Cancer Control in Australia: Into the 21St Century

Jpn J Clin Oncol 2002;32(Supplement 1)S3–S9

Original Articles

Cancer Control in Australia: Into the 21st Century

Robert C. Burton

Anti-Cancer Council of Victoria, Carlton, Victoria, Australia

Received September 11, 2000; accepted August 13, 2001

Australia ranks highly internationally for cancer control, in part because cancer registries provide comprehensive national incidence and mortality and other cancer data annually. This has facilitated short- and long-term planning over the last quarter of a century. For example, male death rates from lung cancer in Australia peaked around 1985 and are now falling steeply, whereas female rates are still rising. In women, cervical cancer incidence and mortality have fallen owing to cervical screening over the last quarter of a century. Breast cancer mortality is now trending downwards after a half-century plateau, probably because of better treatment protocols. Cancer registry based State and national population treatment surveys over the last 15 years have identified both deficiencies and strengths in evidenced-based treatment of cancer in Australia. For example, in the State of Victoria, 5-yearly population based breast cancer treatment surveys have shown that the proportion of women who had a mastectomy versus a lumpectomy as their primary surgical operation fell from 78% in 1986 to 47% 1995. All Austral- ians are covered by a universal health care system. Therefore, socio-economic status is not a barrier to accessing high quality cancer control, whether it be prevention, screening, treatment or education. In addition, non-government organizations (NGOs) specializing in cancer control have been providing free or highly subsidized support services to patients and their families for over half a century in most states. These NGOs have also been very active in public education about cancer, especially cancer prevention.

Key words: cancer – Australia – treatment – prevention – education – statistics

INTRODUCTION Australia, but most of this is disability and very little is due to death. Cancer control is defined internationally as all actions which reduce the burden of cancer in the community: research, prevention, early detection, treatment, palliation, patient and CANCER INCIDENCE AND MORTALITY family support and education. Implicit is continuous monitor- National Australian data on cancer mortality has been avail- ing of population cancer outcomes: incidence, mortality, the able for over a century, based on information in medical certifi- treatments employed and the burden of cancer on the society. cates of causes of death provided to State registrars of births, What is the burden of cancer in Australia? There is a mea- deaths and marriages. State based cancer registries began to be sure developed through the World Health Organization called established over half a century ago (Victoria in 1936) to vali- Disability Adjusted Life Years or DALYs. In Australia, cancer date this death certificate data and to collect cancer treatment caused the second greatest burden of DALYs in 1996 and most and follow-up information from hospitals. Some State cancer of these DALYs were years of life lost due to premature death registries began to collect new case (incidence) data (New (1). Cardiovascular disease is the greatest cause of DALYs in South Wales from 1972) and national and State laws making Australia, but in terms of premature death, cancer is equal. It is cancer a notifiable disease came into effect in 1982. From 1982 worth noting that mental illness ranked third for DALYs in there has been universal registration of new cancer cases in Australia and an Australian Association of Cancer Registries which, together with the Australian Institute of Health and Welfare, produces an annual report: Cancer in Australia (2). For reprints and all correspondence: Robert C. Burton, Anti-Cancer Council of Victoria, 1 Rathdowne Street, Carlton, Victoria 3053, Australia. This network of State cancer registries has made it possible E-mail: [email protected] to undertake State and national population based treatment

Figure 1. The number of new cases and deaths from the 14 most common cancers (excluding non-melanocytic skin cancer) in Australia in 1996. surveys (described later) and other valuable epidemiological which was only the third commonest cause of cancer death in research, in addition to reporting cancer incidence and mortal- men in 1950, became the commonest fatal male cancer by ity. These registry data reveal that for females and males the 1960 (Fig. 2). As night follows day, so lung cancer mortality in world age standardized incidence of all cancers in Australia all populations peaks about 30 years after the peak prevalence peaked around 1995 and is now falling. These data also reveal of cigarette smoking. Australian male lung cancer mortality that male and female world age standardized cancer mortalities peaked around 1980 and 70% of adult Australian males have been falling slowly since the end of the 1980s. Australia smoked cigarettes in 1950. Stomach cancer, which was the is one of the few countries in the world where the whole popu- commonest fatal cancer in 1954, is disappearing and bowel and lation world age standardized cancer incidence and mortality prostate cancer have been on a plateau over the last half rates are both falling. century. The male melanoma mortality rate rose during the last In 1996 there were 14 potentially fatal cancers recorded in half century, plateaued in the early 1990s and is now falling. Australia, for which there were more than 1000 new cases (2). The commonest of these was colorectal cancer; however, it is worth noting that the gender specific cancers, prostate cancer in men and breast cancer in women, ranked second and third (Fig. 1). For many of these cancers, the incidence rate was twice or more the mortality rate, reflecting improvements in treatments that have occurred during the twentieth century. However, for cancers of the lung, pancreas and brain, incidence and mortality were almost the same, reflecting the very poor prognosis that these cancers still have. This pattern of cancer incidence and mortality is typical of westernized developed countries, where lifestyle factors such as cigarette smoking and consumption of diets high in calories and fats and low in vegetables are common.

TRENDS IN CANCER MORTALITY RATES IN AUSTRALIA Between 1990 and 1996 the world age standardized cancer mortality rates in Australia for both males and females fell by 0.4% per year (2). Mortality trends for the five commonest potentially fatal cancers in Australia, together with stomach and cervical cancers, are considered below. Cancer mortality Figure 2. Mortality trends of five common cancers in Australian men, 1950– trends in men over the last half century reveal that lung cancer, 1995. Jpn J Clin Oncol 2002;32(Supplement 1) S5

GOVERNMENT FUNDING OF CANCER CONTROL IN AUSTRALIA The Commonwealth (National or Federal) Government has the major responsibility for funding health in Australia and in respect of cancer there are three major mechanisms. Cancer research is funded via the National Health and Medical Research Council (NHMRC). In 1995 an NHMRC National Breast Cancer Centre was created to improve breast cancer management in Australia. In 1996 government funding from all sources for cancer research was about 71 million dollars (Australian dollars are quoted throughout). There is now a National Cancer Control Initiative, the counterpart of the National Cancer Institutes in other countries, which was established in 1997 to advise the Commonwealth Government about cancer policy, to produce and facilitate the implementation of a national cancer control plan and to develop and manage cancer control projects. A major mechanism for Commonwealth Government funding of health care is via funding grants to the eight State and Territory governments, all of which have health departments. Figure 3. Mortality trends of six common cancers in Australian women, 1950– 1995. There is an Australian Health Ministers Advisory Council, where the Commonwealth and State Health Ministers meet to coordinate their respective government activities in health For women, the commonest potentially fatal cancer is breast care. In respect of cancer most of these funding grants are cancer and it has been on a plateau of mortality for most of the expended on the diagnosis, treatment and palliation of cancer twentieth century (Fig. 3). Mortality from breast cancer is now in public (government) hospitals. In addition, there are the two falling in Australia as of 1995 and this is probably attributable national cancer screening programmes mentioned above and in to adjuvant chemotherapy and hormone drug treatment. The the 2000 Commonwealth Budget a feasibility study on colo- national mammographic screening programme is not expected rectal cancer screening was funded. These National Cancer to have an effect until later this decade. Bowel cancer is screening programmes are largely implemented by the State becoming less fatal in women. In the first half of the twentieth and Territory governments, using Commonwealth funding. century Australian men and women had almost the same bowel The third major mechanism of Commonwealth Government funding is via direct payments to health care professionals on a cancer fatality rates. The fall in the women’s rate, but not in the fee for service basis and via subsidization of the pharmaceuti- men’s, has been attributed to a number of factors. A likely cal drugs which they prescribe. This effectively provides free cause is the introduction of synthetic oestrogens to the treat- primary health care to all Australians via primary health care ment of women beginning in the 1950s and 1960s, in the form physicians (General Practitioners), who can refer patients to of hormone replacement therapy and the contraceptive pill. appropriately qualified specialist medical practitioners. For Stomach cancer is disappearing in women as it is in men and cancer, these are medical practitioners who have specialist for the same reasons. These are the widespread introduction of knowledge in medical oncology, surgical oncology, radiation refrigeration into Australia in the 1950s, together with more oncology, gynaecological oncology, haematology and pallia- hygienic cattle slaughtering practices. This reduced the tive care. Increasingly, cancer is being managed by multi- consumption of preserved meats, where smoking, drying and disciplinary teams and most of these are found in the 20 or salting had been common and reduced contamination with so comprehensive cancer centres in Australia. Most cancer Helicobacter pylori, which is thought to cause about 40% of all patients, however, are still treated in the community through stomach cancer world-wide. Cervical cancer mortality has hospitals and clinics. About 40% of Australians have private halved in the last half century and this is all attributable to cer- health insurance, so these treatments can be administered in vical cancer screening programmes. Melanoma mortality has public or private clinics and hospitals. been falling in women since the mid-1980s. Unfortunately lung cancer, which was ranked seventh in 1950 for mortality, CANCER TREATMENT IN AUSTRALIA is now third. The peak of cigarette smoking in Australian Despite being the commonest cause of premature loss of life in women was around 1980 and so the peak lung cancer mortality Australia, cancer rates only eighth in direct health care expend- rate is not expected until about 2010. So for all these female iture: about 2 billion dollars in 1993–94 (3). Australia is unique cancers there are now declining mortality rates except for lung in the world in spending more on skin cancer than any other cancer and this, unfortunately, has used up most of these gains. cancer. About 5% of the population of almost 20 million Aus- S6 Cancer control in Australia

Table 1. NHMRC levels of evidence supporting clinical practice

Level I Evidence obtained from a systematic review of all relevant randomized controlled trials Level II Evidence obtained from at least one properly designed randomized controlled trial Level III-1 Evidence obtained from well-designed pseudo-randomized controlled trials (alternate allocation or some other method) Level III-2 Evidence obtained from comparative studies with concurrent controls and allocation not randomized (cohort studies), case control studies or interrupted time series with a control group Level III-3 Evidence obtained from comparative studies with historical control, two or more single-arm studies or interrupted time series without a parallel control group Level IV Evidence obtained from case series, either post-test or pre-test and post-test tralians have one or more skin lesions removed every year and The first NHMRC cancer guidelines published were for the for about a third of these Australians, or 260 000 persons, the management of early breast cancer and there was a wealth of lesion is a non-melanocytic skin cancer (4). This is the penalty level 1 and level 2 evidence to support the recommendations we pay for being a white skinned people who have migrated to being made (8). In contrast, the NHMRC guidelines for the a country which is much closer to the equator than our native management of melanoma contain many recommendations lands. If Australia were superimposed on Europe, then most of based on opinion and consensus and the only randomized con- our country would lie over the Sahara Desert, so it is no trolled trials (level 2 evidence) that were available addressed surprise that indigenous Australians (Aboriginals and Torres the excision margins and the efficacy of adjuvant and palliative Strait Islanders) have dark skins. Canberra, our capital city in chemotherapy and immunotherapy (9). Currently in Australia, the south-eastern corner of Australia, is at latitude 35°S; Tokyo there are nine clinical practice guideline publications for the is at latitude 35°N. management of different cancers, of which five have already In Australia in 1900, the cure rate for non-skin cancer was been endorsed by the NHMRC. The remainder should have essentially zero. By 1950, the cure rate for all cancers was been endorsed by the end of 2000. By then most of the about 25% and this was all due to surgery. In 1991–98 the 5- common and serious cancers will have had evidence-based year survival after a diagnosis of a potentially fatal cancer in guidelines available for their treatment. the State of South Australia was 59% and in Queensland 60%; these figures should apply to the whole country (5). Most of NON-GOVERNMENT FUNDING OF CANCER this improvement in the cure rate since 1950 has come from CONTROL IN AUSTRALIA radiotherapy, cytotoxic chemotherapy and hormone treatment (6). The revolution in our understanding of cancer biology over The non-government cancer control organizations are made up the last 25 years has now resulted in thousands of potential new of the State and Territory Cancer Councils with their national pharmaceutical agents for the control of cancer and scores of Australian Cancer Society and additional State and Territory these are currently in clinical trials. So, if we about half con- bodies in the form of cancer institutes, cancer foundations and quered cancer in the twentieth century, it is difficult to believe cancer trials groups. The Australian Cancer Society coordi- that we will not complete the task by the middle of the twenty- nates the activities of the State and Territory Cancer Councils, first century. the Clinical Oncological Society of Australia and Palliative Care Australia, so that national cancer initiatives are developed EVIDENCE BASED TREATMENT logically and advocacy to the Federal Government is coordi- nated and unanimous. Whereas the Commonwealth and State In Australia today, the Commonwealth Government funding of Governments direct most of their expenditure to fund cancer health care is increasingly based on the evidence which sup- diagnosis, treatment and palliation, the non-government organ- ports the diagnostic tests and treatments being used. For cancer izations have focused on funding cancer research, cancer pre- and many other diseases, this is increasingly achieved through vention, patient support and education services. In addition to the implementation of clinical practice guidelines, which are the 71 million dollars a year provided for cancer research by developed according to rules and recommendations set down governments, the non-government cancer organizations pro- by the NHMRC (7). When guidelines are being developed for vide another 15 million dollars per year, most of which comes treatment, levels of evidence must be assigned to each of the from the Cancer Councils. This means that Australia spends recommendations (Table 1). If there is no empirical evidence, about $85 million a year on cancer research or about 4 dollars then the guidelines must state that the recommendations are per head of population and about $2 billion a year on the diag- not evidence-based, but based on opinion or consensus. This nosis, treatment and palliation of cancer. has been a revolution in my lifetime. When I was a medical The non-government organizations have been spectacularly student in the 1960s, the highest level of opinion was the opin- successful in some areas of prevention. As described previ- ion of the great man (there weren’t any great women). ously, about 70% of Australian males smoked cigarettes in Jpn J Clin Oncol 2002;32(Supplement 1) S7

Figure 4. Effect of lung cancer on the death rate from all cancers in Australian men during the twentieth century.

1950 and by 1999 total adult smoking prevalence had fallen to research. Therefore, it is heavily focused on prevention, early 22% (10). A large part of that fall came after 1970 when detection and best evidenced-based practice. One of the many targeted programmes were provided to the public on quitting unique features of this new cancer research institute is the and prevention of uptake of cigarette smoking, many of which Victorian Co-operative Oncology Group (VCOG), which con- were developed by the Anti-Cancer Council of Victoria. sists of 12 sub-speciality committees: Breast, Gastrointestinal, Behaviour change has always been a major activity of Lung, Gynaecological, Urological, Skin, Neuro-Oncology, the Anti-Cancer Council of Victoria and much of cancer Head and Neck, Genetics, Palliative Medicine, Psyco-Oncol- control depends upon changing the behaviour of the general ogy and Trials Data Management. Established in 1976, the public and health care providers. The most recent reduction in VCOG now involves more than 350 cancer specialists who adult cigarette smoking is due to Australia’s National Tobacco meet to design clinical trials, to audit treatment and to provide Campaign, which was largely developed by behavioural scien- advice to the Anti-Cancer Council of Victoria and State and tists at the Cancer Control Research Institute of the Anti-Cancer Commonwealth Governments. In recent years, cooperative Council of Victoria (10). oncology groups, modelled on the VCOG, have been estab- The impact of cigarette smoking on lung cancer mortality in lished in most other States. Australia has been profound. In 1910 about 110 per 100 000 One of the innovations of the VCOG is the population treat- Australian males died of all cancers each year (Fig. 4). If Aus- ment survey, based on cancer registry data. All persons diag- tralian men had not started to smoke cigarettes with the First nosed with a particular cancer in a particular period of time World War, then this total cancer mortality rate would have have their initial treatment documented. The first of these in remained on a plateau for the whole century. Lung cancer pro- Australia was for breast cancer in 1986 in Victoria and this was duced a rise in total cancer mortality to 160 per 100 000 Aus- repeated in Victoria in 1990 and for the whole of Australia for tralian males by the 1980s. Thankfully, with current adult male 1995, by scientists from the Centre for Behavioural Research cigarette smoking rates only one-third of what they were in the in Cancer in the Anti-Cancer Council of Victoria (11). One or 1950s, that mortality curve is now falling steeply. However, more treatment surveys for 10 different cancers have now been more than 100 000 Australian men died of cigarette smoking conducted in Victoria and the technique has been adopted by induced lung cancer in the twentieth century, more than have the cancer registries in Western Australia and New South been killed in all the conflicts in which we have ever been Wales for treatment surveys of breast and colorectal cancer in engaged. those States. All of the State registries have collaborated to complete the Australian survey of breast cancer treatment, mentioned above, and a national colorectal cancer treatment POPULATION CANCER RESEARCH survey is currently under way. The Anti-Cancer Council of Victoria has established the A comparison of the Victorian data from the 1986, 1990 and world’s first multi-disciplinary population Cancer Control 1995 breast cancer surveys, where the doctors who treated all Research Institute. It consists of research units of behavioural women newly diagnosed in a 6-month period in each of those science, epidemiology, clinical trials research and education years were approached via the Victorian Cancer Registry to S8 Cancer control in Australia complete a questionnaire of the treatment the women had most skin cancers begin in childhood and that childhood sun- received, is revealing (12). In 1986, 89% of women had com- burn is a critical factor. Therefore, the Anti-Cancer Council of plete treatment details obtained and by 1995, this had risen to Victoria began a programme entitled Slip (on a T-shirt), Slop 97%, attesting to the cooperation of all oncologists. The pro- (on sunscreen) and Slap (on a hat) in the early 1980s and this portion of women who had a mastectomy versus a lumpectomy was succeeded by SunSmart over the last decade (15). Popula- fell from 78% in 1986 to 47% in 1995. So for the State of tion measures conducted over summer time in Melbourne Victoria, we now know that about half of all newly diagnosed (population 3.3 million), the capital city of the State of Victoria women with breast cancer are having a mastectomy and about (population 4.7 million) have shown that sunburn rates have half are having a lumpectomy. This gives us a measure of how halved in the last decade. Taken together with data which show the whole population is being treated, in contrast to data from that non-melanocytic skin cancer and melanoma incidence is a comprehensive cancer centre where a select group of patients falling in Australians aged less than 50 years, this is good evi- will be treated by the most advanced means. This may have dence that these sun protection campaigns have had an effect. limited application to the ‘real world’. Public health messages to ‘get to know your skin’, in order to These three surveys also reveal that the average number of diagnose melanoma early, began in some States of Australia in breast cancer cases treated per surgeon rose from three to seven the 1950s and have been part of the SunSmart campaign in in that 10-year period and so there is an increasing focus of later years. The early detection of melanoma is almost equiva- breast cancer on specialists. In 1986 less than half of women lent to a cure. These public health campaigns have had their who had a lumpectomy had adjuvant radiotherapy and this effect. More than 75% of invasive melanomas now diagnosed improved only by about 15% in 10 years. The evidenced-based in Australia are less than 1.5 mm in thickness and the overall guidelines for the management of early breast cancer recom- 5-year survival for melanoma exceeds 90%. Australia is the mend that all women who have less than mastectomy should only country in the world where age standardized mortality receive adjuvant radiotherapy. Therefore, at the whole popula- rates for melanoma for men and women are now falling (16). tion level, the women of Victoria are receiving considerably less than best evidenced-based practice. Finally, one would SERVICES FOR PATIENTS AND THEIR FAMILIES expect that all women who have lymph nodes positive for breast cancer would receive chemotherapy or hormonal The burden of cancer in Australia in DALYs is high and it is therapy and adjuvant drug therapy was used in 82% of such also high when measured by the American Cancer Society cri- women in 1986. This indicates that best evidenced-based teria of the number of incident cases in a given year, plus the practice had probably been instituted in Victoria by 1986. number of deaths in a given year, plus the number of patients Hormonal treatment for post menopausal women with breast diagnosed with cancer in the previous 5 years who are still cancer almost tripled between 1986 to 1995 and would now alive. Using that measure, the burden of cancer in Australia in meet the criteria for best evidenced-based practice. These are 1996 was ~300 000 persons. Given that 95% of cancer occurs all positive changes and, in fact, the 5-year survival after diag- after the age of 40, most of these Australians would have had a nosis of breast cancer increased from 72% to 82% between spouse, parents perhaps, brothers and sisters and children. If 1972 and 1995, in our most populous State of New South we assume an average of six close relatives per patient, then about 1.8 million Australians were having a first hand experi- Wales (13). We are currently ranked second in the world for ence of cancer in that year or about 10% of the whole popula- 5-year survival following treatment for breast cancer, after the tion. white female population of the USA. Therefore, support for patients, their families and their carers is a crucial element of cancer control. This has also been an SKIN CANCER PREVENTION AND EARLY DETECTION IN area where non-government agencies have been pre-eminent. AUSTRALIA There is an Australian telephone number, where for the cost of Australia has a very special problem with skin cancer. Non- a local call the caller can contact the cancer information and melanocytic skin cancer has not been registered by our cancer support service in their own State. In Victoria in 1999, the registries, because more than a quarter of a million Australians Anti-Cancer Council of Victoria received over 30 000 calls to are diagnosed with such a cancer every year and the numbers its cancer Helpline and the commonest reason to call was for would simply overwhelm the registries. There have, however, breast cancer. The cancer Helpline operates five days per week been three population surveys, done in 1985, 1990 and 1995 and the staff consists of about two-thirds trained counsellors, (14). These show that non-melanocytic skin cancer incidence most of whom are oncology nurses, and about one-third infor- is increasing in Australians aged more than 50 years, but mation officers. The telephone answering is hands free and the falling in Australians aged less than 50 years. This fall has staff enter details into a database developed for this purpose, so been attributed to a quarter century of strong public health that accurate demographics on the nature of the caller and the campaigns which have stressed protecting children and adults calls can be recorded as part of our quality assurance and from the sun. These have been largely run by non-government tailoring the system to meet needs. Calls are all anonymous, organizations of which the Anti-Cancer Council of Victoria unless the caller wishes to have material mailed, faxed or e- has been a leader. Epidemiological research has shown that mailed to them. This service also provides no-interest loans Jpn J Clin Oncol 2002;32(Supplement 1) S9 and small grants to help patients and their families at times of 3. Mathers C, Penn R, Sanson-Fisher R, Carter R, Campbell E. Health great financial need and together with LaTrobe University, Systems Costs of Cancer in Australia 1993–94. AIHW Cat. No. HWE 4. Canberra: Australian Institute of Health and Welfare and National Cancer runs a breast cancer support nurse training scheme which has Control Initiative (Health and Welfare Expenditure Series No. 4) 1998. had over 200 graduates. This is a distance learning course and 4. Burton R. Aspects of screening for skin cancer. Cancer Forum at least two Japanese nurses have graduated. 1996;20:244–6. 5. South Australian Cancer Registry. Epidemiology, Mortality and Survival, In conclusion, we are somewhat of an anomaly in Asia: a 1977–1999. Incidence and Mortality, 1999. Adelaide: Openbook Publica- white population living in a very alien solar environment tions 2000. which has made skin cancer our commonest and most expen- 6. DeVita VT Jr. Progress in cancer management. Cancer 1983;51:2401–9. sive cancer. We share with the developed northern hemisphere 7. A Guide to the Development, Implementation and Evaluation of Clinical Practice Guidelines. Canberra: National Health and Medical Research countries the outcomes of high calorie, high fat, low vegetable Council, NHMRC, Commonwealth of Australia 1999. diets, past high cigarette consumption and decreasing exercise 8. Clinical Practice Guidelines: the Management of Early Breast Cancer. rates. So heart disease, obesity, diabetes and a range of cancers Canberra: National Health and Medical Research Council, NHMRC, which result in part or in whole from these risk factors are Commonwealth of Australia 1995. 9. Clinical Practice Guidelines: the Management of Cutaneous Melanoma. common: lung, colorectal, breast and prostate cancers. We are, Canberra: National Health and Medical Research Council, NHMRC, however, bringing many cancers under control through effec- Commonwealth of Australia 1999. tive programmes of prevention and early detection. Further, we 10. Hazzard K, editor. Australia’s National Tobacco Campaign: Evaluation Report, Vol. 2. Canberra: Commonwealth of Australia 2000. have one of the highest total population 5-year survival after 11. Surgical Management of Breast Cancer in Australia in 1995. Sydney: treatment for cancer in the world. A consistent approach by NHMRC National Breast Cancer Centre 1999. government and non-government agencies, with comprehen- 12. Hill D, White V, Pruden M, Giles G, Collins J. Surgical Management of sive population measures and effective programmes across the Breast Cancer in Victoria in 1995. Melbourne: Cancer Control Research Institute, Anti-Cancer Council of Victoria 2000. continuum of cancer control, have resulted in Australia being 13. Supramaniam R, Smith DP, Coates MS, Armstrong BK. Survival from one of the most successful countries in controlling cancer for Cancer in New South Wales in 1980 to 1995. Sydney: NSW Cancer Coun- its citizens. cil 1998. 14. Staples M, Marks R, Giles G. Trends in the incidence of non-melanocytic skin cancer (NMSC) treated in Australia 1985–1995: are primary preven- References tion programmes starting to have an effect? Int J Cancer 1998;78:144–8. 15. SunSmart Evaluation Studies Number 6: the Anti-Cancer Council of 1. Mathers CD, Vos ET, Stevenson CE, Begg SJ. The Australian Burden of Victoria’s Skin Cancer Control Program 1996/1997 and 1997/1988 and Disease Study: measuring the loss of health from diseases, injuries and Related Research Evaluation. Melbourne: Anti-Cancer Council of risk factors. Med J Aust 2000;172:592–6. Victoria 1999. 2. Australian Institute of Health and Welfare (AIHW) and Australasian 16. Giles GG, Armstrong BK, Burton RC, Staples MP, Thursfield VJ. Has Association of Cancer Registries (AACR). Cancer in Australia 1996: Inci- mortality from melanoma stopped rising in Australia? Analysis of trends dence and Mortality Data for 1996 and Selected Data for 1997 and 1998. between 1931 and 1994. Br Med J 1996;312:1121–5. AIHW Cat. No. CAN 7. Canberra: AIHW (Cancer Series) 1999.