Cancer advance care planning: development of a screening tool.

Objectives: Advance care discussions are a useful communication tools for medical preferences and beneficial for shared decision-making processes in hospital settings. The present study developed the first screening tool for need for advance care planning (ACP).

Methods: In phase 1 (n=92), items were evaluated using feasibility analysis and item reduction. In phase 2 (n=201), reduced screening items were analysed for predictive value of need for ACP. Statistical analysis included receiver-operating characteristics analysis (area under the curve>0.80), optimal cut-off based on sensitivity and specificity, interpretation of OR and construct validity using correlation with death anxiety, communication avoidance within families and trust based on the relationship with the treating physician.

Results: Participants in both phases were approximately 60 years old with non-curative prognosis. After item reduction, predictive values of four possible items with good item difficulty and discrimination were compared for mild, moderate and great levels of death anxiety. A two-item combination of I am burdened by thoughts of an unfavourable course of the disease and I am burdened by the feeling of being ill-prepared for the end of life showed best prediction of death anxiety and communication avoidance. Clinical cut-off at sum-score ≥6 was of high sensitivity (95%) and specificity (81%). Previous use of social support and readiness for ACP was related to higher chance of interest in ACP.

Conclusion: Screening for need of ACP is possible with two objective items and one subjective item. Positive screening therefore indicates when to offer ACP discussions and provides routine estimation of ACP need in clinical practice.