Correlates of benzodiazepine use in individuals with insomnia

Sleep Medicine 5 (2004) 457–462 www.elsevier.com/locate/sleep

Original article

Correlates of benzodiazepine use in individuals with insomnia Charles M. Morina,b,*, Lynda Be´langera,b, Franc¸ois Berniera b

a Universite´ Laval, Quebec, Canada Centre de Recherche Universite´ Laval-Robert-Giffard, Quebec, Canada

Received 23 October 2003; received in revised form 16 March 2004; accepted 7 April 2004

Abstract Background and purpose: Although benzodiazepines (BZDs) are commonly used in the treatment of insomnia, there is little information about psychological, health, and sociodemographic correlates associated with their use. Objective: This study examined correlates of benzodiazepine use for sleep in a clinical sample of patients seeking treatment for insomnia at a sleep disorders clinic. Patients and methods: The sample consisted of 97 individuals evaluated at a sleep disorders clinic for a presenting complaint of insomnia. Two groups were formed, including one with 61 patients who had been using BZDs for sleep for an average of 56.6 months (SD ¼ 68.0), and another with 36 patients who had insomnia but were not using BZDs or any other sleep aid. Logistic regressions, adjusting for an age difference between the user and non-user groups, were performed to examine variables associated with BZD use. Independent variables included demographic, medical, insomnia-related and psychological parameters and subjective sleep – wake characteristics. Unadjusted regression analyses were performed to identify factors associated with a pattern of long-term use (. 12 months) within the user group. Results: Age was a significant predictor of BZD use (OR ¼ 1.84, P ¼ 0:0002Þ: Significant age-adjusted predictors of BZD use included perceived insomnia severity (OR ¼ 1.17, P ¼ 0:038Þ; depressive symptoms, (OR ¼ 1.108, P ¼ 0:009Þ and state and trait anxiety symptoms (OR ¼ 1.062, P ¼ 0:016; OR ¼ 1.084, P ¼ 0:005; respectively). Significant predictors of long-term use (.12 months) were age of insomnia onset (OR ¼ 0.951, P ¼ 0:0214Þ; more frequent BZD use (OR ¼ 3.284, P ¼ 0:0221Þ; and higher state-anxiety (OR ¼ 1.106, P ¼ 0:0471Þ: Conclusions: Age, psychological variables and perceived sleep disturbances severity, are associated with BZD use in patients with insomnia. q 2004 Elsevier B.V. All rights reserved. Keywords: Insomnia; Benzodiazepines; Hypnotics; Long-term use; Psychosocial factors

1. Introduction Insomnia is a widespread and burdensome health problem that has been associated with daytime fatigue, impaired functioning, and with mood disturbances [1]. Pharmacological management represents the most widely used treatment option for both acute and chronic insomnia [2 – 5]. Among the different drug classes available for the treatment of insomnia, benzodiazepines (BZDs) are still among the most widely prescribed medications [6 –9]. BZDs are considered efficacious and safe in the short term management of insomnia [12]. However, their long-term use is controversial because of potential residual effects [10,13] and risks of tolerance * Corresponding author. Tel.: þ 418-656-3275; fax: 418-656-5152. E-mail address: [email protected] (C.M. Morin). 1389-9457/$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.sleep.2004.04.001

[14] and dependence [15]. According to standard prescription guidelines, their use as sleep promoting agents should be restricted to a maximum of 4 weeks [12,16]. However, according to some reports [8,11] a considerable proportion of individuals use BZDs for sleep on a chronic basis (more than a year). Factors associated with BZD use in patients diagnosed with insomnia are not well known. Studies examining BZD use (mostly for sleep and anxiety) in the general population and in medical patients show that use is more frequent in women and older individuals [2,7,17,18], in individuals who are separated or widowed, in those with a lower income and a lower level of education [18,19]. In two population-based studies, one conducted in the United States and the other in France and Canada, that have specifically examined factors associated with use of sleep-aids in individuals complaining

458

C.M. Morin et al. / Sleep Medicine 5 (2004) 457–462

of insomnia [4,8], female gender, older age (50 –79 years old) and presence of anxiety and depression symptoms were associated with hypnotic use. Health problems and psychological distress were associated with use of sleep medications in one of these studies [4]. However, use of BZD hypnotics per se was not specifically examined in the former study, and in neither study were long-term users differentiated from short-term users. While these reports provide useful information on factors associated with use of sleep medications, additional studies are needed to further document the factors associated specifically with BZD use and long-term use (. 12 months) in patients with insomnia. The aim of this study was to examine predictors of BZD use for sleep among individuals with insomnia and, among these users, identify correlates of long-term use (. 12 months).

2. Method 2.1. Participants Participants were 97 patients (53 women and 44 men) who were evaluated for a primary complaint of insomnia at a sleep disorders center of a teaching hospital. The mean age of patients was 43.7 years (SD ¼ 14.8), and the majority were married (59%) and working (76.3%). Mean level of education was of 15.18 years (SD ¼ 3.2). Patients were either self- or physician-referred. They reported an average duration of insomnia of 10.6 years (SD ¼ 11.34). They were selected from a data base of 345 patients evaluated for insomnia over a period of 7 years at the sleep disorders center. Patients were included in the study if they were 18 years old or older and presented insomnia as the main complaint. They were excluded if they used medications other than BZDs to promote sleep (e.g. antidepressants), if insomnia was secondary to another medical (pain) or psychiatric (depression) condition, or if the main diagnosis was another sleep disorder such as a circadian rhythm disorder, sleep apnea, periodic limb movements, or parasomnia. Patients who used BZDs for sleep were compared to nonusers on demographic, psychological, medical and sleeprelated variables. The non-user group was composed of 36 patients (16 women, 20 men; mean age ¼ 34.8, SD ¼ 10.97) who had either never used BZDs ðn ¼ 29Þ or had used them ðn ¼ 7Þ in the past, but not in the twelve months prior to evaluation (average of 50 months since last use, range 12– 96 months). The user group was composed of 61 patients (37 women, 24 men; mean age ¼ 48.9, SD ¼ 14.3) who had been using BZDs for an average of 56.6 months (SD ¼ 68.0; range ¼ 0.5– 300). Mean frequency of use was of 4.1 times per week (SD ¼ 2.5, range ¼ 0.5 –7). Intermediate-acting agents were used by 63.9% (39/61) of the sample, while 26.2% (16/61) used long acting and 32.8% (20/61) used short acting agents.

Most patients (72.1%, 44/61) used one BZD, 24.6% (15/61) used two, one patient used three, and one patient used four. All participants included in this study were not using any other type of sleep aids (e.g. over-the-counter sleep aids, alcohol or other prescription drugs). 2.2. Measures Insomnia Interview Schedule (IIS) [21]. This semistructured interview was used to get a detailed sleep history and evaluation of the insomnia complaint. It yields information on type of insomnia complaint (e.g. difficulties initiating or maintaining sleep; sleep – wake schedule); use of sleep-promoting medications and other sleep aids (e.g. alcohol); history of sleep problem (e.g. precipitating factors); bedroom environment; diet and exercise habits; presence of other sleep disorders’ symptoms (e.g. snoring as in apnea); medical history and medication use, and presence of current or past psychopathology. Insomnia Severity Index (ISI) [21]. This seven-item selfreport measure yields information on patients’ perception of their sleep difficulties. It assesses the perceived severity of insomnia (initial, middle, late), interference with daytime functioning, satisfaction with sleep. The degree to which individuals believe their insomnia is noticeable to others, and the degree of concerns/worries about their sleep difficulties. Items are rated on 5-point (0 –4) Likert-type scales. The ISI shows good internal consistency ða ¼ 0:88Þ and appropriate test –retest reliability ðr ¼ 0:65Þ [22,23]. Total scores range from 0 to 28 with the following interpretation guidelines: 0 –7, absence of clinically significant insomnia; 8 – 14, sub-threshold insomnia; 15 – 21, moderate insomnia and, 22 – 28, severe insomnia. Sleep diary [21]. This self-report measure is completed daily by patients, preferably at-or around breakfast time. It reflects in number the patients’ perceptions of their previous night’s sleep. It probes for information about naps, medication use, bedtime, sleep latency, number of awakenings, length of awakenings, time of last awakening, rising time, feeling upon rising (1 ¼ exhausted, 5 ¼ rested) and overall sleep quality (1 ¼ very agitated, 5 ¼ very deep). A global index of subjective sleep quality was obtained through mean scores of the ‘feeling upon rising’ and overall ‘quality of sleep’ scales. Analyses were performed on sleep latency, time awake after initial sleep onset, total sleep time, total wake time, number of awakenings and sleep efficiency. Although it does not reflect absolute values obtained from Polysomnography (PSG), the sleep diary represents a reliable and valid index of the insomnia complaint [24]. It is also a convenient, inexpensive, and widely used method for assessing insomnia [25,26]. Mood. The Beck Depression Inventory (BDI) [27] was used to assess symptoms of depression and the State-Trait Anxiety Inventory (STAI) [28] was used for assessing anxiety symptoms. Psychometric properties of these two instruments are well documented.

C.M. Morin et al. / Sleep Medicine 5 (2004) 457–462

Procedures. All participants completed a standard clinical evaluation of insomnia, which was conducted by the first author (CM) or by clinical psychology fellows or interns under his supervision. The assessment included a semi-structured evaluation using the IIS [18] and a psychological screening evaluation using selected questions from the Structured Clinical Interview for DSM-III-R (SCID) [29]. A medical history was obtained and a physical examination was performed by a physician when relevant information was unavailable from the referring physician. PSG was used to rule out the presence of sleep apnea or periodic limb movements when deemed necessary. Participants were asked to complete several questionnaires (e.g. sleep survey, symptoms checklist, psychological inventories) and monitor their sleep with a daily diary for a period of two weeks prior to the initial clinical evaluation. Selfreport questionnaires and sleep diaries were mailed to participants upon making their appointment and they were instructed to bring those materials back at the initial visit. 2.3. Statistical analyses Univariate logistic regressions were performed in order to identify predictors of BZD use. The following variables were examined: (1) demographic variables: age, gender, education, marital status, and occupation; (2) clinical insomnia variables and sleep –wake parameters: insomnia duration, age of insomnia onset, insomnia severity (ISI scores) and type of precipitating factors (health-related, family, work/school, psychological problems), sleep onset latency, time awake after sleep onset, total sleep time, total wake time, number of awakenings, sleep efficiency and sleep quality; (3) psychological and medical variables: depression symptoms (BDI score), anxiety symptoms (STAI state and trait scores), presence of current psychopathology, presence of a medical illness, number of medical illnesses. Given that BZD users were significantly older than non-users, a second set of logistic regressions (controlling for age) was also performed. As a secondary aim of this study was to examine long-term use, individuals using BZDs for more than 12 months ðn ¼ 50Þ were further compared to those using BZDs for less than 12 months ðn ¼ 11Þ: Age was not controlled for since it was not significantly different between those two subgroups.

3. Results Table 1 shows descriptive statistics for insomnia, sleep, medical, and psychological variables. Tables 2 and 3 show logistic regression results for BZD use and use longer than 12 months, respectively. Univariate logistic regressions showed that age (OR ¼ 1.84, P ¼ 0:0002Þ was a significant predictor of BZD usage, with older age associated with increased risk to use BZDs as a sleep aid. Other significant predictors were

459

Table 1 Descriptive characteristics of BZD users and non-users Variable*

Users

Non-users

Insomnia duration (years) Age of insomnia onset (years) ISI score SOL (min.) WASO (min.) TST (min.) TWT (min.) FNA SE (%) SQ

10.6 (11.6) 38.4 (17.5) 20.2 (4.4) 54.0 (38.7) 48.6 (46.9) 335.7 (86.0) 155.2 (87.5) 1.7 (1.1) 68.6 (15.4) 2.7 (0.7)

10.7 (11.0) 24.8 (11.1) 17.8 (3.9) 46.8 (38.5) 43.12 (50.1) 331.3 (93.3) 135.0 (96.3) 1.8 (1.0) 71.2 (18.6) 2.6 (0.5)

Medical illness (%) None At least one Number of medical illness BDI score State anxiety Trait anxiety

37.3 62.7 1.0 (1.0) 14.1 (8.3) 45.96 (9.6) 47.6 (8.9)

76.0 24.0 0.5 (1.1) 10.5 (6.2) 42.1 (11.7) 42.2 (10.5)

Psychopathology (%) None At least one

48.3 51.7

54.5 45.5

Precipitating factor (%) Undetermined Health Family Work/school Psychological Other

18.0 19.7 27.9 18.0 11.5 4.9

37.1 5.7 31.4 17.1 5.7 2.9

*Means and standard deviations are presented unless specified otherwise. Abbreviations: SOL, sleep onset latency; WASO, wake after sleep onset; TST, total sleep time; TWT, total wake time; FNA, frequency of night awakening; SE Sleep efficiency; SQ, sleep quality; BDI, Beck Depression Inventory; min., minutes.

marital status, age of insomnia onset, insomnia severity (ISI total score), type of precipitating factor (health-related), presence of a medical illness, higher depression scores, and higher state and trait anxiety scores. When these later variables were adjusted for age, the following predictors remained significant: perceived insomnia severity (OR ¼ 1.17, P ¼ 0:038Þ; depressive symptoms, (OR ¼ 1.108, P ¼ 0:009Þ and state and trait anxiety symptoms (OR ¼ 1.062, P ¼ 0:016; OR ¼ 1.084, P ¼ 0:005; respectively). Thus, higher perceived insomnia severity (as reflected by ISI scores) was associated with increased risks of BZD use, as were more severe anxiety and depressive symptoms. Subjective sleep –wake parameters as measured by the daily diary, as well as the presence of a current psychopathology, were not significantly associated with BZD use. The following variables predicted long-term BZD use (. 12 months): age of insomnia onset (OR ¼ 0.951, P ¼ 0:0214Þ; frequency of BZD use (OR ¼ 3.284, P ¼ 0:0221Þ; and state and anxiety scores (OR ¼ 1.106, P ¼ 0:0471Þ: Increased risks of prolonged BZD use was associated with

460

C.M. Morin et al. / Sleep Medicine 5 (2004) 457–462

Table 2 Logistic regression analyses results predicting BZD usage Predictor

Age Marital status Presence of a Physical illness Precipitating factor (health-related) Age insomnia onset Insomnia severity BDI scores State anxiety Trait anxiety

Unadjusted analysis

Adjusted analysis

OR

95% CI of OR

P-value

OR

95% CI of OR

P-value

1.91 0.14 5.33 7.0 1.06 1.15 1.07 1.04 1.063

1.05–1.140 0.05–0.38 1.81–15.69 1.30–38.75 1.03–1.10 1.01–1.31 1.01–1.14 0.99–1.08 1.01–1.12

0.0001 0.0001 0.0024 0.0238 0.0004 0.0369 0.0334 0.095 0.0165

0.33 2.134 5.59 1.03 1.17 1.11 1.06 1.08

0.11– 1.06 0.61– 7.48 0.87– 35.88 0.98– 1.08 1.01– 1.35 1.03– 1.20 1.01– 1.12 1.02– 1.15

0.0627 0.2362 0.0696 0.303 0.0378 0.0091 0.0164 0.0052

an onset of insomnia occurring later in life, with more frequent BZD use, and with more severe state-anxiety symptoms.

4. Discussion Results of this study show that an older age, higher perceived insomnia severity, and more severe depressive and anxiety symptomatology were associated with increased risks of benzodiazepine usage. The presence of health problem was also related, but this association was no longer significant after controlling for the effect of age. Subjective sleep –wake parameters, as measured by a daily diary, and the presence of psychopathology were not associated with BZD usage. Predictors of long-term use (. 12 months) included a later age of insomnia onset, more frequent BZD use, and higher state-anxiety. The finding that older age is associated with BZD use among a clinical sample of insomnia patients is consistent with data from population-based studies on BZD use in general [2,7,18,20] and with use of sleep aids in insomniacs [4,8]. This increased use of BZDs and other psychotropic medications with age, which is a consistent finding in the literature [8,11], may be explained by more frequent visits with health-care providers, a greater likelihood among physicians to prescribe such medications to elderly patients [11,18], or closer scrutiny of medical prescriptions for the older population. Another plausible explanation is that individuals who experience insomnia in later life may attribute the problem to the natural aging process and believe that medication is the only intervention that can slow the process. Regardless of the explanation, closer monitoring of BZD use may be necessary among older adults, given that this class of medication has been associated with increased risks of falls and hip fractures [30 –32] and road accidents among the elderly [33,34]. No demographic variable other than age was associated with BZD use. The lack of significant gender difference, for instance, is not consistent with prior reports of BZD use in

the general population [2,7,18] and in insomniacs [4,8], which have generally found a higher rate of use among women. Also, unlike prior studies that have reported greater use in divorced and widowed individuals [18,20], this study did not show any difference in marital status when age was controlled for. The different source of recruitment for the present study sample (i.e. sleep disorders clinic) and previous population-based studies might account for these differences [35]. None of the subjective sleep – wake parameters or insomnia duration was associated with BZD use. This finding is surprising, as one might expect that more severe and more chronic sleep disturbances would increase the likelihood of using BZDs. On the other hand, perceived insomnia severity was significantly associated with BZD use. This observation is not necessarily contradictory, as perceived insomnia severity was based on a retrospective and global evaluation, whereas sleep –wake parameters were estimated with prospective, daily sleep diaries. Interestingly, there was no significant difference between BZD users and non-users on any of the subjective sleep – wake parameters—a possible indication that BZD therapy was no longer effective for this group of patients seeking additional insomnia treatment. However, the retrospective study design and the lack of baseline data prior to initiation of BZD therapy, as well as the lack of PSG data, precludes any definite conclusion on this issue [24,40,41]. Increased anxiety and depressive symptoms were also associated with increased risks of BZD usage. While this finding is concordant with prior reports [36,37], it is unclear whether psychological symptoms are the result of chronic Table 3 Regression analyses predicting BZD usage above 12 months Predictor

OR

95% CI of OR

P-value

Age of insomnia onset Frequency of BZD use State anxiety

0.95 3.28 1.11

0.91–0.99 1.19–9.091 1.0–1.22

0.0214 0.0221 0.0471

C.M. Morin et al. / Sleep Medicine 5 (2004) 457–462

BZD use or are part of the clinical symptomatology that led to patients’ BZD use in the first place. These symptoms may also be related to health problems that were precipitating factors of insomnia. When age was controlled for, they were no longer significant. This result, which is inconsistent with other reports [4,38,39], might be explained by the contextual differences among the clinical case series and population-based studies. The present results must be interpreted with caution for several reasons. First, the sample was relatively small compared to population-based epidemiological surveys. The small sample size may also have reduced statistical power. The lack of objective PSG sleep measures, precluding any conclusion about differences between BZD users and non-users, is another limitation. While this study provides some evidence of a number of insomnia and psychological variables associated with increased risks of BZD use, larger prospective studies using PSG measures are warranted to expand these findings and more precisely identify determinants of hypnotic medication usage among individuals with insomnia.

References [1] Ohayon MM. Epidemiology of insomnia: What we know and what we still need to learn. Sleep Med Rev 2002;6:97–111. [2] Dunbar GC, Perera MH, Jenner FA. Pattern of benzodiazepine use in Great Britain as measured by a general population survey. Br J Psychiatry 1989;155:836 –41. [3] Lacks P, Powlishta K. Improvement following behavioral treatment for insomnia: Clinical significance, long-term maintenance, and predictors of outcome. Behav Ther 1989;20:117–34. [4] Mellinger GD, Balter MB, Uhlenhuth EH. Insomnia and its treatment: prevalence and correlates. Arch Gen Psychiatry 1985; 42:225–32. [5] Quera-Salva M-A, Orluc A, Goldberg F, Guilleminault C. Insomnia and the use of hypnotics: study of a French population. Sleep 1991;14: 386–91. [6] Hohagen F, Ka¨ppler C, Schramm E, Rink K, Weyerer S, Riemann D, Berger M. Prevalence of insomnia in elderly general practice attenders and the current treatment modalities. Acta Psychiatr Scand 1994;90:102– 8. [7] Mant A, Mattick RP, de Burgh S, Donnelly N, Hall W. Benzodiazepine prescribing in general practice: dispelling some myths. Fam Pract 1995;12:37 –43. [8] Ohayon MM, Caulet M. Psychotropic medication and insomnia complaints in two epidemiological studies. Can J Psychiatry 1996;41: 457–64. [9] Walsh JK, Schweitzer PK. Ten-year trends in the pharmacological treatment of insomnia. Sleep 1999;22:371–5. [10] Kripke DF, Klauber MR, Wingard DL, et al. Mortality hazards associated with prescription hypnotics. Biol Psychiatry 1998;43: 687–93. [11] Simon GE, VonKorff M, Barlow W, Pabiniak C, Wagner E. Predictors of chronic benzodiazepine use in a health maintenance organization sample. J Clin Epidemiol 1996;49:1067–73. [12] National Institutes of Health (NIH), Drugs and insomnia: The use of medications to promote sleep. J Am Med Assoc 1984;18: 2410–4.

461

[13] Moskowitz H, Linnoila M, Roehrs T. Psychomotor performance in chronic insomniacs during 14-day use of Flurazepam and Midazolam. J Clin Psychopharmacol 1990;10:44S –55S. [14] Soldatos CR, Dikeos DG, Whitehead A. Tolerance and rebound insomnia with rapidly eliminated hypnotics: a meta analysis of sleep laboratory studies. Int Clin Psychopharmacol 1999;14: 287–303. [15] Lader M. Biological processes in benzodiazepine dependence. Addiction 1994;89:1413– 8. [16] National Institutes of Health, Consensus development conference statement: the treatment of sleep disorders of older people. Sleep 1991;14:169 –77. [17] Egan M, Morides Y, Wolfson C, Monette J. Long-term continuous use of benzodiazepines by older adults in Quebec: prevalence, incidence and risk factors. J Am Ger Soc 2000;48:811 –6. [18] Laurier C, Dumas J, Gre´goire J-P. Factors related to benzodiazepine use in Quebec: a secondary analysis of survey data. J Pharmacoepidemiol 1992;2:73–86. [19] Tamblyn RM, McLoead PJ, Abrahamovitz M, Monette J, Gayton DC, Berkson L, et al. Questionable prescribing for elderly patients in Quebec. Can Med Assoc J 1994;150:1801–9. [20] Swartz M, Landerman R, George LK, Melville M, Blazer D, Smith K. Benzodiazepine anti-anxiety agents: prevalence and correlates of use in a Southern community. Am J Public Health 1991;81: 592–6. [21] Morin CM. Insomnia: psychological assessment and management. New York, NY: Guilford; 1993. [22] Blais FC, Gendron L, Mimeault V. Evaluation de l’insomnie: Validation de trois questionnaires. L’Ence´phale 1997;23:447–53. [23] Bastien CH, Vallie`res A, Morin CM. Validation of the Insomnia Severity Index as a clinical outcome measure for insomnia research. Sleep Med 2001;2:297– 307. [24] Coates TJ, Killen JD, George J, Marchini E, Silverman S, Thoresen C. Estimating sleep parameters: a multitrait-multimethod analysis. J Consult Clin Psychol 1982;50:345–52. [25] Lacks P, Morin CM. Recent advances in the assessment and treatment of insomnia. J Consult Clin Psychol 1992;60:586–94. [26] Lichstein KL, Riedel BW. Behavioral assessment and treatment of insomnia: A review with an emphasis on clinical application. Behav Ther 1994;25:659 –88. [27] Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:561–71. [28] Spielberger CD, Gorshuch RL, Lushene RE. Manual for the StateTrait Anxiety Inventory. Palo Alto, CA: Consulting Psychologist; 1970. [29] American Psychiatric Association, Diagnostic and statistical manual of mental disorders, 3rd ed. revised. Washington, DC: Author; 1987. [30] Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in older people: a systematic review and meta-analysis: psychotropic drugs. J Am Ger Soc 1999;47:30–9. [31] Ray WA, Griffin MR, Schaffner W, Bauth DK, Melton LJ. Psychotropic drug use and the risk of hip fracture. N Engl J Med 1987;316:363–9. [32] Ray WA, Thapa PB, Gideon P. Benzodiazepines and the risk of falls in nursing home residents. J Am Ger Soc 2000;48:682–5. [33] Ray WA, Fought RL, Decker MD. Psychoactive drugs and the risk of injurious motor vehicle crashes in elderly drivers. Am J Epidemiol 1992;136:873 –83. [34] Hemmelgarn B, Suissa S, Huang A, Boivin JF, Pinard G. Benzodiazepine use and the risk of motor vehicle crash in the elderly. J Am Med Assoc 1997;278:27 –31. [35] Bixler EO, Kales A, Soldatos CR, Kales JD, Healy B. Prevalence of sleep disorders in the Los Angeles metropolitan area. Am J Psychiatry 1979;136:1257–62. [36] Neiman J, Persson H, Bergman H. Early benzodiazepine withdrawal in benzodiazepine-dependent subjects: a pilot study. Nord J Psychiatry 1994;48:19 –25.

462

C.M. Morin et al. / Sleep Medicine 5 (2004) 457–462

[37] Rickels K, Schweizer E, Case WG, Greenblatt DJ. Long term therapeutic use of benzodiazepine: effects of abrupt discontinuation. Arch Gen Psychiatry 1990;47:899–907. [38] Brady KT, Johnston AL, Cunningham M, Malcom R. Profiles of hospitalized benzodiazepine abusers. J Psychoactive Drugs 1991;23:71–2. [39] Busto UE, Romach MK, Sellers EM. Multiple drug use and psychiatric comorbidity in patients admitted to the hospital with severe benzodiazepine dependence. J Clin Psychopharmacol 1996;16:51–7.

[40] Vanable PA, Aikens JE, Tadimeti L, Caruana-Montaldo B, Mendelson WB. Sleep latency and duration estimates among sleep disorder patients: variability as a function of sleep disorder diagnosis, sleep history and psychological characteristics. Sleep 2000;23:71–9. [41] Edinger JD, fins AI. The sleep distribution an clinical significance of sleep time misperception among insomniacs. Sleep 1995;18: 232 –9.