Adjudication of Codes for Identifying Sepsis in Hospital Administrative Data by Expert Consensus*
Authors' conclusion
Compared with other code-based algorithms, AlgorithmL includes more infection and organ dysfunction codes. AlgorithmL incidence rates are higher; hospital mortality rates are lower.
Study context
- Population
- Adult hospitalizations in Calgary, Canada (adult population 1.033 million), ICU and non-ICU cohorts
- Participants
- 61,632 eligible hospitalizations
- Exposure
- AlgorithmL vs four previously described code-based algorithms
- Outcome
- Annual sepsis incidence (per 100,000 population)
Methodological detailsexpert consensus adjudication of codes, population-based, administrative data, Experts reached consensus on 1928 infection and 108 organ dysfunction codes
- Expert consensus adjudication of codes
- Population-based
- Administrative data
- Experts reached consensus on 1928 infection and 108 organ dysfunction codes
Main finding
As reported in the abstract
| Group | ICU | non-ICU | ICU (range) | non-ICU (range) |
|---|---|---|---|---|
| AlgorithmL | 91 | 291 | ||
| Comparators | 28-77 | 11-266 |
ICU
AlgorithmL91
Comparators— not stated in the abstract
non-ICU
AlgorithmL291
Comparators— not stated in the abstract
ICU (range)
AlgorithmL— not stated in the abstract
Comparators28-77
non-ICU (range)
AlgorithmL— not stated in the abstract
Comparators11-266
Observational data: associations, not effects.
Further findings
| Outcome | AlgorithmL |
|---|---|
| Number of codes included | |
| Infection codes | 720 |
| Organ dysfunction codes | 50 |
| Hospital sepsis mortality | |
| ICU | 24% |
| non-ICU | 17% |
| Cases meeting CDC Adult Sepsis Event (ASE) criteria | 41% |
- Number of codes included — Comparison algorithms: infection codes
- 42-941
- Number of codes included — Comparison algorithms: organ codes
- 2-36
- Hospital sepsis mortality — Comparators, ICU cohort
- 27-38%
- Hospital sepsis mortality — Comparators, non-ICU cohort
- 18-47%
- Cases meeting CDC Adult Sepsis Event (ASE) criteria — Comparator algorithms
- 42-82%
Number of codes included
- Substantial nonoverlap of codes in AlgorithmL vs comparators
Original publication
Adjudication of Codes for Identifying Sepsis in Hospital Administrative Data by Expert Consensus*. Crit Care Med. Published 10 October 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →