生命末期老人用藥之探討

Abstract

研究背景 過去生命末期照護以癌症族群為主，近年隨著人口老化及非癌症死因的比例增加，顯示生命末期非癌症老年族群的照護日漸受到重視。然而，關於非癌症老年族群於生命末期的用藥適當性文獻探討十分有限。 研究目的 探討台灣地區因癌症或非癌症過世的老人，於生命末期的症狀緩解與共病症藥品的使用與影響因素 研究方法 本研究以台灣健保資料庫為資料來源將2008至2012年間在醫院過世的老年族群，依死因分成癌症、器官衰竭、衰弱、突然死亡與其他，共五組。藥品使用分為症狀緩解與共病症藥品的使用者比例與用藥數目，以及症狀緩解用藥細項數目百分比分布，門診用藥採追蹤死前12個月內用藥紀錄，住院用藥採追蹤死前3次住院用藥紀錄。每位研究對象於生命末期之用藥變化，分為死前1個月與死前第12個月(基準點)門診之用藥差異，及死前1次與死前第3次(基準點)住院之用藥差異。另外，生命末期最後一次住院用藥的影響因素使用multivariate logistic regression分析。 研究結果 本研究共納入70,396位在醫院過世的老年族群，其中癌症組占31.26%、器官衰竭組21.53%、衰弱組29.52%、突然死亡組2.08%與其他組占15.61%。症狀緩解藥品，門診用藥者比例，除了突然死亡組不變外，癌症組下降最少，住院用藥者比例、門診與住院用藥數目變化以癌症組增加最多，[死前1個月與死前第12個月用藥數目差mean (SD)：癌症組1.60 (3.36)，器官衰竭組0.86 (2.88)，衰弱組0.76 (2.95)，突然死亡組0.78 (3.49)，其他組0.84 (2.97)；死前1次住院與死前第3次住院用藥數目差mean (SD)：癌症組0.47 (1.89)，器官衰竭組0.04 (1.51)，衰弱組0.00 (1.46)，突然死亡組0.15 (1.56)，其他組0.08 (1.58)]；共病症藥品，門診用藥者比例與用藥數目變化，以癌症組減少最多，住院用藥者比例，五組皆為增加，住院用藥數目變化以癌症組減少最多，[死前1個月與死前第12個月用藥數目差mean (SD)：癌症組-0.40 (1.30)，器官衰竭組-0.10 (1.27)，衰弱組-0.11 (1.20)，突然死亡組0.05 (0.82)，其他組-0.08 (1.36)；死前1次住院與死前第3次住院用藥數目差mean (SD)：癌症組-0.04 (2.07)，器官衰竭組0.17 (2.28)，衰弱組-0.02 (2.28)，突然死亡組0.15 (1.21)，其他組-0.01 (2.63)]。在癌症、器官衰竭與衰弱三組的症狀緩解藥品分布顯示opioids類止痛藥使用，在死前第12個月與死前1個月門診紀錄：癌症(2%, 15%)、器官衰竭(0%, 0%)、衰弱(6-7 %)；死前第3次到死前1次住院紀錄：癌症(15%, 23%)、器官衰竭(6%, 7%)、衰弱(6%, 6%)。Multivariate logistic regression分析結果為女性、75歲以上、死因為器官衰竭或衰弱或突然死亡或其他疾病、死前住院機構為區域醫院或地區醫院或基層院所之老年人，較不易接收症狀緩解藥品治療；女性、因器官衰竭或衰弱或其他疾病死亡、共病症程度高者之老人，較容易使用共病症藥品治療。 研究結論 本研究顯示台灣地區老年非癌症族群於生命末期，器官衰竭組與衰弱組在症狀緩解用藥如opioids類止痛藥，較癌症組使用偏低；住院共病症用藥者比例在生命末期仍有增加情形。有關前述藥品於生命末期使用的適當性，有待未來更多研究探討。

Background Despite the importance of end of life care in non-cancer group, studies investigating the optimization of drug use in non-cancer aging population at the end of life are very limited. Objectives To investigate the patterns and determinants of drug utilization for symptom-relief and comorbid illnesses in older adults dying with cancer or non-cancer at the end of life in Taiwan. Methods Using Taiwan’s National Health Insurance Research Database (NHIRD), inpatients who aged 65 years old and older died between 2008-2012 were identified and divided into five groups (cancer, organ failure, frailty sudden death and others). Drug utilization includes the proportion of drug user and number of drug use for symptom-relief and comorbid illnesses and distribution of symptom-relief drug use. Drug utilization in outpatient care measured monthly, as for inpatient care, it was measured for 3 hospitalizations prior to death. The incremental change of drug utilization among five groups of elderly decedents were from last 12th month to last month in outpatient care and from last 3rd hospitalization to last hospitalization in inpatient care. Multivariate logistic regression analysis was used to identify the possible determinants of drug utilization for the last hospitalization before death. Results A total of 70,396 elderly decedents (cancer 31.26%, organ failure 21.53%, frailty 29.52%, sudden death 2.08%, others 15.61%) were included in the study. For drug use of symptom-relief, except for sudden death group, the mean changes in proportion of outpatient drug user are reduced most in cancer group, and the mean changed proportion of outpatient drug user and number of outpatient and inpatient drug use are increased most cancer group (Incremental change from last 12th to last month, mean (SD): cancer 1.60 (3.36), organ failure 0.86 (2.88), frailty 0.76 (2.95), sudden death 0.78 (3.49), others 0.84 (2.97); Incremental change from last 3rd to last hospitalization, mean (SD): cancer 0.47 (1.89), organ failure 0.04 (1.51), frailty 0.00 (1.46), sudden death 0.15 (1.56), others 0.08 (1.58)). As for drug use of comorbid illnesses, the mean changes in proportion of outpatient drug user and number of outpatient and inpatient drug use are increased most in cancer group, the mean changes in proportion of inpatient drug user are increased among 5 groups, (Incremental change from last 12th to last month, mean (SD): cancer -0.40 (1.30), organ failure -0.10 (1.27), frailty -0.11 (1.20), sudden death 0.05 (0.82), others -0.08 (1.36); Incremental change from last 3rd to last hospitalization, mean (SD): cancer -0.04 (2.07), organ failure 0.17 (2.28), frailty -0.02 (2.28), sudden death 0.15 (1.21), others -0.01 (2.63)). Symptom-relief drug distribution among 3 groups (cancer, organ failure and frailty) showed opioid use; outpatient (-12 and -1 month: cancer (2%, 15%), organ failure (0%, 0%), frailty (6-7 %)); inpatient (-3 and -1 hospitalization: cancer (15%, 23%), organ failure (6%, 7%), frailty (6%, 6%)). Multivariate logistic regression analysis showed patients were female, older age, dying form organ failure or frailty or sudden death or others, died in local hospital or physician clinics were related to less symptom-relief drug use; patients with female, death from organ failure or frailty or others and higher level of comorbid illnesses were related to more comorbid drug use. Conclusions Our study showed that among non-cancer group, less opioid use in organ failure and frailty group than cancer group at the end of life; the proportion of comorbid inpatient drug user are increased at the end of life. More detailed studies are needed to explore the appropriateness of opioid use in organ failure and frailty group and the optimized timing of withdrawing comorbid drug at the end of life.