Impact of Real-Time Comorbidity Capture on Clinical Escalation and Harm Prevention in Hospitalized Patients: A Benchmarking and Statistical Gap Analysis.

Background Systemic documentation deficiencies in hospitalized patients with high-acuity conditions such as systemic inflammatory response syndrome (SIRS) can compromise clinical decision-making, risk adjustment, and patient safety. Despite well-established evidence linking chronic conditions like chronic kidney disease (CKD), diabetes mellitus (DM), and liver failure to adverse outcomes, these comorbidities are often underrepresented in electronic health records (EHRs). This study introduces a statistically rigorous, informatics-assisted framework to benchmark and quantify real-world documentation gaps, advancing the field of clinical documentation integrity (CDI) through actionable metrics. Objective The objectives of this study are to evaluate the accuracy of comorbidity documentation in adult inpatients with SIRS and acute organ dysfunction by applying a benchmarking model based on peer-reviewed prevalence data and to assess the clinical, operational, and financial implications of these documentation deficits. Methods In this retrospective observational study, 82 adult patients admitted with a principal diagnosis of SIRS (ICD-10-CM R65.11) were analyzed using structured chart abstraction from EPIC EHR data. Six high-impact comorbidities were evaluated. Observed documentation rates were compared against literature-derived expected prevalence benchmarks. Gap scores were calculated as proportional differences, and one-sample z-tests were used to assess statistical significance, stratified by age and sex. Results Five of six comorbidities demonstrated statistically significant underdocumentation, with liver failure (expected: 57%, observed: 6.1%; p < 0.001), DM (53% vs. 18.3%; p < 0.001), and HTN (78% vs. 36.6%; p < 0.001) showing the most severe deficiencies. Gap scores ranged from 0.39 to 0.89, revealing systemic failure in documenting clinically relevant conditions. Age and sex stratification further exposed disparities in documentation behavior. Notably, underdocumentation did not correlate with inpatient mortality (p = 0.89), indicating systemic gaps unrelated to clinical outcome visibility. Conclusion This study provides a validated, reproducible framework for identifying and quantifying clinical documentation gaps in real-time. The comorbidity capture gap model integrates seamlessly with EHRs and aligns with CMS risk adjustment frameworks such as Hierarchical Condition Category (HCC) and Diagnosis-Related Group (DRG). Its application can enhance clinical accuracy, improve institutional quality metrics, and inform scalable, AI-augmented CDI interventions. The findings underscore a critical opportunity to shift from reactive to proactive documentation practices, transforming the reliability of health data for risk stratification, harm prevention, and healthcare performance reporting.