35卷2期

• Link 公衛今與昔 Public Health Now and Then
• 陳拱北與全民健保Dr. Kung-Pei Chen and Taiwan's National Health Insurance
• 江東亮、楊志良、張秀蓉
  Tung-Liang Chiang, Chih-Liang Yaung, Hsiu-Jung Chang
• 無
  none
• 無
  none
• 107 - 112
• 10.6288/TJPH201635105005

• Link 公衛論壇 Public Health Forum
• 台灣藥價差亂象－醫療經濟學角度之分析及建議Pharmaceutical market distortion under the current reimbursement policy-analyses and recommendations from a health economics’ perspective
• 施廷芳、李禮君、陳鴻儀
  Sophy Ting-Fang Shih, Li-Chun Lee, Hung-Yi Chen
• 無
  none
• 無
  none
• 113 - 115
• 10.6288/TJPH201635105017

• Link 綜論 Review Article
• 以系統性文獻回顧探討護理人員職場暴力現況與影響A systematic review of workplace violence in nursing10.6288/TJPH201635104045
• 明金蓮、 黃惠美、洪曉佩、曾麗華、黃美淑、江淑華、董貞吟
  Jin-Lain Ming, Hui-Mei Huang, Shiao-Pei Hong, Li-Hua Tseng, Mei-Shu Huang, Su-Hua Chiang, Chen-Yin Tung
• 系統性文獻回顧 ； 職場暴力 ； 護理職場 ； 護理人員
  systematic review ; workplace violence ; nursing workplace ; nurses
• 為瞭解醫療機構中護理人員遭受職場暴力狀況，本文運用系統性文獻回顧，經納入與排除條件篩選後，共納入37篇符合條件之文章。分別就不同調查期間、不同臨床環境與不同地區進行探討職場暴力盛行率，測量時間以一年時期的盛行率，肢體暴力有6-74%，非肢體暴力-言語暴力15.2-97.6%，職場霸凌11.6-47.6%，性騷擾33.7-34.9%；言語暴力高於肢體暴力，肢體暴力較常發生於急診室及精神科，言語暴力與職場霸凌則普遍發生於各單位，常見施暴者依序為病人、家屬或訪客、工作人員，高風險因素可歸類為工作人員、工作環境與攻擊者，不僅對受暴者身心有負面影響，亦對機構與護理專業造成危害，結論為遭受暴力後之因應方式分為生理、心理掌控及提升自尊等。高達40-80%人員遭受暴力後未向管理者通報，主要是認為不需協助、自己能處理及認為通報是無用的。因此建議提供顧客有表達意見的管道，建構良好護理工作環境，安全有效的硬體設備，支持職場暴力零容忍政策，建立暴力事件處理通報流程，並依不同職場暴力類型、不同臨床情境，發展創新與系統性職場暴力防治課程及因應策略，藉由有效的在職教育，提升護理人員對職場暴力預防與處置技能，以期有效改善工作環境，留任第一線臨床護理人員。
  This systematic review evaluated the workplace violence in nursing. Thirty-seven studies that met the criteria were included, reviewed, and analyzed. The one-year prevalence rates of physical violence ranged from 6-74%, sexual harassment 33.7-34.9%, while nurses experienced nonphysical violence including 15.2-97.6% verbal violence and 11.6-47.6% bullying behavior. Verbal violence is more common than physical violence. The later is commonly happened in emergency room or psychiatry department. The common perpetrators of these types of violence included the patients, family members, visitors and hospital staff. The nurses were apt to affected by these type of violence not only physically and psychologically, but also damaging the nursing profession and the whole hospital organization. Workplace violence victims have several coping strategies including physical mental support and promote self dignity. Unreported violence in nursing ranged from 40-80%, because violence victims believed that they were capable to handle and solve the problems or didn’t trust the reported system. Based on these, the most priority are provide the mutual feedback and communication pathway, and In conclusion, various types of violence in nursing are prevalent. It is a great challenge to establish a safe and healthy nursing environment with zero tolerance of workplace violence. Nursing education deserves courses and strategies to enable nursing professionals in handling the workplace violence with confidence and to promote their professional contribution.
• 116 - 135
• 10.6288/TJPH201635104045

• Link 原著 Original Article
• 剖析台灣民眾的就醫流動：利用引力模式評估就醫距離與醫療資源分布的影響Profiling the patient flow for seeking healthcare in Taiwan: using gravity modeling to investigate the influences of travel distance and healthcare resources
• 林民浩、郭年真、陳威全、溫在弘
  Min-Hau Lin, Raymmond N. Kuo, Benny Wei Chien Chin, Tzai-Hung Wen
• 引力模式 ； 病人流動 ； 就醫距離 ； 醫療資源 ； 空間分析
  gravity model ; patient flow ; travel distance ; health care resources ; spatial analysis
• 目標：過去研究多以行政區或醫療分區的每萬人口醫療資源數作為評估就醫可近性之指標，但分區的概念簡化民眾就醫行為，與實際醫療利用情形有所落差。本研究將以實際就醫流動資料呈現各地區醫療可近性，並量化就醫距離與醫療資源分布對就醫流動的影響。方法：使用全民健康保險研究資料庫2010年承保抽樣歸人檔，依居住地與就醫地建立流動起迄矩陣，計算各鄉鎮市區加權就醫距離作為可近性指標，再以引力模式估計影響就醫流動的因素。結果：加權就醫距離呈現出花東與雲嘉沿海等地有較差的醫療可近性。引力模式分析顯示，就醫距離與醫療資源皆顯著影響民眾的就醫流動。然而，相較於急診與住院的就醫流動，就醫距離對於門診就醫流動影響更為明顯。但於鄰近地區的就醫權衡，醫療機構一般病床數對於急診與住院就醫流動的吸引力，更重於就醫距離的效果。結論：加權就醫距離更能客觀反映一地醫療可近性。求近是台灣民眾就醫的首要考量，但民眾急診與住院就醫，則傾向在鄰近地區中尋求硬體規模最大者。
  Objectives: Township-level physician-to-population and hospital bed-to-population ratios have been widely adopted as indicators of health care accessibility in Taiwan; however, these indicators cannot reflect actual healthcare-seeking behaviors of patients. Therefore, the aims of this study were to establish actual patient flow for seeking healthcare and to clarify the factors influencing preferences for seeking healthcare. Methods: We estimated the residence and collected the locations of healthcare visits for each patient from the nationwide Longitudinal Health Insurance Database in 2010 to establish an origin-destination matrix. The patient flow among townships could be conducted from the matrix. The weighted travel distance of healthcare visits was developed as a new indicator of accessibility. The gravity model was then used to analyze the patient flow and assess the influences of travel distance and healthcare resources. Results: Using the weighted travel distance, eastern Taiwan and the coastal areas of Yunlin and Chiayi were identified as the poorest accessibility for healthcare resources. The results of the gravity model showed that travel distance was the major factor correlated negatively with the volumes of outpatient and inpatient/emergency flow; however, the abundance of healthcare resources positively influenced the preference of inpatient/emergency patients from their residence to neighborhood regions for seeking healthcare. Conclusions: The weighted travel distance could properly reflect the access of healthcare. We conclude that travel distance is the major factor which influences patient flow for seeking healthcare. In contrast, people tend to choose the healthcare facilities with abundant resources for seeking inpatient/emergency healthcare.
• 136 - 151
• 10.6288/TJPH201635104086

• Link 原著 Original Article
• 照護連續性對高齡者急診與非緊急急診之影響探討Effects of continuity of care on emergency department use and non-urgent emergency department visits among the elderly
• 梁亞文、陳文意、張曉鳳
  Yia-Wun Liang, Wen-Yi Chen, Hsiao-Feng Chang
• 照護連續性 ； 高齡者 ； 急診 ； 非緊急急診
  continuity of care ; elderly ; emergency ; non-urgent ED visits
• 目標：本研究旨在探討實施全民健康保險制度下的台灣高齡者照護連續性對高齡者急診及非緊急急診的影響。方法：本研究採用縱向研究法，利用2008-2009年門診就醫資料推算照護連續性指數(COCI)做為自變項，探討高齡者之照護連續性強度對急診利用及非緊急急診之影響。統計分析方法包含負二項式迴歸分析、邏輯斯迴歸分析，及趨勢卡方檢驗。結果：研究結果顯示，照護連續性顯著影響高齡者急診利用及非緊急急診的發生。在控制其他變項後，迴歸分析結果顯示，相較於低度照護連續性高齡者，中度照護連續性及高度照護連續性的高齡者，其急診利用次數的相對風險分別減少25.17%及35.60%；中度照護連續性及高度照護連續性高齡者，非緊急急診發生風險的勝算比分別為低度照護連續性高齡者的0.87(95% C.I.=0.76-0.99)及0.81(95% C.I.=0.69-0.94)。結論：本研究證實實施全民健康保險制度但缺乏轉診制度的台灣，高齡者照護連續性仍然顯著減少高齡者的急診利用，顯示提升高齡者照護連續性，對高齡者及健康照護系統都有助益。
  Objectives: This study determined whether or not continuity of care (COC) is associated with emergency department (ED) visits and non-urgent ED visits by the elderly in Taiwan’s universal health care system. Methods: This study used a longitudinal health insurance database compiled for 2010 from the National Health Insurance Research Database in Taiwan. COC was calculated using the continuity of care index (COCI), which reflects visit concentration with individual clinicians. Negative binominal regression and multivariate logistic regression were performed to determine the effects of COC on the ED and non-urgent ED visits in 2010, respectively. Results: This study showed that lower COC was associated with increased ED and non-urgent ED visits. After adjusting for age, gender, socio-economic status, region, physician visits, Charlson index, and physician density, patients in the medium and high COC groups had 25.17% and 35.60% less ED visits, respectively. In an additional analysis of non-urgent ED visits in which we compared patients in the medium and high COC groups with patients who were in the low COC group, we further showed that the probability of having non-urgent ED visits was reduced for elderly patients with medium and high COC (adjusted odds ratio = 0.87 and 95% CI=0.76-0.99; adjusted odds ratio =0.81 and 95% CI=0.69-0.94, respectively). Conclusions: This study showed that higher COC is associated with decreased ED and non-urgent ED visits by the elderly, even in a universal health care system that lacks a referral system. We conclude that improving the COC is beneficial for patients and the health care system.
• 152 - 163
• 10.6288/TJPH201635104084

• Link 原著 Original Article
• 台灣全民健康保險的減貧效果Effect of National Health Insurance on poverty reduction in Taiwan
• 江宛霖、江東亮
  Wan-Lin Chiang, Tung-Liang Chiang
• 台灣全民健保 ； 貧窮率 ； 醫療財務災難
  Taiwan's National Health Insurance ; poverty rate ; catastrophic health spending
• 目標：全民健保的首要目標就是保障公平就醫的權利，並減少因病而貧的狀況。本研究從收入面及支出面探討全民健保的減貧效果。方法：資料來源為1993年至2012年家庭收支調查。我們以全國家戶均等可支配所得中位數的一半作為貧窮線，計算家戶、兒童與老人貧窮率，並以醫療支出佔扣除生存所需的家戶消費的25%與40%為閾值，估計家戶、兒童與老人的醫療財務災難率。結果：1995年，台灣家戶貧窮率在加計社會移轉收支後由13.7%降為7.5%，減貧效果達6.2%，其中1.1%的效果為全民健保的貢獻，到了2012年全民健保貢獻則提高為3.7%。至於對兒童貧窮率及老人貧窮率，全民健保的貢獻在同一期間亦分別由0.5%及3.3%提高為2.0%及8.0%。其次，全民健保開辦後，家戶醫療財務災難率由1993年的7.0%驟降為1998年的1.3%，而兒童與老人的醫療財務災難率，亦分別由2.3%及17.7%降為0.3%及3.7%，但是進入21世紀以後，無論家戶、兒童或老人的醫療財務災難率，皆微幅上升。結論：自從開辦以來，全民健保對於減輕台灣的貧窮現象有重要的貢獻。
  Objectives: In 1995 National Health Insurance (NHI) was launched in Taiwan to provide equal access to adequate health care for all, and to protect people from financial hardship. This study determined the effect of the NHI on poverty reduction. Methods: Data were obtained from the Family Income and Expenditure Survey (1993-2012). First, we computed the household, child, and elderly poverty rates using 50% of the national median equivalized disposable income as the poverty line. We then defined catastrophic health spending as household health expenditures exceeding a specific fraction (25% or 40%) of household consumption remaining after subsistence needs. Results: In 1995 the household poverty rate fell from 13.7% to 7.5% after social taxes and transfers (a decrease of 6.2%), of which 1.1% was contributed by the NHI transfers. In 2012 the NHI contribution increased to 3.7%. With respect to child and elderly poverty rates, the NHI contribution also increased from 0.5% and 3.3% to 2.0% and 8.0%, respectively, during the same time period. After implementation of the NHI, catastrophic health spending fell from 7.0% in 1993 to 1.3% in 1998 for households, 2.3% to 0.3% for children, and 17.7% to 3.7% for the elderly. Since the beginning of the 21st century, all three types of catastrophic health spending have increased slightly. Conclusions: The NHI in Taiwan has made a significant contribution to poverty reduction since its implementation.
• 164 - 171
• 10.6288/TJPH201635104082

• Link 評論 Commentary
• 評論：台灣全民健康保險的減貧效果Commentary: effect of National Health Insurance on poverty reduction in Taiwan
• 文羽苹
  Yu-Ping Wen
• 無
  none
• 無
  none
• 172 - 172
• 10.6288/TJPH20163510408201

• Link 原著 Original Article
• 「思覺失調症醫療給付改善方案」成效探討－以台灣北部某精神專科醫院為例Effects of a pay-for-performance program for schizophrenia at a psychiatric hospital in northern Taiwan
• 陳俐樺、張鳳琴、簡以嘉、戴桂英
  Li-Hua Chen, Fong-Ching Chang, I-Chia Chien, Guey-Ing Day
• 思覺失調症 ； 論質計酬 ； 醫療利用
  schizophrenia ; pay-for-performance program ; medical utilization
• 目標：本研究旨在評價「思覺失調症醫療給付改善方案」介入，對有參與該方案的病人在精神科門診、住院、急診醫療利用之影響。方法：研究對象為台灣北部某精神專科醫院於2009-2013年參與「思覺失調症醫療給付改善方案」共2,236人，本研究資料來源為該醫院「醫療資訊系統」及「思覺失調症個案管理系統」等資料庫，以廣義估計方程式(Generalized Estimating Equation, GEE)方法進行成效檢定。結果：研究結果顯示：「思覺失調症醫療給付改善方案」介入顯著減少病人強制住院年發生次數、精神科急性病房年住院天數、精神科急性病房年住院次數、急診年使用次數；參與該方案的病人其門診年就診次數有稍下降然仍維持12次以上；然方案介入對病人六個月內精神科急性病房再住院次數無顯著影響。結論：建議持續推動「思覺失調症醫療給付改善方案」長期追蹤思覺失調症病人的疾病自我管理情形，以更積極提升精神醫療照護服務的品質，並預防病人疾病惡化。
  Objectives: This study determined the effects of a ”pay-for-performance program for schizophrenia” on patient utilization of the outpatient department, emergency services, and hospitalization. Methods: This study analyzed 2236 patients from a psychiatric hospital in northern Taiwan who joined the ”pay-for-performance program for schizophrenia.” Two datasets were analyzed, including the 2009-2013 Hospital Information System and the Schizophrenia Case Management System. The Generalized Estimating Equation (GEE) method was used to determine the effects of the program on patient utilization measures. Results: The ”pay-for-performance program for schizophrenia” significantly decreased the yearly incidence of compulsory hospitalization, days of acute psychiatric hospitalization, and annual emergency department visits. In addition, the number of psychiatric outpatient visits decreased significantly, but persisted at > 12 times per year. The number of 6-month readmissions to the acute psychiatric ward did not change significantly after implementing the program. Conclusions: Implementing the ”pay-forperformance program for schizophrenia” enhanced self-management of schizophrenic patients, improved medical care outcomes, and prevented worsening of diseases.
• 173 - 186
• 10.6288/TJPH201635104058

• Link 原著 Original Article
• 國人對雲端照護服務付費意願及其相關因素的探討：2013年全國調查分析Exploring factors related to willingness to pay for telehealth services: a 2013 national survey in Taiwan
• 梁娟娟、章美英
  Chuan-Chuan Liang, Mei-Ying Chang
• 照護服務 ； 系統品質 ； 服務品質 ； 資訊品質
  telehealth services ; system quality ; service quality ; information quality
• 目標：探討台灣民眾對雲端照護服務願意付費使用意願與醫療服務利用行為及雲端照護需求(系統品質、服務品質和資訊品質)之相關性。方法：研究對象為20歲以上台灣民眾，樣本分配以中選機率與人數等比例方式進行抽樣，並以電腦輔助電話訪問系統(Computer Assisted Telephone Interview, CATI)進行資料收集，進行性別、年齡、教育程度與居住地區等資料統計加權後，有效樣本數為1,132位。結果：(1)願意付費者使用雲端照護服務佔68.6%，以男性、20-29歲、無婚姻、獨居、高收入和整體健康較差者意願較高。(2)願意付費者認為雲端照護服務以系統品質最重要，其次為服務品質和資訊品質。(3)願意付費者對於雲端照護服務的預測因子為婚姻狀態、都市化程度、健康狀況、教育程度以及性別，預測率為69.4%。結論：本結果有助於了解使用雲端照護服務的人的經驗和開發能滿足人們的個性化需求的服務。首應考量資料分享，如電腦化健康紀錄的機密性和安全性，未來可研究探討。
  Objectives: The purpose of this research was to explore the correlations between the factors of Taiwanese residents’ willingness to pay for telehealth services and the factors of individuals’1 desired needs from the service (system quality, service quality, and information quality). Methods: The study sample comprised people older than 20 years who were selected using multistage cluster sampling of probability proportional to size sampling method. Surveys were conducted using computer-aided telephone interviews (Computer Assisted Telephone Interview, CATI), and valid responses were collected from a total of 1,132 people. Results: The study results are listed as follows: (a) Of the respondents, 68.6% were willing to pay for telehealth services; those who were male, aged between 20 and 29 years, were single, had a high income, and exhibited poor health demonstrated a relatively higher willingness. (b) System quality was the most crucial factor affecting Taiwanese residents’ willingness to pay for telehealth services, followed by service quality and information quality. (c) The predictors of willingness to pay for telehealth services were marital status, urbanization area, health condition, education level, and sex, which collectively yielded a predicted probability of 69.4%. Conclusions: These results can facilitate understanding the experiences of people using telehealth services and developing services that thoroughly satisfy people’s individual needs. Concerns regarding data sharing, such as the confidentiality and security of computerized health records, will be explored in future studies.
• 187 - 198
• 10.6288/TJPH201635104103

• Link 原著 Original Article
• 室外空氣污染物與學齡前兒童首發氣喘之關聯，2007-2011年Associations between outdoor air pollutants and first occurrence of asthma in pre-school children, 2007-2011
• 鍾慧穎、謝佳容、曾俊傑、 尹立銘
  Hui-Ying Chung, Hui-Ying Chung, Chun-Chieh Tseng, Lih-Ming Yiin
• 全民健康保險資料庫 ； 學齡前兒童 ； 氣喘 ； 空氣汙染物 ； 交通排放
  National Health Insurance Research Database ; pre-school children ; asthma ; air pollutants ; transportation exhaust
• 目標：探討近年來全台灣0-6歲兒童氣喘首次發生率，與環境監測之空氣汙染物濃度的關係。方法：以2007年至2011年之全民健康保險研究資料庫承保抽樣歸人檔40萬人，依據國際疾病分類法第九版(ICD-9-CM)診斷碼，篩選6歲以下第一次發生氣喘兒童為研究對象，並使用2006與2011年間環保署空氣品質監測數據PM_(10)、SO_2、NO_X(NO及NO_2)、CO及O_3之濃度資料，探究首發氣喘與空氣汙染物之相關性。結果：室外的空氣汙染物，以NO_X、CO與學齡前兒童首次氣喘發病關聯性顯著(ρ ＞ 0.5)，而此二者間相關性也高於其他任何汙染物種間的關聯(ρ = 0.892)，推測NO_X及CO主要來源為交通排放。進一步以多變項邏輯斯迴歸分析，發現暴露於高濃度NOX(＞ 16.3 ppb)導致氣喘發病比低濃度具有顯著勝算比，最高可達1.45(95%CI: 1.13- 1.86)。由於並無直接證據顯示NO_X、CO可導致氣喘，因此推論台灣學齡前兒童的氣喘首發率，應與交通排放汙染物相關。結論：交通排放汙染物與氣喘發病相關性高，建議民眾做好個人防護，並時時注意空氣品質資訊，以降低學齡前兒童氣喘發病機率。
  Objectives: This study determined the incidence of the first occurrence of asthma in preschool children (0-6 years of age), and the relationship to air pollutants. Methods: Based on the Longitudinal National Health Insurance database (n=400,000) from 2007-2011, the study identified eligible asthmatic subjects in accordance with the 9th version of the international disease classification (ICD-9-CM), and used the local environmental monitoring data of PM_(10), SO_2, NO_X (NO and NO_2), CO, and O_3 from the Environmental Protection Administration (EPA) to relate to the first incidence of asthma for the eligible subjects. Results: Among the air pollutants, NO_X and CO were significantly associated with asthma, and were highly correlated with each other (? = 0.892), suggesting that NO_X and CO should have originated from transportation exhaust. The multivariate logistic regression analyses showed that high levels of NOX (> 16.3 ppb) resulted in significant odds ratios of asthma exacerbation up to 1.45 (95% CI: 1.13 - 1.86) compared to low levels. Because of no direct evidence showing that NOX or CO triggered asthma, it is suggested that the first incidence of asthma in pre-school children is related to transportation exhaust. Conclusions: Asthma in pre-school children was significantly associated with transportation exhaust. It is recommended that attention should be paid to personal care and information of air quality to lower the chances of asthma occurrence for preschool children.
• 199 - 208
• 10.6288/TJPH201635104038

• Link 實務 Public Health Practice
• 台灣「婦幼健康主題式資料庫」之建構與未來應用Set-up and future applications of the Taiwan Maternal and Child Health Database (TMCHD)
• 李中一、陳麗華、邱孟君、梁富文、呂宗學
  Chung-Yi Li, Lea-Hua Chen, Meng-Jiun Chiou, Fu-Wen Liang, Tsung-Hsueh Lu
• 婦幼衛生 ； 健康資料庫 ； 出生登記 ； 死因統計 ； 生命歷程
  maternal and child health ; health database ; birth registry ; death registry ; life course
• 目標：介紹「婦幼健康主題式資料庫」之建置過程與內容。方法：本研究利用我國出生登記、出生通報、死因統計(以上之資料年代為2004-2011)、與全民健康保險(簡稱健保)(1998-2011)等行政資料，完成親代與子代之健保資料連結；並將新生兒未投保前使用父親或母親身分證字號申報健保給付之紀錄補正回新生兒之就醫紀錄中。結果：本研究建置之「婦幼健康主題式資料庫」，由4種次檔案組成，包括(1)2004-2011年台灣所有活產新生兒之健保門診與住院紀錄及其出生登記的資料，如出生體重、妊娠周數、單/多胞胎等，且每一筆新生兒的健保就醫紀錄中新增該名新生兒父親與母親的身分證字號(加密)；(2)出生通報中每名母親吸菸、飲酒，及新生兒Apgar score；(3)所有新生兒父親與母親的健保門診與住院紀錄(含醫令資料)(1998-2011)；(4)新生兒與父母親之死因統計資料(2004-2011)。結論：「婦幼健康主題式資料庫」可以作為生命歷程研究之材料，也可提供較佳的訊息作為實證衛生政策的參考依據。
  Objectives: To introduce the process of setting up the Taiwan Maternal and Child Health Database (TMCHD) and the content of the TMCHD. Methods: This project utilized the Taiwan Birth Registration Database (BRD), Birth Certificate Application (BCA), National Register of Death (NRD [2004-2011 for the BRD, BCA, and NRD]), and the National Health Insurance Research Database (NHIRD [1998-2011]) to inter-link the medical claims of offspring and their parents. An effort was also made to restore the medical claims of a parent to his/her newborn; medical claims are usually issued to one of the parent within 60 days after birth. Results: The TMCHD includes the information derived from inpatient/outpatient claims of all infants born between 2004 and 2011, in addition to a number of variables recorded in the BRD, including birthweight, gestational age, and single/multiple births. The variables in the BCA, such as maternal smoking, alcohol consumption, and newborn Apgar scores, were identified in a separate file. The medical claims of parents retrieved from the NHIRD, including medical orders (1998-2011) and the information pertaining to mortality retrieved from the NRD (2001-2011), were arranged in separate files. Conclusions: The TMCHD can be used as material for life course studies and may provide better information for evidence-based health policy.
• 209 - 220
• 10.6288/TJPH201635104053

• Link 研究紀要 Research Brief
• 兒童死亡率：台灣與經濟合作暨發展組織國家比較Child mortality: Taiwan compared with OECD countries
• 梁富文、 黃雅莉、吳美環、呂鴻基、江東亮、呂宗學
  Fu-Wen Liang, Ya-Li Huang, Mei-Hwan Wu, Hung-Chi Lue, Tung-liang Chiang, Tsung-Hsueh Lu
• 嬰兒死亡率 ； 兒童死亡率 ； 國際比較
  infant mortality ; child mortality ; international comparison
• 目標：瞭解台灣兒童死亡率與經濟合作暨發展組織(Organization for Economic Cooperation and Development, OECD)國家的差異。方法：OECD國家的出生數、人口數與死亡數資料來自世界衛生組織的死因統計資訊系統，台灣資料來自衛生福利部統計處網站。首先計算嬰兒(＜1歲)，新生兒(＜28天)，後新生兒(28-364天)，1-4歲，5-9歲及10-14歲的死亡率。除了排序各國兒童死亡率外，我們以台灣為基準計算死亡率比與95%信賴區間。最後再比較台灣與日本的死因組成，以及台灣1996-2013年兒童死亡率趨勢。結果：合計2011-2013年資料，台灣嬰兒，新生兒，後新生兒每千活產死亡數分別為3.9、2.4及1.5，1-4歲，5-9歲，10-14歲每十萬人口死亡數分別為25.8、13.2與14.4，與33個OECD國家比較，排名分別為第23，16，24，27，30與27名。經統計檢定死亡率顯著小於台灣的國家數目分別為19，12，21，24，26與22。在嬰兒死亡個案中，源於周產期的病況比例台灣(46%)較日本(25%)高，先天性疾病比例台灣(22%)較日本(36%)低。在1-4歲死亡個案中，台灣(28%)外因比例高於日本(16%)。關於台灣兒童死亡率趨勢，1996-2013年嬰兒死亡率的年下降率(3.4%)最小，1-4歲死亡率下降率(5.5%)最大。結論：台灣兒童死亡率相較於大多數OECD國家及文化背景相似的日本與韓國都表現不佳，尤其五歲以上兒童的死亡率排名相對較差。1-4歲死亡率雖然下降趨勢顯著，但是外因比例還是高於日本，應該是未來的優先族群。
  Objectives: To determine the child mortality rate differences between Taiwan and Organization for Economic Cooperation and Development (OECD) countries. Methods: Data on live births, population, and mortality in OECD countries were derived from the mortality database of the World Health Organization. Taiwan data were derived from the Ministry of Health and Welfare. We first calculated the child mortality rate by age, i.e., infant (<1 year), neonate (<28 days), post-neonate (28-364 days), 1-4 years, 5-9 years, and 10-14 years. We then computed the mortality rate ratio and 95% confidence intervals using the mortality rate of Taiwan as the reference. We also compared the pattern of cause of death between Taiwan and Japan and examined the trends in child mortality of Taiwan between 1996 and 2013. Results: Between 2011 and 2013, the infant, neonate, and post-neonate mortality rate in Taiwan was 3.9, 2.4, and 1.5 deaths per 1,000 live births, respectively. The mortality rates in the 1-4 year, 5-9 year, and 10-14 year age groups in Taiwan were 25.8, 13.2, and 14.4 deaths per 100,000 population, respectively. Compared with 33 OECD countries, the mortality rate in Taiwan ranked 23rd, 16th, 24th, 27th, 30th, and 27th for each age group, respectively. The number of countries with mortality rates significantly lower than Taiwan was 19, 12, 21, 24, 26, and 22 for each age group, respectively. For infant deaths, the proportion attributed to perinatal causes in Taiwan (46%) was greater than Japan (25%); however, the proportion of infant deaths due to congenital causes in Taiwan (22%) was less than Japan (36%). For the 1-4 year age group, the proportion of deaths resulting from external causes in Taiwan (28%) was greater than Japan (16%). The decline in the mortality rate from 1996 through 2013 was greatest amongst children 1-4 years of age (annual reduction of 5.5%) and was less significant for infants (annual reduction of 3.4%). Conclusions: The child mortality rate in Taiwan, especially children ? 5 years of age, was higher than most OECD countries, including Japan and Korea, which have a similar cultural background to Taiwan. Despite a drastic reduction in the mortality rate for children 1-4 years of age in Taiwan, the proportion of deaths attributed to external causes was greater than the corresponding children in Japan, which is a prevention priority.
• 221 - 233
• 10.6288/TJPH201635104087