A Clinician's Guide to Statistics and Epidemiology in Mental Health. Measuring Truth and Uncertainty

Description

Accessible and clinically relevant, A Clinician’s Guide to Statistics and Epidemiology in Mental Health describes statistical concepts in plain English with minimal mathematical content, making it perfect for the busy health professional. Using clear language in favour of complex terminology, limitations of statistical techniques are emphasized, as well as the importance of interpretation - as opposed to ‘number-crunching’ - in analysis. Uniquely for a text of this kind, there is extensive coverage of causation and the conceptual, philosophical and political factors involved, with forthright discussion of the pharmaceutical industry’s role in psychiatric research. By creating a greater understanding of the world of research, this book empowers health professionals to make their own judgments on which statistics to believe - and why.

• An accessible overview of statistics without the jargon or mathematics • Tailored to the needs of mental health professionals • Explains when and when not to believe statistic

Contents

Preface; Acknowledgements; Part I. Basic Concepts: 1. Why data never speak for themselves; 2. Why you cannot believe your eyes: the three C's; 3. Levels of evidence; Part II. Bias: 4. Types of bias: 5. Randomization; 6. Regression; Part III. Chance: 7. Hypothesis testing: the dreaded p-value and statistical significance; 8. The use of hypothesis testing statistics in clinical trials; 9. The alternative effect estimation; Part IV. Causation: 10. What does causation mean?; 11. A philosophy of statistics; Part V. The Limits of Statistics: 12. Evidence-based medicine: defense and criticisms; 13. The alchemy of meta-analysis; 14. Bayesian statistics: why your opinion counts; Part VI. The Politics of Statistics: 15. How journal articles get published; 16. How scientific research impacts practice; 17. Dollars, data, and drugs; 18. Bioethics and the clinician/researcher divide; Appendix. Regression models and multivariable analysis.