Predictors of Adherence to a Brief Behavioral Insomnia Intervention: Daily Process Analysis

Available online at www.sciencedirect.com

ScienceDirect Behavior Therapy 45 (2014) 430 – 442 www.elsevier.com/locate/bt

Predictors of Adherence to a Brief Behavioral Insomnia Intervention: Daily Process Analysis Megan E. Ruiter Petrov Arizona State University Kenneth L. Lichstein University of Alabama Carrie E. Huisingh Center for Clinical and Translational Science, University of Alabama at Birmingham Laurence A. Bradley University of Alabama at Birmingham

Behavioral interventions for insomnia are effective in improving sleep, yet adherence is variable, and predictors of adherence have not been consistently replicated. The relationships between daily variations in state factors at the initiation of treatment and adherence have not been investigated. Using 2-week, self-report online logs, this study determined, among 53 college students with probable insomnia, the associations of pretreatment factors and daily factors during treatment on daily variations in adherence to one session of behavioral treatments for insomnia. These treatments included stimulus control therapy (SCT), sleep restriction therapy (SRT), and sleep hygiene (SH). Low self-efficacy was associated with poorer SCT and SH adherence. Participants with a “bed partner or pet” at least some of the time had better SCT adherence. Greater total sleep time and poorer sleep quality were associated with poor SCT and SRT adherence the following night. Dr. Lichstein is a member of the Merck Insomnia Advisory Board. The other authors have no conflicts of interest to report. We would like to acknowledge the statistical consultation support received by Dr. Gerald McGwin. Dr. Lichstein is a member of the Merck Insomnia Advisory Board. The other authors have no conflicts of interest to report. Address correspondence to Megan E. Petrov, Ph.D., College of Nursing and Health Innovation, Arizona State University, 500 North 3rd Street, MC: 3020, Phoenix, AZ 85004.; e-mail: [email protected]. 0005-7894/45/430-442/$1.00/0 © 2014 Association for Behavioral and Cognitive Therapies. Published by Elsevier Ltd. All rights reserved.

Greater sleep efficiency was related to greater next night SCT and SRT adherence. Alcohol consumption was related to poorer SRT and SH adherence the following night. Future studies should test the replicability of these findings. Adherence trials may want to test whether discouraging alcohol intake, enhancing treatment-related self-efficacy, and monitoring and providing feedback on sleep, early in treatment, affects adherence.

Keywords: adherence; insomnia; behavior therapy; self-efficacy; alcohol

Sleep disorders, such as chronic insomnia, are among medical conditions with the lowest treatment adherence rates (DiMatteo, 2004). Chronic insomnia is a costly public health problem afflicting 6–15% of the general population (Lichstein, Durrence, Riedel, Taylor, & Bush, 2004; Morin, LeBlanc, Daley, Gregoire, & Mérette, 2006; Ohayon, 2002). Stimulus control therapy (SCT) and sleep restriction therapy (SRT) are evidence-based, effective treatments for chronic insomnia (Edinger & Means, 2005; Morgenthaler et al., 2006). However, adherence rates to these interventions have been highly variable and often suboptimal (Matthews, Arnedt, McCarthy, Cuddihy, & Aloia, 2013). Adherence rates were likely variable across studies because study duration and treatment components varied, and

adherence to insomnia intervention different metrics to measure adherence (e.g., often dichotomously vs. continuously) were used (Matthews et al., 2013). Currently there is no gold standard for measuring adherence to behavioral treatments for insomnia. Examples of how adherence has been measured have ranged from calculating the amount of deviance from a prescribed sleep schedule on a continuous scale (Riedel & Lichstein, 2001) to whether participants completed all treatment sessions (Vincent & Lionberg, 2001). Despite differing conceptualizations, the data available suggest that adherence, however it was defined, could be improved. In order to boost adherence, effective identification of patient and behavioral factors that alter adherence rates is crucial. Adherence to SCT, SRT, and general sleep hygiene (SH) recommendations is significantly linked to improvements in sleep (Harvey, Inglis, & Espie, 2002; Riedel & Lichstein, 2001; Vincent & Hameed, 2003; Vincent, Lewycky, & Finnegan, 2008). However, identifying replicable predictors of adherence has been elusive. Factors identified that are associated with improved adherence were acceptance of the intervention (Vincent & Lionberg, 2001), greater self and task-related efficacy (Bouchard, Bastien, & Morin, 2003), higher intentions to change sleep behavior (Hebert, Vincent, Lewycky, & Walsh, 2010; Matthews et al., 2013), lower fatigue (Matthews, Schmiege, Cook, Berger, & Aloia, 2012), less pretreatment sleepiness (Vincent et al., 2008), and greater pretreatment sleep disturbance severity (Hebert et al., 2010; Matthews et al., 2012; Morgan, Thompson, Dixon, Tomeny, & Mathers, 2003). Reported barriers to adherence were heightened psychopathological status (Dashevsky & Kramer, 1997; Hebert et al., 2010; McChargue et al., 2012; Vincent & Hameed, 2003), sleep concerns (Dashevsky & Kramer, 1997), fatigue (Dashevsky & Kramer, 1997), poor physical health (Hohagen et al., 1993; Morgan et al., 2003), and reductions in sleep disturbances (McChargue et al., 2012). One of the major problems with these studies is most if not all predictors were assessed at baseline or at the exit interview. Little information has been documented on the day-to-day factors that affect adherence during the initial treatment engagement. These factors may be more influential than those measured prior to treatment because they may affect subsequent adherence behavior or the decision to continue being engaged in treatment at all. Furthermore, in a changing health care field that increasingly emphasizes cost-effective, brief behavioral interventions, it is important to identify predictors for treatment-planning purposes, especially if provider–patient contact is infrequent or not feasible.

431

In the present study we sought to determine the predictors of variations in daily adherence during initial treatment engagement after one session of behavioral therapy for insomnia among young adults with probable insomnia. To do this we examined the daily relationships between several literature-based and exploratory factors and adherence to SCT, SRT, and SH over a 14-day period using self-reported, online daily diaries. The literature-based daily factors that were measured were fatigue, mood, perceived health status, and sleep quality, total sleep time, and sleep efficiency (Hebert et al., 2010; Matthews et al., 2012; McChargue et al., 2012; Morgan et al., 2003). Exploratory factors were defined as daily experiences known to affect sleep quality and were proposed in the literature as potentially interfering with adherence, yet have not been systematically investigated. The exploratory factors measured were pain (Smith, Perlis, Smith, Giles, & Carmody, 2000; Suh et al., 2011), stress, daily exercise (Baron, Reid, & Zee, 2012; Buman, Hekler, Bliwise, & King, 2011), and alcohol intake (Ebrahim, Shapiro, Williams, & Fenwick, 2013; Singleton & Wolfson, 2009). For replication purposes we also examined whether baseline sleepiness, self-efficacy, acceptance of treatment, intentions to adhere, depressive or anxiety symptoms, general health, insomnia symptom severity, and frequency of sleeping with a “bed partner or pet” were also associated with subsequent adherence rates. Bed partner or pet status was of particular interest because previous literature suggests involving a bed partner affects adherence to treatment for other sleep disorders (Baron et al., 2011; Cartwright, 2008).

Materials and Methods design The primary aim of the study was to identify day-to-day factors that were associated with adherence during behavioral insomnia treatment initiation. We employed a one-group design to determine daily factors experienced in the 2 weeks after one, in-person treatment session. Insomnia severity was assessed at baseline and after 2-week follow-up. participants In an online survey study, 1,678 college students participating in an introductory psychology course research subject pool were screened for probable insomnia. Those meeting study inclusion criteria were invited to take part in one didactic session of behavioral treatment for insomnia (n = 251). Participating in the study partially satisfied a course requirement. Fifty-eight college students agreed to participate in the study of those eligible (i.e., 23.5% response rate). Five participants dropped out during

432

ruiter petrov et al.

the 2-week follow-up period. The final sample size was 53 college students. These participants all completed one session of behavioral treatment for insomnia and at least 9 of the possible 14 daily online diaries. These participants met two major inclusion criteria: reported a clinical cutoff score of ≥ 8 on the Insomnia Severity Index (ISI; Morin, 1993), and met the International Classification of Sleep Disorders–II criteria for an insomnia disorder (ICSD-II; American Academy of Sleep Medicine, 2005). Participants were excluded if they were pregnant, employed in shift work, or had other probable sleep disorders, as determined by the Global Sleep Assessment Questionnaire (GSAQ; Roth et al., 2002). Risk for other sleep disorders was determined by responding “usually” or “always” to items on the GSAQ indicating prominent symptoms of sleep apnea, parasomnias, restless legs syndrome, and/or periodic limb movement disorder. Based on clinical contraindications, participants were also excluded if they reported any disabilities that severely restricted their mobility, any severe unstable medical conditions, or if they had a diagnosis of bipolar disorder, epilepsy, orthostatic hypotension, or a parasomnia (Smith & Perlis, 2006). The final sample was no different in age, sex, ethnicity, or insomnia symptom severity, frequency, or duration than those eligible in the subject pool (data not presented).

met study criteria if they responded to having these experiences at least “sometimes” in the past 4 weeks. Participants also had to endorse experiencing at least one of the following forms of daytime impairment related to their nighttime sleep difficulty: fatigue/ malaise; attention concentration, or memory impairment; social/vocational dysfunction or poor school performance; mood disturbance/irritability; daytime sleepiness; motivation/energy/initiative reduction; proneness for errors/accidents at work or while driving; tension headaches and/or gastrointestinal symptoms in response to sleep loss; and concerns or worries about sleep. Sleep Disorders Screener The GSAQ (Roth et al., 2002) was distributed to all potential participants to screen for individuals with other sleep disorders. The questionnaire consists of 11 items each with four response options (i.e., “never,” “sometimes,” “usually,” “always”). Sensitivity and specificity analyses revealed the GSAQ differentiates insomnia from obstructive sleep apnea, periodic limb movement, and parasomnias (Roth et al., 2002). Participants with scores indicative of no sleep disorder (i.e., responding “never” to all items) or any sleep disorder other than insomnia (i.e., responding “usually” or “always” to screening items) were excluded from the study.

screening measures Insomnia Severity Index The ISI was used as a brief screening measure of insomnia, and as an indicator of treatment outcome 2 weeks after one session of behavioral insomnia treatment. The ISI is a valid and reliable tool that quantifies perceived sleep difficulties and insomnia severity (Morin, 1993; Morin, Belleville, Bélanger, & Ivers, 2011). It is a seven-item instrument with ratings on a 0- to 4-point scale. A total composite score is summed with higher scores indicating greater insomnia severity. A clinical cutoff score of 8 was used as an identifier of threshold insomnia because it has optimal sensitivity and specificity in distinguishing people with insomnia from normal sleepers (Savard, Savard, Simard, & Ivers, 2005). ICSD-II Insomnia Screener Participants responded to questions about ICSD-II criteria for probable insomnia. The questions were “During the past 4 weeks did you have difficulty falling asleep, staying asleep, waking up too early, or feeling poorly rested in the morning?” and “Did you have trouble sleeping despite having adequate opportunity and circumstances to sleep?” Response choices for both questions included “never,” “sometimes,” “usually,” or “always.” Participants

behavioral insomnia treatment SCT (Bootzin, Epstein, & Ward, 1991) was used to reestablish a consistent sleep/wake schedule, and sleep-promoting associations with the bed/bedroom. The participants were encouraged to follow six instructions: discontinue all arousing activities conducted in the bed/bedroom except sex; go to bed only when sleepy, get out of bed and engage in a quiet activity in another room if they did not fall asleep within 15 − 20 minutes, repeat the previous instruction for awakenings during the night, get up at the same time every morning regardless of how much sleep they obtained, and to not nap. SRT (Spielman, Saskin, & Thorpy, 1987) was used to increase sleep consolidation by creating a state of partial sleep deprivation. SRT prescribes a restricted amount of time spent in bed equal to the average total sleep time (see Spielman et al., 1987). In this study the participants were instructed to restrict their time in bed to their typical total sleep time plus a half hour. Time in bed and total sleep time estimates were determined from recall of sleep from the past 3 weekdays or the most recent 3 weekdays typical of their usual sleep pattern. Participants were encouraged to get up at a fixed wake-up time regardless of the amount of sleep they obtained.

adherence to insomnia intervention Participants were instructed to calculate their average sleep efficiency (proportion of time slept to time spent in bed) and were provided with handouts on the method. Participants were told how to alter their time in bed according to their average sleep efficiency (i.e., by 15-minute increments) after the 2-week follow-up period if they chose to do so. SH (Lichstein & Morin, 2000) is a psychoeducational intervention consisting of five instructions meant to discourage behaviors that influence sleep quality and quantity. The instructions were to avoid caffeine after noontime, and to avoid exercise, nicotine, alcohol, and heavy meals within 2 hours of bedtime.

independent variables Potential Daily Assessed Predictors All day-to-day factors were measured using an online sleep diary (Perlis, Jungquist, Smith, & Posner, 2005). Participants were asked to record immediately prior to bed how they felt during the day on the following variables: pain, fatigue, mood, stress, perceived health status, amount of daily exercise (in minutes), and number of alcoholic beverages consumed. Pain and fatigue severity were measured on a 0–5 scale from none to a lot. Mood (affect) was assessed on a 0–5 scale from bad to good. Perceived health status was measured on a 0–5 scale from felt fine to bad. Total sleep time, sleep efficiency, and sleep quality (i.e., very poor, poor, fair, good, or excellent) were recorded immediately after the final awakening of each night. Potential Baseline Predictors Baseline, self-reported variables included perceived health status, treatment-related self-efficacy, acceptance of treatment, intentions to adhere, ISI score, GSAQ-measured excessive daytime sleepiness, GSAQ-measured depressive and anxiety symptoms, and frequency of sleeping with a bed partner or pet. See Table 1 for an overview of how most of these variables were measured. Treatment-related selfefficacy was measured with one item adapted from Reed and Aspinwall (1998). The item read, “If I wanted to follow treatment recommendations it would be easy for me to do so tonight,” and was measured on a 1–9 scale from totally disagree to totally agree. Acceptance of treatment and intentions to adhere were assessed with the following adapted questions from Sherman and colleagues (2000): “How important do you think it is that people engage in the sleep-promoting behaviors that were discussed to avoid the consequences of poor and limited sleep?” and “How likely do you think it is that you personally will actually follow

433

the behavioral recommendations to treat your insomnia?” The questions were answered using a 0–9 scale, anchored from not important to very important.

adherence measures Adherence to sleep recommendations was measured with online sleep diaries and two adherence logs specific to SH and SCT instructions. Sleep diaries are a record of a participant’s sleep and wake time each night for a period of 2 weeks. The sleep diary used in the present study was by Perlis and colleagues (2005). This particular diary was used because it documents common sleep variables and specific variables related to SCT and SRT instructions. The primary dependent variable from the sleep diary was the average of the time deviations (in minutes) from the prescribed wake time. This average was used as a measure of adherence to SRT. This method is similar in approach to that by Riedel and Lichstein (2001) who found that this method of assessing SRT adherence through variations in arising time is superior to other adherence measures in predicting treatment outcome. Specifically they found that measures of time in bed and bedtime reduction did not significantly correlate with treatment outcomes, whereas lower variance in arising time in the morning was significantly correlated. Two adherence logs were used to document whether the participants were adherent to the six instructions of SCT and the five instructions of SH. The proportion of instructions the participants were compliant to each of the 14 days was averaged. This method of measurement is similar to that used in a previous study (McChargue et al., 2012). procedure Didactic sessions were conducted in a university building by a trained undergraduate honors research assistant who was supervised on a weekly basis by a certified behavioral sleep medicine specialist. Sessions were conducted at various times across the 2010 − 2011 academic year, but never between semesters or during spring break. Each session was delivered in an interactive, didactic format to small groups of two to nine participants. The sessions were not structured as group therapy. The treatment session featured sleep and insomnia education, and three behavioral treatment components: SCT, SRT, and SH. The session lasted 1.5 hours. After the treatment session, participants completed 14 days of time-stamped, online sleep and adherence diaries. To dampen any bias in the measurement of the study’s

434

ruiter petrov et al.

Table 1

Sample Characteristics of Baseline and 14-Day Postbehavioral Insomnia Treatment Variables Variable

# Log Entries

Mean or %

SD

Range

1.7

17–25

Baseline Variables

Age (years) Gender (% female) Race/ethnicity Non-Hispanic White Non-Hispanic Black Other Ethnic groups Health status Good, fair, or poor Excellent or very good Insomnia Severity Index GSAQ excessive daytime sleepiness Usually or always Never or sometimes “Bed partner or pet” frequency At least sometimes Never GSAQ depressive or anxiety symptoms frequency Usually or always Never or sometimes Self-efficacy Low (1–5) Moderate (6–7) High (8–9) Acceptance of treatment Lower (5–7) Higher (8–9) Intentions to adhere Lower (3–7) Higher (8–9)

53 46

18.9 86.8

40 10 3

75.5 18.9 5.7

22 31 53

41.5 58.5 12.4

3.4

8.0–21.0

18 35

34.0 66.0

-

-

23 30

43.4 56.6

-

-

20 33

37.7 62.3

-

-

18 19 16

34.0 35.9 30.1

30 23

56.6 43.4

27 26

50.9 49.1

14-Day Postbehavioral Insomnia Treatment Variables

# Log Entries

Mean or %

SD

Range

Total sleep time (in minutes) Sleep efficiency (%) Sleep quality Poor to very poor (1–2) Fair to excellent quality (3–5) Pain None to low levels (0–1) Moderate to high levels (2–5) Fatigue Low (0–1) Moderate (2–3) High (4–5) Mood Bad (0–1) Fine (2–3) Good (4–5) Stress None to low levels (0–2) Moderate to high levels (3–5) Health status Good (0–1) Fair (2–3) Poor (4–5)

700 700 719 109 610 719 521 198 719 161 373 185 719 66 284 369 720 367 353 719 493 159 67

407.4 86.4 3.4 15.2 84.8 0.9 72.5 27.5 2.6 22.4 51.9 25.7 3.3 9.2 39.5 51.3 2.4 51.0 49.0 1.1 68.6 22.1 9.3

130.0 15.0 0.4 1.2 1.3 -

0–857 0–100 1.9–4.5 0–5 0–5 -

1.2 -

0–5 -

1.4 1.4 -

0–5 0–5 -

435

adherence to insomnia intervention Table 1 (continued) Variable

# Log Entries

Mean or %

SD

Range

35.9 1.37

0–300 0–9

Baseline Variables

Exercise time (in minutes) No. alcoholic beverages None ≥1

718 718 631 87

24.8 0.6 87.9 12.1

Note. GSAQ = item derived from the Global Sleep Assessment Questionnaire; SD = standard deviation

objective (i.e., to assess adherence), participants were informed the study was intended to discover what aspects of behavioral insomnia treatment make it effective. Participants were reminded by e-mail every other day to complete the online logs no matter if they utilized treatment recommendations or not. The participants were informed that their ability to satisfy the course requirement was not contingent on if they followed the treatment recommendations. At the completion of the study, the participants were debriefed of the actual intent of the study. The university’s Institutional Review Board reviewed and approved of the study protocol.

Data Analysis We computed descriptive statistics on all predictor variables, covariates, and outcomes. A paired-samples t test was used to compare baseline and posttreatment ISI scores. The continuous data for the adherence outcome measures were not normally distributed; therefore, they were collapsed into ordinal variables. The average proportion of SCT instructions followed over the 14-day follow-up period was categorized into the following groups: less than four out of six (b 0.67); four out of six (0.67); five out of six (0.83); and six out of six (1.00). The average proportion of SH instructions followed over the 14-day follow-up period was categorized into the following groups: 0.80; 0.80– 0.90; and 1.00 (five out of five instructions). The average deviation in minutes from the prescribed wake time in SRT was converted into clinically meaningful increments as follows: 0; 1–15; 16–30; 31–45; 46–60; 61–90; and N 90. A proportional odds model (Scott, Goldberg & Mayo, 1997) using generalized estimated equations was used to analyze the ordinal outcome data and produce odds ratios (ORs) and 95% confidence intervals (CIs) for the association between the baseline and daily assessed predictors with SCT, SRT, and SH adherence scores. The use of generalized estimated equations allowed for the models to account for the clustering within study participants. Daily assessed fatigue, mood, stress, pain, sleep quality, and health status as well as baseline self-efficacy, treatment acceptance, and

intentions to adhere were measured on ordinal scales. Bed partner or pet status, baseline health status, sleepiness, and depressive and anxiety symptoms were measured on categorical scales. The sample sizes for some of the values on these ordinal and categorical scales were low. Low sample sizes per ordinal value or category can lead to wide confidence intervals in generalized estimated equations analysis. Therefore, these variables were collapsed into categories based on the frequency of endorsement for each value in each predictor to improve the stability of the estimates. The number of predictors to be tested with the three adherence outcome measures was large. Rather than using a strict correction factor for multiple comparisons, such as the Bonferroni, which may increase our Type II error rate, the following analyses were based on the stepwise approach used by Fournier and colleagues (2009). This stepwise approach was used to determine the most salient predictors of adherence while minimizing the number of statistical tests conducted to balance concerns about high rates of Type I and Type II errors. First, each predictor variable was assigned to one of seven domains based on content and time of collection as follows: demographics (age, gender, race/ethnicity), baseline sleep and health (ISI score, excessive daytime sleepiness, depressive and anxiety symptoms, and general health status), baseline treatment perceptions (selfefficacy, acceptance of treatment, and intentions to adhere), bed partner or pet status, nightly sleep (total sleep time, sleep efficiency, and sleep quality), daytime symptoms (pain, mood, fatigue, stress, and health status), and daytime behaviors (alcohol consumption and time spent exercising). To check for issues with multicollinearity, the correlation between total sleep time and sleep efficiency was assessed (r = .68). Since the correlation was lower than r b .75, both variables were placed in the same domain. Second, six separate generalized estimated equation models were constructed to identify significant predictors within each domain in a stepwise manner. In Step 1, any predictors within each domain that attained significance values of

436

ruiter petrov et al.

p b 0.20 were retained in the model. At subsequent Steps 2 and 3, predictors were only retained if they achieved significance values of p b .10 and then p b .05, respectively. All remaining significant predictors were then included into a final, full model to determine unique effects of each predictor while controlling for the effects of the other significant predictors. A cumulative logit distribution was used to estimate the model parameters. An alpha level of p b .05 was considered statistically significant. The generalized estimated equation analyses were conducted using the PROC GENMOD statement in SAS 9.2 (SAS Institute, Inc., Cary, North Carolina).

Results sample characteristics Fifty-three participants completed the treatment protocol and at least 9 days of sleep diaries and adherence logs (M = 13.2, SD = 1.8, range: 9 − 14 days). Five additional participants completed the treatment protocol but dropped out of the study before providing any sleep diary or adherence log information. These participants were significantly older, M = 20.8, SD = 2.7 versus M = 18.9, SD = 1.7, t(56) = –2.3, p = .030, and had less treatmentrelated self-efficacy, M = 4.4, SD = 2.9 versus M = 6.3, SD = 1.8, t(56) = 2.1, p = .040, than the final sample. They did not differ on any other baseline variables. The most cited reason for dropping out of the study was time commitment. The vast majority of the final sample were full-time students (98.1%), and 15.4% were employed part-time. None of the participants reported being diagnosed or treated for a mental disorder. Descriptive statistics on all independent variables can be found in Table 1. treatment outcome Baseline ISI scores (M = 12.4, SD = 3.4) compared with 2-week follow-up ISI scores (M = 10.7, SD = 3.7) indicated a statistically significant reduction in insomnia severity, t(50) = 3.51, p = .001. missing data Common to most studies conducting daily measurements, there were some missing data on the daily assessed variables and adherence measures. Missing data for the daily assessed variables from a possible 742 entries (i.e., 53 participants × 14 days) ranged from 3.0 to 5.7%. Missing data for the daily adherence measures ranged from 3.2 to 4.2%. Given the small percentages of missing data, complete case analyses were not employed to assess all relationships.

adherence to online diary entry At the treatment session, participants were instructed to complete online sleep and treatment adherence logs each morning over the following 14 days. Of 727 time-stamped, online diary entries, 75.2% (n = 547) were entered within 1 day, and 85.4% (n = 621) were entered within 2 days of the designated time. Treatment group size was not related to any of the adherence variables (data not presented). treatment adherence Table 2 displays the descriptive information on the adherence scores for each treatment component. Participants, on average, followed four to five of the six SCT instructions and five SH recommendations per night. In contrast, only 46.7% of the participants followed SRT recommendations fairly well (i.e., only deviated from their recommended wake time on average by no more than 30 minutes). relationships of baseline and daily assessed factors to sct adherence Table 3 presents Steps 1, 2, and the final model of the stepwise procedure to identify baseline and daily assessed variables associated with SCT adherence. In the final model, the only baseline Table 2

Level of Adherence of Study Participants Over the 14-Day Follow-Up Period Adherence Outcomes

# Log Entries

Mean or %

SD

Range

Stimulus Control (SC) a b 0.67 0.67 0.83 1.0 Sleep Restriction Therapy (SRT) b 0 1–15 16–30 31–45 46–60 61–90 ≥ 91.0 Sleep Hygiene (SH) c b 0.8 0.8–0.9 1.0

719 106 168 309 136 713

0.77 14.7 23.4 43.0 18.9 81.6

0.18 – – – – 96.5

0–1 – – – – 0–540.0

207 71 55 22 62 55 241 720 122 247 351

29.0 10.0 7.7 3.1 8.7 7.7 33.3 0.85 16.9 34.3 48.8

– – – – – – – 0.18 – – –

– – – – – – – 0–1 – – –

Note. SD = standard deviation. a Adherence is the proportion of the six instructions followed each day/night. b Adherence is the number of minutes discrepant from the recommended wake time. c Adherence is the proportion of the five instructions followed each day/night.

437

adherence to insomnia intervention Table 3

Stepwise Approach to Final Model of Variables Associated With Adherence to Stimulus Control Therapy Step 1: Retain Effects at p b .20 a

95% CI

p

Domain

Predictor

OR

Baseline sleep and health Baseline treatment perceptions

Health (ref: excellent or very good; 0–1); good, fair, or poor (2–4) Self-efficacy (ref: high; 8–9); moderate (6–7) Low (1–5) Bed partner or pet (ref: never) Total sleep time (15-min increments) Sleep efficiency (5% increments) Sleep quality (ref: fair to excellent); poor to very poor Pain (ref: none to low; 0–1); moderate to high (2–5) Health status (ref: good; 0–1); fair health (2–3) Poor health (4–5) Exercise (30-min increments)

1.79 0.70 0.22 2.06 0.95 1.14 0.57 1.46 0.60 0.45 1.16

0.94, 3.41 0.31, 1.59 0.09, 0.55 1.14, 3.69 0.92, 0.98 1.03, 1.27 0.34, 0.98 1.02, 2.09 0.43, 0.85 0.23, 0.87 0.96, 1.41

Health (ref: excellent to very good; 0–1); good, fair, or poor (2–4) Self-efficacy (ref: high; 8–9); moderate (6–7) Low (1–5) Bed partner or pet (ref: never) Total sleep time (15-min increments) Sleep efficiency (5% increments) Sleep quality (ref: fair to excellent); poor to very poor Pain (ref: none to low; 0–1); moderate to high (2–5) Health status (ref: good; 0-1); fair health (2–3) Poor health (4–5)

1.68 0.66 0.21 2.06 0.95 1.14 0.57 1.47 0.58 0.44

0.93, 0.33, 0.10, 1.14, 0.92, 1.03, 0.34, 1.01, 0.40, 0.24,

3.03 1.31 0.44 3.69 0.98 1.27 0.98 2.15 0.84 0.83

.085 .235 b .001 .016 .003 .007 .043 .043 .004 .011

0.79 0.25 2.57 0.95 1.14 0.56

0.39, 0.13, 1.53, 0.92, 1.03, 0.33,

1.58 0.46 4.33 0.98 1.25 0.94

.512 b .001 b .001 .003 .009 .006

Baseline bed partner or pet Nightly sleep

Daytime symptoms

Daytime behaviors

.077 .392 .001 .016 .003 .007 .043 .041 .004 .018 .134

Step 2: Retain Effects at p b .10 Baseline sleep and health Baseline treatment perceptions Bed partner or pet Nightly sleep

Daytime symptoms

Final Model With All Significant Predictors (p b .05) in Step 3: From All Domains Baseline treatment perceptions

Self-efficacy (ref: high; 8–9);

moderate (6–7) Low (1–5)

Bed partner or pet Nightly sleep

Bed partner or pet (ref: none) Total sleep time (15-min increments) Sleep efficiency (5% increments) Sleep quality (ref: fair to excellent); poor to very poor

Note. OR = odds ratio; CI = confidence interval. a Odds ratios greater than 1.0 indicate greater treatment adherence; odds ratios less than 1.0 indicate poorer treatment adherence.

variables that were retained were bed partner or pet status and treatment-related self-efficacy. Better SCT adherence was associated with having a bed partner or pet. Lower SCT adherence was associated with low self-efficacy. Odds of reporting good SCT adherence were also decreased by 44% for reporting poor to very poor sleep quality when compared with fair to excellent sleep quality. An increase in total sleep time by 15 minutes was associated with a 5% decrease in adherence, whereas an increase in sleep efficiency by 5% was related to a 14% increase in adherence.

relationships of baseline and daily assessed factors to srt adherence Table 4 presents Steps 1, 2, and the final model of the stepwise procedure to identify baseline and daily assessed variables associated with adherence to SRT wake time recommendations. None of the

baseline variables assessed from any of the domains were significantly related to nightly variations in SRT adherence. Daily assessed variables that were significantly related to SRT adherence in the final model were alcohol consumption, total sleep time, sleep efficiency, and sleep quality. Any alcoholic beverages consumed were related to greater odds of poorer SRT adherence compared with no alcoholic beverages consumed. Increases in total sleep time and poorer sleep quality were also related to next night poorer SRT adherence, whereas increases in sleep efficiency were related to greater SRT adherence.

relationships of baseline and daily assessed factors to sh adherence Table 5 displays Steps 1, 2, and the final model of the stepwise procedure to identify baseline and daily assessed variables associated with SH

438

ruiter petrov et al.

Table 4

Stepwise Approach to Final Model of Variables Associated With Adherence to Sleep Restriction Therapy Step 1: Retain Effects at p b .20 Domain

Predictor

OR a

95% CI

p

Nightly sleep

Total sleep time (15-min increments) Sleep efficiency (5% increments) Sleep quality (ref: fair to excellent); poor to very poor Mood (ref: good; 4–5); fine (2–3) Bad (0–1) Alcohol consumption (ref: 0 drinks)

0.78 1.51 0.46 0.70 0.62 0.39

0.72, 1.34, 0.29, 0.49, 0.40, 0.22,

0.83 1.70 0.74 0.99 0.96 0.70

b .001 b .001 .001 .043 .033 .002

0.78 1.51 0.46 0.68 0.59 0.38

0.72, 1.34, 0.29, 0.47, 0.37, 0.21,

0.83 1.70 0.73 0.99 0.93 0.68

b .001 b .001 .001 .049 .023 .001

0.77 1.53 0.46 0.38

0.72, 1.36, 0.29, 0.17,

0.83 1.72 0.74 0.84

b .001 b .001 .001 .017

Daytime symptoms Daytime behaviors

Step 2: Retain Effects at p b .10 Nightly sleep

Daytime symptoms Daytime behaviors

Total sleep time (15-min increments) Sleep efficiency (5% increments) Sleep quality (ref: fair to excellent); poor to very poor Mood (ref: good; 0–1); fine (2–3) Bad (4–5) Alcohol consumption (ref: 0 drinks)

Final Model With All Significant Predictors (p b .05) in Step 3: From All Domains Nightly sleep

Daytime behaviors

Total sleep time (15-min increments) Sleep efficiency (5% increments) Sleep quality (ref: fair to excellent); poor to very poor Alcohol consumption (ref: 0 drinks)

Note. OR = odds ratio; CI = confidence interval. a Odds ratios greater than 1.0 indicate greater treatment adherence; odds ratios less than 1.0 indicate poorer treatment adherence.

adherence. In the final model, low treatment-related self-efficacy (compared to high self-efficacy) and any alcohol use were significantly associated with poorer SH adherence.

Discussion Daily variations in adherence, during the 2 weeks after one session of behavioral therapy for insomnia, were associated with several pretreatment participant factors and during treatment state factors. Lower baseline treatment-related self-efficacy was related to lower SCT and SH adherence. Sharing a bed with a partner or pet was related to better SCT adherence. Poor sleep quality, greater total sleep time, and less sleep efficiency were associated with poorer SCT and SRT adherence the following night. Consumption of alcoholic beverages was related to poorer SRT and SH adherence on the following night. The association between low self-efficacy and poorer adherence to SCT and SH recommendations is in concordance with a previous study. That study found that greater global, treatment-related self-efficacy was related to better weekly adherence to cognitive-behavioral therapy for insomnia, particularly during the first week of treatment (Bouchard et al., 2003). In the present study, self-efficacy accounted for more of the variability

in SCT and SH adherence than acceptance of treatment and intentions to adhere, suggesting that gauging or enhancing patients’ self-efficacy may be particularly important during the initial introduction to treatment for successful subsequent treatment adherence and response. Our results suggest that poor sleep quality and sleep efficiency as well as greater total sleep time on one night were predictive of nonadherence to SCT and SRT the next night. The total sleep time result is similar to that found by McChargue and colleagues (2012). Their study found that decreasing sleep disturbances over the course of treatment was related to lower adherence to relaxation therapy for insomnia among women with breast cancer. In the present study it follows that if participants are experiencing greater sleep durations, they may not feel the need to engage in further treatment having perceived that their sleep is improved. However, this relationship also may be seen as evidence that the participant is not engaging in treatment from the very beginning of treatment. Behavioral treatments of insomnia when followed usually result in lower total sleep times during the early treatment phase and then increase over time. Poor sleep quality, low sleep efficiency, and greater total sleep time within the first few nights of treatment may

439

adherence to insomnia intervention Table 5

Stepwise Approach to Final Model of Variables Associated With Adherence to Sleep Hygiene Recommendations Step 1: Retain Effects at p b .20 p

1.10 1.94 0.58 0.58 1.28 0.40 1.67 0.67 0.50 0.61 0.11

0.96, 0.84, 0.26, 0.27, 0.46, 0.12, 0.79, 0.37, 0.30, 0.30, 0.06,

1.25 4.47 1.32 1.25 3.51 1.34 3.54 1.20 0.80 1.23 0.20

.189 .121 .195 .162 .637 .138 .181 .173 .004 .167 b .001

1.11 0.31 0.75 0.51 0.64 0.11

0.42, 0.12, 0.34, 0.30, 0.32, 0.07,

2.95 0.81 1.66 0.87 1.26 0.19

.827 .016 .480 .014 .194 b .001

0.93 0.25 0.11

0.38, 2.30 0.10, 0.63 0.06, 0.21

.882 .004 b .001

Predictor

OR

Baseline sleep and health

Insomnia Severity Index Health (ref: excellent to very good; 0–1); good, fair, or poor (2–4) EDS (ref: never or sometimes); usually or always Intentions (ref: lower; 3–7); higher (8–9) Self-efficacy (ref: high; 8–9); moderate (6–7) Low (1–5) Bed partner or pet (ref: never) Sleep quality (ref: fair to excellent); poor to very poor Health status (ref: good 0–1); fair health (2–3) Poor health (4–5) Alcohol consumption (ref: 0 drinks)

Self-efficacy (ref: high; 8–9);

Baseline treatment perceptions

Bed partner or pet Nightly sleep Daytime symptoms Daytime behaviors

a

95% CI

Domain

Step 2: Retain effects at p b .10 Baseline treatment perceptions Nightly sleep Daytime symptoms Daytime behaviors

moderate (6–7) Low (1–5) Sleep quality (ref: fair to excellent); poor to very poor Health status (ref: good 0–1); fair health (2–3) Poor health (4–5) Alcohol (ref: 0 drinks)

Final Model With All Significant Predictors (p b .05) in Step 3: From All Domains Baseline treatment perceptions

Self-efficacy (ref: high; 8–9);

Daytime behaviors

Alcohol (ref: 0 drinks)

moderate (6–7) Low (1–5)

Note. OR = odds ratio; CI = confidence interval; EDS = excessive daytime sleepiness. a Odds ratios greater than 1.0 indicate greater treatment adherence; odds ratios less than 1.0 indicate poorer treatment adherence.

simply be indicators that the participant is struggling to engage with the treatment, which is predictive of a continuing struggle throughout treatment. Careful monitoring of sleep by the therapist within the first few nights of treatment engagement may be crucial for successful treatment adherence and response thereafter. Several literature-based predictors were not replicated in this study, including pretreatment depressive and anxiety symptoms, general health, daytime sleepiness, and insomnia severity, as well as daily assessed fatigue, mood, pain, exercise, and health status. The baseline-assessed variables likely were not replicated for the same reasons that so many of the literature-based factors have not been consistently replicated in other studies such as differences in the measurement of predictors and adherence outcomes, treatment components and how they were delivered, and populations studied. Without standardized methods of measuring variables in adherence trials and the delivery of behavioral treatments of insomnia it is difficult to

compare results to previous studies. Regarding the daily assessed variables, these factors were only found to be significantly related to treatment adherence in previous studies when they were measured at baseline assessment, which may explain why they were not found to be significantly related in this study. A novel predictor of SCT adherence was sharing a bed with a bed partner or pet at least some of the time. This result may be explained by the likely situation that these participants lived in apartments or homes, rather than a college dormitory where the implementation of SCT may be more difficult. However, an alternative explanation is that merely the physical and social presence of another living creature, whether human or animal in the bed, may improve SCT and SH adherence. These participants may have altered their sleep patterns more readily because they felt accountable to the other person or animal present in the bed. Another explanation, in the case of a human bed partner, is social support received from that bed partner may have influenced

440

ruiter petrov et al.

SCT adherence. Objective sleep data have revealed sleep–wake cycles among bed partners are often interdependent (Meadows, Arber, Venn, Hislop, & Stanley, 2009). We propose that social support from a bed partner may have affected self-efficacy in adhering to behavioral sleep treatment recommendations. The health behavior literature suggests that relationship partners influence health behavior change (Lewis et al., 2006), and a recent review suggests bed partners may be an important component for successful behavioral treatment for insomnia (Rogojanski, Carney, & Monson, 2012). However, extreme caution should be exercised with this interpretation because the assessment of bed partner or pet status was combined, therefore it is unknown how many of these participants were reporting on bed partners as opposed to pets. Future studies should assess for bed partner status separate from pets and reevaluate its relation to treatment adherence. Alcohol consumption was associated with poorer next night SRT and SH adherence. The timing of this consumption was not measured. However, if the consumption occurred in the 2 hours prior to bedtime in this sample, it may have directly or indirectly affected sleep–wake schedules. Alcohol’s detrimental effects on sleep have been documented among healthy, nonalcoholic participants (Roehrs & Roth, 2001). Among college students, greater alcohol consumption is associated with delays in bedtimes and wake times, oversleeping, and desynchrony between weekday and weekend sleep–wake schedules (Singleton & Wolfson, 2009). These previous studies along with our data indicate alcohol may be a barrier to full adherence to behavioral insomnia treatments. However, given that a large proportion of the study participants were under the legal drinking age, another interpretation is that these participants who did not adhere to drinking regulations may also be prone not to adhere to a behavioral treatment program. Nonetheless, clinicians may opt to educate their patients on the effects of alcohol on sleep, and monitor consumption. Future research should investigate whether education and monitoring would promote adherence and lead to more favorable treatment outcomes than not adding these components. The meaningfulness of these data are bolstered by the study’s strengths, including the use of ICSD-II criteria to identify participants with probable insomnia, obtaining daily online data, and analyzing these data from a daily process framework. However, the study has limitations. Notably, our sample size was small and confined to mostly female and non-Hispanic White college undergraduates with mild insomnia, thus limiting generaliz-

ability to other ethnic groups, young adults, and patients with severe insomnia. Therefore, the findings from the final regression models may not generalize to men or to nonstudents with insomnia who are seeking treatment. Nonetheless, this population is important to investigate because many insomnia sufferers report their symptoms began during their older adolescence and young adult years (Bixler, Kales, Soldatos, Kales, & Healey, 1979; Singareddy et al., 2012), and the transition to college often heralds changes in sleep patterns that have been associated with poor academic performance and psychopathology (Gaultney, 2010; Taylor et al., 2011). Further limitations of the study were the lack of psychometrically validated and standardized daily assessed predictor variables and the lack of verified clinical diagnoses of insomnia. While the questions used to determine whether the participants had insomnia conformed to ICSD-II diagnostic criteria, the questions had not been previously validated and therefore the participants could only be determined to have probable insomnia. This is in contrast to the literature that mostly reports on the adherence to behavioral insomnia treatment by participants with verified, chronic insomnia. Future research should work to standardize self-report screening measures for insomnia according to ICSD-II criteria and daytime symptoms and behaviors related to behavioral treatments for insomnia. A final limitation was that the study utilized a didactic, single session of behavioral insomnia treatment adapted from cognitive-behavioral therapy for insomnia, which differs from this evidence-based treatment’s optimal treatment duration of four weekly sessions (Edinger, Wohlgemuth, Radtke, Coffman, & Carney, 2007). However, the original intent of the present study was to capture adherence at the early stages of treatment initiation because delivery of four sessions of behavioral insomnia treatment in a modern health care system is not always feasible.

Conclusions We discovered several pretreatment participant factors as well as daily experienced factors during treatment engagement were related to daily variations in adherence to a brief behavioral intervention for insomnia. The results suggest interventions designed to enhance self-efficacy in conjunction with behavioral therapies for insomnia may be worthy of further testing, particularly if this finding is replicated in future studies. We recommend that in research and clinical settings, the presence of another person or pet in the bedroom ought to be assessed to determine whether it may be an important factor in a given patient’s treatment engagement and response.

adherence to insomnia intervention Comprehensive monitoring of alcohol intake and sleep during the first few nights of treatment may also prove to be clinically important and predictive of adherence if replicated. Conflict of Interest Statement Dr. Lichstein is a member of the Merck Insomnia Advisory Board. The other authors have no conflicts of interest to report. We would like to acknowledge the statistical consultation support received by Dr. Gerald McGwin. Dr. Lichstein is a member of the Merck Insomnia Advisory Board. The other authors have no conflicts of interest to report.

References American Academy of Sleep Medicine. (2005). The international classification of sleep disorders and diagnostic and coding manual (2nd ed.). Westchester, IL: Author. Baron, K. G., Reid, K. J., & Zee, P. C. (2012). Exercise to improve sleep in insomnia: Exploration of the bidirectional effects. Journal of Clinical Sleep Medicine, 9(8), 819–824. http://dx.doi.org/10.5664/jcsm.2930 Baron, K. G., Smith, T. W., Berg, C. A., Czajkowski, L. A., Gunn, H., & Jones, C. R. (2011). Spousal involvement in CPAP adherence among patients with obstructive sleep apnea. Sleep and Breathing, 15(3), 525–534. http://dx.doi.org/ 10.1007/s11325-010-0374-z Bixler, E. O., Kales, A., Soldatos, C. R., Kales, J. D., & Healey, S. (1979). Prevalence of sleep disorders in the Los Angeles metropolitan area. American Journal of Psychiatry, 136, 1257–1262. Bootzin, R. R., Epstein, D., & Ward, J. M. (1991). Stimulus control instructions. In P. Hauri (Ed.), Case studies in insomnia (pp. 19–28). New York, NY: Plenum Press. Bouchard, S., Bastien, C., & Morin, C. M. (2003). Self-efficacy and adherence to cognitive-behavioral treatment of insomnia. Behavioral Sleep Medicine, 1(4), 187–199. http://dx.doi.org/ 10.1207/S15402010BSM0104_2 Buman, M. P., Hekler, E. B., Bliwise, D. L., & King, A. C. (2011). Exercise effects on night-to-night fluctuations in self-rated sleep among older adults with sleep complaints. Journal of Sleep Research, 20(1), 28–37. http://dx.doi.org/ 10.1111/j.1365-2869.2010.00866.x Cartwright, R. (2008). Sleeping together: A pilot study of the effects of shared sleeping on adherence to CPAP treatment in obstructive sleep apnea.Journal of Clinical Sleep Medicine, 4(2), 123–127 (Retrieved from www.ncbi.nlm.nih.gov/pmc/ articles/PMC2335391). Dashevsky, B., & Kramer, M. (1997). Patients who discontinue combined behavioral and medicinal treatment of insomnia. Sleep Research, 26, 350. DiMatteo, R. M. (2004). Variations in patients’ adherence to medical recommendations: A quantitative review of 50 years of research. Medical Care, 42(3), 200–209. http://dx.doi.org/ 10.1097/01.mlr.0000114908.90348.f9 Ebrahim, I. O., Shapiro, C. M., Williams, A. J., & Fenwick, P. B. (2013). Alcohol and sleep I: Effects on normal sleep. Alcoholism: Clinical and Experimental Research, 37(4), 539–549. http://dx.doi.org/10.1111/acer.12006 Edinger, J. D., & Means, M. K. (2005). Cognitive-behavioral therapy for primary insomnia. Clinical Psychology Review, 25(5), 539–558. http://dx.doi.org/10.1016/j.cpr.2005. 04.003 Edinger, J. D., Wohlgemuth, W. K., Radtke, R. A., Coffman, C. J., & Carney, C. E. (2007). Dose-response effects of cognitive-behavioral insomnia therapy: A randomized clinical

441

trial. Sleep, 30(2), 203–212. http://dx.doi.org/10.1016/j.cpr. 2005.04.003. Fournier, J. C., DeRubeis, R. J., Shelton, R. C., Hollon, S. D. Amsterdam, J. D., & Gallop, R. (2009). Prediction of response to medication and cognitive therapy in the treatment of moderate to severe depression. Journal of Consulting and Clinical Psychology, 77(4), 775–787. http://dx.doi.org/ 10.1037/a0015401 Gaultney, J. F. (2010). The prevalence of sleep disorders in college students: Impact on academic performance. Journal of American College Health, 59(2), 91–97. http://dx.doi.org/ 10.1080/07448481.2010.483708 Harvey, L., Inglis, S. J., & Espie, C. A. (2002). Insomniacs’ reported use of CBT components and relationship to long-term clinical outcome. Behaviour Research and Therapy, 40, 75–83. http://dx.doi.org/10.1016/S0005-7967(01)00004-3 Hebert, E. A., Vincent, N., Lewycky, S., & Walsh, K. (2010). Attrition and adherence in the online treatment of chronic insomnia. Behavioral Sleep Medicine, 8, 141–150. http://dx.doi.org/10.1080/15402002.2010.487457 Hohagen, F., Rink, K., Käppler, C., Schramm, E., Riemann, D., Weyerer, S., & Berger, M. (1993). Prevalence and treatment of insomnia in general practice: A longitudinal study. European Archives of Psychiatry and Clinical Neuroscience, 242(6), 329–336. http://dx.doi.org/10.1007/BF02190245 Lewis, M. A., McBride, C. M., Pollack, K. I., Puleo, E., Butterfield, R. M., & Emmons, K. M. (2006). Understanding health behavior change among couples: An interdependence and communal coping approach. Social Science and Medicine, 62(6), 1369–1380. http://dx.doi.org/10.1016/j.socscimed. 2005.08.006 Lichstein, K. L., Durrence, H. H., Riedel, B. W., Taylor, D. J., & Bush, A. J. (2004). Epidemiology of sleep: Age, gender, and ethnicity. Mahwah, NJ: Erlbaum. Lichstein, K. L., & Morin, C. M. (2000). Treatment of late-life insomnia. Thousand Oaks, CA: Sage. Matthews, E. E., Arnedt, J. T., McCarthy, M. S., Cuddihy, L. J., & Aloia, M. S. (2013). Adherence to cognitive behavioral therapy for insomnia: A systematic review. Sleep Medicine Reviews. http://dx.doi.org/10.1016/j.smrv.2013.01.001 Matthews, E. E., Schmiege, S. J., Cook, P. F., Berger, A. M., & Aloia, M. S. (2012). Adherence to cognitive behavioral therapy for insomnia (CBTI) among women following primary breast cancer treatment: A pilot study. Behavioral Sleep Medicine, 10(3), 217–229. http://dx.doi.org/10.1080/ 15402002.2012.666220 McChargue, D. E., Sankaranarayanan, J., Visovsky, C. G., Matthews, E. E., Highland, K. B., & Berger, A. M. (2012). Predictors of adherence to a behavioral therapy sleep intervention during breast cancer chemotherapy. Supportive Care in Cancer: Official Journal of the Multinational Association of Supportive Care in Cancer, 20(2), 245–252. http://dx.doi.org/10.1007/s00520-010-1060-1 Meadows, R., Arber, S., Venn, S., Hislop, J., & Stanley, N. (2009). Exploring the interdependence of couples’ rest-wake cycles: An actigraphic study. Chronobiology International, 26(1), 80–92. http://dx.doi.org/10.1080/07420520802678452 Morgan, K., Thompson, J., Dixon, S., Tomeny, M., & Mathers, N. (2003). Predicting longer-term outcomes following psychological treatment for hypnotic-dependent chronic insomnia. Journal of Psychosomatic Research, 54(1), 21–29. http://dx.doi.org/10.1016/S0022-3999(02) 00569-X Morgenthaler, T., Kramer, M., Alessi C, Friedman, L., Boehlecke, B., Brown, T., … Swick, T. (2006). Practice parameters for the psychological and behavioral treatment of insomnia: An update. An American Academy of Sleep

442

ruiter petrov et al.

Medicine report. Sleep, 29(11), 1415–1419. Retrieved from www.journalsleep.org/Articles/291105.pdf. Morin, C. M. (1993). Insomnia: Psychological assessment and management. New York, NY: Guilford Press. Morin, C. M., Belleville, G., Bélanger, L., & Ivers, H. (2011). The Insomnia Severity Index: Psychometric indicators to detect insomnia cases and evaluate treatment response. Sleep, 34(5), 601–608. Retrieved from www.ncbi.nlm.nih. gov/pmc/articles/PMC3079939. Morin, C. M., LeBlanc, M., Daley, M., Gregoire, J. P., & Mérette, C. (2006). Epidemiology of insomnia: Prevalence, self-help treatments, consultations, and determinants of help-seeking behaviors. Sleep Medicine, 7(2), 123–130. http://dx.doi.org/10.1016/j.sleep.2005.08.008 Ohayon, M. M. (2002). Epidemiology of insomnia: What we know and what we still need to learn. Sleep Medicine Reviews, 6(2), 97–111. http://dx.doi.org/10.1053/smrv. 2002.0186 Perlis, M. L., Jungquist, C., Smith, M. T., & Posner, D. (2005). Cognitive behavioral treatment of insomnia: A session-bysession guide. New York, NY: Springer Science + Business Media. Reed, M. B., & Aspinwall, L. G. (1998). Self-affirmation reduces biased processing of health risk information. Motivation and Emotion, 22(2), 99–131. http://dx.doi.org/ 10.1023/A:1021463221281 Riedel, B. W., & Lichstein, K. L. (2001). Strategies for evaluating adherence to sleep restriction treatment for insomnia. Behaviour research and therapy, 39(2), 201–212. http://dx.doi.org/10.1016/S0005-7967(00) 00002-4 Roehrs, T., & Roth, T. (2001). Sleep, sleepiness, and alcohol use.Alcohol Research and Health, 25(2), 101–109. Retrieved from webapps.ou.edu/alcohol/docs/05SleepSleepinessandEton Us101-109.pdf. Rogojanski, J., Carney, C. E., & Monson, C. M. (2012). Interpersonal factors in insomnia: A model for integrating bed partners into cognitive behavioral therapy for insomnia. Sleep Medicine Reviews, 17, 55–64. http://dx.doi.org/ 10.1016/j.smrv.2012.02.003 Roth, T., Zammit, G., Kushida, C., Doghramji, K., Mathias, S. D., Wong, J. M., & Buysse, D. J. (2002). A new questionnaire to detect sleep disorders. Sleep Medicine, 3(2), 99–108. http://dx.doi.org/10.1016/ S1389-9457(01)00131-9 Savard, M. H., Savard, J., Simard, S., & Ivers, H. (2005). Empirical validation of the insomnia severity index in cancer patients. Psychooncology, 14(6), 429–441. http://dx.doi.org/ 10.1002/pon.860 Scott, S. C., Goldberg, M. S., & Mayo, N. E. (1997). Statistical assessment of ordinal outcomes in comparative studies. Journal of Clinical Epidemiology, 50(1), 45–55. http://dx.doi.org/10. 1016/S0895-4356(96)00312-5 Sherman, D. K., Nelson, L. D., & Steele, C. M. (2000). Do messages about health risks threaten the self? Increasing the

acceptance of threatening health messages via self−affirmation. Personality and Social Psychology Bulletin, 26, 1046–1058. Singareddy, R., Vgontzas, A. N., Fernandez-Mendoza, J., Liao, D., Calhoun, S., Shaffer, M. L., & Bixler, E. O. (2012). Risk factors for incident chronic insomnia: A general population prospective study. Sleep Medicine, 13(4), 346–353. http://dx.doi.org/10.1016/j.sleep.2011.10.033 Singleton, R. A., & Wolfson, A. R. (2009). Alcohol consumption, sleep and academic performance among college students. Journal of Studies on Alcohol and Drugs, 70(3), 355–363 Retrieved from www.jsad.com/jsad/download article/Alcohol_Consumption_Sleep_and_Academic_ Performance_Among_College_Students/4614.pdf. Smith, M. T., & Perlis, M. (2006). Who is a candidate for cognitive-behavioral therapy for insomnia? Health Psychology, 25(1), 15–19. http://dx.doi.org/10.1037/ 0278-6133.25.1.15 Smith, M. T., Perlis, M. L., Smith, M. S., Giles, D. E., & Carmody, T. P. (2000). Sleep quality and presleep arousal in chronic pain. Journal of Behavioral Medicine, 23(1), 1–13. http://dx.doi.org/10.1023/A:1005444719169 Spielman, A. J., Saskin, P., & Thorpy, M. J. (1987). Treatment of chronic insomnia by restriction of time in bed. Sleep, 10(1), 45–56 Retrieved from www.journalsleep.org/ ViewAbstract.aspx?pid=25195. Suh, S., Dowdle, C. L., Willett, E. S., Nowakowski, S., Siebern, A. T., Ong, J., … Manber, R. (2011). Helpful components of group cognitive behavioral therapy for insomnia (CBTI) for patients who perceived pain to interfere with sleep. SLEEP, 34, A180 Retrieved from www.journalsleep.org/Resources/ Documents/2011abstractsupplement.pdf. Taylor, D. J., Gardner, C. E., Bramoweth, A. D., Williams, J. M., Roane, B. M., Grieser, E. A., & Tatum, J. I. (2011). Insomnia and mental health in college students. Behavioral Sleep Medicine, 9(2), 107–116. http://dx.doi.org/10. 1080/15402002.2011.557992 Vincent, N., Lewycky, S., & Finnegan, H. (2008). Barriers to engagement in sleep restriction and stimulus control in chronic insomnia. Journal of Consulting and Clinical Psychology, 76(5), 820–828. http://dx.doi.org/10. 1037/0022-006X.76.5.820 Vincent, N., & Lionberg, C. (2001). Treatment preference and patient satisfaction in chronic insomnia. Sleep, 24(4), 411–417 Retrieved from www.ncbi.nlm.nih.gov/pubmed/ 11403525. Vincent, N. K., & Hameed, H. (2003). Relation between adherence and outcome in the group treatment of insomnia. Behavioral Sleep Medicine, 1(3), 125–139. http://dx.doi.org/ 10.1207/S15402010BSM0103_1

R E C E I V E D : April 10, 2013 A C C E P T E D : January 18, 2014 Available online 28 January 2014