Friday September 26 2008
The researchers used income to determine socioeconomic status
“Middle classes ‘more likely to develop breast and skin cancer’”, is the headline in The Daily Telegraph. A study has shown that this socioeconomic group is significantly more likely to develop breast and skin cancer than the less well off. The newspaper suggests that “career women delaying having children and exposing themselves more to the sun on foreign holidays are thought to be behind the gap”. It also says social deprivation is linked to lung and cervical cancer, “because people from poorer classes are more likely to smoke and to skip smear tests”.
The story is based on a study that compares socioeconomic-specific incidence rates of breast, skin, lung and cervical cancer diagnosed between 1998 and 2003 in England. The results do show variations in the incidence of some cancers but cannot indicate any reasons for the differences. However, the results highlight an important aspect of public health – health inequality. There are differences or 'gaps' in mortality and survival that the researchers say have been looked at in other studies that relate them to variations in access to treatment. Both inequalities (i.e. differences in health need) and inequity (differences in the care provided) should be considered in the design and planning of public health interventions to reduce regional variations in healthcare and, ultimately, cancer burden.
Where did the story come from?
Dr Lorraine Shack and colleagues from Christie Hospital NHS Trust in Manchester, the London School of Hygiene and Tropical Medicine, Kings College London, the Trent Cancer Registry and Cancer Research UK carried out this study. The study was supported by the UK Association of Cancer Registries. It was published in the peer-reviewed medical journal: BMC Cancer.
What kind of scientific study was this?
In this cross-sectional study, researchers obtained information from all cancer registries in the UK for people diagnosed between 1998 and 2003 with invasive breast cancer, lung cancer, cervical cancer and malignant melanoma of the skin.
Socioeconomic status was assigned to patient based on their postcode at the time of diagnosis using a shortened version of the index of multiple deprivation (IMD) system. The IMD is a national measure of deprivation that assigns a ‘deprivation score’ to small areas across the country based on information collected during the census and from other government databases (income support, jobseekers allowance etc.). The score is determined using seven domains: income, employment, health deprivation and disability, education skills and training, barriers to housing and services, crime and living environment.
In this study, the researchers used only the ‘income’ domain to work out the levels of deprivation. They did this because they wanted to exclude health-related domains and say that there is good correlation between income and deprivation. Income was divided into quintiles; five equal groups of national earnings, each containing 20% of the population of England. Quintile one represented the 20% of England who were least deprived (i.e. highest earners) with quintile five representing the most deprived (i.e. lowest earners).
The researchers then compared the incidence of these different cancers in regions and age groups across the socioeconomic classes.
What were the results of the study?
The most deprived groups in England had the highest incidence rates of lung cancer and cervical cancer. The opposite was true for skin cancer and breast cancer.
Men who were classified as ‘most deprived’ were at a 2.5 times greater risk for lung cancer than those who were least deprived. Most deprived women were at a 2.7 times greater risk for lung cancer. There was also double the risk for cervical cancer in women who were most deprived compared with those who were least deprived.
This trend was reversed for breast cancer and skin cancer. Women in the least deprived groups were at greater risk for breast cancer (0.15 times more risk) and both men and women for in the least deprived groups had a greater risk of skin cancer (0.5 times greater risk).
The researchers also found ‘modest differences’ in socioeconomic-specific breast cancer incidence rates between and within regions, and substantial regional variation with cervical cancer, lung cancer and skin cancer. Across the four cancer types, the deprivation gap did not differ by age (between those under 65 years old and those over 65 years old) for breast, cervical or skin cancer. For lung cancer there was a difference between the level of risk depending on age. There was a much higher risk difference between the most and least deprived group in those aged under 65 than in those over 65.
What interpretations did the researchers draw from these results?
The researchers conclude that a reduction in the socioeconomic variations in incidence of cancers could have a substantial impact on the burden of cancer. They note that the regional differences between the socioeconomic-specific incidence of cervical, lung and skin cancer highlights variations in exposure to known risk factors. They say targeted public health interventions could help to reduce regional inequalities in incidence and reduce the future cancer burden.
What does the NHS Knowledge Service make of this study?
This large cross-sectional study has compared how incidence rates of cancer (lung, skin, cervical and breast) differ across socioeconomic groups and whether these differences are consistent across regions in England and age groups (under or over 65 years). There were about 450,000 cancer cases available for analysis in the dataset. The study confirms that there is an association between socioeconomic status and incidence of these cancers. It also finds regional variations in this ‘deprivation gap’. The researchers put forward some suggestions to explain these differences, including variations in recreational sun exposure and lifestyle factors (number of children) between socioeconomic groups.
Some of the problems associated with the data are highlighted by the researchers:
- Relying on areas of residence to determine socioeconomic status (as in the IMD system) has its shortcomings not everyone in a small area may be that alike.
- They also acknowledge that there are likely to be regional variations in the way that the data are collected for entry into the cancer registries.
There is a complex interaction between a person’s genetics, their risk factors, the environment and the care provided by health services in determining these various ’unfairness gaps’. Patterns of inequality shown in the incidence of disease, as in this study, may be mirrored by differences in medical care (inequity) or survival, but not always. Common factors such as access to screening may affect disease incidence, healthcare services or disease outcome, and some of these have been discussed by these researchers. In general, screening is thought to temporarily increase the rates of detection of cancer and to improve survival and so uptake rates are critical to an understanding of inequity.
Although deprived women are believed to have lower uptake levels of breast screening, in this study there was little variation in uptake across socioeconomic groups. The researchers suggest that this may reflect a high awareness among all groups. For cervical cancer, there were uptake differences across socioeconomic groups, which may explain the variations in deprivation gap.
This study and the researchers’ discussion of the results highlights an important area of public health, that of health inequality and health inequity. The findings may be used to target regions for public health programmes that will try to reduce inequality and inequity and, ultimately, the burden of these cancers.