Objective Reducing geographic inequalities in survival from cancer in England was a key aim of the Calman-Hine Report (1995) and the NHS Cancer Plan (2000). In this paper we assess whether regional differences have diminished following these policy developments by analysing the trend in one-year relative survival from six cancers in the 28 Cancer Networks of England.

Methods We estimated population-based relative survival at one year for 1.4 million patients who were diagnosed with cancer of the breast (women), cervix, stomach, oesophagus, lung or colon in England during 1991-2006 and followed up to 31 December 2007. Relative survival is the ratio of the observed survival of cancer patients relative to the expected mortality in the general population (background mortality) and it can be interpreted as the survival of cancer patients after other causes of death have been taken into account. Background mortality was estimated by age, sex, calendar year, deprivation category and Government Office Region. Age-standardised relative survival was estimated by Cancer Network in three calendar periods: 1991–1995, 1996–2000 and 2001–2006. Funnel plots were used to display spatial and temporal variation in survival. The number of Cancer Network, sex and age combinations that were outside of the 99.8% control limits of the England-wide estimate of relative survival was charted over time.

Results One-year relative survival improved over time for all patients except those diagnosed with cervical cancer. There were large regional differences in relative survival for each of the six cancers. Cancer Networks that were low-survival outliers across several cancers were clustered across Northern England and the Midlands. The north-south divide became less marked over time although the overall number of lower outliers compared to the national value remained stable.

Conclusion Policy changes over the past two decades coincided with improved relative survival, without seeing an increase in regional disparity. The north-south divide in the distribution of low-survival Cancer Networks became less pronounced over time but regional disparities persist. Further methodological development is needed to obtain more robust estimates of age-standardised relative survival for small populations, in order to monitor these regional trends.