Measures Of Diease Flashcards
3 types of disease outcome
Continuous
Binary
Binary time-to-event (e.g. time to death or time to remission)
Define prevalence, and what type of study design is usually used to measure this?
- proportion of ppl in pop w disease at a specified time
-cross-sectional studies
What’s the true value and is it possible to get. hence, what’s used instead ?
Infinite sample value
We use 95% confidence interval around estimate of value - we are 95% sure that true value lies within this range.
What are odds, and what study design is used to measure it?
Odds = no. Ppl w disease / no. Ppl w/out disease
- case-control studies
Odds can be calculated for what outcome type?
Can it account for confounders (logistic regression)?
Binary outcomes
Define risk
No. People experiencing outcome in defined period / no, ppl in study pop at start period
What kind of data is person-years at risk measured from ?
Define it
Time-to-event data
Sum of no. Years (from all ppl) in which individuals at risk of event have been followed
What’s rate
No. Ppl experiencing event in defined period / person-years at risk unit = per 100 person-years or 1000
- always positive
E.g. death rate
Why is rate better than risk? What are we enabled to do by using person-years at risk as the denominator to calculate risk?
Define ’censored’ in terms of rate?
What is the relationship between prevalence and risk?
Define exposure and give 2 types of examples?
Define prevalence ratio, give equation and what does ratio of 1 mean?
What’s odds ratio? Give 2 equations and what does ratio of 1 mean?
What’s risk ratio? Give equation, and what does risk ratio of 1 mean?
What does risk ratio below 1 mean?
What’s another name for percent risk reduction?
What does risk ratio above 1 mean?
What’s another name for rate ratio?
Equation ?
Ratio rate of 1 indicates what ?
