Wednesday December 22 2010
Overall, survival rates improved in all six countries
“Cancer survival rates in Britain still lag behind those in comparable Western countries,” reported The Daily Telegraph. The news story is based on a study of more than 2 million people, comparing survival rates in Australia, Canada, Denmark, Norway, Sweden and the UK. Rates were compared for cancers of the bowel, breast, lung and ovary that were diagnosed between 1995 and 2007.
Overall, survival rates improved in all six countries. However, survival was found to be “persistently lower” in Denmark, England, Northern Ireland and Wales. This was particularly the case after the first year of diagnosis and for patients aged 65 and over.
Importantly, cancer survival in the UK has improved significantly. Breast cancer survival at five years, for example, increased from 74.8% in 1995-9, to 81.6% in 2005-7. This 6.8% increase is higher than that which took place in the five other countries during the same period. Despite this, the UK still has the lowest breast cancer survival rate at five years.
This is a large government-funded study and its findings are likely to be reliable. The authors suggest that the lower survival rates in the UK may be partly due to later diagnosis as well as differences in treatment and in access to healthcare. Further research that examines the effect of lifestyle choices, early diagnosis and treatment differences between countries will help to inform national cancer strategies.
Steps to improve survival rates have already been taken. Professor Sir Mike Richards, the government's National Clinical Director for Cancer said: "In England we have already started work on improving early diagnosis, including a new campaign starting next month to alert people to the early signs and symptoms of bowel, lung and breast cancer and plans to give GPs more direct access to key diagnostic tests."
Where did the story come from?
The study was carried out by researchers from a number of institutions in the six countries that were the focus of the study. It was funded by Cancer Research UK and the Department of Health, England. The study was published in the peer-reviewed medical journal The Lancet.
The research was reported accurately in the newspapers and online.
What kind of research was this?
This was an analysis of data on the survival rates of 2.4 million adults with cancer from Australia, Canada, Sweden, Denmark, Norway and the UK. Separate data were provided for England, Northern Ireland and Wales.
The researchers point out that cancer survival is a key measure of the effectiveness of healthcare systems and that persistent regional and international differences in survival represent many avoidable deaths. This is the first study in a new programme to investigate differences in cancer survival rates, with the aim of informing health policy.
What did the research involve?
The researchers selected four cancers on which to base their study: breast (women only), bowel (colorectal cancer), lung and ovary. The study data were obtained from the cancer registries in 12 jurisdictions in the six countries described above. The researchers looked at anonymous, individual cancer registration records for adults aged 15-99 years diagnosed with primary, invasive cancers during the 13 years from 1995 to 2007. Details included date of diagnosis, anatomical site of the cancer, behaviour of the tumour, date of birth, sex and last known status of the patient, as well as the cancer stage at diagnosis and treatment.
Survival rates were estimated using national incidence and mortality rates for 1998-2002. The researchers used relative survival rates, which is a standard approach for looking at population-based cancer survival. Relative survival rate describes the ratio of the survival observed in patients with cancer, compared with the survival that would have been expected in patients without cancer and therefore takes into account other causes of death.
The researchers carried out their analyses and data control using the same methods across all countries and the project was overseen by external experts. The analysis was of age-specific and age-standardised relative survival, at one and five years after diagnosis. The researchers also looked at five-year survival in people who had already survived one year, this allowed them to make international comparisons while minimising the effect of factors, such as very late diagnosis, that mainly affect survival in the first year after diagnosis.
What were the basic results?
The researchers found that 2.4 million adults were eligible for the analysis. Between 1995 and 2007, survival rates improved for all four cancers in all six countries. During this time:
- Survival was “persistently higher” in Australia, Canada and Sweden; intermediate in Norway; and lower in Denmark, England, Northern Ireland and Wales, particularly in the first year after diagnosis and for patients aged 65 years and older.
- For women with breast cancer, the differences in survival rates became smaller (the lines on the graph narrowed and came closer together), from 14% difference to 8% difference in five year survival.
- For lung and ovary cancers, the differences in survival rates did not lessen.
- For colorectal cancer, the differences between countries lessened only for patients aged 65 years and older.
Survival improved in the UK between 1995 and 2007 for all cancers, but not as much as in most of the other countries analysed.
- For bowel cancer, UK survival at five years rose from 47.8% to 53.6%. This compares with a rise from 61.2% to 66.4% in the registry with the best 1995/99 rate: New South Wales.
- For lung cancer, UK survival rates at five years rose from 7.0% to 8.8%. This compares with a rise from 15.7% to 18.4% in the registry with the best 1995/99 rate: Canada.
- For breast cancer, UK survival rates at five years rose from 74.8% to 81.6%. This compares with a rise from 86.7% to 88.5% in the registry with the best 1995/99 rate: Sweden.
- For ovarian cancer, UK survival rates at five years rose from 32.6% to 36.4%. This compares with a rise from 37.2% to 39.7% in the registry with the best 1995/99 rate: Norway.
How did the researchers interpret the results?
The researchers say their analysis shows increases in survival rates for these four cancers, but also persistent differences between countries. Overall, the largest gains in survival were recorded for bowel cancer and the smallest for lung and ovarian cancer.
They say that the quality of cancer registrations in the different countries is high and that poor quality data is unlikely to be the explanation for the lower survival rates found in the UK. They suggest that the differences could be due to treatment delays and later diagnosis, especially in older and less affluent groups. They say that large variations in diagnostic and surgical practice might also contribute, in particular in breast cancer treatment, especially for women aged 65 and older.
This is an important study and its findings show that while the UK has made good progress in some areas such as breast cancer survival, it still lags behind some other developed countries in some areas. The researchers say that monitoring survival trends will help in the future formulation of strategies for cancer control.
The researchers note several features of interest in assessing these types of studies.
- Mortality rates depend on both incidence (the number of new cases in a year) and survival over a number of years. Mortality rates therefore refer to the number of people who die in a specific year. This means that for cancers such as lung cancer, which typically has low five-year survival, the mortality trends largely follow the incidence trends. Addressing the causes of cancer, mostly smoking, is a public health priority for these.
- In cancers that have good five-year survival rates, such as breast cancer, trends in mortality provide a delayed and imprecise picture of any trends in survival. These mortality data are subject to errors in certification of the cause of death, particularly in elderly patients.
- The researchers say that data quality issues were not a concern in their study and that the mortality data were almost complete.
- The prevalence of obesity, physical activity, smoking and other lifestyle risk factors was not included in the analysis. Future studies would benefit from taking these into account.
It is not yet completely clear why the UK has lower survival rates than these other countries. Further analyses would need to look at whether cancers are diagnosed at a later stage in the UK, or if treatment differs. These sorts of analysis will be needed to provide evidence for how best to reduce the differences.
Professor Sir Mike Richards, the government's National Clinical Director for Cancer said:
"In England we have already started work on improving early diagnosis, including a new campaign starting next month to alert people to the early signs and symptoms of bowel, lung and breast cancer and plans to give GPs more direct access to key diagnostic tests."