Advanced Search

Home > Journals > European Journal of Physical and Rehabilitation Medicine > Past Issues > Europa Medicophysica 2003 December;39(4) > Europa Medicophysica 2003 December;39(4):201-4



A Journal on Physical Medicine and Rehabilitation after Pathological Events

Official Journal of the Italian Society of Physical and Rehabilitation Medicine (SIMFER), European Society of Physical and Rehabilitation Medicine (ESPRM), European Union of Medical Specialists - Physical and Rehabilitation Medicine Section (UEMS-PRM), Mediterranean Forum of Physical and Rehabilitation Medicine (MFPRM), Hellenic Society of Physical and Rehabilitation Medicine (EEFIAP)
In association with International Society of Physical and Rehabilitation Medicine (ISPRM)
Indexed/Abstracted in: CINAHL, Current Contents/Clinical Medicine, EMBASE, PubMed/MEDLINE, Science Citation Index Expanded (SciSearch), Scopus
Impact Factor 2,063

Frequency: Bi-Monthly

ISSN 1973-9087

Online ISSN 1973-9095


Europa Medicophysica 2003 December;39(4):201-4


Case-mix classification systems. The specific case for rehabilitation

Tesio L. 1, 2

1 ­Unit of Neu­ro­motor Reha­bil­i­ta­tion Isti­tuto Aux­o­logico Ital­iano, ­IRCCS, ­Milan, ­Italy
2 Depart­ment of Reha­bil­i­ta­tion Med­i­cine and ­Socio-med­ical Ser­vices, ­CRISP Centro di ­Ricerca Inter­uni­ver­sit­ario sui Ser­vizi alla Per­sona di Pub­blica ­Utilità, ­Milan, ­Italy

The ­case-mix clas­sifi­ca­tion con­sists in iden­ti­fying a ­series of clin­ical con­di­tions, so ­that ­they can be uni­vo­cally clus­tered in ­groups ­which are mutu­ally exclu­sive. Homo­ge­neous con­di­tions are ­assigned to the ­same “group” or “­class”. The cri­teria for homo­ge­neity can be ­much dif­ferent, ­depending on the ­goal of the clas­sifi­ca­tion ­itself. ­Many ­grouping ­systems are acknowl­edged inter­na­tion­ally. ­They may ­follow the cri­terion of homo­ge­neity ­with ­respect to aetio­logy (ICD ­system), ­resource con­sump­tion ­during an inpa­tient ­stay in ­acute ­care facil­ties (DRG ­system), in a reha­bil­i­ta­tion ­unit (FIM-FRG ­systems), or in a ­nursing ­home (RUG ­system). ­Each of ­these ­systems ­strives to ­keep ­some dis­tinc­tion ­across ­groups, ­based on clin­ical cri­teria, ­even ­when the dom­i­nant ­grouping cri­terion (e.g. ­cost homo­ge­neity) ­would ­rather sug­gest ­group ­merging. The ­main sta­tis­tical tech­nique ­adopted for ­case-mix ­grouping is the clas­sifi­ca­tion and regres­sion ­tree (­CART). ­CART ­allows to ­find the ­best com­pro­mise ­between the ­need to max­i­mise var­i­ance expla­na­tion of the depen­dent var­i­able (i.e. the homo­ge­neity cri­terion: clin­ical ­severity, ­cost, ­length of ­stay and the ­like), and the ­need to ­retain clin­i­cally iden­ti­fi­able dis­tinc­tions (e.g. ­based on the ­organ ­impaired). ­Using a clas­sifi­ca­tion ­system as a pay­ment ­system ­entails ­other con­straints: easi­ness of account­ability, dis­in­cen­tive ­against ­upcoding, dis­in­cen­tive ­against ­adverse selec­tion of ­more ­demanding ­patients. In reha­bil­i­ta­tion med­i­cine a FIM-FRG ­system is ­already ­adopted as a pay­ment ­system for ­post-­acute reha­bil­i­ta­tion in USA. ­Some ­trials on FIM-FRG clas­sifi­ca­tion ­have ­been suc­cess­fully com­pleted ­in ­Italy, too.

language: English


top of page