背　景 成人健檢資料中，潛藏著一些未曾探索的重要資訊與知識。資料挖掘方法可以做資料類型的發覺與萃取，而對一資料組做最佳的詮釋。
目　的 藉由資料挖掘技術，以找出成人健檢資料庫中，健檢數值異常項目的組合。
方　法 本研究以台灣中部某市府所屬機構員工體檢的21,570筆資料為資料庫，資料分析分成二大階段：一般統計分析與資料挖掘分析。一般統計分析稱為資料的前處理，資料處理後，以供第二階段資料挖掘用。
結　果 近乎三分之二的受檢人，其體脂肪率皆為異常，即體內囤積過多的脂肪。收縮壓偏高者為30.8%，舒張壓偏高者有18.4%；空腹血糖異常者70.9%，膽固醇偏高者有64.0%。SGPT異常佔17.0%，尿酸有11.9%受檢者過高，尿蛋白則有17.6%受檢者呈現異常。運用資料挖掘技術中的關聯規則來描述健檢數值異常項目組合間的關聯性，發現成人健檢異常項目組合的46項關聯規則，可見為代謝症候羣的組合，即體脂肪率異常、收縮壓與舒張壓升高、空腹血糖異常、膽固醇異常；及SGPT的異常會伴隨五成～七成的代謝症候羣症狀出現。
結　論／實務應用 有鑒於代謝症候群與心臟血管疾病、非酒精性脂肪肝、糖尿病皆有相關性，而三高症候羣：高血糖、高血脂、與高血壓這些症狀間又是息息相關，因此建議在健康指導上力求打破其間的某一環節，以有助於成人健康狀態的改善。

Background: Important but non-intuitive information and knowledge in the adult health check-up database can be found, extracted and organized using data mining.
Purpose: This study applied data mining techniques to enhance understanding of abnormal item combinations in adult health check-up data.
Methods: Researchers collected a total of 21,570 health check-up records for use in this study. Data analysis consisted of the two parts of general statistics and data mining. General statistics was a pre-process used to clarify and organize data in preparation for data mining.
Results: Results of this study revealed that nearly two-thirds of subjects had an abnormal percentage of body fat (i.e., were clinically overweight). Those with abnormally high systolic blood pressure, diastolic blood pressure, fasting blood sugar, cholesterol, and SGPT comprised, respectively, 30.8%, 18.4%, 70.9%, 64.0% and 17.0% of all sample subjects. One-tenth (11.9%) was identified as having hyperuricemia, and 17.6% as having proteinuria. Using data mining association rules analysis, we identified a total of 46 rules. Among these included the combination for metabolic syndrome (abnormal readings in body fat percentage, blood pressure, fasting blood sugar and cholesterol). A majority (50-70%) of metabolic syndrome symptoms were associated with elevated SGPT.
Conclusions / Implications for practice: Metabolic syndrome is increasingly recognized as a major cause of cardiovascular disease, nonalcoholic fatty liver disease, and diabetes mellitus. This study suggests the importance of breaking the linkages amongst hyperglycemia, hyperlipidemia and hypertension of the metabolic syndrome to improve adult health.