Prognostic Factors for Survival in Advanced Cancer Patients in Iceland

• Main Author: Sigríður Helgadóttir 1979-
• Other Authors: Háskóli Íslands
• Format: Thesis
• Language: English
• Published: 2012
• Subjects: Líf- og læknavísindi; Lífsgæði; Lífslíkur; Krabbameinssjúklingar; Meier; Iceland
• Online Access: http://hdl.handle.net/1946/13200

Over 150 different factors have been evaluated with regard to prognostication in cancer in various studies. The most common are performance status, cognitive function, quality of life (QoL), physical symptoms and signs, concomitant diseases and biochemical related factors. The aim of this thesis is to investigate the prognostic value of cancer characteristics, prevalent physical symptoms (e.g. pain, fatigue and dyspnoea), opioid treatment, mobility, falls, nutrition, cognitive function and QoL. The objective was also to compare the two datasets used in relation to prognostication and to evaluate survival prediction by health care professionals. A comparable study has not been conducted among Icelandic cancer patients before. Materials and methods: Data consisted of material from two prospective cohort studies, including 266 patients. Firstly, data from Icelandic participants in the European Pharmacogenetic Opioid Study (EPOS) were used (N=150). These patients, who all received opioid treatment due to pain, had cancer and were either on palliative care and/or life-prolonging cancer therapy. Secondly, assessments of patients in palliative care service evaluated with the interRAI Palliative Care (PC) assessment tool, were included (N=116). Statistical analysis was performed with Kaplan Meier survival and Cox regression analyses. Results: Median survival of the patients in the EPOS part of the study was 101 day. In a multivariate Cox regression, variables that significantly predicted shortened survival were greater number of metastases (HR 1.24) and presence of fatigue (HR 1.82). Higher score on Mini Mental State Examination (MMSE) (HR 0.86), the role function scale (HR 0.99) of the EORTC questionnaire and the Brief Pain Inventory (BPI) interference scale (HR 0.91) was significantly related to longer survival. In the study using the interRAI PC assessment tool the median survival was 41 day. The variables significantly associated with shorter survival in a multivariate Cox regression were increased age (HR 1.02), ...