Skip to main content
NCBI home page
VA Author Manuscripts logoLink to VA Author Manuscripts
. Author manuscript; available in PMC: 2023 Aug 9.
Published in final edited form as: Def Peace Econ. 2017 Jul 12;29:10.1080/10242694.2017.1349633. doi: 10.1080/10242694.2017.1349633

Clinical Characterization of Insomnia among Veterans with PTSD: Identifying Risk Factors for Diagnosis and Treatment with Sedative-Hypnotics

Adam D Bramoweth a,b, James Luther a,c, Barbara H Hanusa a, Jon D Walker a, Charles W Atwood Jr b,d, Anne Germain e
PMCID: PMC10411149  NIHMSID: NIHMS1919544  PMID: 37560405

Abstract

Insomnia is prevalent among Veterans with post-traumatic stress disorder (PTSD), it exacerbates PTSD symptoms, and it contributes to impaired functioning and quality of life. To improve treatment outcomes, it is important to identify risk factors for insomnia and sedative-hypnotic use. Classification and regression trees and logistic regression models were used to identify variables associated with insomnia or sedative-hypnotic use. Key findings include low insomnia diagnosis rates (3.5–5.6%) and high rates of sedative-hypnotics (44.2–49.0%). Younger Veterans and those without a breathing-related sleep disorder (BRSD) were more likely to receive an insomnia diagnosis. Veterans with greater service connection and those with an alcohol/substance use disorder were more likely to be prescribed sedative-hypnotics. Interaction terms may have identified potential groups at risk of being under-diagnosed with insomnia (i.e. non-black Veterans with psychiatric co-morbidity, black Veterans without psychiatric co-morbidity) as well as groups at risk for sedative-hypnotic use (i.e. younger Veterans without BRSD). In sum, Veterans with PTSD have high rates of sedative-hypnotic use despite minimal evidence they are effective. This is counter to recommendations indicating behavioral interventions are the first-line treatment. Policy changes are needed to reduce use of sedative-hypnotics and increase access to behavioral insomnia interventions.

Keywords: Insomnia, PTSD, risk factors, sedatives and hypnotics, Veterans

Introduction

Post-traumatic stress disorder (PTSD) is one of the defining health conditions for the men and women who serve in the military, especially those who are deployed to theaters of combat. Estimates of PTSD vary by conflict, yet it remains prevalent for older and younger generations of service members: 30%, Vietnam; 12%, Operation Desert Storm, and 11–20%, Operations Iraqi Freedom (OIF), Enduring Freedom (OEF), and New Dawn (OND; Kang et al. 2003; Kulka et al. 1990; Tanielian and Jaycox 2008). PTSD often results in significant social, emotional, health, and functional consequences, including poor relationships, difficulty obtaining and maintaining employment, reduced health-related quality of life, higher rates of medical problems like cardio-respiratory symptoms, musculoskeletal pain, and gastrointestinal health, as well as reduced executive function (Aupperle et al. 2012; Pacella, Hruska, and Delahanty 2013; Rodriguez, Holowka, and Marx 2012). Individuals with PTSD are also at significant risk of having a co-morbid psychiatric disorder (e.g. depression, substance use) and are at Increased risk for suicide (Sareen et al. 2005, 2007). Another common and challenging aspect of PTSD is the associated sleep problems, particularly insomnia (Gehrman 2016).

Difficulty falling asleep, staying asleep, and poor quality sleep are among the most frequently reported symptoms of PTSD. Sleep dysfunction has even been posited as the hallmark symptom of PTSD (Germain 2013). However, insomnia, in many cases is actually an independent problem, a co-morbid disorder that requires its own focused treatment. Whether a symptom of PTSD or a co-morbid disorder, the vast majority of active duty Service Members and Veterans with PTSD, sub-syndromal PTSD, trauma exposure, or simply an overseas deployment, report insomnia-related sleep difficulties. A study of Vietnam Veterans with PTSD found 44% with sleep onset insomnia and 91% with sleep maintenance insomnia (Neylan et al. 1998). Another study with trauma-exposed OEF/OIF Veterans found that difficulty initiating and maintaining sleep was strongly associated with PTSD symptom severity (Gellis et al. 2010). Furthermore, a recent study that confirmed a PTSD diagnosis in VA medical records with the Structured Clinical Interview for DSM-IV (SCID), found that approximately 90% of OEF/OIF Veterans with PTSD per the SCID had difficulty falling or staying asleep (Holowka et al. 2014). Research findings also indicate insomnia as a risk factor, or prodromal symptom, for PTSD. A study of service members returning from Iraq or Afghanistan found that greater insomnia symptoms upon return from deployment predicted greater PTSD symptom severity 3-months later (McLay, Klam, and Volkert 2010). Lastly, in a study of U.S. OIF combat soldiers, greater insomnia severity at 4-months post-deployment predicted greater PTSD severity 8-months later (Wright et al. 2011).

Co-morbid insomnia impacts PTSD treatment outcomes as well. Treating PTSD with selective serotonin reuptake inhibitors, as well as evidence-based psychotherapies (i.e. prolonged exposure therapy, cognitive processing therapy), often fail to improve insomnia symptoms even when non-sleep PTSD symptoms improve (Davidson et al. 2001; Ursano et al. 2004; Zayfert and DeViva 2004). These findings offer additional evidence that insomnia is often co-morbid with PTSD rather than simply a symptom. Fortunately, several studies have now shown that cognitive behavioral interventions in Veterans with PTSD, that focus solely on insomnia and other sleep-disruptions (e.g. nightmares), were able to significantly improve the sleep-related symptoms (Germain et al. 2014; Talbot et al. 2014; Ulmer, Edinger, and Calhoun 2011). These studies highlight the importance and necessity of treating sleep-related disturbances in Veterans with PTSD with sleep-focused interventions. Unexpectedly, these studies also show preliminary evidence of the potential to improve PTSD symptom severity by focusing on insomnia and sleep disturbance. In a study that compared an insomnia and nightmare intervention to treatment as usual, Ulmer and colleagues found that PTSD symptom severity was significantly reduced, as measured by the PTSD Checklist (Ulmer, Edinger, and Calhoun 2011). Talbot and colleagues also found a significant reduction in PTSD severity, per the Clinician-Administered PTSD Scale, following an insomnia intervention; however, the treatment group did not differ from the waitlist control group (Talbot et al. 2014). While, more research is necessary, these findings suggest that insomnia may be an important treatment target among Service Members and Veterans with PTSD.

Notably, these studies utilized cognitive and behavioral treatments (i.e. cognitive behavioral therapy for insomnia [CBTI] and imagery rehearsal therapy); CBTI is the recommended first line of intervention for insomnia per the National Institutes of Health (NIH) and the American College of Physicians (NIH 2005; Qaseem et al. 2016). However, sedative-hypnotic medications remain the most common treatment for insomnia, including insomnia co-morbid with PTSD. One large study (n = 274,297) analyzing administrative data of Veterans with PTSD found 41% were prescribed a sedative-hypnotic (Mohamed and Rosenheck 2008). The best predictors for receiving a sedative-hypnotic were a diagnosis of an anxiety disorder and disability/service connection rating of 51–100%, Although diagnosis of insomnia was not measured, the authors state that Veterans likely received medications ‘for symptomatic treatment of insomnia’ and that ‘some prescribing is driven by symptoms, such as insomnia’ (Mohamed and Rosenheck 2008, 963–964). Another large study (n = 5,531,379) in a general Veteran sample supports these findings. The authors found that almost 14% of Veterans received a sedative-hypnotic medication yet only 3.4% had an insomnia diagnosis (Hermes and Rosenheck 2014). Furthermore, presence of an insomnia diagnosis greatly increased the likelihood, eightfold, of being prescribed a sedative-hypnotic. A diagnosis of insomnia was also linked to a significant increase, almost double, in the number of any psychotropic medications prescribed (adjusted means: 9.9 [insomnia] vs. 5.5 [no insomnia]). While, there was no sub-analysis of medication use by psychiatric disorder, almost a quarter of Veterans with an insomnia diagnosis had co-morbid PTSD, second only to depressive disorders. Veterans with PTSD also had higher odds of receiving an insomnia diagnosis (OR = 1.50, 95% CI 1.48–1.52). Despite the high use of sedative-hypnotics in Veterans with PTSD, there is limited evidence of the effectiveness of pharmacotherapy to treat PTSD, especially insomnia symptoms (Davidson et al. 2001; Ursano et al. 2004; Zayfert and DeViva 2004). Furthermore, there are risks (e.g. tolerance/dependence; Carson et al. 2008) and adverse effects (e.g. daytime fatigue, impaired cognitive and psychomotor functioning; Riemann and Perlis 2009) related to sedative-hypnotic use.

Based on the limited knowledge of how insomnia is diagnosed and treated in Veterans with PTSD, the goal of this study was better characterize Veterans based on the presence of an insomnia diagnosis and/or treatment with sedative-hypnotics. Specifically, the aims were to (1) measure the prevalence of a documented insomnia diagnosis, (2) measure the frequency of common sedative-hypnotic medications indicated for treatment of insomnia, and (3) identify demographic and disorder variables as potential factors for receiving an insomnia diagnosis and/or treatment with sedative-hypnotics.

Method

This study was approved by the VA Pittsburgh Healthcare System (VAPHS) Institutional Review Board. Data were collected from a cohort of Veterans who received care at VAPHS and affiliated Community Based Outpatient Clinics in the calendar year 2007. Within the archival cohort (n = 38,879), data were collected retrospectively (pre-2007) until the first date the Veteran entered the VAPHS system and prospectively (post-2007) until 12/31/2011 or the Veteran left the system (i.e. stopped receiving care at VAPHS, deceased). Administrative data from the VAPHS electronic medical records was collected on demographic information, psychiatric diagnoses, sleep disorders, and sedative-hypnotic usage. Veterans’ service connection, the level of disability (0–100%) due to illness or injury incurred or aggravated during military service, was also collected (Department of Veterans Affairs 2014). Psychiatric disorders were broadly clustered into DSM-IV-TR diagnostic categories, including: depression, anxiety, PTSD, adjustment, bipolar, and alcohol and substance disorders. Insomnia was its own diagnostic category as was breathing-related sleep disorders (BRSDs). Data collected from electronic medical records for each sleep and psychiatric disorder included the number of times a diagnosis was used during clinical encounters (e.g. five psychology appointments linked to a diagnosis of depression) and the first date of the diagnosis. Information on sedative-hypnotic medications to treat insomnia was collected for: eszopiclone, zaleplon, and zolpidem (z-drugs), trazodone, and temazepam. Data for each medication included dose, first date of a filled prescription, number of filled prescriptions, and duration of use (days).

To create the initial PTSD subsample from the larger cohort (n = 38,879), each Veteran needed ≥2 diagnoses of PTSD in their electronic medical record (n = 2516). For the current analyses, two sub-samples of Veterans were created from those with PTSD. The two sub-samples were based on the timing of the dependent variables (Figure 1): (1) a diagnosis of insomnia (n = 2304) and (2) a prescription for a sedative-hypnotic medication (n = 2068). Due to the temporal relationship between the independent and dependent variables for each analysis, it was possible for a Veteran to be in both sub-samples (n = 2047), which contributed to the different sample sizes. For analyses involving a diagnosis of insomnia as the dependent variable, all potential independent variables (e.g. PTSD and other psychiatric diagnoses, prescription medication) needed to occur prior to the date of the first insomnia diagnosis. For analyses involving sedative-hypnotic medications as the dependent variable, all independent variables (e.g. PTSD, insomnia, other psychiatric diagnoses) needed to occur prior to the first prescription fill date for a sedative-hypnotic medication (i.e. z-drug, trazodone, or temazepam). All psychiatric and sleep disorder variables required ≥2 lifetime diagnoses to be present in the medical record; this is consistent with methods used in other studies of VA administrative data (Frayne et al. 2010; Gravely et al. 2011). Sedative-hypnotics (z-drugs, trazodone, and temazepam) required a prescription for >30 days. These thresholds were established in order to identify chronic cases and emphasize outpatient visits (e.g. a 3 day prescription for zolpidem could be indicative of use during an in-patient hospitalization). This two-sample analysis allowed for identification of risk factors for both insomnia diagnosis and sedative-hypnotic medication use, while still being able to include one as a potential independent variable for the other.

Figure 1.

Figure 1.

Open in a new tab

Flow chart of Veterans with PTSD for each analysis.

Analysis Plan

Analyses were conducted in an exploratory fashion to identify variables and variable interactions that might predict the diagnosis of insomnia and/or treatment of insomnia with sedative-hypnotic medications. Descriptive statistics were calculated as means and standard deviations for continuous measures, and as frequencies and percentages for categorical measures. Tests of association included Student’s t or Mann–Whitney U for continuous measures depending on whether distributional assumptions are met; Chi-Square or Fisher’s Exact were used for categorical measures depending on whether expected cell sizes met distributional assumptions. All demographic and diagnostic variables were analyzed by a classification and regression tree (CART®) model in an effort to identify which variables were most associated with either outcome – insomnia diagnosis or sedative-hypnotic use – and to discover interaction terms. Trees were permitted to grow until all nodes included only 10% of the sample. All variables composing two-way interactions were included in logistic regression models. CART analyses were conducted using CART® Classification and Regression Trees (Basic SPM v7.0, Salford Systems, San Diego, CA, USA) and statistical analyses were conducted using SAS v9 (SAS Institute Inc., Cary, NC, USA).

Results

Insomnia Diagnosis Analysis

The sample size for the insomnia diagnosis analysis, in the context of Veterans with PTSD, was 2304 Veterans. Demographic characteristics are listed in Table 1 and co-morbidities (lifetime) are listed in Table 2. One hundred thirty Veterans were diagnosed with insomnia (5.6%) and 1129 were prescribed a sedative-hypnotic medication (49.0%). There was one significant demographic difference: Veterans with PTSD with no documented service connected disability were significantly more likely to receive an insomnia diagnosis than Veterans with PTSD with a service connected disability (7.6% vs. 5.2%, p = 0.041; Table 1). In regards to psychiatric disorders, the number of psychiatric co-morbidities was significantly related to insomnia diagnosis (Table 2). Veterans with PTSD plus no additional co-morbid psychiatric disorders were more likely to have an insomnia diagnosis than Veterans with PTSD with ≥1 co-morbid psychiatric disorder. Additionally, for Veterans with PTSD and a BRSD (lifetime), the vast majority (98.2%) had obstructive sleep apnea, were significantly less likely to have an insomnia diagnosis (1.8% vs. 6.4%, p < 0.001; Table 2).

Table 1.

Demographic characteristics by insomnia diagnosis.

Measure Total (N = 2304) Insomnia diagnosis Analysis
Yes (N = 130) No (N = 2174) Test statistic p
Age 51.4 ± 11.9 51.0 ± 10.9 51.5 ± 12.0 t(2302) = 0.41 0.685
Sex χ2(1)<0.01 0.974
 Male 2199 (95.4) 124 (5.6) 2075 (94.4)
 Female 105 (4.6) 6 (5.7) 99 (94.3)
Race χ2(2) = 1.60 0.450
 Not Black (white + other)a 1848 (82.6) 111 (6.0) 1737 (94.0)
 Black 300 (13.4) 13 (4.3) 287 (95.7)
Marital status χ2(3) = 1.51 0.679
 Married 1225 (53.2) 71 (5.8) 1154 (94.2)
 Separated/Divorced 643 (27.9) 31 (4.8) 612 (95.2)
 Never married 236 (10.3) 16 (6.8) 220 (93.2)
 Widowed 197 (8.6) 12 (6.1) 185 (93.9)
Service branch χ2(3) = 2.61 0.456
 Army 1421 (62.1) 74 (5.2) 1347 (94.8)
 Marine Corps 493 (21.6) 34 (6.9) 459 (93.1)
 Navy 225 (9.8) 15 (6.7) 210 (93.3)
 Air Force 148 (6.5) 7 (4.7) 141 (95.3)
Service period χ2(2) = 2.38 0.305
 WWII/Korea 379 (16.5) 16 (4.2) 363 (95.8)
 Vietnam/Post-Vietnam 1665 (72.4) 96 (5.8) 1569 (94.2)
 Persian Gulf 257 (11.2) 18 (7.0) 239 (93.0)
Service duration (years) 3.39 ± 3.39 4.09 ± 4.51 3.35 ± 3.31 U(1) = 3.57 0.059
Combat χ2(1) = 1.13 0.288
 Yes 594 (29.1) 29 (4.9) 565 (95.1)
 No 1446 (70.9) 88 (6.1) 1358 (93.9)
Disabled χ2(1) = 4.17 0.041
 Yes 1844 (80.0) 95 (5.2) 1749 (94.8)
 No 460 (20.0) 35 (7.6) 425 (92.4)
Percent disabled 54.3 ± 36.8 49.5 ± 39.1 54.6 ± 36.7 U(1)= 1.81 0.179
Sleep medication χ2(1) = 0.92 0.339
 Yes 1129 (49.0) 69 (6.1) 1060 (93.9)
 No 1175 (51.0) 61 (5.2) 1114 (94.8)
Open in a new tab

Notes: Data are presented as mean ± SD and N (%N). Sleep medication prescribed prior to diagnosis of insomnia.

a

Due to small cell size, other race was combined with white race as they were not significantly different from white race but significantly different from black race.

Table 2.

Psychiatric disorders and breathing-related sleep disorder by insomnia diagnosis.

Comorbidity Total (N = 2304) Insomnia diagnosis Analysis
Yes (N = 130) No (N = 2174) Test statistic p
Psychiatric disorders
Adjustment χ2(1) = 1.03 0.309
 Yes 238 (10.3) 10 (4.2) 228 (95.8)
 No 2066 (89.7) 120 (5.8) 1946 (94.2)
Anxiety χ2(1) = 2.66 0.103
 Yes 365 (15.8) 14 (3.8) 351 (96.2)
 No 1939 (84.2) 116 (6.0) 1823 (94.0)
Bipolar χ2(1) = 2.01 0.156
 Yes 203 (8.8) 7 (3.4) 196 (96.6)
 No 2101 (91.2) 123 (5.9) 1978 (94.1)
Depression χ2(1) = 0.27 0.600
 Yes 430 (18.7) 22 (5.1) 408 (94.9)
 No 1874 (81.3) 108 (5.8) 1766 (94.2)
Alcohol/Substance χ2(1) = 0.04 0.836
 Yes 261 (11.3) 14 (5.4) 247 (94.6)
 No 2043 (88.7) 116 (5.7) 1927 (94.3)
N Axis I Co-morbidities 0.65 ± 0.85 0.52 ± 0.91 0.66 ± 0.85 U(1) = 7.02 0.008
N Axis I Co-morbidities χ2(3) = 9.73 0.021
 0 1266 (54.9) 88 (7.0) 1178 (93.0)
 1 683 (29.6) 25 (3.7) 658 (96.3)
 2 265 (11.5) 12 (4.5) 253 (95.5)
 3+ 90 (3.9) 5 (5.6) 85 (94.4)
Breathing-related sleep disorder χ2(1) = 12.7 <0.001
 Yes 385 (16.7) 7 (1.8) 378 (98.2)
 No 1919 (83.3) 123 (6.4) 1796 (93.6)
Open in a new tab

Notes: Data are presented as mean ± SD and N (%N). All comorbidities diagnosed prior to diagnosis of insomnia.

To identify predictor variables and interactions for inclusion in the logistic regression, a CART® model was conducted. The key variables identified were: race (black vs. not black [i.e. white, Asian, American Indian/Alaska Native, Native Hawaiian or other Pacific Islander, mixed, missing, declined, unknown]), age (18–64 vs. ≥65), any non-PTSD psychiatric co-morbidity (lifetime; 0 vs. ≥1), BRSD (lifetime; yes vs. no), and the interaction of race * psychiatric co-morbidity. Logistic regression analysis indicated that younger Veterans with PTSD (vs. older Veterans with PTSD) and Veterans with PTSD without a BRSD were significantly more likely to receive an insomnia diagnosis (Table 3). However, the analysis also found that compared to non-black Veterans without a psychiatric co-morbidity (46.9%), both non-black Veterans with co-morbidity (37.2%) and black Veterans without co-morbidity (5.5%) were significantly less likely to receive an insomnia diagnosis. Black Veterans with a psychiatric co-morbidity (7.5%) were not significantly different from the reference group (Table 3).

Table 3.

Logistic regression model of insomnia diagnosis.

Measure N Adjusted odds ratio 95% Confidence interval p
Age, 18–64 (reference) 1940 – – –
Age, ≥65 297 0.502 (0.271–0.929) 0.028
No breathing-related sleep disorder (reference) 1853 – – –
Breathing-related sleep disorder 384 0.250 (0.116–0.542) <0.001
Race by any psychiatric co-morbidity
 Not Black, no psychiatric co-morbidity (reference) 1081 – – –
 Not Black, any psychiatric co-morbidity 856 0.424 (0.277–0.649) <0.001
 Black, no psychiatric co-morbidity 127 0.172 (0.042–0.711) 0.015
 Black, any psychiatric co-morbidity 173 0.735 (0.381–1.418) 0.359
Open in a new tab

Insomnia Treatment Analysis

The sample size for the insomnia treatment with a sedative-hypnotic analysis, in the context of Veterans with PTSD, was 2068 Veterans. Nine hundred sixty-eight Veterans were prescribed a sedative-hypnotic (44.2%) and 73 were diagnosed with insomnia (3.5%). Demographic characteristics are listed in Table 4 and co-morbidities are listed in Table 5. Several demographic variables were significantly different for Veterans with or without treatment with a sedative-hypnotic (Table 4). Younger Veterans (vs. older Veterans), and married and separated/divorced Veterans (vs. never married Veterans), were significantly more likely to receive treatment. There was also a significant difference by service period. Vietnam War era Veterans were significantly more likely to receive treatment with a sedative-hypnotic, while World War II/Korean War era Veterans were less likely to receive treatment with a sedative-hypnotic. In contrast to the insomnia diagnosis analysis, Veterans with a service connected disability were significantly more likely to receive treatment than Veterans with no service connection. Furthermore, of those with a service connection, Veterans who received sedative-hypnotic medications had significantly greater percent service connection. There were several differences between groups on co-morbid disorders (Table 5). Veterans with an adjustment disorder were significantly less likely to receive a sedative-hypnotic. Those with an alcohol and/or substance use disorder were significantly more likely to receive a sedative-hypnotic. Last, Veterans with a BRSD were significantly less likely to receive a sedative-hypnotic.

Table 4.

Demographic characteristics by sleep medication.

Measure Total (N = 2068) Sleep medication Analysis
Yes (N = 915) No (N = 1153) Test statistic p
Age 51.6 ± 12.0 50.2 ± 9.77 52.7 ± 13.4 t(2053) = 5.05 <0.001
Sex χ2(1) = 0.19 0.666
 Male 1982 (95.8) 875 (44.1) 1107 (55.9)
 Female 86 (4.2) 40 (46.5) 46 (53.5)
Race χ2(2) = 5.86 0.053
 Not Black 1652 (82.6) 736 (44.6) 916 (55.4)
 Black 266 (13.3) 136 (51.1) 130 (48.9)
Marital status χ2(3) = 11.6 0.009
 Married 1134 (54.9) 506 (44.6) 628 (55.4)
 Separated/Divorced 555 (26.9) 269 (48.5) 286 (51.5)
 Never married 205 (9.9) 76 (37.1) 129 (62.9)
 Widowed 171 (8.3) 64 (37.4) 107 (62.6)
Service branch χ2(3) = 6.20 0.102
 Army 1283 (62.5) 554 (43.2) 729 (56.8)
 Marine Corps 443 (21.6) 219 (49.4) 224 (50.6)
 Navy 197 (9.6) 87 (44.2) 110 (55.8)
 Air Force 129 (6.3) 52 (40.3) 77 (59.7)
Service period χ2(2) = 70.3 <0.001
 WWII/Korea 340 (16.5) 89 (26.2) 251 (73.8)
 Vietnam/Post-Vietnam 1498 (72.5) 744 (49.7) 754 (50.3)
 Persian Gulf 227 (11.0) 80 (35.2) 147 (64.8)
Service duration (years) 3.38 ± 3.39 3.52 ± 3.74 3.26 ± 3.07 U(1) = 0.21 0.644
Combat χ2(1) = 1.75 0.185
 Yes 554 (30.4) 240 (43.3) 314 (56.7)
 No 1266 (69.6) 591 (46.7) 675 (53.3)
Disabled χ2(1) = 12.5 <0.001
 Yes 1674 (80.9) 772 (46.1) 902 (53.9)
 No 394 (19.1) 143 (36.3) 251 (63.7)
Percent disabled 55.4 ± 36.6 60.8 ± 35.0 51.0 ± 37.3 U(1) = 33.2 <0.001
Open in a new tab

Note: Data are presented as mean ± SD and N (%N).

Table 5.

Psychiatric disorders and breathing-related sleep disorder by sleep medication.

Co-morbidity Total (N = 2068) Sleep medication Analysis
Yes (N = 915) No (N = 1153) Test statistic p
Psychiatric disorder
Insomnia χ2(1) = 0.17 0.683
 Yes 73 (3.5) 34 (46.6) 39 (53.4)
 No 1995 (96.5) 881 (44.2) 1114 (55.8)
Adjustment χ2(1) = 4.29 0.038
 Yes 174 (8.4) 64 (36.8) 110 (63.2)
 No 1894 (91.6) 851 (44.9) 1043 (55.1)
Anxiety χ2(1) = 3.01 0.083
 Yes 267 (12.9) 105 (39.3) 162 (60.7)
 No 1801 (87.1) 810 (45.0) 991 (55.0)
Bipolar χ2(1) = 2.04 0.153
 Yes 138 (6.7) 53 (38.4) 85 (61.6)
 No 1930 (93.3) 862 (44.7) 1068 (55.3)
Depression χ2(1) = 0.74 0.389
 Yes 281 (13.6) 131 (46.6) 150 (53.4)
 No 1787 (86.4) 784 (43.9) 1003 (56.1)
Alcohol/Substance χ2(1) = 6.68 <0.010
 Yes 186 (9.0) 99 (53.2) 87 (46.8)
 No 1882 (91.0) 816 (43.4) 1066 (56.6)
N Axis I Comorbidities 0.51 ± 0.75 0.49 ± 0.75 0.52 ± 0.76 U(1) = 0.49 0.483
N Axis I Comorbidities χ2(3) = 0.59 0.900
 0 1295 (62.6) 581 (44.9) 714 (55.1)
 1 550 (26.6) 237 (43.1) 313 (56.9)
 2 178 (8.6) 78 (43.8) 100 (56.2)
 3+ 45 (2.2) 19 (42.2) 26 (57.8)
Breathing-related sleep disorder χ2(1) = 42.8 <0.001
 Yes 232 (11.2) 56 (24.1) 176 (75.9)
 No 1836 (88.8) 859 (46.8) 977 (53.2)
Open in a new tab

Notes: Data are presented as mean ± SD and N (%N). All comorbidities diagnosed prior to prescription of sleep medication.

Similar to the insomnia diagnosis analysis, a CART® model was conducted to identify the key variables and interaction terms to include in the logistic regression analysis. The key variables included: age (18–64 vs. ≥65), BRSD (lifetime; yes vs. no), percent service connected disability (0–50% vs. 51–100%), an alcohol/substance use disorder (lifetime; 0 vs.≥1), and the interaction of age*BRSD. Logistic regression analysis indicated that Veterans with PTSD with 51–100% service connection (vs. Veterans with PTSD with 0–50%) and Veteran with PTSD plus a co-morbid alcohol and/or substance use disorder (vs. no co-morbid alcohol/substance use disorder) were significantly more likely to receive prescription sedative-hypnotics. Also, compared to the reference group of younger Veterans (<65) without BRSD (75.2%), both younger Veterans with BRSD (10.5%) and older Veterans without BRSD (13.6%) were significantly less likely to be prescribed a sedative-hypnotic. Older Veterans with BRSD (0.07%) were not significantly different from the reference group (Table 6).

Table 6.

Logistic regression model of prescribed sleep medicine.

Measure N Adjusted odds ratio 95% Confidence interval p
Percent disabled, 0–50 (reference) 877 – – –
Percent disabled, 51–100 1191 1.931 (1.603–2.326) <0.001
No alcohol/substance use disorder (reference) 1882 – – –
Alcohol/Substance use disorder 186 1.370 (1.001–1.874) 0.049
Age by breathing-related sleep disorder (BRSD)
 <65, no BRSD (reference) 1555 – – –
 <65, BRSD 217 0.285 (0.204–0.397) <0.001
 ≥65, no BRSD 281 0.305 (0.227–0.410) <0.001
 ≥65, BRSD 15 0.362 (0.113–1.155) 0.086
Open in a new tab

Discussion

Despite past evidence that chronic insomnia is one of the most common co-morbid disorders in Veterans with PTSD (Bramoweth and Germain 2013; Maher, Rego, and Asnis 2006; Wright et al. 2011), this sample shows a low proportion of Veterans with PTSD have an insomnia diagnosis in their electronic medical record. The prevalence of 3.5%–5.6% in these subsamples of Veterans with PTSD is consistent with the 4.5% prevalence rate of insomnia found in the full Veteran cohort of nearly 40,000 (Bramoweth, Gregory, and Walker 2013). Veterans with PTSD and service connected >50%, which may be a proxy for overall disability, had lower rates of insomnia diagnosis but higher rates of sedative-hypnotic use compared to Veterans with PTSD and service connected 0–50%. The only sub-group of Veterans with PTSD who had elevated rates of sedative-hypnotic use was Veterans with a lifetime diagnosis of alcohol and/or substance use. As sleep disturbances are common during the recovery of alcohol and/or substance use disorders, sedative-hypnotics may be a common aspect of treatment for these Veterans.

The significant interaction terms may indicate groups to target for further clinical evaluation. Two groups had significantly decreased likelihood of receiving an insomnia diagnosis. Black Veterans with PTSD and no other co-morbid psychiatric disorder as well as non-black Veterans with PTSD plus a co-morbid psychiatric disorder were even less likely to be diagnosed with insomnia than the reference group (non-black Veterans with PTSD and no other co-morbid psychiatric disorder). The decreased likelihood of diagnosis likely results in delayed or even missed opportunities for treatment for those Veterans with an insomnia disorder. Similarly, younger Veterans with PTSD and no BRSD and older Veterans with PTSD and BRSD had an increased likelihood of receiving a sedative-hypnotic medication, which have known risks and side effects associated with their use, most notably in older Veterans (American Geriatrics Society 2015; U.S. Food and Drug Administration 2014). These two groups, due to their increased odds of sedative-hypnotic use, could be potential targets for increasing access to care for cognitive behavioral treatments for insomnia (CBTI).

A highlight of the findings are that a substantial portion (44–49%) of Veterans with PTSD are being prescribed sedative-hypnotic medications indicated to treat insomnia, yet only a fraction are being diagnosed with insomnia (3.5–5.6%). As a secondary analysis of an archival administrative data-set, it is not surprising that the findings differ from prospective epidemiological studies, which show much higher prevalence of insomnia: 22% of US adults (Roth et al. 2011) and upwards of 40% in Veterans (Hoge et al. 2008; Maher, Rego, and Asnis 2006; Mustafa et al. 2005). The data reported here do, however, show an alarming pattern. Insomnia, a disorder highly co-morbid with PTSD, appears to be drastically under-diagnosed and even in the absence of a diagnosis, high percentages of Veterans are receiving sedative-hypnotics intended to treat insomnia. Notably, CBTI is the first line, evidence-based treatment recommended by the American College of Physicians (Qaseem et al. 2016) and the NIH (2005) rather than the aforementioned drugs. Unfortunately there are numerous reasons why CBTI is not delivered as often as it should be in the treatment of insomnia. Challenges to delivering CBTI include provider-level barriers such as clinicians not having the knowledge to refer to CBTI, and a lack of clinicians who are trained to provide this therapy, and patient-level barriers such as lack of transportation, long distance between patient homes and VA Medical Centers, and work schedules. But perhaps the most apparent reason, and modifiable, is the lack of an insomnia diagnosis in the medical records. Preliminary evidence suggests that an insomnia diagnosis in the electronic medical record significantly increases the likelihood of being referred for CBTI rather than prescribed a sedative-hypnotic (OR = 2.09, CI 1.20–3.62; Bramoweth et al. 2017).

We hypothesize that without formal recognition of insomnia disorder (i.e. documentation in the medical record), it may continue to be treated as a secondary disorder to (or a symptom of) other conditions, such as PTSD and other psychiatric and medical disorders, and may prevent treatment in a timely manner with recommended interventions. A diagnosis in the medical record may improve treatment planning and potentially improve treatment outcomes. By recognizing, diagnosing, and documenting insomnia earlier (if at all), Veterans may receive treatment for one of the most common and often distressing health conditions more readily. Another potential benefit to the diagnosis and documentation of insomnia is that it may be viewed as more ‘medical’ than ‘psychiatric,’ especially if diagnosed in a non-mental health setting (e.g. primary care). While stigma is decreasing, diagnosis and treatment of psychiatric disorders is still a barrier for many Veterans. For Veterans with PTSD and co-morbid insomnia, who may be avoiding treatment for PTSD due to stigma, any pathway toward treatment, such as focusing on insomnia, will be beneficial. Research evidence suggests that when insomnia and other sleep-related symptoms of PTSD are the focus of treatment, both insomnia and PTSD outcomes improve (Schmitz, Browning, and Webb-Murphy 2009; Talbot et al. 2014; Ulmer, Edinger, and Calhoun 2011). Veterans with PTSD co-morbid with insomnia, who respond positively to CBTI as delivered by a mental health provider, may development rapport and therapeutic alliance, two important components to treatment outcomes for mental health disorders (Krupnick et al. 1996; Leach 2005), and increase their willingness to engage in treatment for PTSD. Another benefit is that patients who complete CBTI have high-treatment response, as measured by a significant symptom reduction on a validated measure (e.g. Insomnia Severity Index; Bastien, Vallieres, and Morin 2001), often at rates greater than 70% (Germain et al. 2014; Trockel et al. 2014), and they often respond rapidly (within 4–8 weeks), which can give patients a sense of treatment success. Fortunately, prospective randomized clinical trials are underway to investigate the potential benefit of treating insomnia prior to treating trauma in Service Members and Veterans in order to test the hypothesis that CBTI may be a potential ‘gateway therapy’ for the treatment of PTSD.

Policy Implications

These findings lend additional support to the under-diagnosis of insomnia (Bramoweth, Gregory, and Walker 2013; Hermes and Rosenheck 2014; Mohamed and Rosenheck 2008), especially in a population known for high rates of the disorder, Veterans with PTSD. Additionally, the high rates of sedative-hypnotic use in contrast to the low rates of insomnia indicate that symptoms of insomnia are being treated but diagnoses are not being made which, can result in inadequate treatment planning. A recent review of the sedative-hypnotic literature also indicates that, while these medications are better than treatment as usual or placebo, there is a lack of strong evidence supporting their use (Sateia et al. 2017). Meanwhile, organizations like the American College of Physicians (Qaseem et al. 2016), the American Academy of Sleep Medicine (Schutte-Rodin et al. 2008), and the NIH (2005) recommend behavioral interventions such as CBTI. Policy changes that can help with recognition of insomnia and proper diagnosis, increased access to CBTI and similar evidence-based adaptations, and efforts to deprescribe, or at least reduce the chronicity of sedative-hypnotic use are needed. As described above, access to CBTI may be even more important in populations with PTSD considering that treatments focused on insomnia can also improve symptoms of PTSD (Schmitz, Browning, and Webb-Murphy 2009; Talbot et al. 2014; Ulmer, Edinger, and Calhoun 2011). However, even with policy changes, to successfully implement treatments like CBTI and achieve sustained changes in practice, education and training efforts are needed both for the providers who often initially assess and diagnose insomnia (e.g. primary care providers) as well as to develop a new stream of providers who are accessible to treat Veterans with insomnia (e.g. psychologists, social workers, nurses). Policy changes regarding the diagnosis and treatment of insomnia are necessary to adequately provide evidence-based treatments in a sustainable and effective manner.

Limitations

Several limitations need to be considered when interpreting the results. First and foremost, the use of administrative data does not allow for causal pathways to be determined and variables are limited to those available in the VAPHS electronic medical records. There are numerous advantages to administrative data and electronic medical records, including the ability to analyze large amounts of data; however, the data entered may be prone to error. Diagnoses in particular may be used inconsistently among providers. For example, some clinicians may only enter the primary diagnosis (e.g. PTSD) of a visit into the encounter form therefore missing an opportunity to include a new or existing co-morbid disorder (e.g. insomnia) in the medical record even though it was clinically recognized and perhaps even treated. Another limitation is that medications prescribed by a non-VA provider may not be included in the electronic medical record despite efforts of VA clinicians to include all active-medications (VA, non-VA, and over-the-counter drugs). Also, the current analysis focused only on the three most common sedative-hypnotics used among the cohort (Bramoweth, Gregory, and Walker 2013) – z-drugs (i.e. eszopiclone, zaleplon, and zolpidem), trazodone, and temazepam – in the dose ranges indicated for the treatment of insomnia (Clinical Pharmacology 2015). Therefore, overall sedative-hypnotic use was under-measured. However, the specific drugs measured were also those that were most likely to be used for insomnia. Other medications indicated for treatment of insomnia are more often used for other disorders (e.g. mirtazapine, amitriptyline, lorazepam) and without specific instructions in the medical record (e.g. take at bedtime for insomnia), the results with an expanded drug list would likely result in over-estimation of sedative-hypnotic use. Lastly, data collection was limited by Inclusion of Insomnia consults only and not Including treatment, CBTI, Initiation or completion. This was primarily due to the timing of retrospective data collection, which was limited to pre-2012; the VA’s nation-wide rollout of CBTI began In 2011. Furthermore, templated notes for CBTI were not Installed Into the electronic medical records until 2015, which have significantly Improved the ability to collect data on CBTI progress and treatment outcomes.

Future Directions

As previously mentioned, prospective sequencing trials that will evaluate the benefit of treating Insomnia prior to trauma-focused treatments are underway. Also In progress are efforts to collect electronic medical record data that Include CBTI Initiation and outcomes, utilizing an expanded review of progress notes for Veterans who completed an Insomnia consult (2011–2014) as well a review of data In CBTI templated notes (2015–2017). This new data will allow for analyses to Identify potential determinants of treatment outcome as well as treatment dropout. Furthermore, future efforts to Increase clinician knowledge of Insomnia, treatments available, and the benefits and challenges of each are necessary to reduce the current gap between high rate of prescription sedative-hypnotic medications and the low rate of Insomnia diagnoses. Lastly, additional efforts are needed to help Increase access to care and the utilization of evidence-based cognitive-behavioral treatments, especially In populations with Increased prevalence of Insomnia like Service Members and Veterans with PTSD. One area of focus Is exploring treatment options that Involve multiple methods of delivery, alone or In combination, such as brief In-person treatment, telephone, video/telehealth, online and mobile-based treatments. Also to be considered Is who delivers treatment (e.g. psychologist, nurse, social worker) and where (e.g. specialty mental health, sleep medicine, or primary care clinics). The combination of Implementing and disseminating the appropriate evidence-based treatments In the appropriate settings delivered by the appropriate clinicians will result In the most efficient access to care and Improved outcomes for Veterans with PTSD and co-morbid chronic insomnia.

Funding

This work was supported by the Department of Veterans Affairs VISN 4 Mental Illness Research, Education and Clinical Center [XVA 72-904] and a HSR&D Career Development Award [CDA 13-260].

Footnotes

Disclosure Statement

No potential conflict of Interest was reported by the authors. The views expressed In this article are those of the authors and do not represent the views of the University of Pittsburgh, the Department of Veterans Affairs, or the United States Government.

References

  1. American Geriatrics Society. 2015. “American Geriatrics Society 2015 Updated Beers Criteria for Potentially Inappropriate Medication Use In Older Adults.” Journal of the American Geriatrics Society 63 (11): 2227–2246. [DOI] [PubMed] [Google Scholar]
  2. Aupperle RL, Melrose AJ, Stein MB, and Paulus MP. 2012. “Executive Function and PTSD: Disengaging from Trauma” Neuropharmacology 62 (2): 686–694. doi: 10.1016/j.neuropharm.2011.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bastien CH, Vallieres A, and Morin CM. 2001. “Validation of the Insomnia Severity Index as an Outcome Measure for Insomnia Research.” Sleep Medicine 2: 297–307. [DOI] [PubMed] [Google Scholar]
  4. Bramoweth AD, and Germain A. 2013. “Deployment-related Insomnia In Military Personnel and Veterans” Current Psychiatry Reports 15 (10): 401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bramoweth AD, Gregory MP, and Walker JD. 2013. “Clinical Characterization of Insomnia In the VA: Diagnosis and Medication Use.” Sleep 36: A200. [Google Scholar]
  6. Bramoweth AD, Renqvist JG, Hanusa BH, Walker JD, Germain A, and Atwood CW. 2017. May. “Identifying the Sociodemographic and Mental Health Factors That Influence Insomnia Treatment Recommendations within the Veteran Population.” Behavioral Sleep Medicine 2: 1–12. doi: 10.1080/15402002.2017.1318752. [DOI] [PubMed] [Google Scholar]
  7. Carson Susan, Marian McDonagh, Sujata Thakurta, and Yen Po-Yin. 2008. Drug Class Review: Newer Drugs for Insomnia. Portland, OR: Oregon Health & Sciences University. [PubMed] [Google Scholar]
  8. Clinical Pharmacology. 2015. “Drugs indicated for insomnia.” Elsevier. Accessed October 2014, www.clinicalpharmacology.com [Google Scholar]
  9. Davidson JR, Rothbaum BO, van der Kolk BA, Sikes CR, and Farfel GM. 2001. “Multicenter, Double-blind Comparison of Sertraline and Placebo in the Treatment of Posttraumatic Stress Disorder” Archives of General Psychiatry 58 (5): 485–492. doi: 10.1001/archpsyc.58.5.485. [DOI] [PubMed] [Google Scholar]
  10. Department of Veterans Affairs. 2014. “Chapter 2: Service-connected Disabilities.” In Federal Benefits for Veterans, Dependents and Survivors. 2014th ed. Washington, DC: Department of Veterans Affairs. [Google Scholar]
  11. Frayne SM, Miller DR, Sharkansky EJ, Jackson VW, Wang F, Halanych JH, Berlowitz DR, Kader B, Rosen CS, and Keane TM. 2010. “Using Administrative Data to Identify Mental Illness: What Approach is Best?” American Journal of Medical Quality 25 (1): 42–50. doi: 10.1177/1062860609346347. [DOI] [PubMed] [Google Scholar]
  12. Gehrman P 2016. “Sleep Problems in Veterans with PTSD” U.S. Department of Veterans Affairs. Accessed 26 February 2016. http://www.ptsd.va.gov/professional/co-occurring/sleep_problems_veterans_ptsd.asp [Google Scholar]
  13. Gellis LA, Gehrman PR, Mavandadi S, and Oslin DW. 2010. “Predictors of Sleep Disturbances in Operation Iraqi Freedom/Operation Enduring Freedom Veterans Reporting a Trauma” Military Medicine 175: 567–573. [DOI] [PubMed] [Google Scholar]
  14. Germain A 2013. “Sleep Disturbances as the Hallmark of PTSD: Where Are We Now?” American Journal of Psychiatry 170 (4): 372–382. doi: 10.1176/appi.ajp.2012.12040432. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Germain A, Richardson R, Stocker R, Mammen O, Hall M, Bramoweth AD, Begley A, et al. 2014. “Treatment for Insomnia in Combat-exposed OEF/OIF/OND Military Veterans: Preliminary Randomized Controlled Trial.” Behaviour Research and Therapy 61: 78–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gravely AA, Cutting A, Nugent S, Grill J, Carlson K, and Spoont M. 2011. “Validity of PTSD Diagnoses in VA Administrative Data: Comparison of VA Administrative PTSD Diagnoses to Self-reported PTSD Checklist Scores.” The Journal of Rehabilitation Research and Development 48 (1): 21–30. [DOI] [PubMed] [Google Scholar]
  17. Hermes E, and Rosenheck R. 2014. “Prevalence, Pharmacotherapy and Clinical Correlates of Diagnosed Insomnia among Veterans Health Administration Service Users Nationally.” Sleep Medicine 15 (5): 508–514. doi: 10.1016/j.sleep.2013.12.010. [DOI] [PubMed] [Google Scholar]
  18. Hoge CW, McGurk D, Thomas JL, Cox AL, Engel CC, and Castro CA. 2008. “Mild Traumatic Brain Injury in U.S. Soldiers Returning from Iraq.” New England Journal of Medicine 358: 453–463. [DOI] [PubMed] [Google Scholar]
  19. Holowka DW, Marx BP, Gates MA, Litman HJ, Ranganathan G, Rosen RC, and Keane TM. 2014. “PTSD Diagnostic Validity in Veterans Affairs Electronic Records of Iraq and Afghanistan Veterans” Journal of Consulting and Clinical Psychology 82 (4): 569–579. doi: 10.1037/a0036347. [DOI] [PubMed] [Google Scholar]
  20. Kang HK, Natelson BH, Mahan CM, Lee KY, and Murphy FM. 2003. “Post-traumatic Stress Disorder and Chronic Fatigue Syndrome-like Illness among Gulf War Veterans: A Population-based Survey of 30,000 Veterans” American Journal of Epidemiology 157 (2): 141–148. [DOI] [PubMed] [Google Scholar]
  21. Krupnick JL, Sotsky SM, Simmens S, Moyer J, Elkin I, Watkins J, and Pilkonis PA. 1996. “The Role of the Therapeutic Alliance in Psychotherapy and Pharmacotherapy Outcome: Findings in the National Institute of Mental Health Treatment of Depression Collaborative Research Program” Journal of Consulting and Clinical Psychology 64 (3): 532–539. [DOI] [PubMed] [Google Scholar]
  22. Kulka RA, Schlenger WE, Fairbank JA, Hough RL, Jordan BK, Marmar CR, and Weiss DS. 1990. Trauma and the Vietnam War Generation: Report of Findings from the National Vietnam Veterans Readjustment Study. New York: Brunner/Mazel. [Google Scholar]
  23. Leach MJ 2005. “Rapport: A Key to Treatment Success” Complementary Therapies in Clinical Practice 11 (4): 262–265. doi: 10.1016/j.ctcp.2005.05.005. [DOI] [PubMed] [Google Scholar]
  24. Maher MJ, Rego SA, and Asnis GM. 2006. “Sleep Disturbances in Patients with Post-traumatic Stress Disorder” CNS Drugs 20 (7): 567–590. doi: 10.2165/00023210-200620070-00003. [DOI] [PubMed] [Google Scholar]
  25. McLay RN, Klam WP, and Volkert SL. 2010. “Insomnia is the Most Commonly Reported Symptom and Predicts Other Symptoms of Post-traumatic Stress Disorder in U.S. Service Members Returning from Military Deployments.” Military Medicine 175: 759–762. [DOI] [PubMed] [Google Scholar]
  26. Mohamed S, and Rosenheck RA. 2008. “Pharmacotherapy of PTSD in the U.S. Department of Veterans Affairs.” The Journal of Clinical Psychiatry 69 (6): 959–965. [DOI] [PubMed] [Google Scholar]
  27. Mustafa M, Erokwu N, Ebose I, and Strohl K. 2005. “Sleep Problems and the Risk for Sleep Disorders in an Outpatient Veteran Population.” Sleep and Breathing 9: 57–63. [DOI] [PubMed] [Google Scholar]
  28. National Institutes of Health. 2005. NIH State-of-the-science Conference on Manifestations and Management of Chronic Insomnia. Bethesda: National Institutes of Health. http://consensus.nih.gov/2005/insomniastatement.pdf. [Google Scholar]
  29. Neylan T, Marmar C, Metzler T, Weiss D, Zatzick D, Delucchi K, Wu R, and Schoenfeld F. 1998. “Sleep Disturbances in the Vietnam Generation: Findings from a Nationally Representative Sample of Male Vietnam Veterans” American Journal of Psychiatry 155: 929–933. [DOI] [PubMed] [Google Scholar]
  30. Pacella ML, Hruska B, and Delahanty DL. 2013. “The Physical Health Consequences of PTSD and PTSD Symptoms: A Meta-analytic Review.” Journal of Anxiety Disorders 27 (1): 33–46. doi: 10.1016/j.janxdis.2012.08.004. [DOI] [PubMed] [Google Scholar]
  31. Qaseem A, Kansagara D, Forciea MA, Cooke M, Denberg TD, and Clinical Guidelines Committee of the American College of Physicians. 2016. “Management of Chronic Insomnia Disorder in Adults: A Clinical Practice Guideline from the American College of Physicians.” Annals of Internal Medicine 165 (2): 125–133. doi: 10.7326/M15-2175. [DOI] [PubMed] [Google Scholar]
  32. Riemann D, and Perils ML. 2009. “The Treatments of Chronic Insomnia: A Review of Benzodiazepine Receptor Agonists and Psychological and Behavioral Therapies.” Sleep Medicine Reviews 13: 205–214. [DOI] [PubMed] [Google Scholar]
  33. Rodriguez P, Holowka DW, and P Marx B. 2012. “Assessment of Posttraumatic Stress Disorder-related Functional Impairment: A Review.” The Journal of Rehabilitation Research and Development 49 (5): 649–665. [DOI] [PubMed] [Google Scholar]
  34. Roth T, Coulouvrat C, Hajak G, Lakoma MD, Sampson NA, Shahly V, Shillington AC, Stephenson JJ, Walsh JK, and Kessler RC. 2011. “Prevalence and Perceived Health Associated with Insomnia Based on DSM-IV-TR; International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; and Research Diagnostic Criteria/International Classification of Sleep Disorders, Second Edition Criteria: Results from the America Insomnia Survey.” Biological Psychiatry 69: 592–600. [DOI] [PubMed] [Google Scholar]
  35. Sareen J, Cox BJ, Clara I, and Asmundson GJ. 2005. “The Relationship between Anxiety Disorders and Physical Disorders in the U.S. National Comorbidity Survey.” Depression and Anxiety 21 (4): 193–202. doi: 10.1002/da.20072. [DOI] [PubMed] [Google Scholar]
  36. Sareen J, Cox BJ, Stein MB, Afifi TO, Fleet C, and Asmundson GJ. 2007. “Physical and Mental Comorbidity, Disability, and Suicidal Behavior Associated with Posttraumatic Stress Disorder in a Large Community Sample.” Psychosomatic Medicine 69 (3): 242–248. doi: 10.1097/PSY.0b013e31803146d8. [DOI] [PubMed] [Google Scholar]
  37. Sateia MJ, Buysse DJ, Krystal AD, Neubauer DN, and Heald JL. 2017. “Clinical Practice Guideline for the Pharmacologic Treatment of Chronic Insomnia in Adults: An American Academy of Sleep Medicine Clinical Practice Guideline.” Journal of Clinical Sleep Medicine 13 (2): 307–349. doi: 10.5664/jcsm.6470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Schmitz K, Browning K, and Webb-Murphy J. 2009. “Key Facts on Sleep and Operational Stress for Navy & USMC Leaders.” Research Facilitation 1–9. [Google Scholar]
  39. Schutte-Rodin S, Broch L, Buysse D, Dorsey C, and Sateia M. 2008. “Clinical Guideline for the Evaluation and Management of Chronic Insomnia in Adults.” Journal of Clinical Sleep Medicine 4 (5): 487–504. [PMC free article] [PubMed] [Google Scholar]
  40. Talbot LS, Maguen S, Metzler TJ, Schmitz M, McCaslin SE, Richards A, Perlis ML, et al. 2014. “Cognitive Behavioral Therapy for Insomnia in Posttraumatic Stress Disorder: A Randomized Controlled Trial.” Sleep 37 (2): 327–341. doi: 10.5665/sleep.3408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Tanielian T, and Jaycox LH, eds. 2008. Invisible Wounds of War: Psychological and Cognitive Injuries, Their Consequences, and Services to Assist Recovery. Santa Monica, CA: RAND Corporation. http://www.rand.org/pubs/monographs/MG720. [Google Scholar]
  42. Trockel M, Karlin BE, Taylor CB, and Manber R. 2014. “Cognitive Behavioral Therapy for Insomnia with Veterans: Evaluation of Effectiveness and Correlates of Treatment Outcomes.” Behaviour Research and Therapy 53: 41–46. doi: 10.1016/j.brat.2013.11.006. [DOI] [PubMed] [Google Scholar]
  43. Ulmer Christi, Edinger Jack, and Calhoun Patrick. 2011. “A Multi-component Cognitive-behavioral Intervention for Sleep Disturbance in Veterans with PTSD: A Pilot Study.” Journal of Clinical Sleep Medicine 7 (1): 57–68. [PMC free article] [PubMed] [Google Scholar]
  44. Ursano RJ, Bell C, Eth S, Friedman M, Norwood A, Pfefferbaum B, Pynoos JD, et al. 2004. “Practice Guideline for the Treatment of Patients with Acute Stress Disorder and Posttraumatic Stress Disorder.” The American Journal of Psychiatry 161 (11 Suppl): 3–31. [PubMed] [Google Scholar]
  45. U.S. Food and Drug Administration. 2014. “Questions and Answers: Risk of Next-morning Impairment After use of Insomnia Drugs; FDA Requires Lower Recommended Doses for Certain Drugs Containing Zolpidem (Ambien, Ambien CR, Edluar, and Zolpimist).” Accessed October 2015. http://www.fda.gov/drugs/drugsafety/ucm334041.htm
  46. Wright KM, Britt TW, Bliese PD, Adler AB, Picchioni D, and Moore D. 2011. “Insomnia as Predictor versus Outcome of PTSD and Depression among Iraq Combat Veterans.” Journal of Clinical Psychology 67: 1240–1258. [DOI] [PubMed] [Google Scholar]
  47. Zayfert C, and DeViva JC. 2004. “Residual Insomnia following Cognitive Behavioral Therapy for PTSD.”Journal of Traumatic Stress 17 (1): 69–73. doi: 10.1023/B:JOTS.0000014679.31799.e7. [DOI] [PubMed] [Google Scholar]

RESOURCES