample of GPs from the August-September 1994 audit; 11.6% of encounters were for URTI/pharyn gitis and 1.2% for influenza. For new cases of URTI/pharyngitis, an antibiotic prescription was recorded for 57% of urban patient encounters and for 73% of rural patient encounters. Corresponding figures for antibiotic prescribing in new cases of influenza without pneumonia were 30% of urban and 62% of rural patient encounters. The TREND project did not evaluate whether the prescription was to be filled only under certain conditions (e.g., if symptoms had not improved after a certain interval), but some GPs commented that this was their advice to patients. For symptomatic management, the most common choices were decongestants and analgesics.

AMI survey data showed that overall antibiotic usage by age broadly followed the proportional representation of particular age groups in the community, with the exception of the 0-19 years age group. This group received 36.9% of antibiotic prescriptions, but made up only 28.7% of the population. However, for specific conditions the age patterns varied. For example, the TREND project showed that an antibiotic was less often prescribed for children aged 0-5 years with URTI (34.1%) than for adults aged 50 years and over (62.5%).

Figure 3 shows quarterly fluctuations in dispensing of oral antibiotics compared with oral -blockers -- a drug type used on an ongoing basis. Antibiotic dispensing was markedly seasonal, with higher levels in the winter quarters (April to June and July to September), while dispensing of oral -blockers was low at the beginning of the year and high at the end. The second pattern was typical of "safety net" fluctuations.⁷ These result from Pharmaceutical Benefits Scheme (PBS) provisions that provide drugs free, or at lower cost, to the patient if they are dispensed after the patient has incurred a set expenditure on PBS items in a calendar year (the "safety net" threshold). Consequently, patients tend to fill repeat prescriptions towards the end of the safety net year, after they reach this threshold. Changes to reduce these fluctuations were introduced in November 1994.  

Discussion

The comparison between seven major developed countries showed that community use of antibiotics in Australia between 1985 and 1994 was high -- second only to France, with the US a close third -- but did not increase during the 1990s, as it did in most of the other countries.

In 1994, hospitals accounted for only 7% of retail sales of oral antibiotics in Australia, similar to the percentages seen in Canada and the UK. Similarly, a survey of cardiovascular drug use in Australian public hospitals found that hospital use accounted for less than 10% of the total use of all cardiovascular drug groups.⁸ The highest percentage of antibiotic use in the hospital sector was found in West Germany, where capping of GP prescribing budgets was introduced in 1993. It has been argued that these prescribing budgets for reimbursable drugs have increased hospital budgets.⁹

Australia had the highest percentage use of tetracyclines among the seven major developed countries, possibly because of its use in managing acne. In contrast, Australia had the lowest percentage use of fluoroquinolones, most likely because of PBS prescribing restrictions ("authority required").

Amoxycillin remains the antibiotic most dispensed through community pharmacies in Australia, but its use declined between 1990 and 1995, undoubtedly because of transfer prescribing to amoxycillin-clavulanate or other antibiotics. Concerns over an increased risk of severe adverse reactions with trimethoprim-sulfamethoxazole in the elderly¹⁰ and of hepatotoxicity with flucloxacillin¹¹ were probably respon sible for the marked fall in use of these drugs and the rise in use of cephalexin, a substitute for flucloxacillin in skin and soft-tissue infections pending the availability of dicloxacillin (PBS listed in 1997).¹²

The winter increase in dispensing of oral antibiotics in Australia contrasted with the "safety net" fluctuations of the oral -blockers, which are used on an ongoing basis, and probably reflected treatment of respiratory tract infections. In fact, antibiotics were prescribed for 57%-73% of new cases of URTI. This contrasts with peer consensus recommendations on the use of antimicrobial drugs in medical practice. The Antibiotic guidelines¹³ state that for URTI "the cause is almost invariably viral" and "antibiotics are not indicated". Furthermore, TREND data show that rural GPs are more likely to prescribe antibiotics for URTI than urban GPs. The difference may reflect rural GPs' concern for the greater patient travel and inconvenience in visiting the doctor in the country, with prescriptions being supplied for use if symptoms fail to resolve after a period of time. However, it may also reflect differences in access to continuing education and in industry promotion between rural and urban GPs.

The data also provide insight into GPs' perceptions of antibiotic resistance among bacteria. For sinusitis, the most prescribed antibiotics were doxycycline, amoxycillin-clavulanate and cefaclor, which are recommended if resistance to amoxycillin issuspected or proven.¹³ For bronchitis, amoxycillin was most prescribed, followed closely by roxithro mycin and cefaclor, which are recommended if a b -lactamase producing organism is isolated or if the clinical response is slow. Our results indicate a strong perception among doctors that resistant organisms are a significant clinical problem, at least for sinusitis and bronchitis. Reasons for this are unclear and deserve investigation; they may include previous clinical experience of slow resolution of these infections with standard therapy, or awareness of the prevalence of resistant bacteria in the local community.

A corollary to the study of antibiotic use is an examination of the patterns of antibiotic resistance in bacteria, which may be associated with levels of use of particular antibiotics and with total use. This issue is currently being explored in Australia by the Australian Group on Antimicrobial Resistance and by the National Antimicrobial Resistance Surveillance Program and internationally by the Alexander Project.  

Acknowledgement

References

1. Birkett DJ, Mitchell AS, Godeck A, et al. Profiles of antibacterial drug use in Australia and trends from 1987 to 1989. A report from the Drug Utilization Subcommittee of the Pharmaceutical Benefits Advisory Committee. Med J Aust 1991; 155: 410-415.
2. Edmonds DJ, Dumbrell DM, Primrose JG, et al. Development of an Australian drug utilisation database. A report from the Drug Utilization Subcommittee of the Pharmaceutical Benefits Advisory Committee. PharmacoEconomics 1993; 3: 427-432.
3. Nordic Council on Medicines. Nordic Statistics on Medicines 1990-1992. NLN publication number 34. Uppsala, Sweden: NLN, 1993.
4. Hurley SF, McNeil JJ, Berbatis CG. Sources of Australian pharmacoepidemiology data. Community Health Studies 1988; 12: 82-96.
5. Australian Bureau of Statistics. Index of relative socioeconomic disadvantage. Canberra: ABS, 1993. (Catalogue no 1356.0.)
6. Australian Institute of Health and Welfare. Medical labour force 1992-93. National Health Labour Force Bulletin No. 3. Canberra: Australian Institute of Health and Welfare, Jul 1995.
7. McManus P. Drug utilisation [letter]. Med J Aust 1993; 158: 724.
8. Doecke C, Harvey R, Havas L. Cardiovascular drug use in Australian hospitals 1990. Summary report of a survey conducted by the Society of Hospital Pharmacists of Australia and the Australian Institute of Health for the Commonwealth Department of Community Services and Health. Canberra: the Department, Oct 1991.
9. German drug caps boost hospital costs SCRIP 1993; 1877: 4.
10. Adverse Drug Reactions Advisory Committee. Trimethoprim-sulphamethoxazole warning on elderly. Aust Adverse Drug React Bull 1990 Fed.
11. Fairley CK, McNeil JJ, Desmond P, et al. Risk factors for development of flucloxacillin associated jaundice. BMJ 1993; 306: 233-235.
12. Turnidge J. What to use instead of flucloxacillin [editorial]. Aust Prescriber 1995; 18: 54-55.
13. Victorian Medical Postgraduate Foundation. Antibiotic guidelines 1996/97. 9th edition. Melbourne: The Foundation, 1996.