Introduction to Medical Care Statistical Workshops for Health Services Research: Practical, State-of-the-Art Reviews

    loading  Checking for direct PDF access through Ovid


Understanding advances in medical care can lead to better health care policies, decisions, and population programs. As such, evidence-based research on planning, organizing, financing, providing, and evaluating health services requires state-of-the art qualitative and quantitative methods. In this issue of Medical Care, we are pleased to announce a new, recurring feature, Statistical Workshops for Health Services Research: Practical, State-of-the Art Methods and Reviews. The ultimate goal of this series is to improve the quality of health services research reported in the peer-reviewed literature by focusing on common and emerging statistical methods that may be particularly prone to misuse. We will focus on practical applications of statistical methods in health services research where even technically sophisticated but commonly used methods will be displayed to readers in a manner that makes them available and practical. To this end, this series will welcome presentations with practical recommendations and the inclusion of statistical software codes when appropriate to allow readers to apply the discussed methods in their work.
The inaugural manuscript in this series, “Meta-analysis of odds ratios: current good practices,”1 provides a template that matches our vision for future installments, and we would like to point out some important features of this work. First, the manuscript reviews past applications of the statistical methodology under consideration, pointing out important places where the optimal approach was not used. Second, recommendations for best practice are provided based on sound statistical principles. Third, the manuscript offers specific examples of the meaningful distortions that may occur from failure to use the best approach. Fourth, practical approaches to implementing the stated recommendations are offered, complete with code in different statistical languages. In addition, we believe that this manuscript contains the right amount of technical detail, providing, “just enough mathematics to HELP and NOT to HINDER.”2
Therefore, we are issuing a call for future manuscript proposals, with the recommendation that a concept paragraph be prepared for in-house review before proceeding with a full submission. All manuscripts in this series will undergo our usual strenuous process of peer review.
    loading  Loading Related Articles