Abstract
Objectives to examine the construct validity and sensitivity of the Barthel Index (BI) and, thus, to assess its utility as an outcome measure in stroke rehabilitation. Design comparison of BI scores with qualitative assessment of functional status for a cohort of patients recovering from a first stroke. Setting re-analysis of data arising in an acute hospital setting (care of the elderly and medical units) as a result of a nursing and midwifery audit initiative, funded by the Welsh Office. Participants 23 patients with discharge BI scores greater than or equal to 16/20. (12 men, 11 women. Age range 55–89.) Main outcome measure within subject comparison of BI scores, with numbers and types of impairment and/or disability as defined by the International Classification of Impairment, Disability and Handicap. Results for this cohort of patients there was no direct relationship between BI score and numbers of identified impairments and disabilities. Patients with similar BI scores had different patterns of impairment. Patients with similar change scores had different patterns of improvement. Patients with maximum BI scores on discharge had residual impairments and disabilities. Conclusions the construct that BI measures is not clearly defined. Therefore, the meaning of BI scores, in terms of patient functional ability, is unclear, and its construct validity is questionable. Its sensitivity is not demonstrated as it has a definite ceiling effect. There is a need to approach outcome measurement for complex areas, such as stroke rehabilitation, in a way that clearly specifies the construct being studied and allows individual progress and support needs to be assessed.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call