Comparative Performance of the COMHON Index and the Braden Scale for Pressure Injury Risk Assessment in Critically Ill Patients: A Retrospective Observational Study Using Consecutive ICU Sampling
Sandra Mas Serván, Sandra Giao Varela, María Victoria Crespo Torres, Sonsoles Hernández Iglesias, Mercedes Gómez del Pulgar, Miguel Angel Cuevas-Budhart, Johis Ortega, Alina Renghea, Almudena Crespo-CañizaresBackground
Pressure injuries remain a major patient safety concern in critically ill patients admitted to intensive care units (ICUs). Although the Braden Scale is widely used for pressure injury risk assessment, its performance may be limited in highly complex critical care settings. The COMHON Index was developed as an ICU‐specific tool integrating physiological variables related to tissue perfusion and oxygenation.
Aim
To compare the performance of the COMHON Index and the Braden Scale for pressure injury risk assessment in critically ill patients.
Study Design
A retrospective,longitudinal, comparative observational study was conducted in a single ICU inSpain between January 2023 and December 2024, using consecutive nonprobabilitysampling.
Results
A total of 476 critically ill patients were included; ICU‐acquired pressure injuries developed in 1.9% of patients during the first 24 h after ICU admission. Greater disease severitywas associated with a higher risk of pressure injury. Higher total Braden scores showed a protective effect, whereas COMHON domains related to consciousness, mobility, hemodynamic stability, and oxygenation were directly associated with pressure injury development. The COMHON Nutrition subscale did not reach statistical significance in univariable analysis (OR: 1.38; p = 0.303). In multivariable analysisusing Firth’s penalized logistic regression, impaired nutritional status(COMHON) was associated with pressure injury, whereas the Braden sensoryperception and moisture exposure subscales showed inverse associations. At the Youden‐optimized cutoffs (Braden ≤ 10; COMHON ≥ 15), the Braden Scale demonstrated 100.0% sensitivity (95% CI: 66.4–100.0) and 77.1% specificity (95% CI: 73.0–80.8), while the COMHON Index demonstrated 88.9% sensitivity (95% CI: 51.8–99.7) and 73.2% specificity (95% CI: 69.0–77.2). Both instruments showed very high negative predictive values (Braden: 100.0%; COMHON: 99.7%).
Conclusions
Both instruments were associated withpressure injury occurrence. The final multivariable model incorporated physiological domains from both instruments, two Braden Scale subscales (sensory perception and moisture exposure) and one COMHON Index domain (nutritional status), suggesting complementary rather than substitutive clinical utility. These findings should be interpreted cautiously given the retrospective single‐center design and low event rate. Prospective multicenter studies are needed to confirm the clinical utility of ICU‐specific risk assessment tools in pressure injury prevention.
Relevance to Clinical Practice
Integrating ICU‐specific physiological parameters into pressure injury risk assessment may support earlier identification of critically ill patients requiring intensified preventive interventions, including repositioning protocols, advanced pressure‐redistributing surfaces, nutritional optimization, and closer hemodynamic monitoring. These findings may contribute to more individualized nursing care planning and improved allocation of preventive resources in intensive care settings.
Impact Statement
This study supports the use of ICU‐specific risk assessment tools to strengthen patient safety strategies and optimize individualized nursing interventions in critical care settings.