mix Flashcards
(44 cards)
how to calculate Positive likelihood ratio and negative likelihood ratio?
PPV = sensitivity / (100-specificity) NPV = (100-sensitivity) / specificity
what is incidence?
number of new cases of disease in healthy population
How do you calculate relative risk
PPV / NPV
RR >1 treatment increases risk of outcome
RR <1 treatment decreases the risk of outcome
difference between odds and probability/risk?
odds is the the event occurring and not occurring.
Probability is the chance of an event occurring in relation to the options.
E.g. the odds and probability of having a girl or boy
odds 1:1
probability 0.5
what is type 1 error and how can it be reduced?
rejects a null hypothesis when it is true (false positive).
reduced by lower p value (increasing sample size)
what is type 2 error and how can it be best reduced?
failure to reject the null hypothesis when it is wrong (false negatives). it can be reduced by increasing the power of a study - population size, size effect, variance
what car seat should a 4-7 year old be in
back seat, forward facing, booster, seat belt
what car safety seat should a baby be placed in
back seat, rear facing, booster with 5-6 point harness
polysaccharide conjugate vaccines
Hib, meningococcal, pneumococcal
protein subunit vaccines / recombinant
Hep B, influenza, pertussis, HPV
toxoid vaccines
diptheria tenanus
which vaccines are inactivated/killed
polio, Hep A, rabies, Rotateq
what are the phases of clinical trials
phase 0 = pharmacokinetics / pharmacodynamics phase 1 = safety phase 2 = dose range, safety phase 3 = efficacy vs gold standard phase 4 = post marketing surveillance
what is intention to treat?
- analysis irrespective of adherence to treatment
- avoids crossover and dropout which may break random assignment
- provides information on potential effects of treatment.
how can you improve internal validity
randomisation, blinding, intention to treat, appropriate comparator, published protocol
how can you improve external validity
multi-centre, wide inclusion, limit exclusion criteria, selection of patients, differences protocol to routine practice.
how do you reduce performance bias
blinding
how do you reduce detection bias
double blinding
how do you reduce attrition bias
randominsation and intention to treat
what is the Cohen effect size
used in forrest plots
<0.2 small
0.5 mod
0.8 large
wha is chi squared in the forrest plot
assesses heterogeneity between studies
high value >50% = less sure the intervention will consistently have effect
what is an ecological study?
population based, not individual, decriptive
what is a case-control study?
retrospective, good for rare diseases
subject to recall bias and selection bias
confounding
what is a cross-sectional study?
good for prevalence
single time point
