External Validation of the Simple Clinical Score and the HOTEL Score, Two Scores for Predicting Short-Term Mortality after Admission to an Acute Medical Unit

Abstract

BackgroundClinical scores can be of aid to predict early mortality after admission to a medical admission unit. A developed scoring system needs to be externally validated to minimise the risk of the discriminatory power and calibration to be falsely elevated. We performed the present study with the objective of validating the Simple Clinical Score (SCS) and the HOTEL score, two existing risk stratification systems that predict mortality for medical patients based solely on clinical information, but not only vital signs.MethodsPre-planned prospective observational cohort study.SettingDanish 460-bed regional teaching hospital.FindingsWe included 3046 consecutive patients from 2 October 2008 until 19 February 2009. 26 (0.9%) died within one calendar day and 196 (6.4%) died within 30 days. We calculated SCS for 1080 patients. We found an AUROC of 0.960 (95% confidence interval [CI], 0.932 to 0.988) for 24-hours mortality and 0.826 (95% CI, 0.774–0.879) for 30-day mortality, and goodness-of-fit test, χ2 = 2.68 (10 degrees of freedom), P = 0.998 and χ2 = 4.00, P = 0.947, respectively. We included 1470 patients when calculating the HOTEL score. Discriminatory power (AUROC) was 0.931 (95% CI, 0.901–0.962) for 24-hours mortality and goodness-of-fit test, χ2 = 5.56 (10 degrees of freedom), P = 0.234.ConclusionWe find that both the SCS and HOTEL scores showed an excellent to outstanding ability in identifying patients at high risk of dying with good or acceptable precision.

The Patient- And Nutrition-Derived Outcome Risk Assessment Score (PANDORA): Development of a Simple Predictive Risk Score for 30-Day In-Hospital Mortality Based on Demographics, Clinical Observation, and Nutrition

“…To our surprise, we found that functional status [ 39 ], such as mobility and nutrition[ 40 ] were of major importance. Reduced mobility has recently also been included in the Simple Clinical Score for medical emergency patients in addition to vital signs [ 25 ]. Both poor nutrition and reduced mobility could be consequences of the disease process, but they may have health-related effects of their own.…”

“…Prognostic models for acute care situations always include either several vital signs typically obtained from monitors, electrocardiograms or imaging information, biomarker or procedural characteristics. This information is missing in a large proportion of patients even in medical emergency departments [ 25 ]. There is no prognostic model quantifying the risk in the population of hospitalised patients as a whole—excluding ICU-patients—by a single scoring system.…”

End points for validating early warning scores in the context of rapid response systems: a Delphi consensus study

“…8 The fraction of irreversibly ill patients in the populations identified by these end points underlines this problem, as seen in studies from the United Kingdom (UK). Such a problem could become increasingly important with the development of more and more accurate models with c-statistics even higher than ViEWS, such as the c-statistic of 0.96 (0.93-0.99) found in a recent Danish validation study of the Simple Clinical Score.…”

“…Such a problem could become increasingly important with the development of more and more accurate models with c-statistics even higher than ViEWS, such as the c-statistic of 0.96 (0.93-0.99) found in a recent Danish validation study of the Simple Clinical Score. 8 The fraction of irreversibly ill patients in the populations identified by these end points underlines this problem, as seen in studies from the United Kingdom (UK). A review of 1000 deaths in UK hospitals reported 5.2% of the deaths as having a 50% or greater chance of being preventable.…”

Patient‐ and Nutrition‐Derived Outcome Risk Assessment Score as a Predictor of Mortality in Critically Ill Surgical Patients: A Retrospective, Single‐Center Observational Study

“…The AUCs of the ROC analysis models for 30‐day mortality using PANDORA scores were reported to be between 0.79 and 0.82. It is worth noting that ICU patients were excluded in the development of the PANDORA score because the authors noted several “good” (as defined by AUCs in the 0.8–0.9 range) mortality prediction tools validated for use in the ICU . Indeed, widely used ICU‐related mortality prediction scores, such as APACHE, SOFA, Mortality Probability Model, and Simplified Acute Physiology scores, all require extensive physiologic data and may require at least 24 hours of care before they can be calculated .…”