ABSTRACTMeasuring functional activity for elderly at very low functional levels remains a challenge because many functional instruments have not been standardized in a frail elderly population. The Frail Elderly Functional Assessment questionnaire (FEFA) is a 19-item, interviewer-administered questionnaire designed to assess function in frail elderly at a very low activity level. The purpose of this study was to determine the reliability and validity of this instrument in a frail elderly population. Two groups of subjects over 65 yr old were selected to test the reliability and validity of this questionnaire. Test-retest reliability was determined by correlating the responses of 29 homebound (including nursing home-bound) subjects who answered the questionnaire on two occasions 2 wk apart. To assess the validity of the FEFA, the questionnaire was administered to 23 frail, homebound (including nursing home-bound) elderly subjects who had a Mini-Mental State Examination score of ≥18. Validity was determined by correlating patient responses to direct observations by the investigators of tasks addressed in the questionnaire. Correlation was also determined against the Katz's Activity of Daily Living index, Lawton's Instrumental Activity of Daily Living index, and the Barthel index. The reliability coefficient was 0.82. Correlation between the FEFA questionnaire and direct observation of questionnaire task performance was 0.90. Construct validity against the Katz's Activity of Daily Living, Lawton's Instrumental Activity of Daily Living, and the Barthel index showed correlations of 0.86, 0.67 and 0.91, respectively. Initial data indicate that the FEFA is a valid and reliable instrument that may be useful in assessing function in frail elderly people.