台灣社區居民重度憂鬱症與自殺行為之流行病學研究

Epidemiological Studies on Major Depressive Disorder and Suicidal Behaviors among Community Residents in Taiwan

廖士程(Shih-Cheng Liao)

指導教授 ： 陳為堅教授

共同指導教授 ： 李明濱教授(Ming-Been Lee)

國立臺灣大學/公共衛生學院/流行病學與預防醫學研究所/博士(2012年)

https://doi.org/10.6342/NTU.2012.10399

摘要

研究緣起
重度憂鬱症與自殺是當代重大公共衛生問題，依據世界衛生組織預測，西元2020年時全球十大死因，自殺將排名第九，而全球衛生負擔，以「生命損失年數」與「伴隨殘障存活年數」估算，最高的十大「疾病與傷害」當中，重度憂鬱症將排名第二，僅次於缺血性心臟病，而包括重度憂鬱症在內之常見精神疾病，也是自殺行為的重要危險因子。
自1980年代以來，使用結構式診斷工具之臺灣社區樣本精神疾病流行病學研究結果，關於重度憂鬱症與自殺方面，仍有許多待解決的問題，其中包括：(一)台灣曾經是跨國精神疾病流行病學研究中，重度憂鬱症終生盛行率最低的國家；然而隨著時代的演進，西方國家重度憂鬱症盛行率仍高甚至大幅增加，臺灣是否仍舊偏低？而可能的解釋為何？(二)1980年代台灣流行病學區域研究(epidemiological catchment area study)結果顯示重度憂鬱症以及其他常見精神疾病之盛行率有明顯城鄉差距(urban-rural difference)，但方向不一致，近年運用全國代表性樣本(national probability sample)及數量化都市化程度分層之研究設計是否能更清楚地呈現這種差異？為進一步防治工作及研究提供區域與環境因素之線索；(三)與西方國家相較台灣重度憂鬱症及常見精神疾病盛行率偏低，但自殺死亡率相較部分西方國家為高，過去文獻指出包括重度憂鬱症之常見精神疾病是自殺的重要危險因子，是否臺灣常見精神疾病與自殺行為之間的關係與西方不同？
研究目的
目的一：探討台灣成人社區居民重度憂鬱症之盛行率及風險因子，並尋求盛行率偏低之可能解釋。目的二：於全國代表性樣本中描述常見精神疾病於台灣不同都市化程度(urbanicity-related)分層以及地理區域(geographical)分層之盛行率分布。目的三：探討台灣成人社區居民自殺行為之風險因子以及常見精神疾病之於自殺行為之人群歸因成分(population attributable fraction: PAF)。
研究設計與方法
本研究利用2003年至2005年於台灣本島實施之「全國精神疾病流行病學調查」進行資料分析。該調查之抽樣方法採多階段分層隨機抽樣法，且納入都市化程度分層於各初段抽樣單位(primary sampling unit)之分層抽樣。本研究共有10135名18歲以上成人機率樣本(probability sample)完成訪查，診斷及臨床資料蒐集工具採用「世界心理衛生研究」（The World Mental Health Study: WMH）所使用「組合型國際診斷會談」(Composite International Diagnostic Interview) 。本研究針對複雜設計普查(complex design survey) 調整多階段分層抽樣 clustering effect 並調整個體反應機率權值之 design-based analysis 方法，進行各項包括都市化程度分層、地理生活圈分層等之盛行率估計、以及危險因子勝算比計算。此外並利用WMH-CIDI當中對於各項功能障礙之結構性測量、其他常見精神疾病之共病狀態、以及求助行為等變因，與 WMH-CIDI 研究工具及相關抽樣方法發源地之美國調查結果相比較，探討盛行率偏低之可能解釋因子。除利用上述 design-based analysis 之 logistic regression 探討自殺行為的相關因子外，利用原本橫斷面調查當中，疾病發病年齡以及自殺行為發生年齡之間的先後次序及差異，運用 Poisson log-linear regression 估計risk ratio以及計算各類常見精神疾病之於自殺行為的人群歸因成分。
結果
本研究共有10135名18歲以上成人機率樣本(probability sample)完成訪查，估計台灣社區成人居民之重度憂鬱症終生盛行率為1.20% (S.E.=0.16)。重度憂鬱症相關風險因子包括女性、年齡40歲以下、以及失婚或鰥寡。鄉村居民被診斷為重度憂鬱症之風險較都會區(urban)及亞都會區(suburban)居民顯著為低。與在美國實施使用相同診斷工具之全國代表性樣本研究結果相較，重度憂鬱症在台灣不僅盛行率較低，且與其他常見精神疾患共病比例亦偏低。然而臺灣重度憂鬱症患者報告因為該疾病無法執行工作或日常生活常規的日數是美國樣本的一倍以上(台灣個案74.9天；美國個案35.2天)，但曾經尋求各種專業協助的比例卻只有美國個案的一半以下(台灣個案20.0%；美國個案57.3%)。
四大類常見精神疾患終生盛行率分別為：(一)情感疾患為1.54% (S.E.=0.19)，其中78%為重度憂鬱症；(二)焦慮疾患1.20% (S.E.=0.17)；(三)衝動控制疾患0.58% (S.E.=0.12)；(四)物質使用疾患(不包括尼古丁使用)為3.88% ( S.E.=0.38)，其中98%為酒精濫用或依賴。四大類常見精神疾患於各種都市化程度分層分布有所不同，內化型疾患(internalizing亦即焦慮及憂鬱疾患)之終生盛行率，都會區最高，其次亞都會區，鄉村地區最低。而就外化型疾患(externalizing亦即衝動控制與物質使用疾患)，亞都會區最高，其次鄉村地區，都會區最低。對於性別及年齡分層而言，內化型與外化型疾患之分布不同。整體而言，女性較男性有較高之內化型疾患罹病風險，然而都會區40歲以下男性之內化型疾患終生盛行率，甚至比同年齡層之女性為高。就地理分區而言，無論內化型或外化型疾患，與出生世代及性別之關聯性，於某些特定地理分區呈現差異。
台灣社區成年人自殺行為終生盛行率分別為：自殺意念7.52% (S.E.=0.46)、自殺計畫1.31% ( S.E.=0.16)、以及自殺企圖1.29% ( S.E.=0.16)。四大類常見精神疾患對於自殺行為之效應(effect)，隨著意念、計畫、企圖而逐漸增加。對於自殺企圖而言，任何一種常見精神疾病之PAF可達40%，其中物質使用疾患對於自殺企圖之PAF約為21%，高於其他WMH國家調查結果的10%至14%。
結論
台灣成人社區樣本之重度憂鬱症終生盛行率與美國相較偏低，然而個案之失能程度較高、求助行為亦偏低，台灣的重度憂鬱症防治工作，在公共衛生層面仍有極大努力空間。包括重度憂鬱症在內之常見精神疾患，終生盛行率之城鄉差距在性別年齡分層或是地理分層間呈現分布之差異，暗示環境交互作用之可能性，以及因地制宜防治之重要性。雖然常見精神疾病於台灣社區當中之盛行率較西方大多數國家為低，然而對於自殺行為，尤其是自殺企圖，常見精神疾病仍為重要危險因子，而其中物質使用疾患對於自殺企圖之人群歸因成分，甚至比其他開發中或已開發國家為高。從公共衛生的角度出發，對於常見精神疾病，特別是物質使用疾患的預防與治療，應屬於自殺防治工作之重點。

關鍵字

重度憂鬱症 ； 常見精神疾病 ； 自殺 ； 共病 ； 人群歸因成分

並列摘要

Background
Major depressive disorder (MDD) and suicide are two important public health problems in the world. The community-based epidemiological study is of high value to guide the preventive strategies for suicide and MDD.
In Taiwan, previous community surveys of psychiatric disorders have disclosed the epidemiology of MDD and suicide. However, there are some unanswered questions require further investigation. First, lifetime prevalence of MDD in Taiwan has been reported to be at lower bound around the world, especially in comparison with western countries. What is the possible explanation? Second, conflicting results in urban-rural differences of MDD and CMDs has been reported in previous epidemiological catchment area studies. The association between urbanicity and mental disorders required further investigation with new epidemiological method. Third, in contrast to low prevalence of MDD, the suicide mortality rate is higher in Taiwan than in some Western countries. Since MDD is one of the major risk factors of suicide, the effect of MDD on the risk of suicide in Taiwan might differ from its counterpart in Western countries.
Aims
The specific aims of this dissertation including (I) To investigate prevalence and correlates of major depressive disorder among community residents；(II)To describe the distribution of CMDs across level of urbanicity and geographic clusters from a nationally representative sample; (III) To calculate population attributable fractions (PAF) of common mental disorders for suicidal behaviors.
Design and Method
All of the analyses were based on the data from Taiwan Psychiatric Morbidity Survey (TPMS). TPMS was a survey of common psychiatric disorders in a nationally representative sample of individuals who were age 18 or older, available for a face-to-face interview, and non-institutionalized civilians in Taiwan between Feb 2003 and Dec 2005. Stratified multistage sampling with the principle of probability proportional to size was used for the TPMS.
Design-based analyses of proportion, cross tabulation, and multivariable logistic regression analyses were performed using STATA version 10.0 (Texas, U.S.A.). By using of the quantitatively defined level of urbanicity, we demonstrated the association between urbanicity and common mental disorder from data of a nationally representative sample. For calculating the PAF of each category of mental disorders, the associations were estimated using Poisson log-linear regression model in complex survey data to calculate the risk ratios (RRs).The adjusted population attributable fraction (aPAF) for each risk factor of suicidal behaviors was calculated by Bruzzi’s formula.
Results
Among the 10 135 respondents, the lifetime prevalence of MDD was 1.20% (S.E.=0.2%). Individuals who were divorced or widowed, aged f40 years, and female were at increased risk, whereas rural residents were at lower risk for MDD. Only one-third of Taiwanese individuals with MDD sought help despite having twice the number of lost workdays compared with the US sample.
The lifetime prevalence of CMDs was 6.27% (S.E.=0.53) (include internalizing and externalizing disorders) and the 12-month prevalence of CMDs was 2.64% (S.E.=0.32). There were different pattern of distribution of CMDs across level of urbanicity. The point estimation of lifetime prevalence of internalizing disorders was highest in urban and followed by suburban and rural area. People living in suburban area, however, had highest risk to have externalizing disorders than their counterpart in rural and urban area.
The lifetime prevalence of suicide ideation, plan, and attempt are 7.52% (S.E. = 0.46), 1.31% (S.E. = 0.16), and 1.29% (S.E. = 0.16).The adjusted population attributable fraction (aPAF) for suicidal behaviors increased gradually from suicidal ideation to suicidal attempt. The aPAF for suicidal attempt in Taiwan (45.5% for any mental disorder, 33.1%for mood disorders) is comparable with that in developing and developed countries in WMH survey. Substance use disorders even have higher aPAF in Taiwan than in other WMH countries.
Conclusion
Despite the low prevalence of MDD in Taiwanese adults, the pattern of low help-seeking behavior and profound functional impairment indicates much room for improvement in the early detection of and intervention in major depression in this population. In terms of the distribution of CMDs across different level or urbanicity and different geographic or living circles, there is no simple pattern of association between urbanicity and psychopathology. Such complicated environment-individual interactions should be clarified in the future. CMDs, despite of their low prevalence, are still important risk factors of suicidal behaviors in Taiwan and still the important targets for suicide prevention from the aspects of public health and clinical medicine in Taiwan.

並列關鍵字

major depressive disorder ； common mental disorder ； suicidal behaviors ； comorbidity ； population attributable fraction

參考文獻

Lee M-B, Liao S-C (2006): Risk factors and prevention strategy of suicide. Formosan J Med 10:367-375.

Liu CY, Hung YT, Chuang YL, Chen YJ, Weng WS, Liu JS, et al (2006): Incorporating development stratification of Taiwan townships into sampling design of large scale health interview survey. J Health Manag 4:1-22.

American Psychiatric Association (2000): Diagnostic and Statistical Manual of Mental Disorders DSM-IV-TR Washington.

Anderson LM, Fielding JE, Fullilove MT, Scrimshaw SC, Carande-Kulis VG (2003): Methods for conducting systematic reviews of the evidence of effectiveness and economic efficiency of interventions to promote healthy social environments. Am J Prev Med 24:25-31.

Andrade L, Caraveo-Anduaga JJ, Berglund P, Bijl RV, De Graaf R, Vollebergh W, et al (2003): The epidemiology of major depressive episodes: results from the International Consortium of Psychiatric Epidemiology (ICPE) surveys.[erratum appears in Int J Methods Psychiatr Res. 2003;12(3):165]. Int J Methods Psychiatr Res 12:3-21.