How many deaths would be avoidable if socioeconomic inequalities in cancer survival in England were eliminated? A national population-based study, 1996–2006

https://doi.org/10.1016/j.ejca.2011.10.008Get rights and content

Abstract

Aim

Inequalities in survival between rich and poor have been reported for most adult cancers in England. This study aims to quantify the public health impact of these inequalities by estimating the number of cancer-related deaths that would be avoidable if all patients were to have the same cancer survival as the most affluent patients.

Methods

National Cancer Registry data for all adults diagnosed with one of 21 common cancers in England were used to estimate relative survival. We estimated the number of excess (cancer-related) deaths that would be avoidable within three years after diagnosis if relative survival for patients in all deprivation groups was as high as the most affluent group.

Results

For patients diagnosed during 2004–2006, 7122 of the 64,940 excess deaths a year (11%) would have been avoidable if three-year survival for all patients had been as high as in the most affluent group. The annual number of avoidable deaths fell from 8435 (13%) a year for patients diagnosed during 1996–2000. Over 60% of the total number of avoidable deaths occurred within six months after diagnosis and approximately 70% occurred in the two most deprived groups.

Conclusion

The downward trend in the annual number of avoidable deaths reflects more an improvement in survival in England overall, rather than a narrowing of the deficit in cancer survival between poor and rich. The lack of any substantial change in the percentage of avoidable excess deaths highlights the persistent nature of the deficit in survival between affluent and deprived groups.

Introduction

Inequalities in survival between rich and poor have been reported for most adult cancers in England and Wales.1, 2 The origin of these disparities in survival is still not fully understood, but factors such as stage at diagnosis and access to optimal treatment have been implicated.3 Such observations suggest that deprived patients do not benefit equally from health-care services in the United Kingdom (UK), despite a universal health-care system that is free to all at the point of use. Quantifying the public health impact of these inequalities in cancer survival is important to inform health policy. One such approach is to consider the number of deaths that would be avoidable if all patients were to have the same survival from their cancer as that observed for the most affluent patients.

The NHS (National Health Service) Cancer Plan for England, published in late 2000, was designed to improve prevention, early diagnosis and screening, and to provide optimal treatment for all patients. One of the main aims of the Cancer Plan was to tackle inequalities in cancer survival for people from deprived or less affluent backgrounds.4 Recent observations suggest there has been a modest acceleration of the previous upward trend in survival in England since implementation of the NHS Cancer Plan.5 However, there is little evidence that the Cancer Plan has been effective in reducing socioeconomic inequalities in short-term survival in the period up to 2006.2 Inequalities in short-term survival between rich and poor were still large for many cancers among patients diagnosed in 2006.

We set out to update the public health evaluation of socioeconomic inequalities in survival by estimating how many cancer deaths would have been avoidable within three years of diagnosis if relative survival for all patients had been as high as for the most affluent patients. We examined National Cancer Registry data for England in three calendar periods, defined in relation to the NHS Cancer Plan: 1996–2000 (five years; before the Cancer Plan), 2001–2003 (three years; initialisation) and 2004–2006 (three years; implementation). Trends in the annual number of avoidable deaths can be used as a public health measure of progress towards the goals set out in the NHS Cancer Plan.

Section snippets

Relative survival, excess mortality and avoidable deaths

The overall mortality in a group of cancer patients can be divided into two components: the background mortality (or expected mortality, derived from all-cause death rates in the general population), and the excess mortality, attributable to the cancer. Excess (cancer-related) mortality is estimated using the relative survival approach.6, 7 Avoidable deaths are the component of excess mortality that would not occur if relative survival in all deprivation categories was as high as in affluent

Results

For patients diagnosed with one of 21 common cancers in England during 2004–2006, a total of 7122 of the 64,940 excess (cancer-related) deaths a year would have been avoidable within three years since diagnosis if survival for all patients had been as high as the most affluent group. This represents a fall in the number of avoidable deaths within three years since diagnosis from 8435 per year among patients diagnosed during 1996–2000. The percentage of excess deaths that was avoidable fell from

Discussion

The absolute number of avoidable deaths for a particular cancer depends on the deficit in relative survival between affluent and deprived groups (the ‘deprivation gap’), but also on the number of patients diagnosed with that cancer and on the relative survival for that cancer. Our findings show that for adult cancer patients diagnosed in England during 2004–2006, 7122 (11%) of the 64,940 cancer-related deaths that occurred each year within three years since diagnosis would have been avoidable

Role of funding source

This work was supported by the Office for National Statistics (ONS) [NT-04/2355A]; and Cancer Research UK [C1336/A5735]. ONS collated the data from the regional registries and arranged their linkage to data on deaths at the National Health Service Central Register. Neither Cancer Research UK nor ONS had any role in study design, analysis, interpretation of the data, writing of the report, or in the decision to submit the paper for publication. The findings and conclusions in this report are

Author contributions

B.R. and M.P.C. led the study design. L.E., B.R. and M.P.C. carried out the data preparation and quality control. L.E. did the analyses. L.E., B.R. and M.P.C. contributed to interpretation of the findings and drafted the report.

Conflict of interest statement

None declared.

Acknowledgments

We thank the Cancer Registry staff in England: their sustained data collection and quality control have enabled the survival of patients to be analysed and compared in this study. We also thank the Cancer Team at ONS for extensive work in preparing the data in the National Cancer Registry.

References (28)

  • Office of the Deputy Prime Minister

    The English indices of deprivation 2004

    (2004)
  • Information Services Division

    Cancer mortality in Scotland (2009)

    (2010)
  • A. Pokhrel et al.

    Education, survival and avoidable deaths in cancer patients in Finland

    Br J Cancer

    (2010)
  • K.M. Gorey et al.

    An international comparison of cancer survival: Toronto, Ontario and Detroit, Michigan, metropolitan areas

    Am J Public Health

    (1997)
  • Cited by (72)

    • Referral Patterns and Outcome of Patients With Synchronous Brain Metastases From Non-small Cell Lung Cancer Treated With Gamma Knife Radiosurgery in a Third-Line Treatment Centre in The Netherlands – A Retrospective Analysis

      2020, Clinical Oncology
      Citation Excerpt :

      Patients in both groups had a similar number of comorbidities, which might serve, at least partly, as a proxy for overall health. Socioeconomic status was also compared, as a lower socioeconomic status and education level are correlated with a poorer prognosis in cancer patients in general [25–27]. A recent study found that the cultural backgrounds of patients with multiple brain metastases from NSCLC, specifically, had a significant influence on the patients' survival prognosis [28].

    • Loss in life expectancy and gain in life years as measures of cancer impact

      2019, Cancer Epidemiology
      Citation Excerpt :

      We will also consider alternative formats of the metric; both the proportion of life lost and also conditional measures to calculate updated impact on a patient’s life expectancy. A number of investigators have tried to quantify the impact of differences between population groups by calculating the deaths that could be avoided at given timepoints post diagnosis if inequalities in relative survival were removed [8–12]. One alternative would be to calculate the life expectancy for cancer patients should the differences in relative survival be removed [12,13].

    View all citing articles on Scopus
    View full text