Abstract
Purpose:
We reported that physicians’ ability to predict the success of low vision rehabilitation (LVR), relative to patient-reported outcomes, is no better than chance. This study determines if patients’ traits influence ratings physicians provide and if those traits affect patient-reported outcomes.
Methods:
The Activity Inventory (AI), a self-report visual function questionnaire, was administered pre and post-LVR to 270 low vision patients served by 28 LVR centers that participated in a collaborative study. The physical component of the SF-36, GDS, and TICS also were administered pre-LVR to measure physical capability, depression and cognitive status. Following patient evaluation, 38 LVR physicians estimated the probability of outcome success (POS) on a scale of 0 to 100% for each of their patients, using their own criteria for success.The POS ratings and change in functional ability, based on Rasch analysis of AI responses, were used to assess the effects of patients’ baseline traits on predicted and patient-reported outcomes. The distributions of SF-36 physical functioning, GDS, and TICS measures were divided into quartiles and visual acuity (VA) was categorized by severity. Analysis of variance was performed with POS and change in functional ability as the dependent variables and the quartiles for each trait serving as factors.
Results:
Baseline traits with significant effects on POS are: physical functioning (p=0.003), cognitive state (p=0.001), visual acuity (p<0.001), and baseline functional ability (p= 0.005). Depressed mood, age, gender, and number of symptomatic co-morbidities have no significant effect on the POS. Only pre-LVR functional ability (p=0.001) has a significant effect on changes in functional ability, none of the other traits has a significant effect.
Conclusions:
Physicians’ predictions of LVR outcomes appear to be driven by patient’s physical ability, cognitive functioning, VA, and baseline visual function - information physicians acquire from the patient’s history and examination. However, none of these traits, except baseline functional ability, is predictive of changes in functional ability. Although physicians appear to be using reasonable information about the patient’s state to estimate rehabilitation potential, the baseline traits selected are not reliable predictors of patient outcomes, therefore LVR physician predictions of successful outcomes are no better than chance.
Keywords: 584 low vision •
672 reading •
462 clinical (human) or epidemiologic studies: outcomes/complications