Measures of association/ association and causation

• Risk refers to the probability of an untoward event
• Used to refer to the likelihood that persons who are without disease but exposed to certain factors will acquire the disease
• Risk factors are associated with an increased risk of acquiring disease
• Sources of risk factors include:

  • Environmental factors such as toxins, drugs, and temperature
  • Biological factors like infectious agents and vectors of diseases
  • Sociocultural factors like culture and family disruption
  • Behavioral factors such as smoking and lung cancer, alcohol and liver cirrhosis
  • Genetic factors like BRECA and HLA
• Risks can be classified as:

  • Modifiable
  • Non-modifiable
  • Intrinsic
  • Extrinsic
  • Biologic/genetic
  • Chemical
  • Physical
  • Environmental
• Exposure is the contact a person has with the agent that can cause a disease
• Uses of information about risk include prediction of future incidences of a disease, diagnosis, cause determination, and prevention
• Studies for determination of risks include observational studies like case-control studies and cohort studies
• Attributable risk is the incidence of the disease attributable to the exposure
• Relative Risk Ratio indicates how many times more likely exposed persons are to become diseased relative to non-exposed
• Population Attributable Risk is the excess incidence of disease in a population associated with the occurrence of a risk factor
• Population Attributable Fraction is the proportion of the disease in the population which can be ascribed to the risk factor
• The p value indicates the likelihood of obtaining a statistical estimate by chance
• A p value of 0.05 and below shows statistical significance
• The confidence interval gives a range of values where the true estimate of a characteristic or parameter lies
• Interpretation of a confidence interval depends on the parameter estimated and the width of the interval
• Association is the statistical dependence between two variables
• Types of associations include artifactual, interactions, and causal associations
• The concept of causation includes necessary cause, sufficient cause, postulates by Henle Koch and Evan, and the assessment of causality
• Criteria for causality include strength of association, consistency, specificity, temporal sequence, dose-response relationship, biologic plausibility, and experimental evidence
• Analytical epidemiology quantifies the association between exposure and outcome to establish causal relationships and identify effective interventions
• Comparison is key in epidemiologic analysis
• In cohort studies, the difference is between the frequency of disease in exposed persons and non-exposed persons
• In case-control studies, the difference is between the frequency of the risk factor in case participants (persons with the disease) and control participants (persons without the disease)
• Relative measures estimate the strength of association between exposure and outcome
• Relative measures include prevalence ratio, risk ratio, odds ratio, and incidence rate ratio
• Relative risk measures indicate how much more likely an exposed individual is to develop the outcome compared to an unexposed individual
• Relative risk >1 indicates a positive association, =1 means no difference in risk, and <1 indicates a negative association
• A 2x2 table is used for calculations of measures/tests of association and epidemiological data
• The table has two rows and two columns representing exposure and disease status
• Relative risk (Risk Ratio) is calculated as the risk of the outcome in the exposed group divided by the risk in the unexposed group
• Odds ratio is the ratio of the odds of the outcome in the exposed group to the odds in the unexposed group
• Attributable risk measures the additional risk of disease following exposure over that of non-exposed individuals
• Attributable risk is the difference between the incidence rates in exposed and non-exposed groups
• Attributable risk percent expresses the proportion of disease in the exposed group attributable to the exposure
• Population attributable risk estimates the excess rate of disease in the total study population attributable to the exposure
• Population attributable risk is calculated as the rate of disease in the total population minus the rate in the unexposed group
• Population attributable risk percent expresses the proportion of disease in the study population attributable to the exposure