Medicine (RMH) - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemPredicting poor clinical outcomes among older inpatients( 2021)Predicting poor clinical outcomes among inpatients has significant impacts on clinical decision making and treatment plan. A valid prognostic tool allows early interventions and patient-specific therapies by identifying patients who are at a higher risk of poor clinical outcomes. The Comprehensive Geriatric Assessment (CGA), a multidisciplinary diagnostic and treatment process, is implemented in geriatric rehabilitation inpatients and it includes multiple validated clinical assessment tools to evaluate older patients’ medical, psychosocial and functional limitations. The primary objective of this PhD thesis is to evaluate the performance of morbidity measures and frailty assessment tools in predicting poor clinical outcomes among geriatric rehabilitation inpatients. The systematic review included in this thesis shows that morbidity measures, predominantly the Charlson Comorbidity Index (CCI) and the Cumulative Illness Rating Scale (CIRS) and Geriatric Index of Comorbidity (GIC), are predictive for post-discharge mortality among older inpatients and the predictive performances are better in longer follow-up period. However, they are not predictive for functional decline among older inpatients. A total of 1890 geriatric rehabilitation inpatients (median age 83.4 [IQR 77.6-88.4] years, 56% female) were included in this thesis. They were mostly independent two-week prior to hospitalization (median Activities of Daily Living (ADL) score: 6 [IQR 4-6]; median Instrumental (I)ADL score: 5 [IQR 2-7] respectively). At admission to the geriatric rehabilitation wards, they had a median CCI score of 2 [IQR 1-4] and were mostly frail (median Clinical Frailty Scale (CFS) score: 6 [IQR 5-7]). Among these geriatric rehabilitation inpatients, three distinct functional trajectories were identified from two-week prior to hospitalization to three-month post-discharge: remained poor, deteriorated and recovered. Cognitive impairment and greater frailty status, assessed using the CFS, were each associated with deterioration and remaining poor in patients’ functional performance. Greater frailty status, assessed using the CFS, was also shown to be associated with in-hospital, three-month and one-year post-discharge mortality. An increased frailty severity from admission to discharge from geriatric rehabilitation was also associated with three-month post-discharge mortality. This thesis shows that frailty assessment tools, based on either clinical judgment (CFS) or laboratory tests (frailty index-laboratory test (FI-lab)), were poor predictors for one-year mortality among geriatric rehabilitation inpatients. Nonetheless, the clinical judgement-based frailty assessment tool (CFS) was slightly better than the laboratory test-based assessment tools (FI-lab and modified FI-lab) in predicting mortality and that they were all significantly associated with one-year mortality. Furthermore, this thesis includes a study applying the machine learning approach and the characteristics included in the CGA were shown to be poor predictors for geriatric rehabilitation length of stay. Lastly, greater frailty status, assessed using the CFS, was associated with the transitions from premorbid functional performance to institutionalisation and mortality from two-week preadmission to three-month post-discharge. This thesis emphasises the importance of assessing patients’ frailty status at geriatric rehabilitation admission as it reflects those with a higher risk of poor clinical outcomes. Future studies should focus on evaluating the treatment and intervention for frailty and investigate the importance of other CGA components.