Assessing environmentally determined mobility disability: Self-report versus observed community mobility

OBJECTIVES: To examine the test-retest reliability and concurrent validity of a new self-report measure of mobility function by comparing it with observed mobility, self-reported activity of daily living (ADL) function, and performance-based measures of gait and balance. DESIGN: Cross-sectional study involving two groups of older adults. SETTING: Community sites in Seattle, Washington, and Waterloo, Ontario, Canada. PARTICIPANTS: Fifty-four adults aged 70 and older, recruited. MEASUREMENTS: Subjects completed the Environmental Analysis of Mobility Questionnaire (EAMQ), reporting frequency of encounter and avoidance of 24 features of the physical environment, grouped into eight dimensions, on two occasions 1 week apart. Subjects were observed and videotaped during six trips into the community; frequency of encounters with environmental features within the eight dimensions was recorded. EAMQ encounter and avoidance scores were compared with observed environmental encounters, with disability in ADLs and instrumental ADLs (IADLs), and lower extremity functional measures including the Short Physical Performance Battery (SPPB) and the Berg Balance Test. RESULTS: EAMQ test-retest reliability was high for all eight dimensions (intraclass correlation coefficient range=0.81-1.0) and for summary encounter (0.98) and avoidance (0.96) scores. Observed mobility was significantly correlated (Spearman correlation = r) with EAMQ summary encounter (r=0.66) and avoidance (r=-0.58) scores. Moderate correlations were present between the EAMQ (encounter or avoidance) and observed mobility in the distance, temporal, terrain, posture, load, and density dimensions but not in the attention and ambient dimensions. EAMQ encounter/avoidance was significantly associated with ADL and IADL ability and performance on the SPPB and Berg Balance Test. CONCLUSION: Self-reported frequency of encounter and avoidance of specific environmental features appears to be a valid method for determining environmentally specific mobility disability but needs to be confirmed in a larger sample.