The measurement properties of the original Barthel Index : A systematic review

Abstract

Background: Worldwide many countries are experiencing ageing populations. Ageing can result in older adults experiencing compromised health and increased hospitalisations making older adults recurrent users of health care services. The process of ageing can reduce an older adult’s ability to perform everyday tasks and participate in the activities required for daily life. As a result, older adults are often referred to occupational therapists to improve functional capacity and support engagement in activities and occupations. One role occupational therapists perform when working with older adults in a hospital or rehabilitation setting is to enhance function in self-care tasks and plan for discharge. Occupational therapists use functional assessments to determine an older adult’s ability to carry out everyday tasks. Some occupational therapists use standardised assessments whereas others prefer to use non-standardised assessments. The major barriers to occupational therapists using non-standardised assessments are a lack of knowledge and skill. Standardised assessments are used to measure function, monitor change and determine the effectiveness of therapy. The quality of these measurements relies on the validity and reliability of standardised assessments when used with specific patient groups and clinical settings they were designed for. The original Barthel Index (BI) developed by Mahoney and Barthel (1965) is a standardised functional assessment widely used within rehabilitation, hospital and community settings with various patient groups. To date, the validity and reliability of the original BI has not been established with older adults. Aim: The aim of this research is to investigate the measurement properties of the original BI and to determine its suitability as a standardised assessment to measure function with older adults. Method: A comprehensive systematic review of the literature was undertaken and the following databases were searched; Cumulative Index of Nursing and Allied Health (CINAHL), Embase and Medline. Methodological quality of the included studies was assessed using the COnsensus-based Standard for the selection of health Measurement INstruments (COSMIN) checklist. Results: 3273 abstracts were screened and 5 articles included. COSMIN rated structural validity (good), cross-cultural validity (poor), hypothesis testing (fair to poor), internal consistency (poor) and responsiveness (fair). No studies investigated reliability, measurement error, criterion or content validity. Discussion: Limited research has been undertaken on the measurement properties of the original BI, highlighting the need for further robust research to be undertaken on the measurement properties of the original BI, so that the reliability, validity and limitations of the original BI to be established with older adults. Multiple versions of the BI were identified making interpretation of results problematic. This review recommends re-naming these versions to ensure transparency and straightforward interpretation of results. Implications: This and further research will continue to bridge knowledge gaps within the occupational therapy profession and enable occupational therapists to make informed decision when selecting the most appropriate standardised assessment to measure function with older adults. Breaking down the barriers to measuring outcomes through increased knowledge and use of standardised assessments will increase occupational therapists adherence to evidence base practice and professional accountability to the occupational therapy profession. This in turn will provide older adults with accurate measurements of their functional capacity which will ultimately impact on patient recovery, quality of care and earlier discharge.