簡國龍陳秀熙臺灣大學：李依錦Lee, Yi-ChinYi-ChinLee2007-11-282018-06-292007-11-282018-06-292005http://ntur.lib.ntu.edu.tw//handle/246246/59219背景 疲勞（fatigue）不只是臨床病人常見的主述在健康族群中亦然，與疲勞相關的危險因子隨著不同的族群而不同，在台灣的研究所學生應該也是疲勞的高危險群，但是卻很少研究針對這個族群進行探討。 目的 因此本研究的目的為針對研究所學生估計疲勞的盛行率以及找出引起疲勞的顯著危險因子。 方法 國立台灣大學在2004年秋季應有2688位新進研究生接受健康檢查，而實際上為2144位參加，這2144位當中有1806位同意加入我們的疲勞調查研究（回應率是84%），在健康檢查進行的同時我們收集了基本資料、身體活動量問卷、以及由主觀疲勞經驗、工作動機、注意力、和活動力四個面向組成的個人疲勞強度問卷（Checklist Individual strength questionnaire）。Cronbach’s α用來檢定個人疲勞強度問卷的內在一致性。將個人疲勞強度問卷分數大於76分定義為疲勞來估計疲勞的盛行率。研究危險因子時，我們進行兩種方式，一是將疲勞視為二元變項，用邏輯式迴歸模型來估計，另一是將疲勞分成四個等級，分別為正常、輕微、中等以及嚴重，用比例發生比模型(Proportional Odds model)來估計。 結果 男性疲勞的盛行率是45.8%，女性是48.9%。個人疲勞強度問卷的Cronbach´s α值高達0.92，有很好的內在一致性，對於男性問卷上顯示有三個概念，而女性有四個。進行變項之間彼此校正的多變項分析後，統計上仍有顯著的變項有：博士新生比碩士新生低39%危險性發生疲勞(OR = 0.61 (95% CI: 0.45 ~ 0.82)）、有身體疾病但不包括肝炎患者發生疲勞的危險性為無疾病的1.67倍 （OR = 1.61 (95% CI: 1.19 ~ 2.19)）、每周至少失眠一次比未失眠者高2.23倍 （OR = 2.23 (95% CI: 1.79 ~ 2.76)）、三餐規則進食比不規則者降低31%的危險性 （OR = 0.69 (95% CI: 0.51 ~ 0.80)）、有規則運動習慣比沒有運動習慣者降低31%（OR = 0.68 (95% CI: 0.54 ~ 0.87)）、身體活動量從低到高發生的危險性比最低的降低情形分別為28%、50%、以及64%（OR = 0.72 (95% CI: 0.54 ~ 0.95) for Q2, 0.50 (95% CI: 0.38 ~ 0.66) for Q3, and 0.36 (95% CI: 0.26 ~ 0.50) for Q4.），身體活動量部份也符合趨勢檢 定(trend test, p<0.01)。而相似的結果也出現在使用比例發生比模型。 結論 本研究發現使用個人疲勞強度問卷測量研究所學生的疲勞有相當高的盛行率，這樣的高盛行率不只和患有慢性疾病或是失眠有關，缺乏規則的生活型態以及運動也被認為是重要的因素。這樣的發現有助於臨床上面對抱怨疲勞的研究所學生的處置。Background: Fatigue is not only commonly seen in clinical patients but also prevails in healthy population. Risk factors associated with fatigue were reported to vary from population to population. Few studies have been conducted to address fatigue in postgraduate students that are supposed to be a high risk group for being afflicted with fatigue in Taiwan. Objectives: The aims of this thesis are therefore to estimate the prevalence rate of fatigue and to identify significant risk factors responsible for fatigue with the emphasis on postgraduate students. Materials and Methods: In the fall term of 2004, a total of 2688 graduate students newly admitted to National Taiwan University were invited to take health check-ups. Of 2688 invited students, 2144 attended the physical check-up. Of 2144 attendee, 1806 agreed to participate in our fatigue survey with 84% of response rate. During the check-up time , basic information and the Checklist Individual Strength questionnaire (CIS-20) with four dimensions was administered: subjective fatigue, reduced motivation, reduced concentration, and reduced activity and the measurement of physical activity questionnaire were collected. The Cronbach’s α was adopted to test the internal consistency of CIS-20’s. The definition for estimating prevalence rate of fatigue was CIS-20 greater than 76. We investigated risk factors associated with fatigue in two types of outcome, one based on binary outcome and another leaning on fatigue score divided into four levels (normal, mild, moderate, and severe). The former was analyzed by using logistic regression model and the latter using proportional odds model. Results: The prevalence rate of fatigue was 45.8% for males and 48.9% for females. The CIS-20 has good internal consistency with Cronbach´s α high up to 0.92, yielding three constructs for male and four constructs for female. In multivariate analysis with adjusting for variables in each other, variables still remained statistically significant including the identity (OR = 0.61 (95% CI: 0.45 ~ 0.82) doctorate students versus master students), individuals having systemic disease without hepatitis history (OR = 1.61 (95% CI: 1.19 ~ 2.19)), insomnia at least once per week versus none (OR = 2.23 (95% CI: 1.79 ~ 2.76)), sleeping greater than seven hours versus less than seven hours (OR =0.7 (95% CI: 0.56 ~ 0.87)), regular 3 meal per day versus irregular (OR = 0.69 (95% CI: 0.51 ~ 0.80)), regular exercise versus irregular (OR = 0.68 (95% CI: 0.54 ~ 0.87)), physical activity in four levels from the smallest to the greatest ( OR = 0.72 (95% CI: 0.54 ~ 0.95) for Q2, 0.50 (95% CI: 0.38 ~ 0.66) for Q3, and 0.36 (95% CI: 0.26 ~ 0.50) for Q4). There was a statistically significant trend in physical activity by four levels (p < 0.01). Similar findings were reported by using proportional odds model. Conclusion: High prevalence rate of fatigue measured by CIS-20 among graduate students has been demonstrated. Such high risk among these postgraduate students may not be only related to existing chronic disease and insomnia but also attributed to lacking of regular life styles and exercise. These findings have a significant implication for clinical management of postgraduate students with fatigue as a chief complaint.ABSTRACT 1 中文摘要 4 1. INTRODUCTION 6 2. LITERATURE REVIEW 9 2.1. PREVALENCE OF FATIGUE 9 2.2. MEASUREMENT OF FATIGUE (QUESTIONNAIRES OF FATIGUE) 10 2.3. RISK FACTORS ASSOCIATED WITH FATIGUE 13 2.4. MEASUREMENT OF PHYSICAL ACTIVITY 16 2.5. SUMMARY 17 3. MATERIALS AND METHODS 21 3.1. STUDY SUBJECTS 21 3.2. DATA COLLECTION 21 3.3. MEASUREMENT OF VARIABLES 22 3.3.1. Questionnaire of Fatigue (CIS) 22 3.3.2. Measurement of Physical Activity 23 3.3.3. Other Variables 23 3.3.3.1. Health Status: 24 3.4. STATISTICAL ANALYSIS 24 3.4.1. Descriptive Analysis 24 3.4.2. Validity and Reliability of Questionnaire of Fatigue 25 3.4.2.1. Internal Consistency 25 3.4.2.2. Factor Analysis 26 3.4.3. Multiple Logistic Regression 26 3.4.3.1.1. Binary Outcome 26 3.4.3.1.2. Ordinal Outcome 27 4. RESULTS 28 4.1. DESCRIPTIVE FINDING 28 4.1.1. The Distribution of the CIS-20 Score 28 4.1.3. The Prevalence Rate of Fatigue in Each Group 28 4.1.4. Comparison of CIS-20 Score and Physical Activity Level Between Both Genders 29 4.1.5. Comparison of CIS-20 Total Score By Risk Factor 29 4.2. RELIABILITY AND VALIDITY OF QUESTIONNAIRE 30 4.2.1. Internal Consistency of CIS-20（Cronbach´s α） 30 4.2.2. Factor Analysis of CIS-20 32 4.3. ASSOCIATION 34 4.3.1. Multiple Logistic Regression 34 4.3.2. Proportional Odds Model 37 5. DISCUSSION 39 5.1. INTERPRETATION OF MAJOR FINDINGS 39 5.2. COMPARISON WITH OTHER STUDIES 40 5.3. CLINICAL IMPLICATION 43 5.4. LIMITATION 45 6. CONCLUSION 47 APPENDIX 48 A. LISTS OF FIGURES 48 A.1 CIS-20 ITEMS BY FREQUENCY IN BOTH GENDER 48 A.2 CIS-20 BY PERCENTAGE IN BOTH GENDER 48 B. LISTS OF TABLES 49 B.1 THE CHARACTERISTIC OF SUBJECTS BY GENDER 49 B.2 PREVALENCE OF FATIGUE IN EACH GROUP 52 B.3 COMPARISON OF FATIGUE SCORE WITHIN BOTH GENDER GROUP 53 B.4 COMPARISON OF PHYSICAL ACTIVITY LEVEL WITHIN BOTH GENDER GROUP 53 B.5 COMPARISON OF FATIGUE SCORE BY THE RISK FACTOR WITHIN BOTH GENDER 54 B.6 LOGISTIC REGRESSION MODEL 57 B.7 PROPORTIONAL ODD MODEL 59 B.8 INTERNAL CONSISTENCY OF THE CIS QUESTIONNAIRE 60 B.8.1 Total fatigue 60 B.8.2 Subjective Fatigue 61 B.8.3 Reduced Motivation 62 B.8.4 Reduced Concentration 63 B.8.5 Reduced Activity 64 B.9 FACTOR ANALYSIS 65 B.9.1 Rotated Factor Pattern of Male 65 B.9.2 Rotated Factor Pattern of female 66 B.10 TABLES OF LITERATURE REVIEW 67 B.10.1 prevalence of fatigue 67 B.10.2 Risk factors associated with fatigue 68 C. ASSOCIATED TABLES AND PRINTS 72 C.1 CIS-20 72 C.2 QUESTIONNAIRE OF PHYSICAL ACTIVITY 73 C.3 BASIC INFORMATION SHEET 74 REFERENCE 76en-US疲勞, 研究生, 危險因子, 個人疲勞強度問卷, 邏輯式迴歸, 比例發生比模型, 運動, 身體活動量fatigue, graduate student, risk factor, Checklist Individual Strength questionnaire , logistic regression, proportional odds model, exercie, physical activity[SDGs]SDG3thesis