Comparing the Responsiveness of Barthel Index and Longshi Scale in Assessing the ADL of Stroke Survivors

The whole experiment process has been supervised by a research assistant, and all the data collected are recorded on paper case report forms. The research assistant also takes charge of checking the consistency of data enter into the registry against predefined rules. Stroke survivors in Shenzhen Second People's Hospital have been invited to participate in this study and divide into three groups according to their duration of stroke (acute stroke, subacute stroke and chronic strike).

Data collection:

Baseline information was collected within 24 hours of admission. ADL of stroke survivors was assessed on admission, at discharge and 3 months after discharge.

Sample size calculation:

The sample size in this study was calculated according to the sample size calculation formula: n=[(µα+µβ)(µα+µβ)(1+1/k)p(1-p)]/(pe-pc)(pe-pc). Investigations in many places in China showed that varying degrees of dysfunction remain in 70%-80% of stroke survivors, and the disability rate is 30.2%-62.8%. In this reserach，p=(pe+kpc)/(1+k)，pe=0.3，pc =0.6, α=0.05，β=0.1, so µ0.05=1.6449，µ0.10=1.2816，thus the sample size in our study was n=47. We assume that 10% of them loss to follow-up, thus the sample size in our study was 52 in each group.

Statistical analysis:

• Internal responsiveness First, standardized effect size (SES) was calculated by dividing the mean change in score between admission and discharge assessments by the standard deviation (SD) of the admission score. second, the standardized response mean (SRM) was obtained by dividing the mean change scores by the SD of the change score between admission and discharge scores. A value of SES or SRM>0.8 was large, 0.5 to 0.8 moderate, and 0.2 to 0.5 small.
• External responsiveness The pearson correlation coefficient (r) was used to investigate the external responsiveness. An r value >0.75 represented good to excellent association; values of 0.50 to 0.75, moderate to good association; values of 0.25 to 0.50, fair association; and values ≤0.25, little association between the changes in scores on these measures. And the significance threshold was set at 0.05.
• Receiver operating characteristic (ROC) curve For this study, the area under ROC curve (AUC) obtained by comparing subjects who improved by one or more levels on the modified Rankin scale vs. those who exhibit no change on the modified Rankin scale. Larger AUC means higher screening accuracy. In general, an AUC of 0.5 suggests no discrimination, 0.7 to 0.8 is considered acceptable, 0.8 to 0.9 is considered excellent, and more than 0.9 is considered outstanding.

Inclusion criteria

• age 18-80 years
• diagnosis of cerebral infraction or intracerebral hemorrhage
• stable vital signs

Exclusion criteria

• brain tumor, Parkinson's disease or active epilepsy within three months
• impaired cognitive functions
• participation in any other clinical study