ARE PROXY ASSESSMENTS OF HEALTH-STATUS AFTER STROKE WITH THE EUROQOL QUESTIONNAIRE FEASIBLE, ACCURATE, AND UNBIASED

Background and Purpose It is often difficult to determine the health-r elated quality of life (HRQoL) of stroke patients because physical and cognitive problems limit their ability to complete complex questionna ires. A proxy, such as a family member or caregiver, may be able to gi ve an estimate of the patients' health status. We therefore examined t he agreement between the HRQoL as assessed by a series of patients and that assessed by their proxies. Methods We studied the validity of th e EuroQol in a series of 152 patients from our prospective registry of patients with first (or recurrent) stroke. We asked patients to ensur e that a friend or relative (a proxy) who knew them well was available at the time of the interview. We asked each proxy to complete a EuroQ ol questionnaire independently on behalf of the patient. Results Proxi es completed forms for 130 patients (86%). Agreement between responses from the patients and those from their proxies was better for patient s who were able to self-complete the EuroQol than for patients who req uired the EuroQol to be administered by interview. For both groups, ag reement was best for the self-care domain and worst for the domain tha t assessed psychological outcome. For the more severely affected patie nts, agreement was only fair for the pain and social functioning domai ns and no better than chance alone for the psychological functioning d omain (kappa=0.05, 95% confidence interval, 0 to 0.43). Patients tende d to rate their own health status as better than their proxies did (P< .05). Conclusions We found moderate agreement between responses from p atients and those from their proxies for the more directly observable domains of the EuroQol. Proxy agreement was less good for the more sub jective domains. In health surveys, allowing responses by a proxy incr eases response rate. However, the disadvantages inherent in the use of proxy responses must be considered carefully. In general, some domain s of HRQoL information obtained from a proxy may be sufficiently valid and unbiased to be useable in most types of trials and surveys.