for people ages 18-75 (full criteria)
at San Francisco, California and other locations
study started
estimated completion



Many Veterans with posttraumatic stress disorder (PTSD) have trouble sleeping or have frequent nightmares. So far, no medication has been approved for treatment of insomnia in PTSD. The purpose of this research study is to find out if taking medications called trazodone, eszopiclone, or gabapentin can help decrease symptoms of insomnia in patients with PTSD. PTSD is a form of intense anxiety which sometimes results from severe trauma. Symptoms may include nightmares, flashbacks, troublesome memories, difficulty sleeping, poor concentration, irritability, anger, and emotional withdrawal. Insomnia is a disorder that can make it hard to fall sleep, stay asleep or cause a person to wake up too early and not be able to fall back to sleep.

Official Title

CSP #2016 - National Adaptive Trial for PTSD Related Insomnia


VA Cooperative Studies Program #2016 is a double-blind four-arm adaptive clinical trial to compare the efficacy of trazodone hydrochloride, eszopiclone, and gabapentin to placebo, as adjunctive therapies in the treatment of insomnia symptoms among Veterans with military related PTSD, as measured by statistically significant difference in change from baseline in Insomnia Severity Index (ISI) total score at Week 12.

Participants will be approximately 1224 male and female Veterans with PTSD and moderate levels of insomnia as measured on the ISI. Veterans who meet inclusion and exclusion criteria will be randomized within each site to receive trazodone hydrochloride, eszopiclone, gabapentin or placebo. Permuted blocks randomization will be used within each participating site. A mid-point interim analysis will be conducted wherein active treatment arms meeting futility early stopping criteria will be dropped. If all active treatment arms are dropped at the interim analysis, the study is stopped at that time. Otherwise, the study will continue and the remaining sample size will be allocated to the remaining study arms with equal randomization probabilities. Study drug dose will be increased using a flexible dose titration schedule over a period of up to 3 weeks and continued for up to 12 weeks.

The Insomnia Severity Index (ISI) is the primary outcome for this study. The Clinician Administered PTSD Scale for DSM-V (CAPS-5) will be the primary secondary outcome measuring change in PTSD symptoms. Other secondary outcomes that measure PTSD and sleep include the PTSD Checklist (PCL-5) and Pittsburgh Sleep Quality Index Scale-Addendum for PTSD (PSQI-A). Other secondary outcomes include brief questionnaire secondary measures of comorbid depression (PHQ-9), anxiety (GAD-7), quality of life (WHOQOL-BREF), treatment satisfaction questionnaire for medication (TSQM-9), anger and aggression (DAR-5), smoking and alcohol consumption (Timeline Follow-Back, or TLFB), clinical global change (CGI-S), resource utilization (an abbreviated subset of the Service Utilization and Resources Form, or SURF), Columbia Suicide Severity Rating Scale (C-SSRS), optional wearable device (Actiwatch Spectrum Plus by Philips) to measure actigraphy.

This study is designed to serve as a well-powered "screen" for efficacious medications for the treatment of PTSD-related insomnia from among the medications already widely prescribed for this purpose within VA. Thus, this study is powered to detect significant differences between trazodone, eszopiclone, and gabapentin versus placebo. The investigators have powered the study to detect a small effect size, recognizing that the effect size of s-zopiclone is larger ( 0.5) in a small short-term pilot study in PTSD but that the effect size of s-zopiclone declines over time in a well-powered study of primary insomnia, stabilizing at 12 weeks of treatment at a level that is sustained over subsequent months of treatment. Presuming that the widespread prescription rates of these three medications for PTSD patients suggests that they have some efficacy for PTSD-related insomnia, the investigators would expect that the effect size for the comparisons among the active medications would be very small. Therefore, the study is not powered to detect differences among the active medications. The investigators have chosen the ISI as the primary outcome for several reasons: (a) it has excellent psychometric properties, (b) it is feasible (in terms of subject burden and cost) to administer the ISI at multiple timepoints during the 12-week trial, (c) according to the insomnia experts who provided input into the study design it has displaced sleep diary-related measures as the primary outcome in clinical trials for sleep, and d) it has been accepted by the FDA as the primary outcome measure in registration trials.

Based on the literature, an effect size of 0.35 in the primary endpoint (ISI) is plausible and clinically meaningful. The investigators estimated the overall participant dropout rate is between 10% to 15%. With a conservative estimate drop-out rate of 15%, the total sample size will be 1224. From the previous CSP PTSD trials, the investigators anticipate each participating site can randomize on average 1.25 participants per month, or 15 patients per site per year. With 3 years of recruitment, the investigators would need 28 sites to reach sample 1224. The investigators plan to start the study with 32 sites to allow dropping of non-performing sites.

VA bears a unique responsibility for addressing the limited efficacy of current evidence-based pharmacotherapy practices for PTSD. Since 2001, there have been only two FDA-approved medications for PTSD, both serotonin reuptake inhibiting antidepressants (SRIs), and SRIs have limited efficacy for military-related PTSD. This "efficacy gap" results in widespread polypharmacy for PTSD in VA, such that Veterans with antidepressant-resistant symptoms are treated, on average, with more than three psychotropic medications that present risks without clear benefit. In particular, SRI-resistant insomnia in military-related PTSD is a significant problem for VA, with 88% of these patients reporting clinically significant sleep impairment. In PTSD, sleep disturbances contribute importantly to impairments in quality of life, reduced social and vocational function, suicide risk, and poorer health. Effective treatment of persisting insomnia in PTSD is a sufficiently serious unmet need that the 2017 VA/DoD Clinical Practice Guideline for the Management of Posttraumatic Stress Disorder, called for "studies of non-benzodiazepine sedative/hypnotics." The purpose of the study is address this gap through testing the efficacy of three non-benzodiazepine hypnotics in comparison to placebo, representing the three medications or medication classes that are most commonly prescribed to Veterans with PTSD on an off-label basis and have yet to be tested in a definitive clinical trial. A novel aspect of this study is its implementation of an adaptive design in which arms would be dropped for evidence of futility based on pre-specified criteria at a designated interim analysis, intended to increase the efficiency of the trial and thereby improve the feasibility of its ambitious aim. The VA Cooperative Studies Program is uniquely suited to conduct this study.


Insomnia, Sleep Initiation and Maintenance Disorders, Gabapentin, Trazodone, Eszopiclone


You can join if…

Open to people ages 18-75

  1. Ability to comprehend and provision of signed and dated informed consent form
  2. Stated willingness to comply with all study procedures and availability for the duration of the study (approximately 17 weeks from the date of being randomized)
  3. Individuals, between the ages of 18 and 75 years
  4. Allow digital recording of phone interviews
  5. PTSD related to military service
  6. Primary DSM-5 diagnosis of PTSD, assessed by structured interview using the CAPS-5
  7. Total CAPS-5 score 26
  8. ISI >15
  9. Screening clinical laboratory tests without clinically significant abnormalities that would make study participation inappropriate, as determined by the site investigator with input, if needed, from the study chair

    10. Electrocardiogram (ECG) at baseline without clinically significant abnormalities that

    would make study participation inappropriate, as determined by the site investigator with input, if needed, from the study chair and/or contingent upon approval by consulting medical physician.

    11. Females of childbearing potential:

    1. Must have a negative pregnancy test during screening
    2. Must agree not to become pregnant or breastfeed during the course of the study
    3. Must be willing to use a reliable form of contraception for 16 weeks (during study treatment and for 2 weeks after taking the last dose) which includes: barrier contraceptives (male or female condoms with or without a spermicide, diaphragm or cervical cap with spermicide, or intrauterine device) and hormone-based therapy (contraceptive pills, intrauterine devices, or Depo-Provera�)
    4. Birth control for female participants is not necessary if surgically sterile or if with a partner with whom they are not capable of conceiving children (defined as a surgically sterile female by hysterectomy, bilateral oophorectomy, or bilateral tubal ligation; surgically sterile male who has undergone a complete orchiectomy or successful vasectomy; or a same sex partner)
      1. Agree to secure firearms while receiving study treatment
      2. If individuals are undergoing evidence-based psychotherapy (EBT), which includes cognitive behavioral therapy (CBT), cognitive processing therapy (CPT), prolonged exposure therapy (PE), and/or stress inoculation therapy (SIT), they must have started these therapies at least 60 days prior to starting screening. If screening is started, and it is then discovered that EBT was started within 60 days prior to screening, participants must wait at least 60 days since staring the new EBT before they can complete the screening ISI, the screening PCL, and the Phone Assessment. (Supportive individual and group therapy is allowed)
        1. Agreement to adhere to Lifestyle Considerations (see Section 5.3) throughout study participation
        2. Clinical evidence of adequately treated sleep apnea or absence of sleep apnea having a severity that would make study participation problematic, established by meeting one of the criteria below:
      3. Clinical evidence of adequately treated sleep apnea with a continuous positive airway pressure (CPAP) or alternative treatment device (as defined in Section 8.1) (Participants can be reevaluated at least 30 days after screen failure.)
      4. If clinically tested, sleep study negative for sleep apnea or results indicating an Apnea-Hypopnea Index (AHI) < 23 within the past 6 months
      5. If tested for study eligibility, ApneaLink (or equivalent alternative device) result or other sleep study shows AHI < 23.

You CAN'T join if...

Currently enrolled in any other interventional study unless prior approval is provided by the study team (It is a CSP policy that exemptions will be assessed for individual patients on a case-by-case basis. Exemptions require the agreement in writing of the following individuals or groups: (1) the SIs of both studies; (2) the Study Chairs of the involved studies; (3) the appropriate CSP Center Director(s); and (4) the VA Central IRB.) 2. Allergy and/or history of intolerance to trazodone hydrochloride, gabapentin, and/or eszopiclone, or history of experiencing complex sleep behaviors (i.e., sleep walking, sleep driving, and/or engaging in other activities while not fully awake) while taking any sleep medication 3. A comorbid current or lifetime diagnosis of bipolar I disorder, bipolar II disorder, schizoaffective disorder, schizophrenia or delusional disorder, or current comorbid diagnosis of schizophreniform disorder, brief psychotic disorder, psychotic disorder due to a general medical condition, substance-induced psychotic disorder or psychotic disorder not otherwise specified (NOS) according to Structured Clinical Interview for DSM-5 (SCID)-I-RV/P 4. History of moderate or severe traumatic brain injury (TBI) or history of gross structural damage as shown on MRI.

  1. Positive urine test for an illicit substance, excluding cannabis, within the past 90 days prior to screening 6. Substance use meeting DSM-5 criteria for moderate or severe dependence (excluding nicotine) within the past 12 months prior to screening. (Note: Current mild dependence for alcohol and cannabis use is acceptable, but current mild dependence of any other substances is exclusionary.) 7. Inpatient psychiatric hospitalization within 30 days prior to screening 8. Suicidal or homicidal ideation with intent or plan to harm themselves or others within 90 days prior to screening 9. Creatinine clearance (CrCl) less than 60 mL/min (estimated using the Cockcroft-Gault equation, using ideal or adjusted body weight for overweight or obesity), or chronic liver disease with two or more of the following occurring within the past six months: international normalized ratio (INR) greater than or equal to 1.7 (not on warfarin therapy), bilirubin greater than or equal to 2 mg/dL, serum albumin less than or equal to 3.5 g/dL, ascites, or encephalopathy (Participants can be reevaluated in 30 days) 10. Clinical and laboratory evidence of untreated hypothyroidism or hyperthyroidism 11. A corrected QT (QTc) interval greater than 470 ms 12. Unstable, serious medical condition or one requiring acute medical treatment, or planned hospitalization for extended care 13. Dementia, epilepsy, stroke, or current treatment with warfarin for anticoagulation 14. Taking any of the exclusionary medications listed in Appendix A. Note- an individual taking one of these medications for the sole purpose of improving sleep that elects to undergo an adequate wash-out period of at least 5 half-lives of the parent compound or active metabolite (e.g., for medications like diazepam), under the care of the individual's clinical provider, would not be excluded by this criterion.
  2. Under criminal investigation or pending legal charges with potential incarceration 16. Individuals who lack stable contact information (including lack of a telephone number) 17. Participants who anticipate working during the hours of midnight to 6am during the course of study trial 18. Participants with narcolepsy


  • San Francisco VA Medical Center, San Francisco, CA accepting new patients
    San Francisco California 94121 United States
  • VA Palo Alto Health Care System, Palo Alto, CA accepting new patients
    Palo Alto California 94304-1290 United States


accepting new patients
Start Date
Completion Date
VA Office of Research and Development
Phase 3 research study
Study Type
Expecting 1224 study participants
Last Updated