HEALTH-STATUS PREDICTION IN CRITICALLY ILL CHILDREN - A PILOT-STUDY INTRODUCING STANDARDIZED HEALTH RATIOS

Performance of intensive care is usually quantified by means of standa rdized mortality rates, where standardization is directed towards the severity of illness on admission. However, as more critically ill pati ents survive, functional outcome and quality of life of these patients becomes more important. In a prospective study in a 10-bed tertiary p aediatric intensive care unit (ICU), admission and follow-up health st atus were collected for 209 surviving patients. For this cohort of pat ients, health status 1 year after admission was also predicted, using the quantified health-utility-index (HUI), as a Value between 0 and 1. For this purpose, two alternative multiple regression models were con structed. The most important predictors of 1-year health status were t he level of sensation, mobility and cognition on admission to which se lf-care, systolic blood pressure, oxygen, Glascow Coma Scale, glucose and age may be added. The two alternative predictive models performed equally well (R-2 = 0.83 and 0.84 respectively), indicating that healt h status could be predicted to a significant degree, The concept of re lating expected future health status (based on base-line health status ), with actual (observed) health status is denoted with the Standardiz ed Health Ratio (SHR). In combination with the Standardized Mortality Ratios (SMR), such a ratio may become a new comprehensive indicator of performance in intensive care medicine.