ORIGINAL ARTICLES   

Italian Journal of Vascular and Endovascular Surgery 2004 March;11(1):25-33

Copyright © 2005 EDIZIONI MINERVA MEDICA

language: English

Variation of utilization of cardiovascular surgery in Marche Region

Panella M. ¹, Demarchi L. M. ¹, Kozel D. ¹, Luparia M. ¹, Marchisio S. ¹, Renna M. ¹, Sarasino D. ¹, Di Stanislao F. ²

¹ Unit of Hygiene and Public Health, Department of Medical Sciences, “A. Avogadro”, Univesity of Eastern Piedmont, Novara, Italy 2 Unit of Hygiene and Public Health, University of Ancona, Italy

Aim. To ­describe vari­a­tion ­rates of car­diac and vas­cu­lar sur­gi­cal pro­ce­dures in Marche Region and to ­find its deter­mi­nants.
Methods. We col­lect­ed ­data of the car­di­o­vas­cu­lar sur­gi­cal pro­ce­dures per­formed in the 13 Health Authorities (ASL) of the Marche Region dur­ing the peri­od 1997-2000. First we quan­ti­fied the lev­el of vari­a­tion for ­each pro­ce­dure. Then we ­described the pos­sible sit­u­a­tions of under­use and over­use for ­each inter­ven­tion and ASL. Finally we ­built mul­ti­ple regres­sion mod­els in ­order to eval­u­ate the ­effect of ­socio-eco­nom­ic and san­i­tary fac­tors on the ­observed vari­a­tions.
Results. We ­observed the high­est lev­els of vari­a­tion for dis­cre­tion­ary pro­ce­dures. In par­tic­u­lar the North­ern ASLs ­showed a prev­a­lence of over­use, where­as under­use pre­vailed in the south­ern. We ­found a sig­nif­i­cant cor­re­la­tion ­between the under­use and over­use lev­els of the ASLs (r=-0.756; p=0.002). The mul­ti­var­i­ate mod­els ­explained up to 58% of the ­observed vari­a­tions.
Conclusion. We iden­ti­fied as pos­sible deter­mi­nants of vari­a­tion in car­diac and vas­cu­lar sur­gery: the dif­fer­ent sup­ply of Health Services; ­local ­socio-eco­nom­ic fac­tors, pas­sive mobil­ity ­with ­respect to the bor­der­ing Regions and pro­fes­sion­al beha­vi­our. However, ­once we ­have deter­mined the pres­ence of a ­strong vari­a­tion in the ana­lysed ter­ri­to­ry, we ­still ­need to inter­pret our ­results in rela­tion to the ­real lev­el of the appro­pri­ate­ness and of the out­comes of the ana­lysed pro­ce­dures, and to ­find strat­e­gies fit to con­trol ­this phe­nom­e­non.