Validation of a quality of life questionnaire for critically ill patients

Abstract

Objective: Development and validation of quality of life questionnaire for critical care patients.

Design: Prospective study.

Setting: Intensive care unit (ICU) of a general hospital and ICUs of 83 Spanish hospitals.

Sample: Patients admitted to the ICU > 18 years of age; close family members.

Method: A committee of experts designed a questionnaire with characteristics judged essential for intensive care use: easy, quick administration (5-10 min); capable of completion by patient or close family member, by direct or telephone interview. Fifteen items relevant to critical care patients were grouped in three subscales: basic physiological activities, normal daily activities, and emotional state. Reproducibility of interobserver, intraobserver, patient/family member and telephone/direct interviews was analysed and also internal consistency, responsiveness, and main components.

Results: Internal consistency (578 patients): Cronbach's alpha coefficient = 0.85. Reproducibility: intraobserver reproducibility (n = 150): Spearman correlation coefficient = 0.92. Interobserver (n = 85): correlation = 0.92. Patient/family member (n = 81): correlation = 0.92. Telephone/direct interview (n = 54): correlation = 0.96.

Validity: factorial analysis confirmed that the three subscales were fundamental questionnaire components. There was good concordance between questionnaire/subscale and Glasgow Outcome Scale (GOS) results. Responsiveness: quality of life score changes between preadmission and 6 months' postdischarge correlated with GOS findings (weighted kappa index = 0.56).

Conclusions: Questionnaire meets objectives recommended for critical care use, and fulfills essential requirements of validity and reproducibility when applied to critically ill patients.