Breast cancer care compared with clinical Guidelines: an observational study in France
- Marie Lebeau†1,
- Simone Mathoulin-Pélissier†2, 3, 4, 5Email author,
- Carine Bellera3, 4,
- Christine Tunon-de-Lara6,
- Alain Daban7, 8,
- Francis Lipinski9,
- Dominique Jaubert2,
- Pierre Ingrand1,
- Virginie Migeot1 and
- the REPERES group
© Lebeau et al; licensee BioMed Central Ltd. 2011
Received: 5 August 2010
Accepted: 20 January 2011
Published: 20 January 2011
Great variability in breast cancer (BC) treatment practices according to patient, tumour or organisation of care characteristics has been reported but the relation between these factors is not well known. In two French regions, we measured compliance with Clinical Practice Guidelines for non-metastatic BC care management and identified factors associated with non-compliance at clinical and organisational levels.
Eligible patients had invasive unilateral BC without distant metastases and at least two contacts with one of the two regional healthcare systems (2003-2004) in the first year after diagnosis. Medical data were collected from patient medical records in all public and private hospitals (99 hospitals).
The care process was defined by 20 criteria: clinical decisions for treatment and therapeutic procedures. Each criterion was classified according to level of compliance ("Compliant", "Justifiable" and "Not Compliant") and factors of non-compliance were identified (mixed effect logistic regression).
926 women were included. Non-compliance with clinical decisions for treatment was associated with older patient age (OR 2.1; 95%CI: 1.3-3.6) and region (OR 3.0; 95%CI: 1.2-7.4). Non-compliance with clinical decisions for radiotherapy was associated with lymph node involvement or the presence of peritumoural vascular invasion (OR 1.5; 95%CI: 1.01-2.3) and non-compliance with overall treatment (clinical decisions for treatment + therapeutic procedures) was associated with the presence of positive lymph nodes (OR 2.0; 95%CI: 1.2-3.3), grade III versus grade I (OR 2.9; 95%CI: 1.4-6.2), and one region of care versus another (OR 3.5; 95%CI: 1.7-7.1). Finally, heterogeneity of compliance in overall treatment sequence was identified between local cancer units (p < 0.05).
This study provides interesting insights into factors of non-compliance in non-metastatic BC management and could lead to quality care improvements.
The treatment of breast cancer is complex and requires multidisciplinary care [1, 2]. To deal with this complexity and reduce the number of inappropriate interventions, government agencies and specialist societies have developed Clinical Practice Guidelines (CPGs). Nevertheless, many studies [3–5] have highlighted a great variability of practices in breast cancer treatment according to patient or tumour characteristics, and the way care is organised. Other studies in the literature emphasise organisational factors as determinants of compliance with CPGs [6–11]. In 2003, all French regions began to develop local cancer units (LCUs) dedicated to cancer care. The aim at this time was to improve the quality of care by enhancing the quality of therapeutic multidisciplinary committee discussions. These multidisciplinary committees base their decisions on national or regional CPGs in order to harmonise practices. Our objectives in this study were firstly to measure the compliance with CPGs for the management of non-metastatic breast cancer care and secondly to identify factors associated with non-compliance at a clinical and organisational level (LCU).
Study design and population selection
Eligible patients had a pathological diagnosis of invasive unilateral breast cancer without distant metastases (from June 2003 to October 2004), at least two contacts with one of the two regional healthcare systems between diagnosis and their first year of follow-up, no previous malignancy, and provided written informed consent. These eligible patients were reported by public and private hospitals in the two regions (additional file 1: 99 eligible hospitals: 13% with less than 11 BC surgeries per year and 51% with more than 50). All oncologists in both regions reported every patient with a first diagnosis of invasive non-metastatic breast cancer. Following this, a letter explaining the aims of the study was sent to each patient, together with a consent form and a questionnaire collecting personal information. Once the consent form had been received, medical data was collected from the patient's medical record. The logistics of data collection were carried out by a experienced research team specifically dedicated to the project (research assistants and research practitioners). Full details of the present design were published earlier [12, 13]. Clinical data covered the care process from diagnosis to the last sequence of treatment (follow-up was excluded). The care process was divided into six potential steps: initial and complementary surgery, breast or chest radiotherapy, lymph node radiotherapy, chemotherapy, and hormone therapy. This study was approved by the ethics committee.
Evaluation of compliance with practice guidelines
We categorised the available non-metastatic BC management pathways from French national CPGs (additional file 2) into 20 criteria (Additional file 3). These criteria were used to assess compliance with the care process for each patient according to the six steps defined. Each criterion was classified into three levels of compliance: (C) compliance with CPGs; (J) justifiable non-compliance, i.e., not strictly compliant but documented justification due to the patient's general status, preference or a change during the course of care management (for example, chemotherapy interruption related to adverse effects) or other factors; (NC) non-compliance with CPGs and no justification available in the patient's medical record. For each criterion, the conditions needed to define three-level compliance were defined by a specific questionnaire sent to the relevant experts (surgeon, pathologist, radiologist, chemotherapist and radiation oncologist); moreover, these definitions were agreed upon by each expert. For example, we defined the type of breast and lymph node surgery recommended for different tumour sizes (< 1 cm, 1-3 cm, and >3 cm). Finally, the entire expert group reached a consensus on the description of criteria with particular attention to the justifiable category.
To measure the appropriateness of medical decisions, we also defined compliance with criteria which included therapeutic decisions as opposed to diagnostic or organisational elements (criteria 1, 2, 6, 15, 18, 20). Moreover, we studied both the clinical decisions for treatment and therapeutic procedures of the least compliant therapeutic steps (criteria 6-14). In addition to assessing the 20 criteria, overall treatment compliance (clinical decisions for treatment and therapeutic procedures) with CPGs was determined for each patient. This depended on the compliance with criteria 1-20 of clinical care. Because we modelled the probability of non-compliance, the overall treatment sequence compliance was coded '0' if all therapeutic clinical decisions for treatment and therapeutic procedures were compliant (C) or justifiable (J), and '1' if at least one of these was not compliant with standards (NC).
Explanatory variables were selected from the literature review and related to the patient and her social status [1, 14, 15], to the tumour [6, 7, 15–18] and to the healthcare system [1, 6–8]. Patients' characteristics were analysed by age (< 50, 50-69, ≥70), educational level (less than high-school diploma, at least high-school diploma), cohabitation status (living alone, living with other(s)), menopausal status (pre- or postmenopausal). Tumour characteristics were analysed by localisation (central, medial/lateral, several quadrants), nodal status (negative, positive), histological tumour size (≤10 mm, 11-30 mm, >30 mm), hormonal receptor status (oestrogen or progesterone receptors positive, both receptors negative), lymph node and peritumoural vascular invasion (presence/absence), histological grade (I, II, III). Tumour staging was recorded according to the American Joint Committee on Cancer classification . Variables related to the healthcare system were sector of care (public, private, both), surgical hospital status (teaching, non-teaching) and region where patients underwent surgery. Given that variations in care across practices has been documented previously in France , we suspected that non-compliance could vary across the LCUs (15 Units) and thus considered this variable when evaluating non-compliance with CPGs.
We examined factors associated with non-compliance with CPGs. Aside from the LCU, all variables were first fitted in univariate logistic regression models. Variables significant in the univariate analyses (p < 0.20), as well as relevant clinical variables, were then fitted in a classical multivariate logistic model . This process was independently repeated for each outcome variable: non-compliance regarding 1) clinical decision for treatment, 2) radiotherapy, 3) overall treatment sequence.
As the compliance of many patients at the same LCU could depend on local management practices, we expected data clustering to occur at the level of the LCU. We thus relied on a mixed-effect model by including a random effect in our classical multivariate model . By doing so, the LCUs were assumed to give rise to another source of random variation in addition to the residual variation left unexplained by the fixed-effect variables. The resulting mixed-effect model accommodated for correlations within LCUs, or similarly, heterogeneity of non-compliance between LCUs. We tested this heterogeneity of practices across LCUs by comparing the mixed-effect model to the initial classical logistic model using the likelihood ratio test (LRT) . We reported odds ratios (OR) with their 95% confidence intervals (95% CI) and performed all statistical analyses with the SAS 8.2 software.
Description of breast cancer population (926 patients)
Less than baccalaureate (secondary-school diploma)
Baccalaureate or higher
Medial or lateral
Stage 0 and 1
Histological size (mm)
Hormonal receptor status
At least one receptor positive
Two receptors negative
Peritumoural vascular invasion
Care sector itinerary
Both (public and private)
Status of surgical hospital
Region of surgery
Compliance by care management step
Criteria by step of breast cancer care according to the presence of compliance with Clinical Practices Guidelines (C), justifiable non-compliance (J) or non-compliance (NC) (926 patients)
≥ 10 lymph nodes
Breast or chest
Breast or chest
Time to radiotherapy
Time to chemotherapy
Factors associated with treatment compliance
With regard to clinical decisions for treatment, the rate of non-compliance with CPGs was 71% (657 patients). Variables selected from the univariate analyses and first included in a multivariate fixed-effect model were: age (p = 0.001), education level (p = 0.03), cohabitation status (p = 0.06), stage (p = 0.18), hormonal status (p = 0.06), care sector (p < 0.0001), teaching status of surgical hospital (p = 0.0004) and region where patients underwent surgery (p < 0.0001). After fitting the multivariate fixed-effect model, three factors were associated positively with non-compliance (p < 0.05): age, status of surgical hospital, and region where patients underwent surgery. Introducing the LCUs as a random effect appeared to decrease the residual variability (LRT, p < 0.05), which again indicated heterogeneity in compliance with clinical decisions for treatment, across LCUs and emphasised the need to keep this variable in our model. The mixed-effect model suggested that compared to patients under 50 years of age, patients over 70 years had a two-fold risk of non-compliant care management (OR = 2.1; 95% CI, 1.3-3.6). Similarly, there was a three-fold increase in the risk of non-compliance for one region compared to the second one (OR = 3.0; 95% CI, 1.2-7.4). Finally, note that although the teaching status of the surgical hospital had a significant effect in the fixed-effect model (p = 0.0004), it was no longer significant in the mixed-effect model (p = 0.12).
Of all the therapeutic steps considered, radiotherapy was the least compliant step: we observed at least one non-compliant clinical decision for radiotherapy or radiotherapy procedure criterion for 484 patients (52%). Variables selected from the univariate analyses and first included in a multivariate fixed-effect model were: tumour localisation (p = 0.036), nodal status (p = 0.009), histological size (p = 0.018), hormonal receptor status (p = 0.007), peritumoural vascular invasion (p = 0.001), grade (p = 0.015), and region where patients underwent surgery (p = 0.04). Peritumoural vascular invasion, histological grade and region were positively associated with non-compliance in the multivariate fixed-effect model. Including the LCU as a random effect decreased the residual variability (LRT, p < 0.05), suggesting the presence of heterogeneity of compliance in radiotherapy across LCUs. Based on this mixed-effect model, only the presence of peritumoural vascular invasion remained significant at the 5% level, and it was associated with 1.5-fold risk of non-compliance (OR = 1.5; 95% CI, 1.01-2.3).
Multivariate analyses for non-compliance of overall treatment sequence, therapeutic indications and radiotherapy (926 patients)
Logistic mixed model
No of patients
No of patients
Clinical decisions for treatment (n = 856)
50 - 69
Radiotherapy (clinical decision and procedure) (n = 817)
Peritumoural vascular invasion
Overall treatment sequence (n = 805)
This study showed three main results: BC compliance, factors associated with non-compliance and a multidisciplinary and multi-hospital organisation to BC care which explains the variability in BC practices. Together these results can be used to enhance BC care.
The first results of our study concern BC care compliance that ranged from 88% for hormonal therapy to 48% for radiotherapy, essentially due to the non respect of delay to radiotherapy. With respect to overall treatment sequences, the management of non-metastatic breast cancer was fully compliant (20 criteria) with CPGs in only 12% of cases. This proportion increased to 29% when only clinical decisions for treatment steps were considered. To our knowledge, this is the first time that results of overall BC therapeutic care with details on procedures and clinical decision sequences have been reported in a large BC population. Indeed, most recent publications have only focused on single therapeutic care management step (surgery, radiotherapy or chemotherapy) according to BC stage and most do not provide details on overall compliance with clinical decisions. International studies [1, 3, 6, 17, 23–26] have found a higher compliance rate for performing a biopsy before surgery, radiotherapy and chemotherapy, and a lower rate for performing radiotherapy after conservative breast surgery and hormonal therapy. However, caution is required with such comparisons because many studies did not accurately explain the compliance criteria they used. Two other French regional studies have compared breast cancer management with CPG recommendations and found approximately similar results. The first study (100 patients by audits of medical records)  reported the same proportion of non-compliance for clinical decision for lymph node surgery and for clinical decision for radiotherapy. The second  observed strict compliance with clinical decision for treatment in 1995: 92-96% for surgery, 71-85% for chemotherapy, 72-93% for radiotherapy and 83-94% for hormonal therapy. Our results show similar or higher levels of compliance.
We also reported in our study factors related to BC non-compliance. Older age and region were factors to explain non-compliant clinical decisions for treatment and confirm then results reported previously [3, 15]. Concerning radiotherapy, the least compliant step, non-compliance was associated with three prognostic factors (nodal status, peritumoural vascular invasion and histological grade) and the region of surgical treatment. Finally, we showed an association between overall non-compliance and the presence of positive lymph nodes, histological grade III and region. Lymph node involvement and histological grade are known prognostic factors for breast cancer recurrence and survival. These two clinical characteristics are part of CGPs for breast cancer management and are considered, for example, in making decisions regarding the need for chemotherapy or radiotherapy. Our findings emphasise their impact on overall non-compliance. Concerning the region of BC treatment, a factor associated with non-compliance, the region with a higher compliance rate had already implemented breast cancer CPGs (2004) by local specialist involvement in regional guidelines at the time of the study. This implementation could be explained partially by the regional differences. But some disparity for care accessibility (equipment or personnel resources) between the two regions could explain part of these differences, as reported by others . The more recent publications of factors related to the implementation of CGPs [9–11, 29, 30] showed that patient factors, such as comorbidities or very short life expectancy, reduce the chance that guidelines are followed because they do not encourage the physician to prescribe aggressive therapy. Physician factors related to implementation could be seniority, lack of awareness and limited agreement with guidelines. Moreover, organisation factors such as limited time, work pressure and limited support from peers have been described as barriers to change. Our results showed finally a variability in compliance with CPGs that may be due to the LCU. This discrepancy in practices underlines the need to enhance cancer care coordination and multidisciplinary care at a local level, so that variation in care can be reduced. Whether we modelled the non-compliance of practices within the overall treatment sequence, with regard to clinical decision for treatment, or with regard to radiotherapy only, introducing the LCU as a random variable significantly decreased residual variability. These results confirm the presence of heterogeneity across LCUs, or equivalently, the presence of correlation within LCUs.
This study has a number of strengths and limitations. Firstly, concerning the population, data collection and definition of non-compliance, we cannot exclude a population selection bias since a proportion of women were not included (refusal or non response) and we included probably more patients from hospitals with high volumes (but regional data are available for comparison concerning all stages of cancer, metastatic and non-metastatic BC). We proposed 20 criteria that distinguish clinical decisions for treatment from the procedures resulting from the decision. Such analysis of compliance is necessary to understand better the complexity of the care process, particularly in BC management. A recent synthesis  showed that the complexity of CPGs can be one factor of non-compliance and could so explain a low rate of overall compliance. Our results showed that compliance was lowest for radiotherapy lead time and multidisciplinary approach. Long radiotherapy lead times have already been reported in Italy  and in a critical review of the literature  and are partly due to lack of equipment or human resources . The French national Cancer Plan has attempted to respond to this shortcoming . Patient management in multidisciplinary meetings is recommended for every cancer patient, but this recommendation was too recent (2003) at the date of our study and may explain the relatively low compliance rate of 57%.
We grouped compliance (C) and justifiable non-compliance (J) in the same category, particurlaly to ensure the homogeneity of the data within the non-compliance category (NC). Another statistical choice could be an analysis with three levels of compliance (a polychotomous regression). However, our primary interest was to specifically assess determinants of non-compliance/non conformity to CPGs, without distinguishing compliant and justifiable decisions, which we considered as equivalent. Indeed, from a practical point of view, it is best to specifically target those factors associated with non-compliance so that actions can be undertaken. In view of these objectives, logistic regression using two levels was the optimal modelling approach.
Another potential limitation of our results was the availablity of data in medical records to explain non-compliance and our choice of professionals or organisational factors. We could not collect all data on every professional (for instance, experience or age) but BC management is multidisciplinary and several specialists are involved in medical decisions at each BC management step. Concerning organisational factors, we cannot relate overall compliance to the hospital's volume since the breast cancer care steps are often in different hospitals (surgery in one hospital and radiotherapy in another...) and these hospitals may have different volumes. It therefore seemed that the most interesting data was the LCU. However, we use hospital volume in a recent publication focused on surgery step .
We must underline several points concerning the interpretation of our results. Firstly, a negative impact on outcomes of high non-compliance rate with CPGs for non-metastatic breast cancer management should be interpreted carefully. Each criterion received the same weight whatever its prognostic value and some patients had more non-compliant criteria than others. However, more than two thirds of the patients had fewer than three non-compliant criteria. A specific consensus  amongst local breast cancer experts will be required to develop and apply criteria integrating weights according to prognostic impact. This definition of non-compliance of care management could increase heterogeneity between LCUs, even though variations in practice between LCUs are not surprising since similar variations have been reported for larger institutions . In addition, it should be noted that service improvements for local and regional cancer care organisations (including implementation of LCU) was probably too recent at the date of this study to have an homogeneous impact on medical practices. It would be useful to reproduce this study in another cohort of patients to ascertain if compliance with CPGs has improved over time. Currently, the implementation of the regional guidelines cannot be used as the only factor to explain a better compliance in one region. Indeed, it is well known that adherence with these guidelines depends on many other factors [9–11] such as guideline, professional, patient and environmental characteristics. Our study was implemented between 2003 and 2004, and we studied compliance based on CPGs implemented at this time and in these areas.
Secondly, in the presence of heterogeneity across LCUs, it is important to rely on an appropriate statistical model; otherwise misleading conclusions can be derived. We relied on a mixed-effect model which allowed us to account for the presence of clusters. Other approaches have been suggested to test for the heterogeneity of responses in the context of binary variables, such as the estimation of the median odds ratio as suggested recently . Applying this method did not modify our initial conclusions with respect to heterogeneity across practices.
Finally, this study provides interesting insights into factors of non-compliance in non-metastatic BC management and could lead to quality care improvements. It must be followed by feedback to LCUs and to the regional organisation. Feedback is considered as one of the most efficient methods for improving the application of CPGs . Repeated studies will be required to confirm the likely impact of regional service improvement initiatives over time.
Our findings emphasise also the need for the national scientific and local cancer care organisations to provide further clarification of BC clinical practice guidelines (which may currently be too complex for a complete implementation in local cancer units). For decision makers and national health care authorities, priorities need to be defined in the quality of cancer care regarding care coordination and local cancer units as well as their equipment and personnel resources.
Conflict of interest
The authors declare that they have no competing interests.
We acknowledge Gauthier Bouche, Nicolas Mériau, Nadège Lapeyrere, Isabelle Ingrand for their participation in data collection and analysis, as well as the expert physicians who participated in data validation: C Tunon de Lara (surgeon), M Debled (oncologist), F Valentin (radiologist), H Laharie (radiation oncologist), G MacGrogan (pathologist), P Levillain (pathologist), O Renaud (pathologist), D Cambon (surgeon), D Tariel (surgeon), A Bernard (oncologist), L Prié (radiation oncologist), A Daban (radiation oncologist), D Barret (radiologist) and S Duverge (radiologist).
We thank also all the patients who participated in the study and the oncologists from the two regions who included the patients (REPERES Group): Drs S Abdiche, A Aleba, JC Balzon, A Bernard, O Bernard, N Bonichon-Lamichhane, P Bonichon, V Boulanger, H Bourgeois, V Brousse, L Cany, J Carenco, M Carricaburu, I Catry-Thomas, D Célerier, B Chacon, L Chen, S Chieze, F Coulaud-Pasdeloup, C Cronier, C Dagada, J Dauba, JM Dilhuydy, N Dohollou, JP Dujols, M Durand, JP Dutin, E Fleck, A Floquet, T Germain, P Gesta, D Guichard, D Heymans, N Houédé, D Jaubert, D Langlois, D Larregain-Fournier, P Lefevre, M Lepienne, CB Levache, F Lipinski, M Malet, A Marcu, P Marti, O Maton, L Mauriac, MP Meurisse, B Minier, F Mokhtari, B Ng Ying Kin, JF Paitel, F Pariente, L Prié, C Ragot, P Remuzon, S Roche-Forestier, M Rotarsky, D Schlaifer, JL Suel, JM Tourani, R Trouette, N Trufflandier, A Veret.
We thank Ray Cook and Pippa McKelvie-Sebileau for medical editorial assistance in English and Craig Underhill for his helpful suggestions.
- Guadagnoli E, Shapiro CL, Weeks JC, Gurwitz JH, Borbas C, Soumerai SB: The quality of care for treatment of early stage breast carcinoma: is it consistent with national guidelines?. Cancer. 1998, 83: 302-309. 10.1002/(SICI)1097-0142(19980715)83:2<302::AID-CNCR14>3.0.CO;2-X.View ArticlePubMedGoogle Scholar
- Cardoso F, Piccart MJ: The best use of chemotherapy in the adjuvant setting. Breast. 2003, 12: 522-528. 10.1016/S0960-9776(03)00161-9.View ArticlePubMedGoogle Scholar
- Hand R, Sener S, Imperato J, Chmiel JS, Sylvester JA, Fremgen A: Hospital variables associated with quality of care for breast cancer patients. JAMA. 1991, 266: 3429-3432. 10.1001/jama.266.24.3429.View ArticlePubMedGoogle Scholar
- Bloom BS, de Pouvourville N, Chhatre S, Jayadevappa R, Weinberg D: Breast cancer treatment in clinical practice compared to best evidence and practice guidelines. Br J Cancer. 2004, 90: 26-30. 10.1038/sj.bjc.6601439.View ArticlePubMedPubMed CentralGoogle Scholar
- Ray-Coquard I, Philip T, de Laroche G, Froger X, Suchaud JP, Voloch A, Mathieu-Daudé H, Lurkin A, Farsi F, Bertrand P, et al: Persistence of medical change at implementation of clinical guidelines on medical practice: a controlled study in a cancer network. J Clin Oncol. 2005, 23: 4414-4423. 10.1200/JCO.2005.01.040.View ArticlePubMedGoogle Scholar
- Ray-Coquard I, Thiesse P, Ranchère-Vince D, Chauvin F, Bobin JY, Sunyach MP, Carret JP, Mongodin B, Marec-Bérard P, Philip T, et al: Conformity to clinical practice guidelines, multidisciplinary management and outcome of treatment for soft tissue sarcomas. Ann Oncol. 2004, 15: 307-315. 10.1093/annonc/mdh058.View ArticlePubMedGoogle Scholar
- Schaapveld M, de Vries EG, Otter R, de Vries J, Dolsma WV, Willemse PH: Guideline adherence for early breast cancer before and after introduction of the sentinel node biopsy. Br J Cancer. 2005, 93: 520-528. 10.1038/sj.bjc.6602747.View ArticlePubMedPubMed CentralGoogle Scholar
- White J, Morrow M, Moughan J, Owen J, Pajak T, DesHarnais S, Winchester DP, Wilson JF: Compliance with breast-conservation standards for patients with early-stage breast carcinoma. Cancer. 2003, 97: 893-904. 10.1002/cncr.11141.View ArticlePubMedGoogle Scholar
- Francke AL, Smit CM, de Veer JEA, Mistiaen P: Factors influencing the implementation of clinical guidelines for health care professionals: A systematic meta-review. BMC Medical Informatics and Decision Making. 2008, 8: 38-49. 10.1186/1472-6947-8-38.View ArticlePubMedPubMed CentralGoogle Scholar
- Saillour-Glénisson F, Domecq S, Pouchadon M-L, Jacques B, Sibé M: [Analyse qualitative et quantitative des déterminants à l'application de recommandations professionnelles (RP) par les médecins.]. Rev Epidemiol Sante Publique. 2008, 56: 207-219.View ArticleGoogle Scholar
- Grimshaw JM, Thomas RE, MacLennan G, Fraser C, Ramsay CR, Vale L, Whitty P, Eccles MP, Matowe L, Shirran L, et al: Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technology Assessment. 2004, 8:Google Scholar
- Bouche G, Migeot V, Mathoulin-Pélissier S, Salamon R, Ingrand P: Breast cancer surgery: do all patients want to go to high-volume hospitals?. Surgery. 2008, 143: 699-705. 10.1016/j.surg.2008.03.013.View ArticlePubMedGoogle Scholar
- Defossez G, Mathoulin-Pelissier S, Ingrand I, Gasquet I, Sifer-Riviere L, Ingrand P, Salamon R, Migeot V: Satisfaction with care among patients with non-metastatic breast cancer: development and first steps of validation of the REPERES-60 questionnaire. BMC Cancer. 2007, 7: 129-10.1186/1471-2407-7-129.View ArticlePubMedPubMed CentralGoogle Scholar
- Neal RD, Allgar VL: Sociodemographic factors and delays in the diagnosis of six cancers: analysis of data from the 'National Survey of NHS Patients: Cancer'. Br J Cancer. 2005, 92: 1971-1975. 10.1038/sj.bjc.6602623.View ArticlePubMedPubMed CentralGoogle Scholar
- Hébert-Croteau N, Brisson J, Pineault R: Review of organizational factors related to care offered to women with breast cancer. Epidemiol Rev. 2000, 22: 228-238.View ArticlePubMedGoogle Scholar
- Hébert-Croteau N, Brisson J, Latreille J, Blanchette C, Deschênes : Compliance with consensus recommendations for the treatment of early stage breast carcinoma in elderly women. Cancer. 1999, 85: 1104-1113.View ArticlePubMedGoogle Scholar
- Kingsmore D, Hole D, Gillis C: Why does specialist treatment of breast cancer improve survival? The role of surgical management. Br J Cancer. 2004, 90: 1920-1925. 10.1038/sj.bjc.6601846.View ArticlePubMedPubMed CentralGoogle Scholar
- Roila F, Ballatori E, Patoia L, Palazzo S, Veronesi A, Frassoldati A, Cetto G, Cinieri S, Goldhirsch A: Adjuvant systemic therapies in women with breast cancer: an audit of clinical practice in Italy. Ann Oncol. 2003, 14: 843-848. 10.1093/annonc/mdg256.View ArticlePubMedGoogle Scholar
- American Joint Committee on Cancer (AJCC): Revision of the American Joint Committee on Cancer, VIe edition. Staging System for Breast Cancer. J Clin Oncol. 2002, 20: 3628-3636. 10.1200/JCO.2002.02.026.View ArticleGoogle Scholar
- Direction de la Recherche des Etudes de l'Evaluation et des Statistiques: Les actes chirurgicaux liés au cancer du sein en 1997 à travers le PMSI. Études et résultats. 1999, [http://www.sante-jeunesse-sports.gouv.fr/IMG/pdf/er018.pdf]Google Scholar
- Hosmer DW, Lemeshow S: Applied logistic regression. 2000, New York: Wiley, 2View ArticleGoogle Scholar
- Brown H, Prescott R: Applied mixed models in medicine. 1999, New York: WileyGoogle Scholar
- Wyld L, Garg DK, Kumar ID, Brown H, Reed MW: Stage and treatment variation with age in postmenopausal women with breast cancer: compliance with guidelines. Br J Cancer. 2004, 90: 1486-1491. 10.1038/sj.bjc.6601742.View ArticlePubMedPubMed CentralGoogle Scholar
- Ma M, Bell J, Campbell S, Basnett I, Pollock A, Taylor I: Breast cancer management: is volume related to quality? Clinical Advisory Panel. Br J Cancer. 1997, 75: 1652-1659. 10.1038/bjc.1997.281.View ArticlePubMedPubMed CentralGoogle Scholar
- Malin JL, Kahn KL, Adams J, Kwan L, Laouri M, Ganz PA: Validity of cancer registry data for measuring the quality of breast cancer care. J Natl Cancer Inst. 2002, 94: 835-844.View ArticlePubMedGoogle Scholar
- McEvoy SP, Ingram DM, Byrne MJ, Joseph DJ, Dewar J, Trotter J, Harper C, Haworth C, Harvey JM, Sterrett GF, et al: Breast cancer in Western Australia: clinical practice and clinical guidelines. Med J Aust. 2004, 181: 305-309.PubMedGoogle Scholar
- Calon E: Prise en charge du cancer du sein: évaluation des pratiques en Bretagne. Oncologie. 2004, 6: 583-587. 10.1007/s10269-005-0135-y.View ArticleGoogle Scholar
- Hébert-Croteau N, Freeman CR, Latreille J, Brisson J: Delay in adjuvant radiation treatment and outcomes of breast cancer: a review. Breast Cancer Res Treat. 2002, 74: 77-94.View ArticlePubMedGoogle Scholar
- McCarthy M, Bore J: Treatment of breast cancer in two teaching hospitals: a comparison with consensus guidelines. Eur J Cancer Clin Oncol. 1991, 27: 579-582. 10.1016/0277-5379(91)90222-Y.View ArticleGoogle Scholar
- Rosato R, Sacerdote C, Pagano E, Di Cuonzo D, Baldi I, Bordon R, Ponti A, Bertetto O, Segnan N, Merletti F, et al: Appropriateness of early breast cancer management in relation to patient and hospital characteristics: a population based study in Northern Italy. Breast Cancer Res Treat. 2009, 117: 349-56. 10.1007/s10549-008-0252-6.View ArticlePubMedGoogle Scholar
- Pagano E, Di Cuonzo D, Bona C, Baldi I, Gabriele P, Ricardi U, Rotta P, Bertetto O, Appiano S, Merletti F, et al: Accessibility as a major determinant of radiotherapy underutilization: A population based study. Health Policy. 2007, 80: 483-491. 10.1016/j.healthpol.2006.05.006.View ArticlePubMedGoogle Scholar
- Bouche G, Ingrand I, Mathoulin-Pelissier S, Ingrand P, Breton-Callu C, Migeot V: Determinants of variability in waiting times for radiotherapy in the treatment of breast cancer. Radiother Oncol. 2010, 97: 541-547.View ArticlePubMedGoogle Scholar
- Steimle S: French government launches National Cancer Institute. J Natl Cancer Inst . 2005, 97: 1116-1117.View ArticlePubMedGoogle Scholar
- Linstone HA, Turoff M: The Delphi method: techniques and applications. 1975, Reading (Mass): Addison-WesleyGoogle Scholar
- Edge SB, Stewart A, Patel-Parekh L, Gay G, Palis B, Ko C: Measuring the quality of breast cancer care: National variation in the use of radiation among women under age 70. ASCO Annual Meeting Proceedings Part I. J Clin Oncol. 2004, 25: 599-Google Scholar
- Larsen K, Merlo J: Appropriate assessment of neighborhood effects on individual health: integrating random and fixed effects in multilevel logistic regression. Am J Epidemiol. 2005, 161: 81-88. 10.1093/aje/kwi017.View ArticlePubMedGoogle Scholar
- Durieux P, Ravaud P, Dosquet P, Durocher A: Effectiveness of clinical guideline implementation strategies: systematic review of systematic reviews. Gastroenterol Clin Biol. 2000, 24: 1018-1025.PubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2458/11/45/prepub