Practical Statistics for Medical Research
Most medical researchers, whether clinical or non-clinical, receive some background in statistics as undergraduates. However, it is most often brief, a long time ago, and largely forgotten by the time it is needed. Furthermore, many introductory texts fall short of adequately explaining the underlying concepts of statistics, and often are divorced from the reality of conducting and assessing medical research.
Practical Statistics for Medical Research is a problem-based text for medical researchers, medical students, and others in the medical arena who need to use statistics but have no specialized mathematics background.
The author draws on twenty years of experience as a consulting medical statistician to provide clear explanations to key statistical concepts, with a firm emphasis on practical aspects of designing and analyzing medical research. The text gives special attention to the presentation and interpretation of results and the many real problems that arise in medical research.
What people are saying - Write a review
Types of data
Using a computer
Preparing to analyse data
Principles of statistical analysis
Comparing groups continuous data
Analysis of survival times
Some common problems in medical research
The medical literature
Appendix A Mathematical notation
Appendix B Statistical tables
Answers to exercises
allocation analysis of variance approach assess association assumption average bilirubin Binomial distribution blood glucose blood pressure calculate cancer case-control study centiles Chapter clinical trials comparison confidence interval correlation coefficient data in Table degrees of freedom described in section discussed disease effect estimate example expected frequencies formula give given hazard ratio histogram hypothesis is true hypothesis test important indicate individuals interest interpretation linear logrank test mean and standard measurements median multiple regression non-parametric Normal distribution Normal plot null hypothesis obtained outcome paired patients placebo population possible predict prediction interval probability problem proportion random ranks reasonable regression analysis regression line regression model relation residuals risk scores serum albumin shown in Figure shows skewed squared distribution squared test standard deviation standard error statistical analysis statistical methods statistically significant test statistic tion Total transformation treatment trend usually values variables variation women
Page 581 - Martin TR, Bracken MB. The association between low birth weight and caffeine consumption during pregnancy. Am J Epidemial.
Page 584 - The use of aspirin to prevent pregnancy-induced hypertension and lower the ratio of thromboxane A2 to prostacyclin in relatively high risk pregnancies.