No abstract

No abstract

No abstract

There is debate as to whether cohort studies are valid when they are based on a source population that is non-representative of a given general population. This baseline selection may introduce collider bias if the exposure of interest and some other outcome risk factors affect the probability of being in the source population, thus altering the associations between the exposure and those risk factors. We argue that this mechanism is not specific to ‘selected cohorts’ and also occurs in ‘representative cohorts’ due to the selection processes that occur in any population. These selection processes are for example linked to the life status, immigration and emigration, which, in turn, may be affected by environmental and social determinants, lifestyles and genetics. We provide real-world examples of this phenomenon using data on the population of the Piedmont region, Italy. In addition to well-recognised mechanisms, such as shared common causes, the associations between the exposure of interest and the risk factors for the outcome of interest in any source population are potentially shaped by collider bias due to the underlying selection processes. We conclude that, when conducting a cohort study, different source populations, whether ‘selected’ or ‘representative’, may lead to different exposure–outcome risk factor associations, and thus different degrees of lack of exchangeability, but that one approach is not inherently more or less biased than the other. The key issue is whether the relevant risk factors can be identified and controlled.

Administrative data may provide valuable information for monitoring the quality of care at population level and offer an efficient way of gathering data on individual patterns of care, and also to shed light on inequalities in access to appropriate medical care. The aim of the study was to investigate the role of patient and hospital characteristics in the initial treatment of early breast cancer using administrative data. Incident breast cancer patients were identified from hospital discharge records and linked to the radiotherapy outpatient database during 2000-2004 in the Piedmont region of Northwestern Italy. Women treated with breast-conserving surgery followed by radiotherapy (BCS ? RT) were compared to those treated with BCS without radiotherapy (BCS w/o RT) or mastectomy using multinomial logistic regression models. Out of 16,022 incident cases, 46.2% received BCS ? RT, 20.3% received BCS w/o RT, and 33.5% received a mastectomy. Compared to BCS ? RT, the factors associated with BCS w/o RT were: increased age (OR = 1.54; 95% CI = 1.29-1.85, for ages 70-79 vs. \50), being unmarried (1.24; 1.13-1.36), presence of co-morbidities (1.32; 1.10-1.58), being treated at hospitals with low surgical volume (1.31; 1.07-1.60 for hospitals with less than 50 vs. C150 interventions/year), and living far from radiotherapy facilities (1.75; 1.39-2.20 for those at a distance of [45 min). These same factors were also associated with mastectomy. During the 5-year period observed, there was a trend of reduced probability of receiving a mastectomy (0.70; 0.56-0.88 for 2004 vs. 2000). The presence or absence of nodal involvement was positively associated with mastectomy (2.28; 1.83-2.85) and negatively associated with BCS w/o RT (0.65; 0.56-0.76). After adjustment for potential confounders, education level did not show any association with the type of treatment. Social and geographical factors, in addition to hospital specialization, should be considered to reduce inappropriateness of care for breast cancer.

No abstract

No abstract

BackgroundIt has been documented that variations exist in breast cancer treatment despite wide dissemination of clinical practice guidelines. The aim of this population-based study was to evaluate the impact of regional guidelines (Piedmont guidelines, PGL) for breast cancer diagnosis and treatment on quality-of-care indicators in the Northwestern Italian region of Piedmont.MethodsWe included two samples of women aged 50–69 years with incident breast cancer treated in Piedmont before and after the introduction of PGL: 600 in 2002 (pre-PGL) and 621 in 2004 (post-PGL). Patients were randomly selected among all incident breast cancer cases identified through the hospital discharge records database. We extracted clinical data on breast cancer cases from medical charts and ascertained vital status through linkage with town offices. We assessed compliance with 14 quality-of-care indicators from PGL recommendations, before and after their introduction in clinical practice.ResultsAmong patients with invasive lesions, 77.1% (N = 368) and 77.5% (N = 383) in the pre-PGL and post-PGL groups, respectively, received breast conservative surgery (BCS) as a first-line treatment. Following BCS, 87.7% received radiotherapy in 2002, compared to 87.9% in 2004. Of all patients at medium-to-high risk of distant metastasis, 65.5% (N = 268) and 63.6% (N = 252) received chemotherapy in 2002 and in 2004, respectively. Among the 117 patients with invasive lesions and negative estrogen receptor status in 2002, hormonal therapy was prescribed in 23 of them (19.6%). The incorrect prescription of hormonal therapy decreased to 10.8% (N = 10) among the 92 estrogen receptor-negative patients in 2004 (p < 0.01).Compliance with PGL recommendations was already high in the pre-PGL group, although some quality-of-care indicators did not reach the standard. In the pre/post analysis, 8 out of 14 quality-of-care indicators showed an improvement from 2002 to 2004, but only 4 out of 14 reached statistical significance. We did not find any change in the risk of mortality in the post-PGL versus the pre-PGL group (adjusted hazard ratio 0.94, 95%CI 0.56–1.56).ConclusionsThese results highlight the need to continue to improve breast cancer care and to measure adherence to PGL.