Effect of improved data collection on breast cancer incidence and survival: reconciliation of a registry with a clinical database
Variations in the way breast cancer is registered in hospitals is leading to missed cases and, as a result, may be distorting national cancer statistics - currently a principal means of identifying cancer "black spots" around the county, according to a study in this week's BMJ.
Researchers in Leicestershire compared breast cancer records for 1997 at Trent Cancer Registry with a local hospital database. They found that the recorded date of diagnosis was, on average, 26 days earlier on the hospital database than the date of registration, and 70 patients were assigned to a previous year. By the end of 1998, 134 patients listed on the hospital database were still not registered. Therefore, although only 535 breast cancers were recorded for 1997 at the cancer registry, 599 patients were finally identified - a 12% increase.
Adding the missing cases to the register increased the level of cancer and reduced the overall survival in the Trent region because the missing cases tended to be older with a lower life expectancy, explain the authors. These findings, they say, highlight the important influence of data collection on national cancer statistics, and stress that those who use them must bear this in mind, particularly when attempting to compare cancer incidence and survival in different centres and even countries.
Contact:
Anne Stotter, Consultant Surgeon, Glenfield Hospital NHS Trust, Leicester LE3 9QP Email: anne.stotter@glenfield-tr.trent.nhs.uk