SummaryThe aim of this study was to assess the acceptability, validity and reliability of the Short Form 36 quality of life questionnaire in 166 adult patients following discharge from a general intensive care unit. Reliability was quantified by measuring internal consistency using correlation among items and Cronbach's alpha coefficient. Reliability coefficients were calculated from two-way analysis of variance. Construct validity was tested by examining differences in scores between sex and age groups. Content validity was reflected by the spread of dimension scores. Acceptability to patients appeared reasonable, although considerable nursing time was required to administer the questionnaire. The measures of reliability exceeded recognised statistical standards in all but two instances. Construct validity was confirmed by lower scores being reported by women and older age groups. The scores of six of the eight dimensions were spread throughout the entire range of possible scores suggesting acceptable content validity. A. Ridley Accepted: 4 September 1996 Assessing the outcome of health care interventions is one of the key challenges of this decade. Mortality is a classic outcome indicator and has undoubtedly played a valuable role in the evaluation and comparison of intensive care units (ICUs). However, in the current climate, there is increasing pressure to consider other aspects of patient outcome, two of the most important being quality of life and functional ability. Mortality, although a vital measure, is too crude an indicator for this purpose. Patients who survive critical illness should be evaluated in terms of their ongoing health and the degree to which they are able to return to their previous lifestyle.Identifying standardised outcome measures is problematical in health care, but further complexities arise when dealing with patients obtunded by critical illness. Diseasespecific tools are available, such as alteration in quality of life following the onset of rheumatoid arthritis [1], but such tools can only be applied to a select group of patients.A general tool, reflecting the wide range of critical illness seen in ICU, may be more appropriate. An ideal general tool should be easy to administer and not present too great a burden for the patient; it should have wide application and yet be sensitive to modest changes in quality of life. Converting patient perceptions into a score, which can be analysed, makes assumptions about the nature and internal relationships of the scales or scores used. These assumptions may change if the tool is used on a different patient population. Therefore, it is most important that the tool is shown to be reliable and valid for use in its new setting before changes in quality of life as a result of illness are investigated. If a measure can provide reliable and accurate results, then comparisons between health care programmes, groups of patients and ICUs could be made.The Short Form 36 (SF-36), developed in the early 1990s and containing 36 questions, is a self-co...

SummaryThe Short Form 36 was used to compare critically ill patients' premorbid quality of life with normal values and investigate any changes following 6 months convalescence. One hundred and sixty-six survivors completed the Short Form 36 at discharge from intensive care. The answers given by survivors were significantly lower than normal for all dimensions. However, 21 patients who suffered from acute life-threatening conditions were identified and their overall scores were similar to normal values. After 6 months, 95 questionnaires were returned. Patients who had suffered acute pathologies reported significant decreases in quality of life whilst other patients with pre-existing ill health reported significant improvement with reduced pain and better mental health, vitality and social function. This study suggests that quality of life of most patients admitted to intensive care is not as good as in the normal population but does not deteriorate except for those patients admitted after acute life-threatening events.

SummaryThe aim of this study was to determine the reliability and validity of relatives' assessment of patients' quality of life and to measure the agreement between patients' and relatives' responses to the Short Form 36 quality of life questionnaire, at discharge from and 6 months following intensive care treatment. Ninety-nine patient-relative pairs were studied. Reliability was quantified by using measures of internal consistency (Cronbach's alpha and correlation coefficients) and reliability coefficients. Relatives' responses met the required standards of reliability and validity, but reliability was consistently weaker in the mental health dimension. Relatives' and patients' scores differed significantly in six dimensions at discharge; however, agreement between patients' and relatives' responses, as measured by the Kappa statistic, was fair, improved over 6 months, and was greatest in aspects concerning physical health. We conclude that relatives are able to give a good proxy assessment of functional aspects of quality of life.