Informatics Support of Outcome Measures

Abstract

Outcome measures enable the comparison of a patient’s health status before and after an intervention. Such interventions can include surgical procedures, drug treatments, vaccines or changes in nursing care. Outcome measures are used heavily in support of quality improvement initiatives and comparative effectiveness research. Quality improvement involves the setting of goals and then comparing the performance of that organization (or clinic or provider) with those goals, identifying gaps in expectations. The knowledge of existing gaps is used to improve the processes of care. With comparative effectiveness research, the outcomes of several interventions are compared across large populations. Examples include the study of efficacy of two or more drugs and the comparison of frequency of adverse reactions to a drug in patients with different genotypes for drug metabolizing enzymes. To allow the widespread use of outcome measures in the age of the electronic health record, it is necessary to automate the process of computing them. This task is more challenging than it may seem. To accurately compute an outcome measure, it is necessary to create an operational definition that states precisely the algorithm used to compute the value of the measure. This involves the clear definition of specified data elements, data sources, and logic on how to handle unexpected conditions. It also involves close consultation with clinicians in order to ensure that the measure is actually calculating something meaningful. Only then will they be useful for quality improvement and comparative effectiveness research.