## Ultra HighYield Review

Step By Step Guide To Mastering The Osce

Get Instant Access

Most USMLE Step 1 candidates probably spend no more than 3 to 5 hours reviewing biostatistics. In this short time, the candidate should be able to memorize the ultra-high-yield items in this checklist. Together with a background understanding from the previous chapters in this book, these items should equip the candidate to pick up a good number of points in a subject area that is neglected by many students and medical schools, which should mean that biostatistics in itself will be relatively a high-yield subject on the examination for the candidate. After this list and the referenced material in this book have been reviewed, a self-test can be administered by using the USMLE-style exercise questions at the end of each chapter.

The USMLE Step 1 candidate should:

□ be able to use the addition and multiplication rules of probability (page 3).

□ be able to find and use the three measures of central tendency (page 11).

□ understand the standard deviation (page 13).

□ know and be able to use the proportions of the normal distribution which are within or beyond 1, 2, or 3 standard deviations from the mean (page 13).

□ understand and be able to use ^-scores (page 15).

□ understand confidence limits and be able to find 95% confidence limits (page 25). C understand precision and accuracy (the dartboard analogy) (page 26).

□ understand how sample size relates to precision (page 26).

□ know exactly how to increase precision and reduce the width of the confidcncc interval (page 26).

□ know how to be 95% confident about the true mean of a population (page 26).

□ know the meaning and limitations of p values and statistical significance (page 37).

□ know the meaning of type I and type II errors in hypothesis testing and in diagnostic testing (page 37).

□ know how to avoid type I and type II errors in hypothesis testing (page 37).

□ know the meaning of a test's power, how to increase it, and the dangers of a lack of power (the radar screen analogy) (pages 37-38).

□ know the meaning of main and interaction effects in ANOVA (the beards and lipstick analogy) (page 44).

□ know the meaning and use of correlation coefficients and r values (page 50).

□ be able to avoid the temptation to infer causation from correlation (page 52).

□ be able to interpret scattergrams of bivariate distributions (page 51).

□ calculate and know the meaning of the coefficient of determination (r2) (page 52).

□ know what regression techniques do (pages 52-53).

□ memorize Table 4-1 to be able to choose the appropriate basic test for a given research question (page 54).

□ know the purposes of the features of clinical trials: (pages 59-60)

□ randomization or matching.

□ cohort studies (the Framingham study and chimney sweeps example) (pages 62-63).

□ case-control studies (DES and vaginal carcinoma study) (page 63).

□ be able to choose the appropriate type of research study for a given question (pages

□ know the meanings of and be able to find: (pages 68-69)

□ and the relationships between them (the epidemiologist's bathtub) (page

□ know the meaning of case-fatality and attack rates (page 70).

□ know the meanings of and be able to find: (pages 72-73)

□ relative risk and relative risk reduction.

□ number needed to treat.

□ know the meaning of validity (including internal and external validity) and reliability (page 78-79).

□ know the meanings of and be able to find:

□ positive predictive value (page 83).

□ negative predictive value (page 83).

□ know how changing a test's cutoff point will affect its sensitivity and specificity (page 82).

□ know what kind of test to use to rule in or rule out a disease (mnemonics "Snout" and "Spin") (page 81). 