go back (01aug2003/27jan05)
DrTim homepage link

CH1 Introduction

Intro to Clinical Statistics by Timothy Bilash MD
August 2003

based on:
Review of Basic & Clinical Biostatistics by
Beth Dawson, Robert Trapp (2001) CH1

  1. Three types of investigation1*

    1. Scientific Investigation (Experiment)
      1. test hypothesis for consistency
      2. control conditions (expose all) or use control groups (compare to non-exposed)
      3. randomize to improve accuracy when cannot control exposure
      4. all exposure risks must be known (controlled and/or randomized)
      5. all outcomes results must be observed (controlled and/or randomized)
      6. descriptive statistics summarize results within known group.
      7. look for changes in outcome for an individual within group
      8. most accurate means we have to determine causes and effects

    2. Evidence-Based Medicine (BioStatistics)
      1. uses sampling statistics to determine average behavior and correlations
      2. exposure (risk) is not known for sure (controlled), but determined statistically
      3. outcome (result) is not known for sure (controlled), but determined statistically
      4. uses inferential statistics to determine which variables are exposures and which are outcomes
      5. not necessarily causal, because unable to control after the fact and non-random for many factors. in particular, there is expsoure bias, sampling bias and time-dependent effects.
      6. uses statistics to determine behavior of the group as if they were all the same individual

    3. Consensus (Popularity)
      1. determines what is correct by how many believe it, or what is assumed to be how many believe it (Gossip)
      2. bias very common, creates consistency by arbitrarily excluding data, very unreliable
      3. those who control the information and promote the information determine its accuracy

  2. Biostatistics
    1. mathematics
    2. using statistics to
      1. summarize data
      2. reach certain conclusions or absence of conclusions
      3. determine the reliablity of results thru statistical techniques

  3. Epidemiology
    1. study of health and illness in human populations (groups)
      1. patterns of health or disease and the factors that influence these patterns
    2. evidence-based medicine is NOT science
      1. also called clinical epidemiology
      2. the application of population-based information to decision making about individuals
    3. evidence-based medicine is mathematics
      1. math functions and summaries
      2. outcomes not directly linked causally to an individual, but to the population summary of many individuals
        1. requires proper summary
        2. requires inferring application to an individual via membership in the group
      3. uses bio-statistics

  4. Errors in published reports
    1. 1986 Williamson et al
      1. assessed 30 journals
      2. only about 20% of 4235 research reports met criteria for validity
      3. subset of 8 articles
        1. inadequately designed and analyzed reported 80% positive findings
        2. properly designed and analyzed had only 25% positive findings
        3. thus poor studies more likely to find effect
    2. more recent articles indicate that the problems have not improved
      1. Avram

  5. Newer statistical concepts
    1. number needed to treat (to prevent one undesirable outcome)
    2. likelihood ratio (analyze risk factors to outcomes)
    3. odds ratio (predict outcomes from risk factors)
    4. power (number of subjects needed to detect and effect)
    5. risk
    6. multivariate methods
      1. statistically (mathematically) complex
        1. logistic regression
        2. cox proportional hazard model

    (*1 Author's Note revised 01.27.2005)



    [an error occurred while processing this directive]page views since Sept2007