Conformity with clinical practice guidelines for Management of patients with sarcoma: a costeffectiveness assessment in two European regions.

Objectives: Sarcomas are rare tumors (1-2% of all cancers) originating from connective tissue, skin, retroperitoneum, bone, and viscera. The rarity of the disease, the variety of histological types or locations and the heterogeneity of prognostic factors account for the difficulty in acquiring sufficient personal experience. Moreover, the lack of graduate or post-graduate medical teaching makes it difficult for physicians outside centres of excellence to provide appropriate sarcoma management. Therefore, a cost-effectiveness analysis was undertaken in Veneto and Rhône-Alpes in order to evaluate the impact of conformity of clinical pathway to Clinical Practice Guidelines on prognosis of sarcoma. The general objective of this research is to analyze during the whole clinical pathway of patients within the healthcare system the relationships between (i) the management strategies, including conformity with clinical practice guidelines; (ii) the health outcomes, e.g. the disease free survival; and (iii) the resources used. Methodology : Study sample: patients (age 15 years) with histological diagnosis of primary malignant sarcoma of all histological subtypes except osteosarcoma; sarcomatoid carcinoma; mesothelioma; neuroblastoma; paraganglioma. Other inclusion criteria were (i) in Veneto: patients diagnosed and treated in the public hospitals of the region from January 2007 to December 2007; (ii) in Rhône-Alpes: patients diagnosed and treated patients at the University Hospital of Lyon and/or at the Cancer Centre of the Rhône-Alpes region between March 2005 and February 2006; Source of information: hospital records (patients’ age, sex, comorbidities, etc.). Conformity to Clinical Practice Guidelines (CPGs): assessed by two physicians from the Cancer Center Léon Bérard and the University of Padova. Costs assessment: hospital’s point of view; time horizon ranging from diagnosis to the end of follow-up or at the event’s date (relapse, diagnosis of another cancer, death). Costs expressed in € 2009; discounting rate: 4%. Effectiveness: Overall survival and disease free survival. Incremental cost-effectiveness Ratios (ICERs): ICERs was expressing the cost per life year gained and per relapse free year gained. Statistical analysis: descriptive statistics were used to analyze patient and tumor characteristics and costs; a Cox-regression model was used to fit conformity to CPGs on survival after adjusting for grade (II and III vs I); the uncertainty surrounding the ICERs was captured by a probabilistic sensitivity analysis: 1000 replications were obtained by non parametric bootstrap methods. In each quadrant of the cost-effectiveness plane, the probability that the true ICER was in the quadrant under consideration was specified by a percentage. The external ellipse defines the confidence region at level 95%, and the internal ellipse at level 50%. The relationship between the cost of each treatment and a range of explanatory variables including conformity to CPGs was estimated using standard linear models with Ordinary Least Squares (OLS) when the cost variable was continuous, and Tobit regression when the cost variable was censored All statistical analysis were performed using STATA software (version 11.0) and Gauss software (version 9.0). ain content: Conformity with CPG: 118 out of 219 patients (54%) have an overall phase conformed with CPG. The conformity of single phase ranged from 74% for chemotherapy to 63% for follow-up. Differences about conformity group were detected for grade (p=0.01) and histological subtype (p=0.03). Outcomes: overall survival reached 2,3 years when treatments were conformed with CPG versus 2,4 years when treatments were not conformed with CPG (p=0.79, log-rank test). Disease free survival reached 2,1 years when treatments were conformed with CPG versus 2,1 years when treatments were not conformed with CPG (p=0.91, log-rank test). Costs: The average costs of overall sarcoma treatment reached €23,571 (n=118) and 27,019€ (n=101) when treatments were conformed and not conformed with CPG respectively. Incremental cost-effectiveness Ratios: As treatments not conformed with CPG were more costly and less effective than treatments conformed with CPG, no conformity with CPG was strictly dominated for disease free survival. When handling uncertainty, since the origin of the cost-effectiveness plane was included in the 95% confidence ellipses, the probabilistic sensitivity analysis did not permit to confirm these results. The probability that the true ICER fell in the southeast quadrant (i.e. treatment conformed with CPG is less costly and more effective compared to treatment not conformed with CPG) is 48%. Econometrics results: conformity of diagnosis decreased the cost surgery (p=0.03). Conformity of surgical resections increases the cost of surgery (p=0.01) and decreased the cost of initial treatment (p=0.05) and overall treatment (p=0.05). Conformity of chemotherapy decreased the cost of radiation therapy (p=0.05). Expected original contribution: In the context that few studies have examined patient clinical outcomes and costs in relation to conformity with CPG, this cost-effectiveness analysis shows that conformity of the initial treatment of sarcoma with CPG seems to be more expensive in the short run but reduce hospital care expenditures at three years post-diagnosis. This study has never been neither presented nor published previously.