Our study provides time trends for cancers of individual sub-sites according to the ICD 10 classification for those head and neck neoplasms which have, to a degree, common risk factors. Joinpoint analysis provides a much clearer picture of time trends, in different segments of time, within the overall period concerned. As such we were able to show that significant changes in trends have taken place during this period.
There has been an encouraging decline in lip, oral cavity and pharyngeal cancers overall (ICD10, C00-06, C09, C10 and C14) in Australia in recent decades. However, when cancers of the oropharynx are considered separately, rising trends have been shown, particularly among men, from 1982 to 2008.
We have shown biphasic trends for Lip, Oral Cavity and for Oropharyngeal cancers combined for both men and women. An increasing trend of 0.9% p.a. from 1982 to 1992 and a decline of 1.6% p.a. from 1992 to 2008 were observed in males. Similar biphasic trends were demonstrated among females with a steady increase of 2.0% pa from 1982–1997 and a sharp decline of 2.8% pa thereafter. In a study covering the State of New South Wales, Macfarlane et al., in 1994, reported a similar pattern:. They found increasing incidence of “oral and pharyngeal” cancer from 6.5/100,000 pa to 9.3/100,000 pa among males, and from 2.1/100,000 pa to 3.0/100,000 pa among females, from 1974 to 1986: However, this trend declined thereafter . Another epidemiological study on “lip and oral cavity” (which unfortunately included cancers of the major salivary glands), based on data from the Cancer Registry of Western Australia, reported increasing trends in “oral and pharyngeal cancer” from 1982–1990 at the rate of 14.6/100,000 p.a. for males and 6.2/100,000 p.a. for females and observed declining trends thereafter .
Since the second half of the last century reports from many parts of the world on the incidence of “oral cancer” have described declining, stable or increasing rates in different regions or countries [4, 9–12, 15, 26]. Because of the impossibility of linking cause and effect directly, there is no unambiguous explanation for the causes of these trends. However, the most conceivable explanations are life-style changes, particularly changes in smoking rates  and spread of HPV infections [10–12].
Overall per capita tobacco consumption in Australia has declined steadily since the latter part of the last century. Among males the estimated prevalence of tobacco use declined from 58% in 1964 to 18% in 2007. In contrast, among females the prevalence of tobacco smoking increased from 28% in 1964 to a peak of 31% in 1980, with a subsequent decline to 15.2% in 2007 .
The National Drug Strategy Household Survey 2010 revealed a substantial – almost 40% - reduction in the prevalence of daily smokers in Australia for people aged 14 years or older from 24.3% in 1991 to 15.1% in 2010, . However, increased smoking among females from 1964 to 1980 may have contributed to the statistically significant increase of lip, oral cavity and pharyngeal cancers observed in the present analyses during the 1982–1992 period.
The synergistic effect of alcohol consumption and smoking has been well established . Overall per capita alcohol consumption in Australia in 1960 was estimated at 9.4 L pa. This gradually increased to 13.0 L in 1980 and slowly declined to 10.1 L in 2009 . In 2010, 1 in 5 people in Australia at or over the age of 14 years consumed alcohol at harmful levels .
A recent report from France indicated a considerable decrease of upper aero-digestive tract cancers in men, while the same were increased in women over the 25 year period from 1980 to 2005, especially oropharyngeal, palatal and hypopharyngeal cancers: world-standardised incidence rates of lip, oral cavity and pharynx cancers combined declined by 42.9% in men while females showed an increase by 48.6% . Decreasing prevalence of smoking among men in the general population and slightly increasing tobacco smoking in women were suggested as accounting for these changes . Significant declines in the incidence of oral cavity and pharyngeal cancers for both men and women in all races in the USA have been observed over the period from 1977 to 2007, reflecting the steady decline in smoking and alcohol consumption in that nation . Another recent study from the USA reported decreasing trends of oral plus pharyngeal cancers for women with APC of −1.0 from 1992 to 2008. In contrast to women, although men showed a decreasing trend with APC of −1.4 from 1982 to 2006, this turned to a rise of 3% pa from 2006 . Unfortunately it is not possible to separate sub-sites in these data.
Lip cancer has been the dominant site in the oral and oropharyngeal region in Australia, contributing over 36% of cases, of which 90% are cancers of the lower lip. These are more common in males with ASI 4.93/100,000 and 3.34/100,000 in 1982 and 2008 respectively. Although this cancer has shown an increasing trend from 1982 to 1996 with an annual change of 1.5%, a decline of 4.2% per year was observed thereafter. Compared to males the overall ASI was small, but a comparable biphasic trend pattern was observed among females, with 6.1% annual increase and 4.3% annual decrease for the same periods.
Contrary to this Tan (1971), in a countrywide hospital based survey, reported a declining trend in lip cancer (upper and lower lip combined) from 6.5/100,000 in 1959 to 4.9/100,000 in 1964 . He found that cancers of the lower lip were 9.9 times more common among males compared to females, whereas the present study revealed a male to female ratio of only 3:1. However, these are quite old data and, in a hospital-based study, under-reporting is likely.
In a state-based study on lip cancer in Western Australia, Abreu et al. (2009) reported an upward trend with 8.9/100,000 and 2.7/100,000 pa for males and females respectively from 1982 to 2006 . These figures are high compared to the national data reported here, variations in incidence in different states in Australia probably being attributable to differences in the rural/urban population mix and in exposure to risk factors. Lip cancer is much more common in those who live or work outdoors, with direct exposure to sunlight [33, 34]. High incidence has long been associated with prolonged exposure to solar radiation, especially in people with fair complexion [33–37]. The lower lip receives considerably more direct sunlight than the upper lip . In contrast, the comparatively low incidence of lower lip cancer among females could be attributed to the protective effect of cosmetics and lower outdoor exposures .
The present study revealed that cancer of the upper lip was higher among females than males. Moreover, there was a sharp increase in incidence from 1982 to 1990, an annual change of 13.1%, which started declining thereafter with annual change of 1.3%. In contrast, a slight decline in cancer of the upper lip was observed amongst males over the whole period from 1982 to 2008, with an annual change of 0.1%. As with the present study, significant female predilection for cancers of the upper lip was reported from Western Australia . An almost equal sex distribution of upper lip cancers was reported in another Australia-wide hospital-based study in 1971 . Although no unambiguous explanation for a higher incidence of upper lip cancers among females can be given, differences in exposure factors, particularly increasing prevalence of tobacco smoking among females from the early 1960’s until 1980, may have contributed .
In contrast to “lip and oral cavity” cancers, when “oropharyngeal” cancers are considered separately, we have observed increasing trends, throughout the period analysed here. Oropharyngeal cancer has been reported to be increasing significantly and quickly in several countries, particularly in the developed world, and is widely regarded as associated with infection with humanpapillomaviruses of known high oncogenic potential – especially HPV-16 and −18 [10, 13, 14, 39–41]. A recent report from Australia described a significant increase in potentially HPV-associated head and neck cancers in both males and females between 1982 and 2005, with an annual percentage increase of 1.04% and 1.42% for females and males respectively . Our findings are comparable with this, but changes in life-style risk factors, especially smoking and heavy alcohol consumption, and their synergism, will have confounding effects in understanding the causes of cancers in these sites. Varying degrees of exposure may partly explain differences between males and females.
Although overall rates of lip, oral cavity and pharyngeal cancer are currently declining in Australia, these are still high in comparison with many other countries. Efforts to reduce the burden of these cancers remain vital. Further reductions in exposure to lifestyle risk factors: ultraviolet light; all forms of tobacco; excessive alcohol use/abuse; and the consumption of diets rich in antioxidants and minerals, need to be promoted. Sexual hygiene needs to be promoted to reduce the carriage of HPVs in the upper aero-digestive tract: it will be interesting to examine the extent to which current vaccination programs against oncogenic HPVs, at present focused on young women for the prevention of cancer of the uterine cervix, lead to reductions in oropharyngeal cancer in the long term [42, 43].
Limitations of our study include the small number of cases in certain sub-sites and subgroups, an inevitability in a nation with a small population (20 million and less during the period under study) spread over a vast geographical area, and that we have not been able to explore differences by ethnic group as we were not permitted access to this information for ethical reasons.