38卷1期

• Link 政策論壇 Policy Forum
• 從永續發展目標看藥物濫用防治：善用全國性流行病學調查的發現The prevention and treatment of drug abuse in sustainable development goals: making good use of the findings of national surveys
• 陳為堅
  Wei J. Chen
• 無
  none
• 無
  none
• 1-4
• 10.6288/TJPH.201902_38(1).PF01

• Link 公衛論壇 Public Health Forum
• 尊嚴善終：我國病人自主權利法施行之整備End of life in dignified way: preparedness for implementation of Patient Autonomy Right Act
• 陳再晉、林廷育、陳冠文、鐘珮純
  Tzay-Jinn Chen, Ting-Yu Lin, Kuan-Wen Chen, Pei-Chun Chung
• 無
  none
• 無
  none
• 5-8
• 10.6288/TJPH.201902_38(1).107103

• Link 綜論 Review Article
• 台灣青少年選手專項化問題與肌力訓練之角色Specialization problems of youth athletes in Taiwan and the role of strength training
• 林政東
  JENQ-DONG LIN
• 過早專項化訓練；運動傷害；肌肉骨骼系統失衡；結締組織
  early specialization training?;?sports injuries?;?musculoskeletal imbalance?;?connective tissue
• 台灣升學制度及一些措施鼓勵青少年運動競賽，使青少年過早投入專項化訓練與比賽，此一作法較不符合運動訓練原則。本文以相關關鍵字搜尋資料庫並進行文獻探討，發現過度重視青少年運動成績，將促使過早進行專項化訓練，雖在國際分齡運動賽事上屢創佳績，但選手卻易運動傷害，錯失強化骨骼與肌腱的最佳時期，甚至可能影響20歲後運動表現。建議考量青少年從事多樣化（diversified sports）運動，並以肌力訓練為其發展平衡的肌肉骨骼解剖結構，建立強?的骨骼、肌腱…等等結締組織，減少青少年運動傷害，亦為20－30歲高強度訓練及巔峰表現作準備。現行強調運動成績的趨勢，除了不利於運動表現的發展，亦容易導致青少年遭受運動傷害，建議政策上應編訂一套有利於青少年運動生涯發展的措施。
  Taiwan's further education system and several measures that encourage youth sports and competitions have caused teenagers to enter specialized training programs and competitions prematurely; this practice does not seem to conform to the principles of sports training. This article searched databases using relevant keywords to conduct a literature review, discovering that Taiwan places too much emphasis on teenagers' athletic performance, resulting in early specialization training. Despite achieving positive results at international sports events of various age levels, the athletes experience sports injuries or miss the most suitable period for strengthening their bones and tendons even if they are not injured. In addition, their athletic performance after the age of 20 is affected. These are all caused by measures that encourage early specialization training. Such practices are not merely detrimental to the development of the sports in Taiwan, but also cause the youth athletes to experience sports injuries throughout their lifetime. The most satisfactory practice is for the youth athletes to engage in diversified sports as well as use strength training to develop a balanced musculoskeletal anatomy, and build strong connective tissue such as bones and tendons to reduce their sports injuries. Such practices prepare them for high-intensity training and peak performance aged 20-30 years, thereby achieving the greatest well-being for these youth athletes.
• 9-18
• 10.6288/TJPH.201902_38(1).107036

• Link 原著 Original Article
• 雲嘉南地區大氣細懸浮微粒濃度與交通污染源相關性分析Correlation of atmospheric fine particulate matter with traffic sources in Yun-Chia-Nan area in Taiwan
• 黃淑倫、林裕清、蕭光佑、林玠模、紀妙青、黃友珊
  SU-LUN HWANG, YU-CHING LIN, KUANG-YU HSIAO, CHIEH-MO LIN, MIAO-CHING CHI, YU-SHAN HUANG
• 雲嘉南；細懸浮微粒；機動車密度
  Yun-Chia-Nan?;?fine particulate matter?;?motor vehicles density
• 目標：大氣細懸浮微粒（fine particulate matter, PM_(2.5)）濃度可能受交通污染源的影響，本研究以台灣雲嘉南地區為研究地區，探討大氣PM_(2.5)與交通污染源的關係。方法：分析行政院環境保護署空氣品質監測站之PM_(2.5)數據及機動車密度登記資料。研究區域為雲林縣、嘉義縣市與台南市。以敘述性統計（平均值、中位數、最小值和最大值）分析PM_(2.5)濃度、氣象因子與機動車密度；以斯皮爾曼等級相關（Spearman's Rank Correlation Coefficient）探討PM_(2.5)濃度、氣象因子與機動車密度的相關性；以複迴歸分析機動車密度對PM_(2.5)濃度的影響。資料分析期間為2015年1至12月。結果：2015年雲嘉南地區PM_(2.5)平均濃度為26.4μg/m^3，冷季（11～4月，35.0μg/m^3）高於熱季（5～10月，17.8μg/m^3）。機動車平均密度分別為380.1（汽車）與734.8（機車）輛／平方公里。機動車密度雖與PM_(2.5)濃度呈正向關係但未達統計顯著，溫度為影響雲嘉南地區PM_(2.5)濃度之重要因子。機動車密度對PM_(2.5)的影響可能存在區域差異性。台南市：汽車密度每增加1輛／平方公里，PM_(2.5)增加0.762μg/m^3。嘉義市：機車密度每增加1輛／平方公里，PM_(2.5)增加0.362μg/m^3。嘉義縣：汽車密度每增加1輛／平方公里，PM_(2.5)增加0.977μg/m^3。雲林縣：汽車密度每增加1輛／平方公里，PM_(2.5)增加1.030μg/m^3。結論：溫度為影響雲嘉南地區大氣PM_(2.5)濃度的重要因子。交通污染源對雲嘉南地區PM_(2.5)的影響可能存在區域差異性，值得後續深入探討。
  Objectives: Levels of atmospheric fine particulate matter (PM_(2.5)) may be affected by the traffic sources. This article explores correlations between the PM_(2.5) level and traffic sources in Yun-Chia-Nan area in Taiwan. Methods: Data regarding PM_(2.5) levels and motor vehicle density in 2015 were obtained from Environmental Protection Administration monitoring stations and Environmental Resource Database, respectively. The study area covers four southwestern districts of Taiwan: Yunlin County, Chiayi City, Chiayi County, and Tainan City. Mean, median, minimum, and maximum values for PM_(2.5) levels, meteorological variables, and motor vehicle densities were calculated. The relationship between PM_(2.5), meteorological variables, and motor vehicle density was estimated using the Spearman's rank correlation coefficient. A multiple linear regression model was used to explore the associations of PM_(2.5) with motor vehicle density. The study duration was January-December 2015. Results: In 2015, the annual mean PM_(2.5) level was 26.4 µg/m^3; PM_(2.5) levels were higher in the colder season (November-April; 35.0 µg/m^3) than in the hotter season (May-October; 17.8 µg/m^3). The motor vehicle densities were 380.1 car/km^2 and 734.8 motorcycle/km^2. The relationship between motor vehicle density and PM_(2.5) level was positive, but nonsignificant. Temperature was the major factor influencing the PM_(2.5) levels. Moreover, the effects of motor vehicle density on the PM_(2.5) level varied geographically: for every 1-car/km^2 increase, the PM_(2.5) level increased by 0.762, 0.362, 0.977, and 1.030 µg/m^3 in Tainan City, Chiayi City, Chiayi County, and Yunlin County, respectively. Conclusions: Temperature is the major factor influencing PM_(2.5) levels in Yun-Chia-Nan area in Taiwan. The effect of traffic sources on PM_(2.5) level might have geographic variation. Further relevant research is warranted.
• 19 - 30
• 10.6288/TJPH.201902_38(1).107104

• Link 原著 Original Article
• 經濟景氣與生命末期照護相關字詞搜尋之關係探討－以Google趨勢分析Macroeconomic conditions and Internet search for end-of-life care: an analysis using Google Trends
• 吳丕玉、王鶴偉、陳端容
  PI-YU WU, HO-WEI WANG, DUAN-RUNG CHEN
• Google趨勢 ； 臨終照護 ； 總體經濟 ； 經濟衰退
  Google Trends?;?end-of-life care?;?macroeconomics?;?economic recession
• 目標：照護生命末期的家人對許多照護者而言會造成龐大的經濟壓力，經濟景氣變壞是否與增加人們關心生命末期照護的訊息，尚未有探討。本研究藉由Google趨勢收集的資料，探討生命末期照護相關字詞搜尋狀況與經濟景氣之關係。方法：自雅虎財經網站下載2008年1月至2013年12月間的每週台灣加權股價指數作為經濟景氣的指標。從Google趨勢的網站下載台灣2008年1月至2015年12月間生命末期照護相關字詞的搜尋強度。分析方式為對各個關鍵字搜尋指數進行與台股大盤指數間的相關性分析，除了同期間指數與關鍵字搜尋指數的比較外，也分別對股價指數與各個關鍵字3,6,9個月後的關鍵字搜尋指數做相關分析，以探討股價對關鍵字搜尋具有同期效應或領先預測的效應。結果：整體而言，生命末期照護相關字詞的Google搜尋趨勢與反映經濟景氣的股市大盤指數間呈現負相關。股市大盤的股價走勢與其後的6個月後或9個月後的末期照護相關字詞相關性較為顯著，顯示當大環境經濟不理想時，人們對於生命末期病人照護的關切增加，人們想要了解安寧緩和療護的照護方式的搜尋行為也越多。結論：總體經濟景氣的變化與人們的網路搜尋行為有一定的關係，間接可幫助了解經濟環境與個人的臨終照護決策的相關性。
  Objectives: End-of-life (EOL) care imposes a significant financial burden on most caregivers, but the relationship between macroeconomic conditions and people's EOL care awareness remains unknown. This study employed Google Trends (GT) to explore the association between Internet search for EOL care and macroeconomic conditions. Methods: The weekly average Taiwan Stock Exchange (TWSE) weighted index from January 2008 to December 2013 was used as the macroeconomic index. Local Google searches for EOL care-related keywords from January 2008 to December 2015 were downloaded from GT. A series of correlation analyses was performed on the TWSE and the GT for each keyword at the concurrent time and at 3, 6, and 9 months later to ascertain the nowcasting and forecasting effect. Results: The GT indices for EOL care-related keywords were negatively correlated with the concurrent and previous TWSE indices. For each keyword, either the 6-month or 9-month forecasting effect was higher than the concurrent nowcasting effect, suggesting that worse economic conditions (lower TWSE) might be a factor associated with people's awareness of advanced EOL care. Conclusions: To better understand how people may be more sensitive to circumstances related to EOL care decisions, macroeconomic conditions and the implications of Internet search behavior should be further considered.
• 31 - 41
• 10.6288/TJPH.201902_38(1).107051

• Link 原著 Original Article
• 護理人員留任意願之相關因素探討：護理工作環境與性格特質的交互作用Factors related to nurses' intentions to stay: the interaction between nurses' work environment and personality traits
• 曾旭民、陳慧玲、莊月娥、黃瑋婷
  HSU-MIN TSENG, HUI-LING CHEN, YUEH-O CHUANG, WEI TING HUANG
• 留任意願；護理工作環境；五大性格；職場壓力
  ?retention intention?;?nursing work environment?;?big five personality?;?work stress
• 目標：全球化護理人員短缺已成為健康照護領域重要的課題。雖然有相當多的研究探討護理人員留任的相關課題，但多數研究大多關注於組織政策或個別護理人員的態度等相關因素，較少針對個體與環境交互作用之影響。本研究的主要目的乃依據個人與環境互動的理論架構，了解影響護理人員留任意願之相關因素，並特別著重於臨床工作環境、性格特質與其交互作用。方法：採橫斷式問卷調查研究設計，取樣北部某醫學中心之正職護理人員，研究工具包括基本與工作相關特質、護理工作環境、五大性格、護理職場壓力、與留任意願等量表。結果：本研究有效回收170份。階層迴歸分析結果顯示，工作環境與護理留任意願具顯著關係，而性格特質中之情緒穩定性對工作環境中的工作量合理性構面因素與留任意願之顯著關係具有交互效果。結論：研究結果能提供醫院重要的應用價值，護理管理者應考量採用差異化人力資源管理策略，依據不同性格特質進行工作特性安排的適配性，進而提升護理人員留任意願。
  Objectives: The current worldwide nursing shortage represents a major problem for the health sector. Despite extensive investigation into predictors of nurses' intention to stay, most studies have focused on the attitudinal and organizational aspects concerning individual nurses. This study, however, employed a framework of person-environment interactions as the basis for an investigation of the interaction effects of personal (i.e., Big Five personality traits) and contextual factors (i.e., perceived practice work environment) on nurses' intentions to stay. Methods: A cross-sectional survey was conducted on registered nurses at a medical center in northern Taiwan. The nurses completed a questionnaire with items covering demographics and the following topics: work, Big Five personality traits, work environment, work stress, and intention to stay. Results: One hundred seventy nurses completed the questionnaire. Hierarchical regression analysis results indicated that the nurses' intentions to stay were largely influenced by how they perceived their work environment as well as the interaction between perceived workload and the emotional stability personality trait. Conclusions: This study's findings have critical implications for hospitals because they can be used to promote an effective work environment using differential human resource strategies to reduce nursing turnover.
• 42 - 52
• 10.6288/TJPH.201902_38(1).107082

• Link 原著 Original Article
• 居家醫療資訊科技服務平台之需求探討Exploring the requirements of home medical carers and carees for a technological information platform
• 陳雅美、王映捷、黃國晉、黃松共、楊昆澈、 姜遠萍、鍾國彪
  YA-MEI CHEN, YING-CHIEH WANG, KUO-CHIN HUANG, SONG-KONG HUANG, KUEN-CHEH YANG, YUAN-PING CHIANG, KUO-PIAO CHUNG
• 居家醫療資訊科技服務平台；焦點團體訪談；內容分析法
  ?technological information platform of home medical care services?;?focus group?;?content analysis
• 目標：本研究旨在了解居家醫療服務過程之主要參與者對於資訊科技平台的需求，以作為國家相關單位及長照產業之參考。方法：本研究以居家醫療服務的主要參與者為研究對象，包含三類型：（1）醫師、（2）護理師，及（3）長照服務的使用者與潛在使用者，採立意抽樣方式取得研究樣本，運用焦點團體訪談進行資料收集。後續，使用內容分析法（content analysis）進行資料分析，將訪問內容進行主題編碼歸類，進而了解臨床人員及民眾對於居家醫療資訊科技服務平台的需求及期待。結果：臨床人員及民眾對於居家醫療資訊科技服務平台的需求大致可歸納為四大主題，分別為：（1）提升行政效率、（2）照護連續性、（3）長期照護服務諮詢與整合、及（4）照護品質。醫護端期望透過科技平台使作業流程更為精簡順暢，以協助行政效率有所提昇，並希冀科技平台可輔助進行資訊交換共享，以掌握個案的生理資訊，進而增進照護連續性及照護品質；長照服務的使用者與潛在使用者則期望透過平台追蹤專業照顧人員的服務狀況並與其進行即時互動，同時亦希望平台協助進行相關評估且將服務進行整合，並透過單一窗口獲得高品質的服務。然而，目前應用於我國居家醫療服務過程的相關資訊系統，因具有（1）照護資訊分散於各資訊系統；（2）行政管理作業支援功能不夠全面；（3）無法及時提供照護資源的相關資訊；（4）無法達成雙向交流等不足之處，故無法滿足上述需求。結論：雖然台灣政府已有開發居家醫療服務的相關資訊平台，以促進居家醫療服務的發展，然這些資訊系統皆無法完全滿足使用者的需求。因此如何發展具備（1）提升行政效率、（2）增進照護連續性、（3）協助長期照護服務諮詢與整合、及（4）促進照護品質等功能的整合資訊平台將會是我國為未來相關單位應優先考量的面向。
  Objectives: The purpose of this study was to explore the requirements of home medical carers and carees for a technological information platform. Methods: This was a qualitative study. The research participants, recruited through purposive sampling, were individuals with experience of home medical services, including doctors, nurses, and users and potential users of long-term care services. Content analysis was used to analyze the text. Results: Four themes emerged regarding the requirements that a technological information platform should meet: (1) high administration efficiency; (2) promotion of continuity of care; (3) integration of long-term care services and service consultation; and (4) promotion of quality of care. The doctors and nurses who administered home medical care stated that the platform should promote continuity and quality of care by improving work efficiency and simplifying communication between health care professionals and caregivers (or care recipients). The users and potential users of home care services hoped that they could obtain information about their service status and communicate with health care professionals. In addition, the users and potential users expected that the platform could integrate long-term care services and provide related health and welfare information; moreover, they wished to have access to high quality services through one single platform. Based on our review, the current information technology system of home medical care in Taiwan faces the following four challenges in meeting users’ needs: (1) care information is held in various information systems; (2) the administrative support functions are not sufficiently comprehensive; (3) information on care resources is not provided in a timely manner; and (4) communication is monodirectional, from carers to carees. Conclusions: Although the Taiwanese government has developed several platforms to facilitate the development of home medical care services, these platforms do not meet users’ needs. In an integrated information platform, the following functions must be prioritized: (1) high administrative efficiency; (2) promotion of continuity of care; (3) integration of long-term care services with service consultations; and (4) promotion of quality of care.
• 53 - 65
• 10.6288/TJPH.201902_38(1).107064

• Link 評論 Commentary
• 評論：居家醫療資訊科技服務平台之需求探討Commentary: exploring the requirements of home medical carers and carees for a technological information platform
• 劉立凡
  Li-Fan Liou
• 無
  none
• 無
  none
• 66 - 66
• 10.6288/TJPH.201902_38(1).10706401

• Link 原著 Original Article
• 老人定點用餐服務現況及參與用餐者之自評效益－以台北市為例Current status of congregate meal services for elderly people and self-perceived benefits of the participants by using Taipei City as an example
• 王映捷、 陳雅美、李玉春、楊銘欽
  YING-CHIEH WANG, YA-MEI CHEN, YUE-CHUNE LEE, MING-CHIN YANG
• 無
  none
• 無
  none
• 67 - 80
• 10.6288/TJPH.201902_38(1).107070

• Link 原著 Original Article
• 台灣偏頭痛病患發生睡眠障礙之風險：一項全國性之世代研究The risk of incident sleep disorder in migraine patients in Taiwan: a nationwide cohort study
• 李建瑩、蔡雅芳、張雯婷、蔡東翰、黃光華
  CHIEN-YING LEE, YA-FANG TSAI, WEN-TING CHANG, TUNG-HAN TSAI, KUANG-HUA HUANG
• 偏頭痛；睡眠障礙；用藥型態；全民健保資料庫；世代研究
  ?migraine?;?sleep disorder?;?prescription patterns?;?National Health Research Institutes Database?;?cohort study
• 目標：偏頭痛與睡眠障礙之相關性是複雜的，本研究探討偏頭痛與睡眠障礙之相關性，同時瞭解其他共病與偏頭痛之用藥是否會影響其相關性。方法：本研究為一回溯性世代研究，以承保抽樣歸人檔作為分析資料庫，以新發生之偏頭痛病患作為研究組，且排除曾出現睡眠障礙者，並利用傾向分數配對法，取得無偏頭痛之病患作為對照組。取得研究對象後，進一步控制相關變項，以Cox比例風險模式探討偏頭痛與發生睡眠障礙之相關性。本研究之控制變項為共病情況及偏頭痛之用藥型態。結果：偏頭痛患者之睡眠障礙發生率為6.86%。偏頭痛患者相較於無偏頭痛患者，其發生睡眠障礙之風險比為2.71倍（95% CI=2.50-2.94）。高血壓患者（HR=1.45; 95% CI=1.29-1.64）、重鬱症（單純發作）患者（HR=1.94; 95% CI=1.29-2.92）與精神官能性憂鬱症患者（HR=1.61; 95% CI=1.23-2.11），皆有較高之發生睡眠障礙之風險比。偏頭痛用藥與睡眠障礙則無相關性。結論：偏頭痛與發生睡眠障礙確實具有相關性，高血壓疾病、重鬱症（單純發作）及精神官能性憂鬱症，亦為睡眠障礙之危險因子。
  Objectives: The correlation between migraine and sleep disorders is complex and may be causal in nature. This study investigated the correlation between migraine and sleep disorders and discussed whether other comorbidities or prescription patterns influence their relevance. Methods: This retrospective cohort study used data from the Longitudinal Health Insurance Database of Taiwan. The study subjects were patients with new-onset migraine who had not been diagnosed as having a sleep disorder prior to migraine diagnosis. To avoid selection bias, we applied propensity score matching to select patients without migraine as the control group. We used the Cox proportional hazard model to analyze differences in the risk of sleep disorder among patients with migraine after controlling for related variables, including comorbidities and migraine prescription patterns. Results: The rate of incident sleep disorders was 6.86% in patients with migraine. The hazard ratio (HR) in patients with migraine was higher than that in patients without migraine (HR = 2.71; 95% confidence interval [CI] = 2.50-2.94). Patients with hypertension (HR = 1.45; 95% CI = 1.29-1.64), major depression (single episode) (HR = 2.38; 95% CI = 1.47-3.84), or neurotic depression (HR = 1.93; 95% CI = 1.41-2.64) had a higher risk of sleep disorders. Furthermore, migraine medication was not related to sleep disorders. Conclusions: Patients with migraine have a higher risk of sleep disorders than do those without migraine. The rate of sleep disorders is highest in patients with hypertension, major depression (single episode), or neurotic depression.
• 81 - 91
• 10.6288/TJPH.201902_38(1).107092

• Link 研究紀要 Research Brief
• 微生物螢光檢測儀於醫院公共空間空氣細菌篩檢與定量分析Preliminary study on monitoring airborne bacteria in hospital environments using a microbe fluorescence sensor
• 簡誌良、 蘇富榮、 曠永銓、徐嘉彬、 楊宗儒、 施志恆
  CHIH-LIANG CHIEN, FU-JUNG SU, YUNG-CHUAN KUANG, CHIA-PIN HSU, TSUNG-JU YANG, CHIH-HENG SHIH
• 醫院室內空氣品質；微生物螢光檢測；快速巡檢工具
  ?indoor air quality in hospitals?;?microbe fluorescence detection?;?fast screening tool
• 目標：細菌為室內空氣品質管理法中醫療機構的管制項目之一，公共空間細菌量檢測須以培養方式取得，尚無直讀儀器用於較快速的巡檢作業。本研究評估微生物螢光分析法作為快速巡檢工具之可行性與適宜性，針對細菌超標醫院建立可快速篩檢室內空間細菌濃度分布的輔助工具。方法：在8處列管醫院的管制空間、空調機房內及外氣引入口，以微生物螢光檢測儀與可培養性細菌採樣器進行平行採樣，其中可培養性細菌的採樣及分析，乃依照環檢所的標準檢測方法進行。結果：就檢測儀在高細菌濃度下的表現，當測得螢光係數大於2.82 V/s時，有極大機會細菌濃度可能高於1,500 CFU/m^3的法規限值。結論：對於已知細菌超標的醫院，可利用螢光檢測儀作為輔助，研判疑似污染源位置之污染程度，快速確認細菌濃度空間分佈潛勢並對可能超標區域進行改善，其結果應具有一定程度的參考價值。待累積更多比對數據後，可利用巡檢的螢光係數來推估細菌濃度，以利醫療機構即時現場評估室內空間細菌的可能分布。
  Objectives: In accordance with indoor air quality management regulations, the bacteria concentrations of public spaces in healthcare facilities are determined by using microbial cultures, a method that is time-consuming and inefficient. This study examines the feasibility and applicability of using a semi-portable fluorescence microbe sensor in hospital settings to quickly obtain the potential spatial distribution of airborne bacteria. Methods: Eight hospitals were selected as sampling sites. Both the fluorescence microbe sensor and microbial culture of air samplers were used. Parallel sampling was performed in hospital areas accessible to the public, air conditioning facilities, and fresh air inlets. Results: Fluorescence intensities obtained by microbe fluorescence detection were found to be highly correlated to the bacterial concentrations at 21 sampling locations. When the measured fluorescence intensity is higher than 2.82 V/s, the estimated bacterial concentration has a high probability of exceeding 1,500 CFU/m^3, which is the maximum legal limit. Conclusions: In hospitals where levels of bacteria exceed the legal threshold, the microbe sensor can be used as a fast screening tool to identify the potential spatial distribution of bacteria in real time. Upon completing further studies and comparing more data, using fluorescence intensity to estimate bacteria concentrations is expected to become a more feasible and effective method of monitoring indoor air quality in hospitals.
• 92 - 101
• 10.6288/TJPH.201902_38(1).107105

• Link 研究紀要 Research Brief
• 森林圖繪製的探討應用：以MS-Excel繪製為例Feasibility of using Microsoft Excel to draw forest plots
• 蔡崇煌、 蔡濬莉、王俊堯
  CHUNG-HUANG TSAI, JUN-LI TSAI, JIUN-YAO WANG
• 森林圖；刺胳針；統合分析；新英格蘭醫學雜誌
  ?forest plot?;?Lancet?;?meta-analysis?;?N Engl J Med
• 目標：免費之RevMan 5.3套裝軟體所畫之森林圖不易做修改，本研究嘗試用微軟Excel 2016重新製作常用統合分析軟體的分析結果及新英格蘭醫學雜誌（NEJM）發表的文獻森林圖。方法：以PubMed搜尋題目關鍵詞為“Excel”AND”meta analysis或meta-analysis”、“Excel”AND”forest plot”，中文資料由思博網中文電子期刊服務搜尋，關鍵字同上。並在Google搜索引擎搜尋關鍵詞為“森林圖或forest plot”作圖法。已發表文獻所呈現的森林圖，主要參考NEJM、刺胳針等期刊的森林圖呈現法。結果：共找到4篇文獻，皆是探討如何使用Excel計算效應量及95% CI，畫森林圖皆非其重點。由期刊上搜尋到的圖形、多元迴歸分析及統合分析套裝軟體計算的結果，皆可用Excel直接畫出，且無論在線條或標記形狀，皆具有極大的修改自由性，更重要為Excel相對較常見，且不難學習。結論：使用Excel製作森林圖可以符合國際學術期刊的呈現法，且容易修改，適合研究及教學的統計實際應用。
  Objectives: Forest plots drawn using the free software package RevMan 5.3 are difficult to modify. This study sought to recreate forest plots from the results of meta-analyses in published articles of the New England Journal of Medicine (NEJM) by using Microsoft Excel 2016. Methods: The topic keywords ”Excel AND meta-analysis OR meta-analysis” and ”Excel AND forest plot” were searched for in PubMed and a Chinese-language electronic journal service. In addition, the keyword ”forest plot” was searched for using Google. Example forest plots were collected from journals such as the NEJM and Lancet. Results: Four articles were found, all of which explored how to use Excel to calculate the effect size and 95% confidence interval without particular focus on drawing forest plots. All forest plots found in the journal search could be plotted using Excel; furthermore, Excel could be used to calculate the results of the multivariate regressions in the meta-analyses. Excel provides great freedom to modify forest plots regardless of the line or shape required and is a commonly used piece of software that is easy to learn. Conclusions: Excel can be used to draw forest plots that are consistent with presentations used in journals; moreover, such plots are easily modifiable and are therefore suitable for practical application in medical research and by teaching personnel.
• 102 - 110
• 10.6288/TJPH.201902_38(1).107108