OUTCOMES

Outcomes may be viewed differently from the perspective of the patient, provider, payer, and society. The standard outcome parameters are outlined in Table 1. Although survival (live/die) seems straightforward, the endpoint in time may vary considerably, including in hospital, 30 days, 6 months, 1 year, and time to death. These variations project imprecision into many of the existing severity scoring and mortality prediction models.

Table 1 Outcome Parameters in Medicine

The use of morbidity as an outcome requires distinguishing between a complication and a pre-existing condition (i.e., a comorbidity) and providing a precise definition of a complication. A wound infection requires a precise diagnosis using the NISS classification; however, many of the infections are discovered and treated as outpatients and are never recognized by registries. Distinguishing between pre-existing renal or pulmonary disease and subsequent renal or pulmonary dysfunction after trauma can be challenging.

Hospital length of stay is a gross parameter of quality or outcome because of the variety of practices in institutions relative to the use of intensive care unit, step-down, floor, and outpatient care. The effect of early discharge is widely unknown relative to the effect on the patient’s family, visiting nurses, physician’s offices, and unanticipated hospital readmissions.

As with all outcome parameters, the cost of trauma care can be in the eye of the beholder. The payer is likely to know his/her cost with some precision, which usually reflects what he/she paid plus administrative costs. The patient usually perceives the cost as out-of-pocket expenses plus lost wages. However, the cost to the providers (i.e., emergency medical services, physician, and hospital) has many confounding variables, and is much less precise. The cost to society is even more abstract and studies have revealed substantial variation in estimates.

Quality of life has been recognized by researchers as desired outcome measures. The Functional Independence Measure (FIM), Glasgow Outcome Scale (GOS), Functional Capacity Index (FCI), Quality of Well-Being Scale, Sickness Impact Profile (SIP), and the SF-36 Survey are among the most popular in trauma-related outcome studies. Some are labor intensive and become impractical except for focused studies.

Many commercially available survey tools are religiously utilized by hospital and system administrators as measures of patient satisfaction. Goals are frequently set to meet target scores, suggesting either improvement or decline in outcome.

Finally, measuring compliance with evidence-based guidelines can provide a measurement of outcomes. Studies of compliance with Advanced Trauma Live Support (ATLS) guidelines, as well as head injury guidelines, have provided several outcome studies. Using guideline compliance as an outcome itself assumes that the desired outcome is compliance, inferring improvement in other standard patient outcomes based on the evidence on which the guidelines were developed. Therefore, caution is required in interpreting this outcome parameter.

EVIDENCE-BASED MEDICINE

The discipline of evidence-based medicine has emerged over the past decade to allow investigators to quantify the power of scientific studies based on the certainty of the scientific methods employed. Although simplistic, a classification system has been well received by the medical community that allows dialog among investigators (Table 2). Fortunately, the trend of dominance of class III data in the world of trauma outcomes is slowly giving way to significant class II and class I outcome studies in trauma and critical care.

Table 2 Evidence-Based Classification of Outcome Studies

In general, class I and class II studies provide the fuel for the strongest evidence-based guidelines on which to base the processes of trauma care (Table 3). The value of class III studies is to point the researcher toward an area of need and to help formulate the appropriate null hypothesis for a higher-power study. However, there are many clinical questions that have prohibitive barriers that prevent class I and class II studies, thereby augmenting the value of class III evidence.

Table 3 Trauma-Related Evidence-Based Guidelines