Glossary
Terms used in this document are defined in the NICE glossary and below.
Atherosclerotic cardiovascular disease (ASCVD)
Current or previous history of peripheral arterial disease, coronary artery disease such as myocardial infarction and unstable angina, cerebrovascular disease such as transient ischaemic attack and ischaemic stroke or unspecified stroke. Definition obtained from NICE's guideline on type 2 diabetes in adults. The health state 'ASCVD other' refers to current or previous history of peripheral arterial disease or transient ischaemic attack.
Bariatric procedures
A term used to encompass surgical (invasive or minimally invasive) and non-surgical procedures (minimally invasive).
Baseline
1). The starting time-point of the model, usually the decision-point with respect to choice of treatment strategy. The opposite of the 'time horizon', baseline characteristics include the age and comorbidities of patients at the outset.
2). 'Baseline risks' are the event rates in 1 arm of the model. The event rates in another model arm might be determined by multiplying the baseline risk with a relative treatment effect.
Cohort model
A type of health economic model that follows a group of people with shared characteristics through defined health states over time, using average transition probabilities to estimate cost effectiveness of interventions.
Direct evidence
When direct evidence of an effect is required, this should be evidence from randomised controlled trials (RCTs) for the specific treatment. When RCT evidence is unavailable or inappropriate, direct evidence from a real-world setting may be used, subject to appropriate assessment of bias and confounding factors.
Individual patient-level simulation
A modelling approach that simulates outcomes for individual patients rather than a whole cohort. Each simulated patient has unique characteristics and pathways through health states, allowing variability in risks, events and treatment effects to be captured.
Partial waning
Used when treatment effectiveness declines over time to a lower level.
Recommended statements
Where there is some uncertainty, an approach is 'recommended' within this reference case extension. These statements should be adhered to, as appropriate, and when possible. When deviations occur, these should be discussed with the committee and NICE quality assurance staff.
Required statements
Where there is strong evidence and it is considered a best approach, this reference case extension makes it clear that a specific element is 'required'. These statements must be adhered to unless there is a strong justification for not doing so, such as new evidence that contradicts the guidance in this document. When deviations occur, these should be discussed and agreed with the committee and NICE quality assurance staff.
Risk equations
Risk equations are mathematical formulas used to estimate the probability of a health event based on patient characteristics. These equations are often derived from regression models using clinical trial or cohort data and are commonly used in economic models to simulate disease progression or treatment impact.
Risk prediction tools
These use 1 or more risk equations and other types of approaches, such as algorithms and scores, to provide an estimate of a person's risk of a health event.
State transition model
A modelling approach in which a population or individual moves between defined health states over certain time periods. State transition models can be implemented as cohort models or individual-level models.
Stratum
A stratum is a group of people that is pre-defined at baseline based on specific characteristics. Stratification is used to ensure that each group is adequately represented, and each stratum is mutually exclusive. The reference case extension defines 8 strata for use in NICE economic evaluations of weight management. Cost-effectiveness results should not be pooled across different strata.
Subgroup
A subgroup is a set of participants, used to explore differences in health outcomes or effects across specific characteristics (for example, sex or comorbidity status). Subgroups can overlap and are often used in subgroup analyses.
Threshold analysis
An analysis used to identify the value of a key input at which an intervention becomes cost-effective.
Tunnel state
A temporary health state in a state transition (Markov) model used to capture time-dependent events or short-term effects that cannot be represented by standard Markov states.