The use of survivorship analysis requires an assumption that patients who are lost to follow-up are no more or less likely to be at risk of failure of an operation or a procedure than are patients who are still being followed. This is a major assumption in long-term orthopaedic studies, in which a high percentage of patients are usually lost to follow-up. We compared the survivorship curve for the first 100 Tharies replacements done at our institution (which were completed by September 1977), using data that were collected in the standard way up to 1985, through a letter requesting a follow-up visit, with the curve for the same patients that was based on almost complete follow-up data that were gathered by telephone from 1985 on. The similarity of the two curves suggested that the assumptions that are necessary for the validity of survivorship analysis are reasonable, even in the orthopaedic setting, in which many patients are lost to follow-up. The usefulness of the survivorship curve for prediction was also evaluated by comparing the curve based on the first forty-six of the 100 Tharies replacements (before 1977) with the curve based on the last fifty-four such operations (from January 1977 to September 1977). The results of these two comparisons suggest that survivorship analysis is a valid technique to use in the long-term evaluation of patients who have had a joint replacement.