P.Mean: Tentative training schedule (created 2009-08-31, revised 2009-09-12).

I've been asked to develop a series of training classes, which I may also use as a series of monthly webinars (web seminars). Here's a first draft. All classes are two hours long. Classes are self-contained and do not need to be taken in the order listed below. There are no prerequisites other than the ability to use a pocket calculator.

October 2009 : What do all these numbers mean? P-values and confidence intervals and the Bayesian alternative. Based on Chapter 6 of Statistical Evidence and on www.cmh.edu/stats/training/hand22.asp and new material. In this class you will learn how to:

• distinguish between statistical significance and clinical significance;
• define and interpret p-values;
• explain the ethical issues associated with inadequate sample sizes.
• explain the difference between informative and diffuse priors;
• interpret statistics from a posterior distribution.

November 2009: What do all these numbers mean? Sensitivity, specificity, and the likelihood ratio. Based largely on Chapter 6 of Statistical Evidence and on www.cmh.edu/stats/training/hand21.asp and www.cmh.edu/stats/training/hand24.asp. In this class, you will learn how to:

• compute sensitivity and specificity;
• identify the problems with diagnosing a rare disease;
• understand which tests are useful for ruling in or ruling out a disease.

Based largely on Chapter 6 of Statistical Evidence and on www.cmh.edu/stats/training/hand21.asp and www.cmh.edu/stats/training/hand24.asp.

December 2009: What do all these numbers mean? Odds ratios, relative risks, and number needed to treat. Based largely on Chapter 6 of Statistical Evidence and on www.cmh.edu/stats/training/hand23.asp. In this class you will learn how to:

• list the advantages of the odds ratio;
• list the advantages of the relative risk;
• compute and interpret the number needed to treat.

January 2010: What do all these numbers mean? Linear and logistic regression coefficients. Based largely on Chapter 6 of Statistical Evidence and on www.cmh.edu/stats/training/hand25.asp. In this class you will learn how to:

• Interpret the slope coefficient in a linear regression model;
• Identify problems with extrapolation in linear regression;
• Explain why the logistic regression model uses log odds units.

February 2010: Evidence Based Medicine: Randomization. Based largely on Chapter 1 of Statistical Evidence and on www.cmh.edu/stats/training/hand32a.asp. In this class you will learn how to:

• create a randomized list of treatments;
• understand the practical and ethical limitations to randomized studies;
• describes three variations on randomization.

March 2010: Evidence Based Medicine: Observational studies. Based largely on Chapter 1 of Statistical Evidence and on www.cmh.edu/stats/training/hand32b.asp. In this class you will learn how to:

• contrast the selection process in a cohort design with a case-control design;
• describe the limitations of a case-control design for evaluating a diagnostic test;
• appraise the extent of temporal bias in the historical-controls design.

April 2010: Evidence Based Medicine: Statistical adjustments to control for bias. Based largely on Chapter 1 of Statistical Evidence and on www.cmh.edu/stats/training/hand32c.asp. In this class you will learn how to:

• apply matching and stratification to prevent covariate imbalance;
• discuss the strengths and weaknesses of crossover studies;
• explain how covariate adjustment and weighting work to reduce or remove bias.

May 2010: Evidence Based Medicine: Exclusions and dropouts in clinical trials. Based largely on Chapter 2 of Statistical Evidence and on www.cmh.edu/stats/training/hand33.asp. In this class you will learn how to:

• assess the impact of dropouts in a research study;
• describe how intention to treat is used in studies with compliance issues;
• explain how the ethical need for informed consent research can limit generalizability.

June 2010: Evidence Based Medicine: Clinical significance. Based largely on Chapter 3 of Statistical Evidence and on www.cmh.edu/stats/training/hand34.asp. In this class you will learn how to:

• explain how a narrow focus provides better quality evidence;
• assess the impact of post hoc protocol changes;
• evaluate the quality of outcome measures;
• recognize the ethical and practical limitations of a study when the sample size is too small;

September 2010: Evidence Based Medicine: Bias and fraud in research studies. Based on new material and on Chapter 4 of Statistical Evidence and on www.cmh.edu/stats/training/hand36.asp and www.cmh.edu/stats/training/hand72.asp. In this class you will learn how to:

• document examples where conflicts of interest have led to biased research results;
• list the important elements of a conflict of interest statement;
• identify measures to protect against fraudulent research.

October 2010: Evidence Based Medicine: Publication bias in systematic overviews. Based largely on Chapter 5 of Statistical Evidence and on www.cmh.edu/stats/training/hand35.asp. In this class you will learn how to:

• recognize sources of heterogeneity in meta-analysis;
• explain graphical methods for identifying heterogeneity;
• understand methods to control for heterogeneity.

November 2010: Evidence Based Medicine: Heterogeneity in systematic overviews. Based largely on Chapter 5 of Statistical Evidence and on www.cmh.edu/stats/training/hand35.asp. In this class you will learn how to:

• define publication bias;
• identify methods to control for publication bias;
• explain the ethical concerns with failure to publish and with duplicate publication.

December 2010: Your first three steps in writing a research protocol. Based largely on www.cmh.edu/stats/training/hand42.asp. In this class you will learn how to:

• identify various research designs and their practical limitations;
• implement blinding and randomization in a research study;
• calculate an appropriate sample size;
• identify ethical issues associated with randomization and blinding.

January 2011: Your first three steps in selecting a sample size. Based largely on www.pmean.com/09/AppropriateSampleSize.html. In this class you will learn how to:

• define a testable hypothesis;
• find an appropriate estimate of variation for your outcome measure;
• identify the minimum difference that still has clinical significance.

February 2011: Your first three steps in data entry and documentation. Based largely on www.cmh.edu/stats/training/hand01.asp. In this class you will learn how to:

• structure your data so that is amenable for data analysis;
• identify special problems with date values.

March 2011: Your first three steps in a descriptive data analysis. Based largely on www.cmh.edu/stats/training/hand02.asp. In this class you will learn how to:

• distinguish between categorical and continuous variables;
• compute ranges and frequencies;
• examine relationships among variables.

April 2011: Your first three steps in a linear regression analysis. Based largely on www.cmh.edu/stats/training/hand03.asp. In this class you will learn how to:

• interpret the slope and intercept in a linear regression model;
• compute a simple linear regression model;
• make statistical adjustments for covariates.

May 2011: Your first three steps in a logistic regression analysis. Based largely on www.cmh.edu/stats/training/hand04.asp. In this class you will learn how to:

• compute and interpret simple odds ratios;
• relate the output of a logistic regression model to these odds ratios;
• examine the assumptions behind your logistic model.

June 2011: Your first three steps in a survival data analysis. Based largely on www.cmh.edu/stats/training/hand05.asp. In this class you will learn how to:

• define censoring;
• apply the Kaplan-Meier estimate of the survival function;
• compare survival times using the log-rank test.