WEEK 6 disease prevention, strategies Flashcards
In what levels is Caplan’s prevention model divided? Explain?
- Primary prevention
Avois development of disease - Secondary prevention
Early disease detection, to prevent progression and symptoms - Tertiary prevention
(reduce complications, intensity, severity of current disease/care of dieases)
What is the target group of universal prevention?
Total population
What is the target group of selective prevention?
High risk groups
What is the target group of indicated prevention?
Early diagnosis
What is the target group of care-related prevention?
Diseased persons
Explain the prevention paradox
- Small absolute risk for individuals may have large impact on population health
- Large risk to population health may be a small risk for individual
What are 3 divisions in population prevention strategy? Examples?
Health promotion
- Healthy diet
- Smoking cessation
- Fluoride toothpaste
Enforcement/regulated
- Product modification
- Bans
- Vaccination
Health protection provisions:
- Safe roads
- Clean water
Explain the high risk approach, give example.
Prevention approach targeted at those who are at highest risk of getting a disease
Tailored to needs of the high risk group
- Weight loss programs to prevent diabetes
- Drug treatment of high cholesterol
What happens with the RR in the population strategy?
The RR is equally reduced among the whole population.
Distribution shifts towards the lower risk side
What happens with the RR in the high risk strategy?
The RR is only reduced among the group at highest risk
Distribution narrows (you get more people in the average risk)
For what type of disease is population strategy good vs high risk strategy?
Population:
Useful for common diseases that have a large impact on the general population
High risk:
Useful for targeting spcific exposures
Formula of sensitivity for diagnostic test? MEaning?
people true positives/# people with a disease
= probability of a positive test, given that the patient has the disease of interest
Formula of specificity of diagnostic test? Meaning?
people true negatives/# people without a disease
= probability of a negative test, given that the patient does not have the disease of interest
What questions to ask yourself to determine what makes your test best in terms of sensitivity vs specificity?
- Rather miss few ‘true positives’ or incorrectly identify people as diseased while they are not?
What is the positive vs negative predictive value? (PPV, NPV)
Positive predictive value: probability of disease in a patient with a positive test result
Negative predictive value: probability that a patient does not have the disease with a negative test result
PPV formula?
people true positives/# people tested positive
NPV formula?
people that are true negatives/# people with a negative test
What does the predictive value depend on?
- Depends on sensitivity and specificity of a test
- Depends on the prevalence of the disease in the population tested
What happens to the PPV when the prevalence is very low?
If the prevalence is low, the positive predictive value (PPV) may be very low.
Even when having high specificity and sensitivity, PPV can still be very low, making it not feasible to do this in practice.
What is the middle road prevention strategy? For what disease is it often used?
policy between population and high risk, where you want to focus on the population but also want the high-risk approach (e.g. focus on the top half of the population), often used in non-communicable chronic disease risk
In what situ is high-risk appr better choice?
When you do not want to subject whole population with a treatment to prevent something that is rare (benefit-to-harm trade-off)
- Rare disease
- Difficult prevention
Difference ecological vs cross-sectional study?
you look at group vs individuals
Analytical cross-sectional study = ?
study to inform ‘causality of risk factors’
What is block randomization?
Block randomization = stratified randomization, especially useful in small study populations where by chance alone an unequal distribution of a risk factor can occur
(separate by gender (stratify) > randomly assign treatment)
Reasons to use stratification?
- See whether there is confounding
- To see whether there is effect measure modification
How can you see from OR and adjusted ORthat something = confounder?
When adj OR is same as crude OR = no confounding
