Predictive Nomogram for ICU Pressure Injury Deterioration Risk

A retrospective study of 278 ICU patients with Stage 1 pressure injuries, split into training (n=195) and validation (n=83) sets. The dataset was used to develop a nomogram predicting injury progression based on three admission factors. The model was created by Chenyan Zhang and last updated in May 2026.

Use Cases

• Developing predictive models for clinical deterioration based on admission variables like diabetes status and medication dose.
• Validating logistic regression or LASSO regression techniques for feature selection in medical datasets.
• Creating clinical decision support tools for early intervention based on risk scores for pressure injury progression.

Strengths

• Model performance is validated with reported AUC scores of 0.800 (training) and 0.785 (validation).
• The dataset is derived from a specific clinical study with a defined cohort of 278 patients.
• Predictors are clearly identified with odds ratios and confidence intervals (e.g., diabetes OR: 2.451).

Limitations

• Row count for the underlying patient data is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is small in size (48.0 KB), indicating limited scope.

Provenance

Source

figshare

Collection Method

Retrospective cohort study of eligible ICU patients.

Freshness

Last updated 2026-05-18 05:47:46; freshness should be verified.