standard approaches to assess trauma outcomes. The challenges are summarized below.
The Functional Independence Measure (FIM) (Keith et al., 1987) has gained wide acceptance in the field of medical rehabilitation as a measure of the "burden of care" associated with an illness or injury. The FIM score is generally assigned by a caregiver on the basis of direct observations of performance, although more recently developed telephone versions of the FIM rely on the individual's own assessment of his or her performance. Although further testing of the FIM is warranted, it holds promise as an effective tool for assessing the impact of inpatient rehabilitation on outcomes following serious trauma. However, the FIM does not encompass broader issues of outcome related to role activity, psychological well-being, and general health perceptions. It has also been criticized for its lack of sensitivity to the range of disabilities associated with traumatic brain injury.
Less frequently applied in the evaluation of trauma care and rehabilitation have been the wide array of health status and HRQOL measures. Although these measures vary widely in form and content, they share two important characteristics. First, they all measure function across multiple domains, including not only physical health, but also mental and cognitive health, social function, role function, and general health perceptions. Second, and even more important, health status and HRQOL measures assess outcomes from the patient's or consumer's perspective through the use of well-constructed questionnaires. One of the more important developments in health care over the past several years has been the recognition that the patient's point of view is important in evaluating the success of alternative therapies (Ware, 1995). Increasingly, patient-oriented measures of health status are playing a central role in health care evaluations (Relman, 1988). Examples of health status measures that have been used in measuring outcomes following trauma include the Sickness Impact Profile (Bergner et al., 1985), the General Health Status Measure Short Form-36 (Ware and Sherbourne, 1992), the Quality of Well-Being Scale (Kaplan et al., 1989; Holbrook et al., 1998), and the Rand Health Insurance Study Measures for Child Health Status (Eisen et al., 1980). However, their application has not been widespread, and important questions remain regarding their discriminate validity and their responsiveness to different treatments and to changes over time. Few studies have compared and contrasted available measures across the wide range of types and severities of injuries. Methodological research is critical in identifying and promoting the use of appropriate measures. An important issue is the extent to which available measures are sensitive to the cognitive deficits and behavioral changes that often accompany head injury.
Broad application of appropriate health status and HRQOL measures using standard protocols is essential for developing benchmarks for trauma outcomes. Substantial progress has been made over the past decade in establishing hospital-based trauma registries and defining the minimal data set needed for quality improvement activities. Typically, however, registries are not designed to collect