失能者居家護理使用之研究

Abstract

居家護理提供民眾在家能持續使用醫療照護服務，實現在地老化的老年生活。台灣自1995年全民健康保險提供居家護理付費之來源，因此使得居家護理服務市場成長快速。居家護理使用者常伴隨高醫療耗用之風險，因此分析使用者現況及問題已成為重要課題。本研究期望能了解全國居家護理使用現況，其次探討具技術性護理需要之失能者居家護理使用之影響因素，並且了解其醫療使用情形，以及居家護理使用對醫療服務使用之影響，期能提供政策規劃實證參考。 本研究之樣本為具有任一項ADL或IADL之失能者，樣本主要來源為「全國長期照護需要評估」失能者機率性樣本。依據研究目的共選取13,979名50歲以上具任一項ADL或IADL之失能者進入研究中，並串聯健保「門診處方及治療明細檔」、「住院醫療費用清單明細檔」及「醫令檔」，以了解失能者之在受訪後一年內之居家護理及醫療服務使用現況。本研究利用「全國長期照護需要評估」取得個人層級之變項，利用「2002年台閩地區長期照護資源名冊」，以及行政院衛生署2002年之衛生統計資料以取得區位層級變項資料。利用修改後之安德遜行為模式(Andersen Behavioral Model)為研究架構，使用階層線性模式(Hierarchical Linear Model)進行多變項分析，以了解個人與區位層級因素對居家與機構樣本使用居家護理服務之影響因素。為了解居家護理使用對醫療使用之影響，利用傾向分數(Propensity Score)進行居家護理使用配對，並以pair t 檢定、卡方檢定及複迴歸進行分析，以了解有無使用居家護理對醫療服務使用之影響。 研究顯示具任一項ADL或IADL之失能者共13,979人，其中有14.7%有一項以上技術性護理需要；其需要項目之前三位為自我照顧指導、留置鼻胃管護理以及留置導尿管護理，比率分別為10.0%、5.9%與4.5%。在受訪後一年內曾使用過居家護理者共835人，佔全部樣本之6.0%；使用者平均使用7.8次；使用者平均使用6.7個月，居家護理申報費用平均為14,155元。 在控制其他變項後，發現個人層級的因素之中，居家樣本之巴氏量表分數越高，則居家護理使用率及使用時間較少；較之資源耗用群組第一類者，資源耗用群組分類越高，則越傾向使用居家護理，使用時間及費用也較高；教育程度越高，則居家護理使用率、使用時間與費用也增加；年齡越大與未婚者則居家護理使用費用較少；區域層級中變項中發現當居住區域之長期照護機構越多，則居家護理使用率及使用時間越多。在投入所有個人及區位層次變項後，仍無法完全解釋居住區域間之居家護理使用率、使用時間與費用之差異。 在機構樣本部份，巴氏量表分數越高，則居家護理使用率及使用時間較少；較之資源耗用群組第一類者，資源耗用群組分類第二類，其居家護理使用率及使用時間較高；教育程度越高，則居家護理使用率、使用時間與費用皆增加；男性之居家護理使用時間與費用較女性少；區域層級變項中發現當居住區域之長期照護機構越多，則居家護理使用率及使用時間越高。在投入所有個人及區位層次變項後，仍無法完全解釋機構樣本不同居住地區間之居家護理使用率、使用時間與費用之差異。 經居家護理使用傾向分數配對後，1,160名配對樣本之人口特性及需要因素分佈皆無差異，居家護理使用機率範圍為0.010至0.697之間。相依樣本平均值檢定後發現使用居家護理樣本在急診、門診的使用率、使用次數及費用較未使用居家護理者多，但有使用居家護理者較未使用居家護理者平均住院天數少37.3天，全年住院費用與整體醫療費用相差高達142,596元與120,065元，達顯著統計差異。因此可知，使用居家護理可以藉由減少住院天數與住院費用，降低整體急性醫療之支出。 本研究之樣本來自「全國長期照護需要評估」失能者機率樣本，並結合健保資料以了解失能者之居家護理使用，因此資料具全國代表性，提供具有技術性護理需要者之居家護理使用資料，及影響使用的原因，並提供居家護理對醫療資源使用之實證資訊。使用居家護理確有節省醫療費用之功效，實為值得發展之照護模式，目前我國健保制度已導入DRG給付制度，因此建議修正居家護理之收案與給付標準，並增加服務次數及內容，以切合民眾日益增加之亞急性及長期照護需要。未來在老年及失能人口照護需要繼續成長的趨勢下，本研究結果可以提供規劃亞急性與長期照護服務模式規劃之實證參考，有助於提供未來政策制定之依據。

Purpose: Home health care services have grown rapidly since the launch of the National Health Insurance in 1995. The determinants of health and social service needs of the elderly have been widely studied. However, little is known about the home health care needs and utilization in Taiwan. Research has demonstrated that home health care users usually were associated with high medical costs. More attention has been paid to the relationship between home health care and health care. The purposes of this study were to understand the needs and utilization of home health care in Taiwan, and to identify the factors which determine the use of home health care services, so as to probe into the relations between home health care and health care service utilization. Method: The design of this study is conducted on the database from “National Long-term Care Evaluation” program of the Department of Health, Executive Yuan. Proportional to Size (PPS) sampling was used to recruit disabled adults who were over 50 years old from the national population in each county/city in Taiwan. During the second stage of evaluation, 14,094 disabled adults were interviewed. A total of 13,979 adults who required assistance in at least one item of ADL (Activities of Daily Living) or IADL (Instrumental Activities of Daily Living) were included in the study. The study also incorporated database from the 2002~2003 National Health Insurance claims for utilization and costs of home health care and health care services (for individual level data), and the 2002 health statistics of Department of Health and 2002 list of Long-term Care resources of Taiwan area (for area level data). The modified “Andersen Behavioral Model” was used as a research framework. Hierarchical linear model was used to determine the influence of the use of home health care in area and individual level for those who were taken care of in homes or institutions. Propensity Score was used for home health care user and non-user matching, in order to compare the health care service utilization with costs. Outcome: Results showed that 14.7% of the 13,979 disabled adults required at least one item of skill nursing. The most needed services on skill nursing were self-care management (10.0%), insertion of nasogastric tube (5.9%), and urinal indwelling catheterlization (4.5%). A total of 835 (6%) disabled adults have used home health care within one year after survey. The average use of home health care were 7.8 times and 6.7 months. The expenditure of home health care averaged $14,155 NTD per user. Using hierarchical linear analyses, the significant predictors of home health care use were scores of Barthel index, classification of resource utilization groups (RUGs), age, marital status, education level, and number of long-term care facility in living area after controlling for other variables. For disabled adults who were taken care in their homes, those with higher Barthel index scores had less use of home health care and required less time and home health care costs. Participants who were classified as RUG level 2, 3, or 4 had higher odds ratios for the use of home health care (OR=2.9, 3.5, 3.7, respectively) and tended to require more time and costs, compare to those who were classified as RUG level 1. Participants with junior high school education or above had higher probability to use home health care (OR=1.6, 95% CI=1.1~2.4) and required more time (OR=1.5) and costs (t=2.7) for home health care than illiterates. Older and unmarried disabled adults spent less home health care costs than their counterparts. The area level resources showed that the number of long-tem care facility was the most significant predictor for the use of home health care. The more long-tem care facility in living area, the more probability to use home health care (OR=1.003) and use longer (OR=1.003). The total differences between areas for home health care use cannot be explained by all the covariates. For disabled adults who were taken care in institutions, those with higher Barthel index scores had less use of home health care (OR=0.97) and required less time (OR=0.98). Participants who were classified as RUG level 2 and 3 had higher odds ratios for the use of home health care (OR=3.0 and 2.2, respectively) and tended to require more time (OR=3.0 and 1.9, respectively) compare to those who were classified as RUG level 1. More educated disabled adults were more likely to use home health care, but required more time and costs for home health care than illiterates. Male participants required less time and costs for home health care than female participants. At the area level resources, the most significant predictor for using home health care was the number of long-tem care facility. The more long-tem care facility in living area, the more probability to use home health care (OR=1.003) and use longer (OR=1.003). The total differences between areas for home health care use cannot be explained by all the covariates. A total of 1,160 matched samples were matched by home health care use and the probability score ranging from 0.010 to 0.697 (home health care user = experimental groups; non-home health care user = reference groups). The health care utilization difference between home health care user and non-user was great. Home health care users had high usage rates, frequency and costs for outpatient and emergency services. Non-users spent more hospital days (approximately 37.3 days) and extra costs ($142,596 NTD) for hospitalization than home health care users. Home health care user saved approximately $120,065 NTD in health care costs than non-users. Conclusion: Home health care was proved to have a positive effect on reduced health care costs because of decreased hospital days and costs in this study. However, research showed that there was a gap between home health care need and utilization. Less use of home health care can be contributed to the limitations of NHI payment system. Because of rising aging population and the new payment system in hospitalization-Taiwan version of Diagnosis Related Groups (Tw-DRGs) has been entered into the NHI, long-term care and post acute care needs are expected to increase rapidly in the future. In response, enrollment criteria, service intensity, contents, and flexibility of home health care services should be modified to reduce the unmet needs of care. The study was based on national proportional samples for disabled adults. The study results may provide statistical data for utilization profile and determinants of home health care in Taiwan. Promoting of home health care services is needed.