AIM: To validate a model to estimate the nursing workload required by trauma victims on intensive care unit (ICU) discharge. BACKGROUND: Identifying a reliable model of nursing workload measurement that can assist professionals. METHODS: A prospective cohort study conducted in 2010 (first phase) and 2015 (second phase) with 342 trauma victims admitted into the ICU of four hospitals located in São Paulo, Brazil. The original model was created during the first phase and included the Simplified Acute Physiology Score and New Injury Severity Score variables. Coefficients of determination (R2 ) were calculated to identify the reliability. RESULTS: The original model presented high reliability (R2 = 44%) in the hospital of origin, and unsatisfactory performance (R2 < 4%) in the other institutions. An improvement in R2 was observed after adjusting the coefficients for each hospital. CONCLUSION: For the Nursing Activities Score prediction of survivors for ICU discharge, trauma centres must validate the original model with coefficient adjustments for their population, or preferably derive their own models. IMPLICATIONS FOR NURSING MANAGEMENT: Although models are useful in predicting discharge of these ICU patients, for treatment continuity after critical care and for organising services, the study showed that they should be evaluated prior to use for nursing management.
AIM: To validate a model to estimate the nursing workload required by trauma victims on intensive care unit (ICU) discharge. BACKGROUND: Identifying a reliable model of nursing workload measurement that can assist professionals. METHODS: A prospective cohort study conducted in 2010 (first phase) and 2015 (second phase) with 342 trauma victims admitted into the ICU of four hospitals located in São Paulo, Brazil. The original model was created during the first phase and included the Simplified Acute Physiology Score and New Injury Severity Score variables. Coefficients of determination (R2 ) were calculated to identify the reliability. RESULTS: The original model presented high reliability (R2 = 44%) in the hospital of origin, and unsatisfactory performance (R2 < 4%) in the other institutions. An improvement in R2 was observed after adjusting the coefficients for each hospital. CONCLUSION: For the Nursing Activities Score prediction of survivors for ICU discharge, trauma centres must validate the original model with coefficient adjustments for their population, or preferably derive their own models. IMPLICATIONS FOR NURSING MANAGEMENT: Although models are useful in predicting discharge of these ICU patients, for treatment continuity after critical care and for organising services, the study showed that they should be evaluated prior to use for nursing management.