Interpreting Association Flashcards
What are the 6 Bradford Hill criteria for evaluation
Temporality (most important)- exposure occurs before outcome
Dose-response- if u do more exposure you get more outcome
Strength- high relative risk (less likely to be due to chance)
Reversibility- if u take away the exposure then you’ve got less outcome
Consistency- association seen across different areas, different study designs, in different people
Biological plausibility- there is a way we could explain the association using sceinec
Also
Coherence (logical consistency with other info)
Analogy- similar to other cause and effect rships
Specificity - rship is specific to outcome of interest (not just all 40 yr old men get lung cancer, it’s all 40 yr old men who smoke get lung cancer
What 5 things can association be due to
Bias- systematic difference in comparison groups which could misrepresent the association
Chance
Confounding factors
Reverse causality (outcome leads to exposure)
True association
What are the three types of bias
Selection bias
Information bias
Publication bias
What is selection bias
A systematic error in selection of study participants or allocation of participants to different study groups
What is info bias
Error in measurement or classification of exposure or outcome eg observer bias, recall bias, wrongly calibrated instruments
What is publication bias
Trials with negative results less likely to be published
What is bias
Systematic error resulting in deviation from the true effect of exposure on outcome
What’s the difference between lead and length time bias
Lead time- early identification doesn’t alter outcome but appears to increase survival as it looks like they have the disease longer
Length time- diseases that progress more slowly are more likely to be picked up by screening so it looks like the screening prolongs live
What is a confounder
Apparent association between exposure and outcome is actually the result of another factor
