3 questions to... Zeynep Or on the occasion of the publication of Issues in Health Economics (226), March 2017: Variations in Surgical Practices in Breast Cancer Treatment in France, in collaboration with Mobillion V. (Upec, IRDES), Touré M. (IRDES), Mazouni C. (Gustave Roussy), Bonastre J. (Gustave Roussy, INSERM-CESP).

May 2017

1/ What does the expression "variation in practices" mean? Why did you choose to study breast cancer surgery?

The variation in practice refers to the fact that patients with comparable characteristics can be treated very differently depending on the physician or healthcare facility, their place of residence and/or socio-economic status. This phenomenon of variations in practices has been observed in the literature on breast cancer treatment, which, it should be recalled, is the leading cause of cancer incidents among women (31% in metropolitan France).

Studying this phenomenon in the context of surgical practices in breast cancer treatment is all the more justified since there are several surgical treatment options for women who require surgery. Three options are studied here: conservative surgery, sentinel lymph node technique and immediate breast reconstruction after a total mastectomy. We compare the evolution between 2005 and 2012 of these treatments, less invasive than others, in order to draw up an inventory of the situation before and after the introduction of the T2A (Activity-based funding, or Tarification à l'activité) and the introduction of minimum activity thresholds in cancerology. The literature has shown that while the prevention and treatment of these cancers has improved over time, variations in the treatment of patients with similar pathologies have been observed. It is important to identify and analyze these variations in order to be able to act to standardize practices and improve the overall quality of care.

2/ How do you measure these variations, both territorial and between hospitals?

To compare the variation in practices nationally, we measured practice rates for these three procedures in different types of healthcare facilities, and depending on where the patient live, by department. The use of these surgical procedures can be influenced by a variety of factors such as health status, patient and practitioner preferences, but also the availability and organization of technical facilities, and local medical habits.

To identify the determinants of the use of these interventions, we conducted a multi-level modeling by monitoring the observable characteristics of patients and healthcare facilities.

3/ What can we learn from these results?

Our analysis shows that surgical practices in the treatment of breast cancer have undergone major changes between 2005 and 2012: tumorectomy has become the reference treatment, with a rate of use exceeding 70% in most hospitals in 2012, and the sentinel node technique has spread everywhere, with the number of patients involved tripling over the period. Immediate breast reconstruction after a total mastectomy, on the other hand, remains infrequent, although there has been an increase in the use of this technique between 2005 and 2012.

Our analysis also highlights the existence of territorial inequalities in the surgical treatment of breast cancer. Territorial disparities in practice suggest that the likelihood of receiving these treatments varies according to the patient's place of residence.

In addition, some of these variations seem to be related to variations in practices between hospitals: all other things being equal, the probability of benefitting from sentinel node techniques or immediate breast reconstruction is higher in Cancer centres (Centres de Lutte contre le Cancer, or CLCC), Regional teaching hospitals (Centres Hospitaliers Régionaux, or CHR) and hospitals with a high patient volume.

Interview by Anna Marek