5. Screening and sampling Flashcards
Why is screening important?
one health and public health are focused on prevention
screening tests desirable qualities
condition screened needs to be important
screening test should have a high benefit cost ratio
condition needs to be sufficiently prevalent
an effective treatment should be avalible
screening test should have precision and accuracy
precision
ability of a measuring instrument to give consistent results on repeated trials
reliability
all shots in the same space not on bulls eye
accuracy
the ability of the measuring instrument to give a true interpretation of reality
validity
All shots hit bulls eye
precision and accuracy
its not possible to be imprecise and accurate
is possible for a measure to be precise and inaccurate
how to measure accuracy
sensitivity = ability to detect disease
specificity = ability to detect the absence of disease
positive predictive value = likelihood of disease giving positive result amount of true positive among positive results
negative predictive value = no disease giving negative results true negatives among all negative results
characteristic of good screening test
high in sensitivity high in specificity high in positive predicted value and high in negative predictive value
unealistic
realistic characteristics of screening test
tend to be very sensitive so you can be confident in negative result
realistic characteristic of confirmatory test
tend to be very specific minimizes false positive
can be confident in positive results
what is the gold standard?
the truth which we compare our tests to
definitive diagnostic test that considered best available
chart for gold standard
sensitivity equation
probability of the test being positive when the disease is actually present
true positives/ true positives + false negatives
specificity equation
the probability of a test correctly classifying an individual as disease free
true negatives / true negatives + false positives
positive predictive value equation
probability of a patient actually having a disease when the test is positive
true positives / true positives + false positives
negative predictive value
probability of a patient being healthy when the test is negative
true negatives / true negatives + false negatives
diagnoses are made based on what characteristics
clinical signs, history
regional prevalence of disease
presence of vectors
test attributes
high PPV
common disease –> positive result –> probability true positive = high PPV
low PPV
rare disease –> positive results –> probably false positive = Low PPV
why sample?
compared to using entire population at risk sampling
costs less
less field time
acceptable accuracy
when its impossible to study entire population
two sampling schemes
random sampling
non randome sampling
random sampling
every element from the entire population has an equal probability of being selected
non random population
probability of being chosen is unknown
less expensive
makes statistical analysis and error id more defficult
types of random sampling
simple random sampling
systemic random sampling
cluster sampling
multi stage sampling
types of non random sampling
convenience sample = sample is selected because they are easily accessible
snowball = friend
purposive sampling = you choose who you think
quota sampling
components of total deviation
sampling bias
sampled population is not representative of overall population
failure to allow for statistical weighing of samples of unequal popbability
sampling error
difference between survey result and population value due to random nature of sample
influenced by size and scheme of sample
measures of sample error
confidence limits
standard error
coefficient of variance
p values
others
sampling error used to measure
sample size prior to sample
determine how sure you are of analytic results
total deviation
total deviation = bias +error
bias vs sampling error
bias = is due to mistake than cane be avoided cannot be precisely meausred
sampling error = unavoidable when sampling can be calculated after sample
