背景：高血壓病人未遵從醫師指示服藥，不但影響血壓控制，還會增加合併症發生與死亡。目的：探討影響高血壓病人服藥遵從行為因素，並建立血壓與遵從行為之關係。方法：以臺灣東部某醫院546位門診高血壓病人為研究對象進行橫斷研究，採結構式問卷與實測血壓進行調查。研究使用SPSS20.0統計軟體進行資料分析，除進行描述性統計外，遵從行為佳與不佳兩組在人口學變項、過去病史與血壓類別變項差異進行卡方檢定和費雪精確檢定，連續變項差異進行獨立樣本t檢定，並以複迴歸分析探討遵從行為與血壓關係。結果：30%高血壓病人為遵從行為不佳；遵從行為佳與不佳二組個案在藥物種類數、藥物副作用與藥物取得來源呈顯著差異；但遵從行為佳與不佳二組個案在藥物顆數、服藥頻率、合併症、罹病年數與有無使用另類療法，都無顯著差異；根據實測血壓值，遵從行為佳與不佳二組個案的舒張壓與平均動脈壓呈顯著差異，收縮壓無顯著差異；但當分別考慮收縮壓與舒張壓時，遵從行為佳與不佳二組個案的收縮壓與舒張壓控制優劣人數呈顯著差異；藥物副作用與藥物取得來源對遵從行為有顯著預測力（p＜.05）。遵從行為分別對收縮壓與舒張壓有顯著預測力（p＜.05）。結論：藥物副作用與藥物取得來源，是影響臺灣病人服藥遵從行為的危險因素之一，而服藥遵從行為優劣可用來預測病人血壓值。

Background: Uncontrolled blood pressure (BP) attributed to medication non‐adherence may increase the risks of complications and of death. Purpose: This study aimed to identify factors that influence poor medication adherence in hypertensive patients and to establish the relationship between BP and medication adherence. Methods: This cross‐sectional study enrolled 546 outpatients with hypertension from a regional hospital in eastern Taiwan. Data collected from participants included BP measurements and answers to a structured questionnaire. Descriptive statistics were performed for all measures as appropriate. Assessments of differences in adherence were based on demographic, medical history, and BP. Chi‐Square test and Fisher's exact test for categorical variables or independent t‐test for continuous variables were performed. The relationship between adherence and BP was analyzed using multiple regression analysis. All data were analyzed using SPSS 20.0 statistical analysis software. Results: A total of 30% of patients were classified in the poor adherence category. A significant difference was found between adherence and poor adherence groups in terms of: type of medication, side effects, and sources of medications. No significant difference, however, was found across the two adherence groups in terms of: number of pills, medication schedule, duration of hypertension and its related complications, and use of complementary‐and‐alternative medicine (CAM). The diastolic blood pressure (DBP) and mean arterial pressure (MAP) differed significantly between the two groups. However, no significant difference in systolic blood pressure (SBP) was identified between the two groups. When SBP and DBP were considered separately, the number of patients who had good and poor control of SBP and DBP differed significantly between the two groups. Side effects and sources of medications were significant predictors of adherence (p ＜ .05). Adherence was a significant predictor of SBP and DBP (p ＜ .05). Conclusions: Side effects and sources of medications were identified as factors that influence the medication adherence of Taiwanese patients, which was identified as a reliable predictor of BP outcome.