Scribd Logo
Upload

Welcome to Scribd!

  • 287 views

    Errors in Epidemiology

    Uploaded by

    Dilshan Wickramanayaka
    This document discusses types of errors in epidemiological studies, including random error and systematic errors. Random error is due to chance and can be measured, while systematic errors include bias and confounding. Bias is a systematic error that results in an incorrect estimate of the association between exposure and disease risk. Confounding is when a variable is independently associated with both the exposure and outcome. Strategies to minimize confounding include restriction, matching, randomization, stratification, and regression analysis.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Errors in Epidemiology

    Uploaded by

    Dilshan Wickramanayaka
    287 views13 pages
    This document discusses types of errors in epidemiological studies, including random error and systematic errors. Random error is due to chance and can be measured, while systematic errors include bias and confounding. Bias is a systematic error that results in an incorrect estimate of the association between exposure and disease risk. Confounding is when a variable is independently associated with both the exposure and outcome. Strategies to minimize confounding include restriction, matching, randomization, stratification, and regression analysis.

    Original Description:

    ERRORS IN EPID

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses types of errors in epidemiological studies, including random error and systematic errors. Random error is due to chance and can be measured, while systematic errors include bias and confounding. Bias is a systematic error that results in an incorrect estimate of the association between exposure and disease risk. Confounding is when a variable is independently associated with both the exposure and outcome. Strategies to minimize confounding include restriction, matching, randomization, stratification, and regression analysis.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    287 views13 pages

    Errors in Epidemiology

    Uploaded by

    Dilshan Wickramanayaka
    This document discusses types of errors in epidemiological studies, including random error and systematic errors. Random error is due to chance and can be measured, while systematic errors include bias and confounding. Bias is a systematic error that results in an incorrect estimate of the association between exposure and disease risk. Confounding is when a variable is independently associated with both the exposure and outcome. Strategies to minimize confounding include restriction, matching, randomization, stratification, and regression analysis.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 13

    Errors in

    Epidemiological
    studies
    Dr. Jag
    ath Am
    Consu arasek
    ltant E era
    Epidem pi de m
    iology iologis
    Unit t
    Types of errors
    Random error
    - Due to chance
    - Can be measured

    Systematic errors
    - Bias
    - Confounding (sometimes called
    confounding bias)
    Why random error???
    Normal distribution of variables
    in the population
    Inference given for population
    based on studying a sample
    Can be calculated
    Role of chance
    In g
    eneral
    sampl , samp
    e size ling er
    S am increa ror de
    ple siz s e s c reases
    e c as the
    Pro alculati
    babilit o n is im
    (P va y of a va portant
    lue) l u e o c cu
    Gen rring d
    erally ue to c
    P valu hance
    e is ca
    lculate
    d for 5
    % and
    1%
    Systematic errors

    Bias and confounding


    Role of systematic errors
    cannot be measured
    They can be minimized
    Bias

    Systematic error in an epidemiological


    study that results in an incorrect estimate of
    the association between exposure and risk of
    disease.
    Possible bias should be identified and minimized
    when designing a study
    of B i as
    Examp l es
    n ti a l
    e si de s
    n pr kd a y
    e r ic a w e e
    m
    : A study o n
    e g s)
    b ia s o u s e r b ia
    c ti o n to h e r ve
    le s e b s
    Se er, hou s ( e g : o
    win n n b ia
    a ti o
    fo r m
    In l b i a s
    e c al w u p
    R o fo l lo
    s t
    Los
    Confounder
    Confounder is a variable which is
    independently associated with the
    exposure (risk factor) and the outcome
    Minimize confounding
    Design stage:
    Restriction
    Matching
    Randomization (in intervention studies only)
    for unknown confounders as well
    Analysis stage:
    Stratification
    Multiple variable regression analysis
    Cohort study assessing obesity as a risk factor for CVD

    Age50 Age50
    CVD NoCVD Total CVD NoCVD Total
    Obese 10 90 100 Obese 36 164 200
    Not Not
    35 465 500 25 175 200
    Obese Obese
    Total 45 555 600 Total 61 339 400
    Effect modifier
    Effect modification : a variable that
    differentially (positively and
    negatively) modifies the observed
    effect of a risk factor on disease
    status.
    Different groups have different risk
    estimates when effect modification is
    present
    Define validity???
    Define reliability???

    You might also like