This paper reports on a pilot study that focused on the feasibility of conducting longitudinal research on the health status of intensive care unit (ICU) survivors. Twenty patients completed a short ICU specific health status survey while in ICU and then were invited to complete the same survey--with the addition of the Medical Outcome Study Short Form-36 (SF-36) health status survey--at 6 and 12 months after their hospital discharge. While the patients were in the ICU, 20 family members completed the same survey on the patients' behalf as proxy respondents. Most patients were willing to be in the study for its duration. There was a wide variation in patients' responses to the ICU specific instrument at all data collection points. Patient and proxy responses were very similar for almost all of the physical items but there was a lack of agreement for the more subjective psychosocial items. The SF-36 was found to be more reliable and easier to score and analyse in this small study than the ICU specific health status instrument. The three SF-36 scales ranked lowest at 6 months were physical functioning, role-physical and role-emotional, while at 12 months the three lowest were role-physical, role-emotional and vitality. Total attrition at 12 months was 40 per cent. Using the results of the SF-36, sample size estimates were conducted in preparation for a larger study. It was concluded that it is feasible to conduct longitudinal research on ICU survivors but proxy responses should not be used indiscriminately.