Concept Architecture
Concept
Theoretically, Surface Under the Cumulative Ranking Curve (SUCRA) is a treatment ranking measure used in network meta-analysis to summarise the probability that an intervention is among the most effective treatments in a network. It transforms cumulative ranking probabilities into a single numerical value ranging from 0 to 1, where higher values indicate greater certainty that a treatment ranks favourably. SUCRA exists to facilitate interpretation and comparison of treatment rankings across multiple competing interventions.
Mathematically, SUCRA is calculated from the cumulative ranking probabilities generated by a network meta-analysis. It represents the area under the cumulative ranking curve relative to the theoretical maximum area and therefore quantifies the average ranking performance of each treatment across all possible ranks. Bayesian network meta-analysis commonly derives SUCRA from posterior ranking probabilities, although analogous measures exist within frequentist analyses.
In practice, SUCRA is routinely reported alongside comparative treatment-effect estimates in network meta-analysis conducted for health technology assessment, comparative effectiveness research and clinical guideline development. Although useful for summarising treatment rankings, SUCRA values should always be interpreted together with treatment-effect estimates, uncertainty and clinical relevance rather than in isolation.
Purpose
Used to summarise treatment rankings in network meta-analysis by quantifying the probability that each intervention is among the most effective options within a treatment network.
Mathematical Formulae
Primary Formula
SUCRA = (1 / (m ? 1)) ? ??????? (1 ? CPr(r))
where:
- SUCRA = Surface Under the Cumulative Ranking Curve
- m = number of treatments
- CPr(r) = cumulative probability that a treatment is among the r least effective treatments
Equivalent formulation:
SUCRA = (Mean rank ? 1) / (m ? 1)
using the appropriate transformation of the expected treatment rank.
Supporting Formulae
Expected rank:
E(R) = ?? r ? Pr(R = r)
where:
- R = treatment rank
- Pr(R = r) = probability that the treatment occupies rank r
Related Mathematical Methods
- Network Meta-Analysis
- Bayesian Network Meta-Analysis
- Frequentist Network Meta-Analysis
- P-Score
- Mixed Treatment Comparison
- Treatment Ranking
- Rank Probability Analysis
Example
A Bayesian network meta-analysis compares eight treatments for severe asthma. Treatment A has a SUCRA value of 0.96, Treatment B 0.81 and Treatment C 0.42. These values indicate that Treatment A has the highest probability of being among the most effective interventions in the network, although reimbursement decisions are based primarily on comparative treatment effects, uncertainty and cost-effectiveness rather than ranking alone.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| SUM | =SUM(B2:B8) | Sum cumulative ranking probabilities |
| AVERAGE | =AVERAGE(C2:C8) | Calculate mean treatment rank |
| RANK.AVG | =RANK.AVG(D2,$D$2:$D$8,0) | Rank treatments according to SUCRA values |
| SORT | =SORT(A2:B8,2,-1) | Order treatments from highest to lowest SUCRA |
VBA (Optional)
Automate calculation of cumulative ranking probabilities and SUCRA values from network meta-analysis output and generate treatment ranking tables for health technology assessment reports.
Sources
- Salanti G, Ades AE, Ioannidis JPA. Graphical Methods and Numerical Summaries for Presenting Results from Multiple-Treatment Meta-Analysis. Journal of Clinical Epidemiology. 2011.
- Chaimani A, Caldwell DM, Li T, et al. Chapter 11: Undertaking Network Meta-Analyses. Cochrane Handbook for Systematic Reviews of Interventions.
- Dias S, Welton NJ, Sutton AJ, Ades AE. NICE Decision Support Unit Technical Support Documents: Evidence Synthesis for Decision Making.
- NICE. Health Technology Evaluation Manual.
- ISPOR Good Practice Task Force Report on Network Meta-Analysis.
Related Concepts (2)
Frequently Asked Questions (6)
What is SUCRA?
An abbreviation for the surface under the cumulative ranking curve, ranking treatments in a Bayesian network meta-analysis by probability of being most effective.
Source: Salanti, Ades & Ioannidis 2011
What does SUCRA summarise about a treatment in a network?
SUCRA, the surface under the cumulative ranking curve, summarises in a single number how highly a treatment ranks across a network meta-analysis, given the uncertainty in the estimates. It captures the probability that a treatment is among the better options overall, with a value near one marking a treatment likely to be best and one near zero likely to be worst. This condenses a complex set of rankings into one accessible figure. Reducing a treatment's overall standing to a number is its purpose. Salanti and colleagues (2011) describe this measure.
Source: Salanti et al. 2011
How is SUCRA calculated?
SUCRA is calculated from the ranking probabilities produced by a Bayesian network meta-analysis: for each treatment, the probabilities of achieving each possible rank are accumulated across ranks to form a cumulative ranking curve, and SUCRA is the area under that curve, standardised to lie between zero and one. A treatment that is always ranked best would have a SUCRA of one, and one always ranked worst a SUCRA of zero. So SUCRA is calculated as the standardised area under the cumulative ranking curve derived from the posterior ranking probabilities, summarising the treatment's ranking performance.
Source: Salanti, Ades & Ioannidis 2011
What does SUCRA indicate?
SUCRA indicates a treatment's overall ranking in the network, with a higher value meaning the treatment tends to occupy better ranks and be more effective relative to the others, and a lower value meaning it tends to rank worse. It summarises the full distribution of ranking probabilities into one number, allowing treatments to be ordered. So SUCRA indicates how well each treatment ranks across the network, providing a concise measure to compare treatments' relative performance, though it reflects ranking rather than the magnitude of differences or the certainty of the evidence, which must be considered separately.
Source: Salanti, Ades & Ioannidis 2011
What are the limitations of SUCRA?
The limitations of SUCRA include that it summarises ranking without conveying the magnitude of the differences between treatments, so treatments with similar effects can have different SUCRA values; that rankings can be unstable when estimates are imprecise or the network sparse; and that a high SUCRA does not establish clinical importance or account for the certainty of the evidence. So SUCRA is interpreted alongside the relative effect estimates, their uncertainty, and the quality of the evidence, since rankings alone can mislead if taken without regard to the size of the differences and the reliability of the network, making it one output among several.
Source: Rücker & Schwarzer 2015
How does SUCRA relate to the P-score?
SUCRA relates to the P-score as its Bayesian counterpart: both summarise a treatment's ranking in a network meta-analysis into a single value between zero and one, with higher values indicating better relative performance, and the two give very similar results. SUCRA is computed from the posterior ranking probabilities of a Bayesian analysis, while the P-score is derived analytically from frequentist estimates without simulation. So SUCRA and the P-score are analogous ranking measures for Bayesian and frequentist network meta-analysis respectively, providing equivalent summaries of how treatments rank, and both are interpreted with the same cautions about ranking measures.
Source: Salanti, Ades & Ioannidis 2011
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 4 Dec 2025
Content version: 1.0.0
Canonical Identity
- Persistent URI
- https://healtheconomics.wiki/concept/sucra
- Term code
- HE-ES-ESM-057
Stable URI · Machine-readable · Resolvable · CC BY 4.0