Abstract
Although Blair's 7-Day Activity Interview (AI) has been widely used in epidemiological studies among adults, this instrument lacked a level of sensitivity when used among older adults with low functional capacities. To resolve this limitation, the AI was modified by adjusting the activity intensity categories to assess low-intensity activities. Concurrent and construct validity of the Modified AI were evaluated in groups of older adults known to differ in activity The Active Group (N = 30) exercised 3 days per week, whereas the Inactive Group (N = 30) did not participate in extracurricular exercise. Energy expenditures calculated using the Modified AI discriminated between the active and inactive groups (p < .0001). The Modified AI was found to have concurrent validity, that is, older adults who exercised regularly had significantly higher energy expenditures (p < .0001). The Modified AI assesses low-intensity activity and was shown to have concurrent and construct validity among these groups of older adults.