Bias and Causality

Random error: error that results from the measurement of data due to imperfections in the measurement system

  • To reduce random error, use statistics to understand error

Inherent Variability: natural differences that exist between individuals in a population

Bias: systematic errors that disproportionally affect the data in one direction or another

  • can’t use statistics to account for, can only try to minimize it

Internal validity: affected by bias, and is the methodology design of the study (variables, controls, samples, tools, etc.)

External validity: affected by bias, and is the generability of the study (is our sample reflective of our larger population)

Incidence: # of new cases during a period of time / # of people at risk at the start of the time period

Prevalence: # of cases at a certain time / total population at that time

Good practices for study design:

  • Make it short

  • Start with broad questions, then go to personal questions

  • Don’t ask leading questions

  • Don’t ask two questions in one

  • Include a good range of answers

Selection bias: impacts external (can manage by changing the target population) and internal validity (can’t manage, have to start the study all over again)

  • Ex: People who work are generally healthier than the overall population

Misclassification bias: measuring things incorrectly/ participants put in the wrong group

  • Ex: Putting the exposed or unexposed in the wrong group

Causation: A causes B

Correlation: A and B share a relationship

Principles of disease causality:

  • All cases of a disease have multiple cases

  • Not all causes act the same

  • Many collections of exposures that, when taken together, can cause disease

Correlation vs Causation:

  • Look at surveillance data to determine the issue

  • Identify if there are any potential correlations

  • Get statistics to help develop causation, and then conduct a controlled study

  • Check for association (is the association real or why isn’t it real?)

Sir Austin Hill’s Causation Considerations (best for infectious disease):

  • strength: strong correlation between variables

  • temporality: cause comes before the outcome

  • biological agents: as the level of exposure increases, so does the amount of disease

  • consistency: relationship is consistent across different studies, populations, times, etc.

  • specificity: single cause and effect

  • plausality: cause/effect relationship is biologically reasonable

  • coherence: relationship is consistent with previous knowledge

  • analogy: similar relationships observed with similar exposure/disease

  • experiment: interventions modify outcomes