Delays in diagnosis and bladder cancer mortality

Authors:
  • DWDavid P. Wood
    David P. Wood
Open Access
Publisher website
Google Scholar
Abstract
BACKGROUND: Mortality from invasive bladder cancer is common, even with high‐quality care. Thus, the best opportunities to improve outcomes may precede the diagnosis. Although screening currently is not recommended, better medical care of patients who are at risk (ie, those with hematuria) has the potential to improve outcomes.METHODS: The authors used the Surveillance, Epidemiology, and End Results‐Medicare linked database for the years 1992 through 2002 to identify 29,740 patients who had hematuria in the year before a bladder cancer diagnosis and grouped them according to the interval between their first claim for hematuria and their bladder cancer diagnosis. Cox proportional hazards models were fitted to assess relations between these intervals and bladder cancer mortality, adjusting first for patient demographics and then for disease severity. Adjusted logistic models were used to estimate the patient's probability of receiving a major intervention.RESULTS: Patients (n = 2084) who had a delay of 9 months were more likely to die from bladder cancer compared with patients who were diagnosed within 3 months (adjusted hazard ratio [HR], 1.34; 95% confidence interval [CI], 1.20‐1.50). This risk was not markedly attenuated after adjusting for disease stage and tumor grade (adjusted HR, 1.29; 95% CI, 1.14‐1.45). In fact, the effect was strongest among patients who had low‐grade tumors (adjusted HR, 2.11; 95% CI, 1.69‐2.64) and low‐stage disease (ie, a tumor [T] classification of Ta or tumor in situ; adjusted HR, 2.02; 95% CI, 1.54‐2.64).CONCLUSIONS: A delay in the diagnosis of bladder cancer increased the risk of death from disease independent of tumor grade and or disease stage. Understanding the mechanisms that underlie these delays may improve outcomes among patients with bladder cancer. Cancer 2010. © 2010 American Cancer Society.

This publication has 28 references indexed in Scilit: