It is imperative to incorporate current research evidence into all aspects of physical therapy, including clinical practice, education, and administration. A major difficulty in performing this task is efficiently searching and selecting relevant research articles from the current literature. Research evidence is one factor used in evidence-based decision making along with clinical expertise and patient preferences.1 The components of evidence-based decision making are illustrated in Figure 1 . Systematic review articles and other forms of preappraised evidence are extremely helpful in efficiently gathering research information.2,3 Systematic reviews “…synthesize the results of multiple primary investigations by using strategies that limit bias and random error. These strategies include a comprehensive search of all potentially relevant articles and the use of explicit, reproducible criteria in the selection of articles for review.”2 The purpose of this article is to provide an overview of methods for efficiently searching for systematic reviews and other types of preappraised evidence.

Regardless of the specific scenario, evidence-based decision making involves several steps. The first step is to identify a gap in current knowledge. The next steps are to formulate a clinical question and search for the best evidence. Then the acquired evidence must be evaluated individually and as a whole. The final step in evidencebased decision making is to apply the evidence to patient care.2,3 The steps in evidence-based decision making are illustrated in Figure 2. There are several techniques and unique databases that facilitate searching for systematic reviews and other types of preappraised evidence.

IDENTIFY GAP IN CURRENT KNOWLEDGE

Identifying gaps in current knowledge can originate from different sources and have different applications. Recognizing need for information can come from several sources including professional practice, professional trends, existing published research, and existing theory.6 In addition, identifying gaps in current knowledge may apply to one of 3 areas of patient management, diagnosis, treatment, or prognosis. Consider this example, “Is physical therapy beneficial for patients with pulmonary disease?” The source of this gap in knowledge is professional practice and it applies to patient treatment or intervention.

FORMULATE A CLINICAL QUESTION

A clinical question must contain 4 components to be answerable. These components are patient population (P), intervention or exposure (I), comparison (C), and outcome (O).1 The components of a clinical question are illustrated in Figure 3. The previously posed question needs to be refined in order to contain and narrow all the needed components of a good clinical question. First, patient population must be defined and narrowed, for example, ‘patients with cystic fibrosis! Next, an intervention must be defined and narrowed, for example, ‘exercise training! A comparison group must be included in the clinical question, which could be either a different intervention group or a no treatment/control group. Lastly, an outcome must be defined and narrowed, for example, ‘exercise tolerance.’ Once a clinical question has been refined a literature search can be undertaken.

SEARCH FOR BEST EVIDENCE

Databases

General databases that include both primary and preappraised literature citations can be used to gather best evidence. PubMed and Ovid are examples of database search tools that include both primary and preappraised citations. PubMed is a search tool that contains several databases including MEDLINE. MEDLINE contains over 11 million records from over 4,000 journals dating back to 1966. MEDLINE is a very inclusive database that contains original trials and preappraised articles and is considered one of the best sources of recent evidence. The primary disadvantage of using MEDLINE to obtain evidence is that it does contain a large number of citations and therefore many limits and strategies are needed to refine a search. Using primary literature can be time consuming to individually appraise quality of research studies and draw conclusions across multiple studies. There are several search strategies that can be used to locate systematic reviews and other preappraised sources of information in MEDLINE. The United States Library of Medicine provides free access to PubMed at www.ncbi.nlm.nih.gov/pubmed. or http://www.ncbi.nlm.gov/entrez.7 Ovid is another database provider for a collection of health and medical subject databases. It includes MEDLINE, Cumulative Index to Nursing and Allied Health (CINAHL), HealthStar, and Evidence Based Medicine (EBM) reviews. CINAHL contains over 7,000 records from more than 1,200 medical and health-related journals, including the Cardiopulmonary Physical Therapy journal.6 Ovid has similar advantages and disadvantages as PubMed, with the additional feature that it allows simultaneous searching of up to 5 databases. Many of the citations found in Ovid also have full-text access. Ovid is provided by subscription at gateway2.