|
|
|
| Assessing the Validity of Medical Information | |
|
|
|
|
When we retrieve or are presented with medical literature, it is critical that we assess its validity. Developing a uniform approach to critical appraisal is a challenge because of the different types of published research. To simplify matters, we have classified published medical research as follows:
Studies are considered descriptive if they "describe" individual variables (eg, the prevalence or characteristics of a particular disease), or analytic if they "analyze" the association between two or more variables (eg, an intervention and mortality). The majority of studies that we evaluate are of the analytic type. Chapter 8 presents a simple, novel, and comprehensive five-step study guide. Its uniform approach may be used to evaluate descriptive studies and all types of analytic studies (ie, studies of risk, diagnosis, intervention or prognosis). The unified approach is rooted in an explanation of a number of key concepts in epidemiology, medical decision-making and designing clinical research. Chapter 9 shows how this single study guide may be easily applied to an example of a study of a diagnostic test, an intervention, and prognosis. Reprints of the published articles representing each of these study types are included. Integrative literature is distinct in that it "integrates" information derived from individual studies. Integrative literature saves the clinician considerable time because it frames the clinical question, retrieves all the relevant information, appraises it, and integrates it within a particular framework to provide a summary finding or recommendation that may then be applied to the care of the patient. Each of these steps comprises the individual steps of evidence-based medicine. To ensure that the integrative literature is indeed evidence-based, it is critical that each of these steps be scrutinized to ensure that they were performed in the spirit of EBM. Chapter 10 provides an easy-to-use five-step integrative literature guide to evaluate overviews, meta-analyses, practice guidelines, decision analyses, and cost-effectiveness analyses. |