Abstract Introduction Most countries including Australia are experiencing an ageing of their population, with an increasing proportion of frail older persons requiring hospitalisation from acute illness. The aging process places the older person at risk of geriatric syndromes, such as falling, dependency in performance of Activities of Daily Living and instrumental Activities of Daily Living, confusion, bladder and bowel incontinence. New or deteriorating geriatric syndromes are a frequent occurrence among hospitalized older patients. Hospital associated factors associated with these outcomes include complications of medical therapies; polypharmacy and excessive bed rest. Few studies have been conducted into factors predicting risk of negative outcomes in older patients admitted to medical units of acute care teaching hospitals. If available, a screening tool with few predictive factors, able to be administered close to the time of admission could be used to identify patients at lower and higher risk. It is imperative that such a tool is developed empirically and tested for its accuracy in identifying patients at high risk. Aims of the research The first aim was to identify the proportion of patients aged ≥ 70 years, admitted to acute care medical units that experienced a negative outcome. These outcomes included falls during hospitalisation, presence of new or a significant decline in existing pressure ulcers, significant decline in independently performing Activities of Daily Living (ADLs), requiring increased care needs at discharge, readmission to hospital with 28 days of the index hospitalisation, bladder and bowel incontinence, and delirium. The second aim was to identify factors predicting the risk of two of these negative outcomes: requiring a higher level of care at discharge, and experiencing a decline in independently performing ADLs. Based on the predictive factors, two screening tools to identify patients at risk were developed and validated. Method A prospective cohort study of 413 acute general medical patients, aged ≥ 70 years and consecutively admitted to an acute care metropolitan 700-bed teaching hospital was conducted. Consenting patients expected to remain in hospital for more than 48 hours were included. Patients were excluded if they were admitted to intensive or coronary care units, admitted for terminal care only or were transferred from a general medical to another unit within 24 hours of admission to the ward. Trained research nurses assessed patients and used the interRAI Acute Care instrument to collect information on candidate predictive variables and negative outcomes. Patients were assessed within 36 hours of admission and at discharge to obtain information on predictive variables and negative outcomes. Patients were also followed daily to identify any instances of transient negative outcomes during hospitalisation and at 28 days following discharge to identify any instances of readmission to hospital. The 413 cases were randomly split into 309 cases in the development cohort and 104 cases in validation cohort. Logistic regression models were used to identify the predictive factors independently associated with two negative outcomes, requiring a higher level of care at discharge and experiencing a decline in independently performing ADLs. Findings At least one negative outcome was experienced by 53% of the development and 63% of the validation cohort. The most common negative outcomes experienced were: delirium (27%; 23%), a significant decline in ADLs (19%, 22%), requiring a higher level of care at discharge (16%, 16%), and readmission to hospital within 28 days of discharge (17%, 28%) in the development and validation cohorts respectively. The logistic regression analysis identified four independent factors associated with requiring higher levels of care at discharge: ‘short term memory problems’ (OR 4.21, 95% CI 1.79, 9.89; p=0.001); ‘dependence in toilet use’ (OR 3.51, 95% CI 1.14, 10.84; p=0.029); ‘dependence in hygiene’ (OR 2.76, 95% CI 1.16, 6.56; p=0.021), and ‘use of community services prior to admission’ (OR 2.41, 95% CI 1.12, 5.16; p= 0.024). A screening tool developed to assess patients at lower and higher risk had a sensitivity, specificity, positive predicted value (PPV) and negative predictive value (NPV) of 77.27%, 73.66%, 36.56% and 94.29% respectively. Reasonable accuracy was evident when tested in the validation sample. Sensitivity, specificity, PPV and NPV were 60%, 76.32%, 33.33% and 90.63% respectively. Predictive factors associated with a significant decline in ADLs were: ‘history of falling’(OR 2.21, 95% CI 1.12, 4.36; p= 0.023), ‘no interest in things enjoyed normally’ (OR 4.30, 95% CI 1.92, 9.64; p=0.000), ‘dependence in management of finances’ (OR 3.93, 95% CI 1.63, 9.48; p =0.002) and ‘hearing problems’ (OR 2.38, 95% CI 1.05, 5.39; p =0.038). The screening tool had sensitivity, specificity, PPV and NPV in the development cohort of 74.55%, 69.13%, 36.6% and 92% respectively and 45%, 65.79%, 25.7% and 82% respectively in the validation sample. Conclusion The tools require further validation in larger samples in diverse settings. Future research should focus on developing a screening tool that could predict risk of a number of negative outcomes to enhance the provision of quality patient care.
| Identifer | oai:union.ndltd.org:ADTP/254283 |
| Creators | Prabha Lakhan |
| Source Sets | Australiasian Digital Theses Program |
| Detected Language | English |
Page generated in 0.0117 seconds