Reliability and responsiveness of the activities of daily living computerized adaptive testing system in patients with stroke

Consistency between two subjective assessments of activities of daily living: Patient- and occupational therapist-reported judgments

Purpose

To examine the consistency between patient- and occupational therapist-reported judgments of patients' ability and change in ADL abilities.

Materials

Patient- and therapist-reported ADL abilities were assessed using a Visual Analogue Scale, whilst the changes in patients' ADL abilities were reported by patients and therapists using a 15-point Likert-type scale.

Methods

Repeated assessments at a 3-week interval were used. 88 inpatients with stroke and 16 occupational therapists were recruited from rehabilitation wards in a medical center.

Results

Moderate correlations (rs = .53-.56) were found between the patient- and therapist-reported ADL abilities. The patient-reported scores were significantly lower (ds = .45; ps < .001 at follow-up) than the therapist-reported scores. Only low correlation (r = .33) was found for the change scores.

Conclusions

Our findings indicated that there was only a moderate to low correlation between the patients' reports and the therapists' judgments regarding the patients' ADL ability and its change. Because both patients' reports and therapists' judgments affect decisions on rehabilitation, frequent communication may be beneficial for reaching consensus and helpful in managing the interventions.

Interrater and Intrarater Reliability of the Balance Computerized Adaptive Test in Patients With Stroke

OBJECTIVE

To examine the interrater and intrarater reliability of the Balance Computerized Adaptive Test (Balance CAT) in patients with chronic stroke having a wide range of balance functions.

DESIGN

Repeated assessments design (1wk apart).

SETTING

Seven teaching hospitals.

PARTICIPANTS

A pooled sample (N=102) including 2 independent groups of outpatients (n=50 for the interrater reliability study; n=52 for the intrarater reliability study) with chronic stroke.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Balance CAT.

RESULTS

For the interrater reliability study, the values of intraclass correlation coefficient, minimal detectable change (MDC), and percentage of MDC (MDC%) for the Balance CAT were .84, 1.90, and 31.0%, respectively. For the intrarater reliability study, the values of intraclass correlation coefficient, MDC, and MDC% ranged from .89 to .91, from 1.14 to 1.26, and from 17.1% to 18.6%, respectively.

CONCLUSIONS

The Balance CAT showed sufficient intrarater reliability in patients with chronic stroke having balance functions ranging from sitting with support to independent walking. Although the Balance CAT may have good interrater reliability, we found substantial random measurement error between different raters. Accordingly, if the Balance CAT is used as an outcome measure in clinical or research settings, same raters are suggested over different time points to ensure reliable assessments.

Intrarater and Interrater Reliability of the Hierarchical Balance Short Forms in Patients With Stroke

OBJECTIVE

To examine the intrarater and interrater reliability of a quick balance measure, the Hierarchical Balance Short Forms (HBSF), in outpatients with stroke receiving rehabilitation.

DESIGN

A repeated-assessments design (1wk apart) was used to examine the intrarater and interrater reliability of the HBSF. The HBSF was administered by a single rater in the intrarater reliability study and by 2 raters in the interrater reliability study. The raters had sufficient working experience in stroke rehabilitation.

SETTING

Seven teaching hospitals.

PARTICIPANTS

Two independent groups of outpatients (N=106; each group, n=53) with stroke in chronic stages and in stable medical condition were recruited.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURE

HBSF.

RESULTS

For the intrarater reliability study, the values of the intraclass correlation coefficient (ICC), minimal detectable change (MDC), and percentage of minimal detectable change (MDC%) were .95, 1.02, and 16.3%, respectively, for the HBSF. The 95% limits of agreement (LOA) of the HBSF ranged from -.69 to 1.19. For the interrater reliability study, the values of the ICC, MDC, and MDC% were .91, 1.22, and 18.3%, respectively, for the HBSF. The 95% LOA of the HBSF ranged from -1.01 to 1.35.

CONCLUSIONS

Our results suggest that the HBSF has satisfactory intrarater and interrater reliability for assessing balance function in outpatients with stroke. The MDC values of the HBSF are useful for both researchers and clinicians to determine whether the change in balance function of an individual patient is real when administered by an individual rater or by different raters.