Send to:

Choose Destination
See comment in PubMed Commons below
Eur J Surg Oncol. 2011 Jul;37(7):571-5. doi: 10.1016/j.ejso.2011.02.009. Epub 2011 Mar 27.

Clinical prediction of survival by surgeons for patients with incurable abdominal malignancy.

Author information

  • 1Department of Surgical Gastroenterology, Oslo University Hospital, Aker, 0514 Oslo, Norway.



Accurate prognosis facilitates decision-making and counselling in incurable cancer. However, predictions of survival are frequently inaccurate and survival is consistently overestimated. The prognostic skills of surgeons are sparsely documented, and the present study was undertaken to assess their prognostic accuracy for patients with advanced abdominal malignancy.


Clinical predictions of survival were made by three consultant surgeons independently in consecutive patients with incurable abdominal cancer. Survival was predicted in intervals ranging from <1 week to 18-24 months. Prognoses were considered accurate when actual survival fell within the expected range. Performance status was classified according to the Eastern Cooperative Oncology Group (ECOG).


243 assessments were made in 178 patients. Prognoses were accurate in 27%, over-optimistic in 42% and over-pessimistic in 31%. Accuracy was inversely related to length of actual survival and did not differ between surgeons (P = 0.466). The proportion of over-optimistic prognoses differed significantly between surgeons (P < 0.001). Prognostic accuracy was 44% in gastric cancer patients, 29% in pancreatic cancer patients and 22% in colorectal cancer patients (P = 0.052). ECOG performance status correlated well with survival.


Surgeons' accuracy in determining prognosis is poor. There are considerable individual differences between surgeons, and accuracy is reduced in cases with prolonged life expectancy.

Copyright © 2011 Elsevier Ltd. All rights reserved.

[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Write to the Help Desk