Responsiveness of different rating instruments in spinocerebellar ataxia patients.

OBJECTIVE

To determine the longitudinal metric properties of recently developed clinical assessment tools in spinocerebellar ataxia (SCA).

METHODS

A subset of 171 patients from the EUROSCA natural history study cohort (43 SCA1, 61 SCA2, 37 SCA3, and 30 SCA6) were examined after 1 year of follow-up. Score changes and effect size indices were calculated for clinical scales (Scale for the Assessment and Rating of Ataxia [SARA], Inventory of Non-Ataxia Symptoms [INAS]), functional tests (SCA Functional Index [SCAFI] and components), and a patient-based scale for subjective health status (EQ-5D visual analogue scale [EQVAS]). Responsiveness was determined in relation to the patient's global impression (PGI) of change and reproducibility described as retest reliability for the stable groups and smallest detectable change.

RESULTS

Within the 1-year follow-up period, SARA, INAS, and SCAFI but not EQVAS indicated worsening in the whole group and in the groups with subjective (PGI) worsening. SCAFI and its 9-hole pegboard (9HPT) component also deteriorated in the stable groups. Standardized response means were highest for 9HPT (-0.67), SARA (0.50), and SCAFI (-0.48) with accordingly lower sample size estimates of 143, 250, or 275 per group for a 2-arm interventional trial that aims to reduce disease progression by 50%. SARA and EQVAS performed best to distinguish groups classified as worse by PGI. All scales except EQVAS reached the criterion for retest reliability.

CONCLUSION

While both the Scale for the Assessment and Rating of Ataxia and the SCA Functional Index (SCAFI) (and its 9-hole pegboard component) had favorable measurement precision, the clinical relevance of SCAFI and 9-hole pegboard score changes warrants further exploration. The EQ-5D visual analogue scale proved insufficient for longitudinal assessment, but validly reflected patients' impression of change.