目的：台灣因少子化、家庭結構轉變，家庭成員照顧功能降低，長照機構成為老年人照顧的另一種選擇。本研究旨在瞭解長照機構照服員的工作壓力源及職業疲勞，並探討二者間的關係。 方法：本研究為一橫斷性研究，資料取自「台灣地區長期照護機構照顧服務員工作壓力、工作適應與健康結果調查」。依全台分層比例抽樣法，抽取64間長照機構，共計 580位照服員，有效問卷為507份，有效問卷回收率為87.41%。研究工具包含中文版工作內容問卷、職場疲勞量表、及受訪者資料。統計方法包含描述性統計、皮爾森積差相關及多層次線性迴歸分析。 結果：507位照服員平均年齡43.19歲(SD=11.15)，以女性居多(87.38%)。迴歸結果顯示，照服員年齡高者(β=-0.19, p<0.001)、外國籍者(β=-0.12, p=0.050)、任職機構開業年數長者(β=-0.18, p<0.001)、任職機構規模大者(β=-0.22, p<0.001)、心理負荷低者(β=-0.35, p<0.001)及社會支持高者(β=-0.13, p=0.010)，其職業疲勞較低。 結論：本研究證實工作壓力源中的心理負荷及社會支持顯著影響照服員的職業疲勞。管理者應定期瞭解照服員之工作壓力源及職業疲勞，並針對年輕、本國籍、自覺心理負荷高及社會支持低等高風險照服員，主動協助調整其壓力感受，增加與其人際互動，建立彼此的情誼、信任及支持，並發展紓壓策略，減輕工作壓力及職業疲勞。

OBJECTIVES. Affected by the lower birth-decline and family dysfunctions, institutional long-term care facilities have become an option for many elderly populations. This study aims to examine the effect of job stressors on burnout for nursing assistants in Taiwan long-term care facilities. METHODS. This cross-sectional study uses a structured-questionnaire with job stressors assessed by C-JCQ and burnout assessed by C-CBI. Data was retrieved from the Survey of Nurse Assistants’ Job Stress, Job Adjustment, and Health Outcomes in Long-Term Care Facilities in Taiwan.” Descriptive analysis, Pearson correlation, and linear regression were applied to explore the relationships between the variables and outcomes. RESULTS. Nursing assistants who were younger (β=-0.19, p<0.001), foreign (β=-0.12, p=0.05), working in a longer operation years of facility (β=-0.18, p<0.001), and working in a bigger facility (β=-0.22, p<0.001) and who had lower psychological demands (β=-0.35, p<0.001) and higher social support (β=-0.13, p=0.01) were less likely have job burnout. CONCLUSION. This study demonstrated job stressors influence nursing assistant burnout. Administrators of long-term care facilities should routinely monitor nursing assistants’ job stressors and burnout, help nursing assistants reduce stress, establish mutual trust, and provide stress-relief techniques if job stress and burnout are to be reduced in high-risk nursing assistants.