STUDY DESIGN: Commentary. OBJECTIVE: To provide guidance for the design and interpretation of predictive studies of whiplash associated disorders (WAD). SUMMARY OF BACKGROUND DATA: Numerous studies have sought to define and explain the clinical course and response to treatment of people with WAD. Design of these studies is often suboptimal, which can lead to biased findings and issues with interpreting the results. METHODS: Literature review and commentary. RESULTS: Predictive studies can be grouped into four broad categories; studies of symptomatic course, studies that aim to identify factors that predict outcome, studies that aim to isolate variables that are causally responsible for outcome, and studies that aim to identify patients who respond best to particular treatments. Although the specific research question will determine the optimal methods, there are a number of generic features that should be incorporated into design of such studies. The aim of these features is to minimize bias, generate adequately precise prognostic estimates, and ensure generalizability of the findings. CONCLUSION: This paper provides a summary of important considerations in the design, conduct, and reporting of prediction studies in the field of whiplash.
AD - The George Institute for Global Health, University of Sydney, Sydney, Australia. skamper@george.org.au AN - 22020594 BT - Spine (Phila Pa 1976) ET - 2011/10/25 LA - eng M1 - 25 Suppl N1 - Kamper, Steven JHancock, Mark JMaher, Christopher GResearch Support, Non-U.S. Gov'tReviewUnited StatesSpineSpine (Phila Pa 1976). 2011 Dec 1;36(25 Suppl):S268-74. N2 -STUDY DESIGN: Commentary. OBJECTIVE: To provide guidance for the design and interpretation of predictive studies of whiplash associated disorders (WAD). SUMMARY OF BACKGROUND DATA: Numerous studies have sought to define and explain the clinical course and response to treatment of people with WAD. Design of these studies is often suboptimal, which can lead to biased findings and issues with interpreting the results. METHODS: Literature review and commentary. RESULTS: Predictive studies can be grouped into four broad categories; studies of symptomatic course, studies that aim to identify factors that predict outcome, studies that aim to isolate variables that are causally responsible for outcome, and studies that aim to identify patients who respond best to particular treatments. Although the specific research question will determine the optimal methods, there are a number of generic features that should be incorporated into design of such studies. The aim of these features is to minimize bias, generate adequately precise prognostic estimates, and ensure generalizability of the findings. CONCLUSION: This paper provides a summary of important considerations in the design, conduct, and reporting of prediction studies in the field of whiplash.
PY - 2011 SN - 1528-1159 (Electronic)0362-2436 (Linking) SP - S268 EP - 74 T2 - Spine (Phila Pa 1976) TI - Optimal designs for prediction studies of whiplash VL - 36 ER -