Summary Time trends in therapeutic approaches and in the prognosis of colon cancer for patients aged 75 years and above have been investigated in comparison with corresponding trends for younger patients using a population-based series of 2089 colon cancer patients diagnosed between 1976 and 1990 in the C6te-d'Or area (478 000 inhabitants), Burgundy, France. Significant progress has been achieved in the management of patients with colon cancer in both age groups, but trends have been more noticeable in patients aged 75 years and above. In the elderly, the proportion of cancers limited to the digestive tract wall showed a 3-year average increase of 2.8% (P = 0.02) and the frequency of curative surgery an average increase of 8.6% (P < 0.001), so that it was performed in 80% of cases in the last 3-year period. Operative mortality decreased by 2.5% between 3-year periods (P < 0.004). Crude 5-year survival rates in elderly patients increased from 15% in the 1976-78 period to 29% in the 1985-87 period (P < 0.001), the corresponding figures being 36% and 44% (P > 0.10) in younger patients.Keywords: colon cancer; stage at diagnosis; time trends; survival Colon cancer is a major problem in elderly patients. Incidence rates rise with age, and over 40% of cases occur in subjects over the age of 74. Recent studies have demonstrated an increase in the incidence of colon cancer in several areas of the Western world (Coleman et al, 1993) and, given the increasing life expectancy of Western populations, an ever-growing number of aged people is exposed to the risk of colon cancer. Age has often been considered as a negative factor in the prognosis for this cancer. However, raw survival data, from which such conclusions are usually drawn, overestimate mortality due to the malignancy under investigation, especially in elderly patients, for whom mortality owing to other causes is high, and tend to conceal the progress that has been achieved in the perioperative management of elderly patients over the past 15 years (Pillon et al, 1991). Data on therapeutic approaches and on the prognosis of colon cancer patients have mostly been provided by specialized hospital units, with unavoidable selection bias, especially for elderly patients.
Outcome measuresThe spread of each malignancy at the time of diagnosis was classified, for resected cancers, according to Dukes (1932), as: limited to the digestive wall (Dukes A) (I, see tables); extension beyond the digestive wall (Dukes B) (II); and lymph node involvement (Dukes C) (Ill). In the absence of resection, cancers were classified as either metastatic (IV) or of undetermined stage (i.e. absence of detectable metastasis) (V). Treatment procedures were defined as: surgery for cure, i.e. complete tumour removal with tumour-free margins (1); palliative resection (II); palliative surgery with no tumour resection (i.e. colostomy or explorative laparotomy) (III); and medical treatment without surgery, i.e. chemotherapy, radiotherapy or purely palliative treatment (IV). As the impact of age and oth...