• 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • br values for last contact days


    values for last contact, days from surgery, or 30-day mortality. To ensure available for analysis. After data processing, the analytic
    The Journal of Thoracic and Cardiovascular Surgery c Volume 158, Number 2 571
    Thoracic: Lung Cancer Moore et al
    TABLE 1. Unadjusted 30-day and 90-day mortality descriptive statistics
    30-d Mortality
    90-d Mortality
    n (%) Mortality P value*
    Mortality P value*
    Insurance status
    Not insured
    Private insurance
    Other government
    Median income
    Year of diagnosis
    Primary sitey
    C340—Main bronchus
    Analytic stage
    572 The Journal of Thoracic and Cardiovascular Surgery c August 2019
    Moore et al Thoracic: Lung Cancer
    TABLE 1. Continued
    30-d Mortality
    90-d Mortality
    Surgical procedure
    Facility type
    Surgery Approach
    Urban vs rural
    Average annual procedure volume (cases)
    Overall LOSx
    SCC, Squamous cell cancer; BAC, bronchioloalveolar cancer; Well diff, well differentiated; Mod diff, moderately differentiated; Poorly diff, poorly differentiated; Undiff, anaplastic, undifferentiated, anaplastic; NeoAdj, neoadjuvant; VATS, video-assisted thoracic surgery; Open (inc conv), open, including conversion to PER977 open; LOS, length of stay. *Calculated using c2 analyses and Kruskal-Wallis testing. yAnatomic location of the tumor. The coding is derived from the ICD-O-3 topographical classification.10 zDefined as a tumor that PER977 overlaps the boundaries of 2 or more subsites and the point of origin is unknown. xValues are presented as median (interquartile range).
    The Journal of Thoracic and Cardiovascular Surgery c Volume 158, Number 2 573
    Thoracic: Lung Cancer Moore et al
    Patient Characteristics and Unadjusted Analyses
    Overall, the 90-day mortality was 5.7% compared with a
    30-day mortality of 3.0%. In both groups, higher mortality was significantly associated with increasing age, male gender, increasing number of comorbidities, Medicare or uninsured status, lower income, earlier year of diagnosis, primary site of tumor, analytic stage, tumor histology, tu-mor grade, receipt of neoadjuvant therapy, surgical proced-ure type, facility type, surgical approach, facility location (city size), facilities that perform fewer than 20 procedures per year, and increased length of stay (Table 1).
    Adjusted 30-Day and 90-Day Overall Mortality
    After adjusting for variables in Model 4, 30-day and 90-day mortality regression models demonstrated significance for several patient, tumor, and hospital factors (Table 2). Because clinically relevant variables were selected for regression modeling, most were significantly associated with mortality at each time point. For example, across both models, increasing age was associated with increased mortality. Patients older than age 80 years had a 4 times increased risk of dying compared with patients younger 
    than age 50 years. Male gender increased odds of mortality by more than 50%. Compared with patients without insur-ance, private insurance reduced the risk of death in each mortality group approximately 30%. For each unit increase in median income more than $38,000, there was a reduction of mortality odds by almost 10% in both mortality groups. Squamous cell carcinoma histology was also associated with an almost 40% increased risk of death in both mortal-ity groups. The biggest reduction in risk was gained when comparing academic facilities to community cancer pro-grams (30-day mortality OR, 0.68; 95% CI, 0.60-0.077 and 90-day mortality OR, 0.72; 95% CI, 0.65-0.80). Compared with facilities that performed 1 to 3 procedures annually, those facilities that performed more than 20 pro-cedures annually were associated with 32% and 22% reduction of risk at 30-day and 90-day mortality, respec-tively. After adjusting for other covariates, differences in neither surrounding population size near a facility nor race were significantly associated with increased mortality.
    In addition to the many similarities mentioned above and shown in Table 2, there was a difference between the 2 models. Receipt of neoadjuvant therapy was not signifi-cantly associated with mortality in the 30-day mortality model; however, it was significant in the 90-day mortality model. For example, although use of neoadjuvant therapy was not significantly associated with mortality at 30-days, it was associated with a 10% increased risk of mortality at 90 days (90-day mortality OR, 1.10; 95% CI, 1.03-1.18).