Predictive validation of a questionnaire for the assessment of social care needs in hospitalized older adults, the geriatric discharge complexity score: a diagnostic accuracy study

Background Anticipating discharge challenges in the elderly population is essential to support effective care planning and reduce risks during care transitions. However, there is a lack of targeted assessment tools specifically designed for use in hospital wards to address this need. Aims To evaluate the predictive value of the Geriatric Discharge Complexity Score (GDCS) and Blaylock Risk Assessment Screening Score (BRASS) with respect to discharge difficulties in a population of hospitalized older adults. Methods The study was conducted on a sample of 416 subjects (175 females) with mean age of 88.2 ± 5.7 years. All subjects underwent evaluation with GDCS, Barthel scale, BRASS and count of total days of hospitalization and days of hospitalization related to social-welfare problems. Results The GDCS showed sensitivity of 97.9% and specificity of 69.8% with an area under the curve (AUC) of 0.936. The BRASS showed sensitivity of 86.3% and specificity of 20.2% with an AUC of 0.518. Total and excess days of hospitalization for social problems were significantly higher in subjects with elevated GDCS. Conclusion The GDSC seems more predictive of social care issues that may lead to a prolongation of hospital stay than the BRASS.