Barthel Index information elicited over the telephone. Is it reliable?

This study examined the comparability of estimates of functional status elicited through a telephone interview and a face-to-face interview. The Barthel Index, a commonly used measure to assess activities of daily living, was administered over the telephone and then again in the home to 366 individuals, up to 5 yr after their discharge from a rehabilitation hospital. One-half of the telephone interviews were performed by health professionals and the other half by trained lay interviewers; all of the home interviews were performed by health professionals. Proxy-respondents provided information for those unable to respond for themselves. The percent agreement between the scores on the telephone and on the home interview was always greater than 90%; the intraclass correlation coefficient for the telephone/home comparison was 0.89. Responses between the modes of interview were more consistent when provided by self-respondents than when provided by proxies. The telephone assessment worked well in identifying those who did not have functional disabilities; all individuals who scored 100 on the home interview, scored 95 or better on the telephone. When differences arose, they were always in those considered to have moderate to severe impairment and were most often (23 of 29 times) in the direction of higher scores, indicative of less disability, on the telephone. The results of this study suggest that, with the exception of a small subgroup of patients, functional status can be elicited reliably over the telephone by both lay persons and health professionals.