Robustness of reliable change indices to variability in Parkinson's disease with mild cognitive impairment

Appl Neuropsychol Adult. 2016 Nov-Dec;23(6):399-402. doi: 10.1080/23279095.2016.1160907. Epub 2016 May 4.

Abstract

Ability to identify change is crucial for measuring response to interventions and tracking disease progression. Beyond psychometrics, investigations of Parkinson's disease with mild cognitive impairment (PD-MCI) must consider fluctuating medication, motor, and mental status. One solution is to employ 90% reliable change indices (RCIs) from test manuals to account for account measurement error and practice effects. The current study examined robustness of 90% RCIs for 19 commonly used executive function tests in 14 PD-MCI subjects assigned to the placebo arm of a 10-week randomized controlled trial of atomoxetine in PD-MCI. Using 90% RCIs, the typical participant showed spurious improvement on one measure, and spurious decline on another. Reliability estimates from healthy adults standardization samples and PD-MCI were similar. In contrast to healthy adult samples, practice effects were minimal in this PD-MCI group. Separate 90% RCIs based on the PD-MCI sample did not further reduce error rate. In the present study, application of 90% RCIs based on healthy adults in standardization samples effectively reduced misidentification of change in a sample of PD-MCI. Our findings support continued application of 90% RCIs when using executive function tests to assess change in neurological populations with fluctuating status.

Keywords: Problem solving; test construction; tests; treatment.

Publication types

  • Randomized Controlled Trial

MeSH terms

  • Adrenergic Uptake Inhibitors / therapeutic use
  • Aged
  • Atomoxetine Hydrochloride / therapeutic use
  • Cognition Disorders / complications*
  • Cognition Disorders / drug therapy
  • Disease Progression
  • Double-Blind Method
  • Executive Function / drug effects
  • Executive Function / physiology*
  • Female
  • Humans
  • Male
  • Middle Aged
  • Neuropsychological Tests*
  • Parkinson Disease / complications*
  • Psychometrics
  • Reproducibility of Results
  • Treatment Outcome*

Substances

  • Adrenergic Uptake Inhibitors
  • Atomoxetine Hydrochloride