Bias

• patients sometimes must be exposed to a risk factor for aprolonged period of time before they develop a clinically detectable result
  • e.g., patients must smoke for many pack-years before bronchogenic carcinoma develops

• a third factor is either positively or negatively associated with both the exposure and outcome
• confounders arenot in the causal pathway
  • if not adjusted for, can distort true association
  • either towards or away from the null hypothesis

• randomization
  • ensures similar baseline characteristics between control and exposure/experimental groups
  • use intention-to-treat analysis to preserve randomization even if participants change study treatments
• matching
  • group similar participants into study pairs
• stratification
  • analyze in separate subgroups determined by a potential confounder
• restriction
  • only include groups with specific features in the sample
• adjustment
  • can only adjust for confounders that areknown and measureable
• crossover studies
  • subject acts as own control

• sampled population is not representative of the population researchers are trying to study
  • due to non-random selection of study participants
    
  • sampling (ascertainment) bias
    • certain individuals are more or less likely to be selected for a study group, leading to incorrect conclusions
    • non-response bias
      • e.g., participants who pick up the phone may be less sick than participants who don't
    • healthy worker effect
      • samples with employed subjects only may be healthier
    • volunteer bias
      • people who volunteer for a study may be different in some fundamental way from those who do not volunteer
  • late-look bias
    • patients with severe disease are less likely to be studied, because they die or are otherwise unavailable, making a disease look less severe
      • e.g., a group of HIV+ individuals are all asymptomatic
    • also can have opposite effect
      • e.g.,people with more mild disease are cured before the study takes place and only persistently sick folks are included in the study, making a disease seem more severe
  • Berkson bias
    • hospitalized study subjects are more likely to have a greater burden of illness than other possible subjects
  • attrition bias
    • those lost to follow-up may be different from those who remain in the study

• randomization
• include patients in multiple settings (outpatient, hospitalized)
• study designs that are longitudinal in nature rather than cross-sectional
• gather maximal information on participants

Measurement Bias

• information is gathered in a way that distorts the information or misclassifies study participants
  • interviewer bias
    • subjects in one group are interviewed in a different way than another
      • differences due to interviewing style disrepencies are falsely attributed to group differences

• standardize data collection

• subjects with the disease are more likely to recall the exposure of interest
  • e.g., parents of children with cancer recall exposure to a chemical

• reducing follow-up time in retrospective studies

• researchers treat groups differently or subjects alter their behavior/responses due to study group awareness
  • Hawthorne effect
    • subjects alter their behavior when they know they are being studied
  • procedure bias
    • researcher decides assigment of treatment versus control and assigns particular patients to one group or the other nonrandomly
    • patient decides assigment of treatment versus control

Lead-Time Bias

• subjects appear to survive longer when in reality their disease was detected earlier
  • common with improved screening
• e.g.,a cancer screening test is deemed to increase survival when in reality the disease was picked up earlier, increasing the time from detection to death

• use mortality rate instead of survival time in screening studies
• estimate lead time and add that to survival in unscreened group

• the control group is inappropriately non-comparable to the intervention group
  • allocation bias
    • difference in the way participants are placed in control versus experimental groups
    • e.g.,all zebras in control group and all lions in exposure group

• randomization
• matching

• observer bias (pygmalion effect)
  • investigator inadvertently conveys her high expectations to subjects, who then produce the expected result
    • a "self-fulfilling prophecy"
    • golem effect is the opposite: study subjects decrease their performance to meet low expectations of investigator
• confirmation bias
  • researcher ignores results that do not support their hypothesis
• response bias
  • participants do not respond accurately because they are concerned about the social desirability of their responses or misinterpret the question

• double blinding
• include positive and negative results

• outcomes are more likely to be detected in certain groups because of increased monitoring
  • e.g.,a certain skin disease being detected more often in hypertensive patients because they have more physician visits than non-hypertensive patients
  • researchers may falsely attribute hypertension to causing the skin disease