Aim and methods: The PaP score combines symptoms, performance status (PS), and lymphocyte counts with the physician's survival estimate (PSE) to reliably predict the short term survival of patients with advanced cancer referred to palliative care services. Its usefulness in other patient populations has not yet been determined. The study aim was to test the predicitive power of the PaP score in the hands of Medical and Radiation Oncologists when applied to 100 consecutive hospitalised patients with advanced cancer prior to palliative care referral (evaluation 180 days post-inclusion).
Results: 66 were still receiving systemic anti-cancer treatment (chemo- and/or radiotherapy). 62 reported anorexia, 35 dyspnea. Karnofsky PS score was>50 in 65. The PSE was>3 months in 55, B/1 month in 9. Severe lymphopenia (B/12% total white cell count) occurred in 60. By April 15 2003, all had reached the end of the study; the PaP score distinguished three distinct groups (log rank test for trend 50.90, d.f.=1, PB/0.0001) with estimated median survival 121 (n=64), 46 (n=32) and 2.5 (n=4) days respectively. These oncologists tended to overestimate the patients' survival which may lead to possibly inadequate diagnostic and therapeutic decisions.
Conclusions: PaP appears to be helpful for oncologists to group heterogenous patient samples into isoprognostic groups.