- Email: [email protected]
insomnia and their effects on daytime functioning
Sleep Medicine 13 (2012) 1115–1121
Contents lists available at SciVerse ScienceDirect
Sleep Medicine journal homepage: www.elsevier.com/locate/sleep
Original Article
A two-year follow-up study on the symptoms of sleep disturbances/insomnia and their effects on daytime functioning Yoko Komada a,b, Takashi Nomura c, Masayoshi Kusumi c, Kenji Nakashima c, Isa Okajima a,b, Taeko Sasai a,b, Yuichi Inoue a,b,⇑ a b c
Department of Somnology, Tokyo Medical University, Japan Japan Somnology Center, Neuropsychiatric Research Institute, Japan Department of Neurology, Institute of the Neurological Sciences, Tottori University Faculty of Medicine, Japan
a r t i c l e
i n f o
Article history: Received 19 December 2011 Received in revised form 22 May 2012 Accepted 23 May 2012 Available online 25 July 2012 Keywords: Quality of life Chronic insomnia Longitudinal study SF-8 Natural course Mental quality of life Physical quality of life
a b s t r a c t Objective: This study attempts to identify changes in the symptoms of sleep disturbances/insomnia over a two-year course and their effects on daytime functioning. Methods: We administered two population-based epidemiological surveys in 2005 and 2007 to participants from rural Japan. Results: In the first survey, 30.7% of the subjects reported sleep disturbances/insomnia. Among them, 60.9% reported sleep problems at the two-year follow-up. A comparison of sleep disturbances/insomnia, and subjective daytime functioning measures between the new incident cases and persistent poor sleepers revealed that the total score of persistent poor sleepers was significantly lower than that of new incident cases on the Pittsburgh Sleep Quality Index and physical quality of life (QoL) but not mental QoL. Longitudinal comparisons of the symptoms of sleep disturbances/insomnia in persistent poor sleepers revealed that sleep efficiency was significantly worse at follow-up. Exacerbation of the symptoms of sleep disturbances/insomnia at follow-up was observed in mild but not severe cases. Conclusions: Sleep efficiency progressively worsens over time, and physical QoL can deteriorate as sleep disturbances/insomnia become chronic. Since the symptoms of sleep disturbances/insomnia and their daytime effects are exacerbated even in mild cases, early intervention and treatment are necessary. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Insomnia is a common disorder [1,2], with an estimated prevalence of about 20% among the general population [3,4]. Poor nocturnal sleep and consequent impairment in daytime functioning are the core symptoms of insomnia [5]. Indeed, both daytime impairment and night-time sleep difficulties have been established as essential items in the diagnostic criteria of insomnia by the International Classification of Sleep Disorders [6]. Several reports have described the natural course of insomnia and have revealed that a considerable number of people with insomnia exhibit a chronic course [7,8]. However, specific changes in the severity and the symptoms of the disorder as it becomes chronic have yet to be ascertained.
⇑ Corresponding author at: Department of Somnology, Tokyo Medical University, 6-7-1 Nishishinjuku, Shinjuku-ku, Tokyo 160-0023, Japan. Tel.: +81 0 3 3342 6111; fax: +81 0 3 3342 7083. E-mail address: [email protected] (Y. Inoue). 1389-9457/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.sleep.2012.05.015
Previous studies on clinical populations have reported that patients with chronic insomnia commonly complain of subjective daytime impairments, including mood disturbances, concentration problems, easy fatigue, and sleepiness [5,9]. Objective measures revealed that insomnia patients show impairment in tasks that evaluate vigilance, working memory, and motor control [10,11]. These daytime dysfunctions attributed to insomnia are assumed to negatively affect sufferers’ quality of life (QoL), which is a measure of general daytime functioning [12,13]. We previously reported that insomnia was generally associated with depressed mood and low QoL scores in both the mental and physical component among participants from rural Japan [14,15]. However, the precise impact of the chronicity of insomnia on QoL remains to be determined. In order to investigate these issues, we performed a longitudinal study where a two-part questionnaire was administered at the start and end of a two-year interval to a cohort taking part in sleep studies in a single rural community [14,16,17]. Through this study, we hope to (1) elucidate the changes in insomnia symptoms over a two-year course and (2) determine the effects of chronic insomnia on daytime functioning.
1116
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
2. Methods 2.1. Participants and procedures This survey was conducted as part of the sleep studies mentioned above in a rural town (Daisen, Tottori Prefecture, Western Japan) [14,16,17]. The Ethics Committee of Tottori University approved this study, and all participants provided written informed consent. In 2004 the total population of the town was 6643, with 5528 residents aged 20 years and above (2521 men, 3007 women), who had a mean age of 55.2 years. The first part of the questionnaire survey was administered between November 2005 and January 2006 (baseline), and the second part was administered between November 2007 and December 2007 (follow-up). With the cooperation of local public health nurses, questionnaires tagged with serial numbers were delivered to all residents aged 20 years and older during both periods. In the current study we did not distinguish nursing home respondents from hospital respondents. Responses to the questionnaire were received from 2822 people at baseline (response rate: 51%; 1222 men, 1600 women; mean age = 57.4, SD = 17.7). Two years later, the follow-up questionnaires were sent to the people who had responded at baseline, with 1577 of them responding to the follow-up questionnaire (response rate: 56%; 683 men, 894 women; mean age = 58.6, SD = 16.1) (Fig. 1). The responses to the two surveys were matched using the serial numbers. 2.2. Measures The contents of the questionnaires at both baseline and followup were as follows:
quality (category 1 [C1]), sleep latency (C2), sleep duration (C3), habitual sleep efficiency (C4), sleep disturbance (C5), use of sleeping medication (C6), and daytime dysfunction (C7). Although the PSQI evaluates sleep disturbances rather than insomnia, the cut-off score for insomnia has been established at 5.5 points according to a previous study [18], and we have followed that convention. Consequently, respondents with PSQI scores of 5.5 or higher were classified as people with sleep disturbances/insomnia in this study. (3) The standardized eight-item Short Form Health Survey of the Medical Outcomes Study (SF-8) [19] was used to assess QoL. The SF-8 measures vitality, social functioning, mental health condition, emotional state, general health, physical functioning, physical state, and bodily pain. The mental component summary (MCS) scale of the SF-8 was used to evaluate mental QoL and the physical component summary (PCS) scale was used to evaluate physical QoL. The average scores for both scales for the general population were set at 50 points. There were overlapping questions regarding daytime dysfunction (C7) on the PSQI and SF-8. However, the PSQI measured daytime dysfunction with regard to sleep disturbances/insomnia, whereas the SF-8 does not make this distinction. Therefore, we analyzed the scores from the PSQI and SF-8 independently of each other. (4) The 12-item version of the Center for Epidemiological Studies Depression Scale (CES-D) [20] was used to measure depressive symptoms. The scale has four response options: ‘‘never or rarely’’ (0), ‘‘sometimes’’ (1), ‘‘often’’ (2), and ‘‘always’’ (4). We used the total scores of CES-D as parameters of depression. 2.3. Statistical analysis
(1) Demographic variables: The participants were asked about their age, sex, the disease currently being treated (‘‘Please tell us the disease you are currently treated for.’’), family situation (‘‘Do you currently live with your family?’’), smoking habits (‘‘Do you currently smoke?’’), and alcohol consumption (‘‘Do you drink regularly?’’). (2) The Japanese version of the Pittsburgh Sleep Quality Index (PSQI) [18] was used to estimate the level of subjective sleep disturbance. The PSQI includes sub-items that evaluate sleep
Student’s paired t-tests were used to compare the PSQI, CES-D, MCS, and PCS scores between the baseline and follow-up surveys for participants who responded to both surveys. On the basis of the results of these tests, the participants were divided into four categories: good sleepers (no insomnia symptoms at both survey periods), new incident cases (no insomnia symptoms at baseline but symptoms present at follow-up), remitted cases (insomnia symptoms at baseline but not at follow-up), and
Survey at baseline Eligible subjects: n = 5,528 2,521 men (46%) 3,007 women (54%)
Non-responders: n = 2 2,706 706 Responders: n = 2 2,822 (51%) 1,222 men (43%) 1 600 women (57%) 1, Sleep disturbances/insomnia patients (PSQI ≥ 6) n = 641 (22.7%)
Non-responders: p n = 1,247 ,
Survey at follow up Responders: p 1,577 , (56%) ( ) 683 men (43%) 894 women (57%) Sleep disturbances/insomnia patients (PSQI ≥ 6) n = 476 (30.2%) Fig. 1. Survey flowchart.
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
persistent poor sleepers (insomnia symptoms present at both survey points) [21,22]. A two-way repeated-measures analysis of variance (ANOVA) (category survey period) was used to compare the scores of the PSQI, CES-D, and the MCS and PCS of the SF-8. The Bonferroni–Dunn test was used for post hoc analysis, and the chi-square test was performed to compare categorical variables. Student’s paired t-tests were used to compare the PSQI, CES-D, MCS, and PCS scores between baseline and follow-up for persistent poor sleepers to reveal any changes in symptoms of sleep disturbances/insomnia and daytime functioning among people with chronic insomnia over the two-year period. A series of logistic regression analyses were conducted for subjects with insomnia/sleep disturbances at baseline to elucidate the factors associated with their persistence. Persistent poor sleepers were further divided into two groups according to their median PSQI value at baseline (at least six but less than nine, and nine or more respectively) to determine whether the changes in the course of insomnia/sleep disturbances differed between mild and severe cases. Changes in scores of PSQI sub-items, total PSQI, CES-D, PCS, and MCS over the two-year interval were examined in these two groups by using Student’s unpaired t-tests. All statistical analyses were performed with SPSS Version 11.5 (SPSS Japan, Inc., Tokyo, Japan) with the alpha value set at 0.01.
3. Results When the demographic data of the respondents who answered only at baseline (n = 1247) and those who responded at both baseline and follow-up (n = 1577) were compared, a significant difference was observed in age (t [2394] = 3.56, p < 0.01), with participants in one group having a mean age of 55.9 (SD 19.6) and the other with a mean age of 58.6 (SD = 16.1). However, the difference in ages between the two groups was only 2.7 years. There were also significant differences in the number of participants who were currently receiving treatment for any disease (31.3% vs. 38.8%; X2 [1] = 17.4; p < 0.01), and in smoking habits (26.6% vs. 18.0%; X2 [1] = 30.0; p < 0.01). No other significant differences were found between the demographics of the two groups. In all participants, the PSQI, CES-D, and MCS scores were slightly but significantly worse at follow-up than at baseline (PSQI: t [1417] = 2.9; CES-D: t [1390] = 2.9; MCS: t [1373] = 2.7; p < 0.01 for all); there was no significant difference for PCS scores (t [1373] = 2.3; not significant [ns]). Among the participants, 56.4% were classified as good sleepers, 12.9% as new incident cases, 12.0% as remitted cases, and 18.7% as persistent poor sleepers. Table 1 shows the demographic data of the participants and the PSQI scores in each survey, and the results of the one-way ANOVA or chi-square test for each parameter of the four insomnia categories. There was no significant difference in age among the four categories (F [3, 1429] = 3.9, ns). A significant difference was observed in the number of participants with insomnia who were undergoing treatment at both baseline and follow-up between the four categories (baseline: X2 [3] = 15.6; follow-up: X2 [3] = 18.1, p < 0.01 for both). Residual analysis revealed that the number of persistent poor sleepers who were undergoing treatment was significantly higher than that of other categories at both data collection periods, and that the number of good sleepers was significantly lower than that of other categories at followup. There was also a significant difference in the number of participants using sleep medication once or more per week at both baseline and follow-up among the four categories (baseline: X2 [3] = 262.3; follow-up: X2 [3] = 243.1; p < 0.01 for both). Residual analysis revealed that remitted cases and persistent poor sleepers
1117
had significantly higher rates of sleep medication use at baseline, and that new incident cases and persistent poor sleepers used sleep medication more often than other participants belonging to categories did at follow-up. A two-way repeated measures ANOVA (category survey period) was performed to determine differences in total PSQI score. There were main effects for both category and survey period, and a significant interaction (main effect of category: F [3, 1414] = 1188.6, p < 0.01; main effect of the survey period: F [1, 1414] = 16.8, p < 0.01; and interaction between category and survey period: F [3, 1414] = 429.2, p < 0.01). The Bonferroni– Dunn post hoc test showed that the total PSQI score was significantly worse at follow-up than at baseline for new incident cases, and significantly improved at follow-up for remitted cases, with no significant differences in the scores between the two survey points for good sleepers and persistent poor sleepers. The total PSQI score for persistent poor sleepers was significantly worse than that for good sleepers, new incident cases, and remitted cases both at baseline and at follow-up (Table 1). 3.1. Comparison of insomnia symptoms in the follow-up survey between new incident cases and persistent poor sleepers PSQI scores at follow-up were compared between new incident cases and persistent poor sleepers to determine the differences in the severity of sleep disturbances/insomnia and the characteristics of the symptoms. We found that the sleep latency (C2), habitual sleep efficiency (C4), use of sleep medication (C6), and total PSQI scores of persistent poor sleepers were significantly worse than those of new incident cases (C2: t [442] = 4.9; C4: t [436] = 2.7; C6: t [442] = 3.5; PSQI total score: t [451] = 6.4, respectively, p < 0.01 for all) (Table 2). The factors associated with persistent insomnia/sleep disturbances were examined with a series of univariate logistic regression analyses, performed for the 11 independent variables: age; sex; disease currently being treated; alcohol consumption; smoking habits; living status (alone or with co-habitants); sleep medication use; and CES-D, MCS, PCS, and PSQI scores at baseline. Two of these variables (living status and PSQI score) were significantly correlated with persistent poor sleepers. The adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were 3.8 (1.1–13.2) for living alone (p < 0.01), and 1.4 (1.3–1.6) for PSQI score (p < 0.01), respectively. These two variables were then analyzed with a multivariate model, which revealed that chronic insomnia/sleep disturbances were significantly associated with a higher PSQI score at baseline (OR = 1.4, 95% CI: 1.2–1.6; p < 0.01). 3.2. Comparison of daytime consequences among categories A two-way repeated measures ANOVA (category survey period) was conducted to compare the scores of daytime consequence measures (CES-D, MCS and PCS) at both survey periods. There were main effects of both category and survey period, and an interaction was found for CES-D (main effect of category: F [3, 1387] = 119.6, p < 0.01; main effect of survey period: F [1, 1387] = 6.8, p < 0.01; interaction between category and survey period: F [3, 1387] = 13.8, p < 0.01). There was a main effect of category and an interaction between both variables for MCS and PCS (MCS: main effect: F [3, 1370] = 48.5, p < 0.01; interaction: F [3, 137] = 13.3, p < 0.01; PCS: main effect: F [3, 1370] = 31.9, p < 0.01; interaction: F [3, 137] = 3.9, p < 0.01). Post-hoc analysis confirmed that CES-D and MCS scores were significantly worse at follow-up than at baseline for good sleepers, that CES-D, MCS, and PCS scores were significantly worse at follow-up than at baseline for new incident cases, and that MCS was better at follow-up than at baseline for remitted
1118
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
Table 1 Demographic data and PSQI scores of participants. At baseline
Total (n = 1434)
Drinking habits, n (%) Smoking habits, n (%) Living alone, n (%) Sleep medication usea, n (%)
PSQIb total score, mean (SD)
With sleep disturbances/insomnia (n = 440)
Without sleep disturbances/ insomnia (good sleepers, n = 809)
With sleep disturbances/ insomnia (new incident cases, n = 185)
Without sleep disturbances/ insomnia (remitted cases, n = 172)
With sleep disturbances/ insomnia (persistent poor sleepers, n = 268)
p-Values of chi-square test or ANOVA and post hoc test among 4 sleep disturbances/insomnia groups
Baseline
633/801 60.2 (16.0) 557 (38.9%)
370/439 59.6 (15.9) 292 (36.1%)
72/113 59.5 (17.5) 63 (34.1%)
81/91 59.9 (15.8) 72 (41.9%)
110/158 62.8 (15.5) 130 (48.5%)
Follow up
611 (42.6%)
314 (38.8%)
86 (46.5%)
69 (40.1%)
142 (53.0%)
Baseline Follow up Baseline Follow up Baseline Follow up Baseline
540 (37.7%) 399 (27.8%) 263 (18.4%) 247 (17.2%) 59 (4.1%) 62 (4.3%) 106 (7.5%)
309 (38.2%) 227 (28.1%) 147 (18.2%) 136 (16.8%) 33 (4.1%) 31 (3.8%) 2 (0.3%)
70 (37.8%) 46 (24.9%) 34 (18.4%) 27 (14.6%) 6 (3.2%) 7 (3.8%) 1 (0.5%)
72 (41.9%) 58 (33.7%) 37 (21.5%) 39 (22.7%) 3 (1.7%) 5 (2.9%) 26 (15.1%)
89 68 45 45 17 19 77
Follow up
120 (8.5%)
6 (0.8%)
31 (17.1%)
4 (2.4%)
79 (30.0%)
Baseline Follow up
4.6 (3.0) 4.8 (3.2)
2.9 (1.5) 2.9 (1.5)
3.7 (1.3)* 7.7 (1.9)*
7.2 (1.7)* 3.8 (1.2)*
8.8 (2.6) 9.2 (2.7)
ns ns Persistent poor sleepers (+) (p < 0.01) good sleepers ( ) (p < 0.01), persistent poor sleepers (+) (p < 0.01) ns ns ns ns ns ns Good sleepers , new incident cases ( ) (p < 0.01), remitted cases, persistent poor sleepers (+) (p < 0.01) Good sleepers, remitted cases ( ) (p < 0.01), new incident cases, persistent poor sleepers (+) (p < 0.01) Insomnia subcategories, survey period, interaction: p < 0.01, respectivelyc
At follow up (2 years later)
Sex (M/F) Age, mean (SD) Disease currently treated, n (%)
Without sleep disturbances/insomnia (n = 994)
(33.2%) (25.4%) (16.8%) (16.8%) (6.3%) (7.1%) (28.7%)
SD = standard deviation, M = male, F = female, ns = not significant. a Sleep medication use of once or more per week; calculated using the PSQI subitem (C6). b PSQI = Pittsburgh Sleep Quality Index. c Results of two-way repeated measurements ANOVA and post hoc tests. * p < 0.01.
cases (p < 0.01 for all). There were no significant differences between these measures in persistent poor sleepers between baseline and follow-up. The Bonferroni–Dunn post hoc test also showed that the scores of the CES-D and PCS at follow-up in persistent poor sleepers were significantly worse than those of the other categories. In addition, MCS scores at follow-up in persistent poor sleepers were significantly worse than those of good sleepers and remitted cases. However, there was no significant difference in MCS scores at follow-up between persistent poor sleepers and new incident cases (Table 3).
Table 2 Comparison of sleep disturbances/insomnia symptoms at follow up between new incident cases and persistent poor sleepers.
C1: sleep quality C2: sleep latency C3: sleep duration C4: habitual sleep efficiency C5: sleep disturbance C6: use of sleeping medication C7: daytime dysfunction PSQI total score
New incident cases
Persistent poor sleepers
pValue
1.5 1.5 1.5 0.8
1.6 1.9 1.7 1.1
(0.6) (0.9) (0.8) (1.1)
ns <0.01 ns <0.01
1.2 (0.5) 0.5 (1.1)
1.3 (0.5) 0.9 (1.3)
ns <0.01
0.8 (0.7) 7.7 (1.9)
0.9 (0.7) 9.2 (2.7)
ns <0.01
(0.6) (0.9) (0.8) (0.9)
PSQI = Pittsburgh Sleep Quality Index, Mean (standard deviation, SD), Student’s ttest; ns = not significant.
3.3. Changes in the symptoms of sleep disturbances/insomnia and depression and quality of life between the two survey points in persistent poor sleepers The PSQI, CES-D, and QoL scores of persistent poor sleepers between baseline and follow-up were compared in order to determine changes over time in the symptoms of sleep disturbances/ insomnia and their effects on daytime functioning. The results are presented in Table 4. Habitual sleep efficiency (C4) at followup was significantly worse than that at baseline (t [255] = 3.2, p < 0.01). However, there were no significant differences in scores for the other sub-items or total PSQI between the two survey periods. There were also no significant differences in CES-D, MCS, and PCS scores. At baseline, 440 subjects reported sleep disturbances/insomnia, with 286 having a PSQI score of at least 6 but less than 9 and 154 having a PSQI score of 9 or more. The remission rate, indicated by a PSQI score of less than 6 at follow-up, was 50.3% in the former and 18.2% in the latter. The difference in remission rate between the two groups was significant (X2 [1] = 43.5; p < 0.01). In order to investigate whether or not a baseline severitydependent difference in the longitudinal course of sleep disturbances/insomnia was present, changes in PSQI, CES-D, and QoL scores were examined using the Student’s unpaired t-test in the two groups of subjects with chronic sleep disturbances/insomnia – namely, the mild insomnia group (with a PSQI score of more than 6 but less than 9) and the severe insomnia group (with a PSQI score of 9 or more at baseline). There were significant differences be-
1119
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121 Table 3 Scores of CES-D, MSC and PCS of participants. At baseline
Total (n = 1,434)
At follow up (2 years later) CES-D score, mean (SD) MCS score, mean (SD) PCS score, mean (SD)
Without sleep disturbances/insomnia (n = 994)
With sleep disturbances/insomnia (n = 440)
Without sleep disturbances/ insomnia (good sleepers, n = 809)
With sleep disturbances/insomnia (new incident cases, n = 185)
Without sleep disturbances/ insomnia (remitted cases, n = 172)
With sleep disturbances/insomnia (persistent poor sleepers, n = 268)
Results of two-way repeated measurements ANOVA and post hoc tests Insomnia subcategories, survey period, interaction: p < 0.01, respectively Insomnia subcategories, interaction: p < 0.01, respectively
Baseline Follow up
8.5 (4.8) 8.9 (4.7)
7.0 (3.9)* 7.3 (3.8)*
8.7 (4.6)* 10.8 (4.9)*
10.0 (5.1) 8.9 (4.3)
11.8 (5.2) 12.1 (5.2)
Baseline Follow up
49.9 (6.5) 49.4 (6.5)
51.4 (5.4)* 50.8 (5.7)*
49.7 (6.6)* 47.0 (7.3)*
48.3 (7.2)* 50.0 (6.2)*
46.8 (7.4) 46.8 (7.0)
Baseline Follow up
47.7 (7.0) 47.4 (7.0)
49.0 (6.3) 48.6 (6.2)
48.1 (7.0)* 46.3 (7.4)*
46.5 (7.4) 47.1 (7.1)
44.6 (7.6) 44.6 (7.8)
Insomnia subcategories, interaction: p < 0.01, respectively
SD = standard deviation, M = male, F = female, ns = not significant, CES-D = Center for Epidemiologic Studies Depression Scale, MCS = Mental Component Summary, PCS = Physical Component Summary. * p < 0.01.
tween the two groups in terms of C1, C2, C4, C5, and total PSQI scores (C1: t [262] = 3.2, C2: t [259] = 3.8, C4: t [254] = 4.0, C5: t [261] = 3.5, total PSQI score: t [2566] = 6.4; p < 0.01 for all). However, there were no significant differences in CES-D, MCS, or PCS scores between the two groups.
4. Discussion We conducted a longitudinal study over a two-year period on a rural population cohort in Japan and examined the course of sleep disturbances/insomnia symptoms and their effects on daytime functioning. The percentage of subjects with sleep disturbances/ insomnia was 30.7% at baseline and 31.6% at follow-up and the number of subjects using sleep medication corresponded to the presence or absence of symptoms of sleep disturbances/insomnia at the studied points. Among the study population, the 18.7% of the subjects reported chronic sleep disturbances/insomnia (a PSQI score of greater than or equal to 5.5 in both surveys). The majority of subjects with sleep disturbances/insomnia at baseline still experienced the symptoms of sleep disturbance at follow-up (60.9%), which is consistent with previous reports [7,8]. For instance, Katz et al. showed that even in subjects with mild insomnia, 59% had sleep problems at a two-year follow-up [8]. Previous longitudinal studies on general populations showed that 40% of the individuals with insomnia at baseline had persistent symptoms [23–25]. Another study on the development of insomnia indicated that 74% of the subjects reported having insomnia symptoms for at least one year [7]. These results indicate that insomnia has a generally persistent course. In the present study, persistent poor sleepers had the highest CES-D score among the four categories, a finding in accordance with those of previous studies [26]. A bidirectional causal relationship between depression and insomnia has been firmly established; that is, insomnia can lead to depression and vice versa [22,27,28]. Therefore, a higher CES-D score could be either a cause or a consequence of sleep disturbances/insomnia. A recent study suggested that the persistence of insomnia was associated with the female sex, lower education level, and daytime symptoms at baseline [29]. However, these associations were not found in our study. The reason for this discrepancy is not clear. However, the fact that our sample consisted of people from a rural area and engaged in agriculture is a possible explanation [15].
In this study, the persistence of sleep disturbances/insomnia was associated with a higher PSQI score at baseline. This result suggests that severe insomnia runs a chronic course. Moreover, the sub-item scores of sleep latency, habitual sleep efficiency, sleep medication use, and total PSQI scores at follow-up were significantly worse for persistent poor sleepers than for new incident cases. In addition, habitual sleep efficiency slightly but significantly worsened over the two-year course. These findings indicate that sleep disturbances/insomnia symptoms worsen with time. In particular, the results of our cross-sectional and longitudinal analyses suggest that sleep efficiency worsens over time. Leger et al. reported that chronic insomnia patients showed lower SF-36 scores than patients without insomnia did in all eight measures used in their study, and that the more severe the insomnia symptoms were, the worse the QoL was [30]. To the best of our knowledge, ours is the first study to describe the natural course of changes in QoL due to sleep disturbances/insomnia. Our findings revealed that persistent poor sleepers had the worst physical QoL at follow-up among the four categories and worse mental QoL compared to good sleepers and remitted cases but not new incident cases. These results suggest that physical QoL continues to deteriorate as sleep disturbances/insomnia becomes chronic, while mental QoL deteriorates when sleep disturbances/insomnia first develops but stabilizes subsequently. When we examined longitudinal QoL data among persistent poor sleepers, neither the MCS nor PCS scores differed significantly over the 2-year course. Thus, a modest aggravation of sleep disturbances/insomnia symptoms may not necessarily lead to a deterioration of daytime functioning over time. A longitudinal comparison revealed significant differences in the sub-item scores of C1 (sleep quality), C2 (sleep latency), C4 (habitual sleep efficiency), C5 (sleep disturbance), and total PSQI scores between those with mild sleep disturbances/insomnia and those with severe sleep disturbances/insomnia. This suggests that mild cases can gradually worsen over time. Therefore, it is important to start an early intervention or treatment in not only patients with severe insomnia but also those with mild insomnia. This study has several limitations. First, the response rate was approximately 50% at both survey points, with only 28.5% of the total population (baseline) responding to the follow-up questionnaire. Older, unhealthier, and non-smoking respondents may have been more compliant, thus leading to a selection bias. However, considering that the prevalence of insomnia is relatively high in el-
1120
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
Table 4 Changes in the scores of PSQI, CES-D, MSC and PCS in subjects with PSQI P 6 but <9 at baseline and subjects compared with PSQI P 9 at baseline.
C1: sleep quality mean (SD) C2: sleep latency mean (SD) C3: sleep duration mean (SD) C4: habitual sleep efficiency mean (SD) C5: sleep disturbance mean (SD) C6: use of sleeping medication mean (SD) C7: daytime dysfunction mean (SD) PSQI total score mean (SD) CES-D mean (SD) MCS score mean (SD) PCS score mean (SD)
Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up
Total (n = 268)
Results of repeated t-test
PSQI P 6 but <9 at baseline (n = 142)
PSQI P 9 at baseline (n = 126)
Results of unpaired t-test*1
1.6 (0.6) 1.6 (0.6) 1.8 (0.8) 1.9 (0.9) 1.6 (0.8) 1.7 (0.8) 0.8 (1.0) 1.1 (1.1) 1.3 (0.6) 1.3 (0.5) 0.8 (1.2) 0.9 (1.3) 0.9 (0.7) 0.9 (0.7) 8.8 (2.6) 9.2 (2.7) 11.8 (5.2) 12.1 (5.2) 46.8 (7.4) 46.8 (7.0) 44.6 (7.6) 44.6 (7.8)
ns
1.4 (0.5) 1.5 (0.6) 1.5 (0.8) 1.7 (0.8) 1.5 (0.8) 1.6 (0.8) 0.4 (0.6) 0.9 (1.0) 1.1 (0.4) 1.2 (0.5) 0.4 (0.9) 0.5 (1.0) 0.7 (0.6) 0.8 (0.7) 6.9 (0.8) 8.1 (2.1) 10.3 (4.4) 10.7 (4.4) 49.3 (5.8) 48.2 (6.1) 45.7 (7.5) 45.6 (7.8)
1.8 (0.6) 1.7 (0.7) 2.2 (0.7) 2.1 (0.8) 1.8 (0.7) 1.7 (0.9) 1.3 (1.1) 1.3 (1.2) 1.6 (0.6) 1.5 (0.6) 1.3 (1.4) 1.4 (1.4) 1.1 (0.8) 1.0 (0.7) 11.0 (2.1) 10.4 (2.9) 13.6 (5.5) 13.5 (5.7) 44.1 (8.0) 45.2 (7.6) 43.9 (7.4) 43.4 (7.6)
p < 0.01
ns ns p < 0.01 ns ns ns ns ns ns ns
p < 0.01 ns p < 0.01 p < 0.01 ns ns p < 0.01 ns ns ns
PSQI = Pittsburgh Sleep Quality Index, CES-D = Center for Epidemiologic Studies Depression Scale, MCS = Mental Component Summary, PCS = Physical Component Summary, SD = standard deviation, ns = not significant. *1 Results of unpaired t-test for differences in scores at follow up between the group with baseline total PSQI score P6 but <9 and the group with the score P9.
derly people and in those who have a disease, this bias might not necessarily imply an underrepresentation of insomnia patients. Second, the questionnaire did not include any measurements of socioeconomic background. Third, in the current study, the definition of sleep disturbances/insomnia was based on the PSQI cutoff established in previous studies [14–16], and the PSQI is frequently used across different ethnic groups. However, unlike the DSM-IVTR and ICSD-II, the PSQI does not provide a clear description of the relationship between nocturnal symptoms and daytime consequences. Fourth, we classified participants into four categories on the basis of PSQI scores obtained at two survey points. However, because the follow-up survey was conducted after a relatively long interval of two years, the changes in sleep disturbances/insomnia may not have been assessed accurately as the severity of the symptoms may have fluctuated in the interval between the two data collection periods. Therefore, future studies should have closer interval assessments so that they can obtain more reliable data regarding the course of insomnia [7]. Fifth, although the follow-up duration was two years, it might have been inadequate in providing conclusive information when unaccompanied by detailed data regarding the duration of sleep disturbances/insomnia. Hence, a study over a longer period, coupled with more frequent assessments, may be necessary to draw conclusions that are more definite. In conclusion, about 30% of our subjects had sleep disturbances/ insomnia at both survey points. Among subjects with sleep
disturbances/insomnia at baseline, 60.9% had chronic sleep disturbances/insomnia at follow-up. Insomnia symptoms, especially the problem of sleep efficiency, were exacerbated over time. Mental QoL was found to deteriorate when insomnia first develops but stabilizes subsequently, whereas physical QoL declines as sleep disturbances/insomnia become chronic. These findings emphasize the importance of an early intervention and treatment in not only populations with severe insomnia, but also in those with mild symptoms.
Conflict of interest The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: 10.1016/j.sleep.2012.05.015.
Acknowledgements The authors are indebted to Mr. Roderick J. Turner; Associate Professor Edward F. Barroga; and Professor J. Patrick Barron, Chairman of the Department of International Medical Communications of Tokyo Medical University for their editorial review of this manuscript.
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
References [1] Ohayon M. Epidemiological study on insomnia in the general population. Sleep 1996;19:S7–S15. [2] Weyerer S, Dilling H. Prevalence and treatment of insomnia in the community: results from the Upper Bavarian Field Study. Sleep 1991;14:392–8. [3] Liu X, Uchiyama M, Kim K, Okawa M, Shibui K, Kudo Y, et al. Sleep loss and daytime sleepiness in the general adult population of Japan. Psychiatry Res 2000;93:1–11. [4] Ohayon MM. Prevalence of DSM-IV diagnostic criteria of insomnia: distinguishing insomnia related to mental disorders from sleep disorders. J Psychiatr Res 1997;31:333–46. [5] Moul DE, Nofzinger EA, Pilkonis PA, Houck PR, Miewald JM, Buysse DJ. Symptom reports in severe chronic insomnia. Sleep 2002;25:553–63. [6] American Academy of Sleep Medicine. International classification of sleep disorders. 2nd ed. Diagnostic and coding manual. IL, Westchester: American Academy of Sleep Medicine; 2005. [7] Morin CM, Belanger L, LeBlanc M, Ivers H, Savard J, Espie CA, et al. The natural history of insomnia: a population-based 3-year longitudinal study. Arch Intern Med 2009;169:447–53. [8] Katz DA, McHorney CA. Clinical correlates of insomnia in patients with chronic illness. Arch Intern Med 1998;158:1099–107. [9] Varkevisser M, Van Dongen HP, Van Amsterdam JG, Kerkhof GA. Chronic insomnia and daytime functioning: an ambulatory assessment. Behav Sleep Med 2007;5:279–96. [10] Edinger JD, Means MK, Carney CE, Krystal AD. Psychomotor performance deficits and their relation to prior nights’ sleep among individuals with primary insomnia. Sleep 2008;31:599–607. [11] Varkevisser M, Kerkhof GA. Chronic insomnia and performance in a 24-h constant routine study. J Sleep Res 2005;14:49–59. [12] Leger D, Scheuermaier K, Raffray T, Metlaine A, Choudat D, Guilleminault C. HD-16: a new quality of life instrument specifically designed for insomnia. Sleep Med 2005;6:191–8. [13] Zammit GK, Weiner J, Damato N, Sillup GP, McMillan CA. Quality of life in people with insomnia. Sleep 1999;22(Suppl. 2):S379–85. [14] Sasai T, Inoue Y, Komada Y, Nomura T, Matsuura M, Matsushima E. Effects of insomnia and sleep medication on health-related quality of life. Sleep Med 2010;11:452–7. [15] Komada Y, Nomura T, Kusumi M, Nakashima K, Okajima I, Sasai T, et al. Correlations among insomnia symptoms, sleep medication use and depressive symptoms. Psychiatry Clin Neurosci 2011;65:20–9. [16] Nomura T, Inoue Y, Kusumi M, Uemura Y, Nakashima K. Prevalence of restless legs syndrome in a rural community in Japan. Mov Disord 2008;23:2363–9.
1121
[17] Kagimura T, Nomura T, Kusumi M, Nakashima K, Inoue Y. Prospective survey on the natural course of restless legs syndrome over two years in a closed cohort. Sleep Med 2011;12:821–6. [18] Doi Y, Minowa M, Uchiyama M, Okawa M, Kim K, Shibui K, et al. Psychometric assessment of subjective sleep quality using the Japanese version of the Pittsburgh Sleep Quality Index (PSQI-J) in psychiatric disordered and control subjects. Psychiatry Res 2000;97:165–72. [19] Tokuda Y, Okubo T, Ohde S, Jacobs J, Takahashi O, Omata F, et al. Assessing items on the SF-8 Japanese version for health-related quality of life: a psychometric analysis based on the nominal categories model of item response theory. Value Health 2009;12:568–73. [20] Poulin C, Hand D, Boudreau B. Validity of a 12-item version of the CES-D used in the National Longitudinal Study of Children and Youth. Chronic Dis Can 2005;26:65–72. [21] Ford DE, Kamerow DB. Epidemiologic study of sleep disturbances and psychiatric disorders. An opportunity for prevention? JAMA 1989;262:1479–84. [22] Okajima I, Komada Y, Nomura T, Nakashima K, Inoue Y. Insomnia as a risk for depression: a longitudinal epidemiologic study on a Japanese rural cohort. J Clin Psychiatry 2012;73:377–83. [23] Kim JM, Stewart R, Kim SW, Yang SJ, Shin IS, Yoon JS. Insomnia, depression, and physical disorders in late life: a 2-year longitudinal community study in Koreans. Sleep 2009;32:1221–8. [24] Jansson-Frojmark M, Linton SJ. The course of insomnia over one year: a longitudinal study in the general population in Sweden. Sleep 2008;31:881–6. [25] Zhang J, Lam SP, Li SX, Li AM, Lai KY, Wing YK. Longitudinal course and outcome of chronic insomnia in Hong Kong Chinese children: a 5-year followup study of a community-based cohort. Sleep 2011;34:1395–402. [26] Johnson EO, Roth T, Breslau N. The association of insomnia with anxiety disorders and depression: exploration of the direction of risk. J Psychiatr Res 2006;40:700–8. [27] Chang PP, Ford DE, Mead LA, Cooper-Patrick L, Klag MJ. Insomnia in young men and subsequent depression. The Johns Hopkins Precursors Study. Am J Epidemiol 1997;146:105–14. [28] Breslau N, Roth T, Rosenthal L, Andreski P. Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry 1996;39:411–8. [29] Zhang J, Lam SP, Li SX, Yu MW, Li AM, Ma RC, et al. Long-term outcomes and predictors of chronic insomnia: A prospective study in Hong Kong Chinese adults. Sleep Med 2012;13:455–62. [30] Leger D, Scheuermaier K, Philip P, Paillard M, Guilleminault C. SF-36: evaluation of quality of life in severe and mild insomniacs compared with good sleepers. Psychosom Med 2001;63:49–55.
Contents lists available at SciVerse ScienceDirect
Sleep Medicine journal homepage: www.elsevier.com/locate/sleep
Original Article
A two-year follow-up study on the symptoms of sleep disturbances/insomnia and their effects on daytime functioning Yoko Komada a,b, Takashi Nomura c, Masayoshi Kusumi c, Kenji Nakashima c, Isa Okajima a,b, Taeko Sasai a,b, Yuichi Inoue a,b,⇑ a b c
Department of Somnology, Tokyo Medical University, Japan Japan Somnology Center, Neuropsychiatric Research Institute, Japan Department of Neurology, Institute of the Neurological Sciences, Tottori University Faculty of Medicine, Japan
a r t i c l e
i n f o
Article history: Received 19 December 2011 Received in revised form 22 May 2012 Accepted 23 May 2012 Available online 25 July 2012 Keywords: Quality of life Chronic insomnia Longitudinal study SF-8 Natural course Mental quality of life Physical quality of life
a b s t r a c t Objective: This study attempts to identify changes in the symptoms of sleep disturbances/insomnia over a two-year course and their effects on daytime functioning. Methods: We administered two population-based epidemiological surveys in 2005 and 2007 to participants from rural Japan. Results: In the first survey, 30.7% of the subjects reported sleep disturbances/insomnia. Among them, 60.9% reported sleep problems at the two-year follow-up. A comparison of sleep disturbances/insomnia, and subjective daytime functioning measures between the new incident cases and persistent poor sleepers revealed that the total score of persistent poor sleepers was significantly lower than that of new incident cases on the Pittsburgh Sleep Quality Index and physical quality of life (QoL) but not mental QoL. Longitudinal comparisons of the symptoms of sleep disturbances/insomnia in persistent poor sleepers revealed that sleep efficiency was significantly worse at follow-up. Exacerbation of the symptoms of sleep disturbances/insomnia at follow-up was observed in mild but not severe cases. Conclusions: Sleep efficiency progressively worsens over time, and physical QoL can deteriorate as sleep disturbances/insomnia become chronic. Since the symptoms of sleep disturbances/insomnia and their daytime effects are exacerbated even in mild cases, early intervention and treatment are necessary. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Insomnia is a common disorder [1,2], with an estimated prevalence of about 20% among the general population [3,4]. Poor nocturnal sleep and consequent impairment in daytime functioning are the core symptoms of insomnia [5]. Indeed, both daytime impairment and night-time sleep difficulties have been established as essential items in the diagnostic criteria of insomnia by the International Classification of Sleep Disorders [6]. Several reports have described the natural course of insomnia and have revealed that a considerable number of people with insomnia exhibit a chronic course [7,8]. However, specific changes in the severity and the symptoms of the disorder as it becomes chronic have yet to be ascertained.
⇑ Corresponding author at: Department of Somnology, Tokyo Medical University, 6-7-1 Nishishinjuku, Shinjuku-ku, Tokyo 160-0023, Japan. Tel.: +81 0 3 3342 6111; fax: +81 0 3 3342 7083. E-mail address: [email protected] (Y. Inoue). 1389-9457/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.sleep.2012.05.015
Previous studies on clinical populations have reported that patients with chronic insomnia commonly complain of subjective daytime impairments, including mood disturbances, concentration problems, easy fatigue, and sleepiness [5,9]. Objective measures revealed that insomnia patients show impairment in tasks that evaluate vigilance, working memory, and motor control [10,11]. These daytime dysfunctions attributed to insomnia are assumed to negatively affect sufferers’ quality of life (QoL), which is a measure of general daytime functioning [12,13]. We previously reported that insomnia was generally associated with depressed mood and low QoL scores in both the mental and physical component among participants from rural Japan [14,15]. However, the precise impact of the chronicity of insomnia on QoL remains to be determined. In order to investigate these issues, we performed a longitudinal study where a two-part questionnaire was administered at the start and end of a two-year interval to a cohort taking part in sleep studies in a single rural community [14,16,17]. Through this study, we hope to (1) elucidate the changes in insomnia symptoms over a two-year course and (2) determine the effects of chronic insomnia on daytime functioning.
1116
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
2. Methods 2.1. Participants and procedures This survey was conducted as part of the sleep studies mentioned above in a rural town (Daisen, Tottori Prefecture, Western Japan) [14,16,17]. The Ethics Committee of Tottori University approved this study, and all participants provided written informed consent. In 2004 the total population of the town was 6643, with 5528 residents aged 20 years and above (2521 men, 3007 women), who had a mean age of 55.2 years. The first part of the questionnaire survey was administered between November 2005 and January 2006 (baseline), and the second part was administered between November 2007 and December 2007 (follow-up). With the cooperation of local public health nurses, questionnaires tagged with serial numbers were delivered to all residents aged 20 years and older during both periods. In the current study we did not distinguish nursing home respondents from hospital respondents. Responses to the questionnaire were received from 2822 people at baseline (response rate: 51%; 1222 men, 1600 women; mean age = 57.4, SD = 17.7). Two years later, the follow-up questionnaires were sent to the people who had responded at baseline, with 1577 of them responding to the follow-up questionnaire (response rate: 56%; 683 men, 894 women; mean age = 58.6, SD = 16.1) (Fig. 1). The responses to the two surveys were matched using the serial numbers. 2.2. Measures The contents of the questionnaires at both baseline and followup were as follows:
quality (category 1 [C1]), sleep latency (C2), sleep duration (C3), habitual sleep efficiency (C4), sleep disturbance (C5), use of sleeping medication (C6), and daytime dysfunction (C7). Although the PSQI evaluates sleep disturbances rather than insomnia, the cut-off score for insomnia has been established at 5.5 points according to a previous study [18], and we have followed that convention. Consequently, respondents with PSQI scores of 5.5 or higher were classified as people with sleep disturbances/insomnia in this study. (3) The standardized eight-item Short Form Health Survey of the Medical Outcomes Study (SF-8) [19] was used to assess QoL. The SF-8 measures vitality, social functioning, mental health condition, emotional state, general health, physical functioning, physical state, and bodily pain. The mental component summary (MCS) scale of the SF-8 was used to evaluate mental QoL and the physical component summary (PCS) scale was used to evaluate physical QoL. The average scores for both scales for the general population were set at 50 points. There were overlapping questions regarding daytime dysfunction (C7) on the PSQI and SF-8. However, the PSQI measured daytime dysfunction with regard to sleep disturbances/insomnia, whereas the SF-8 does not make this distinction. Therefore, we analyzed the scores from the PSQI and SF-8 independently of each other. (4) The 12-item version of the Center for Epidemiological Studies Depression Scale (CES-D) [20] was used to measure depressive symptoms. The scale has four response options: ‘‘never or rarely’’ (0), ‘‘sometimes’’ (1), ‘‘often’’ (2), and ‘‘always’’ (4). We used the total scores of CES-D as parameters of depression. 2.3. Statistical analysis
(1) Demographic variables: The participants were asked about their age, sex, the disease currently being treated (‘‘Please tell us the disease you are currently treated for.’’), family situation (‘‘Do you currently live with your family?’’), smoking habits (‘‘Do you currently smoke?’’), and alcohol consumption (‘‘Do you drink regularly?’’). (2) The Japanese version of the Pittsburgh Sleep Quality Index (PSQI) [18] was used to estimate the level of subjective sleep disturbance. The PSQI includes sub-items that evaluate sleep
Student’s paired t-tests were used to compare the PSQI, CES-D, MCS, and PCS scores between the baseline and follow-up surveys for participants who responded to both surveys. On the basis of the results of these tests, the participants were divided into four categories: good sleepers (no insomnia symptoms at both survey periods), new incident cases (no insomnia symptoms at baseline but symptoms present at follow-up), remitted cases (insomnia symptoms at baseline but not at follow-up), and
Survey at baseline Eligible subjects: n = 5,528 2,521 men (46%) 3,007 women (54%)
Non-responders: n = 2 2,706 706 Responders: n = 2 2,822 (51%) 1,222 men (43%) 1 600 women (57%) 1, Sleep disturbances/insomnia patients (PSQI ≥ 6) n = 641 (22.7%)
Non-responders: p n = 1,247 ,
Survey at follow up Responders: p 1,577 , (56%) ( ) 683 men (43%) 894 women (57%) Sleep disturbances/insomnia patients (PSQI ≥ 6) n = 476 (30.2%) Fig. 1. Survey flowchart.
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
persistent poor sleepers (insomnia symptoms present at both survey points) [21,22]. A two-way repeated-measures analysis of variance (ANOVA) (category survey period) was used to compare the scores of the PSQI, CES-D, and the MCS and PCS of the SF-8. The Bonferroni–Dunn test was used for post hoc analysis, and the chi-square test was performed to compare categorical variables. Student’s paired t-tests were used to compare the PSQI, CES-D, MCS, and PCS scores between baseline and follow-up for persistent poor sleepers to reveal any changes in symptoms of sleep disturbances/insomnia and daytime functioning among people with chronic insomnia over the two-year period. A series of logistic regression analyses were conducted for subjects with insomnia/sleep disturbances at baseline to elucidate the factors associated with their persistence. Persistent poor sleepers were further divided into two groups according to their median PSQI value at baseline (at least six but less than nine, and nine or more respectively) to determine whether the changes in the course of insomnia/sleep disturbances differed between mild and severe cases. Changes in scores of PSQI sub-items, total PSQI, CES-D, PCS, and MCS over the two-year interval were examined in these two groups by using Student’s unpaired t-tests. All statistical analyses were performed with SPSS Version 11.5 (SPSS Japan, Inc., Tokyo, Japan) with the alpha value set at 0.01.
3. Results When the demographic data of the respondents who answered only at baseline (n = 1247) and those who responded at both baseline and follow-up (n = 1577) were compared, a significant difference was observed in age (t [2394] = 3.56, p < 0.01), with participants in one group having a mean age of 55.9 (SD 19.6) and the other with a mean age of 58.6 (SD = 16.1). However, the difference in ages between the two groups was only 2.7 years. There were also significant differences in the number of participants who were currently receiving treatment for any disease (31.3% vs. 38.8%; X2 [1] = 17.4; p < 0.01), and in smoking habits (26.6% vs. 18.0%; X2 [1] = 30.0; p < 0.01). No other significant differences were found between the demographics of the two groups. In all participants, the PSQI, CES-D, and MCS scores were slightly but significantly worse at follow-up than at baseline (PSQI: t [1417] = 2.9; CES-D: t [1390] = 2.9; MCS: t [1373] = 2.7; p < 0.01 for all); there was no significant difference for PCS scores (t [1373] = 2.3; not significant [ns]). Among the participants, 56.4% were classified as good sleepers, 12.9% as new incident cases, 12.0% as remitted cases, and 18.7% as persistent poor sleepers. Table 1 shows the demographic data of the participants and the PSQI scores in each survey, and the results of the one-way ANOVA or chi-square test for each parameter of the four insomnia categories. There was no significant difference in age among the four categories (F [3, 1429] = 3.9, ns). A significant difference was observed in the number of participants with insomnia who were undergoing treatment at both baseline and follow-up between the four categories (baseline: X2 [3] = 15.6; follow-up: X2 [3] = 18.1, p < 0.01 for both). Residual analysis revealed that the number of persistent poor sleepers who were undergoing treatment was significantly higher than that of other categories at both data collection periods, and that the number of good sleepers was significantly lower than that of other categories at followup. There was also a significant difference in the number of participants using sleep medication once or more per week at both baseline and follow-up among the four categories (baseline: X2 [3] = 262.3; follow-up: X2 [3] = 243.1; p < 0.01 for both). Residual analysis revealed that remitted cases and persistent poor sleepers
1117
had significantly higher rates of sleep medication use at baseline, and that new incident cases and persistent poor sleepers used sleep medication more often than other participants belonging to categories did at follow-up. A two-way repeated measures ANOVA (category survey period) was performed to determine differences in total PSQI score. There were main effects for both category and survey period, and a significant interaction (main effect of category: F [3, 1414] = 1188.6, p < 0.01; main effect of the survey period: F [1, 1414] = 16.8, p < 0.01; and interaction between category and survey period: F [3, 1414] = 429.2, p < 0.01). The Bonferroni– Dunn post hoc test showed that the total PSQI score was significantly worse at follow-up than at baseline for new incident cases, and significantly improved at follow-up for remitted cases, with no significant differences in the scores between the two survey points for good sleepers and persistent poor sleepers. The total PSQI score for persistent poor sleepers was significantly worse than that for good sleepers, new incident cases, and remitted cases both at baseline and at follow-up (Table 1). 3.1. Comparison of insomnia symptoms in the follow-up survey between new incident cases and persistent poor sleepers PSQI scores at follow-up were compared between new incident cases and persistent poor sleepers to determine the differences in the severity of sleep disturbances/insomnia and the characteristics of the symptoms. We found that the sleep latency (C2), habitual sleep efficiency (C4), use of sleep medication (C6), and total PSQI scores of persistent poor sleepers were significantly worse than those of new incident cases (C2: t [442] = 4.9; C4: t [436] = 2.7; C6: t [442] = 3.5; PSQI total score: t [451] = 6.4, respectively, p < 0.01 for all) (Table 2). The factors associated with persistent insomnia/sleep disturbances were examined with a series of univariate logistic regression analyses, performed for the 11 independent variables: age; sex; disease currently being treated; alcohol consumption; smoking habits; living status (alone or with co-habitants); sleep medication use; and CES-D, MCS, PCS, and PSQI scores at baseline. Two of these variables (living status and PSQI score) were significantly correlated with persistent poor sleepers. The adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were 3.8 (1.1–13.2) for living alone (p < 0.01), and 1.4 (1.3–1.6) for PSQI score (p < 0.01), respectively. These two variables were then analyzed with a multivariate model, which revealed that chronic insomnia/sleep disturbances were significantly associated with a higher PSQI score at baseline (OR = 1.4, 95% CI: 1.2–1.6; p < 0.01). 3.2. Comparison of daytime consequences among categories A two-way repeated measures ANOVA (category survey period) was conducted to compare the scores of daytime consequence measures (CES-D, MCS and PCS) at both survey periods. There were main effects of both category and survey period, and an interaction was found for CES-D (main effect of category: F [3, 1387] = 119.6, p < 0.01; main effect of survey period: F [1, 1387] = 6.8, p < 0.01; interaction between category and survey period: F [3, 1387] = 13.8, p < 0.01). There was a main effect of category and an interaction between both variables for MCS and PCS (MCS: main effect: F [3, 1370] = 48.5, p < 0.01; interaction: F [3, 137] = 13.3, p < 0.01; PCS: main effect: F [3, 1370] = 31.9, p < 0.01; interaction: F [3, 137] = 3.9, p < 0.01). Post-hoc analysis confirmed that CES-D and MCS scores were significantly worse at follow-up than at baseline for good sleepers, that CES-D, MCS, and PCS scores were significantly worse at follow-up than at baseline for new incident cases, and that MCS was better at follow-up than at baseline for remitted
1118
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
Table 1 Demographic data and PSQI scores of participants. At baseline
Total (n = 1434)
Drinking habits, n (%) Smoking habits, n (%) Living alone, n (%) Sleep medication usea, n (%)
PSQIb total score, mean (SD)
With sleep disturbances/insomnia (n = 440)
Without sleep disturbances/ insomnia (good sleepers, n = 809)
With sleep disturbances/ insomnia (new incident cases, n = 185)
Without sleep disturbances/ insomnia (remitted cases, n = 172)
With sleep disturbances/ insomnia (persistent poor sleepers, n = 268)
p-Values of chi-square test or ANOVA and post hoc test among 4 sleep disturbances/insomnia groups
Baseline
633/801 60.2 (16.0) 557 (38.9%)
370/439 59.6 (15.9) 292 (36.1%)
72/113 59.5 (17.5) 63 (34.1%)
81/91 59.9 (15.8) 72 (41.9%)
110/158 62.8 (15.5) 130 (48.5%)
Follow up
611 (42.6%)
314 (38.8%)
86 (46.5%)
69 (40.1%)
142 (53.0%)
Baseline Follow up Baseline Follow up Baseline Follow up Baseline
540 (37.7%) 399 (27.8%) 263 (18.4%) 247 (17.2%) 59 (4.1%) 62 (4.3%) 106 (7.5%)
309 (38.2%) 227 (28.1%) 147 (18.2%) 136 (16.8%) 33 (4.1%) 31 (3.8%) 2 (0.3%)
70 (37.8%) 46 (24.9%) 34 (18.4%) 27 (14.6%) 6 (3.2%) 7 (3.8%) 1 (0.5%)
72 (41.9%) 58 (33.7%) 37 (21.5%) 39 (22.7%) 3 (1.7%) 5 (2.9%) 26 (15.1%)
89 68 45 45 17 19 77
Follow up
120 (8.5%)
6 (0.8%)
31 (17.1%)
4 (2.4%)
79 (30.0%)
Baseline Follow up
4.6 (3.0) 4.8 (3.2)
2.9 (1.5) 2.9 (1.5)
3.7 (1.3)* 7.7 (1.9)*
7.2 (1.7)* 3.8 (1.2)*
8.8 (2.6) 9.2 (2.7)
ns ns Persistent poor sleepers (+) (p < 0.01) good sleepers ( ) (p < 0.01), persistent poor sleepers (+) (p < 0.01) ns ns ns ns ns ns Good sleepers , new incident cases ( ) (p < 0.01), remitted cases, persistent poor sleepers (+) (p < 0.01) Good sleepers, remitted cases ( ) (p < 0.01), new incident cases, persistent poor sleepers (+) (p < 0.01) Insomnia subcategories, survey period, interaction: p < 0.01, respectivelyc
At follow up (2 years later)
Sex (M/F) Age, mean (SD) Disease currently treated, n (%)
Without sleep disturbances/insomnia (n = 994)
(33.2%) (25.4%) (16.8%) (16.8%) (6.3%) (7.1%) (28.7%)
SD = standard deviation, M = male, F = female, ns = not significant. a Sleep medication use of once or more per week; calculated using the PSQI subitem (C6). b PSQI = Pittsburgh Sleep Quality Index. c Results of two-way repeated measurements ANOVA and post hoc tests. * p < 0.01.
cases (p < 0.01 for all). There were no significant differences between these measures in persistent poor sleepers between baseline and follow-up. The Bonferroni–Dunn post hoc test also showed that the scores of the CES-D and PCS at follow-up in persistent poor sleepers were significantly worse than those of the other categories. In addition, MCS scores at follow-up in persistent poor sleepers were significantly worse than those of good sleepers and remitted cases. However, there was no significant difference in MCS scores at follow-up between persistent poor sleepers and new incident cases (Table 3).
Table 2 Comparison of sleep disturbances/insomnia symptoms at follow up between new incident cases and persistent poor sleepers.
C1: sleep quality C2: sleep latency C3: sleep duration C4: habitual sleep efficiency C5: sleep disturbance C6: use of sleeping medication C7: daytime dysfunction PSQI total score
New incident cases
Persistent poor sleepers
pValue
1.5 1.5 1.5 0.8
1.6 1.9 1.7 1.1
(0.6) (0.9) (0.8) (1.1)
ns <0.01 ns <0.01
1.2 (0.5) 0.5 (1.1)
1.3 (0.5) 0.9 (1.3)
ns <0.01
0.8 (0.7) 7.7 (1.9)
0.9 (0.7) 9.2 (2.7)
ns <0.01
(0.6) (0.9) (0.8) (0.9)
PSQI = Pittsburgh Sleep Quality Index, Mean (standard deviation, SD), Student’s ttest; ns = not significant.
3.3. Changes in the symptoms of sleep disturbances/insomnia and depression and quality of life between the two survey points in persistent poor sleepers The PSQI, CES-D, and QoL scores of persistent poor sleepers between baseline and follow-up were compared in order to determine changes over time in the symptoms of sleep disturbances/ insomnia and their effects on daytime functioning. The results are presented in Table 4. Habitual sleep efficiency (C4) at followup was significantly worse than that at baseline (t [255] = 3.2, p < 0.01). However, there were no significant differences in scores for the other sub-items or total PSQI between the two survey periods. There were also no significant differences in CES-D, MCS, and PCS scores. At baseline, 440 subjects reported sleep disturbances/insomnia, with 286 having a PSQI score of at least 6 but less than 9 and 154 having a PSQI score of 9 or more. The remission rate, indicated by a PSQI score of less than 6 at follow-up, was 50.3% in the former and 18.2% in the latter. The difference in remission rate between the two groups was significant (X2 [1] = 43.5; p < 0.01). In order to investigate whether or not a baseline severitydependent difference in the longitudinal course of sleep disturbances/insomnia was present, changes in PSQI, CES-D, and QoL scores were examined using the Student’s unpaired t-test in the two groups of subjects with chronic sleep disturbances/insomnia – namely, the mild insomnia group (with a PSQI score of more than 6 but less than 9) and the severe insomnia group (with a PSQI score of 9 or more at baseline). There were significant differences be-
1119
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121 Table 3 Scores of CES-D, MSC and PCS of participants. At baseline
Total (n = 1,434)
At follow up (2 years later) CES-D score, mean (SD) MCS score, mean (SD) PCS score, mean (SD)
Without sleep disturbances/insomnia (n = 994)
With sleep disturbances/insomnia (n = 440)
Without sleep disturbances/ insomnia (good sleepers, n = 809)
With sleep disturbances/insomnia (new incident cases, n = 185)
Without sleep disturbances/ insomnia (remitted cases, n = 172)
With sleep disturbances/insomnia (persistent poor sleepers, n = 268)
Results of two-way repeated measurements ANOVA and post hoc tests Insomnia subcategories, survey period, interaction: p < 0.01, respectively Insomnia subcategories, interaction: p < 0.01, respectively
Baseline Follow up
8.5 (4.8) 8.9 (4.7)
7.0 (3.9)* 7.3 (3.8)*
8.7 (4.6)* 10.8 (4.9)*
10.0 (5.1) 8.9 (4.3)
11.8 (5.2) 12.1 (5.2)
Baseline Follow up
49.9 (6.5) 49.4 (6.5)
51.4 (5.4)* 50.8 (5.7)*
49.7 (6.6)* 47.0 (7.3)*
48.3 (7.2)* 50.0 (6.2)*
46.8 (7.4) 46.8 (7.0)
Baseline Follow up
47.7 (7.0) 47.4 (7.0)
49.0 (6.3) 48.6 (6.2)
48.1 (7.0)* 46.3 (7.4)*
46.5 (7.4) 47.1 (7.1)
44.6 (7.6) 44.6 (7.8)
Insomnia subcategories, interaction: p < 0.01, respectively
SD = standard deviation, M = male, F = female, ns = not significant, CES-D = Center for Epidemiologic Studies Depression Scale, MCS = Mental Component Summary, PCS = Physical Component Summary. * p < 0.01.
tween the two groups in terms of C1, C2, C4, C5, and total PSQI scores (C1: t [262] = 3.2, C2: t [259] = 3.8, C4: t [254] = 4.0, C5: t [261] = 3.5, total PSQI score: t [2566] = 6.4; p < 0.01 for all). However, there were no significant differences in CES-D, MCS, or PCS scores between the two groups.
4. Discussion We conducted a longitudinal study over a two-year period on a rural population cohort in Japan and examined the course of sleep disturbances/insomnia symptoms and their effects on daytime functioning. The percentage of subjects with sleep disturbances/ insomnia was 30.7% at baseline and 31.6% at follow-up and the number of subjects using sleep medication corresponded to the presence or absence of symptoms of sleep disturbances/insomnia at the studied points. Among the study population, the 18.7% of the subjects reported chronic sleep disturbances/insomnia (a PSQI score of greater than or equal to 5.5 in both surveys). The majority of subjects with sleep disturbances/insomnia at baseline still experienced the symptoms of sleep disturbance at follow-up (60.9%), which is consistent with previous reports [7,8]. For instance, Katz et al. showed that even in subjects with mild insomnia, 59% had sleep problems at a two-year follow-up [8]. Previous longitudinal studies on general populations showed that 40% of the individuals with insomnia at baseline had persistent symptoms [23–25]. Another study on the development of insomnia indicated that 74% of the subjects reported having insomnia symptoms for at least one year [7]. These results indicate that insomnia has a generally persistent course. In the present study, persistent poor sleepers had the highest CES-D score among the four categories, a finding in accordance with those of previous studies [26]. A bidirectional causal relationship between depression and insomnia has been firmly established; that is, insomnia can lead to depression and vice versa [22,27,28]. Therefore, a higher CES-D score could be either a cause or a consequence of sleep disturbances/insomnia. A recent study suggested that the persistence of insomnia was associated with the female sex, lower education level, and daytime symptoms at baseline [29]. However, these associations were not found in our study. The reason for this discrepancy is not clear. However, the fact that our sample consisted of people from a rural area and engaged in agriculture is a possible explanation [15].
In this study, the persistence of sleep disturbances/insomnia was associated with a higher PSQI score at baseline. This result suggests that severe insomnia runs a chronic course. Moreover, the sub-item scores of sleep latency, habitual sleep efficiency, sleep medication use, and total PSQI scores at follow-up were significantly worse for persistent poor sleepers than for new incident cases. In addition, habitual sleep efficiency slightly but significantly worsened over the two-year course. These findings indicate that sleep disturbances/insomnia symptoms worsen with time. In particular, the results of our cross-sectional and longitudinal analyses suggest that sleep efficiency worsens over time. Leger et al. reported that chronic insomnia patients showed lower SF-36 scores than patients without insomnia did in all eight measures used in their study, and that the more severe the insomnia symptoms were, the worse the QoL was [30]. To the best of our knowledge, ours is the first study to describe the natural course of changes in QoL due to sleep disturbances/insomnia. Our findings revealed that persistent poor sleepers had the worst physical QoL at follow-up among the four categories and worse mental QoL compared to good sleepers and remitted cases but not new incident cases. These results suggest that physical QoL continues to deteriorate as sleep disturbances/insomnia becomes chronic, while mental QoL deteriorates when sleep disturbances/insomnia first develops but stabilizes subsequently. When we examined longitudinal QoL data among persistent poor sleepers, neither the MCS nor PCS scores differed significantly over the 2-year course. Thus, a modest aggravation of sleep disturbances/insomnia symptoms may not necessarily lead to a deterioration of daytime functioning over time. A longitudinal comparison revealed significant differences in the sub-item scores of C1 (sleep quality), C2 (sleep latency), C4 (habitual sleep efficiency), C5 (sleep disturbance), and total PSQI scores between those with mild sleep disturbances/insomnia and those with severe sleep disturbances/insomnia. This suggests that mild cases can gradually worsen over time. Therefore, it is important to start an early intervention or treatment in not only patients with severe insomnia but also those with mild insomnia. This study has several limitations. First, the response rate was approximately 50% at both survey points, with only 28.5% of the total population (baseline) responding to the follow-up questionnaire. Older, unhealthier, and non-smoking respondents may have been more compliant, thus leading to a selection bias. However, considering that the prevalence of insomnia is relatively high in el-
1120
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
Table 4 Changes in the scores of PSQI, CES-D, MSC and PCS in subjects with PSQI P 6 but <9 at baseline and subjects compared with PSQI P 9 at baseline.
C1: sleep quality mean (SD) C2: sleep latency mean (SD) C3: sleep duration mean (SD) C4: habitual sleep efficiency mean (SD) C5: sleep disturbance mean (SD) C6: use of sleeping medication mean (SD) C7: daytime dysfunction mean (SD) PSQI total score mean (SD) CES-D mean (SD) MCS score mean (SD) PCS score mean (SD)
Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up Baseline Follow up
Total (n = 268)
Results of repeated t-test
PSQI P 6 but <9 at baseline (n = 142)
PSQI P 9 at baseline (n = 126)
Results of unpaired t-test*1
1.6 (0.6) 1.6 (0.6) 1.8 (0.8) 1.9 (0.9) 1.6 (0.8) 1.7 (0.8) 0.8 (1.0) 1.1 (1.1) 1.3 (0.6) 1.3 (0.5) 0.8 (1.2) 0.9 (1.3) 0.9 (0.7) 0.9 (0.7) 8.8 (2.6) 9.2 (2.7) 11.8 (5.2) 12.1 (5.2) 46.8 (7.4) 46.8 (7.0) 44.6 (7.6) 44.6 (7.8)
ns
1.4 (0.5) 1.5 (0.6) 1.5 (0.8) 1.7 (0.8) 1.5 (0.8) 1.6 (0.8) 0.4 (0.6) 0.9 (1.0) 1.1 (0.4) 1.2 (0.5) 0.4 (0.9) 0.5 (1.0) 0.7 (0.6) 0.8 (0.7) 6.9 (0.8) 8.1 (2.1) 10.3 (4.4) 10.7 (4.4) 49.3 (5.8) 48.2 (6.1) 45.7 (7.5) 45.6 (7.8)
1.8 (0.6) 1.7 (0.7) 2.2 (0.7) 2.1 (0.8) 1.8 (0.7) 1.7 (0.9) 1.3 (1.1) 1.3 (1.2) 1.6 (0.6) 1.5 (0.6) 1.3 (1.4) 1.4 (1.4) 1.1 (0.8) 1.0 (0.7) 11.0 (2.1) 10.4 (2.9) 13.6 (5.5) 13.5 (5.7) 44.1 (8.0) 45.2 (7.6) 43.9 (7.4) 43.4 (7.6)
p < 0.01
ns ns p < 0.01 ns ns ns ns ns ns ns
p < 0.01 ns p < 0.01 p < 0.01 ns ns p < 0.01 ns ns ns
PSQI = Pittsburgh Sleep Quality Index, CES-D = Center for Epidemiologic Studies Depression Scale, MCS = Mental Component Summary, PCS = Physical Component Summary, SD = standard deviation, ns = not significant. *1 Results of unpaired t-test for differences in scores at follow up between the group with baseline total PSQI score P6 but <9 and the group with the score P9.
derly people and in those who have a disease, this bias might not necessarily imply an underrepresentation of insomnia patients. Second, the questionnaire did not include any measurements of socioeconomic background. Third, in the current study, the definition of sleep disturbances/insomnia was based on the PSQI cutoff established in previous studies [14–16], and the PSQI is frequently used across different ethnic groups. However, unlike the DSM-IVTR and ICSD-II, the PSQI does not provide a clear description of the relationship between nocturnal symptoms and daytime consequences. Fourth, we classified participants into four categories on the basis of PSQI scores obtained at two survey points. However, because the follow-up survey was conducted after a relatively long interval of two years, the changes in sleep disturbances/insomnia may not have been assessed accurately as the severity of the symptoms may have fluctuated in the interval between the two data collection periods. Therefore, future studies should have closer interval assessments so that they can obtain more reliable data regarding the course of insomnia [7]. Fifth, although the follow-up duration was two years, it might have been inadequate in providing conclusive information when unaccompanied by detailed data regarding the duration of sleep disturbances/insomnia. Hence, a study over a longer period, coupled with more frequent assessments, may be necessary to draw conclusions that are more definite. In conclusion, about 30% of our subjects had sleep disturbances/ insomnia at both survey points. Among subjects with sleep
disturbances/insomnia at baseline, 60.9% had chronic sleep disturbances/insomnia at follow-up. Insomnia symptoms, especially the problem of sleep efficiency, were exacerbated over time. Mental QoL was found to deteriorate when insomnia first develops but stabilizes subsequently, whereas physical QoL declines as sleep disturbances/insomnia become chronic. These findings emphasize the importance of an early intervention and treatment in not only populations with severe insomnia, but also in those with mild symptoms.
Conflict of interest The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: 10.1016/j.sleep.2012.05.015.
Acknowledgements The authors are indebted to Mr. Roderick J. Turner; Associate Professor Edward F. Barroga; and Professor J. Patrick Barron, Chairman of the Department of International Medical Communications of Tokyo Medical University for their editorial review of this manuscript.
Y. Komada et al. / Sleep Medicine 13 (2012) 1115–1121
References [1] Ohayon M. Epidemiological study on insomnia in the general population. Sleep 1996;19:S7–S15. [2] Weyerer S, Dilling H. Prevalence and treatment of insomnia in the community: results from the Upper Bavarian Field Study. Sleep 1991;14:392–8. [3] Liu X, Uchiyama M, Kim K, Okawa M, Shibui K, Kudo Y, et al. Sleep loss and daytime sleepiness in the general adult population of Japan. Psychiatry Res 2000;93:1–11. [4] Ohayon MM. Prevalence of DSM-IV diagnostic criteria of insomnia: distinguishing insomnia related to mental disorders from sleep disorders. J Psychiatr Res 1997;31:333–46. [5] Moul DE, Nofzinger EA, Pilkonis PA, Houck PR, Miewald JM, Buysse DJ. Symptom reports in severe chronic insomnia. Sleep 2002;25:553–63. [6] American Academy of Sleep Medicine. International classification of sleep disorders. 2nd ed. Diagnostic and coding manual. IL, Westchester: American Academy of Sleep Medicine; 2005. [7] Morin CM, Belanger L, LeBlanc M, Ivers H, Savard J, Espie CA, et al. The natural history of insomnia: a population-based 3-year longitudinal study. Arch Intern Med 2009;169:447–53. [8] Katz DA, McHorney CA. Clinical correlates of insomnia in patients with chronic illness. Arch Intern Med 1998;158:1099–107. [9] Varkevisser M, Van Dongen HP, Van Amsterdam JG, Kerkhof GA. Chronic insomnia and daytime functioning: an ambulatory assessment. Behav Sleep Med 2007;5:279–96. [10] Edinger JD, Means MK, Carney CE, Krystal AD. Psychomotor performance deficits and their relation to prior nights’ sleep among individuals with primary insomnia. Sleep 2008;31:599–607. [11] Varkevisser M, Kerkhof GA. Chronic insomnia and performance in a 24-h constant routine study. J Sleep Res 2005;14:49–59. [12] Leger D, Scheuermaier K, Raffray T, Metlaine A, Choudat D, Guilleminault C. HD-16: a new quality of life instrument specifically designed for insomnia. Sleep Med 2005;6:191–8. [13] Zammit GK, Weiner J, Damato N, Sillup GP, McMillan CA. Quality of life in people with insomnia. Sleep 1999;22(Suppl. 2):S379–85. [14] Sasai T, Inoue Y, Komada Y, Nomura T, Matsuura M, Matsushima E. Effects of insomnia and sleep medication on health-related quality of life. Sleep Med 2010;11:452–7. [15] Komada Y, Nomura T, Kusumi M, Nakashima K, Okajima I, Sasai T, et al. Correlations among insomnia symptoms, sleep medication use and depressive symptoms. Psychiatry Clin Neurosci 2011;65:20–9. [16] Nomura T, Inoue Y, Kusumi M, Uemura Y, Nakashima K. Prevalence of restless legs syndrome in a rural community in Japan. Mov Disord 2008;23:2363–9.
1121
[17] Kagimura T, Nomura T, Kusumi M, Nakashima K, Inoue Y. Prospective survey on the natural course of restless legs syndrome over two years in a closed cohort. Sleep Med 2011;12:821–6. [18] Doi Y, Minowa M, Uchiyama M, Okawa M, Kim K, Shibui K, et al. Psychometric assessment of subjective sleep quality using the Japanese version of the Pittsburgh Sleep Quality Index (PSQI-J) in psychiatric disordered and control subjects. Psychiatry Res 2000;97:165–72. [19] Tokuda Y, Okubo T, Ohde S, Jacobs J, Takahashi O, Omata F, et al. Assessing items on the SF-8 Japanese version for health-related quality of life: a psychometric analysis based on the nominal categories model of item response theory. Value Health 2009;12:568–73. [20] Poulin C, Hand D, Boudreau B. Validity of a 12-item version of the CES-D used in the National Longitudinal Study of Children and Youth. Chronic Dis Can 2005;26:65–72. [21] Ford DE, Kamerow DB. Epidemiologic study of sleep disturbances and psychiatric disorders. An opportunity for prevention? JAMA 1989;262:1479–84. [22] Okajima I, Komada Y, Nomura T, Nakashima K, Inoue Y. Insomnia as a risk for depression: a longitudinal epidemiologic study on a Japanese rural cohort. J Clin Psychiatry 2012;73:377–83. [23] Kim JM, Stewart R, Kim SW, Yang SJ, Shin IS, Yoon JS. Insomnia, depression, and physical disorders in late life: a 2-year longitudinal community study in Koreans. Sleep 2009;32:1221–8. [24] Jansson-Frojmark M, Linton SJ. The course of insomnia over one year: a longitudinal study in the general population in Sweden. Sleep 2008;31:881–6. [25] Zhang J, Lam SP, Li SX, Li AM, Lai KY, Wing YK. Longitudinal course and outcome of chronic insomnia in Hong Kong Chinese children: a 5-year followup study of a community-based cohort. Sleep 2011;34:1395–402. [26] Johnson EO, Roth T, Breslau N. The association of insomnia with anxiety disorders and depression: exploration of the direction of risk. J Psychiatr Res 2006;40:700–8. [27] Chang PP, Ford DE, Mead LA, Cooper-Patrick L, Klag MJ. Insomnia in young men and subsequent depression. The Johns Hopkins Precursors Study. Am J Epidemiol 1997;146:105–14. [28] Breslau N, Roth T, Rosenthal L, Andreski P. Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry 1996;39:411–8. [29] Zhang J, Lam SP, Li SX, Yu MW, Li AM, Ma RC, et al. Long-term outcomes and predictors of chronic insomnia: A prospective study in Hong Kong Chinese adults. Sleep Med 2012;13:455–62. [30] Leger D, Scheuermaier K, Philip P, Paillard M, Guilleminault C. SF-36: evaluation of quality of life in severe and mild insomniacs compared with good sleepers. Psychosom Med 2001;63:49–55.