BackgroundAn increasing number of older patients are hospitalised. Prognostic uncertainty causes hospital doctors to be reluctant to make the switch from cure to care. The Supportive and Palliative Care Indicators Tool (SPICT) has not been validated for prognostication in an older hospitalised population.AimTo validate SPICT as a prognostic tool for risk of dying within one year in older hospitalised patients.DesignProspective multicentre study. Premorbid SPICT and 1-year survival and survival time were assessed.Setting/participantsPatients 75 years and older admitted at acute geriatric (n=209) and cardiology units (CUs) (n=249) of four hospitals.ResultsIn total, 59.3% (124/209) was SPICT identified on acute geriatric vs 40.6% (101/249) on CUs (p<0.001). SPICT-identified patients in CUs reported more functional needs and more symptoms compared to SPICT non-identified patients. On acute geriatric units, SPICT-identified patients reported more functional needs only.The HR of dying was 2.9 (95% CI 1.1 to 8.7) in SPICT-identified versus non-identified after adjustment for hospital strata, age, gender and did not differ between units. One-year mortality was 24% and 22%, respectively, on acute geriatric versus CUs (p=0.488). Pooled average sensitivity, specificity and partial area under the curve differed significantly between acute geriatric and CUs (p<0.001), respectively, 0.82 (95%CI 0.66 to 0.91), 0.49 (95%CI 0.40 to 0.58) and 0.82 in geriatric vs 0.69 (95% CI 0.42 to 0.87), 0.66 (95% CI 0.55 to 0.77) and 0.65 in CUs.ConclusionsSPICT may be used as a tool to identify older hospitalised patients at risk of dying within 1 year and who may benefit from a palliative care approach including advance care planning. The prognostic accuracy of SPICT is better in older patients admitted at the acute geriatric versus the CU.