Send to

Choose Destination
Oncologist. 2013;18(1):64-72. doi: 10.1634/theoncologist.2012-0241. Epub 2013 Jan 3.

Feasibility of computer-based self-administered cancer-specific geriatric assessment in older patients with gastrointestinal malignancy.

Author information

Dana-Farber Cancer Institute, Department of Medical Oncology, Gastrointestinal Oncology, 450 Brookline Avenue, Boston, MA 02215, USA.



The Cancer-Specific Geriatric Assessment (CSGA) is a primarily self-administered paper survey of validated measures.


We developed and tested the feasibility of a computer-based CSGA in patients ≥70 years of age who were receiving treatment for gastrointestinal malignancies at the Dana-Farber Cancer Institute. From December 2009 to June 2011, patients were invited to complete the CSGA at baseline (start of new treatment) and follow-up (at the first of 4 months later or within 4 weeks of completing treatment). Feasibility endpoints were proportion of eligible patients consented, proportion completing CSGA at baseline and follow-up, time to complete CSGA, and proportion of physicians reporting CSGA results that led to a change in clinical decision-making.


Of the 49 eligible patients, 38 consented (76% were treatment naive). Median age was 77 years (range: 70-89 years), and 48% were diagnosed with colorectal cancer. Mean physician-rated Karnofsky Performance Status was 87.5 at baseline (SD 8.4) and 83.5 at follow-up (SD 8). At baseline, 92% used a touchscreen computer; 97% completed the CSGA (51% independently). At follow-up, all patients used a touchscreen computer; 71% completed the CSGA (41% independently). Mean time to completion was 23 minutes at baseline (SD 8.4) and 20 minutes at follow-up (SD 5.1). The CSGA added information to clinical assessment for 75% at baseline (n = 27) and 65% at follow-up (n = 17), but it did not alter immediate clinical decision-making.


The computer-based CSGA feasibility endpoints were met, although approximately half of patients required assistance. The CSGA added information to clinical assessment but did not affect clinical decision-making, possibly due to limited alternate treatment options in this subset of patients.

[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for HighWire Icon for PubMed Central
Loading ...
Support Center