Level of Palliative Care Complexity in Advanced Cancer Patients: A Multinomial Logistic Analysis.

The current treatment approach for patients in palliative care (PC) requires a health model based on shared and individualised care, according to the degree of complexity encountered. The aims of this study were to describe the levels of complexity that may be present, to determine their most prevalent elements and to identify factors that may be related to palliative complexity in advanced-stage cancer patients. An observational retrospective study was performed of patients attended to at the Cudeca Hospice. Socio-demographic and clinical data were compiled, together with information on the patients' functional and performance status (according to the Palliative Performance Scale (PPS)). The level of complexity was determined by the Diagnostic Instrument of Complexity in Palliative Care (IDC-Pal©) and classified as highly complex, complex or non-complex. The impact of the independent variables on PC complexity was assessed by multinomial logistic regression analysis. Of the 501 patients studied, 44.8% presented a situation classed as highly complex and another 44% were considered complex. The highly complex items most frequently observed were the absence or insufficiency of family support and/or caregivers (24.3%) and the presence of difficult-to-control symptoms (17.3%). The complex item most frequently observed was an abrupt change in the level of functional autonomy (47.6%). The main factor related to the presence of high vs. non-complexity was that of performance status (odds ratio (OR) = 10.68, 95% confidence interval (CI) = 2.81-40.52, for PPS values < 40%). However, age was inversely related to high complexity. This study confirms the high level of complexity present in patients referred to a PC centre. Determining the factors related to this complexity could help physicians identify situations calling for timely referral for specialised PC, such as a low PPS score.