Abstract
Purpose
The aim of this study was to evaluate the prevalence and clinical correlates of impaired sleep quality and excessive daytime sleepiness among patients receiving methadone for opioid use disorder (OUD).
Methods
Patients receiving methadone (n = 164) completed surveys assessing sleep quality (Pittsburgh Sleep Quality Index [(PSQI]), daytime sleepiness (Epworth Sleepiness Scale [ESS]), and related comorbidities. We used bivariate and multivariable linear regression models to evaluate correlates of sleep quality and daytime sleepiness.
Results
90% of patients had poor sleep quality (PSQI > 5), and the mean PSQI was high (11.0 ± 4). 46% reported excessive daytime sleepiness (ESS > 10). In multivariable analyses, higher PSQI (worse sleep quality) was significantly associated with pain interference (Coefficient = 0.40; 95% CI = 0.18–0.62; β = 0.31), somatization (Coefficient = 2.2; 95% CI = 0.75–3.6; β = 0.26), and negatively associated with employment (Coefficient = −2.6; 95% CI = −4.9- −0.19; β = −0.17). Greater sleepiness was significantly associated with body mass index (Coefficient = 0.32; 95% CI = 0.18–0.46; β = 0.33), and there was a non-significant association between sleepiness and current chronic pain (Coefficient = 1.6; 95% CI = 0.26–3.5; β = 0.13; p-value = 0.09).
Conclusions
Poor sleep quality and excessive daytime sleepiness are common in patients receiving methadone for OUD. Chronic pain, somatization, employment status, and obesity are potentially modifiable risk factors for sleep problems for individuals maintained on methadone. People with OUD receiving methadone should be routinely and promptly evaluated and treated for sleep disorders.
Keywords: Opioid, methadone, sleep quality, sleepiness, pain, somatization
1. INTRODUCTION
Sleep is critical for preserving normal brain function, and impaired sleep quality is known to disrupt many of the same neural processes altered in opioid use disorder (OUD) and other addictive disorders. For example, sleep deprivation is known to alter activity of dopaminergic brain reward systems, a key mediator of addiction and drug craving.[1, 2] Thus, impaired sleep quality could be an important risk factor in causing relapse and continued drug use in people with existing drug use disorders.[3] Impairment in sleep quality and its associated conditions including insufficient sleep duration, sleep-disordered breathing, circadian rhythm disruption, and excessive daytime sleepiness can all adversely affect otherwise healthy individuals. The effects of impaired sleep quality on people with OUD may be amplified and therefore have a clinically significant impact on OUD treatment outcomes, providing an opportunity to implement sleep interventions that improve both sleep and addiction-related outcomes. Given the similarities of neurocognitive deficits induced from sleep impairment and OUD, the effects of both conditions may be reinforced. It is therefore critical to more fully understand the nature of sleep impairments deficiency in people with OUD to provide potential targets for interventions.
Prior literature has identified a high prevalence of sleep disturbances in patients receiving methadone for OUD, reaching nearly 85% in some studies [4]. Circadian rhythm misalignment and changes in sleep architecture result in lower total sleep time, less slow wave sleep, less REM sleep, and more frequent nighttime awakenings in patients receiving methadone compared with healthy controls.[5, 6] Both obstructive and central sleep apnea are common [7], and the risk of central sleep apnea in particular is elevated in people receiving opioids chronically [8]. Among many factors contributing to impaired sleep quality in OUD patients, pain and comorbid psychiatric conditions may be particularly important [9].
The aim of this study was to determine the prevalence and correlates of impaired sleep quality and sleepiness among people with OUD receiving methadone treatment. Correlates included demographics, psychiatric comorbidities, and pain. Understanding the nature of sleep disorders in people with OUD might lead to development of novel sleep-related interventions and harm reduction strategies in this population to improve OUD outcomes.
2. METHODS
2.1. Patient Population, Setting, and Procedures
Participants were recruited between January 2014 and March 2015 by fliers posted at three different urban opioid treatment facilities operated by the APT Foundation, a Connecticut-based not-for-profit community organization, affiliated with the Yale School of Medicine. The APT Foundation, one of the largest providers of opioid agonist treatment in Southern New England, uses an “open-access model” whereby eligible patients are enrolled rapidly in treatment, irrespective of their ability to pay, and are provided real-time access to multiple voluntary drop-in treatment options[10]. Prospective participants needed to be currently receiving methadone at an APT clinic and English-speaking to be eligible for the study. There were no other exclusions for study participation. Participants were remunerated $15 for their study participation. The study was approved by the APT Foundation Board of Directors and Human Investigations Committee at the Yale School of Medicine.
2.2. Survey instruments
Participants provided information on basic demographics, duration of methadone treatment, current methadone dose, history of hypertension, height, and weight. Subjective sleep quality over the preceding 30 days was measured using the Pittsburgh Sleep Quality Index (PSQI) [11]. This tool provides a global score (0–21) that is the sum of seven component scores (0–3) of subjective sleep quality. As in prior reports, a global PSQI score of > 5 was used to identify respondents with poor sleep quality [12]. Subjective daytime sleepiness was measured by the Epworth Sleepiness Scale (ESS) [13]. This tool asks about the chance of dozing (0–3) while engaged in eight common situations (total score of 0–24). Higher scores represent increasing daytime sleepiness, with ESS > 10 defined as excessive daytime sleepiness [14]. The PSQI and ESS have clinical application to patients with suspected or confirmed sleep disorders. Risk of sleep apnea was assessed by the Berlin Questionnaire.[15] This is a validated instrument to predict sleep apnea with 89% positive predictive value.
The Brief Pain Inventory (BPI)[16] is a 11-item self-report measure that uses 0 to 10 numerical rating scales to measure pain intensity and pain interference (i.e., decrement in functioning attributable to pain). Participants completed the Brief Symptom Inventory (BSI), a well-validated and widely used 18-item self-reported scale that measures psychological distress during the past 7-days [17]. The BSI-18 contains 3 subscales (Somatization, Anxiety, Depression) as well as an overall scale (Global Severity Index). Using gender-specific norms from the scoring manual, summed subscale scores were transformed into T-scores. Caseness or clinical significance on the subscales or overall scale corresponds to a T-score of 63 or higher. The Perceived Stress Scale (PSS) is a widely used, 10-item self-administered measure of perceived stress in the past month [18]. Participants rate each item on a 5-point scale ranging from 0 (Never) to 4 (Very Often). High scores indicate higher perceived stress. The UCLA Loneliness Scale (Version 3) is a 20-item measure subjective feelings of loneliness and social isolation [19]. Participants rate each item on a 5-point scale ranging from 1 (Never) to 4 (Often). Higher scores indicate greater loneliness. The Sexual Desire Inventory (SDI) is a widely used, 14-item measure of sexual desire.[20] Participants rate the 10 desire items on a 9-point scale ranging from 0 (no desire) to 8 (strong desire). The four frequency items are scored from 0 (not at all) to 7 (more than once a day). Higher scores indicate higher sexual desire. Compared with an online sample, the SDI scores of study patients were numerically lower [21].
2.3. Statistical Analysis
Descriptive analyses characterized demographics, clinical and treatment characteristics, sleep quality, and sleepiness of the population. We performed linear regression modeling to determine correlates of the two primary study outcomes: (1) impaired sleep quality (measured by PSQI) and (2) daytime sleepiness (measured by ESS). Each independent variable was first tested using bivariate regression. A hierarchical multivariable regression model was then performed using variables from the bivariate analyses with associations of p ≤ 0.1, and additionally included key demographic characteristics (age, sex, race, education, and housing stability). Multicollinearity was assessed by calculating correlation coefficients between related groups of variables (i.e., demographics, pain, and psychiatric). If a high degree of correlation (i.e., r > 0.3) was noted between variables of a given group, only the variable with the strongest relationship in the bivariate model was included in the multivariate model. Data were analyzed using Stata SE Version 15 statistical software (StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC).
3. RESULTS
Demographic characteristics of the study population are summarized in Table 1. The mean age was 44 ± 10 years. The sample was majority Caucasian (59%), and male (59%); 76% had at least a high school level of education, 12% were married, and 10% had any employment. The mean methadone dose was 80 ± 31 mg, and the median duration of treatment was 24 months (IQR: 8–48 months).
Table 1:
Demographics
| Demographics (n = 164) | |
|---|---|
| Age, mean (SD) | 44 (10) |
| Female, No. (%) | 67 (41) |
| Hispanic, No. (%) | 18 (11) |
| Race, No. (%) | |
| American Indian or Alaskan native | 2 (1.2) |
| Asian | 2 (1.2) |
| Black or African American | 48 (29) |
| White/Caucasian | 96 (59) |
| Multiracial or other | 10 (6.1) |
| Highest Education, No. (%) | |
| 8th grade or less | 6 (4) |
| Some high school | 33 (20) |
| High school graduate / GED | 74 (45) |
| Some college | 26 (16) |
| 4-year college graduate, vocational school, or higher | 25 (15) |
| Full time employment, No. (%) | 7 (4) |
| Any employment, No. (%) | 16 (9.8) |
| Homeless/shelter past month, No. (%) | 28 (17) |
| Any history of homelessness, No. (%) | 112 (68) |
| Duration of treatment (mo), med (IQR) | 24 (8-48) |
| Methadone dose (mg), mean (SD) | 80 (31) |
| BMI (kg/m2), mean (SD) | 27.9 (6.4) |
Sleep characteristics are described in Table 2. The vast majority of respondents had poor sleep quality (90%) as measured by PSQI (11.0 ± 4). 46% reported excessive daytime sleepiness (mean ESS 11±7). Short sleep duration (57% with less than 7 hours slept per night), sleep latency longer than 30 minutes (37%), and frequent early awakenings (42%) were all common. Sleep efficiency was low (78%±28%). Despite these findings, a majority subjectively reported fairly good or very good sleep (56%) within the PSQI questionnaire.
Table 2:
Sleep Characteristics
| Sleep Characteristics (n = 164) | |
|---|---|
| Sleep Quality | |
| Pittsburgh Sleep Quality Index, Mean (SD) | 11.0 (4.2) |
| Poor Sleep Quality (PSQI > 5), No. (%) | 145 (90) |
| Good Sleep Quality (PSQI ≤ 5), No. (%) | 17 (10) |
| Overall Subjective Sleep Quality Rating | |
| Very bad, No. (%) | 29 (17.8) |
| Fairly bad, No. (%) | 43 (26.4) |
| Fairly good, No. (%) | 76 (46.6) |
| Very good, No. (%) | 15 (9.2) |
| Bedtime | |
| 6PM-9PM, No. (%) | 22 (13.4) |
| 9PM-1 AM, No. (%) | 120 (73.2) |
| 1AM or later, No. (%) | 22 (13.4) |
| Waking time | |
| 3AM-6AM, No. (%) | 49 (29.9) |
| 6AM-9AM, No. (%) | 104 (63.4) |
| 9AM or later, No. (%) | 11 (6.7) |
| Total sleep time < 7 hours, No. (%) | 93 (57) |
| Frequent sleep onset > 30 minutes, No. (%) | 60 (37) |
| Frequent early awakenings, No. (%) | 69 (42) |
| Sleep Efficiency, Mean (SD) | 78 (28) |
| Sleepiness | |
| Epworth Sleepiness Scale, Mean (SD) | 10.6 (6.7) |
| Sleepy (ESS > 10), No. (%) | 75 (46) |
| Sleep Disordered Breathing | |
| High Risk of Sleep Apnea, No. (%) | 67 (41) |
| Frequent snoring, No. (%) | 53 (32) |
| Frequent witnessed apneas, No. (%) | 17 (10) |
Comorbidities are summarized in Table 3. A majority (54%) reported current chronic pain, and nearly 70% reported a lifetime history of chronic pain. Clinically significant somatization (51%), anxiety (37%), and depression (38%) were common. Subjective pain interference with sleep was significant (6.2± 3.2). Medically, 26% had hypertension and 31% were obese (i.e., BMI ≥30).
Table 3:
Comorbidities
| Comorbidities (n = 164) | |
|---|---|
| Psychiatric | |
| History of chronic pain, No. (%) | 114 (69.5) |
| Current Chronic Pain No. (%) | 89 (54.2) |
| Pain interference with sleep (mean, SD) | 6.2 (3.2) |
| Additional use of sedating analgesics, No. (%) | 16 (9.8) |
| BSI Somatization, mean (SD) Clinically significant, No. (%) |
60.0 (10.1) 83 (50.6) |
| BSI Anxiety, mean (SD) Clinically significant, No. (%) |
58.2 (12.1) 63 (37.2) |
| BSI Depression, mean (SD) Clinically significant, No. (%) |
57.7 (11.5) 62 (37.8) |
| Global severity index, mean (SD) Clinically significant, No. (%) |
60.0 (11.3) 72 (43.9) |
| Perceived stress, mean (SD) | 21.3 (5.8) |
| Social isolation, mean (SD) | 47.5 (9.6) |
| Medical | |
| Hypertension, No. (%) | 42 (26) |
| Obesity (BMI ≥ 30), No. (%) | 51 (31) |
| Physical Functioning (SF-12), mean (SD) | 46 (11) |
| Sexual desire scale, mean (SD) | 51.8 (20.8) |
The correlates of sleep quality impairment (PSQI) are summarized in Table 4. In the bivariate model, the following variables were positively associated with higher PSQI (worse sleep quality): history of homelessness (Coefficient = 1.5; 95% CI = 0.11–2.9; β = −0.17); pain variables including history of chronic pain, current chronic pain, and pain interference with sleep (Coefficient = 0.49; 95% CI = 0.29–0.70; β = 0.38); associated psychiatric variables included clinically significant somatization (Coefficient = 3.4; 95% CI = 2.2–4.6; β = 0.40), anxiety (Coefficient = 2.6; 95% CI = 1.3–3.9; β = 0.30), depression (Coefficient = 2.1; 95% CI = 0.74–3.4; β = 0.24), global severity index (Coefficient = 2.5; 95% CI = 1.2–3.8; β = 0.29), perceived stress (Coefficient = 0.21; 95% CI = 0.10–0.32; β = 0.29), and social isolation (Coefficient = 0.11; 95% CI = 0.04–0.17; β = 0.24), and increased daytime sleepiness as measured by ESS (Coefficient = 0.12; 95% CI = 0.02–0.23; β = 0.18). Methadone dose was not significantly associated with higher PSQI (Coefficient = 0.01; 95% CI = −0.01–0.04; β = 0.10).
Table 4:
Predictors of Worse Sleep Quality (PSQI)
| PSQI (linear) | Coefficient (Bivariate) | 95% CI | Beta weight | Coefficient (Multivariate) | 95% CI | Beta weight |
|---|---|---|---|---|---|---|
| BIVARIATE MODEL | MULTIVARIATE MODEL | |||||
| Demoaraphics | ||||||
| Age | 0.01 | −0.05–0.08 | 0.04 | 0.04 | −0.03–0.11 | 0.09 |
| Male gender | 0.26 | −1.1–1.6 | 0.03 | 0.34 | −1.1–1.7 | 0.04 |
| White race | 0.21 | −1.1–1.6 | 0.02 | −0.45 | −1.99–1.1 | −0.05 |
| College education or higher | 0.38 | −1.0–1.8 | 0.04 | 0.50 | −0.90–1.9 | 0.06 |
| Any current employment | −2.1 | −4.4–0.27 | −0.14 | −2.56 | −4.9–0.19 | −0.17 |
| History of homelessness | 1.5 | 0.11–2.9 | 0.17 | 0.29 | −1.3–1.8 | 0.03 |
| Methadone dose | 0.01 | −0.01–0.04 | 0.10 | |||
| BMI | 0.02 | −0.08–0.13 | 0.03 | |||
| Pain | ||||||
| History of chronic pain | 2.28 | 0.90–3.7 | 0.25 | |||
| Current Chronic Pain | 2.0 | 0.72–3.3 | 0.24 | |||
| Sleep interference | 0.49 | 0.29–0.70 | 0.38 | 0.40 | 0.18–0.62 | 0.31 |
| Use of additional pain medications | 0.67 | −1.54–2.88 | 0.05 | |||
| Psvchiatric | ||||||
| BSI somatization | 3.4 | 2.2–4.6 | 0.40 | 2.2 | 075–3.6 | 0.26 |
| BSI Anxiety | 2.6 | 1.3–3.9 | 0.30 | |||
| BSI Depression | 2.1 | 0.74–3.4 | 0.24 | |||
| Global severity index | 2.5 | 1.2–3.8 | 0.29 | |||
| TMT-A | 0.03 | −0.02–0.07 | 0.09 | |||
| TMT-B | −0.01 | −0.02–0.01 | −0.08 | |||
| Perceived stress | 0.21 | 0.10–0.32 | 0.29 | |||
| Social isolation | 0.11 | 0.04–0.17 | 0.24 | |||
| Sexual desire scale | −0.01 | −0.03–0.03 | −0.02 | |||
| Sieeoiness | ||||||
| Epworth Scale | 0.12 | 0.02–0.23 | 0.18 | −0.02 | −0.13–0.08 | −0.04 |
PSQI = Pittsburgh Sleep Quality Index
ESS = Epworth Sleepiness Scale
Italics only indicate p < 0.1 in bivariate model, but not significant when included in multivariable model.
Bold italics indicate p < 0.1 in bivariate model and p < 0.05 in multivariable model
In the multivariable model, higher PSQI was positively associated with pain interference with sleep (Coefficient = 0.40; 95% CI = 0.18–0.62; β = 0.31), clinically significant somatization (Coefficient = 2.2; 95% CI = 0.75–3.6; β = 0.26), and negatively associated with any current employment (Coefficient = −2.6; 95% CI = −4.9– −0.19; β = −0.17).
The correlates of increased daytime sleepiness (higher ESS) are summarized in Table 5. In the bivariate model, the following variables were positively associated with higher ESS (increased sleepiness): BMI (Coefficient = 0.35; 95% CI = 0.21–0.49; β = 0.37); pain variables including a history of chronic pain (Coefficient = 1.7; 95% CI = −0.28–3.8; β = 0.13), current chronic pain (Coefficient = 2.4; 95% CI = 0.55–4.2; β = 0.20), and the use of additional sedating pain medications (i.e. short-acting opioids, anticonvulsants, etc.) (Coefficient = 3.3; 95% CI = 0.22–6.5; β = 0.16); and higher PSQI score (Coefficient = 0.25; 95% CI = 0.03–0.47; β = 0.18). Higher methadone dose was not associated with increased sleepiness.
Table 5:
Predictors of Increased Sleepiness (Epworth)
| ESS (linear) | Coefficient (Bivariate) | 95% CI | Beta weight | Coefficient (Multivariate) | 95% CI | Beta weight |
|---|---|---|---|---|---|---|
| BIVARIATE MODEL | MULTIVARIATE MODEL | |||||
| Demographics | ||||||
| Age | −0.03 | −0.12–0.06 | −0.05 | −0.03 | −0.13–0.07 | −0.05 |
| Male gender | −1.1 | −3.0–0.8 | −0.09 | −0.71 | −2.6–1.2 | −0.06 |
| White race | −0.04 | −1.9–1.9 | −0.003 | −0.23 | −2.3–1.9 | −0.02 |
| College or higher | −0.49 | −2.5–1.5 | −0.04 | −0.53 | −2.4–1.4 | −0.04 |
| Any current employment | −2.70 | −6.0–0.64 | −0.12 | −2.5 | −5.6–0.73 | −0.11 |
| History of homelessness | 1.2 | −0.78–3.2 | 0.09 | 0.45 | −1.7–2.6 | 0.03 |
| Methadone dose | 0.005 | −0.03–0.04 | 0.02 | |||
| BMI | 0.35 | 0.21–0.49 | 0.37 | 0.32 | 0.18–0.46 | 0.33 |
| Pain | ||||||
| History of chronic pain | 1.7 | −0.28–3.8 | 0.13 | |||
| Current Chronic Pain | 2.4 | 0.55–4.2 | 0.20 | 1.6 | −0.26–3.5 | 0.13 |
| Sleep interference | 0.21 | −0.11–0.53 | 0.11 | |||
| Use of additional pain medications | 3.3 | 0.22–6.5 | 0.16 | 2.4 | −0.64–5.5 | 0.12 |
| Psychiatric | ||||||
| BSI somatization | 1.77 | −0.09–3.6 | 0.15 | 0.82 | −1.2–2.8 | 0.07 |
| BSI Anxiety | 0.96 | −0.98–2.9 | 0.08 | |||
| BSI Depression | 1.33 | −0.60–3.3 | 0.11 | |||
| Global severity index | 1.5 | −0.34–3.4 | 0.13 | |||
| TMT-A | 0.01 | −0.05–0.08 | 0.04 | |||
| TMT-B | 0.004 | −0.02–0.03 | 0.03 | |||
| Perceived stress | 0.06 | −0.10–0.22 | 0.06 | |||
| Social isolation | 0.08 | −0.02–0.17 | 0.12 | |||
| Sexual desire scale | 0.02 | −0.02–0.07 | 0.08 | |||
| Sleep Quality | ||||||
| PSQI | 0.25 | 0.03–0.47 | 0.18 | 0.12 | −0.11–0.35 | 0.08 |
PSQI = Pittsburgh Sleep Quality Index
ESS = Epworth Sleepiness Scale
Italics only indicate p < 0.1 in bivariate model, but not significant when included in multivariable model.
Bold italics indicate p < 0.1 in bivariate model and p < 0.05 in multivariable model
In the multivariable model, higher ESS was significantly associated with body mass index (Coefficient = 0.32; 95% CI = 0.18–0.46; β = 0.33) and a trend towards association with current chronic pain (Coefficient = 1.6; 95% CI = −0.26–3.5; β = 0.13; p-value = 0.09).
4. DISCUSSION
We found a high prevalence of poor sleep quality as measured by PSQI (90%) and excessive daytime sleepiness (46%) among people with OUD who are receiving methadone treatment. Although 90% of individuals had a PSQI score above 5, which is suggestive of poor sleep quality, 56% of individuals endorsed a subjective sleep quality of fairly good or very good sleep within the PSQI. These findings suggest that recognition of impaired sleep quality among people with OUD may be limited. This lack of recognition is a novel finding that could have significant implications for understanding and treating sleep disorders in people with OUD. The strongest predictors of impaired sleep quality in the study population were pain interference, clinically significant somatization, and employment status. Obesity was highly predictive of daytime sleepiness, and current chronic pain trended with higher sleepiness. Our data are consistent with prior studies that have demonstrated poor sleep quality in this patient population [22, 23]. While not studied directly here, sleep apnea may be undiagnosed and, if present, could contribute to a perception of sleep deficiency and/or sleepiness. It is important to note that our population had a low level of employment and high rate of history of housing instability, which likely reflects a sicker subset of the broader OUD population typically served in our treatment center.
Sleep disturbance is extremely common in patients receiving methadone treatment, reaching nearly 85% in some studies.[4, 24] Our findings are consistent with this prior literature and build upon the evidence base that impaired sleep quality is a critical comorbidity in many patients with OUD with potentially significant consequences. Furthermore, our data suggest that impaired sleep quality may go unnoticed among patients, which could lead to a longer time to diagnosis and fewer people receiving standard-of-care for sleep disorders. Alternatively, people with OUD and untreated sleep disorders might simply compensate in other ways. For example, it is known that the prevalence of cigarette smoking and nicotine use approaches 90% in the OUD population.[25] The stimulant effects of nicotine might help to overcome the effects of chronic sleep restriction and impaired sleep quality, which could partially explain why smoking quit rates are very low in people with OUD. The causes of sleep quality impairment in patients receiving methadone are likely complex and reflect comorbidities in addition to the direct effects of methadone or other substances on the nervous and respiratory systems. Methadone and buprenorphine both have known effects of sleep architecture including sleep fragmentation, frequent awakenings, decreased EEG arousals, reduced adenosine levels (buprenorphine, a partial opioid agonist treatment for OUD), inhibition of REM sleep, and less slow wave sleep [26–29]. Associated conditions predicting sleep impairment in prior studies included pain, anxiety and depressive symptoms, higher levels of nicotine dependence, methadone dose, benzodiazepine use, and unemployment [4, 24].
Subjective sleep impairments including inadequate or poor quality sleep were previously noted in more than 40% of OUD patients receiving either methadone or buprenorphine and did not differ by medication type, although limited statistical power may have affected the observed effects [23]. Similarly, polysomnogram studies revealed no differences in the presence or severity of sleep apnea and hypoxemia in OUD patients receiving either methadone or buprenorphine [30]. Sleep disordered breathing is also common in patients without OUD who are prescribed opioids long-term for pain (39% with obstructive sleep apnea and 24% with central sleep apnea) [7]. The risk of having central sleep apnea in the setting of chronic opioid exposure is particularly significant and appears to occur in a dose-dependent manner [8, 31]. In the present study, we noted that obesity was the major risk factor for higher daytime sleepiness, which in some cases may reflect the presence of undiagnosed sleep apnea. Indeed, our findings suggest that obese people with OUD might be at higher risk to have combined OSA and CSA and should routinely be referred for a comprehensive sleep evaluation including polysomnography.
In addition to the observed effects of pain and depressive symptoms on impaired sleep quality, we did not observe an association related to methadone dose and did not have specific data on the level of nicotine dependence or benzodiazepine use. The cross-sectional nature and statistical power of this study may have limited the evaluation of dose effects, which could be important. The literature characterizing the physiologic nature of abnormalities contributing to impaired sleep quality in patients receiving methadone compared with healthy controls has been mixed, but more commonly shows lower total sleep time, less slow wave sleep, less REM sleep, and more frequent nighttime awakenings [5, 6]. Chronic pain may be particularly problematic in this population, as suggested by our findings. While inadequately treated pain may contribute to insomnia and poor sleep quality, adequately or excessively treated pain might increase daytime sleepiness. It was interesting to note that the minority of our patients (10%) who were consuming sedating pain medications such as short-acting opioids (i.e. oxycodone) and anticonvulsants (i.e. gabapentin) in addition to receiving methadone had a higher risk of daytime sleepiness in the bivariate model but were not more likely to have a higher PSQI. Balancing pain control with optimized daytime arousal and functioning may pose a particularly significant clinical challenge in this population. In the scenario where pain is inadequately controlled, a downward cycle of increased pain medication leading to sleep impairment and subsequent higher pain sensitivity could complicate treatment efforts.
Similarly, somatization was an important predictor of worse sleep quality in our study. Some of the symptoms capturing somatization in the administered survey included nausea, weakness, chest pain, shortness of breath, faintness, and numbness. In prior literature, the somatization subscale of the BSI correlated highly with general distress [32]. Though the etiology of these symptoms is unclear, they may reflect underlying psychiatric or medical illnesses as well as medication side effects. Indeed, it is important to note that in the current study, somatization was highly correlated with anxiety, depression, and the global severity index subscales. This suggests that the presence of any of these symptoms may predict worse sleep quality.
Despite high rates of impaired sleep quality in the OUD population, the mechanisms and directionality underlying sleep disorders in people with OUD remains incompletely understood and is a current active area of scientific inquiry. One key observation has been that self-reported sleep complaints in patients receiving methadone correlate well with objective polysomnography measures including sleep duration and quality [22]. However, subjective poor sleep quality does not predict the presence of OSA or CSA [33], indicating that a failure to screen OUD patients for sleep-disordered breathing may miss an opportunity to treat these conditions. Given the significant effects that sleep deficiency has on cognitive performance and impulsivity [34], the importance of treating sleep disorders in people with OUD is heightened. The links between impaired sleep quality and relapse risk in drug addiction are significant [3]. Thus, addressing these factors and improving sleep quality in OUD patients may be a novel strategy to improve treatment outcomes, including potentially relapse and death.
The literature regarding the effects of opioids themselves on sleep quality has been mixed. Some of this discrepancy likely reflects differences in the underlying patient populations who use opioids. People with chronic pain syndromes without OUD likely have fewer complicating conditions that affect sleep quality compared with patients with OUD. As demonstrated in the present study, patients with OUD receiving methadone commonly suffer a range of comorbid conditions that may affect sleep including mood disorders, pain, and psychosocial stress. Elucidating which of these factors links closely with impaired sleep remains difficult, though it suggests that a comprehensive and integrated approach to care is necessary. Understanding potential benefits of modalities such as cognitive-behavioral therapy, mindfulness mediation, or biofeedback approaches might be useful in future investigations.
There are some limitations of the study to note. First, we recruited patients from three clinics operated by a single treatment organization, which uses a low-barrier to treatment-entry model; thus, the results may not generalize to other populations and settings. Nonetheless, we note similarities between our patient population and others with OUD that have been previously studied [4]. Second, the cross-sectional data in the current analysis prevented examination of longitudinal changes in sleep quality and daytime sleepiness, limiting our ability to evaluate cause and effect. Self-report measures are subject to recall bias, although sleep complaints in this population have previously been linked to objective sleep impairments as described above [22]. Third, other medical comorbidities (i.e., history of cardiovascular disease, cigarette smoking), methadone dose time and take-home status, and medications that might affect sleep were also unavailable in the dataset and may have been important variables to consider. Finally, our sample size may have been insufficiently large to detect statistically significant relationships between variables affecting sleep quality. Accumulating larger datasets on this population will increase understanding of mechanisms affecting sleep disorders in OUD patients.
5. CONCLUSIONS
Impaired sleep quality and excessive daytime sleepiness are very common among people with OUD receiving methadone. Pain, somatization, employment status, and obesity are critical risk factors for sleep disorders in this population. Many individuals reported a subjective experience of good sleep despite sleep quality impairment. Our findings indicate that people with OUD receiving methadone should be evaluated and treated for sleep disorders. Interventions targeting sleep disorders in this population may be critical to improving OUD treatment outcomes and may be effective to help reduce morbidity and mortality in this growing patient population.
Acknowledgments
We thank the APT Foundation research staff for the administration of the survey and data collection. This effort was supported by the National Institute on Drug Abuse (K23DA045957; K23 DA024050). The content of the manuscript solely reflects the views of the authors and does not necessarily represent the views of the NIH or authors’ affiliated institutions.
Funding: This study was funded by the National Institute on Drug Abuse (K23DA045957; K23 DA024050).
Footnotes
Compliance with Ethical Standards
Conflicts of Interest: The authors declare that they have no conflict of interest.
Ethical approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent: Informed consent was obtained from all individual participants included in the study.
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of a an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
REFERENCES
- 1.Volkow ND, Wang G-J, Telang F, Fowler JS, Logan J, Wong C, et al. Sleep deprivation decreases binding of [11C] raclopride to dopamine D2/D3 receptors in the human brain. Journal of Neuroscience. 2008;28:8454–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Volkow ND, Tomasi D, Wang G-J, Telang F, Fowler JS, Logan J, et al. Evidence that sleep deprivation downregulates dopamine D2R in ventral striatum in the human brain. Journal of Neuroscience. 2012;32:6711–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Brower KJ, Perron BE. Sleep disturbance as a universal risk factor for relapse in addictions to psychoactive substances. Medical hypotheses. 2010;74:928–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Stein MD, Herman DS, Bishop S, Lassor JA, Weinstock M, Anthony J, et al. Sleep disturbances among methadone maintained patients. Journal of Substance Abuse Treatment. 2004;26:175–80. [DOI] [PubMed] [Google Scholar]
- 5.Sharkey KM, Kurth ME, Corso RP, Brower KJ, Millman RP, Stein MD. Home polysomnography in methadone maintenance patients with subjective sleep complaints. The American journal of drug and alcohol abuse. 2009;35:178–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wang D, Teichtahl H, Drummer O, Goodman C, Cherry G, Cunnington D, et al. Central sleep apnea in stable methadone maintenance treatment patients. Chest. 2005;128:1348–56. [DOI] [PubMed] [Google Scholar]
- 7.Webster LR, Choi Y, Desai H, Webster L, Grant BJB. Sleep-Disordered Breathing and Chronic Opioid Therapy. Pain Medicine. 2007;9:425–32. doi: 10.1111/j.1526-4637.2007.00343.x. [DOI] [PubMed] [Google Scholar]
- 8.Walker JM, Farney RJ, Rhondeau SM, Boyle KM, Valentine K, Cloward TV, et al. Chronic opioid use is a risk factor for the development of central sleep apnea and ataxic breathing. Journal of Clinical Sleep Medicine. 2007;3:455–61. [PMC free article] [PubMed] [Google Scholar]
- 9.Moldofsky H. Sleep and pain. Sleep medicine reviews. 2001;5:385–96. [DOI] [PubMed] [Google Scholar]
- 10.Madden LM, Farnum SO, Eggert KF, Quanbeck AR, Freeman RM, Ball SA, et al. An investigation of an open-access model for scaling up methadone maintenance treatment. Addiction (Abingdon, England). 2018;113:1450–8. doi: 10.1111/add.14198. [DOI] [PubMed] [Google Scholar]
- 11.Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28:193–213. doi: 10.1016/0165-1781(89)90047-4. [DOI] [PubMed] [Google Scholar]
- 12.Buysse DJ, Hall ML, Strollo PJ, Kamarck TW, Owens J, Lee L, et al. Relationships between the Pittsburgh Sleep Quality Index (PSQI), Epworth Sleepiness Scale (ESS), and clinical/polysomnographic measures in a community sample. J Clin Sleep Med. 2008;4:563–71. [PMC free article] [PubMed] [Google Scholar]
- 13.Johns MW. Sleepiness in different situations measured by the Epworth Sleepiness Scale. Sleep. 1994;17:703–10. doi: 10.1093/sleep/17.8.703. [DOI] [PubMed] [Google Scholar]
- 14.Johns M, Hocking B. Daytime sleepiness and sleep habits of Australian workers. Sleep. 1997;20:844–9. doi: 10.1093/sleep/20.10.844. [DOI] [PubMed] [Google Scholar]
- 15.Netzer NC, Stoohs RA, Netzer CM, Clark K, Strohl KP. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Annals of internal medicine. 1999;131:485–91. [DOI] [PubMed] [Google Scholar]
- 16.Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain Inventory. Annals, Academy of Medicine, Singapore: 1994. [PubMed] [Google Scholar]
- 17.Derogatis LR. BSI 18, Brief Symptom Inventory 18: Administration, scoring and procedures manual: NCS Pearson, Incorporated; 2001. [Google Scholar]
- 18.Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. Journal of health and social behavior. 1983:385–96. [PubMed] [Google Scholar]
- 19.Russell DW. UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure. Journal of personality assessment. 1996;66:20–40. [DOI] [PubMed] [Google Scholar]
- 20.Spector IP, Carey MP, Steinberg L. The Sexual Desire Inventory: Development, factor structure, and evidence of reliability. Journal of sex & marital therapy. 1996;22:175–90. [DOI] [PubMed] [Google Scholar]
- 21.Holmberg D, Blair KL. Sexual desire, communication, satisfaction, and preferences of men and women in same-sex versus mixed-sex relationships. Journal of Sex Research. 2009;46:57–66. [DOI] [PubMed] [Google Scholar]
- 22.Sharkey KM, Kurth ME, Anderson BJ, Corso RP, Millman RP, Stein MD. Assessing sleep in opioid dependence: a comparison of subjective ratings, sleep diaries, and home polysomnography in methadone maintenance patients. Drug and alcohol dependence. 2011;113:245–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Dunn KE, Finan PH, Tompkins DA, Strain EC. Frequency and correlates of sleep disturbance in methadone and buprenorphine-maintained patients. Addict Behav. 2018;76:8–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Peles E, Schreiber S, Adelson M. Variables associated with perceived sleep disorders in methadone maintenance treatment (MMT) patients. Drug and Alcohol Dependence. 2006;82:103–10. [DOI] [PubMed] [Google Scholar]
- 25.Baldassarri SR, Fiellin DA, Savage ME, Madden LM, Beitel M, Dhingra LK, et al. Electronic cigarette and tobacco use in individuals entering methadone or buprenorphine treatment. Drug and Alcohol Dependence. 2019;197:37–41. doi: 10.1016/j.drugalcdep.2018.12.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gauthier EA, Guzick SE, Brummett CM, Baghdoyan HA, Lydic R. Buprenorphine disrupts sleep and decreases adenosine concentrations in sleep-regulating brain regions of Sprague Dawley rat. Anesthesiology: The Journal of the American Society of Anesthesiologists. 2011;115:743–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Staedt J, Wassmuth F, Stoppe G, Hajak G, Rodenbeck A, Poser W, et al. Effects of chronic treatment with methadone and naltrexone on sleep in addicts. European archives of psychiatry and clinical neuroscience. 1996;246:305–9. [DOI] [PubMed] [Google Scholar]
- 28.Cronin A, Keifer JC, Baghdoyan HA, Lydic R. Opioid inhibition of rapid eye movement sleep by a specific mu receptor agonist. BJA: British Journal of Anaesthesia. 1995;74:188–92. [DOI] [PubMed] [Google Scholar]
- 29.Xiao L, Tang Y-l, Smith AK, Xiang Y-t, Sheng L-x, Chi Y, et al. Nocturnal sleep architecture disturbances in early methadone treatment patients. Psychiatry research. 2010;179:91–5. [DOI] [PubMed] [Google Scholar]
- 30.Farney RJ, McDonald AM, Boyle KM, Snow GL, Nuttall RT, Coudreaut MF, et al. Sleep disordered breathing in patients receiving therapy with buprenorphine/naloxone. European Respiratory Journal. 2013;42:394–403. [DOI] [PubMed] [Google Scholar]
- 31.Hassamal S, Miotto K, Wang T, Saxon AJ. A narrative review: the effects of opioids on sleep disordered breathing in chronic pain patients and methadone maintained patients. The American journal on addictions. 2016;25:452–65. [DOI] [PubMed] [Google Scholar]
- 32.Meijer RR, de Vries RM, van Bruggen V. An evaluation of the Brief Symptom Inventory–18 using item response theory: Which items are most strongly related to psychological distress? Psychological assessment. 2011;23:193. [DOI] [PubMed] [Google Scholar]
- 33.Sharkey KM, Kurth ME, Anderson BJ, Corso RP, Millman RP, Stein MD. Obstructive sleep apnea is more common than central sleep apnea in methadone maintenance patients with subjective sleep complaints. Drug and alcohol dependence. 2010;108:77–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Drummond SPA, Paulus MP, Tapert SF. Effects of two nights sleep deprivation and two nights recovery sleep on response inhibition. Journal of sleep research. 2006;15:261–5. [DOI] [PubMed] [Google Scholar]