背景:睡眠問題被證實會影響健康結果，並與生活品質相關。尤其，老人睡眠型態有別於其他族群，其睡眠障礙常多重因子及共病相關，服用安眠藥是常用治療方式。然而，過去研究甚少同時關注老人睡眠時間、睡眠品質與安眠藥的使用是否影響死亡風險且未考慮相關變項隨時間的變化對結果的影響。故本研究目的以時間相依共變項存活分析與探討老人睡眠時間、睡眠品質與安眠藥使用對死亡風險的影響。
方法:本研究使用1993-2007年台灣地區中老年長期追蹤調查資料庫，針對1993年65歲以上老人共追蹤5波的調查，並連結2007年12月31日前國家死亡登記檔，以時間相依共變項存活分析，控制社會人口、健康和疾病狀態變動後，探討睡眠時間、睡眠品質與安眠藥使用對14年死亡風險比之影響。
結果:有效樣本共2488名老人(1391名男性、1097名女性)，平均追蹤11.81±3.78年。整體樣本控制社會人口與健康變項後，睡眠時間過長者(≧9)的死亡風險較睡眠時間正常(7-8小時)者平均高38%，其中女性高達62%，男性則為25%。安眠藥使用僅在男性達顯著增加死亡風險，亦即在控制社會人口變項後，男性使用安眠藥者相較於未用藥者死亡風險高39%，但加入健康相關變項後顯著性消失。女性在控制社會人口與健康相關變項後，僅睡眠時間過長顯著增加死亡風險62%，但睡眠時間過短與睡眠品質不好反而減少死亡風險。進一步分析發現，使用安眠藥的比例隨年代快速增加，從1993年8%至2007年16.41%，1993年使用安眠藥者的特性為:女性、睡眠品質不好、自覺健康差、多種慢性病、憂鬱、及ADL有障礙者；然而至2007年僅剩下女性及多重慢性病達顯著差異，睡眠時間與睡眠品質兩項因素未達顯著差異。
結論: 隨時間演變與年齡增長，老人使用安眠藥的比例快速成長。睡眠時間、睡眠品質與使用安眠藥對死亡風險之影響有性別差異。男性睡眠時間過長、睡眠品質不好與使用安眠藥都顯著增加死亡風險，但控制健康相關變項後，安眠藥與睡眠品質不好的顯著性消失，僅睡眠時間過長仍維持顯著。女性僅睡眠時間過長顯著增加死亡風險。

The sleep patterns in older adults were very different from younger groups. The purpose of this study was to analyze the hazard ratio of sleep duration, sleep quality, and the use of hypnotic drugs among older adults in Taiwan. The data of Taiwan longitudinal survey of aging (TLSA) from 1993 to 2007 was used and the death certificates were linked to the national death registry in Taiwan. We used time varying survival analysis to conduct the Cox proportional hazard models stratified by sex to calculate the risk of mortality and hazard ratios after adjust the covariates in raw data. Among 2488 participants, the average of follow-up years was 11.81±3.78 years. This study found that the use of hypnotics dramatically increased from 8.28% in 1993 to 16.41% in 2007. Use of hypnotics was found to be significant only in men; however, after controlling for health related factors, the significance disappeared. Both men and women, the hazard ratio (HR) of longer sleep (≧9) was 38% greater than normal sleep hours (7-8). Gender difference was found in sleep behavior and the use of hypnotic drugs on the risk of mortality. For men, longer sleep duration, poor sleep quality and the use of hypnotics were significantly associated with lower survival rate; however, after controlling other covariates, the differences of poor sleep quality and the use of hypnotics disappeared. For women, only longer sleep duration was significantly associated with higher hazard ratio.

INTRODUCTION

The world population is rapidly ageing. Sleep is a daily essential physiological phenomenon for human beings to maintain normal physiological and psychological functions. However, aging is often accompanied by sleep disturbances.

Studies have indicated that the sleep patterns in older adults were very different from younger groups. According to the estimation, more than half of the elderly indicated that they suffer from sleep disturbances. Sleep disturbances in the elderly are caused by multiple factors, and are related to comorbidities and drugs. Sleeping pills are the most rapid approach for improving sleep disturbance in the elderly.

Past studies on sleeping pills are mainly cross-sectional review studies. For example, 5 literature reviews on the prevalence of use of sleeping pills in various countries, including Finland, Sweden, Australia, the U.S., etc. found that the prevalence was approximately 4.7-31.3%.Longer sleep period and poor sleep quality in elderly were associated with higher risk of mortality. Previous research showed that, however, the survival rates of using hypnotics and sedative drugs were not consistent. To date, there is a lack of study using adjusting time varying covariates survival analysis to analyze and investigate the influence of three factors, sleep duration, sleep quality, and use of sleeping pills, on risk of death.

The purpose of this study is to take Taiwan as an example, use representative national samples from the database of Taiwan Longitudinal Study on Aging (TLSA), and take into account different social and demographic factors and health-related factors to perform the follow-up on changes in 14 years, as well as to use time varying covariates survival analysis to investigate the influence of sleep duration, sleep quality, and use of sleeping pills in the elderly on their risk of death.

MATERIALS AND METHODS

This study used the database of “Taiwan Longitudinal Study on Aging (TLSA)” to select the male and female elderly people over the age of 60 whose nationality is Taiwan since late 1988 as the research population, and used a 3-stage stratified random sampling to perform a long-term follow-up at an interval of every 3 to 4 years. To date, a total of 6 waves of survey data were collected in 1989, 1993, 1996, 1999, 2003, and 2007, respectively.Because the measurement of sleep duration in 1993 was more complete, this study used the 5 waves of survey data from 1993 to 2007.

The inclusion criteria of this study are: respondents over the age of 65 since 1993 and who completed the survey on their own and provided complete research variable data. The exclusion criteria are: (1) the elderly who were under the age of 65 (N=150) in 1993; (2) the elderly who did not complete the survey on their own in 1993 (N=173); (3) the elderly who did not provide complete information on sleep duration, sleep quality, and use of sleeping pills in 1993 (N=36); (4) the elderly who passed away before 1996 (N=308) [to reduce the deviation caused by death of other reasons or diseases]. This study performed a total of 4 follow-ups in 1996, 1999, 2003, and 2007, respectively, investigated the changes in socio-demographic variables factors and health-related factors, and probed into whether sleep duration, sleep quality, and use of sleeping pills affected the respondents’ survival in 2007.

The research data were the 5 waves of survey data of the 14-year follow-up on 2,488 elderly over the age of 65 from 1993 to 2007. In addition, this study also referred to the National Death Registration to calculate the number of years of survival of every respondent during the study. Because repeatedly measured data from the same one respondent would be collected, there were still data dependencies. It was necessary to take into account the correlation among various data to perform statistical analyses. Therefore, this study used the statistical model where the repeated measurements of multiple data can be controlled – Time Varying Survival Analysis, and referred to Your Survival Guide to Using Time Dependent Covariates(Teresa, Powell, & Melissa, 2012) to construct the Cox proportional risk model according to the time dependent covariates. After all of the confounding factors which might have an influence were taken into account, this study analyzed whether sleep duration, sleep quality, and use of sleeping pills had a statistically significant effect on 14-year survival. Because there was a difference in risk of death of use of sleeping pills between male and female respondents, this study re-ran the 5 models according to gender difference to calculate the hazard ratio for death of various variables in male and female respondents. This study used SAS 9.3 statistical software to arrange and analyze data. All statistics were two-tailed, P<0.05, reaching a statistically significant difference.

RESULTS AND DISCUSSION

Among the 2,488 elderly people enrolled in this study, 1,391 of them were male, and 1,097 of them were female. The average follow-up period was 11.81±3.78 years. The age range of samples in 1993 was 65-93, and the average age was 71.17 years old. The proportion of unemployed elderly people whose educational background was lower and who were not partnered increased year by year. Approximately 25-33% of the elderly slept more than 9 hours per day. With the increase of age, the proportion of elderly people who perceived poor sleep quality increased from 28.6% to 35.2%. For the use of sleeping pills, the proportion of the elderly who used sleeping pills increased by 2 times from 8.3% in 1993 to 16.4% in 2007.

Both men and women, the hazard ratio (HR) of longer sleep (≧9) was 38% greater than normal sleep hours (7-8). The HR of longer sleep in males was 25% greater than normal sleep hours (7-8).Use of hypnotics was found to be significant only in men; however, after controlling for health related factors, the significance disappeared. The hazard of longer sleepers in females were surprising increased 62%, after controlling socio-demographic and health related factors. In addition, this study found that the use of hypnotics dramatically increased from 8.28% in 1993 to 16.41% in 2007. We found the characters of hypnotic users were the same between the begin and the end of this study such as female, poor sleep quality, and the greater number of diseases.

CONCLUSION

More and more participants used hypnotic drugs when the time passed and were getting older. Gender difference was found in sleep behavior and the use of hypnotic drugs on the risk of mortality. For men, longer sleep duration, poor sleep quality and the use of hypnotics were significantly associated with greater survival rate; however, after controlling other covariates, the differences of poor sleep quality and the use of hypnotics disappeared.The association between longer sleep and greater HR was still significance. For women, only longer sleep duration was significantly associated with higher hazard ratio.

Andresen, E. M., Malmgren, J. A., Carter, W. B., & Patrick, D. L. (1994). Screening for depression in well older adults: Evaluation of a short form of the CES-D. American journal of preventive medicine.
Baker, M., Manheimer, D., Mellinger, G., & Uhlenhuth, E. (1984). A cross-national comparison of anti-anxiety/sedative drug use. Current Medical Research and Opinion, 8(s4), 5-20.
Baldwin, D. S., Aitchison, K., Bateson, A., Curran, H. V., Davies, S., Leonard, B., . . . Wilson, S. (2013). Benzodiazepines: risks and benefits. A reconsideration. J Psychopharmacol, 27(11), 967-971. doi: 10.1177/0269881113503509
Cappuccio, F., D'Elia, L., Strazzullo, P., & Miller, M. A. (2010). Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep, 33(5), 585-592.
Castro-Costa, E., Dewey, M. E., Ferri, C. P., Uchoa, E., Firmo, J. O., Rocha, F. L., . . . Stewart, R. (2011). Association between sleep duration and all-cause mortality in old age: 9-year follow-up of the Bambui Cohort Study, Brazil. J Sleep Res, 20(2), 303-310. doi: 10.1111/j.1365-2869.2010.00884.x
Cohen-Mansfield, J., & Perach, R. (2012). Sleep duration, nap habits, and mortality in older persons. Sleep, 35(7), 1003-1009. doi: 10.5665/sleep.1970
Crowley, K. (2011). Sleep and sleep disorders in older adults. Neuropsychology review, 21(1), 41-53.
Desplenter, F., Caenen, C., Meelberghs, J., Hartikainen, S., Sulkava, R., & Bell, J. S. (2011). Change in psychotropic drug use among community-dwelling people aged 75 years and older in Finland: repeated cross-sectional population studies. Int Psychogeriatr, 23(8), 1278-1284. doi: 10.1017/S1041610211000718
Ensrud, K. E., Blackwell, T. L., Ancoli-Israel, S., Redline, S., Cawthon, P. M., Paudel, M. L., . . . Stone, K. L. (2012). Sleep disturbances and risk of frailty and mortality in older men. Sleep Med, 13(10), 1217-1225. doi: 10.1016/j.sleep.2012.04.010
Gangwisch, J. E., Heymsfield, S. B., Boden-Albala, B., Buijs, R. M., Kreier, F., Opler, M. G., . . . Malaspina, D. (2008). Sleep duration associated with mortality in elderly, but not middle-aged, adults in a large US sample. Sleep, 31(8), 1087.
Garde, A. H., Hansen, A. M., Holtermann, A., Gyntelberg, F., & Suadicani, P. (2013). Sleep duration and ischemic heart disease and all-cause mortality: prospective cohort study on effects of tranquilizers/hypnotics and perceived stress. Scand J Work Environ Health, 39(6), 550-558. doi: 10.5271/sjweh.3372
Giron, M. S. T., Forsell, Y., Bernsten, C., Thorslund, M., Winblad, B., & Fastbom, J. (2002). Sleep Problems in a Very Old Population Drug Use and Clinical Correlates. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 57(4), M236-M240.
Habte-Gabr, E., Wallace, R. B., Colsher, P. L., Hulbert, J. R., White, L. R., & Smith, I. M. (1991). Sleep patterns in rural elders: demographic, health, and psychobehavioral correlates. Journal of Clinical Epidemiology, 44(1), 5-13.
Hamer, M., Batty, G. D., & Kivimaki, M. (2012). Sleep loss due to worry and future risk of cardiovascular disease and all-cause mortality: the Scottish Health Survey. Eur J Prev Cardiol, 19(6), 1437-1443. doi: 10.1177/1741826711426092
Harris, M. G., Burgess, P. M., Pirkis, J., Siskind, D., Slade, T., & Whiteford, H. A. (2011). Correlates of antidepressant and anxiolytic, hypnotic or sedative medication use in an Australian community sample. Aust N Z J Psychiatry, 45(3), 249-260. doi: 10.3109/00048674.2010.531459
Hausken, A. M., Skurtveit, S., & Tverdal, A. (2007). Use of anxiolytic or hypnotic drugs and total mortality in a general middle-aged population. Pharmacoepidemiol Drug Saf, 16(8), 913-918. doi: 10.1002/pds.1417
Hublin, C., Partinen, M., Koskenvuo, M., & Kaprio, J. (2007). Sleep and mortality: a population-based 22-year follow-up study. Sleep, 30(10), 1245.
Ikehara, S., Iso, H., Date, C., Kikuchi, S., Watanabe, Y., Wada, Y., . . . Group, J. S. (2009). Association of sleep duration with mortality from cardiovascular disease and other causes for Japanese men and women: the JACC study. Sleep, 32(3), 295.
Jaussent, I., Ancelin, M. L., Berr, C., Pérès, K., Scali, J., Besset, A., . . . Dauvilliers, Y. (2013). Hypnotics and mortality in an elderly general population: a 12-year prospective study. BMC medicine, 11(1), 212-223.
John, U., Baumeister, S. E., Volzke, H., Meyer, C., Ulbricht, S., & Alte, D. (2007). Sedative, hypnotic, anxiolytic and opioid medicament use and its co-occurrence with tobacco smoking and alcohol risk drinking in a community sample. BMC Public Health, 7, 337. doi: 10.1186/1471-2458-7-337
Johnell, K., & Fastbom, J. (2011). Gender and use of hypnotics or sedatives in old age: a nationwide register-based study. Int J Clin Pharm, 33(5), 788-793. doi: 10.1007/s11096-011-9536-8
Jung, K. I., Song, C. H., Ancoli-Israel, S., & Barrett-Connor, E. (2013). Gender differences in nighttime sleep and daytime napping as predictors of mortality in older adults: the Rancho Bernardo study. Sleep Med, 14(1), 12-19. doi: 10.1016/j.sleep.2012.06.004
Kohout, F. J., Berkman, L. F., Evans, D. A., & Cornoni-Huntley, J. (1993). Two shorter forms of the CES-D depression symptoms index. Journal of Aging and Health, 5(2), 179-193.
Kojima, M., Wakai, K., Kawamura, T., Tamakoshi, A., Aoki, R., Lin, Y., . . . Ohno, Y. (2000). Sleep patterns and total mortality: a 12-year follow-up study in Japan. Journal of Epidemiology/Japan Epidemiological Association, 10(2), 87-93.
Kripke, D. F., Garfinkel, L., Wingard, D. L., Klauber, M. R., & Marler, M. R. (2002). Mortality associated with sleep duration and insomnia. Archives of general psychiatry, 59(2), 131-136.
Kripke, D. F., Klauber, M. R., Wingard, D. L., Fell, R. L., Assmus, J. D., & Garfinkel, L. (1998). Mortality hazard associated with prescription hypnotics. Biological psychiatry, 43(9), 687-693.
Kripke, D. F., Langer, R. D., Elliott, J. A., Klauber, M. R., & Rex, K. M. (2011). Mortality related to actigraphic long and short sleep. Sleep Med, 12(1), 28-33. doi: 10.1016/j.sleep.2010.04.016
Kripke, D. F., Langer, R. D., & Kline, L. E. (2012). Hypnotics' association with mortality or cancer: a matched cohort study. BMJ Open, 2(1), e000850. doi: 10.1136/bmjopen-2012-000850
Kripke, D. F., Simons, R. N., Garfinkel, L., & Hammond, E. C. (1979). Short and long sleep and sleeping pills: is increased mortality associated? Archives of general psychiatry, 36(1), 103-116.
Kronholm, E., Laatikainen, T., Peltonen, M., Sippola, R., & Partonen, T. (2011). Self-reported sleep duration, all-cause mortality, cardiovascular mortality and morbidity in Finland. Sleep Med, 12(3), 215-221. doi: 10.1016/j.sleep.2010.07.021
Lan, T.-Y., Lan, T.-H., Wen, C.-P., Lin, Y.-H., & Chuang, Y.-L. (2007). Nighttime sleep, Chinese afternoon nap, and mortality in the elderly. Sleep, 30(9), 1105.
Li, Y., Sato, Y., & Yamaguchi, N. (2013). Potential biochemical pathways for the relationship between sleep duration and mortality. Sleep Med, 14(1), 98-104. doi: 10.1016/j.sleep.2012.08.020
Mallon, L., Broman, J. E., & Hetta, J. (2009). Is usage of hypnotics associated with mortality? Sleep Med, 10(3), 279-286. doi: 10.1016/j.sleep.2008.12.004
Patel, S. R., Ayas, N. T., Malhotra, M. R., White, D. P., Schernhammer, E. S., Speizer, F. E., . . . Hu, F. B. (2004). A prospective study of sleep duration and mortality risk in women. SLEEP-NEW YORK THEN WESTCHESTER-, 27(3), 440-444.
Phillips, B., & Mannino, D. M. (2005). Does insomnia kill? SLEEP-NEW YORK THEN WESTCHESTER-, 28(8), 965.
Qiu, L., Sautter, J., Liu, Y., & Gu, D. (2011). Age and gender differences in linkages of sleep with subsequent mortality and health among very old Chinese. Sleep Med, 12(10), 1008-1017. doi: 10.1016/j.sleep.2011.04.014
Shankar, A., Koh, W. P., Yuan, J. M., Lee, H. P., & Yu, M. C. (2008). Sleep duration and coronary heart disease mortality among Chinese adults in Singapore: a population-based cohort study. Am J Epidemiol, 168(12), 1367-1373. doi: 10.1093/aje/kwn281
Suzuki, E., Yorifuji, T., Ueshima, K., Takao, S., Sugiyama, M., Ohta, T., . . . Doi, H. (2009). Sleep duration, sleep quality and cardiovascular disease mortality among the elderly: a population-based cohort study. Prev Med, 49(2-3), 135-141. doi: 10.1016/j.ypmed.2009.06.016
Tamakoshi, A., & Ohno, Y. (2004). Self-reported sleep duration as a predictor of all-cause mortality: results from the JACC study, Japan. SLEEP-NEW YORK THEN WESTCHESTER-, 27(1), 51-54.
Teresa, M., Powell, M. S., & Melissa, E. (2012). Your Survival Guide to Using Time Dependent Covariates. SAS Global Forum 2012, 168-178.
Vgontzas, A. N., Liao, D., Pejovic, S., Calhoun, S., Karataraki, M., Basta, M., . . . Bixler, E. O. (2010a). Insomnia with short sleep duration and mortality: the Penn State cohort. Sleep, 33(9), 1159-1164.
Vgontzas, A. N., Liao, D., Pejovic, S., Calhoun, S., Karataraki, M., Basta, M., . . . Bixler, E. O. (2010b). Insomnia with short sleep duration and mortality: the Penn State cohort. Sleep, 33(9), 1159.
Wang, J.-P. (2013). Excessive Use of Sedatives/Hypnotics/Anxiolytics in Taiwan. Taiwanese Journal of Psychiatry (Taipei), 27(2), 95-98.
Webb, W., & Agnew, H. (1970). Sleep stage characteristics of long and short sleepers. Science, 168(3927), 146-147.
WHO, W. H. O. (2012). Ageing and life course: Interesting facts about ageing.
吳京一, & 童麗珠. (2006). 睡眠你知多少事 (3)-每晚你睡幾小時? 科學教育(286), 40-50.
吳易謙, 韓良誠, & 林錫璋. (2008). 長壽與成功老化. 內科學誌, 19(5), 394-400.
吳曼阡, & 宋惠娟. (2010). 光線治療在老人照護的運用. 志為護理-慈濟護理雜誌, 9(3), 63-70.
李月. (2013). 睡眠新知. 百科知识(8), 36-38.
李志宏, & 施肇榮. (2013). 安眠藥品的是與非 (下)─ 管制藥品的管理實務. 臺灣醫界, 56(8), 26-31.
李庚霖, 區雅倫, 陳淑惠, & 翁儷禎. (2009). “臺灣地區中老年身心社會生活狀況長期追蹤調查” 短版 CES-D 量表之心理計量特性. 中華心理衛生學刊, 22(4), 383-410.
李政育, & 江漢光. (2004). 中西醫會診-失眠: 書泉出版社.
良介. (2011). 世界睡眠日. 实用糖尿病杂志, 7(3), 64-64.
孟欣宏, 吳春樺, & 林文德. (2013). 退休對心理健康之影響――以鎮定安眠藥之使用為指標.
林建甫. (2008). 存活分析: 雙葉書廊.
張文蓓, & 林佳靜. (2010). 失眠之概念分析. 新臺北護理期刊, 12(1), 63-70.
郭淑珍, 陳怡君, & 楊雪華. (2012). 醫師的安眠藥物處方行為之初探: 一項質性研究. 臺灣公共衛生雜誌, 31(6), 556-569.
陳崇賢, 陳昭源, & 林忠順. (2010). 失眠之評估與治療. 家庭醫學與基層醫療, 25(3), 92-99.
陳惠真, & 王香蘋. (2005). 台灣老人健康狀況初探: 1989 年至 1999 年. 臺灣家庭醫學雜誌, 15(1), 25-35.
黃玉如, 張意宜, & 蘇渝評. (2012). 某教學醫院抗焦慮作用 BZDs 之處方合理性分析. 臺灣臨床藥學雜誌, 20(2), 103-113.
溫劭宜. (2007). 台灣失能老人安眠藥使用情形與相關因素探討. 臺灣大學衛生政策與管理研究所學位論文, 1-81.
經建會. (2010). 年至 2060 年臺灣人口推計. 台北: 行政院經濟建設委員會人力規劃處.
趙瑩, 方喬玲, 李敏慈, 樓亞洲, & 陳立奇. (2011). 失眠患者衛教介入之成效. 北市醫學雜誌, 8(3), 33-47.
劉惠瑚, 陳玉敏, & 李月萍. (2005). 老人睡眠問題之探討. 長庚護理, 16(4), 408-413.
蔡崇煌, 陳宇嘉, 王雪鳳, 劉金明, 吳萬慶, & 林高德. (2000). 老人的睡眠障礙及其相關因子之探討-區域醫院之經驗. 臺灣家庭醫學雜誌, 10(3), 119-128.
鄭淑文. (2012). 精神專科醫院門診鎮靜安眠藥處方之探討. 臺灣臨床藥學雜誌, 20(3), 181-194.
賴奕菁. (2004). 門診老年失眠症患者之安眠藥處方型態及處方問題. 撰者.
謝昌成, & 蕭雅尤. (2013). 老年人的失眠. 家庭醫學與基層醫療, 28(7), 180-185.