台灣60歲以上婦女尿失禁之研究

Urinary incontinence in Taiwanese women aged 60 or older

謝卿宏(Ching-Hung Hsieh)

指導教授 ： 李孟智

中山醫學大學/醫學院/醫學研究所/博士(2009年)

https://doi.org/10.6834/CSMU.2009.00022

摘要

研究背景與目的: 對老年婦女來說，尿失禁是一個會導致醫學上、社交與心理問題的常見症狀，而且因為尿失禁在年長女性非常常見，所以它常被誤認為是老化的自然過程。但是，老年人尿失禁盛行率的增加，卻意味身體功能的惡化和伴隨而來的殘障的可能，故老年人的尿失禁較特殊，不僅與年輕人的不同，而且治療時會比較棘手。20-59歲台灣婦女的尿失禁盛行率為18.7％，年齡、結婚、喝酒、會藥物過敏、糖尿病、高血壓、體重增加和曾開過婦科刀都是尿失禁的危險因子，都會增加尿失禁的盛行率，由於對台灣高齡女性尿失禁的相關流行病學全無所悉，故本研究旨在評估60歲以上台灣女性的尿失禁盛行率與尿失禁的知識、態度暨行為與其危險因子。
研究方法: 本研究是行政院衛生署家庭計畫研究所(原台灣省家庭計畫研究所)自1989年就已開始規劃、進行的一系列「台灣中老年人保健與生涯規劃」研究中1999年之追蹤調查的一部份，由行政院衛生署家庭計畫研究所與美國密西根大學人口研究中心技術合作，該調查係以多階段取樣法隨機抽樣，依據戶籍登記資料，採用三階段分層隨機抽樣方法抽出之等機率隨機樣本。三階段抽樣過程如下：（1）抽出初抽單位：將全台331個平地鄉鎮市區依據行政區劃、教育程度與總生育率等三項，區分為27層，以最小群的母群人數作為區距，系統隨機抽出散佈在台灣的56個鄉鎮市區，作為第一階段初抽單位，即樣本地區。（2）抽出區段：按鄉鎮市區欲抽樣本人數的一定比例（1/2）抽出若干鄰，做為樣本鄰。（3）抽出個案：以系統隨機抽樣方法從每個樣本鄰中抽出二名老人做為樣本個案。本次調查問卷內容及設計，主要參照台灣省家庭計畫研究所第一次、第二次和第三次老人保健與生活問題調查之問項內容，並參考美、日等相關研究之問卷設計。至於基本背景、過去病史、婦產科病史、尿失禁、尿失禁的知識、態度暨行為與其他下泌尿道症狀等內容以用來研究尿失禁盛行率和尿失禁危險因子的問卷，也一併加入這次的調查研究中，問卷內容大多從Prof. D. R. Ostergard所設計的“URG-data base system. Urodynamics and gynecologic urology. HISTORY” 翻譯修正而來的。為提高調查問卷的量度能力，在正式訪查之前，由行政院衛生署家庭計畫研究所人員就問卷初稿進行三次試查訪問，並參酌國內外專家、學者意見修訂，經多次討論後定案，本調查以面對面訪查方式收集資料，由特約調查員持問卷先進行一對一訪問，除了被選之樣本個案本人因意識不清，重病或聾啞等無法自由回答，可因瞭解個案狀況之家人或照顧者代答問卷中合適於代答者代答之問項外，一律由樣本個案本人接受訪問，問卷中每一個別問題的統計都是依據回答該問題的所有受訪人數來分析。所有資料都使用電腦建檔並以SAS軟體(SAS Institute, Cary, NC)分析處理，並依處理資料之性質，分別使用卡方檢定、費雪正確概率檢定與羅吉斯回歸分析來檢定變數間的關聯性，且用勝算比(OR)與其95%信賴區間(CI)來估算風險，P值小於0.05表示統計上有意義。
研究結果: 本研究以多階段取樣法隨機抽樣取出一個包含655個60-69歲和1,755個70歲以上婦女的樣本，追蹤時已有621人死亡，其中有590人是自1989年即開始研究追蹤世代之成員，另31人則是1996年加入的世代中60至69歲的婦女，這兩個世代中分別各有181人與91人因拒絕訪查、失蹤或未能回答問卷而無法完成調查，故完訪率為84.8% (1,517/1,789)。本研究顯示60歲以上台灣婦女尿失禁的盛行率是29.8% (485/1,517, SE 0.012, 95%CI 0.274-0.322);在自訴有尿失禁的受訪者中，有93人因無法指出何時發生尿失禁，故沒有列入統計分析，有69.9% (250/367)的尿失禁受訪者報告她們不到一個月就會發生一次尿失禁。至於尿失禁的知識、態度暨行為調查則顯示54.3% (822/1,514)的受訪者已知道尿失禁的意思，大多數(84.61%, 1,281/1,514)的婦女報告說，一旦她們有尿失禁，她們就會告訴醫師， 89.21% (1,384/1,511)的受訪者則說，如果她們有尿失禁，她們會尋求治療，有30.3% (147/485)的尿失禁婦女已去找醫師診治，至於有尿失禁而未求醫的主要原因則是害羞。尿失禁的危險因子則有年齡(OR=1.04, 每增加一歲, 95%CI 1.02-1.05)，糖尿病(盛行率39.8%，P=0.002)、高血壓(盛行率39.5%，P=0.001)、曾有婦科手術(盛行率41.4%，P=0.001)、子宮切除手術(盛行率42.4%，P=0.003)、會藥物過敏(盛行率41.3%，P=0.001)、抽煙者(盛行率45.5%，P=0.010)、荷爾蒙治療(盛行率41.5%，P=0.026)與身體質量指數(OR=1.05, 每增加一單位, 95%CI 1.02-1.09)。至於生產數、生產途徑、喝酒與否和是否結婚，都不會有增加尿失禁的風險。
研究結論: 台灣60歲以上婦女尿失禁的盛行率與西方社會女性相近，由於害羞，或認為尿失禁是自然的老化現象(尤其是曾生產者)，故僅有少數的婦女會就醫，其結果則會造成尿失禁治療不足的情形。也因此，加強尿失禁的衛教，對幫助大眾瞭解尿失禁，並使大家知道尿失禁是可以治療的，是絕對有其需要；本研究顯示糖尿病、高血壓、會藥物過敏、較肥胖者、抽煙者、有荷爾蒙治療者、曾有婦科手術者和子宮切除，都是台灣老年女性形成尿失禁的危險因子；此外，老化不僅是尿失禁的危險因子，更是某些尿失禁危險因子的危險因子，這個發現於是獲致建議:如果可能的話，預防與治療老化可能會有效降低尿失禁的危險。

關鍵字

知識態度暨行為調查 ； 下泌尿道症狀 ； 停經 ； 盛行率 ； 尿失禁的危險因子 ； 台灣婦女 ； 尿失禁

並列摘要

Background and Objectives: Urinary incontinence (UI) is a common condition with significant medical, social, and psychological problems in older women. It is so common in older adults that it is often viewed mistakenly as a natural process of aging. The increased prevalence of UI in late life reflects functional deterioration and coincidental disability. This represents an older population that has a greater burden of UI and as such their concerns may differ from a younger population, which makes them a unique population to investigate. The prevalence of UI among Taiwanese women aged between 20 and 59 years is 18.7%. It showed that age was risk factor for UI among Taiwanese women of that age group. The other report also showed that the prevalence of UI in Taiwanese women aged 20-59 increased significantly with marriage, alcohol intake, drug allergy, diabetes mellitus (DM), hypertension, body mass index (BMI) and a previous gynecological operation. The data of the prevalence and risk factors of UI in elderly Taiwanese women aged was not available. This study is to evaluate the prevalence and attitudes toward UI and to investigate the risk factors in the development of UI in Taiwanese women aged 60 or over.
Methods: The study was part of the 1999 “Survey of Health and Living Status of the Middle Aged and Elderly in Taiwan” which has been conducted by the National Institute of Family Planning of the Department of Health, The Executive Yuan, ROC, since 1989. In coordination with the National Institute of Family Planning of the Department of Health, supervised by Population Studies Center, University of Michigan, U.S.A, the sample was selected by using a multistage random sampling design. Three-stage random sampling was carried out to select a stratified probability sample. The three-stage selection process is described as below: (1) Primary sampling units (PSU): The first stage sample was drawn from the 331 administrative units of Taiwan, excluding the mountainous areas with largely indigenous populations, and stratified by administrative level (districts of large cities and entire smaller cities, towns and townships), three levels of education and three levels of total fertility rate into 27 strata. Selection was made by systematic random sampling with the interval of selection equal to the size of the smallest sample. This procedure resulted in the selection of 56 PSUs out of the 331 administrative units in Taiwan. (2) Blocks (lins): A fixed proportion of 1/2 the city district, town and township sample was used to select a certain number of lins according to the desired sample size. (3) Cases: Two eligible respondents were selected by systematic random sampling from each selected block (lin). The contents of the survey questionnaire are mainly based on the questions and contents of the first, second and third Survey of the Health and Living Status of the Elderly in Taiwan. In addition, the National Institute of Family Planning referred to the questionnaire design of related research conducted in the U.S. and Japan. The questionnaire, which had been devised for the evaluation of the prevalence of UI and the risk factors related to UI, including general background, past history, obstetric and gynecologic history, UI, and knowledge, attitudes, and practices of lower urinary tract symptoms were incorporated in the survey. It has been modified from “URG-data base system. Urodynamics and gynecologic urology. HISTORY.” To increase the measurement accuracy of the survey questionnaire, before the formal implementation of the survey, staff of the National Institute of Family Planning conducted three trial surveys using a preliminary draft of the questionnaire. The final questionnaire was completed following numerous discussions and edits based on the opinions provided by domestic and foreign experts and scholars. The project employed 86 interviewers who are specially designated interviewers of the National Institute of Family Planning of the Department of Health. Two short-term interviewer training sessions of four days each were carried out. Most of the interviewers completed one of these two sessions. This survey was conducted via face-to-face interviews to gather data. The interviewee was the selected sample case, except when the subject was too mentally impaired, too physically ill or too hard of hearing and/or mute to be interviewed directly. An analysis of individual items was performed and based only on the number of subjects who answered those particular questions. All data were entered into a computer database and analyzed using SAS software (SAS Institute, Cary, NC). The χ2 test was used to test for differences in the percentage of UI among participants with different dichotomous explanatory variables. A Fisher’s Exact test was used to test for differences in the frequency of UI among participants with different dichotomous explanatory variables. Simple logistic regression models were used as univariate analyses to evaluate the association between UI and each concerned variable. Multivariable logistic regression analyses were performed to assess the interaction of independent or dependent variables on the risk for UI. To adjust for possible confounders, we dropped out the non-significant variables step-by-step from the multivariable model. Risk was estimated using odds ratio (OR) with 95% confidence intervals (CI). A P value less than 0.05 was regarded as statistically significant.
Results: A sample of 655 women aged betweens 60 and 69 years and 1,755 women aged 70 years or older were taken by multistage random sampling design. A total of 621 women had passed away, 590 from the older group selected in 1989, and 31 from the 60-69 years group. The numbers of women who refused the interviews, could not be found, or did not answer the questionnaire for the two groups are 181 and 91, respectively. The overall response rate was 84.8% (1,517/1,789). The overall proportion of those with UI was 29.8% (485/1,517, SE 0.012, 95% CI 0.274-0.322). A total of ninety-three subjectively incontinent women were excluded from the analysis because, although they reported experiencing intermittent incontinence, they did not know when it occurred. 69.9% (250/367) of urinary incontinence women had incontinence less than once a month. Regarding knowledge, attitudes, and practices to urinary incontinence, 54.3% (822/1,514) women knew the meaning of urinary incontinence, Most women (1,281/1,514, 84.61%) said they would tell a doctor if they had UI, while 89.21% (1,348/1,511) would seek treatment if they had UI, and 30.3% (147/485) of women with urinary incontinence had seen a physician for help. The main reason for not seeking medical help was shyness. Factors and their prevalence associated with UI were age (odds ratio=1.04 per year, 95% CI 1.02-1.05), diabetes mellitus (39.8%, P=0.002), hypertension (39.5%, P=0.001), abdominal gynecological surgery (41.4%, P=0.001), hysterectomy (42.4%, P=0.003), history of drug allergy (41.3%, P=0.001), smoking (45.5%, P= 0.010), hormone replacement therapy (41.5%, P=0.026) and high body mass index (OR=1.05 per unit, 95% CI 1.02-1.09). Parity, the routes of delivery, alcohol consumption and marriage did not increase the risk of UI.
Conclusions: The prevalence of UI of Taiwanese women aged 60 years and older was similar to those in Western societies. But few women with UI seek treatment, primarily because of either embarrassment or an assumption that it is a natural part of life, particularly after delivery. Women with this disorder are therefore undertreated. It is apparent that better health education is needed to improve understanding of this disorder and increase awareness of the treatment that is available. The results of this study show relative high risk of UI development in the elderly Taiwanese women with diabetes mellitus, hypertension, and drug allergy history, and high BMI, and those smoking, on hormone replacement therapy and after abdominally gynecological operation or hysterectomy. Besides, aging is an important risk factor of both UI and some risk factors of UI. The finding suggests that prevention and treatment of aging if it is possible might be effective in reducing the risk of UI.

並列關鍵字

KAP (knowledge, attitudes and practices) survey ； Lower urinary tract symptoms ； Menopause ； Prevalence ； Risk factors of urinary incontinence ； Taiwanese women ； Urinary incontinence

參考文獻

1. Samuelsson E, Victor A, Tibblin G: A population study of urinary incontinence and nocturia among women aged 20-59 years. Prevalence, well-being and wish for treatment. Acta Obstet Gynecol Scand 1997; 76: 74-80.

3. Jolleys JV: Reported prevalence of urinary incontinence in women in a general practice. Br Med J Clin Res Ed 1988; 296:1300-1302.

4. National Institutes of Health Consensus Development Conference Statement. Natl Inst Health Consensus Dev Conf Consensus Statement 1988; 7:1-11.

5. Hsieh CH, Su TH, Chang ST. Prevalence of and attitude toward urinary incontinence in Taiwanese women. Int J Gynecol Obstet 2005; 88:152-153.

6. Hsieh CH, Lee MS, Lee MC, Kuo TC, Hsu CS, Chang ST. Risk factors for urinary incontinence in Taiwanese women aged 20 to 59 years. Taiwanese J Obstet Gynecol 2008; 47:197-202.