Primary or specialist care at home: multicenter validation study of the “CP.BA.SPE 2025” questionnaire for the identification of the level of palliative care in advanced cancer patients

Reliable tools to discriminate between primary and specialist home palliative care are scarce, and no questionnaire has yet been validated in domiciliary oncology practice. The aim of the study was to validate the 17-item CP.BA.SPE 2025 questionnaire for identifying advanced cancer patients who need specialist home palliative care. Cross-sectional multicentre study (pilot+multicentre validation) conducted in eight Home Palliative Care Units (HPCU) across Emilia-Romagna, Italy (March 2022-June 2024). The CP.BA.SPE 2025 was completed together with the HPCU’s independent judgement of the appropriate care level (gold standard). Item weights were derived with multivariable logistic regression; internal consistency (Kuder-Richardson-20), inter-rater reliability (Cohen’s κ), calibration (Hosmer–Lemeshow) and discrimination (ROC Area Under Curve-AUC) were calculated. A total score ≥14 defined “complex” cases. 543 consecutive adults with metastatic or locally-advanced cancer and Australia-modified Karnofsky Performance Status ≤50 were assessed at first home visit. The CP.BA.SPE 2025 questionnaire scores ranged 0-36.5. With the 14-point cut-off, sensitivity was 0.79 (95% CI 0.74-0.84) and specificity 0.82 (95% CI 0.78-0.87), AUC 0.88 (95% CI 0.85-0.91), Brier score 0.14. Inter-rater agreement among individual professionals was excellent (κ 0.83-0.92); agreement between HPCU and an external comparison team was moderate (κ 0.43). Internal consistency was acceptable (Kuder-Richardson-20 = 0.65). The tool classified 251/543 patients (46.3%) as complex, aligning with international estimates of specialist-care need. CP.BA.SPE 2025 is quick to administer, reproducible and shows good diagnostic accuracy for directing advanced cancer patients to the appropriate level of home palliative care. Its adoption could foster equitable resource allocation within home-care networks; external and longitudinal validation are recommended.