• 2022-09
  • 2022-08
  • 2022-07
  • 2022-05
  • 2022-04
  • 2021-03
  • 2020-08
  • 2020-07
  • 2018-07
  • br There are several limitations to this study that we


    There are several limitations to this study that we would like to acknowledge. For the sake of analytic feasi-bility and ease of presentation, we incorporated a reduc-tionist approach in conducting our analyses. A more expansive approach, by including all possible variables, might have yielded different probabilities. In addition, refinement of our model, for example, by using ICD-10 instead of ICD-9 codes, has the potential to ferret out the individual codes rather than comorbidity groupings, which have the largest influence on mortality. In addition, interactions between comorbid conditions (eg obesity) and the risk of dying from breast cancer were not exam-ined in this model. We limited our patient ABT-263 to stage I breast cancer and it is quite likely that our calcu-lated probabilities of competing risks of death would change with increasing breast cancer stages. We did not consider the influence of adjuvant chemotherapy, radia-tion treatment, or hormone treatment on our calculated probabilities. Our data are retrospective and therefore subject to selection bias, misclassification bias, and 
    missing variables. In particular, the use of the cause of death variable, although validated by studies, is only as good as the information available on the death certificate.
    Table 3. Competing Risk Sub-Distribution Hazards for Breast Cancer Death
    36 Wasif et al Discussion J Am Coll Surg
    Quantifying competing risks of mortality, primarily those of breast cancer death, death from other cancers, and non-cancer death, in patients with stage I breast cancer is possible using registry data. This information can be used to develop a predictive mortality model using patient age, comorbidity groupings, and ER status to aid in clin-ical decision making. Risk stratification of patients in clin-ical trials for efficacy should also consider incorporating comorbidity information. In addition, reporting of long-term cancer survival when censoring patients who die from comorbid conditions has the potential to artifi-cially inflate disease-free survival estimates.15 Additional development of this model, including incorporation of treatment variables, such as adjuvant chemotherapy or ICD-10 codes and external validation, can be used to develop online calculators that take into account competing risks of death when calculating survival bene-fits from treatment.
    Author Contributions
    Study conception and design: Wasif, Neville, Pockaj Acquisition of data: Wasif, Neville, Pockaj
    Analysis and interpretation of data: Wasif, Neville, Gray, Cronin, Pockaj
    Drafting of manuscript: Wasif, Neville, Gray, Cronin, Pockaj Critical revision: Wasif, Neville, Gray, Cronin, Pockaj
    1. American Cancer Society. Breast Cancer Facts and Fig-ures 2017-2018. Available at: dam/cancer-org/research/cancer-facts-and-statistics/breast-cancer-facts-and-figures/breast-cancer-facts-and-figures-2017-2018. pdf. Accessed March 20, 2019.
    2. Hughes KS, Schnaper LA, Bellon JR, et al. Lumpectomy plus tamoxifen with or without irradiation in women age 70 years
    3. Martelli G, Miceli R, Daidone MG, et al. Axillary dissection versus no axillary dissection in elderly patients with breast can-
    4. Society of Surgical Oncology. Choosing Wisely. Don’t
    routinely use sentinel node biopsy in clincally node negative women 70 years of age with hormone receptor positive inva-sive breast cancer. Available at: http://www.choosingwisely. org/clinician-lists/sso-sentinel-node-biopsy-in-node-negative-women-70-and-over/. Accessed March 20, 2019.
    5. Warren JL, Klabunde CN, Schrag D, et al. Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population. Med Care 2002;40:IV-3-18.
    6. Howlader N, Ries LA, Mariotto AB, et al. Improved estimates