VOLUME 55, NO. 7, 2002 GASTROINTESTINAL ENDOSCOPY 847 The prevalence of dyspepsia in western countries has been estimated at between 32% and 40%. 1-5 Many people with dyspepsia seek medical advice, with 42% to 56% consulting medical practitioners about symptoms. 1,2,6 Upper endoscopy has been widely adopted as the preferred diagnostic test for patients with persistent, acute, or alarming upper GI symptoms. Surprisingly, there is little documen- tation regarding trends in the use of upper endos- copy. Between 1984 and 1993 the upper endoscopy rate in the United States increased by 117% among Medicare beneficiaries, and in 1993 the endoscopy rate surpassed that for upper GI contrast radiogra- phy. 7 In Australia, diagnostic upper endoscopy was found in one study to be the most frequently per- formed of all medical procedures in people over 64 years of age. 8 With evidence of increasing utilization rates, concerns have been raised regarding unex- plained geographic variations and use of endoscopy for inappropriate indications. 9-11 Little is known about the extent of upper endos- copy use, changes in procedure rates over time, or the demographic characteristics of patients who undergo endoscopy. Trends in diagnostic testing reflect clinical practice and may be used to examine the impact of changes in health policy. In addition there is interest in the extent to which patients with private health insurance may have greater access to endoscopy compared with those who rely on govern- ment-funded health services. Australia has a universal health insurance system similar to those of Canada and the United Kingdom. In addition, around one third of Australians have private health insurance. Between 1988 and 1998 pri- vate health insurance coverage fell from approximate- ly 46% to 31%. 12 The advantages of private insurance include choice of physician for hospital care and often include faster access to diagnostic testing and elective surgery. However, individuals with private insurance may be public-insured patients some of the time by choosing to receive care as such. For example, of people with private insurance admitted to public hospitals, 60% were admitted as publicly insured patients. 13 This occurs because in many situations private insurance offers no advantage, and care as a publicly insured patient does not result in out-of-pocket expenses. Publicly insured patients may elect treatment as pri- vate patients without having private insurance cover- age (i.e., an individual can personally pay all costs Received April 10, 2001. Accepted January 25, 2002. Current affiliation: Centre for Health Informatics, University of New South Wales, Kensington, Australia. Reprint requests: Johanna I. Westbrook, PhD, Associate Professor Medical Informatics, Centre for Health Informatics, Faculty of Medicine, University of New South Wales, Kensington 2052, Australia. Copyright © 2002 by the American Society for Gastrointestinal Endoscopy 0016-5107/2002/$35.00 + 0 37/1/124634 doi:10.1067/mge.2002.124634 Trends in the utilization of diagnostic upper GI endoscopy in New South Wales, Australia, 1988 to 1998 Johanna I. Westbrook, PhD Kensington, Australia Background: Upper endoscopy is frequently performed, yet there are few published data regard- ing utilization trends. Such data provide an indication of changes in clinical practice over time and identify subpopulations who may be underexposed or overexposed to the procedure. Methods: Time series data were analyzed for all diagnostic upper endoscopies performed on res- idents of an Australia state between mid 1988 and mid 1998. Endoscopy rates by age, gender, and patient status (public vs. privately insured) were calculated. Results: In 1997/98 the endoscopy rate was 179 of 10,000.The majority were performed on publicly insured patients and women. Over the decade the rate increased by 128%. Rate increases were not constant across time, age, gender, or patient (insurance) status. The greatest increases occurred in the early to mid 1990s, and in patients less than 65 years of age. There was a greater increase (126%) among women 55 to 64 years of age compared with other age categories for both genders. Rates for publicly insured patients increased more than those for privately insured patients, par- ticularly among individuals in the age range 25 to 34 years. Rates for private patients increased more for women than men (81% vs. 59%). Conclusions: Rates of utilization of endoscopy increased dramatically during the period from 1988 to 1998, particularly during the earlier part of this decade. The cause(s) of the differences in sub- population rates warrant investigation, as do their relationship to patient management and out- comes. These data provide opportunities for international comparisons. (Gastrointest Endosc 2002;55:847-53.)