Studies estimate that Veterans diagnosed with post-traumatic stress disorder (PTSD) have an insomnia prevalence rate of 70-91%. Conventional treatments for PTSD-related insomnia include medications, psychotherapy, and cognitive behavioral therapy. While some of these treatments do improve PTSD-related insomnia, many have limitations (e.g., medication effects, lengthy time commitments, psycho-social stigma). Because of this, many Veterans are increasingly turning to complementary and alternative medicine (CAM) to relieve their symptoms. There is a growing body of research that shows that acupuncture may improve many health symptoms including depression, PTSD, addiction, headaches, pain, and insomnia. However, to date, no study has explored how acupuncture may affect PTSD-related insomnia. Because so many Veterans with PTSD experience PTSD-related insomnia, and because the current conflicts in Southwest Asia are producing a new generation of combat Veterans, it is critical that the VA explore innovative treatments for PTSD-related insomnia.
A major goal of this study was to evaluate the feasibility of the intervention and get preliminary estimates about efficacy. The first objective was to examine how group auricular acupuncture (GAA) may influence some of the maladaptive perpetuating factors associated with PTSD-related insomnia in Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF) Veterans. A second objective was to evaluate the degree of acceptance among OEF/OIF veterans for GAA for PTSD-related insomnia. Long-term objectives were: 1) determine barriers to the successful implementation of GAA for PTSD-related insomnia; and 2) describe characteristics of Veterans who are most likely to respond to GAA.
The study aims were to address two research questions:
Primary: How does GAA influence the perpetuating factors associated with chronic PTSD-related insomnia compared to sham acupuncture and true control? Perpetuating factors were defined as: a) perceived sleep quality; b) fragmented sleep patterns; and c) hypnotic medication use.
Secondary: What is the degree of acceptance for GAA for PTSD-related insomnia? This was measured by patient satisfaction scores and attrition rates.
Spielman's 3-P Model of Insomnia provided the conceptual framework for this single-site, randomized, controlled trial. OEF/OIF Veterans with PTSD-related insomnia were assigned to one of three treatment groups: GAA, sham control, or true control. The GAA group received twice weekly, 5-point GAA for two months, as well as conventional care for PTSD-related insomnia. The sham group received services identical to the GAA group except that five non-acupuncture points were used. The true control group received conventional care for PTSD-related insomnia. Both the sham and the true control groups were eligible for true GAA once the study period was complete.
Inclusion criteria were: OEF/OIF Veteran; diagnosed with PTSD per DSM-IV criteria; have insomnia as indicated by score equal to or greater than 8 on Insomnia Severity Index (ISI); diagnosis of insomnia made after PTSD diagnosis; and stable on psychotropic medications for one month prior to study enrollment. Exclusion criteria were: history of moderate or severe traumatic brain injury; start use of continuous or bilevel positive airway pressure therapy during the study; history of substance abuse dependence during the year preceding enrollment or history of illicit substance use for 3 months prior to study enrollment or positive Audit C score at study enrollment or during course of study enrollment; received acupuncture during past 3 months; on Coumadin/Heparin/Lovenox; or pregnancy.
Dependent variables were: 1) perceived sleep quality ratings (primary outcome variable); 2) fragmented sleep patterns - defined as disruptive sleep patterns in which four sleep parameters were analyzed (total sleep time, sleep latency, sleep efficiency, and naps); 3) hypnotic medication use; 4) patient satisfaction; and 5) attrition rates. Study covariates included: 1) baseline values for above outcome variables; and 2) predisposing and precipitating factors, including a) demographic characteristics (age, gender, marital status, employment status, shift work, race, education level, military rank), and b) clinical characteristics (PTSD severity and duration, duration of PTSD-related insomnia, chronic physical/mental illnesses, medication use, use of non-pharmacological therapies, and substance use).
Data collection instruments were Morin sleep diaries, ISI, wrist actigraphs, PCL-17 (PTSD Check List), demographic questionnaires, patient satisfaction surveys, and chart reviews.
Means were compared for the three groups at the 2-month time point using one-way Analysis of Variance. Effect size for this association was indicated by R2. Differences in change over time between treatment groups were assessed by examining the group x time interaction term in repeated measures analyses of variance. An independent variable accounting for 10% of the variance in the dependent variable (R2 = .10), or an interaction accounting for 10% of total variance, can be considered to be clinically meaningful. Because this was a pilot study, probability values of p<.20 were considered indicative of trends that should be followed up with further study, while those with p<.05 were considered as likely to be true positives. A total sample of 25 Veterans completed the study (n=8 true GAA, n=8 sham control, n=9 true control).
Using self-report data, the difference between groups in nap time at post-treatment was significant (p<.05), with a large effect size (R2 = .38). The GAA group had higher nap time than the other two groups. In addition, the group x time interaction was significant for self-reported nap time (p<.05), with a large effect (interaction explained 37.4% of total variance). This indicates that the amount of change over time differed between groups, with the GAA group increasing from pre- to post-treatment, while the other two groups declined. Nap time using wrist actigraph data showed a trend at the post-treatment assessment with moderate effect size (R2 = .19), with the GAA and true Control groups having higher mean nap time than the Sham group (p=.12).
At post-treatment, there was a trend toward reduced self-reported total sleep time in the GAA group compared to the other two groups (p=.14), with a moderately strong effect (R2 = .22). The interaction was significant for self-reported total sleep time (p<.05), indicating that the GAA group dropped while the other two groups increased from pre- to post-treatment. This was a strong effect (interaction explained 33.6% of total variance). Other findings had effect sizes > .10, but are not discussed here because they had p > .10.
For those who completed the study, acupuncture quality satisfaction scores were higher for Veterans in the GAA group compared to Veterans in the sham control (respective scores on 5 point scale: mean 4.88, SD 0.23 vs. mean 4.38, SD 0.64). Attrition rates were lower for the GAA group compared to sham control, as measured by higher 2 month average attendance rates for acupuncture sessions (mean GAA attendance rate 70.7%, SD 18.2 vs. mean sham control attendance rate 59.0%, SD 21.2). Analyses are underway to look at further subcomponents of patient satisfaction surveys and to further examine reasons for non-completion for those who did not complete study.
To our knowledge, this is the first RCT that has examined the impact of a GAA intervention on PTSD-related insomnia. Findings from this study have provided important information on attrition rates, estimates of effect sizes, patient satisfaction scores, and recruitment strategies that may be most effective for CAM studies that are targeting OEF/OIF Veterans with PTSD-related health symptoms.
External Links for this Project
- Prisco MK, Jecmen MC, Bloeser KJ, McCarron KK, Akhter JE, Duncan AD, Balish MS, Amdur RL, Reinhard MJ. Group Auricular Acupuncture for PTSD-Related Insomnia in Veterans: A Randomized Trial. Medical acupuncture. 2013 Dec 17; 25(6):407-422. [view]