The cost of treating COPD patients with long-term oxygen therapy in a French population.

OBJECTIVE In greater Paris and its surrounding (as it is in all France), oxygen is home delivered by not-for-profit (NP) associations or profit-making (PM) health organizations. Both are financed by the national health insurance. This dual context and the current economic climate justify an economic evaluation of all respiratory care for patients with COPD receiving long-term oxygen therapy (LTO). This pragmatic approach identifies the variables that have the greatest impact on direct medical costs and estimates the annual cost for respiratory care per COPD patient. DESIGN Retrospective study. SETTING Health insurance scheme for self-employed professionals (CANAM). PATIENTS AND METHODS Between July 1985 and March 1994, 234 patients registered in CANAM files began LTO, 24% in the PM sector, 76% in the NP sector, mainly using concentrator (78%), mean age of 74 +/- 10 years, male predominance (74%), PaO2 of 56.2 +/- 10.5 mm Hg, FEV1/FVC of 43 +/- 15%, and 51% having 1 or more severe illness(es) associated. The economic appraisal was performed on a representative sample of 61 patients and measured the total resources consumption for respiratory care per COPD patient and per year (physician visits and tests, drugs, physiotherapy, oxygen therapy, hospitalizations for acute respiratory failure, transport costs). RESULTS A quarter of the patients in each sector did not meet the LTO prescription guidelines (PaO2 > 60 mm Hg). For patients having their oxygen delivered by NP sector, the total ambulatory cost for respiratory care was lower ($4,506 per patient and per year vs $5,399) because they mainly used concentrator, all the other direct ambulatory costs being equal. The total annual cost for respiratory care of a COPD patient receiving LTO amounted to $11,672 (NP and PM sectors merged). Oxygen therapy represented 73% of the total ambulatory cost. In a multiple linear regression model, hospitalization represented the largest share of cost, significantly higher when PaO2 was 55 mm Hg or less ($2,287 per patient per year vs $8,717). In contrast, none of the covariates (age, sex, PaO2, FEV1/FVC) influenced at a significant level the total cost of visits, tests, drugs, and physiotherapy, amounting to $1,507. CONCLUSION As oxygen treatment plays an important role in the variation of costs, further pragmatic studies should help to better understand what are the real motivations to choose one mode of oxygen administration more than another and should determine factors that may lead physicians sometimes not to comply with clinical guidelines.