Authors' Affiliation: College of Nursing, The Catholic University of Korea, 505 Banpo-dong, Seocho-gu, Seoul, Korea 137-701.
Cancer Nurs. 2010 Jul-Aug;33(4):310-9. doi: 10.1097/NCC.0b013e3181cd2844.
Thyroid cancer incidence in Korean women has increased radically and is the highest in all cancer types. However, the rate of cancer screening among women is very low.
The aim of the study was to determine the risk factors for thyroid cancer and to develop a predictive model based on these risk factors.
The study design comprised a literature review and a case-control study. To construct a predictive model, the participants selected were 260 female outpatients diagnosed with malignant neoplasm of thyroid gland who had undergone thyroid removal surgery. A total of 259 people for the control group were selected by adopting a 5-year age-matching method.
From the literature review, 6 categories of risk factors were identified. Nine variables, including occupation, live(d) in coastal region, family history of thyroid cancer, history of benign thyroid tumor, menopause status and weight gain, number of full-term deliveries, abortion, exercise intensity, and stress, remained as statistically significant risk factors in the stepwise regression model. Regarding the predictive power of the model, the area under the receiver operating characteristic curve was .79, accuracy was .77, sensitivity was .89, specificity was .65, positive predictive value was .72, and negative predictive value was .85.
The predictive power of the model was relatively good, so it can be used to identify individuals at high risk for thyroid cancer.
The predictive model can be used in promoting to participate in early cancer-screening tests. Thus, it will be possible to detect thyroid cancer in its earliest stage, diminish mortality, and improve quality of life.
韩国女性甲状腺癌的发病率呈急剧上升趋势，在所有癌症类型中发病率最高。然而，女性癌症筛查率却非常低。
本研究旨在确定甲状腺癌的危险因素，并基于这些危险因素建立预测模型。
研究设计包括文献回顾和病例对照研究。为了构建预测模型，选择了 260 名经甲状腺切除术诊断为甲状腺恶性肿瘤的女性门诊患者作为病例组。采用 5 年年龄匹配法选择了 259 名对照组。
从文献回顾中确定了 6 类危险因素。在逐步回归模型中，9 个变量，包括职业、居住在沿海地区、甲状腺癌家族史、良性甲状腺肿瘤史、绝经状态和体重增加、足月分娩次数、流产、运动强度和压力，仍然是统计学上显著的危险因素。该模型的预测能力方面，受试者工作特征曲线下面积为 0.79，准确率为 0.77，敏感度为 0.89，特异度为 0.65，阳性预测值为 0.72，阴性预测值为 0.85。
该模型的预测能力相对较好，因此可以用于识别甲状腺癌高危个体。
预测模型可用于促进人们参与早期癌症筛查试验。因此，有可能在早期发现甲状腺癌，降低死亡率，并提高生活质量。
