Katsoulis Michail, Kyrozis Andeas, Trichopoulou Antonia, Bamia Christina, Trichopoulos Dimitrios, Lagiou Pagona
Hellenic Health Foundation, 13 Kaisareias Street, 115 27, Athens, Greece,
Cancer Causes Control. 2014 Nov;25(11):1565-70. doi: 10.1007/s10552-014-0460-9. Epub 2014 Aug 22.
To examine whether the documented association of suboptimal cognitive function with total and cardiovascular (CVD) mortality also applies to cancer mortality and probe whether the explanation for this association is biomedical or health care related.
In a subsample of 733 participants of the EPIC-Greece cohort from Athens and surrounding area, we assessed cognitive function at age 65 or older in the period 2004-2006, using the Mini-Mental State Examination (MMSE). Incidence of cancer, mortality from cancer and CVD, and overall mortality were ascertained through active follow-up for a median of 4 years after MMSE assessment using Cox proportional hazards models.
A total of 86 participants died during follow-up. A 2-point decrease in MMSE score was associated with increase in overall (hazard ratio (HR) 1.26, 95 % confidence interval (CI) 1.11-1.43), CVD (HR 1.26, 95 % CI 1.02-1.56), and cancer (HR 1.32, 95 % CI 1.02-1.70) mortality. In contrast, there was no noticeable difference in cancer incidence associated with a 2-point decrease in MMSE score (HR 1.07, 95 % CI 0.79-1.45).
Cognitive function appears to be inversely associated not only with CVD and overall, but also with cancer mortality. Although for CVD mortality there is a biomedical explanation invoking vascular mechanisms, for cancer mortality we may need to focus on socially conditioned factors, such as compromised ability to identify early signs and suboptimal compliance to treatment. Our hypothesis-generating results need to be confirmed in larger studies, as the issue is of major importance, since cognitive decline is not uncommon among the elderly.
探讨已被记录的认知功能欠佳与全因死亡率及心血管疾病（CVD）死亡率之间的关联是否也适用于癌症死亡率，并探究这种关联的解释是与生物医学相关还是与医疗保健相关。
在来自雅典及周边地区的EPIC - 希腊队列研究的733名参与者的子样本中，我们于2004 - 2006年期间使用简易精神状态检查表（MMSE）对65岁及以上参与者的认知功能进行了评估。通过积极随访，利用Cox比例风险模型确定了MMSE评估后中位4年期间的癌症发病率、癌症和CVD死亡率以及全因死亡率。
共有86名参与者在随访期间死亡。MMSE评分降低2分与全因死亡率（风险比（HR）1.26，95%置信区间（CI）1.11 - 1.43）、CVD死亡率（HR 1.26，95% CI 1.02 - 1.56）和癌症死亡率（HR 1.32，95% CI 1.02 - 1.70）的增加相关。相比之下，MMSE评分降低2分与癌症发病率之间没有明显差异（HR 1.07，95% CI 0.79 - 1.45）。
认知功能似乎不仅与CVD和全因死亡率呈负相关，还与癌症死亡率呈负相关。虽然对于CVD死亡率有一个援引血管机制的生物医学解释，但对于癌症死亡率，我们可能需要关注社会条件因素，如识别早期症状的能力受损和对治疗的依从性欠佳。由于认知衰退在老年人中并不罕见，这个问题至关重要，我们这些产生假设的结果需要在更大规模的研究中得到证实。
