Lecture 9 - Review Flashcards
What is incidence?
Number of new cases of disease in a population over a period of time
What is prevalence?
Total number of people who have a particular disease or attribute over a specified period of time
Is case-control studies retrospective or prospective?
Retrospective
(Start with the case then work backwards to see if they had exposure)
What are some feartures of high quality stduueoos?
Large sample size
Prospective studies
Blinded
Random allocation of subjects to groups
Control group
Low loss to follow up
What is confounding?
External environmental variables which are independently associated with the exposuuure and the outcome but are not in the causal pathway
What is bias?
Systematic error in an epidemiological study that results in an incorrect estimate of the association between the exposure and the condition
How does the timing of case-control and cohort studies differ?
Case-control = retrospective
Cohort = prospective
How does the timing of case-control and cohort studies differ?
Case-control = retrospective
Cohort = prospective
How do case-control and cohort studies differ in terms of rare diseases?
Case-control = good for rare diseases
Cohort = good for rare exposure
How do case-control and cohort studies differ in terms of calculating incidence?
Case control = cannot calculate incidence
Cohort = can calculate incidence
How do case-control and cohort studies differ in terms of risk of bias?
Case control = risk of recall bias
Cohort = less risk of bbias
What is the positive predicted value?
Proportion of people with a positive test result that actually have the condition
You test positive, what are the chances you actually are positive
What is the negative predicted value?
Proportion of people with a negative test result that do not have the condition
You test negative, what are thee chances you actually are negative
What is sensitivity?
How good a test is at detecting the specific condition (So the TRUE POSITIVE RESULT)
So you have the condition, what is the chance the test says you have it
What is specificity?
How good a test is at only detecting the specifc condition (so TRUE NEGATIVE RESULT)
You are negative for the condition, what is the chances that the test says you are negative
What is risk ratio?
Risk of disease in the exposed/risk of disease in the unexposed
What type fo study are risk ratios used in?
Cohort studies
What is odds ratio?
Disease in the exposed/ disease in unexposed
What studies are odds ratios used in?
Case-control studies
What studies are odds ratios used in?
Case-control studies
What is meant by p<0.05?
Paper is 95% confident that the risk ratio lies in between the confidence interval
What is the null value for risk ratios?
1
What is the significance of p<0.05 and the value 1 is not in the 95% confidence interval?
We can reject the null hypothesis so the data is statistically significant
Look at slide 14 , then the confidence interval, is this statistically significant?
The value 1 lies outside of the confidence interval, with 95% certainty we can reject the null hypothetical and women aged 18-45 are at a 2.5x greater risk of DVT over 5 years if they take the COCP compared to those who dont
Comment on, reducing risk of MU in patients
RR - 0.32 0.17 - 0.79 (p<0.05)
The results are statistically significant since theres the null value of 1 is not within the 95% confidence interval and p<0.05 so Atorvastatin reduces risk of MI in patients witht high cholesterol by 68%
