RATIONALE, AIMS AND OBJECTIVES: The improvement of survival with time for cancers has been observed worldwide. However, whether and to what extent such an improvement is attributed to early detection or the advance in medical technology is barely addressed. METHOD: We developed a novel graphic method to calculate attributable proportions related to two factors. This graphic method was first to calculate adjusted survival curve by the application of direct standardized epidemiological method to tumour staging. The crude survival curve was then compared with the adjusted survival curve. RESULTS: Cumulative survival curves of two time epochs, 1989-1993 (period 1) and 1994-1998 (period 2), on six common cancers collected from cancer registry of National Taiwan University between 1989 and 1998, were compared. The attributable proportions due to early detection and medical care were 77% and 23% for breast cancer, 50% and 50% for cervical cancer, 52% and 48% for colorectal cancer, 76% and 24% for gastric cancer, 66% and 34% for liver cancer, and 30% and 70% for prostate cancer respectively. CONCLUSION: Distinguishing between early detection from advance in medical care associated with the improvement of survival trend using our graphic method has a significant implication for secondary and tertiary prevention of common cancers.
RATIONALE, AIMS AND OBJECTIVES: The improvement of survival with time for cancers has been observed worldwide. However, whether and to what extent such an improvement is attributed to early detection or the advance in medical technology is barely addressed. METHOD: We developed a novel graphic method to calculate attributable proportions related to two factors. This graphic method was first to calculate adjusted survival curve by the application of direct standardized epidemiological method to tumour staging. The crude survival curve was then compared with the adjusted survival curve. RESULTS: Cumulative survival curves of two time epochs, 1989-1993 (period 1) and 1994-1998 (period 2), on six common cancers collected from cancer registry of National Taiwan University between 1989 and 1998, were compared. The attributable proportions due to early detection and medical care were 77% and 23% for breast cancer, 50% and 50% for cervical cancer, 52% and 48% for colorectal cancer, 76% and 24% for gastric cancer, 66% and 34% for liver cancer, and 30% and 70% for prostate cancer respectively. CONCLUSION: Distinguishing between early detection from advance in medical care associated with the improvement of survival trend using our graphic method has a significant implication for secondary and tertiary prevention of common cancers.