Prevalence and Correlates of Insomnia and its Impact on Quality of Life in Chinese Schizophrenia Patients

Abstract

Study Objectives:

This study aimed to determine the prevalence and the sociodemographic and clinical correlates of insomnia in Chinese schizophrenia outpatients and its impact on patients' quality of life (QOL).

Design:

Two hundred fifty-five clinically stable schizophrenia outpatients were randomly selected in Hong Kong and their counterparts matched according to sex, age, age at onset, and length of illness were recruited in Beijing, China. All subjects at both sites were interviewed by the same investigator using standardized assessment instruments.

Setting:

Hong Kong and Beijing, China.

Patients or Participants:

Clinically stable schizophrenia outpatients.

Interventions:

N/A.

Measurements and Results:

In the combined Beijing-Hong Kong sample the frequency of at least one type of insomnia over the previous 12 months was 36.0%; the rates of difficulty initiating sleep (DIS), difficulty maintaining sleep (DMS), and early morning awakening (EMA) were 21.2%, 23.6%, and 11.9%, respectively. Poor sleep was significantly associated with advanced age, older age at onset, fewer psychiatric admissions, severity of positive symptoms, anxiety, extrapyramidal symptoms (EPS) and depressive symptoms, less frequent use of atypical antipsychotic medications (AP), and more frequent use of benzodiazepines (BZD) and hypnotics. Poor sleepers had significantly poorer QOL in all domains than patients without insomnia. After controlling for the potential confounding effects of sociodemographic and clinical factors, a significant difference remained between the 2 groups with regard to the physical QOL domain. A multiple logistic regression analysis found that advanced age, fewer psychiatric admissions, severity of depressive symptoms and use of hypnotics were significant contributors to poor sleep.

Conclusion:

Insomnia is independently associated with poor QOL. More attention should be paid in clinical practice to the high rate of insomnia in Chinese schizophrenia patients.

Citation:

Xiang YT; Weng YZ; Leung CM; Tang WK; Lai KYC; Ungvari GS. Prevalence and correlates of insomnia and its impact on quality of life in Chinese schizophrenia patients. SLEEP 2009;32(1):105-109.

INSOMNIA IS A MAJOR PUBLIC HEALTH CHALLENGE DUE TO ITS HIGH PREVALENCE AND IMPACT ON HEALTH.^1,2 IN RECENT YEARS IT HAS RECEIVED increasing attention from policymakers and health care providers. Several studies have examined various aspects of insomnia both among the general population and in patients with psychiatric disorders including schizophrenia.^3–5

In recent years the concept of quality of life (QOL) has become an increasingly important outcome measure in psychiatric practice, because it purports to offer a more comprehensive view of the effectiveness of pharmacotherapeutic and psychosocial interventions.⁶ A number of factors, including sociodemographic characteristics,⁷ negative psychotic symptoms,⁸ positive psychotic symptoms,⁹ depressive symptoms,¹⁰ antipsychotic drug-induced side effects,¹¹ and length of illness¹² are related to poor QOL in schizophrenia.

Although poor QOL and insomnia are both common in schizophrenia patients, only one study has so far examined their relationship in schizophrenia. Ritsner et al.¹³ measured sleep quality by the Pittsburgh Sleep Quality Index (PSQI) in 59 schizophrenia inpatients and 86 outpatients and found that 45.4% suffered from poor sleep (“poor sleepers”). Poor sleepers had poorer QOL, were more depressed and distressed, and had more medication-induced side effects than patients without sleep disturbances (“good sleepers”). However, the study of Ritsner et al. included a heterogeneous sample (inpatients and outpatients) and small sample size, which significantly limited its power.

There is preliminary evidence that cross-cultural or ethnic differences exist in both QOL¹⁴ and insomnia.^15,16 Findings reported from Western countries may not therefore be applicable to patients from different ethnic and cultural backgrounds. To the best of our knowledge, no study in China has addressed insomnia in schizophrenia and its relationship with QOL.

Schizophrenia outpatients account for more than 90% of the total schizophrenia population in China, over 5 million people,¹⁷ and constitute a particularly important group for study. The objectives of this study, therefore, were (1) to determine the frequency of various types of DSM-IV-defined insomnia in schizophrenia outpatients in 2 major Chinese cities, Beijing and Hong Kong; (2) to investigate the sociodemographic and clinical correlates of insomnia in this patient population; and (3) to examine the impact of insomnia on schizophrenia patients' QOL.

METHODS

The study was part of a project on QOL in schizophrenia outpatients in China conducted between January 2005 and June 2006. A detailed description of the study design and data collection has been reported elsewhere.¹⁷ Subjects in Hong Kong were randomly selected from the schizophrenia patients who attended the outpatient department (OPD) of a university-affiliated general hospital serving a population of approximately 800,000. Their Beijing counterparts, matched according to sex, age, age at onset, and length of illness, were recruited from schizophrenia patients who attended the Adult Psychiatric OPD at Beijing Anding Hospital, a teaching hospital serving a population of about 3,000,000 in Beijing.

Patients who met the following inclusion criteria were invited to participate in the study: (1) DSM-IV diagnosis of schizophrenia; (2) age between 18 to 60 years; (3) length of illness ≥ 5 years; and (4) outpatients who had been clinically stable for ≥ 3 months before recruitment according to Lobana's criteria.¹⁸ Exclusion criteria were: (1) history of or ongoing major chronic medical or neurological condition(s); and (2) past or current significant drug/alcohol abuse other than nicotine.

The course of recruitment was as follows. (1) At the Hong Kong site, all schizophrenia patients meeting the study criteria were identified prior to their attendance at the OPD. (2) Study subjects were selected randomly from the total number of eligible patients and invited to participate in the study. (3) In Beijing, once schizophrenia patients registered at the OPD of Anding Hospital, their medical records were screened to establish their eligibility for the study. The first patient who matched his or her Hong Kong counterpart was invited to participate. (4) The principal author spent 2 months alternatively in Hong Kong and Beijing selecting, recruiting, and assessing the subjects.

The study protocol was approved by the Joint Chinese University of Hong Kong and New Territories East Cluster (CUHK-NTEC) Clinical Research Ethics Committee in Hong Kong and the Human Research and Ethics Committee of Beijing Anding Hospital. Written consent was obtained from all subjects.

To enable direct comparisons to be made with earlier epidemiologic surveys, patients were asked to reply “yes” or “no” to 3 standard statements on sleep disturbances lasting ≥ 2 weeks or longer in the past 12 months, as specified in DSM-IV. These statements, designed respectively to identify difficulties initiating sleep (DIS), difficulties maintaining sleep (DMS) and early morning awakening (EMA), were worded as follows:

1. “It took you two hours or longer nearly every night before you could fall asleep.”

2. “You woke up nearly every night and took an hour or more to get back to sleep.”

3. “You woke up nearly every morning at least two hours earlier than you wanted to.”

Subjects who reported at least one of these sleep difficulties were defined for the purpose of this study as poor sleepers (“poor sleep”), while those who did not report any type of insomnia were regarded as good sleepers (“good sleep”). Long-term benzodiazepine (BZD) treatment was defined as continuous use of any drug of that class for > 12 weeks.¹⁹ Doses of antipsychotic drugs (AP) were converted to chlorpromazine equivalents.²⁰ Zopiclone, zolpidem, the 2 non-BZD hypnotic drugs prescribed for schizophrenia patients, and any herbal remedies used for promoting sleep were categorized as “hypnotics.” The chemical composition of Chinese herbal remedies taken by the subjects was unknown.

The principal author assessed all subjects throughout the study and conducted interviews on the day when the subjects attended the OPD. Sociodemographic and clinical data were extracted from medical notes and confirmed during the interview. Psychotic symptoms were measured with the Brief Psychiatric Rating Scale (BPRS).²¹ The following 3 mean symptom scores of the scale were used: (1) Positive (conceptual disorganization, suspiciousness, hallucinatory behavior, and unusual thought content); (2) Negative (emotional withdrawal, motor retardation, blunted affect, and disorientation);²² (3) Anxiety and tension.¹⁰ The 17-item Hamilton Depression Rating Scale (HAMD)²³ was used to assess the severity of depressive symptoms.

Extrapyramidal side effects (EPS) were evaluated with the Simpson-Angus Scale of Extrapyramidal Symptoms (SAS)²⁴ and the Barnes Akathisia Rating Scale (BARS).²⁵ The sum scores of these scales were entered in the statistical analysis.²⁶

QOL was defined as “an individual's perception of one's position in life in relation to goals, expectations, standards and concerns in the context of the culture and value systems in which one lives.”²⁷ QOL is a broad concept that consists of 4 main domains: physical health, psychological condition, social relationships, and relationship with the environment. QOL was assessed with the Hong Kong (WHOQOL-BREF-HK) and mainland Chinese (WHOQOL-BREF-CHN) versions of the World Health Organization Quality of Life Schedule-Brief²⁸ in Hong Kong and Beijing, respectively. The 2 QOL schedules are very similar and both cover 4 domains: physical and psychological health, social relationships, and environmental factors.

The data were analyzed using SPSS 13.0 for Windows. The comparison between poor and good sleepers with respect to socio-demographic and clinical characteristics was performed by independent sample t-test, Mann-Whitney U test, and chi-square test as appropriate. To explore the independent associations of each QOL domain with insomnia, analysis of covariance (ANCOVA) was used to adjust for potential confounding effects of other sociodemographic and clinical variables, including categorical variables (sex, marital and employment status, prescription of typical and atypical AP, BZD, and hypnotics) and continuous variables (age, monthly income, age at onset, number of admissions, dose of APs in chlorpromazine equivalents, severity of positive, negative, depressive and anxiety symptoms, and EPS).

Stepwise multiple logistic regression analysis was used to adjust for relevant covariates and to determine the independent correlates of poor sleep. Poor sleep was the dependent variable while the independent variables were those which significantly differed between poor and good sleepers in the univariate analyses.

The one-sample Kolmogorov-Smirnov test was used to check the normality of distribution for continuous variables. The level of significance was set at 0.05 (2-tailed).

RESULTS

A total of 298 patients in Hong Kong and 288 patients in Beijing were invited to participate in the study. Forty-three patients in Hong Kong and 38 in Beijing declined to take part in the study. There was no significant difference between those who agreed and those who declined to take part in the study in terms of age, sex, age at onset, or length of illness. The 2 samples in Hong Kong and Beijing were combined, since they were both drawn from a population of clinically stable outpatients and were matched according to basic sociodemographic and clinical variables. The frequency of all types of insomnia in the past 12 months was 36.0% in the combined Beijing-Hong Kong sample; the rates of DIS, DMS, and EMA were 21.2%, 23.6%, and 11.9%, respectively.

Table 1 shows the sociodemographic and clinical characteristics of the sample as a whole and for poor and good sleepers respectively, and also provides a comparison between the 2 groups in terms of sociodemographic and clinical data and QOL. Poor sleep was significantly associated with advanced age, older age at onset, fewer number of admissions, severity of positive, depressive, and anxiety symptoms, EPS, less frequent use of atypical APs and more frequent use of BZDs and hypnotics. Poor sleepers had significantly poorer QOL in all domains than good sleepers. After controlling for the potential confounding effects of sociodemographic and clinical factors with ANCOVA, a significant difference remained between the 2 groups with regard to the physical QOL domain (F_(1,485) = 6.6, P = 0.01). The differences between poor and good sleepers in the psychological (F_(1,485) = 0.07, P = 0.78), social (F_(1,485) = 0.37, P = 0.55), and environmental domains (F_(1,485) = 0.02, P = 0.90) of QOL did not reach a statistically significant level.

Table 1.

Basic DEmographic and Clinical Characteristics of the Study Sample

	The whole sample		Poor sleepers (n = 182)		Good sleepers (n = 323)		Statisticsb
	N	%	N	%	N	%	X2	df	P
Men	243	48.1	85	46.7	166	51.4	1.0	1	0.3
Married/cohabitating	228	45.1	90	49.5	138	42.7	2.1	1	0.1
Being employed	165	32.7	50	27.5	115	35.6	3.4	1	0.06
On typical AP(s) only	244	48.3	97	53.3	147	45.5	2.8	1	0.09
On atypical AP(s) only	219	43.4	66	36.3	153	47.4	5.8	1	0.01
On BZD	151	29.9	78	42.9	73	22.6	22.7	1	< 0.001
On hypnotics	125	24.8	82	45.1	43	13.3	62.9	1	< 0.001
	Mean	SD	Mean	SD	Mean	SD	T / Z	df	P
Age (year)	43.0	8.4	44.1	8.4	42.3	8.4	2.3	503	0.02
Monthly income (HK$)	2022	2511	1919	2335	2080	2608	−0.4a		0.6
Age at onset (year)	26.8	7.6	27.8	7.4	26.3	7.7	2.1	503	0.03
Number of admission	2.3	2.2	2.1	2.1	2.4	2.2	−2.1a		0.03
CPZeq (mg)	273	188	275	189	271	187	−0.1a		0.8
BPRS positive score	1.4	0.7	1.6	0.8	1.3	0.5	5.4	503	< 0.001
BPRS negative score	1.4	0.5	1.4	0.5	1.3	0.5	1.4	503	0.2
BPRS anxiety score	1.4	0.6	1.6	0.6	1.3	0.5	5.8	503	< 0.001
EPS sum score	0.6	1.6	1.1	2.1	0.4	2.1	−3.3a		0.001
HAMD score	4.4	4.0	7.3	4.4	2.7	2.4	−13.0a		< 0.001
Physical QOL	14.2	2.4	12.8	2.4	14.9	2.1	−10.1	503	< 0.001
Psychological QOL	13.6	2.6	12.4	2.6	14.2	2.4	−7.7	503	< 0.001
Social QOL	13.1	2.6	12.2	2.6	13.6	2.5	−5.9	503	< 0.001
Environmental QOL	13.6	2.2	13.0	2.1	14.0	2.2	−4.9	503	< 0.001

a=Mann-Whitney U test; b=poor sleepers vs good sleepers; AP=antipsychotic drug; BZD=benzodiazepine; EPS=extrapyramidal side effects; QOL=quality of life; HAMD=Hamilton Depression Rating Scale; BPRS=Brief Psychiatric Rating Scale

In stepwise multiple logistic regression analysis, the prescription of atypical APs, BZD and hypnotics, age, age at onset, number of admissions, severity of positive, anxiety, and depressive symptoms and EPS, and all QOL domains were entered as potential contributors to poor sleep. Advanced age, fewer admissions, severity of depressive symptoms and use of hypnotics were independently associated with poor sleep (Table 2).

Table 2.

Factors Associated with Poor Sleep (Multiple Logistic Regression Analysis)

	P value	Odds ratio	95% C.I.
Age	0.007	1.04	1.01–1.07
Number of admissions	0.039	0.89	0.79–0.99
Depressive symptoms	< 0.001	1.56	1.42–1.69
Use of hypnotics	< 0.001	4.67	2.76–7.90

DISCUSSION

The rate of any type of insomnia (36.0%) found was within the range (6% to 76.3%) reported in the general population,^16,29 but lower than the pooled figures in patients with chronic illness including hypertension, diabetes, congestive heart failure, myocardial infarction as well as in depression (50%)³ and schizophrenia (45.4%).¹³ However, the degree of variation in the definitions of insomnia, interview procedures (telephone interviews vs face-to-face interviews) and time-frames (e.g., current, past week, past year, or lifetime) used in previous studies make it difficult for direct comparisons to be made between this study and earlier studies. In a recent epidemiologic survey using the same definition of insomnia and the time-frame as this study, the prevalence of DIS, DMS, EMA, and any type of insomnia in a Chinese population were found to be 7.0%, 8.0%, 4.9%, and 9.2%, respectively (Xiang et al., submitted). These figures are considerably lower than the corresponding 21.2%, 23.6%, and 11.9%, and 36.0% found in this study, which clearly indicates that schizophrenia patients report more disturbed sleep than the general population even if they are clinically stable and live in the community.

This study replicated earlier findings showing that insomnia was significantly associated with depressive^4,13 and anxiety symptoms³⁰ and EPS.³¹ The correlation between atypical APs and lack of sleep disturbances probably reflects the better sleep-promoting effects of this class of drugs than traditional APs.³¹ The association between insomnia and the long-term use of BZD indicates that these drugs are frequently prescribed, but are not necessarily effective, for insomnia.³² According to the UK guidelines for the prescription of BZDs as hypnotics, these agents should only be prescribed for transient or short-term insomnia and should be limited to occasional or intermittent use not exceeding two weeks.³³ Long-term use of BZDs increases the risk of secondary dependence.³⁴ Similarly, the more frequent use of hypnotics by the poor sleepers suggests that these drugs may be taken for insomnia, but with little effect.

Ritsner et al.¹³ found no significant relationship between insomnia and either positive or negative psychotic symptoms in clinically stable schizophrenia patients. In contrast, insomnia was inversely associated with severity of positive symptoms in this study. The small sample size (n = 145) might have limited the power of Ritsner's study.

In multivariate analysis, only advanced age, fewer admissions, severity of depressive symptoms, and use of hypnotics were independently associated with poor sleep. The link between insomnia and advanced age is well known from studies in the general population.^16,35 The association between depressive symptoms and insomnia was expected, as sleep disturbances are core features of depressive illness.³² However, we could offer no explanation for the relationship between insomnia and fewer psychiatric admissions. This factor signals a good prognosis in schizophrenia, and we are not aware of any data suggesting an association between good prognosis and insomnia in schizophrenia. Longitudinal studies may shed light on this unexpected and inexplicable connection.

Poor sleepers had poorer QOL in all domains than good sleepers, and the difference between the two groups in the physical QOL domain remained significant after adjusting for sociodemographic and clinical variables. This suggests that there is an independent association between insomnia and aspects of QOL in schizophrenia patients, though insomnia has not been included as an independent negative contributor in the theoretical models of QOL in psychiatry.^36–41

The strengths of this study are the large, randomly selected and homogeneous sample and the standardized questions on insomnia from DSM-IV, which facilitate direct comparisons with other epidemiologic surveys. However, the results of this study should be interpreted with caution due to certain methodological limitations. First, the study's coverage was confined to clinically stable schizophrenia patients from two major Chinese cities. Its results may not be applicable to other parts of China and to schizophrenia patients with a different clinical condition. Second, the study was cross-sectional, therefore the causality of relationship between insomnia and sociodemographic and clinical variables could not be explored. Third, insomnia was assessed only by self-report, and although we used a relatively simple definition of insomnia, we cannot absolutely rule out recall bias. Fourth, more information on insomnia, such as sleep duration and the level of awareness of sleep hygiene measures, should have been collected and investigated. Finally, the conversion of different APs into chlorpromazine equivalents is a rough approximation at best and lacks a sound scientific basis, particularly for atypical APs.⁴² Using the same conversion standard when comparing the two samples (good vs. poor sleepers) could mitigate this limitation as it may reveal relative trends in prescription practices.⁴³

In conclusion, given the harmful consequences of insomnia coupled with its high rate found in this survey, attempts to address insomnia in schizophrenia patients should be made even if these patients are clinically stable. The inverse association between atypical AP and insomnia warrants the more widespread prescription of this class of agents to promote sleep and improve the QOL of schizophrenia patients. Particular attention should be paid to Chinese schizophrenia patients who are older, present with anxiety and depressive symptoms and experience fewer admissions, because this group of patients is significantly correlated with insomnia. The long-term use of BZDs should be discouraged in clinical practice. Further surveys should be conducted to explore the incidence and characteristics of insomnia in other parts of China, particularly in rural areas. Longitudinal studies should also be carried out to investigate the sociodemographic and clinical predictors of insomnia.

DISCLOSURE STATEMENT

This was not an industry supported study. The authors have indicated no financial conflicts of interest.