Valutazione dell’appropriatezza organizzativa degli ospedali pubblici abruzzesi tramite PRUO, Disease Staging, e mediante il calcolo dei DRG ad alto rischio di inappropriatezza secondo i “nuovi” e “vecchi” LEA: risultati e grado di concordanza tra le diverse metodiche.

Currently, more than one instrument has been found to be reliable and valid for the assessment of hospital admission appropriateness. However; data on the level of agreement among these methodologies are extremely scarce. The study was aimed at evaluating whether the percentages of organizational (in)appropriateness resulting from some of the most diffused instruments (Italian Appropriateness Evaluation Protocol--AEP/PRUO; Disease Staging; Essential Levels of Care--LEA, version 2001 and 2008) are substantially concordant, or they largely depend upon the methodology. For each public hospital of Abruzzo, Italy, the amount of inappropriateness has been computed using six indicators: inappropriate days of care (PRUO1); totally inappropriate admissions (PRU02); early admissions DRGs according to the first Law on Italian LEA (LEA01); admissions assigned to one of the 108 potentially inappropriate DRGs according to the second Law on Italian LEA, currently inactive (LEA08). The sample was composed by all ordinary admissions made in 2006 in the Region, with the exception of PRUO indicators, which were based upon the manual revision of 2% of all admissions that could be assessed using PRUO methodology. We found a good correlation among most indicators based upon administrative discharge data (DS1, DS2 and LEA01), whereas the results obtained using PRUO and new LEA (LEA08) were discordant, and marked differences were observed also between the two PRUO indicators. Although the limitations of the study permit only preliminary conclusions, in future appropriateness evaluations it may be reasonable to use more than one indicator--allowing the creation of combined scores--and rank hospitals in large categories--avoiding excessively precise scores--as such rankings might relevantly differ depending upon the used instrument. (Less)