Relationship Between Post-Traumatic Stress Disorder Symptoms and Chronic Pain-Related Symptom Domains Among Military Active Duty Service Members

Abstract

Objective

This study examined the relationships between symptom domains relevant to post-traumatic stress disorder (PTSD) diagnosis, PTSD screening, and chronic pain-related symptoms (pain intensity, pain interference, physical function, fatigue, depression, anxiety, anger, satisfaction with social roles) experienced by active duty military service members with chronic pain.

Design

ross‐sectional study.

Setting

This study was conducted at the Interdisciplinary Pain Management Center (IPMC) at Madigan Army Medical Center between 2014 and 2018.

Subjects

Active duty service members receiving care at IPMC (n = 2745) were included in this study.

Methods

Independent sample t test was conducted to compare pain intensity and pain-related measures of physical, emotional, and social functioning among patients with and without a PTSD diagnosis or PTSD positive screen (≥3 symptoms). Relative weight analysis was used to identify the relative importance of each PTSD symptom cluster (e.g., intrusion, avoidance, hyperarousal, emotional numbness) to pain and related domains.

Results

Approximately 27.9% of the patients had a positive screen for PTSD, and 30.5% of the patients had a PTSD diagnosis. Patients with PTSD diagnosis and positive screening had higher pain interference and lower physical function and social satisfaction scores (P < 0.001) and had increased anger, anxiety, fatigue, and depression scores (P < 0.001). Emotional numbness accounted for the largest proportion of variance in average pain intensity, pain interference, and psychological functioning, and avoidance accounted for the largest proportion of variance in physical function.

Conclusion

To improve treatment effectiveness and overall functioning for active duty military patients, integrated treatment and therapies targeted to reducing chronic pain and PTSD symptoms (focus on emotional numbness and avoidance) are recommended.

Introduction

Post-traumatic stress disorder (PTSD) is a mental health disorder that may develop if an individual experiences a traumatic event that causes significant distress or interference with functioning more than 1 month later [1]. PTSD occurs with regular frequency among active duty military service members [2], with prevalence rates higher in military and veteran populations relative to civilian populations. In the late 1980s, 30.9% of veterans of the Vietnam War were diagnosed with PTSD [3]. According to the National Academy of Medicine (formerly the Institute of Medicine), the prevalence of PTSD in military members who served in Iraq or Afghanistan ranges from 13% to 20% [4].

PTSD symptoms can negatively impact functioning across several domains (e.g., physical, emotional, and social health domains) [5], regardless of whether an individual meets criteria for a PTSD diagnosis or not [2, 6, 7]. Some evidence indicates that severity of physical symptoms (including pain impact) following a combat-related traumatic injury is associated with severity of subsequent PTSD symptoms in the immediate post-deployment period [8, 9]. Physical symptoms, such as pain interference, also predicted worse PTSD symptoms among veterans injured during deployment [10]. These relationships may explain the high comorbidity between PTSD and chronic pain conditions [11], as well as the compounding impact of pain and PTSD on opioid use disorder [12]. Research shows that higher pain intensity, PTSD, and emotional concerns can compromise functional outcomes among active duty service members serving in Iraq and Afghanistan [13]. Active duty members with comorbid PTSD and pain conditions are less likely to continue military service and are more likely to have medical discharge [14]. It is imperative to understand the relationship between PTSD symptoms and impairment in pain-related physical, emotional, and social functioning among active duty service members.

PTSD symptoms are grouped into four domains: 1) intrusive re-experiencing, 2) avoidance of reminders, 3) hyperarousal, and (4) negative changes in cognition and mood such as self-blame or anger [15]. Different PTSD symptoms play different roles in pain-related physical, emotional, and social function. For example, in one study of almost 1,500 US military veterans, emotional numbness, sleep problems, and psychogenic amnesia explained a large proportion of the variance in physical function, whereas concentration problems, negative thoughts, and positive affect difficulties explained a large proportion of the mental health variance [16]. A different study of 4,150 US military veterans indicated that the arousal symptom (e.g., sleep problems, anger, irritability, concentration problems) accounted for a large proportion of variance in physical health conditions, regardless of whether participants screened positive for subthreshold versus full PTSD [7]. Taken together, it is important to understand which PTSD symptoms account for effects on pain-related and additional functioning domains so that targeted interventions result in more efficient treatment and improved outcomes overall for each patient.

To date, most studies of PTSD and pain-related health outcomes have been conducted in civilian or veteran populations; few studies evaluated the comorbidity of chronic pain and PTSD and related symptoms among active duty service members. The objectives of this study were to 1) examine the relationship between PTSD diagnosis, PTSD symptoms, and symptom domains related to chronic pain (pain intensity, pain interference, physical function, fatigue, depression, anxiety, anger, satisfaction with social roles) in a sample of active duty service members with chronic pain referred for interdisciplinary pain care in a military treatment facility, and (2) determine the relative importance of specific PTSD symptom domain for each pain-related symptom domain.

Methods

Study Design, Setting, and Sample

We describe a cross-sectional study of patients receiving care at the Interdisciplinary Pain Management Center (IPMC) at Madigan Army Medical Center (MAMC) between 2014 and 2018. This study has been approved by the institutional review board of MAMC. MAMC is located on Joint Base Lewis-McChord, which has an active duty population of more than 25,000 service members and was one of eight Army sites selected for an IPMC. The MAMC IPMC was established in 2011. It has a staff of approximately 40 health care providers including physicians board certified in anesthesiology, pain medicine, physical medicine, and family medicine; a nurse practitioner; a physician assistant; and providers of psychology, nurse case management, physical therapy, occupational therapy, clinical pharmacology, chiropractic, acupuncture, medical massage therapy, and yoga therapy.

As part of standard of care, IPMC patients are instructed to complete the Pain Assessment Screening Tool and Outcomes Registry (PASTOR) on a computer or smart phone, at home or at the clinic, prior to appointments and on a regular basis (e.g., approximately monthly while engaged in treatment). The MAMC IPMC has used PASTOR continuously since 2014 to assess treatment response and guide pain management treatment decisions [17]. PASTOR leverages several National Institutes of Health Patient-Reported Outcomes Measurement Information System (PROMIS) computer-adaptive testing scales that examine a range of pain-related outcomes and include additional scales and questionnaires relevant for interdisciplinary pain care [17]. PASTOR has been used in several military IPMCs, including the IPMC described here, to track treatment outcomes as standard of care [18]. The IPMC’s patient population includes primarily active duty service members, with a small percentage of retired veterans and family members of military beneficiaries.

Measures

We collected patient-level factors including military status (duty status, branch of service), injury related to deployment, and demographic characteristics (age, household income, gender). We used the four-item Primary Care PTSD (PC-PTSD) Screen [19], which is a screening tool designed for use in both primary care and other medical settings and was embedded in the original version of PASTOR. For the subset of patients who completed the five-item PC-PTSD Screen [19], we only used four of five items for this analysis. The additional item is the only difference between the Diagnostic and Statistical Manual of Mental Disorders (DSM)-4 and DSM-5 PTSD screening criteria. The PC-PTSD Screen is currently used to assess exposure to traumatic events and screen for PTSD in service members and veterans. It assesses whether the respondent has experienced (yes or no) nightmares or unwanted thoughts (intrusion symptoms), avoidance of trauma-related thoughts and situations (avoidance symptoms), hypervigilance (hyperarousal symptoms), or feeling emotionally numb or detached (emotional numbness symptoms) in the past month. A positive screen indicates probable PTSD, warranting further assessment if at least three symptom domains are endorsed. The PC-PTSD Screen has demonstrated good test-retest reliability (r = 0.83) and predictive validity [20]. PTSD diagnosis was identified using International Classification of Diseases (ICD)-9 and ICD-10 diagnostic codes from the Military Health System Mart (M2) database, which included all medical encounters in the Military Health System that occurred between 1 year before and 1 year after PASTOR registration.

Pain-related functioning domains included two pain-intensity items (average pain and worst pain over the past 7 days) assessed using the Defense and Veterans Pain Rating Scale (DVPRS). The DVPRS is a 0–10 scale that anchors each response with language related to functioning (e.g., 0 = no pain at all, 5 = interrupts some activities, 10 = as bad as it could be, nothing else matters). The scale has demonstrated good test-retest reliability (r = 0.77) among military populations [21] and had good internal consistency reliability (Cronbach alpha = 0.91) in the study sample. Additionally, the PROMIS scales included domains related to pain (pain interference), physical function (physical function, fatigue, sleep-related impairment), emotions (depression, anxiety, anger), and social life (satisfaction with social roles and activities, or social satisfaction). All PROMIS scales have been validated in both clinical and nonclinical samples [22–26]. PROMIS scale scores are interpreted per their names. For example, higher scores in physical function and social satisfaction scales indicate better functioning, whereas higher scores across the remaining PROMIS scales (e.g., depression, fatigue) indicate worse functioning.

Data Analysis

Descriptive statistics were calculated for demographic and military information. Continuous variables were described with mean and standard deviation. Categorical variables were described using frequency and percentage. Independent sample t test was conducted to compare pain intensity and related symptoms between patients with and without a PTSD diagnosis or PTSD positive screening (≥3 symptoms). Separate independent sample t tests were conducted among male and female patients. Pairwise deletion was used to handle the missing data. The alpha level was set at 0.05 for two-sided hypothesis testing. We used SPSS version 25 (IBM, Armonk, NY) for this analysis.

Relative weight analysis (RWA) was used to identify the relative importance of each PTSD symptom domain (e.g., intrusion, avoidance, hyperarousal, emotional numbness) to pain-related symptom domains. The weights in the RWA model reflect the decomposition of the total model R² and the proportion of variance in dependent variables (DVPRS and PROMIS scores) that is attributed to each predictor variable (the four PTSD symptoms) [27]. We used RWA-Web to perform RWA (http://relativeimportance.davidson.edu; an example of RWA-Web for RWA can be found here [27]). The dataset was saved in csv format and sent to R server. Pairwise deletion was used for missing data. We applied 10,000 bootstrapping replications with an alpha level of 0.05 for the test.

Results

Sample Characteristics

Our study sample comprised a total of 2,745 active duty service patients. Demographic characteristics are shown in Table 1. Patients were predominantly in the army (86.9%), younger than 30 years (41.0%), male (81.2%), married or partnered (75.9%), with less than a 4-year college education (73.7%), and a household income less than $49,000 (46.5%). Nearly 32% reported that the injury related to their pain occurred during deployment. The average number of PTSD symptom domains reported was 1.3 ± 1.6 (range = 0–4); 27.9% of the patients reported three or more PTSD symptom domains, indicating a positive screening for PTSD; and 30.5% of these patients had a PTSD diagnosis within 1 year. Among male active duty patients (n = 2230), the prevalence of PTSD diagnosis was 29.2% and the prevalence of PTSD positive screen was 27.9%. Among female active duty patients (n = 515), the prevalence of PTSD diagnosis was 36.3% and the prevalence of PTSD positive screening was 32.2%. Approximately 17% of the patients had both PTSD diagnosis and PTSD positive screening, 13% of the patients had PTSD diagnosis but not PTSD positive screening, and 11% of the patients had PTSD positive screening but not PTSD diagnosis (see Table 2).

Table 1.

Demographic and clinical characteristics (n = 2,745)

Characteristic	No. (%) or Mean (SD)
Gender
Female	515 (18.8%)
Male	2,230 (81.2%)
Age, y
<30	1,127 (41.0%)
30–40	1,027 (37.5%)
>40	590 (21.5%)
Highest education
<4-year college	1,997 (73.7%)
College degree	476 (17.6%)
Advanced degree	238 (8.8%)
Household income
<$49,999	1,128 (46.5%)
50,000–$99,999	991 (40.9%)
>$100,000	376 (12.6%)
Married/partnered
Yes	2,056 (75.9%)
No	652 (24.1%)
Military status
Active military	2,745 (100%)
Branch
Air Force and Air Guard	148 (5.4%)
Army	2,386 (86.9%)
Navy	168 (6.1%)
Other	10 (0.4%)
Setting of injury
During deployment	709 (31.8%)
Not deployment related	1,477 (68.2%)
PC-PTSD (No. of endorsed symptoms)	1.3 (1.6)
Positive screen for PTSD (≥3 symptoms)
Yes	767 (27.9%)
No	1,978 (72.1%)
PTSD diagnosis
Yes	837 (30.5%)
No	1,907 (69.5%)

PC = primary care; PTSD = post-traumatic stress disorder.

Table 2.

Frequency table of PTSD diagnosis and PTSD positive screening

		PTSD Positive Screening		Total
		No	Yes
PTSD diagnosis	No	1,611	296	1,907
	% within total patients	58.7%	10.8%	69.5%
	Yes	366	471	837
	% within total patients	13.1%	17.1%	30.5%
Total	Count	1,977	767	2,744
	% within total patients	72.0%	28%	100%

PTSD = post-traumatic stress disorder.

Associations Between PTSD Diagnosis, PTSD Positive Screening, and Pain-Related Symptoms

The associations between PTSD diagnosis, PTSD positive screening, pain intensity (e.g., worst pain, average pain), pain interference, physical function, fatigue, anger, depression, anxiety, and satisfaction with social roles are shown in Table 3. Results of separate analyses for male and female active duty patients are shown in Supplementary Data (available online). A t test demonstrated a significant relationship between PTSD diagnosis, PTSD positive screening, and pain-related symptoms for both male and female active duty patients. Compared with patients who did not have PTSD diagnosis and PTSD positive screening, patients with PTSD diagnosis and positive screening had higher pain interference and lower physical function and social satisfaction scores (P < 0.001) and had increased anger, anxiety, fatigue, and depression scores (P < 0.001).

Table 3.

Associations between PTSD diagnosis, positive screening, and pain-related psychosocial symptoms (n = 2,745)

Symptom	PTSD Diagnosis			PTSD Positive Screening (≥3 Symptoms)
	Yes (30.5%, n = 837)	No (69.5%, n = 1907)		Yes (27.9%, n = 767)	No (72.1%, n = 1978)
	Mean (SD)	Mean (SD)	P Values	Mean (SD)	Mean (SD)	P Values
Worst pain	7.43 (1.48)	6.95 (1.61)	<0.001	7.52 (1.42)	6.93 (1.62)	<0.001
Average pain	6.17 (1.49)	5.62 (1.52)	<0.001	6.27 (1.40)	5.60 (1.54)	<0.001
Pain interference	66.59 (5.37)	63.97 (5.63)	<0.001	66.90 (5.24)	63.94 (5.63)	<0.001
Physical function	37.91 (5.62)	40.46 (6.22)	<0.001	37.73 (5.54)	40.44 (6.21)	<0.001
Anger	60.04 (11.00)	52.96 (10.68)	<0.001	62.64 (9.86)	52.20 (10.38)	<0.001
Fatigue	62.72 (9.02)	57.86 (9.66)	<0.001	63.91 (8.58)	57.57 (9.58)	<0.001
Depression	57.98 (10.04)	50.53 (9.61)	<0.001	59.76 (9.25)	50.10 (9.42)	<0.001
Anxiety	61.70 (9.89)	53.59 (9.65)	<0.001	64.20 (8.48)	52.90 (9.33)	<0.001
Social satisfaction	36.95 (6.89)	40.11 (7.57)	<0.001	36.76 (6.52)	40.08 (7.66)	<0.001

PTSD = post-traumatic stress disorder.

Relative Weight Analysis

Figure 1 displays the rescaled relative weight and R² for each symptom from RWA-Web. These rescaled relative weights provide estimates of relative importance using the metric of percentage of predicted variance attributed to each symptom. For example, the emotional numbness domain explained 42.4% of the predicted variance in PROMIS depression scores. The total R² represents the proportion of variance in each outcome that was explained by the weighted linear combination of the four symptom domains. For example, the four PTSD symptoms together explain 23.5% of the variance in depression. All models were statistically significant (95% confidence intervals did not contain zero). Briefly, emotional numbness had the largest contribution to the total variance for depression (42.4%), fatigue (34.7%), anxiety (35.7%), anger (38.5%), average pain in the past 7 days (31.6%), pain interference (32.2%), and social satisfaction (34.2%) scores. Intrusion accounted for the largest amount of predicted variance for worst pain in the past 7 days (29.9%). Avoidance accounted for the largest amount of predicted variance for physical function (35.2%).

Figure 1.

Results of relative weight analysis

Discussion

This study’s results suggest that active duty service members with comorbidity of chronic pain and PTSD diagnosis or positive screening are more likely to experience worse pain intensity, pain interference, physical function, and worse emotional and social symptoms. Female active duty service members have higher prevalence of PTSD diagnosis and PTSD positive screening. Only a minority of patients with positive PTSD screen were diagnosed with PTSD. The results also emphasize the importance of symptoms of emotional numbness and avoidance to the manifestation of pain and its related symptoms among our sample of patients engaged in interdisciplinary care in a military treatment facility.

In the present study, approximately 28% and 30% active duty patients referred to a pain specialty center had positive screens for PTSD and PTSD diagnosis, respectively. However, only 17% had both positive screen and diagnosis. The discordance between positive screen and diagnosis may be due to a number of scenarios: 1) PTSD was present at the time of screening but never diagnosed; 2) PTSD was present at the time of diagnosis but improved over time and was no longer present or detectable at the time of screening; 3) PTSD was not present at the time of screening but patients had false-positive screens due to pain-related symptoms; 4) PTSD was not present at the time of screening but developed during the following 1-year period; 5) the diagnostic sensitivity (0.78) and specificity (0.87) of the screening instrument are limited [28]. The 30% 2-year prevalence rate is higher than PTSD prevalence among veterans primarily seeking care within the Veterans Health Administration (23%) [29], but this may be due to differences in methodology in measuring prevalence. However, the prevalence was somewhat lower than expected in this sample of active duty service members referred to a tertiary care clinic for treatment of pain that did not respond to first- and second-line therapies. Helmer et al. [30] reported that 53% of Operations Enduring and Iraqi Freedom veterans who experienced chronic widespread pain had positive PTSD screen. Patients in this study were referred to the IPMC, given a lack of treatment success in primary care or when receiving one or more secondary care therapies (e.g., only physical therapy). The PTSD positive screen prevalence and presence of a PTSD diagnosis may not reflect all PTSD typologies among active duty service members. Moreover, there is stigma of reporting PTSD symptoms, and service members may perceive that reporting PTSD symptoms will adversely impact their career [31]. Among military service members who had positive PTSD symptoms, only 23% to 40% sought mental health care [31]. Some clinicians may not record the PTSD diagnosis in the electronic health record to reduce stigma and to protect military service members’ future careers [32]. Thus, educational programs and interventions should be provided to both active duty service members and clinicians to reduce the stigma of reporting mental health issues and improve treatment engagement. Lastly, we note that screening measures and diagnoses may not correlate with actual functioning, and instead, careful examination of PTSD symptoms are recommended in future studies.

The coexistence of PTSD symptoms and chronic pain is common in the military population. In a study in a northeastern US Veterans Health Administration medical center, approximately 42% of Operations Enduring and Iraqi Freedom reported both pain and PTSD symptoms [33]. Given the important association between pain and PTSD symptoms, integrated treatments of PTSD symptoms and chronic pain among military patients are highly recommended. Otis et al. [34] conducted a 12-session integrated model, a combination of cognitive processing therapy for PTSD and cognitive behavioral therapy for chronic pain. Plagge et al. [35] also designed a collaborative care management model utilizing behavioral activation from the treatment of PTSD among returning veterans. These integrated treatments showed positive results for addressing both chronic pain and PTSD symptoms, however, the dropout rates were high, and there were no comparison groups for either study. More research is needed on how to integrate these treatments in terms of timing, components, dose, and study design to avoid participant dropout, in both veteran and active duty service member populations.

PTSD is a multifaceted condition, and it includes symptoms such as emotional numbing and fear that also appear in depression, anxiety, and other conditions. Consistent with other studies [36, 37], we found that patients who had PTSD diagnosis or positive screening reported higher fatigue, anger, anxiety, and depression than those who did not have PTSD diagnosis or positive screening. These findings are consistent with previous studies, indicating that subdiagnostic PTSD is associated with functional difficulties at a similar level as diagnosed PTSD [38]. As we discussed above, diagnostic criteria may not be a sensitive indicator of functioning; our study highlighted the importance of evaluating symptom presentations and carefully elucidating the relationship between PTSD symptoms and pain and related symptoms to better understand PTSD and chronic pain in a military population. An increasing body of literature suggests that increased allostatic load, oxidative stress, dysregulation of the hypothalamic–pituitary–adrenal axis, and increased inflammation are associated with PTSD symptoms, pain, depression, and fatigue [39, 40]. Future studies are needed to track these symptoms longitudinally and explore the causal relationships between them.

Findings from the RWA suggested that while emotional numbness accounted for the greatest variance in average pain intensity, pain interference, and emotional status (depression, fatigue, anxiety, anger), avoidance accounted for the greatest variance in physical function. Avoidance and numbness are two distinct PTSD symptom domains with different mechanisms [41]. Avoidance is regulated by effortful processes (both cognitive and behavioral) to avoid trauma-related stimuli, while numbing is regulated by automatic processes that reduce arousal. Avoidance and emotional numbness may have different patterns of correlation with psychosocial outcomes. This is consistent with previous findings that emotional numbness was significantly associated with pain outcomes in a civilian sample [5]. Such evidence supports the updated DSM-5 criteria for PTSD, which separates numbness and avoidance into two distinct features of PTSD [15]. Therefore, for patients with PTSD and chronic pain that avoid painful activities, pain neuroscience education may be an effective intervention to reduce maladaptive beliefs about pain [42] to improve physical function among active duty service members.

There are limitations of this study. First, our sample is characterized by several relevant demographic but few clinical characteristics. This study did not measure all factors that may affect development of PTSD or chronic pain, which include combat exposure, number of deployments, duration of deployments, prior experience of trauma, and previous mental health care, among others. Also, we studied a population of active duty service members referred to a pain specialty clinic, so cannot be generalized to all service members. Further, the DVPRS is a relatively new measure with limited psychometric data to measure pain. Nevertheless, it is the standard pain assessment tool required in primary care and pain specialty clinics throughout the Military Health System. In addition, we used PC-PTSD Screen to provisionally screen PTSD symptoms. Future studies can apply more robust methods and gold standard of PTSD assessments and truly elucidate the associations of PTSD symptom clusters and chronic pain. For example, clinician-administered PTSD scale for DSM-5 can be used for actual confirmation of a PTSD diagnosis. Lastly, we assumed that all patients in this study experienced chronic pain as most were referred by their primary care provider to a specialty pain center after initial treatments provided insufficient response. We did not measure how long these patients had primary pain conditions. Future studies conducted in the MAMC IPMC will incorporate the measurement of the chronicity of pain.

Conclusion

This study suggests that active duty service members with PTSD diagnosis and/or positive PTSD screens report more severe pain intensity; higher fatigue, depression, anger, and anxiety; and worse physical function in a sample of patients engaged in tertiary interdisciplinary care in a military treatment facility. Emotional numbness appeared to account for more variance in pain intensity and emotional and social functioning, whereas avoidance accounted for more variance in physical functioning. Integrated evidence-based PTSD treatment and pain neuroscience education are recommended for active duty service members with comorbidity of chronic pain and PTSD symptoms. Future research is needed to explore treatment processes and sequencing of symptom changes [43].

 

Disclosure and conflicts of interest: The authors have no conflict of interest to report.

Disclaimer: The views expressed are those of the authors and do not reflect the official policy or position of the US Army Medical Department, the Uniformed Services University, Department of the Army, Department of Defense, the US government or the Henry M. Jackson Foundation for the Advancement of Military Medicine Inc, or the National Institutes of Health. The investigators adhered to the policies for protection of human subjects as prescribed in 45 CFR 46.

Funding: This work was supported by the National Institute of Nursing Research of the National Institutes of Health (award number K24NR015340).