other Flashcards
precision - aka
accuracy - aka
precision = consistency = reproducibility = reliability accuracy = truness = validity
accuracy vs precision according to errors
accuracy –> systemic
precision –> random (explain by chance - unpredicable)
Metanalysis disadvantage
while it pools together the data from many studies, it also pools together the biases and limitations of those individuals studies
case fatality rate
divide number of fetal cases by the the total number of people with this disease
Hawthorne effect
(aka observer effect) is the tendency of study subjects to change their behavior as a result of their awareness that their being studied
how to minimize Hawthorne effect
study subjects can be kept unaware that they are being studied –> this can occasionally pose ethical problems
Health promotion?
according to WHO –> the process of enabling people to increased control over their health + its determinants and thereby improve their health
examples: dietary habits, exercising regularity, no smoking, lossing weight if needed
FALLS UNDER PRIMARY PREVENTION
Detection bias
the risk factor itself may lead to extensive diagnostic investigation and increase the probability that a disease is identified
Sampling bias
non-random sampling selection
Ecological study
like cross sectional but in populations (not in individuals)
t-test types (explain)
- 2 sample T test –> aka Student’s test –> compare means of 2 independed groups
- Paired T test –> depended –> compare 2 means from the same individuals
Fisher’s exact test
like X square test but in smaller sample
less than 10 people in each cell
anova - types (explain)
- 1 way analysis –> 1 variable (weight loos mean in 3 dif groups)
- 2 ways analysis –> 2 variables (eg. weight loss man in 3 dif groups + men vs women
ROC (receiver operating characteristic) def + expl
graphic with sensitiviy at y + 1-specif (FP rate) at x for a diagnostic test
explanation –> the closer the curve to the diagonia, the less discrimination ability of the test. The closer to the y axis, the better discrimination
variables - types and definition
independent –> an experimenter can change it (salt in diet)
depended –> outcomes that reflex to the change (blood pressure)
counfounding bias strategy of reduction
- multiple repeated studies
- crossover studies
- matching
- restriction
- randomization
- standard deviation vs standard error of a mean (SEM)
- how much variability from the mean in a set of values
2. how much variability exits between the sample mean + the true population
case series studies
descriptive study that tracks patients with a known condition (eg. particular exposure, risk factor, disease) to document natural history or response to treatment (qualifying study that cannot quantify statistical significance)
interquartile range
difference between the values corresponding to the 25th and 7th percentile
Standardised mortality ratio
observed number of deaths / expected number of deaths
maternal mortality rate
maternal deaths / live births
cause specific mortality rate
number of deaths from a particular disease / total population
- crude birth rate
2. crude mortality
- number of live birth / total population
2. number of live death / total population
attrition bias?
in prospective studies, if loss to follow-up occurs disproportionately between the exposed + unexposed groups –> attrition bias can result if the lost subjects differ in their risk of developing the outcome compared to the remaining sugjects
(IT IS A FORM OF SELECTION BIAS)
attrition bias does not occur when the losses happen randomly between the exposed + unexposed groups