The inclination for conscious motor control after stroke: validating the Movement-Specific Reinvestment Scale for use in inpatient stroke patients

Abstract

Purpose: Stroke survivors are inclined to consciously control their movements, a phenomenon termed “reinvestment”. Preliminary evidence suggests reinvestment to impair patients’ motor recovery. To investigate this hypothesis, an instrument is needed that can reliably assess reinvestment post-stroke. Therefore, this study aimed to validate the Movement-Specific Reinvestment Scale (MSRS) within inpatient stroke patients.

Method: One-hundred inpatient stroke patients (<1 year post-stroke) and 100 healthy peers completed the MSRS, which was translated to Dutch for the study purpose. To assess structural validity, confirmatory factor analysis determined whether the scale measures two latent constructs, as previously reported in healthy adults. Construct validity was determined by testing whether patients had higher reinvestment than controls. Reliability analyses entailed assessment of retest reliability (ICC), internal consistency (Cronbach’s alpha), and minimal detectable change.

Results: Both structural and construct validity of the MSRS were supported. Retest reliability and internal consistency indices were acceptable to good. The minimal detectable change was adequate on group level, but considerable on individual level.

Conclusions: The MSRS is a valid and reliable tool and suitable to assess the relationship between reinvestment and motor recovery in the first months post-stroke. Eventually, this may help therapists to individualize motor learning interventions based on patients’ reinvestment preferences.

Implications for rehabilitation

• This study showed that the Movement-Specific Reinvestment Scale (MSRS) is a valid and reliable tool to objectify stroke patients’ inclination for conscious motor control.

• The MSRS may be used to identify stroke patients who are strongly inclined to consciously control their movements, as this disposition may hinder their motor recovery.

• Eventually, the MSRS may enable clinicians to tailor motor learning interventions to stroke patients’ motor control preferences.

Keywords:

• CVA
• motor control
• patient-reported outcome measure
• physical therapy
• rehabilitation
• reinvestment
• stroke

Acknowledgements

We would like to thank Christa de Jonge, Charlotte Postma, Nynke Bos, and Mette van Kruijsbergen for their contributions to the data collection, Jacinta Kal and Steven Barker for their aid in the translation process, and Dr. Wouter Weeda for his valuable statistical advice.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Notes

¹ This procedure was justified, as skewness and kurtosis of each item was well below the recommended [21] values (M_skew =0.62 < 2, M_kurt = 0.25 < 7).

² All three variables were normally distributed in the patient group, but somewhat positively skewed in the control group (M_skew = 0.9). As ICC is highly robust to slight deviations from normality [29] we chose to use the original (non-transformed) data for this analysis.

³ Additional analysis of the inter-item correlation matrix revealed that item 1 (“I remember the times when my movements have failed me”) correlated poorly with items 3 (r = 0.16) and 9 (r = 0.11), and demonstrated weak item-total correlation (i.e. r < 0.3). However, it was decided not to remove this item, considering that confirmatory factor analysis showed item 1 to have satisfactory factor loading (0.61), and since removal of this item would only slightly improve Cronbach’s alpha of the CMP subscale (α = 0.67).