ADVERTISEMENT

Issue Navigator

Volume 15 No. 03
Earn CME
Accepted Papers





Scientific Investigations

Effects of a Workplace-Based Sleep Health Program on Sleep in Members of the German Armed Forces

Cornelia Sauter, PhD1; Jens T. Kowalski, PhD2; Michael Stein, PhD2; Stefan Röttger, PhD2; Heidi Danker-Hopfe, PhD1
1Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Competence Center of Sleep Medicine at Campus Benjamin Franklin, Berlin, Germany; 2German Armed Forces Office, Applied Military Psychology and Research Group, Hamburg, Germany

ABSTRACT

Study Objectives:

To develop and evaluate a brief manual-based sleep health program within the workplace health promotion of the German Armed Forces.

Methods:

The sleep health program comprised four weekly group sessions. Sixty-three members (48 males) were randomly allocated to either a treatment group or a waiting control group matching for age, sex, and baseline Pittsburgh Sleep Quality Index (PSQI). The control group had to wait before participating in the sleep health program until the treatment group finished the intervention. Sleep was assessed by ambulatory polysomnography (PSG) as well as with evening and morning protocols at baseline (t0), directly after the treatment group participated in the sleep health program (t1), and after the control group finished participation (t2). The PSQI, the Insomnia Severity Index (ISI), and the Epworth Sleepiness Scale (ESS) were applied at the same three time points, and during a 3-month follow-up evaluation (t3).

Results:

Fifty-seven out of the 63 randomized individuals (42 males, mean age = 40.6 years; complete PSG data: n = 36; complete questionnaire data: n = 39) participated in the sleep health program. Objective wake after sleep onset, sleep efficiency, latency to persistent sleep, self-reported sleep latency, restfulness, PSQI, and ISI scores improved with medium or large effects in both groups. ESS scores decreased with moderate effects in the treatment group only.

Conclusions:

The sleep health program had a positive and stable effect on objective and self-reported sleep parameters, and it is suitable as a preventive measure in members of the German Armed Forces.

Clinical Trial Registration:

Registry: ClinicalTrials.gov; Title: Development and Evaluation of a Sleep-coaching Program; Identifier: NCT02896062; URL: https://clinicaltrials.gov/ct2/show/record/NCT02896062

Citation:

Sauter C, Kowalski JT, Stein M, Röttger S, Danker-Hopfe H. Effects of a workplace-based sleep health program on sleep in members of the German Armed Forces. J Clin Sleep Med. 2019;15(3):417–429.


BRIEF SUMMARY

Current Knowledge/Study Rationale: Sleep health programs are rarely offered in workplaces, although the benefit of good sleep on physical and psychological health is unequivocal. The brief manual-based sleep health program presented here addressed persons with minor to moderate sleep impairments to improve sleep and prevent future sleep disorders.

Study Impact: This workplace-based group intervention was beneficial in improving objective and self-reported measures of sleep in members of the German Armed Forces. The program might increase health resilience by providing helpful strategies in the case of impaired sleep quality and strengthen self-efficacy.

INTRODUCTION

Studies in German soldiers have shown high prevalence rates of impaired sleep quality as measured by the Pittsburgh Sleep Quality Index (PSQI), ranging between 29% to 38% in non-deployed young soldiers, up to 35% during deployment,1 and 41% after deployment.2 Recently, high rates of sleep apnea (47%) and insomnia (26%) in veterans of the United States were reported,3 emphasizing the importance of early prevention of these conditions. Impaired sleep and insomnia symptoms prior to deployment are predictors for the development of mental disorders such as posttraumatic stress disorder (PTSD), depression, or suicide.36 Furthermore, in addition to an increased risk of mental health problems, Seelig et al. demonstrated in a study on sleep and health resilience in a large United States military cohort that insomnia symptoms were associated with lower self-rated health and negative effects on work and deployment.7 Given these negative consequences of impaired sleep in the military context, several attempts to treat sleep disorders in military personnel have been made, but most of them have focused on the treatment of sleep disorders after a return from deployment or in veterans810 or in those with chronic insomnia.11 Pedersen et al.12 suggested evaluating sleep health programs to promote physical and psychological health, and resilience to stress after deployment. In the current study, this type of preventive approach was followed. A short-term manual-based sleep health program was developed and evaluated to improve the sleep of military personnel of the German Armed Forces. In a pilot study, the feasibility of the sleep health program in the workplace health promotion of the German Armed Forces was proven and positive effects on well-being and self-care were observed.13 The sleep health program takes into consideration the participation of military personnel with different causes and levels of impaired sleep. In the current study, the effects of the sleep health program on objective and self-reported measures of sleep were evaluated in a randomized crossover field study.

METHODS

Procedures

Prior to any study procedures, information sessions were held at four different military sites across Germany to describe the sleep health program and to recruit participants for the study. Participants were randomized either to a treatment group or to a waiting control group for each site separately. After the treatment group finished the sleep health program, the waiting control group received the same intervention (Figure 1). The number for each group was restricted to eight participants.

Study design and time schedule.

Both groups underwent evaluations of objective and self-reported sleep parameters at the same time points (t0, t1, t2). The treatment group participated in the sleep health program following t0, and the control group participated following t1. PSG = polysomnography.

jcsm.15.3.417a.jpg

jcsm.15.3.417a.jpg
Figure 1

Study design and time schedule.

(more ...)

Assessments of objective and self-reported sleep parameters occurred in parallel in the treatment and in the waiting control groups: baseline examinations were conducted prior to any intervention (t0), effects of the course were evaluated following participation of the treatment group in the sleep health program (t1), and directly after the end of the participation of the waiting control group in the sleep health program (t2). Questionnaires were applied at the same three time points, and during a 3-month follow-up evaluation (t3).

The study was approved by the Ethics Committee of the Charité – University Medicine Berlin (EA4/115/14). Written informed consent was obtained from all participants prior to any study intervention.

Participants

In total, 81 self-selected employees of the German Armed Forces at four different sites were interested in taking part in the study and signed the informed consent. Inclusion criteria comprised (1) being an employee of the German Armed Forces, (2) adult age (18 years or older), (3) being interested in improving sleep, and (4) being willing and able to attend all four sessions of the sleep health program and undergo the evaluation procedures. Participants with severe depressive and/or somatic symptoms according to the Patient Health-Questionnaire (PHQ)14,15 were excluded.

Intervention: Face-to-Face Group Sleep Health Program

The sleep health program was provided during regular working hours and consisted of four 90-minute sessions, which were delivered with 1-week intervals between sessions.

The four sessions addressed different sleep-related topics and comprised theoretical parts and practical instructions (Table 1). The practical instructions had to be carried out as a type of “homework” between sessions. Guided group discussions, the exchange of experiences between participants and individualized support by a sleep expert, were essential parts of the intervention. All sessions at all sites were conducted by CS, a clinical and health psychologist, who is a certified expert somnologist (European Sleep Research Society and German Sleep Research Society).

Sleep health program sessions and content.

jcsm.15.3.417.t01.jpg

table icon
Table 1

Sleep health program sessions and content.

(more ...)

Assessment Procedures

Patient Health Questionnaire for Depressive Symptoms

Depressive symptoms were evaluated at baseline by the nine-item subscale for depressive symptoms PHQ-9 of the German version16 of the Patient Health Questionnaire.17 Total scores ranging from 0 to 14 represent no to moderate symptoms, whereas scores above 14 are indicative of severe symptoms of depression.14

Patient Health Questionnaire for Somatic Symptoms

The self-administered subscale PHQ-15 of the German version16 of the Patient Health Questionnaire17 was applied at baseline to screen for somatic symptoms. The total score as well as a cutoff score ≤ 14 for “minimal” to “medium” somatic symptom severity versus > 14 indicative of “high” levels of somatic symptom severity15 were considered for further analyses.

Measurement of Objective Sleep Parameters by Ambulatory Polysomnography

Sleep was recorded and evaluated according to the standard of the American Academy of Sleep Medicine (AASM).18 Objective sleep parameters were measured by ambulatory polysomnography (PSG; Somnoscreen plus, Somnomedics, Randersacker, Germany) at three time points (t0, t1, t2) for 2 consecutive nights each. Electrodes and sensors were attached to the participants at the military sites, and participants slept in their familiar surroundings either at home or at the barracks. The recordings were scored by an external and independent competence center for sleep analysis (The Siesta Group Schlafanalyse GmbH, Vienna, Austria), which provided a validated and Food and Drug Administration-approved sleep scoring under expert human supervision according to the AASM standard criteria.19 The expert scorer was blind to the group condition. The PSG outcome parameters comprised time in bed (TIB; from “lights out” until “lights on” in minutes), sleep latency (SL; “lights out” to first epoch of any sleep in minutes), latency to persistent sleep (LPS; latency from lights out to the first 10 minutes of consecutive epochs of sleep), stage R sleep latency (sleep onset to the first epoch of stage R sleep in minutes), total sleep time (TST; any sleep stage different from wake; in minutes), wake after sleep onset (WASO; wakefulness during TIB, minus TST and minus SL, in minutes), and percentage sleep efficiency ([TST / TIB] × 100).

Measurement of Self-Reported Sleep Parameters by Evening-Morning Protocols

Each time participants underwent ambulatory PSG, they filled out the standard version of the evening and morning protocol of the German Sleep Society.20 These protocols are an extended version of the Consensus Sleep Diary, which is recommended by Carney et al.21 and is intended for use parallel to a polysomnographic recording. The following self-reported sleep parameters, which correspond to the appropriate objective target variables, were analyzed: self-reported time in bed (sTIB in minutes), self-reported sleep latency (sSL in minutes), self-reported total sleep time (sTST in minutes), self-reported wake after sleep onset (sWASO, in minutes), and self-reported percentage sleep efficiency (sTST / sTIB) × 100. In addition, the item restfulness was considered for further analyses (five categories from 1 = very restful to 5 = not at all restful).

To monitor their sleep-wake schedule between assessments and to document the effect of the applied exercises, a short version of the evening morning protocol (sleep diary) of the German Sleep Society22 was given to the participants. The sleep diary comprised six short questions in the evening and eight questions in the morning. Participants of both groups filled in their sleep diary in parallel, 1 week before and during the time of the sleep health program, resulting in approximately 10 weeks of self-reported data per participant (Figure 1). Data are not presented here.

Measurement of Self-Reported Sleep Quality by the Pittsburgh Sleep Quality Index

Self-reported sleep quality was assessed by the PSQI.23 The 19-item PSQI was applied at all four time points (t0, t1, t2, t3). For further analyses, the global score as well as categorical data (total score > 5 indicative of “poor sleep quality” versus ≤ 5 indicating that sleep quality is “good”) were assessed.

Measurement of Self-Reported Insomnia Symptoms by the Insomnia Severity Index

The Insomnia Severity Index (ISI)24 was applied four times (t0, t1, t2, t3) throughout the study to assess insomnia symptoms. A cutoff score ≥ 825 was applied to discriminate participants with “no insomnia” from those with “subthreshold” insomnia to moderate or severe insomnia.

Measurement of Self-Reported Daytime Sleepiness by the Epworth Sleepiness Scale

Self-reported daytime sleepiness was evaluated with the Epworth Sleepiness Scale26 (ESS) at all four times of evaluation. In addition to the total score, the cutoff score > 10 was applied to classify participants with increased daytime sleepiness.

Statistical Analyses

Analysis of Baseline Characteristics

Baseline characteristics between different subsamples were compared by applying unpaired t tests in case of normally distributed values and the Mann-Whitney U test for nonparametric data. Chi-square tests or Fisher exact test were used for comparisons of categorical data dichotomized at a cutoff score.

Analysis of Objective and Self-Reported Sleep Parameters

All participants, for whom data were available from the first and second night at t0 and from at least 1 night for t1 and t2 were included in the analysis, resulting in a total number of 36 participants with data (treatment group; n = 20; waiting control group: n = 16). The mean of 2 nights was analyzed. The number of participants and available data for all analyses as well as numbers and reasons for dropouts are presented in Figure 2.

Participant flowchart.

PSG = polysomnography.

jcsm.15.3.417b.jpg

jcsm.15.3.417b.jpg
Figure 2

Participant flowchart.

(more ...)

All objective and self-reported sleep parameters were individually standardized into z-values to control for interindividual differences. Because almost all data were non-normally distributed, nonparametric statistics were applied. Friedman tests were used to test for possible treatment effects within the treatment and the control group separately. Post hoc Wilcoxon matched-pairs signed-rank tests were performed to look for pairwise changes between three different times of measurement. In addition to tests of significance (P values), effects sizes were considered. Bravais-Pearson correlation coefficients r were calculated for comparisons of two groups. In the case of comparisons between three measurement times within the groups (Friedman test), Kendall W was calculated as an effect size estimate.27 All effect sizes were interpreted according to Cohen,28 with values of ≥ 0.1 indicating a “small” effect, ≥ 0.3 “medium” effect, and ≥ 0.5 a “large” effect.

Analysis of Self-Reported Data From Questionnaires

For the longitudinal data analysis of four time points (t0, t1, t2, t3), complete questionnaires from 39 participants were analyzed. Possible differences between the treatment and the control group at baseline were analyzed by using t tests, or by Mann-Whitney U tests in the case of nonparametric distribution of data. Because of the non-normal distribution of the variables derived from the questionnaires at time points t1 to t3, all further data were analyzed with nonparametric tests. For within-group analyses Friedman tests were performed to measure effects across time, and Wilcoxon matched-pairs signed-rank tests for pairwise comparisons between time points. Chi-square tests or Fisher exact tests were used for the categorical data derived by the corresponding cutoff scores. In addition, Bravais-Pearson r, Kendall W, and Cramers V were calculated to measure magnitude of effects. All effect sizes were interpreted according to Cohen.28

Because of the small sample sizes, only effect sizes are reported for objective and self-reported sleep parameters and for questionnaire data. Only results representing medium to large effects across time points are reported and depicted in the figures.

All tests were performed with a double-sided significance level of P < .05. Data were processed statistically using IBM SPSS 23 (IBM Corp., Armonk, New York, United States).

RESULTS

Baseline Participant Characteristics

From 81 originally screened potential participants, 63 were randomized, and finally 57 employees (42 men, 15 women; mean ± standard deviation age: 40.9 ± 10.7 years; range: 18–58 years) participated in the sleep health program and underwent the accompanying evaluation. The 57 participants did not differ from all dropouts (n = 24) with regard to age, PSQI, ESS, and ISI and other baseline scores (Table S1 supplemental material). Reasons for dropouts are presented in Figure 2, and basic sample characteristics of the participants are shown in Table 2. In none of the analysis groups did the treatment group differ significantly from the control group in any of the baseline characteristics. In all analysis groups, sleep quality measured by the mean scores of the PSQI and the ISI was impaired at t0 according to the applied cutoffs (Table 2). Mean values of self-reported daytime sleepiness in the ESS were just below the cutoff score of 10. Mean scores of PHQ-9 and PHQ-15 lay below the cutoff values according to the inclusion cutoff criteria of 14 and even below 10, which indicates “no” or “mild symptoms” of depression or somatic symptoms, respectively.29 Distribution of military ranks and levels of civil servants between treatment group and control group did not differ significantly (Table S2 in supplemental material).

Baseline (t0) characteristics for all participants (n = 57) and for subsamples of the PSG analysis (n = 36) and of the questionnaire analysis (four time points: n = 39).

jcsm.15.3.417.t02.jpg

table icon
Table 2

Baseline (t0) characteristics for all participants (n = 57) and for subsamples of the PSG analysis (n = 36) and of the questionnaire analysis (four time points: n = 39).

(more ...)

Adherence to Sleep Health Program

Although the criteria required the inclusion of only individuals who were able to participate in all four sessions, the participation rate ranged between 91.2% (52 of 57) in the first session, 87.7% (n = 50) in the second, 89.5% (n = 51) in the third, and 84.2% (n = 48) in the last session. The reasons for absence were “unforeseen business matters” (51.9%, n = 14), “vacations” (25.9%, n = 7), or “illness” (18.5%, n = 5). One person simply forgot the date of the first session (3.7%, n = 1).

Objective Sleep Parameters

Results of descriptive statistics of z-values and comparisons across time points of objective sleep parameters are summarized in Table 3. As seen from Table S3 in the supplemental material, median absolute values do not indicate impaired sleep. Nevertheless, improvements occurred in both groups. As expected, the treatment group showed stronger and earlier improvements in most of the objective sleep parameters than the control group: WASO and sleep efficiency already improved from t0 to t1 with medium effects (Table 3, Figure 3A, and Figure 3B). Both parameters further improved throughout the observation period with large effect sizes from t0 to t2. In addition, SL and LPS decreased with medium effects in this group (Figure 3D and Figure 3E).

Results of objective sleep parameters for three time points and results of comparisons between different time points for treatment and control groups separately.

jcsm.15.3.417.t03.jpg

table icon
Table 3

Results of objective sleep parameters for three time points and results of comparisons between different time points for treatment and control groups separately.

(more ...)

Objective sleep parameters.

This figure illustrates the objective sleep parameters (median and interquartile range) that improved with medium or large effects in the treatment group (A, B, C, D, E) and in the control group (A, B, C, E). The treatment group (n = 20) is depicted with a solid line and the control group (n = 16) with a dashed line. Corresponding effect sizes are shown in Table 3.

jcsm.15.3.417c.jpg

jcsm.15.3.417c.jpg
Figure 3

Objective sleep parameters.

(more ...)

In the control group, a reduction in the LPS with a large effect was observed after participation in the sleep health program compared to baseline (t2–t0; Figure 3E). Furthermore, sleep efficiency increased and WASO decreased from t0 to t2―both with moderate effects. Stage R sleep latency decreased with a strong effect in the treatment group from t0 to both t1 and to t2 (Figure 3C), whereas in the control group a moderate effect occurred from t0 to t2 (Figure 3C).

Self-Reported Sleep Parameters

Results of comparisons of self-reported sleep parameters (z-values) across time points are shown in Table 4. Descriptive statistics of absolute values summarized in Table S4 in the supplemental material indicate unimpaired sleep in both groups. Nevertheless, in the treatment group sSL decreased with strong effects across time points (Table 4, Figure 4A), and sWASO and sTST changed from t1 to t2 with medium effects (Figure 4B and Figure 4C). In the control group, sSL decreased with a moderate effect from t0 to t2 (Figure 4A). Although self-reported restfulness improved with a strong effect in the treatment group from t0 to t1, restfulness declined at t1 in the control group and improved from t1 to t2 (Figure 4D).

Results of self-reported sleep parameters for three time points and results of comparisons between different time points for treatment and control groups separately.

jcsm.15.3.417.t04.jpg

table icon
Table 4

Results of self-reported sleep parameters for three time points and results of comparisons between different time points for treatment and control groups separately.

(more ...)

Self-reported sleep parameters.

This figure illustrates the self-reported sleep parameters (median and interquartile range) that improved with medium or large effects in the treatment group (A, B, C, D) and with medium effects in the control group (A, D). The treatment group (n = 20) is depicted with a solid line and the control group (n = 16) with a dashed line. Further corresponding effect sizes are shown in Table 4.

jcsm.15.3.417d.jpg

jcsm.15.3.417d.jpg
Figure 4

Self-reported sleep parameters.

(more ...)

Results From Questionnaires

Pittsburgh Sleep Quality Index

Self-reported sleep quality improved across the four time points in both groups (Table 5, Figure 5A). Whereas effects between t0 and the three measurement times (t1, t2, t3) were all large in the treatment group, in the control group moderate effects were seen from t0 to both, t1 and t3 and large effects were observed after participation in the sleep health program (t2–t0). The median of the PSQI reached a normal level (total score ≤ 5) already at the end of the sleep health program in the treatment group and dropped below the cutoff at the 3-month follow-up in the control group (Figure 5A, Table 5).

Results of total scores of questionnaires for four time points and results of comparisons between different time points for treatment and control groups separately.

jcsm.15.3.417.t05.jpg

table icon
Table 5

Results of total scores of questionnaires for four time points and results of comparisons between different time points for treatment and control groups separately.

(more ...)

Self-reported sleep quality.

This figure illustrates self-reported sleep quality (Pittsburgh Sleep Quality Index (A) and Insomnia Severity Index (B)) and self-reported daytime sleepiness (Epworth Sleepiness Scale (C)) across four time points (treatment group: n = 22, control group: n = 17). Median and interquartile ranges are depicted for the total scores. The results of the treatment group are shown with a solid line and with a dashed line for the control group. Corresponding effects sizes are depicted in Table 5.

jcsm.15.3.417e.jpg

jcsm.15.3.417e.jpg
Figure 5

Self-reported sleep quality.

(more ...)

Insomnia Severity Index

The total score of the ISI declined across the four time points (Table 5 and Figure 5B). The improvements from t0 to t1 were moderate in both groups. In the course of the observation period, further decreases in the scores with moderate to large effects were observed in both groups at t2 and t3 compared to t0. The median total scores reached “normal” level (ISI ≤ 7) at t1 in the treatment group and at t3 in the control group (Table 5; Figure 5B). Comparisons of cutoff scores at all of the four different time points revealed one significant result at t1: the number of participants with insomnia symptoms was higher in the control group (76.5%) than in the treatment group (40.9%) with a moderate effect (χ2 = 4,932; P = .026; Cramer V = 0.36).

Epworth Sleepiness Scale

It is important to note that the median values of the total score are within the normal range of ≤ 10 in both groups at all four time points, respectively (Table 5, Figure 5C). Nevertheless, the ESS scores decline with increasing effects from t0 to t2 and t3 in the treatment group only (Table 5, Figure 5C).

DISCUSSION

In the current study, a 4-week sleep health program has been shown to be effective in improving objective and self-reported sleep with stable and long-lasting effects in members of the German Armed Forces. Although the study population was rather heterogeneous and comprised individuals with different levels and causes of sleep impairment, a significant positive effect of this new preventive method was observed. To the best of our knowledge, studies on sleep health programs in the military context have not yet been published. Furthermore, several sleep health programs have been implemented as part of sleep education programs in children and adolescents,30 but only a few programs are available for the adult population. The workplace-based, face-to-face studies published focus on sleep knowledge and sleep hygiene education31,32 or on cognitive behavioral therapy for insomnia (CBT-I).33,34 Only one of these studies applied objective measures of sleep by using wrist actigraphy in a study on workplace-based group CBT-I,34 but none implemented PSG recordings. A sleep health program with 1,189 firefighters, which comprised a mandatory educational session, a voluntary sleep disorders screening, diagnosis, and treatment if indicated, revealed a reduction in injuries and disability, but did not evaluate objective parameters of sleep.32

Objective Sleep Parameters

From the objective sleep parameters measured by polysomnography, SL (treatment group only), LPS, stage R sleep latency, sleep efficiency, and WASO improved with medium to strong effects in both groups. Furthermore, these improvements persisted at later time points. The results thus are similar to effects observed in a meta-analysis on the efficacy of CBT-I on diary measures and/or PSG in chronic insomnia; in this earlier study, similar significant improvements with medium to large effect sizes with a decrease in SL and WASO were reported, as well as an increase in sleep efficiency at the posttreatment time point.35

Both the treatment group and the control group, which turned into a treatment group after t1, benefited from the sleep health program. In light of the tight time schedule, it is reasonable that the effects on objective sleep parameters were strongest in the treatment group not directly after the end of their participation in the program, but at the following time point (t2), which was approximately 8 weeks later. Not surprisingly, these results clearly indicate that it takes more time than just 4 weeks to practice and change sleep-related behaviors to achieve noticeable improvements in sleep quality. The results of the current study indicate that four sessions are sufficient to induce long-lasting changes; this is in line with a study on dose-response effects of CBT-I, which indicated that four individual, biweekly sessions were most effective.36 There are no studies available on dose-response effects of sleep health programs on PSG outcome parameters.

Interestingly, stage R sleep latency decreased with a strong effect across time in the treatment group and with a moderate effect in the control group. Stage R sleep latency has been demonstrated to be susceptible to first-night effects in healthy individuals in the laboratory.3739 In-home PSG performed in young healthy individuals, the dynamics of habituation of stage R sleep evolved across 4 nights, expressed by a continuous decrease of stage R sleep latency.40 Analyses on possible first-night effects and on variability of sleep parameters in the current study (not reported here) did not confirm an effect on stage R sleep latency. Stage R sleep alterations have been repeatedly described in depressed patients,41 including a shortened stage R sleep latency. Because severe depressive symptoms were part of the exclusion criteria, it is very unlikely that shorter stage R sleep latency was due to depression in the current study. Pharmacological studies on the treatment of chronic insomnia also revealed a decrease of stage R sleep latency.42,43 Hoever et al. in their study hypothesized that this effect might indicate a kind of REM sleep rebound in individuals who were susceptible to chronic partial REM sleep deprivation.42 However, approximately one-fourth of participants did not report insomnia symptoms (baseline ISI < 7). Though it was not the aim of the study, we analyzed respiration and periodic leg movements of the second PSG night and found obstructive sleep apnea (OSA) according to the diagnostic criteria of the AASM (2014)44 in 12.3% of the 57 study participants. OSA was homogenously distributed across the treatment and the control groups (Fisher exact test: P = .439). Despite the inclusion of participants with OSA, the sleep health program still showed good effects on outcome parameters. None of the participants met the AASM criteria for periodic limb movement disorder (2014).44

Self-Reported Sleep Parameters

A decrease in the sSL, which improved in the treatment group after participating in the program with large effects, was also observed in an 8-week worksite-based healthy sleep program, which provided knowledge about normal sleep and common sleep disorders, and key concepts of traditional CBT-I33 to 53 adult members of an employee wellness center. The program was very similar to the one presented here, aside from the additional topics of yoga, tai chi, and qi gong. In a pre-post assessment with a 23-item questionnaire, almost all ratings improved significantly, with the largest changes seen in knowledge about sleep, followed by the number of nights with “poor quality,” initiating and maintaining sleep and competence towards dealing with individual sleep problems.

In the current study, the self-reported feeling of restfulness in the morning and the sSL improved in both groups after their participation in the sleep health program, reaching moderate to strong effects and corroborating a treatment effect on restfulness at t1. Ohayon et al.45 found two main predictors of daytime consequences in individuals reporting sleep disturbances: global sleep dissatisfaction and nonrestorative sleep. In a study by Harvey et al.,46 individuals with and without insomnia rated feeling alert upon waking and during the day as well as feeling rested and restored on waking as the most important factors when judging sleep quality. Therefore, the improvements in restfulness point to a crucial effect of the sleep health program on self-reported sleep quality.

A further treatment effect was observed with regard to insomnia severity in the ISI. The improvements with large effects in the ISI after the participation in the sleep health program in both groups may be accounted for by the contents of one session on cognitive strategies, which are also applied in CBT-I, (eg, stimulus control and relaxation techniques). These results are in line with the repeatedly shown positive effects of CBT-I in different meta-analyses.35,47

The control group improved only in one of the seven objective and in none of the self-reported sleep parameters at t1, which was expected because the group was still waiting for the intervention to begin. Nevertheless, moderate improvements were already observed at t1 in the PSQI and ISI. Considering that the control group already started to fill in the sleep diary at baseline in parallel with the treatment group, a possible self-monitoring effect of completing a sleep diary every evening and morning might have led to insights concerning potentially disturbing sleep-related behavior48 resulting in behavioral modifications in some of the participants. After participation in the sleep health program, large improvements occurred in the control group in the PSQI and ISI, and in the LPS. The further improvements in all questionnaires in both groups in the 3-month follow-up, except for the ESS in the control group, indicate a stable effect, which became larger across time. Although excessive daytime sleepiness and strategies against it were the main topic of the last session, median ESS scores changed with moderate effects only in the treatment group. Regardless, the median ESS scores lay within the normal range in both groups at any time point.

Limitations and Strengths

One of the limitations of the study was that participants could not be completely randomized by an automatic or predefined algorithm. For reasons related to time and personnel restrictions, sites could only participate in the study consecutively, not concurrently, hampering overall randomization. For the stated reasons, it was also not possible to delay the entry of the control subjects into the program to the end of t2 or even later at the 3-month follow-up, as would be desirable.

Furthermore, the amount of complete PSG data of all planned 6 nights per person was very small; therefore, only a reduced data set could be analyzed. Nevertheless, the data regarding objective and self-reported sleep parameters are encouraging for the further implementation of the sleep health program within workplace health promotion efforts.

One advantage of the sleep health program is that there is no need for comprehensive medical diagnostic procedures in the run-up to the intervention. As seen from the PSG data (Table S3 and Table S4) and the only slightly elevated scores in the PSQI and ISI at baseline, the sleep quality of most of the study population was only minimally impaired. The sleep health program is, therefore, suitable for individuals with subsyndromal sleep disturbances and might have preventive effects. Nevertheless, at the end of the sleep health program further diagnostic or treatment steps might be recommended for some of the participants (eg, OSA). Therefore, trained personnel should offer the sleep health program to identify those who might have sleep disorders requiring further diagnostics and/or treatment. One strength of the study was that all contents were imparted by a clinical and health psychologist and certified expert in sleep medicine (CS), ruling out the effect of having different coaches.

The total amount of time participants spend with the sleep health program comprises approximately 8 hours distributed over a time period of 3 weeks (4 sessions: 90 min/wk; keeping a sleep diary: maximum 5 min/d during participation in the intervention; practicing of techniques if appropriate: several min/d). Most of the members of the German Armed Forces are used to almost-daily exercise, which may help with adherence to the program (“military discipline”). In the current study, participation was voluntary, and as in nonmilitary samples, it became clear that compliance mainly depends on personal motivation and/or psychological strain. The opportunity to engage with the sleep health program during worktime within the workplace health promotion definitely facilitated participation.

CONCLUSIONS

The workplace-based sleep health program on sleep has positive effects on objective and self-reported measures of sleep in members of the German Armed Forces. Meanwhile, the sleep health program is disseminated by psychologists of the German Armed Forces, who received an intensive training by HDH and CS in a 4-day “train-the-coaches” seminar. Although there were only four sessions to be attended during working hours at each site, quite a high percentage of individuals were not able to take part in the sleep health program due to (unforeseen) business matters or duty-related travels. Therefore, and for the purpose of broader and easier accessibility, an electronic-based version of the program is currently being implemented. By analogy with the “stepped care model”49 in insomnia treatment, these different steps of applying the sleep health program50 should improve the availability for most of the military service members in need of support to improve their sleep.

DISCLOSURE STATEMENT

Institution where work was conducted: Charité – Universitätsmedizin Berlin, Competence Centre of Sleep Medicine, Hindenburgdamm 30, 12203 Berlin, Germany and at the German Armed Forces Office, Applied Military Psychology and Research Group, Holstenhofweg 85, 22043 Hamburg, Germany. All authors have read and approved this manuscript. This study was funded by the German Federal Ministry of Defense (M/SAKE/EA002). The authors report no conflicts of interest.

ABBREVIATIONS

AASM

American Academy of Sleep Medicine

CBT-I

cognitive behavioral therapy for insomnia

ESS

Epworth Sleepiness Scale

ISI

Insomnia Severity Index

LPS

latency to persistent sleep

OSA

obstructive sleep apnea

PHQ-15

Patient Health Questionnaire for Somatic Symptoms

PHQ-9

Patient Health Questionnaire for Depressive Symptoms

PSG

polysomnography

PSQI

Pittsburgh Sleep Quality Index

PTSD

posttraumatic stress disorder

REM

rapid eye movement

SL

sleep latency

sSL

self-reported sleep latency

sTIB

self-reported time in bed

sTST

self-reported total sleep time

sWASO

self-reported wake after sleep onset

t0

baseline

t1

directly after the intervention was finished in the treatment group

t2

directly after the intervention was finished in the control group

t3

3-month follow-up

TIB

time in bed

TST

total sleep time

WASO

wake after sleep onset

ACKNOWLEDGMENTS

The authors thank all participants in the sleep health intervention. The authors would like to acknowledge all members of the German Armed Forces who supported the study, especially the site coordinators. The authors also thank the following research staff members of the Competence Center of Sleep Medicine: Dipl. Biol. Juliane Schlee, MTA-F Konstanze Küchler and Cand. B. Sc. Martin Rätzsch for conducting the ambulatory polysomnography and Dr. Marie-Luise Hansen and Dr. Anita Peter for their medical supervision.

REFERENCES

1 

Danker-Hopfe H, Sauter C, Kowalski JT, et al. Sleep quality of German soldiers before, during and after deployment in Afghanistan-a prospective study. J Sleep Res. 2017;26(3):353–363. [PubMed]

2 

Heinrich A, Knappe S, Trautmann S, Schönfeld S, Hauffa R, Wittchen HU. Schlafprobleme bei Soldaten und die Rolle traumatischer Ereignisse bei Auslandseinsätzen. [Sleeping problems of German soldiers and the role of deployment-related traumatic events.] (in German). Z Klein Psychol Psychother. 2015;44(2):121–130

3 

Alexander M, Ray MA, Hebert JR, et al. The National Veteran Sleep Disorder Study: descriptive epidemiology and secular trends, 2000-2010. Sleep. 2016;39(7):1399–1410. [PubMed Central][PubMed]

4 

Gehrman P, Seelig AD, Jacobson IG, et al. Predeployment sleep duration and insomnia symptoms as risk factors for new-onset mental health disorders following military deployment. Sleep. 2013;36(7):1009–1018. [PubMed Central][PubMed]

5 

Koffel E, Polusny MA, Arbisi PA, Erbes CR. Pre-deployment daytime and nighttime sleep complaints as predictors of post-deployment PTSD and depression in national guard troops. J Anxiety Disord. 2013;27(5):512–519. [PubMed]

6 

Bramoweth AD, Germain A. Deployment-related insomnia in military personnel and veterans. Curr Psychiatry Rep. 2013;15(10):401[PubMed]

7 

Seelig AD, Jacobson IG, Donoho CJ, Trone DW, Crum-Cianflone NF, Balkin TJ. Sleep and health resilience metrics in a large military cohort. Sleep. 2016;39(5):1111–1120. [PubMed Central][PubMed]

8 

Bootzin RR. Implementing CBTI at the Veterans Health Administration. J Clin Sleep Med. 2012;8(2):219–220. [PubMed Central][PubMed]

9 

Haynes PL, Kelly M, Warner L, Quan SF, Krakow B, Bootzin RR. Cognitive behavioral social rhythm group therapy for veterans with posttraumatic stress disorder, depression, and sleep disturbance: results from an open trial. J Affect Disord. 2016;192:234–243. [PubMed]

10 

Phelps AJ, Varker T, Metcalf O, Dell L. What are effective psychological interventions for veterans with sleep disturbances? A rapid evidence assessment. Mil Med. 2017;182(1):e1541–e1550. [PubMed]

11 

Taylor DJ, Peterson AL, Pruiksma KE, et al. Internet and in-person cognitive behavioral therapy for insomnia in military personnel: a randomized clinical trial. Sleep. 2017;40(6)

12 

Pedersen ER, Troxel WM, Shih RA, Pinder E, Lee D, Geyer L. Increasing resilience through promotion of healthy sleep among service members. Mil Med. 2015;180(1):4–6. [PubMed Central][PubMed]

13 

Röttger S, Maier J, Krex-Brinkmann L, et al. The benefits of sleep coaching in workplace health promotion. J Public Health. 2017;25(6):685–691

14 

Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613. [PubMed Central][PubMed]

15 

Kroenke K, Spitzer RL, Williams JB. The PHQ-15: validity of a new measure for evaluating the severity of somatic symptoms. Psychosom Med. 2002;64(2):258–266. [PubMed]

16 

Gräfe K, Zipfel S, Herzog W, Löwe B. Screening psychischer Störungen mit dem “Gesundheitsfragebogen für Patienten (PHQ-D)“. [Screening for psychiatric disorders with the Patient Health Questionnaire (PHQ). Results from the German validation study.] (in German). Diagnostica. 2004;50(4):171–181

17 

Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary care evaluation of mental disorders. Patient Health Questionnaire. JAMA. 1999;282(18):1737–1744. [PubMed]

18 

Silber MH, Ancoli-Israel S, Bonnet MH, et al. The visual scoring of sleep in adults. J Clin Sleep Med. 2007;3(2):121–131. [PubMed]

19 

Anderer P, Moreau A, Woertz M, et al. Computer-assisted sleep classification according to the standard of the American Academy of Sleep Medicine: Validation study of the AASM version of the Somnolyzer 24 x 7. Neuropsychobiology. 2010;62(4):250–264. [PubMed]

20 

Hoffmann RM, Müller T, Hajak G, Cassel W. Abend-Morgenprotokolle in Schlafforschung und Schlafmedizin — ein Standardinstrument für den deutschsprachigen Raum. [Sleep logs in sleep research and sleep medicine.] (in German). Somnologie. 1997;1(3):103–109

21 

Carney CE, Buysse DJ, Ancoli-Israel S, et al. The consensus sleep diary: Standardizing prospective sleep self-monitoring. Sleep. 2012;35(2):287–302. [PubMed Central][PubMed]

22 

Liendl S, Hoffmann RM. Compliance-Probleme bei der Bearbeitung von Abend-Morgen-Protokollen—Entwicklung einer Kurzversion der Standardprotokolle der DGSM. [Sleep logs: About problems with compliance — Development of a short version of standard questionnaires.] (in German). Somnologie - Schlafforschung und Schlafmedizin. 1999;3(2):73–77

23 

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(2):193–213. [PubMed]

24 

Bastien CH, Vallieres A, Morin CM. Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med. 2001;2(4):297–307. [PubMed]

25 

Morin CM, Belleville G, Belanger L, Ivers H. The Insomnia Severity Index: psychometric indicators to detect insomnia cases and evaluate treatment response. Sleep. 2011;34(5):601–608. [PubMed Central][PubMed]

26 

Johns MW. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep. 1991;14(6):540–545. [PubMed]

27 

Tomczak M, Tomczak E. The need to report effect size estimates revisited. An overview of some recommended measures of effect size. Trends Sport Sci. 2014;1(21):19–25

28 

Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates Inc.; 1988.

29 

Kroenke K, Spitzer RL, Williams JB, Lowe B. The Patient Health Questionnaire somatic, anxiety, and depressive symptom scales: a systematic review. Gen Hosp Psychiatry. 2010;32(4):345–359. [PubMed]

30 

Gruber R. School-based sleep education programs: a knowledge-to-action perspective regarding barriers, proposed solutions, and future directions. Sleep Med Rev. 2017;36:13–28. [PubMed]

31 

Chen PH, Kuo HY, Chueh KH. Sleep hygiene education: efficacy on sleep quality in working women. J Nurs Res. 2010;18(4):283–289. [PubMed]

32 

Sullivan JP, O'Brien CS, Barger LK, et al. Randomized, prospective study of the impact of a sleep health program on firefighter injury and disability. Sleep. 2017;40(1)

33 

Steffen MW, Hazelton AC, Moore WR, Jenkins SM, Clark MM, Hagen PT. Improving sleep: Outcomes from a worksite healthy sleep program. J Occup Environ Med. 2015;57(1):1–5. [PubMed]

34 

Schiller H, Soderstrom M, Lekander M, Rajaleid K, Kecklund G. A randomized controlled intervention of workplace-based group cognitive behavioral therapy for insomnia. Int Arch Occup Environ Health. 2018;91(4):413–424. [PubMed Central][PubMed]

35 

Trauer JM, Qian MY, Doyle JS, Rajaratnam SM, Cunnington D. Cognitive behavioral therapy for chronic insomnia: a systematic review and meta-analysis. Ann Intern Med. 2015;163(3):191–204. [PubMed]

36 

Edinger JD, Wohlgemuth WK, Radtke RA, Coffman CJ, Carney CE. Dose-response effects of cognitive-behavioral insomnia therapy: a randomized clinical trial. Sleep. 2007;30(2):203–212. [PubMed]

37 

Mendels J, Hawkins DR. Sleep laboratory adaptation in normal subjects and depressed patients (“first night effect”). Electroencephalogr Clin Neurophysiol. 1967;22(6):556–558. [PubMed]

38 

Morgenthaler T, Kramer M, Alessi C, et al. Practice parameters for the psychological and behavioral treatment of insomnia: an update. An American Academy of Sleep Medicine report. Sleep. 2006;29(11):1415–1419. [PubMed]

39 

Toussaint M, Luthringer R, Schaltenbrand N, et al. First-night effect in normal subjects and psychiatric inpatients. Sleep. 1995;18(6):463–469. [PubMed]

40 

Le Bon O, Staner L, Hoffmann G, et al. The first-night effect may last more than one night. J Psychiatr Res. 2001;35(3):165–172. [PubMed]

41 

Palagini L, Baglioni C, Ciapparelli A, Gemignani A, Riemann D. REM sleep dysregulation in depression: state of the art. Sleep Med Rev. 2013;17(5):377–390. [PubMed]

42 

Hoever P, Dorffner G, Benes H, et al. Orexin receptor antagonism, a new sleep-enabling paradigm: a proof-of-concept clinical trial. Clin Pharmacol Ther. 2012;91(6):975–985. [PubMed Central][PubMed]

43 

Bettica P, Squassante L, Groeger JA, Gennery B, Winsky-Sommerer R, Dijk DJ. Differential effects of a dual orexin receptor antagonist (SB-649868) and zolpidem on sleep initiation and consolidation, SWS, REM sleep, and EEG power spectra in a model of situational insomnia. Neuropsychopharmacology. 2012;37(5):1224–1233. [PubMed Central][PubMed]

44 

American Academy of Sleep Medicine. International Classification of Sleep Disorders. 3rd ed. Darien, IL: American Academy of Sleep Medicine; 2014.

45 

Ohayon MM, Riemann D, Morin C, Reynolds CF 3rd. Hierarchy of insomnia criteria based on daytime consequences. Sleep Med. 2012;13(1):52–57. [PubMed]

46 

Harvey AG, Stinson K, Whitaker KL, Moskovitz D, Virk H. The subjective meaning of sleep quality: a comparison of individuals with and without insomnia. Sleep. 2008;31(3):383–393. [PubMed Central][PubMed]

47 

Koffel EA, Koffel JB, Gehrman PR. A meta-analysis of group cognitive behavioral therapy for insomnia. Sleep Med Rev. 2015;19:6–16. [PubMed]

48 

Mairs L, Mullan B. Self-monitoring vs. implementation intentions: a comparison of behaviour change techniques to improve sleep hygiene and sleep outcomes in students. Int J Behav Med. 2015;22(5):635–644. [PubMed]

49 

Espie CA. “Stepped care”: A health technology solution for delivering cognitive behavioral therapy as a first line insomnia treatment. Sleep. 2009;32(12):1549–1558. [PubMed Central][PubMed]

50 

Danker-Hopfe H, Kowalski J, Stein M, Röttger S, Sauter C. Development, implementation, and evaluation of a sleep coaching program for the German armed forces. Somnologie. 2018;22(1):36–44


Supplemental Material

Supplemental Material

(.pdf | 183 KB)