BIOSTATS Flashcards
The likelihood ratio (LR)
The probability of a patient with the disease having a particular finding (eg, positive or negative test result) divided by the probability of a patient without the disease having the same finding.
LR+ = sensitivity / (1 − specificity)
LR− = (1 − sensitivity) / specificity
Screening test
A screening test must have a high sensitivity. This high sensitivity helps to ‘RULE OUT’ the disease by decreasing the number of false-negative results, and byincreasing the negative predictive value (SnNOut)
More specific diagnostic tests have lower false-positive rates than less specific tests.
COHORT
A cohort study is an analytical study design that estimates and compares the risk of developing different health outcomes between exposed and nonexposed groups. It is particularly useful in studying rare exposures.
CROSS SECTIONAL
A cross-sectional study takes a snapshot of a population to estimate the prevalence of risk factors and disease simultaneously. This design cannot estimate and compare the incidence of different health outcomes.
OBSERVER BIAS
Observer bias generally occurs in the absence of blinding, when observers misclassify data due to individual differences in interpretation or preconceived expectations regarding the treatment. Observer bias is particularly likely when the studied outcome is qualitative.
Observer bias can be effectively reduced by using the blinding technique.
External validity
answers the question, “How generalizable are the results of a study to other populations?” For example, a study in middle-aged women would not be necessarily generalizable to elderly men.
Internal validity
Relates to conclusions regarding cause and effect in a study and answers the question, “Are we observing/measuring what we think we are observing/measuring?” The major threat to internal validity is confounding. In this study, randomization improves internal validity by balancing the distribution of confounding variables among the groups.
Post-marketing surveillance is the practice of
monitoring the safety of medications or devices after they have been released on the market. This is due to the fact that clinical trials often have short follow-up times, underrepresented populations, and too few participants to detect rare and serious side effects = inadequate power
The basic premise of the intention-to-treat
principle is that participants in trials should be analyzed in the groups to which they were randomized, regardless of whether they received or adhered to the allocated intervention and regardless of whether they withdrew from treatment.
type 2 error
The failure to detect a difference between groups when it exists is referred to as type II error. The probability of type II errors (β) is related to the power of a study (calculated as 1 − β). Power is increased with larger sample sizes, so a smaller study would be less effective in detecting differences between groups, increasing the chance of a type II error.
type 1 error
Type I errors (false positives) occur when a study incorrectly rejects a null hypothesis that is true. The rate of type I errors is denoted by α and usually reflects the significance level of a test. A higher α increases the likelihood of a type I error and decreases the likelihood of a type II error. The main effect of a smaller sample size is to increase the probability of a type II error rather than a type I error.
In strongly skewed distributions,
the median is a better measure of central tendency than the mean.
Positive predictive value
is the probability that an individual has a disease given a positive test result. Negative predictive value is the probability that an individual does not have a disease given a negative test result.
Case-control studies
begin by identifying participants based on outcome status (eg, cases who have a disease of interest versus controls who do not have the disease of interest). Once identified, cases and controls are then assessed for past exposure to ≥1 risk factors of interest.
The number needed to treat (NNT)
is the number of patients that need to be treated in order to prevent or cure one disease or medical condition. NNT isa useful measure to evaluate the efficacy of a given therapy. It is calculated by getting the inverse of the absolute risk reduction (ARR); in other words, NNT =1/ARR.
Publication bias
occurs when studies with positive results are published but studies with negative/null results are not. Meta-analysis yields an accurate synthesis of the studies included in the analysis, but a biased sample of all relevant studies in the literature distorts the true mean effect computed by the meta-analysis and leads to publication bias.
