Statistical Power
Statistical power is the probability a trial gives a significant result when a true effect of set size exists, used to size clinical and economic trials.
Explore comprehensive, evidence-informed explanations of key health economics concepts, including their development, application and relationships to other concepts. Published entries are validated through human expert review.
A concept page is a published entry on the wiki. Some concepts also have a full encyclopaedia article. The Health Economics Ontology (HEO) holds every concept, including dictionary terms without an article, so it is larger than the list of published pages. The knowledge graph shows how concepts relate to each other.
Statistical power is the probability a trial gives a significant result when a true effect of set size exists, used to size clinical and economic trials.
Statistical significance is a threshold-based conclusion that an observed result is sufficiently incompatible with a prespecified null hypothesis under a stated statistical model to meet the chosen significance level.
A steady state in a Markov model is the stationary distribution of a cohort across health states, unchanged from cycle to cycle under fixed transitions.
Stochastic programming is optimisation that includes parameter uncertainty, used in health care to choose a vaccination policy before outcomes are known.
A stock and flow model tracks health quantities, such as people with a disease, as stocks that change over time only through inflow and outflow rates.
Strict dominance in cost-effectiveness analysis is when one option both costs less and gives more health than another, so the other option is ruled out.
A structural equation model (SEM) is a set of linked regression equations, often with latent variables, that tests hypothesised paths to health outcomes.
Structural uncertainty is doubt about how a health economic model represents a decision problem, such as its health states, pathways or extrapolation.
Subgroup analysis estimates treatment effects, costs or cost-effectiveness separately for defined groups of patients to test whether they differ.
The Suits index measures how progressive a tax or health financing source is by comparing cumulative payment shares with cumulative income shares.
A summary ROC (SROC) curve pools diagnostic accuracy studies to show how a test's sensitivity trades off against specificity across positivity thresholds.
Supplier-induced demand (SID) is use of health care that providers create through their information advantage, beyond what informed patients would choose.
Supply is the amount of a good or service, such as health care, that producers are willing and able to provide at each price over a given period.
Survival analysis is the statistical study of time to an event such as death when follow-up is incomplete, giving survival curves for economic models.
The survival function, S(t), is the probability that a patient stays alive or free of a defined event beyond time t, used to model life years and QALYs.
Survival regression estimates how treatment and patient characteristics change the hazard of, or time to, death or progression in censored data for HTA.
Synthetic control estimates a health policy's effect on one region or provider using a weighted blend of untreated units that tracked it before the policy.
A system dynamics model simulates a health system in aggregate as stocks, flows and feedback loops, projecting how policies play out over time.
The t-distribution is a heavy-tailed, symmetric curve used for intervals and tests on mean costs or effects when variance is estimated from a small sample.
The Tarone-Ware test is a weighted log-rank test comparing survival between trial arms, weighting each event time by the square root of the number at risk.