Abstract

Objectives: To understand the treatment goals of Alzheimer’s disease (AD) patients, carers, and physicians; to estimate whether clinically important goals are met during treatment with donepezil; and to compare a measure of goal attainment with standard measures used to evaluate AD treatment.

Methods: In a 12 month phase IV trial, 108 patients with mild to moderate AD, their primary carers, and treating physicians set goals assigned to five domains, using Goal Attainment Scaling (GAS) as the primary outcome. Goal attainment was assessed quarterly. GAS scores were correlated with standard outcomes, including the Alzheimer’s Disease Assessment Scale-Cognitive (ADAS-cog), and the Clinician’s Interview-Based Impression of Change-Plus (CIBIC-plus).

Results: Physicians set fewer goals (342, mean (SD) per patient=3 (1)) than patients/carers (855, mean=9 (3)), particularly in leisure (20% by physicians compared with 76% by patients/carers), and social interaction (24% versus 49%). Physicians observed statistically significant improvement in global goal attainment for six months, and patients/carers for nine months. Patients/carers described consistent goal attainment, whereas physicians observed variable effects, such as decline in cognition but improved social interaction and behaviour. Physician global GAS scores correlated highly with the CIBIC-plus at weeks 12 (r= −0.82) and 52 (r=−0.80), but not with the ADAS-cog (r=0.12 and r=−0.45, respectively). Patient/carer global GAS scores correlated moderately with the physician’s CIBIC-plus (week 12 r=−0.51; week 52 r=−0.56), and nominally with the ADAS-cog.

Conclusions: Patients/carers and physicians differ in their expectations and impressions of treatment effects. Clinically important changes correlated only modestly with psychometric tests. Attainment of treatment goals does not accord with a simplistic model in which successful AD treatment means that all declines uniformly improve.