According to the Australian Institute of Health and Welfare (AIHW), which relies on data from the Australian Bureau of Statistics, you have to go back to the late 1950s to find the death rate in men from Chronic Obstructive Pulmonary Disease (COPD, a progressive, incurable set of diseases that include emphysema and chronic bronchitis) as low as it is today.
About 73% of deaths from COPD are attributable to smoking, so the main reason why today’s rates are where they are now is because of Australia’s internationally acknowledged success in reducing smoking. The fall in smoking began in the 1960s and has continued ever since. The fall in COPD deaths in men began about 10 years later. Because COPD cannot be cured, the dramatic downturn reflects falling smoking rates. It is an astonishing success story, but one that is far from over.
This is important background when considering an Access Economics report on COPD which hit the headlines last week. COPD is widely acknowledged as one of the most serious but least appreciated consequences of smoking. The National Quit campaign has run explicit ads simulating emphysema in an effort to raise awareness. So, by one assessment, the Access report is important grist to the mill of increasing public and political interest in the disease.
But the report quickly attracted the critical interest of some who work in calculating the extent and impact of disease in Australia. Among the foundational claims it makes is that there are “around” 16,004 deaths “due to COPD in 2008”.
Smoking of course causes many other diseases such as lung cancer and other cancers, heart disease and stroke. The most recent report published by the AIHW estimated that in 2003, total tobacco deaths (from all causes) attributable to smoking was 15,511 including 4,175 deaths due to COPD.
Using similar methodology, a report using figures from 2004-05 prepared by economists Collins and Lapsley for the Department of Health and Ageing put the figure at 14,901, including 3,870 deaths due to COPD. Both estimates of total smoking deaths from all diseases are far below the total deaths. Access Economics claims are caused by just one disease, COPD. A third estimate of deaths due to tobacco in Australia, that of Oxford University’s Sir Richard Peto and colleagues which puts deaths due to tobacco use at 19,184 only included 3,937 deaths due to smoking caused COPD.
Get Crikey FREE to your inbox every weekday morning with the Crikey Worm.
So is Access the only one who’s got it right here? Access acknowledges this massive difference between its own and AIHW estimates (“the actual AIHW reported figure of deaths from COPD is significantly lower than the attributable mortality estimate here”). Access notes also that AIHW lists COPD “7,219 times as an associated cause of death, most commonly in relation to cases where a circulatory disease was listed as the underlying cause of death”. This would be where a person (typically a smoker) had other smoking-caused diseases in addition to COPD, which could include conditions such as coronary heart disease or stroke.
In such cases, the doctor who wrote the death certificate decided that the person was known to have COPD but that another disease was the most proximal cause of death. In some cases this would artificially elevate numbers of the precipitating cause of death (say, stroke) and relegate an important contributory cause like COPD. But no one would regard the full 7,219 figure as a reasonable additional estimate of COPD as the underlying cause of death.
Access goes on to state that “due to under-diagnosis and complications with comorbidities” the AIHW reported “lower prevalence overall than estimated in this report, since their data are based on self-reported sources rather than epidemiological sources.” This is news to Professor Theo Vos from the University of Queensland, an author of the AIHW report. He says that his study’s prevalence estimates for COPD are not based on self-report at all, but rather are derived from trends applied to spirometry data obtained from the Western Australian Busselton study and then extrapolated to the Australian population. Spirometry provides objective data on lung function and plays a key role in the clinical diagnosis of COPD and other lung disease. The Busselton Study has been collecting and analysing data on lung and other diseases since 1966.
Access reports — like this one — typically list an expert panel of people who have provided guidance and comments on their reports. This is a form of peer review, but not one that would be acknowledged in research circles because of its lack of independence from the agency writing the report. The essence of peer review is that it is mediated through independent editors or competitive grant reviewing agencies. Even this process has been heavily criticised for its lack of transparency. Few journals publish the reviews they receive.
Open peer review is the process of throwing reports and papers open for public scholarly review, thereby allowing readers to gain others’ perspectives on the merits or otherwise of research. Access’s report is open to the public. The claims it makes, and the huge public costs that would be involved should the Lung Foundation’s recommendations below be taken up, are startling enough to warrant close inspection.
The Australian Lung Foundation which commissioned the report, has called for “publicly funded testing of lung functions for all current and ex-smokers over 35”. This would see about 5.6 million people being persuaded to have a test for diseases that the Foundation acknowledges can’t be cured and where the main advice that can be given is to stop smoking. If everyone eligible came forward and doctors charged the Medicare scheduled fee for spirometry of $17.75, and we assume that half of these tests were not performed during visits to the doctor for other purposes, and that these consultations lasted 40 minutes (attracting a Medicare rebate of $62.30) then we would be looking at a minimum of $448.28 million … and that would be just in the first year.
Crikey invites those with expertise in public health, epidemiology, biostatistics, respiratory medicine, and economics as well as the public at large, to provide comments on the report here.