EH1 week 4 Flashcards
Restriction is?
only have people in a certain age group, or only being male or female, or ethnic subpopulation, etc.
Matching is?
match a person in the study group to a person in the control group, except for the disease. (Mainly used in case- control studies).
Randomisation is?
allocating people to the study or control groups purely by chance. This equally distributes all potential confounding factors (those we know about, and those we don’t) across both arms of the study.
Selection bias is?
Refers to the way people are selected into the study.
Information bias is?
Occurs when information is collected
differently between two groups leading to
an error in making an association.
What prevents information bias in intervention studies?
Blinding (single-blinding, double-blinding [the better one])
Incidence is?
How is it measured?
looks at the number of new cases arising in a given period of time
number of new cases occurring over a period of time/ number of people at risk
Prevalence is?
how is it measured?
looks at the number of existing cases with a health factor in a given period of time
total number of cases/ total population
Interventional studies vs Observational studies
Interventional (The researcher intervenes to change reality and then observes what happens. Less subject to bias, but less reflective of what happens in real life)
Observational (The researcher studies, but does not alter what occurs. No intervention. More subject to bias, but more reflective of what happens in real life)
3 types of Descriptive studies
1) Case study (look at one study)
2) case series (look at multiple cases)
3) cross-sectional survey (look at exposure and disease at the same time. Report the prevalence [not incidence] of disease or risk factors)
2 categories of observational studies
Descriptive and Analytical observational
2 types of Analytical observational studies
Cohort studies: rare exposure -> Disease
(start with exposure)
Case-control studies: rare disease -> Exposure
(start with disease)
Confounding is?
A variable that is independently associated with both the presumed exposure and outcome (disease)
Confounding only happens if the confounding variables are unevenly distributed in both groups
(Ex: coffee drinking -> pancreatic cancer [confounding smoking])
how does Age-standardisation work?
how does Age-standardisation work?
Age-standardisation can apply the age-specific disease/event frequencies in two different populations to another standard (or reference) population. This will derive an age-adjusted summary measure of the disease/event frequency for each of the two different populations. These two age-standardised (or adjusted) measures can then be directly compared, as any differences in the age distribution between the two comparison populations has been removed as a confounding factor.
