Cronin-Fenton D P, Nørgaard M, Jacobsen J, Garne J P, Ewertz M, Lash T L, Sørensen H T
Department of Clinical Epidemiology, Aarhus University Hospital, Ole Worms Allé 1150, Aarhus C 8000, Denmark.
Br J Cancer. 2007 May 7;96(9):1462-8. doi: 10.1038/sj.bjc.6603717. Epub 2007 Apr 3.
Comorbid diseases can affect breast cancer prognosis. We conducted a population-based study of Danish women diagnosed with a first primary breast cancer from 1995 to 2005 (n=9300), using hospital discharge registry data to quantify comorbidities by Charlson score. We examined the influence of comorbidities on survival, and quantified their impact on relative mortality rates. The prevalence of patients with a Charlson score='0' fell from 86 to 81%, with an increase in those with Charlson score='1-2' from 13 to 16%, and score='3+' from 1 to 2%. One- and five-year survival for patients with Charlson score='0' and '1-2' was better for those diagnosed in 1998-2000 than in 1995-1997. Overall, patients diagnosed in 2001-2004 (mortality ratio (MR)=0.80, 95% CI=0.68-0.95) and 1998-2000 (MR=0.92, 95% CI=0.78-1.09) had lower 1-year age-adjusted mortality compared to those diagnosed in 1995-1997 (reference period). Patients with Charlson scores '1-2' and '3+' had higher age-adjusted 1-year mortality than those with a Charlson score='0' in each time period (2001-2004: MR('1-2')=1.76, 95% CI=1.35-2.30, and MR('3+')=3.78, 95% CI=2.51-5.68; and 1998-2000: MR('1-2')=1.60, 95% CI=1.36-1.88 and MR('3+')=2.34, 95% CI=1.65-3.33). Similar findings were observed for 5-year age-adjusted mortality. Additional analyses, adjusted for stage, indicated that confounding by stage could not explain these findings. Despite continued improvements in breast cancer survival, we found a trend of poorer survival among breast cancer patients with severe comorbidities even after adjusting for age and stage. Such poorer survival is an important public health concern and can be expected to worsen as the population ages.
合并症会影响乳腺癌的预后。我们对1995年至2005年诊断为原发性乳腺癌的丹麦女性进行了一项基于人群的研究（n = 9300），利用医院出院登记数据通过查尔森评分来量化合并症。我们研究了合并症对生存率的影响，并量化了它们对相对死亡率的影响。查尔森评分为“0”的患者比例从86%降至81%，查尔森评分为“1 - 2”的患者比例从13%增至16%，评分为“3 +”的患者比例从1%增至2%。1998 - 2000年诊断的查尔森评分为“0”和“1 - 2”的患者的1年和5年生存率高于1995 - 1997年诊断的患者。总体而言，与1995 - 1997年（参考期）诊断的患者相比，2001 - 2004年诊断的患者（死亡率比（MR）= 0.80，95%可信区间= 0.68 - 0.95）和1998 - 2000年诊断的患者（MR = 0.92，95%可信区间= 0.78 - 1.09）的1年年龄调整死亡率较低。在每个时间段（通过调整年龄），查尔森评分为“1 - 2”和“3 +”的患者的1年年龄调整死亡率高于查尔森评分为“0”的患者（2001 - 2004年：MR（“1 - 2”）= 1.76，95%可信区间= 1.35 - 2.30，MR（“3 +”）= 3.78，95%可信区间= 2.51 - 5.68；1998 - 2000年：MR（“1 - 2”）= 1.60，95%可信区间= 1.36 - 1.88，MR（“3 +”）= 2.34，95%可信区间= 1.65 - 3.33）。5年年龄调整死亡率也观察到类似结果。经分期调整的进一步分析表明，分期混淆不能解释这些结果。尽管乳腺癌生存率持续提高，但我们发现即使在调整年龄和分期后，患有严重合并症的乳腺癌患者的生存趋势仍较差。这种较差的生存是一个重要的公共卫生问题，并且随着人口老龄化可能会恶化。
