WhAT CAn Be expeCTed fRom The depARTmenT of mediCAl infoRmATion To impRove The peRfoRmAnCe index of The AveRAge lengTh of STAy ?

Abstract

introduction: The performance index of the average length of stay (PIALS) depends, among others, on coding. The reference average lengths of stay (ALS) are trimmed, and calculated with the stays of the previous year. Our primary objective was to study the impact of coding on the PIALS. Our secondary objectives were to calculate the PIALS of the University Hospital of Strasbourg (UHS) in 2013, taking into account the trimming of the ALS, and to calculate the PIALS of the French national hospital stays data base in 2013. method: The hospital stays of the UHS in 2013 were studied. The coding quality was assessed using the Standardized Severity Index (SSI). We simulated modified codings, including a normalization of the SSI, and calculated the resulting PIALS. The profit obtained through the recoding in 2014 of the hospital stays of 2013, was compared to the profit which could have been obtained through an increase in activity, with the same hospital bed occupancy rate and with the same change of the PIALS. The PIALS of the UHS in 2013 was recalculated by exluding the stays whose length was not comprised within the trimming limits. The PIALS of the French national hospital stays data base was calculated, using the trimmed ALS in 2013 as a reference. Results: The SSI of the UHS was brought from 1.060 to 1 and the PIALS from 1.089 to 1.133. The profit of the recoding was 3 765 169 euros, versus 5 130 128 euros if the same change of the PIALS had been obtained through an increase in activity, with the same hospital bed occupancy rate. The PIALS of the UHS, excluding outliers, was 1.043. The PIALS of the French national hospital stays data base was 1.026. discussion/conclusion: The impact of coding on the PIALS is significant, but it is limited and must be completed by a reduction of the hospital length of stay and an increase in activity. The role of the physicians of the department of medical information is also to analyse this indicator. Indeed, the way the PIALS is calculated is questionnable, because the outliers are excluded from the calculation of the reference ALS, and it emphasizes artificially the bad performance of the hospitals.