An analysis of the Short-Form 36 (SF-36) and Oxford Hip questionnaires, were used to assess 2 randomized groups, by either mail or interview, at a minimum 10-year follow up after total hip arthroplasty. Ninety-nine patients (median age 77 years) were reviewed at a median 11 years after total hip arthroplasty. There was a 91% response rate to participation in the study. There was no significant difference between the groups for missing values. The mode of administration did not affect the mean Oxford scores (P > .1), but significant differences were noted in SF-36 health scales Role Emotional and Role Physical (P = .01). Analysis of other demographic variables revealed unexpectedly that comorbidity affected the Pain score in the Oxford questionnaire (P = .002) and that age had no effect on scores obtained in either questionnaire (P > .05). The uses of both general health and disease-specific questionnaires complement each other in the assessment of such groups. The SF-36 and Oxford questionnaires give a more accurate reflection of health status when self-completed while accepting higher missing values in an elderly population.