Comparison of different proxy approaches to determine the need for specialized palliative care in patients with...

Patients suffering from cancer can benefit from a timely integration of palliative and end-of-life care. In the literature different approaches are discussed that can be used by health care professionals (as proxies) to determine cancer patients in need for specialist palliative care. Until now data on comparing different tools is scarce. This study compared published methods for detecting patients with advanced and incurable cancer in need for specialist palliative care.

Data of three hundred and sixteen patients with incurable cancer-collected during a study validating the German version of a screening tool based on NCCN guidelines (Glare) - were used for secondary analysis. The data were used to test the performance of different tools in detecting patients with palliative care needs: two disease-specific classifications (Gaertner, Benthien), the Eastern Cooperative Oncology Group Performance Status (ECOG), the Surprise Question, as well as a combination of the Surprise Question and the German NCCN tool and the Surprise Question and the ECOG score. To quantify which tool performed best, survival, Integrated Palliative Outcome Scale (IPOS - staff version) (one or more items ≥ 3), and the information of a preexistent contact to palliative care served as indicators of real SPC needs in this patient group.

The combination of Surprise Question and the German NCCN Screening tool showed a sensitivity between 71.5%-94.3% and specificity between 56.0%-91.3%, while the combination of Surprise Question and ECOG score had a sensitivity between 37.4%-75.7% and specificity between 86.2%-100%. Benthien's classification performed a fair sensitivity (74.8%-91.5%) and a weak specificity (27.3%-39.4%), whereas the guidelines by Gaertner showed high sensitivity (92.2%-100%), but very low specificity in all standards (0.0%-9.9%).

While the combination of the Surprise Question and the German NCCN screening tool showed the best results in terms of sensitivity and specificity overall, a combination of the Surprise Question and ECOG score proved to be highly specific and as time-efficient in identifying patients in need of SPC, which may be beneficial.