Different statistical techniques dealing with confounding in observational research: measuring the effect of breast-conserving therapy and mastectomy on survival

Abstract

Purpose Propensity trimming, hierarchical modelling and instrumental variable (IV) analysis are statistical techniques dealing with confounding, cluster-related variation or confounding by severity. This study aimed to explain (dis)advantages of these techniques in estimating the effect of breast-conserving therapy (BCT) and mastectomy on 10-year distant metastasisfree survival (DMFS). Methods All women diagnosed in 2005 with primary T1-2N0-1 breast cancer treated with BCT or mastectomy were selected from the Netherlands Cancer Registry. We used multivariable Cox regression to correct for confounding. Propensity trimming was used to create a more homogeneous population for which the treatment choice was not self-evident. Hospital of surgery was used as hierarchical level to handle hospital-related variation, and as IV to deal with unmeasured confounding. Results Multivariable Cox regression showed higher 10-year DMFS for BCT than mastectomy ]. Propensity trimming on the 10-90th and the 20-80th percentile of the propensity score distribution and hierarchical modelling showed similar HRs. IV analysis showed no significant difference between BCT and mastectomy. Conclusion Unmeasured confounding is very difficult to eliminate in observational research. We cannot conclude that BCT or mastectomy has a causal relationship with 10-year DMFS. It is crucial to critically evaluate all model's assumptions, and to be careful in drawing firm conclusions.

Improved Survival after Breast-Conserving Therapy Compared with Mastectomy in Stage I-IIA Breast Cancer

“…This finding is in agreement with the reports of other studies, which are all convincing and uniformly show DFS and/or OS benefits for BCT compared to mastectomy in patients with stage I-II breast cancer (Table 6). Equivalent or improved outcomes with BCT as compared with mastectomy in terms of locoregional control, BCSS, DFS, or OS have been reported regardless of age [14,15], intrinsic breast cancer subtype [48][49][50][51], pathological tumor stage [11,14,15,51], overall breast cancer stage [52], and grade [51]. Recently, a Swedish cohort study using prospectively collected data of women with stage T1-2 N0-2 breast cancer confirmed better survival with BCT vs. mastectomy (irrespective of radiation therapy) even when taking into account comorbidity and socioeconomic status in both node-negative and node-positive disease, pointing out that offering more extensive surgery to patients who are suitable for either breast-conserving surgery or mastectomy is not saving lives [51].…”

Effect of omission of surgery on survival in patients aged 80 years and older with early-stage hormone receptor-positive breast cancer

“…Although this was sufficient to demonstrate a survival difference in the primary analysis, the findings for the stratified analysis suggest a lack of power. Although confounding by unmeasured factors can theoretically be avoided using the instrumental variable approach, an instrument that meets all of the required assumptions is not always available in clinical data 23,34 . There was a small age difference across the groups in the present study.…”

Implementing a breast cancer patient decision aid: Process evaluation using medical files and the patients’ perspective

“…For most women diagnosed with early‐stage breast cancer (stage I or II), breast‐conserving therapy (BCT) and mastectomy are equally effective in terms of survival (Fisher et al, 2002 ; Litiere et al, 2012 ; van Maaren et al, 2019 ; McGuire et al, 2009 ), which indicates that the patient's preference should probably dominate the decision. Preferences that are often mentioned by patients are related to the level of fear and anxiety, perceived survival odds, body image values, attitudes towards radiotherapy, the preference or perceived preference of the surgeon, the communication between patient and professional, and the patient's involvement in the decision‐making process (Caldon et al, 2011 ; Hershman et al, 2009 ).…”

“…For most women diagnosed with early-stage breast cancer (stage I or II), breast-conserving therapy (BCT) and mastectomy are equally effective in terms of survival (Fisher et al, 2002; Litiere et al, 2012;van Maaren et al, 2019;McGuire et al, 2009), which indicates that the patient's preference should probably dominate the decision.…”