Join Our Mailing List

Building on Hope

Visit our SunSmart Shop for all your sun protection needs

Australia's Biggest Morning Tea ABMT 2012Community Speakers Program

Follow Us

 

Methodology

QCSOL Home QCSOL Methodology Help with QCSOL

Data Sources

Methodology

Other

Data Sources

Queensland Cancer Registry (QCR)

Details of all cancers diagnosed in Queensland are legally required to be included in the QCR under the Public Health Act 2005. Notifications of patients with cancer are received from all public and private hospitals and nursing homes. Queensland pathology laboratories are also required to provide copies of pathology reports for cancer specimens. Information regarding the deaths of persons with cancer is provided to the QCR from the Registrar of Births, Deaths and Marriages.

Non-melanoma skin cancers were not included in the comparisons of cancer types. This is because non-melanoma skin cancers are not registered by the QCR (similar to the practice in most other cancer registries), since many are treated in doctors’ surgeries using techniques that preclude histological confirmation.

The information available also does not include any adjustment for stage of cancer (a measure of how far the cancer has spread at the time of diagnosis). As is the case for all cancer registries in Australia, complete clinical staging data are not routinely collected by the QCR (although New South Wales collects a measure of the degree of spread). The absence of information on cancer stage makes it difficult to distinguish between early/late diagnosis and better/worse management of the cancer as possible reasons for any observed differences in cancer survival.

Australian Bureau of Statistics (ABS)

De-identified unit record mortality data for all causes of death for Queensland residents were obtained from the Australian Bureau of Statistics. These data were used in relative survival calculations (see Survival). Permission was required from the Registrar of Births, Deaths and Marriages in every State and Territory in Australia to access these data, since some Queensland residents die interstate. Note that cancer mortality data are available from both the Australian Bureau of Statistics and the Queensland Cancer Registry. Differences in coding practices and residential criteria can result in slight differences in the counts and rates calculated from the two data sources. These differences are generally small.

Methodology

Due to confidentiality requirements, data for which there were less than 5 cases per year, or less than 5 cases in total over a specified time period (e.g. 2004-2008) have been replaced in tables with an asterisk, and are not displayed in the graphs.

Age-standardised rates

Age-standardised rates attempt to adjust for variation in age structures in different populations (either different geographical areas or the same population across time). There are two methods of age-standardisation – direct and indirect.

All incidence and mortality trends were calculated using directly standardised rates. The method involves applying age-specific rates from the population of interest (i.e. Queensland) to a standard population, which on this website is the Australian Standard Population 2001. Five-year age groups up to 85 years and over were used for all of the age-standardisation.

Annual percentage change (APC)

This is the annual increase or decrease in the incidence or mortality trends over the specified period. Negative APC values describe a decreasing trend and positive APC values describe an increasing trend. A trend is taken to be statistically significant if the 95% confidence interval does not include zero.

APC values were calculated using a statistical method called joinpoint analysis, with software developed by the Statistical Research and Applications Branch of the National Cancer Institute [1]. The joinpoint method evaluates changing trends (both the direction and the magnitude of the trend) over successive segments of time. A joinpoint is the point at which the linear segment changes significantly.

The analysis begins with the assumption of constant change over time (i.e. no joinpoint). Up to three joinpoints were tested in each model, depending on the number of years of data available and the stability of the yearly estimates. The selected trend line was the one with the fewest joinpoints which provided the best fit to the observed data, based on Monte Carlo permutation tests [1].

Note that trends were calculated with data for all years included, prior to removal of data which did not meet the confidentiality requirements.

Australian Standard Population (2001)


The standard population currently used for direct age-standardisation within Australia is the 2001 Australian resident population, which is released by the Australian Bureau of Statistics [2].

Conditional survival

Conditional survival is the probability of surviving an additional y years given the person has already survived x years [4]. It is calculated by dividing the relative survival at (x+y) years after diagnosis by the relative survval at x years after diagnosis, while confidence intervals were calculated using a variation of Greenwood's formula [5].


Hazard ratio

These are calculated for survival, and indicate the difference between the survival curves. The excess mortality of the group of interest (e.g. females) is divided by the excess mortality of the reference group (e.g. males). A hazard ratio significantly greater than 1 corresponds to increased death (and lower survival) compared to the reference group, while a hazard ratio significantly less than 1 indicates lower death (and higher survival). A ratio of 1.00 with no confidence interval indicates the reference group

Confidence intervals

All estimates are calculated with some degree of uncertainty. This uncertainty is typically reported in terms of a confidence interval, which specifies a range of values in which the true data point is expected to occur with a given level of certainty. For example, a 5-year survival rate may be estimated as 11.1% with a 95% confidence interval of 10.3%-12.0%. This means that there is a 95% probability that the true survival rate will be somewhere between 10.3% and 12.0%.

Incidence

The incidence of a particular disease is the number of new cases diagnosed in a specified population during a given time period (usually one year). Incidence is also commonly expressed as a rate (e.g. per 100,000 population). Since the risk of most cancers varies with age, it is common practice to age-standardise incidence rates to allow for more valid comparisons between populations (see Age-standardised rates).

Lifetime risk

Initially cumulative rates were calculated for ages 0-74 years, 0-79 years and 0-84 years. The cumulative rate is used as an estimate of the risk of developing or dying of cancer up to the age of 75, 80 or 85, respectively, and is expressed as a percentage.
For example, the cumulative rate for up to age 85 is calculated as follows: 


Cumulative rate

Cumulative risk is another measure of risk derived from the cumulative rate. It is more precise, and takes into account the removal of persons from the population of interest who have already been diagnosed with or died from cancer. Expressed as a ‘1 in n’ proportion, the cumulative risk is calculated as: 

Cumulative risk

QCSOL provides the cumulative risk up to the ages of 75, 80 and 85 years as an approximation of lifetime risk. The calculation assumes that the person experiences the current age-specific risk rates up to the age specified (e.g. 85), and is unable to factor in individual risk factors (such as smoking). 

Mortality

Mortality measures the number of deaths caused by a given condition within a specified population over a defined time period (usually one year). Similar to incidence, mortality can also be expressed as a rate (per 100,000 population), and these rates are often age-standardised to account for variation in the age structures of different populations (see Age-standardised rates).

Premature mortality

Premature mortality (measured by years of life lost, or YLL) is based on how much of their expected lifetime a person loses when they die. For example, a person who dies at 40 years of age would lose a greater number of years of (expected) life than a person who dies at age 70.

The calculation of premature mortality was based on the average YLL per death by age group and sex that were used in the 2003 Australian Burden of Disease and Injury study (using a 3% discount rate and no age weighting) [3]. This information was then applied to mortality data from the Queensland Cancer Registry to ascertain the total YLL per year and the average YLL per death by type of cancer.

Prevalence

Prevalence represents the number of people who had a diagnosis of cancer in the past and are still alive at a specified point in time. It is impacted by both the number of new cancers (incidence) and the length of time patients survive after being diagnosed. Even though two types of cancer might have similar incidence, if one cancer has low survival rates and another cancer has higher survival rates, then the prevalence of the second cancer will be greater.

This website presents “limited duration” prevalence, which counts cases who remain alive at a given time point (e.g. 31st December 2008) as prevalent when they were diagnosed within a specific time period. Limited duration prevalence estimates are presented for 1-, 5-, 10-, 15-, 20- and 25-year time periods. Note that persons diagnosed with cancer before 1982 (when the Queensland Cancer Registry began operating) are not included in any prevalence estimates. 

Survival

Survival time is defined as the length of time between when a person is diagnosed with a disease and when they die. However, since the eventual survival time of everyone diagnosed with cancer is not known (for example they may still be alive), statistical adjustments are required to take into account those unknown or censored survival times.

Relative survival was used to estimate the proportion of people who survived for different lengths of time. Relative survival compares the survival of people who have a particular disease or condition against the expected survival of a comparable group from the general population, taking into account age, sex and year of diagnosis. The method does not require knowledge of the specific cause of death, only knowledge of whether the patient has died. Relative survival is the most commonly presented measure of cancer survival when using data from population-based cancer registries [6]. Patients who were still alive at 31st December 2008 were considered censored.

Relative survival estimates can be calculated using either the period or cohort methods [7]. Relative survival estimates shown were produced using the period approach, which is recognised as providing more up-to-date survival estimates [8].

A suite of programs developed by Paul Dickman from the Karolinska Institutet in Sweden [9] were used to generate the relative survival estimates. These programs use a life table (or actuarial) method for calculating observed survival. This approach involves dividing the total period of observation into a series of discrete time intervals. The survival probabilities were then calculated for each of these intervals, and these were multiplied together to get the estimate for observed survival. Expected survival (based on total Queensland mortality data obtained from the Australian Bureau of Statistics) was calculated based on the Ederer II method [10]. Three-year averages for expected survival were used to minimise the effects of year to year variation. Relative survival was then obtained from the ratio of observed survival to expected survival.

Other

Cancer ICD-O3 codes used

All invasive cancers = C00 to C80 (excluding C44 (M805 to M811))
Anus & anal canal cancer = C21
Bladder cancer = C67
Bone cancer = C40 to C41
Brain cancer = C70 to C72
Breast cancer = C50
Cervical cancer = C53
Chronic myeloproliferative diseases = M995 to M996
Colon cancer = C18
Colorectal cancer = C18 to C20, C218
Connective tissue & peripheral nerves cancer = C47, C49
Endocrine glands cancer = C74 to C75
Eye cancer = C69
Floor of mouth cancer = C04
Gallbladder cancer = C23 to C24
Gum cancer = C03
Gynaecological cancers = C51 to C58
Head and neck cancers = C01 to C14, C30 to C32
Hodgkin lymphoma = M965 to M966
Kaposi sarcoma = M914
Kidney cancer = C64 to C66, C68
Larynx cancer = C32
Leukaemia = M980 to M994
Lip cancer = C00
Liver cancer = C22
Lung cancer = C33 to C34
Lymphoid leukaemia = M982 to M983
Lymphoma = M959 to M972
Melanoma = C44 (only M872 to M879)
Mesothelioma = M905
Myelodysplastic diseases = M998
Myeloid leukaemia = M984 to M993
Myeloma = M973
Nasal cavity cancer = C30 to C31
Nasopharynx cancer = C11
Non-Hodgkin lymphoma = M959, M967 to M972
Oesophageal cancer = C15
Other female genital organs cancer = C52, C55, C57 to C58
Other lip, oral cavity & pharynx cancer = C14
Other lymphatic cancers = M974 to M976
Other major salivary glands cancer = C07 to C08
Other parts of mouth cancer= C05 to C06
Other skin cancer = C44 (excluding M805 to M811, M872 to M879)
Other specified leukaemia = M994
Ovarian cancer = C56
Pancreatic cancer = C25
Penile cancer = C60, C63
Prostate cancer = C61
Pyriform sinus & hypopharynx cancer = C12 to C13
Rectosigmoid junction & rectal cancer = C19 to C20
Retroperitoneum & peritoneum cancer = C48
Small intestine cancer = C17
Stomach cancer = C16
Testicular cancer = C62
Thymus, heart, mediastinum & pleura cancer = C37 to C38
Thyroid cancer = C73
Tongue cancer = C01 to C02
Tonsil & oropharynx cancer = C09 to C10
Unknown primary site cancers = C26, C39, C76 to C77, C80
Unspecified leukaemia = M980
Uterine cancer = C54
Vulva cancer = C51

In 2007 an alternate method of classifying bladder cancers as invasive or in-situ was adopted. This new method is consistent with other Australian Registries, and is based on the layer of the bladder involved. This has resulted in a decreased number of invasive bladder cancers, and only affects data from 2007 onwards. However, the bladder cancer ICD-O3 codes have not changed.


 References

  1. National Cancer Institute, 2005. Joinpoint Regression Program, Version 3.0. Retrieved 16 July 2007, from http://srab.cancer.gov/ joinpoint/.
  2. Australian Bureau of Statistics, 2003. Population by age and sex - 2001 census edition. ABS Cat. No. 3201.0. ABS: Canberra. Retrieved 23 July 2007, from http://www.ausstats.abs.gov.au/ausstats.
  3. Begg S, Vos T, Barker B, et al., 2007. The burden of disease and injury in Australia 2003. (AIHW Cat. No. PHE 82). Australian Institute of Health and Welfare: Canberra. Retrieved 25 May 2007, from http://www.aihw.gov.au/publications/index.cfm/title/10317.
  4. Baade PD, Youlden DR, Chambers SK, 2011. When do I know I am cured? Using conditional estimates to provide better information about cancer survival prospects. Med J Aust, 194(2):73-77.
  5. Skuladottir H, Olsen JH, 2003. Conditional survival of patients with the four major histologic subgroups of lung cancer in Denmark. J Clin Oncol, 21(16):3035-3040.
  6. Dickman PW, Sloggett A, Hills M, et al., 2004. Regression models for relative survival. Statistics in Medicine, 23(1):51-64.
  7. Brenner H, 2002. Long-term survival rates of cancer patients achieved by the end of the 20th century: a period analysis. Lancet, 360(9340):1131-1135.
  8. Brenner H, Gefeller O, Hakulinen T, 2004. Period analysis for 'up-to-date' cancer survival data: theory, empirical evaluation, computation realisation and applications. European Journal of Cancer, 40:326-335.
  9. Dickman PW, 2004. Estimating and modelling relative survival using SAS. Retrieved 24 July 2007, from http://www.pauldickman.com/rsmodel/sas_colon/.
  10. Ederer F, Axtell LM, Cutler SJ, 1961. The relative survival rate: a statistical methodology. National Cancer Institute Monographs, 6:101-121.

Back to top