背景 : 主要照顧者的照顧負荷過大不僅危害照顧者的身心健康，更波及病患的照護品質，瞭解消化道癌末病患主要照顧者之照顧負荷與其相關性因素有助於提高末期醫療照護品質。 目的 : 本研究目的為探討消化道癌末病患特性及主要照顧者的基本屬性、社會支持、照顧自我效能、死亡恐懼、自覺病患症狀困擾與照顧負荷的關係。 方法 : 本研究為橫斷式相關性研究，以方便取樣結構式問卷資料收集嘉南地區之消化道末期癌症病患之居家主要照顧者共 178位。 結果 : 安寧居家消化道末期癌症病患之照顧負荷顯著高於門診收案者；肝癌胰臟癌之照顧負荷顯著高於結腸直腸癌者；配偶照顧者之照顧負荷顯著高於直系血親；自覺健康狀況變得較差者在照顧負荷顯著高於自覺沒有變差之照顧者。消化道末期癌症病患主要照顧者之照顧負荷的重要解釋因子依序為社會支持、自覺病患症狀困擾、自覺健康狀況、收案場所、死亡恐懼、照顧關係共可解釋 63.8%之變異數。社會支持為最重要的解釋因子，可單獨解釋 37.2%變異數。 結論 : 建議末期醫療團隊成員能加強對消化道末期癌症病患主要照顧者之社會支持的評估，對社會支持程度較低之照顧者儘早提供協助；加強照顧者的生死教育、協助主要照顧者正向面對死亡恐懼。注意消化道末期癌症病患症狀控制，以減輕照顧者的自覺病患症狀困擾。而對於安寧居家、配偶及自覺健康狀況較差之主要照顧者應加強其照顧負荷之評估。

Background: Heavy caregiving burdens can harm the physical and mental health of primary caregivers and reducepatient care quality. Understanding caregiving burden and its associated factors among primary caregivers of terminally ill patients with gastrointestinal cancer can help improve holistic terminal healthcare quality. Purpose: The authors explore in this paper the relationship between caregiving burden and terminally ill gastrointestinal cancer patient disease characteristics, demographic backgrounds, level of social support, self-care efficacy, fear of death and self-perceived symptom distress in both patients and primary caregivers. Methods: This was a cross-sectional, descriptive, and correlational study that used convenience sampling and structured questionnaires. Data were collected from 178 family caregivers of terminally ill patients with gastrointestinalcancer in the Tainan and Chiayi areas of Southern Taiwan. Results: The caregiving burden of caregivers of terminally ill gastrointestinal cancer patients in hospice homecarewas significantly higher than that of those recruited from outpatient departments. Caregiving burden for liver and pancreatic cancer patients was significantly higher than for colorectal cancer patients. The caregiving burden of spousalcaregivers was significantly higher than that of lineal blood relatives. The caregiving burden of caregivers with worseself-perceived health status was significantly higher than that of those with better self-perceived health status. The most important explanatory factors of caregiving burden among primary caregivers terminally ill gastrointestinal cancer patients were (in descending order) social support, self-perceived symptom distress in patient, self-perceived healthstatus, location of study subject recruitment, fear of death, and relationship with patient; these factors explained 63.8%of the total variation. Social support was the most important explanatory factor, explaining 37.2% of total variance. Conclusions: We recommend that terminal health care teams better assess the social support given primary caregivers of terminally ill gastrointestinal cancer patients, that assistance be provided to caregivers with less social support,that caregiver life-and-death education be improved, and that primary caregivers be taught how to accept and positively handle the death of the loved one in their care. More attention should be paid to controlling symptoms of terminalstage cancer patients in order to reduce caregiver self-perceived symptom distress. Evaluation of caregiving burdenis especially important for those primary caregivers who are hospice homecare workers, spouses, and of lower self-perceived health status.