MJA: Easthope et al., Acupuncture in Australian general practice: practitioner characteristics

Acupuncture in Australian general practice: practitioner characteristics

Gary Easthope, Justin J Beilby, Gerard F Gill and Bruce K Tranter

MJA 1998; 169: 197-200
For editorial comment, see Komesaroff
See also Acupuncture in Australian general practice: patient characteristics

Abstract - Introduction - Methods - Results - Discussion - Acknowledgements - References - Authors' details
Make a comment - Register to be notified of new articles by e-mail - Current contents list - ©MJA1998

Abstract
Objectives: To ascertain the extent of the use of acupuncture and the characteristics of general practitioners using acupuncture.
Design: Secondary analysis of 1996 Health Insurance Commission data on claims by all non-specialist medical practitioners for Medicare Benefits Schedule items for an attendance where acupuncture was performed by a medical practitioner.
Main outcome measures: Use of acupuncture by general practitioners and the practitioners' sex, age, place of primary medical qualification, and the socioeconomic disadvantage index of the practitioners' practice.
Results: 15.1% of general practitioners claimed for acupuncture. Acupuncture was more likely to be provided by male practitioners, by those aged 35-54 years, and by practitioners who have an overseas primary medical qualification. The socioeconomic index of the practice did not significantly affect the number of claims for acupuncture.
Conclusion: Acupuncture is used by about one in seven general practitioners. Its use is associated with middle-aged practitioners, who presumably have more clinical experience. This level of use by experienced doctors suggests that a critical review of the appropriate role of acupuncture in general practice should be considered.
Introduction
There is interest among general practitioners (GPs) in therapies that are not part of traditional medical practice: alternative or complementary therapies.^1-4 A recent MJA editorial stated that: "If 50% of Australians are using alternative treatments each year, it is essential that doctors recognise this fact and be prepared to discuss such use with their patients in a non-judgemental manner",⁴ while American physicians are exhorted "to begin now to integrate some aspects of alternative medicine into family practice".⁵
General practitioners are heeding this exhortation; alternative treatments are used by 16% of GPs in Britain^6,7 and Canada,⁸ and by 30% in New Zealand.^9,10 In Germany, which has a tradition of incorporating alternative therapies into orthodox medicine, 95% of GPs use them.¹¹ Even doctors who do not use such therapies refer patients to medical and non-medical practitioners, at rates ranging from 41% of doctors in Israel¹² to 80% in New Zealand.^9,10 Non-medical practitioners report that about 10% of their patients are referred from doctors.¹³
Acupuncture is frequently reported in studies of GPs' use of alternative therapies overseas.^6,9,10-12 Meta-analyses on acupuncture,¹⁴ homoeopathy¹⁵ and spinal manipulation¹⁶ do not provide evidence for their efficacy, apart from acupuncture for adult postoperative and chemotherapy nausea and postoperative dental pain,¹⁷ and spinal manipulation for lower-back pain.¹⁸ Other explanations for adoption of these therapies by doctors must therefore be sought. The fact that medicine is a profession in which clinical judgement is considered paramount is important. Interviews with doctors using alternative therapies in Queensland found they justified such use on the basis of clinical experience.¹⁹
We studied use of acupuncture to examine doctors' use of alternative therapies. The primary reason for selecting acupuncture was that information on use was available from Medicare data; a specific item in the Medicare Benefits Schedule (now, item 173; before 1991, item 980) has been used at any attendance where acupuncture was performed by a medical practitioner since 1984. Other alternative therapies do not attract a rebate or are not clearly specified. Acupuncture is of interest in that its cost to Medicare is calculable.
Methods
The Health Insurance Commission (HIC) provided data on all acupuncture claims by GPs in Australia in 1996. Information was provided on the GPs' sex, age, HIC classification (Box 1) and whether their primary medical qualification was Australian or not. The postcode of each GP's major practice was also identified. To ensure no individual was identifiable, the postcode data were classified into 62 categories (later consolidated to 12 for presentation) derived from an Australian Bureau of Statistics list which allots each postcode an index of socioeconomic disadvantage (SDI).²⁰ The postcode data are used as a surrogate measure of patient characteristics in relation to socioeconomic disadvantage on the assumption that patients generally visit doctors in their home area. Our data include the total population of GPs, and therefore any differences between categories are substantively significant.
Box 1

A logistic regression model was constructed using SPSS.²¹ Fifty-eight GPs who could not be classified either as vocationally registered or as non-vocationally registered were excluded from the logistic regression analysis.
The HIC also provided data on the number of acupuncture claims made each year since 1984, and further data were extracted from the statistics on general practice provided by Medicare.²²
The University of Tasmania Human Research Ethics Committee approved the study.
Results
Since acupuncture became a Medicare benefit item in 1984, claims have risen from 655 000 in the financial year 1984-85 to 960 000 in 1996-97 (Figure). Medicare reimbursements have increased from $7.7 million to $17.7 million.

In 1996, 2997 (15.1%) of the 19 783 GPs in Australia claimed at least once for acupuncture, and 62 (0.3%) had acupuncture as their major claimed therapy (Box 2). However, acupuncture claims constituted only 0.7% of all claims by vocationally registered GPs²² and pertained only to 1.2% of patients.
Box 2

Doctors who provided acupuncture had a higher number of non-acupuncture claims than non-providing doctors (population mean, 5029; acupuncture providers' mean, 6632; non-providers' mean, 4743).
Non-vocationally registered GPs were less likely than vocationally registered GPs to claim for acupuncture (odds ratio [OR], 0.5), while acupuncture providers were more likely to be men (OR, 1.7), aged 35-54 years (OR, 1.7) and hold primary medical qualifications from outside Australia (OR, 1.6) (Box 3). There was no significant linear or quadratic association between the odds of providing acupuncture and the SDI of the doctor's major practice.
The sex and age differentials persisted, with marginally different odds ratios, when GPs were distinguished by whether they provided acupuncture infrequently (less than 1% of services), frequently (1%-25% of services) or extensively (25%-100% of services). However, for frequent and extensive providers, there was no significant difference in the odds of providing acupuncture between those qualified in Australia and those in another country.
Discussion
There were 960 000 Medicare claims for acupuncture in the 1996-97 financial year, costing $17.7 million in reimbursements. As the HIC data do not include acupuncture services funded by agencies such as Workers' Compensation and the Department of Veterans' Affairs or services provided in public hospitals, these figures understate the total use of acupuncture in Australia.
In 1996, 15.1% of GPs in Australia claimed for acupuncture. The proportion of GPs using acupuncture in Australia is comparable to the proportion of German (15%)¹¹ and New Zealand doctors (18%)^9,10 who provide this service, but much higher than in the United Kingdom (3%).⁶ Acupuncture may be more accessible to patients in Australia than these results suggest because about 70% of GPs work in group practices.²² A survey of general practices in Hobart found that, although only 15% (27) of the GPs provided acupuncture, it was available in 31% (20) of practices.²³
Interestingly, British research found that the use of alternative therapies was more likely in single or one-partner practices than in group practices.⁷ This difference between the United Kingdom and Australia may be a function either of the 10-year gap between surveys or of different payment systems. In the United Kingdom doctors are paid a set capitation fee; in Australia a fee-for-service model operates. Choice of therapy is independent of patient demand in the United Kingdom, whereas in Australia patients can "shop around". In Australia, offering a choice of conventional and alternative treatments in group practices may simply be good marketing, so that patients can select one or the other but the practice will not lose income.
Doctors who provided acupuncture had a greater provision of all other medical services. This suggests that acupuncture is associated with a propensity to provide more services and a greater variety of services. Whether this means acupuncture is associated with practices that have a more complex mix of patients or presenting complaints can only be resolved by a detailed study.
Men were more likely than women to provide acupuncture, and this was not a function of age (the difference remained at each age category except the eldest). In this, as in other respects, women appear to practise medicine differently from men.²⁴
The greater likelihood of doctors aged 35-54 years providing acupuncture may be a function of experience. These practitioners may have had many patients with complaints that have not responded to conventional treatment. Given this experience, they are able to try other therapies and trust their clinical judgement as to efficacy. Those older than 65 years may have been less willing to try alternative therapies, or may have decided such therapies are not effective.
There are a number of possible reasons why GPs qualified outside Australia may be more likely to provide acupuncture. These doctors may have studied acupuncture as part of their initial medical training, they may have practices in immigrant areas with high patient demand for acupuncture, or they may be subject to less peer pressure to conform to orthodox practice. Without a more detailed study no one explanation can be confirmed or disproved.
Previous reports on consumers of alternative therapies suggest these therapies are the choice of young (aged 25-35 years), well-educated and financially secure people, particularly women.²⁵ If acupuncture is an index of the use of alternative therapies, it should be provided in areas of least disadvantage (SDI categories 11 and 12). This is not the case: there was no clear difference between SDI categories (the least-disadvantaged areas, 11 and 12, having only seven frequent or extensive providers). One explanation may be that acupuncture is not claimed through Medicare in these areas. A survey of acupuncture-providing practices in Hobart suggests this is likely, as most providers (60%; 26) and practices (85%; 20) were in relatively affluent suburbs.²³ Residents of such areas may also be receiving acupuncture from non-medical practitioners. Additionally, acupuncture may not be a reliable index of use of alternative therapies because it is available on Medicare. Low-income patients may try acupuncture in preference to orthodox treatments, such as physiotherapy, that they would have to pay for.
Acupuncture services in Australia have risen steadily since 1984, and 2997 GPs are using acupuncture throughout Australia. Given this level of use, a study of the appropriate role of acupuncture in Australian primary care medical practice should be considered.
Acknowledgements
The researchers wish to thank the Government Employees Medical Research Fund, which provided funds to conduct the research.
References

Wardwell WI. Alternative medicine in the United States. Soc Sci Med 1994; 38: 1061-1068.
Paterson C. Complementary medicine [letter]. Br J Gen Pract 1996; 46: 440.
Murray RH, Rubel AJ. Physicians and healers -- unwitting partners in health care. N Engl J Med 1992; 326: 61-64.
Shenfield GM, Atkin PA, Kristoffersen SS. Alternative medicine: an expanding health industry [editorial]. Med J Aust 1997; 166: 516-517.
Gordon JS. Alternative medicine and the family physician. Am Fam Physician 1996; 54: 2205-2212.
Wharton R, Lewith G. Complementary medicine and the general practitioner. BMJ 1986; 292: 1498-1500.
Anderson E, Anderson P. General practitioners and alternative medicine. J R Coll Gen Pract 1987; 37: 52-55.
Verhoef MJ, Sutherland LR. Alternative medicine and general practitioners. Opinions and behaviour. Can Fam Physician 1995; 41: 1005-1011.
Hadley CM. Complementary medicine and the general practitioner: a survey of general practitioners in the Wellington area. N Z Med J 1988; 101: 766-768.
Marshall RJ, Gee R, Israel M, et al. The use of alternative therapies by Auckland general practitioners. N Z Med J 1990; 103: 213-215.
Himmel W, Schulte M, Kochen MM. Complementary medicine: are patients' expectations being met by their general practitioners? Br J Gen Pract 1993; 43: 232-235.
Borkan J, Neher JO, Anson O, Smoker B. Referrals for alternative therapies. J Fam Pract 1994; 39: 545-550.
Fulder SJ, Munro RE. Complementary medicine in the United Kingdom: patients, practitioners, and consultations. Lancet 1985; 2: 542-545.
ter Riet G, Kleijnen J, Knipschild P. Acupuncture and chronic pain: a criteria-based meta-analysis. J Clin Epidemiol 1990; 43: 1191-1199.
Kleijnen J, Knipschild P, ter Riet G. Clinical trials of homoeopathy. BMJ 1991; 302: 316-323.
Shekelle PG, Adams AH, Chassin MR, et al. Spinal manipulation for low-back pain. Ann Intern Med 1992; 117: 590-598.
Acupuncture. NIH Consensus statement online 1997 November 3-5 [4 February 1998]; 15. In press.
Ernst E. Complementary medicine -- doing more good than harm? Br J Gen Pract 1996; 46: 60-61.
Eastwood H. General medical practice, alternative medicine and the globalisation of health [doctoral thesis]. Brisbane: University of Queensland, 1997.
Castles I, Information paper: 1991 census. Socio-economic indexes for areas. Canberra: AGPS, 1994. (Catalogue no. 2912.0.)
SPSS^x statistical package for the social sciences [computer program]. Chicago, Ill: SPSS Inc, 1997.
Commonwealth Department of Health and Family Services. General practice in Australia: supplementary tables 1997. Canberra: AGPS, 1997.
Bombardieri D. Convergence between orthodox and alternative therapies in Hobart, Tasmania [Honours thesis]. Hobart: University of Tasmania, 1997.
Britt H, Bhasale A, Miles DA, et al. The sex of the general practitioner: a comparison of characteristics, patients, and medical conditions managed. Med Care 1996; 34: 403-415.
MacLennan AH, Wilson DH, Taylor AW. Prevalence and cost of alternative medicine in Australia. Lancet 1996; 347: 569-573.

(Received 7 Jan, accepted 17 Apr 1998)

Authors' details

Department of Sociology and Social Work, University of Tasmania, Hobart, TAS.
Gary Easthope, MA, PhD, Associate Professor; Bruce K Tranter, BA, PhD, Lecturer.

Department of General Practice, University of Adelaide, SA.
Justin J Beilby, MPH, FRACGP, Senior Lecturer.

Division of Community and Rural Health, University of Tasmania, Launceston, TAS.
Gerard F Gill, MAE, FRACGP, Clinical Senior Lecturer.

Reprints will not be available from the authors.
Correspondence: Dr G Easthope, Department of Sociology and Social Work, University of Tasmania, GPO Box 252C-17, Hobart, TAS 7001.
E-mail: Gary.EasthopeATutas.edu.au

The Medical Journal of Australia eMJA

Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company
Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>".
<URL: http://www.mja.com.au/> © 1998 Medical Journal of Australia.
We appreciate your comments.


Abstract	*Objectives:* To ascertain the extent of the use of acupuncture and the characteristics of general practitioners using acupuncture. Design: Secondary analysis of 1996 Health Insurance Commission data on claims by all non-specialist medical practitioners for Medicare Benefits Schedule items for an attendance where acupuncture was performed by a medical practitioner. Main outcome measures: Use of acupuncture by general practitioners and the practitioners' sex, age, place of primary medical qualification, and the socioeconomic disadvantage index of the practitioners' practice. Results: 15.1% of general practitioners claimed for acupuncture. Acupuncture was more likely to be provided by male practitioners, by those aged 35-54 years, and by practitioners who have an overseas primary medical qualification. The socioeconomic index of the practice did not significantly affect the number of claims for acupuncture. Conclusion: Acupuncture is used by about one in seven general practitioners. Its use is associated with middle-aged practitioners, who presumably have more clinical experience. This level of use by experienced doctors suggests that a critical review of the appropriate role of acupuncture in general practice should be considered.
Introduction	There is interest among general practitioners (GPs) in therapies that are not part of traditional medical practice: alternative or complementary therapies.^1-4 A recent MJA editorial stated that: "If 50% of Australians are using alternative treatments each year, it is essential that doctors recognise this fact and be prepared to discuss such use with their patients in a non-judgemental manner",⁴ while American physicians are exhorted "to begin now to integrate some aspects of alternative medicine into family practice".⁵ General practitioners are heeding this exhortation; alternative treatments are used by 16% of GPs in Britain^6,7 and Canada,⁸ and by 30% in New Zealand.^9,10 In Germany, which has a tradition of incorporating alternative therapies into orthodox medicine, 95% of GPs use them.¹¹ Even doctors who do not use such therapies refer patients to medical and non-medical practitioners, at rates ranging from 41% of doctors in Israel¹² to 80% in New Zealand.^9,10 Non-medical practitioners report that about 10% of their patients are referred from doctors.¹³ Acupuncture is frequently reported in studies of GPs' use of alternative therapies overseas.^6,9,10-12 Meta-analyses on acupuncture,¹⁴ homoeopathy¹⁵ and spinal manipulation¹⁶ do not provide evidence for their efficacy, apart from acupuncture for adult postoperative and chemotherapy nausea and postoperative dental pain,¹⁷ and spinal manipulation for lower-back pain.¹⁸ Other explanations for adoption of these therapies by doctors must therefore be sought. The fact that medicine is a profession in which clinical judgement is considered paramount is important. Interviews with doctors using alternative therapies in Queensland found they justified such use on the basis of clinical experience.¹⁹ We studied use of acupuncture to examine doctors' use of alternative therapies. The primary reason for selecting acupuncture was that information on use was available from Medicare data; a specific item in the Medicare Benefits Schedule (now, item 173; before 1991, item 980) has been used at any attendance where acupuncture was performed by a medical practitioner since 1984. Other alternative therapies do not attract a rebate or are not clearly specified. Acupuncture is of interest in that its cost to Medicare is calculable.
Methods	The Health Insurance Commission (HIC) provided data on all acupuncture claims by GPs in Australia in 1996. Information was provided on the GPs' sex, age, HIC classification (Box 1) and whether their primary medical qualification was Australian or not. The postcode of each GP's major practice was also identified. To ensure no individual was identifiable, the postcode data were classified into 62 categories (later consolidated to 12 for presentation) derived from an Australian Bureau of Statistics list which allots each postcode an index of socioeconomic disadvantage (SDI).²⁰ The postcode data are used as a surrogate measure of patient characteristics in relation to socioeconomic disadvantage on the assumption that patients generally visit doctors in their home area. Our data include the total population of GPs, and therefore any differences between categories are substantively significant. A logistic regression model was constructed using SPSS.²¹ Fifty-eight GPs who could not be classified either as vocationally registered or as non-vocationally registered were excluded from the logistic regression analysis. The HIC also provided data on the number of acupuncture claims made each year since 1984, and further data were extracted from the statistics on general practice provided by Medicare.²² The University of Tasmania Human Research Ethics Committee approved the study.
Results	Since acupuncture became a Medicare benefit item in 1984, claims have risen from 655 000 in the financial year 1984-85 to 960 000 in 1996-97 (Figure). Medicare reimbursements have increased from $7.7 million to $17.7 million.

	In 1996, 2997 (15.1%) of the 19 783 GPs in Australia claimed at least once for acupuncture, and 62 (0.3%) had acupuncture as their major claimed therapy (Box 2). However, acupuncture claims constituted only 0.7% of all claims by vocationally registered GPs²² and pertained only to 1.2% of patients. Doctors who provided acupuncture had a higher number of non-acupuncture claims than non-providing doctors (population mean, 5029; acupuncture providers' mean, 6632; non-providers' mean, 4743). Non-vocationally registered GPs were less likely than vocationally registered GPs to claim for acupuncture (odds ratio [OR], 0.5), while acupuncture providers were more likely to be men (OR, 1.7), aged 35-54 years (OR, 1.7) and hold primary medical qualifications from outside Australia (OR, 1.6) (Box 3). There was no significant linear or quadratic association between the odds of providing acupuncture and the SDI of the doctor's major practice. The sex and age differentials persisted, with marginally different odds ratios, when GPs were distinguished by whether they provided acupuncture infrequently (less than 1% of services), frequently (1%-25% of services) or extensively (25%-100% of services). However, for frequent and extensive providers, there was no significant difference in the odds of providing acupuncture between those qualified in Australia and those in another country.
Discussion	There were 960 000 Medicare claims for acupuncture in the 1996-97 financial year, costing $17.7 million in reimbursements. As the HIC data do not include acupuncture services funded by agencies such as Workers' Compensation and the Department of Veterans' Affairs or services provided in public hospitals, these figures understate the total use of acupuncture in Australia. In 1996, 15.1% of GPs in Australia claimed for acupuncture. The proportion of GPs using acupuncture in Australia is comparable to the proportion of German (15%)¹¹ and New Zealand doctors (18%)^9,10 who provide this service, but much higher than in the United Kingdom (3%).⁶ Acupuncture may be more accessible to patients in Australia than these results suggest because about 70% of GPs work in group practices.²² A survey of general practices in Hobart found that, although only 15% (27) of the GPs provided acupuncture, it was available in 31% (20) of practices.²³ Interestingly, British research found that the use of alternative therapies was more likely in single or one-partner practices than in group practices.⁷ This difference between the United Kingdom and Australia may be a function either of the 10-year gap between surveys or of different payment systems. In the United Kingdom doctors are paid a set capitation fee; in Australia a fee-for-service model operates. Choice of therapy is independent of patient demand in the United Kingdom, whereas in Australia patients can "shop around". In Australia, offering a choice of conventional and alternative treatments in group practices may simply be good marketing, so that patients can select one or the other but the practice will not lose income. Doctors who provided acupuncture had a greater provision of all other medical services. This suggests that acupuncture is associated with a propensity to provide more services and a greater variety of services. Whether this means acupuncture is associated with practices that have a more complex mix of patients or presenting complaints can only be resolved by a detailed study. Men were more likely than women to provide acupuncture, and this was not a function of age (the difference remained at each age category except the eldest). In this, as in other respects, women appear to practise medicine differently from men.²⁴ The greater likelihood of doctors aged 35-54 years providing acupuncture may be a function of experience. These practitioners may have had many patients with complaints that have not responded to conventional treatment. Given this experience, they are able to try other therapies and trust their clinical judgement as to efficacy. Those older than 65 years may have been less willing to try alternative therapies, or may have decided such therapies are not effective. There are a number of possible reasons why GPs qualified outside Australia may be more likely to provide acupuncture. These doctors may have studied acupuncture as part of their initial medical training, they may have practices in immigrant areas with high patient demand for acupuncture, or they may be subject to less peer pressure to conform to orthodox practice. Without a more detailed study no one explanation can be confirmed or disproved. Previous reports on consumers of alternative therapies suggest these therapies are the choice of young (aged 25-35 years), well-educated and financially secure people, particularly women.²⁵ If acupuncture is an index of the use of alternative therapies, it should be provided in areas of least disadvantage (SDI categories 11 and 12). This is not the case: there was no clear difference between SDI categories (the least-disadvantaged areas, 11 and 12, having only seven frequent or extensive providers). One explanation may be that acupuncture is not claimed through Medicare in these areas. A survey of acupuncture-providing practices in Hobart suggests this is likely, as most providers (60%; 26) and practices (85%; 20) were in relatively affluent suburbs.²³ Residents of such areas may also be receiving acupuncture from non-medical practitioners. Additionally, acupuncture may not be a reliable index of use of alternative therapies because it is available on Medicare. Low-income patients may try acupuncture in preference to orthodox treatments, such as physiotherapy, that they would have to pay for. Acupuncture services in Australia have risen steadily since 1984, and 2997 GPs are using acupuncture throughout Australia. Given this level of use, a study of the appropriate role of acupuncture in Australian primary care medical practice should be considered.
Acknowledgements	The researchers wish to thank the Government Employees Medical Research Fund, which provided funds to conduct the research.
References	Wardwell WI. Alternative medicine in the United States. Soc Sci Med 1994; 38: 1061-1068. Paterson C. Complementary medicine [letter]. Br J Gen Pract 1996; 46: 440. Murray RH, Rubel AJ. Physicians and healers -- unwitting partners in health care. N Engl J Med 1992; 326: 61-64. Shenfield GM, Atkin PA, Kristoffersen SS. Alternative medicine: an expanding health industry [editorial]. Med J Aust 1997; 166: 516-517. Gordon JS. Alternative medicine and the family physician. Am Fam Physician 1996; 54: 2205-2212. Wharton R, Lewith G. Complementary medicine and the general practitioner. BMJ 1986; 292: 1498-1500. Anderson E, Anderson P. General practitioners and alternative medicine. J R Coll Gen Pract 1987; 37: 52-55. Verhoef MJ, Sutherland LR. Alternative medicine and general practitioners. Opinions and behaviour. Can Fam Physician 1995; 41: 1005-1011. Hadley CM. Complementary medicine and the general practitioner: a survey of general practitioners in the Wellington area. N Z Med J 1988; 101: 766-768. Marshall RJ, Gee R, Israel M, et al. The use of alternative therapies by Auckland general practitioners. N Z Med J 1990; 103: 213-215. Himmel W, Schulte M, Kochen MM. Complementary medicine: are patients' expectations being met by their general practitioners? Br J Gen Pract 1993; 43: 232-235. Borkan J, Neher JO, Anson O, Smoker B. Referrals for alternative therapies. J Fam Pract 1994; 39: 545-550. Fulder SJ, Munro RE. Complementary medicine in the United Kingdom: patients, practitioners, and consultations. Lancet 1985; 2: 542-545. ter Riet G, Kleijnen J, Knipschild P. Acupuncture and chronic pain: a criteria-based meta-analysis. J Clin Epidemiol 1990; 43: 1191-1199. Kleijnen J, Knipschild P, ter Riet G. Clinical trials of homoeopathy. BMJ 1991; 302: 316-323. Shekelle PG, Adams AH, Chassin MR, et al. Spinal manipulation for low-back pain. Ann Intern Med 1992; 117: 590-598. Acupuncture. NIH Consensus statement online 1997 November 3-5 [4 February 1998]; 15. In press. Ernst E. Complementary medicine -- doing more good than harm? Br J Gen Pract 1996; 46: 60-61. Eastwood H. General medical practice, alternative medicine and the globalisation of health [doctoral thesis]. Brisbane: University of Queensland, 1997. Castles I, Information paper: 1991 census. Socio-economic indexes for areas. Canberra: AGPS, 1994. (Catalogue no. 2912.0.) SPSS^x statistical package for the social sciences [computer program]. Chicago, Ill: SPSS Inc, 1997. Commonwealth Department of Health and Family Services. General practice in Australia: supplementary tables 1997. Canberra: AGPS, 1997. Bombardieri D. Convergence between orthodox and alternative therapies in Hobart, Tasmania [Honours thesis]. Hobart: University of Tasmania, 1997. Britt H, Bhasale A, Miles DA, et al. The sex of the general practitioner: a comparison of characteristics, patients, and medical conditions managed. Med Care 1996; 34: 403-415. MacLennan AH, Wilson DH, Taylor AW. Prevalence and cost of alternative medicine in Australia. Lancet 1996; 347: 569-573. (Received 7 Jan, accepted 17 Apr 1998)