Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs.
Joseph L. Goulet, PhD, MS, VA Connecticut Healthcare System, 950 Campbell Ave, West Haven, CT 06516, USA. Tel: 203-932-5711 ext. 5325; Fax: 203-937-4926; E-mail: Joseph.Goulet@VA.gov.
Objective. To evaluate sex differences in the prevalence of overall pain, moderate-severe pain, and persistent pain among Veterans of Operations Enduring Freedom and Iraqi Freedom seen at VA outpatient clinics, and to evaluate sex differences in pain assessment.
Design. The observational cohort consisted of Veterans discharged from the U.S. military from October 1, 2001 to November 30, 2007 that enrolled for Veterans Administration (VA) services or received VA care before January 1, 2008. We limited the sample to the 153,212 Veterans (18,481 female, 134,731 male) who had 1 year of observation after their last deployment.
Results. Pain was assessed in 59.7% (n = 91,414) of Veterans in this sample. Among those assessed, 43.3% (n = 39,591) reported any pain, 63.2% (n = 25,028) of whom reported moderate-severe pain. Over 20% (n = 3,427) of Veterans with repeated pain measures reported persistent pain. We found no significant difference in the probability of pain assessment by sex (RR = 0.98, 95% CI 0.96, 1.00). Female Veterans were less likely to report any pain (RR 0.89, 95% CI 0.86, 0.92). Among those with any pain, female Veterans were more likely to report moderate–severe pain (RR 1.05, 95% CI 1.01, 1.09) and less likely to report persistent pain (RR 0.90, 95% CI 0.81, 0.99).
Conclusions. As the VA plans care for the increasing numbers of female Veterans returning from Iraq and Afghanistan, a better understanding of the prevalence of pain, as well as sex-specific variations in the experience and treatment of pain, is important for policy makers and providers who seek to improve identification and management of diverse pain disorders.
Pain is frequently reported by Veterans returning from war . Among Veterans of recent conflicts, including Operations Desert Shield/Desert Storm, as well as Operation Enduring Freedom/Operation Iraqi Freedom (OEF/OIF), several painful conditions are highly prevalent [2–4]. Although studies in civilian populations have reported a higher prevalence of many pain syndromes among women compared with men [5–10], sex differences in the prevalence and characteristics of pain in Veterans returning from OEF/OIF have not yet been investigated.
In civilian populations, women more commonly report specific pain syndromes including migraine headaches , oral-facial pain , fibromyalgia , and abdominal pain . Women also report more severe and longer lasting pain than men [15–17]. In some experimental human and animal models [18–20] females experience higher pain severity at lower thresholds and have less tolerance to noxious stimuli than males, but other studies have not consistently supported these findings . Few studies have addressed gender differences in pain among Veterans. A preliminary study of pain in women Veteran primary care patients  found a higher prevalence of pain compared with a similar study in male Veteran primary care patients . Another study of gender differences in health related quality of life found that, among Veterans with mental illness, women reported more pain than men .
The proportion of women in active military service (currently 15%) is increasing and is expected to double in the next 5 years . Female Veterans of OEF/OIF are using VA services more frequently than any previous female cohort . These new female Veterans are younger, more likely to identify as belonging to a racial minority, and are less likely to be married than their male counterparts . They have a high prevalence of mental health disorders , higher rates of exposure to combat trauma than previous cohorts of women Veterans, and may have high rates of exposure to sexual trauma. These factors place them at risk for chronic pain syndromes [28,29].
Routine pain assessment has become part of health screening at the VA  but previous studies have not addressed whether pain screening varies by sex. No studies have reported whether women returning from war have more severe or persistent pain than their male counterparts. An understanding of the epidemiology of pain in the growing population of female Veterans will be essential as the VA plans pain treatment services.
We hypothesized that female OEF-OIF Veterans would have more overall pain, more moderate–severe pain, and more persistent pain than their male counterparts. We designed this study to evaluate sex differences in the prevalence, severity, and persistence of pain among OEF/OIF Veterans seen at VA outpatient clinic visits during the year after returning from deployment. We also evaluated whether the probability of pain assessment at outpatient visits differed by sex, and whether any association differed by the presence of post-traumatic stress disorder (PTSD) or depression.
The population consisted of the list of Veterans obtained from VA's OEF/OIF roster provided by Defense Manpower Data Center—Contingency Tracking System Deployment File. The roster contains information on all personnel discharged from the U.S. military from October 1, 2001 to November 30, 2007 who enrolled for VA services or received VA care before January 1, 2008 (N = 406,802). Because we were interested in pain assessment, we limited our sample to Veterans who had at least one visit to a VA clinic most likely to obtain pain scores (primary care, women's clinics, and medical or surgical subspecialty clinics). We limited the sample to Veterans who had 1 year of observation after their last deployment end date in order to standardize the time between potential exposures and pain measurement. The analytic sample thus included 153,212 Veterans.
The VA OEF/OIF roster includes information on sex, date of birth, race, education, marital status, military rank, branch of service (e.g., Army, Marine Corps), and deployment start and end dates. Information on eligible visits, ICD-9 codes used to determine medical and psychiatric conditions, and pain scores was ascertained from the VA Corporate Data Warehouse.
The pain numeric rating score is recorded along with vital signs according to VA's “pain as a fifth vital sign” campaign . At each relevant visit, Veterans were asked to rate the intensity of their current pain on a scale of 0–10, where 0 is no pain and 10 is the worst possible pain. We retained only scores recorded at an outpatient visit. A missing or invalid score was considered as not assessed (<1% responses). When multiple scores were recorded on one day, we retained only the highest score.
We defined pain assessment as having any valid pain score in the 1-year observation period. Then, using the first valid pain score in the observation period, we defined any pain as a score ≥1. Consensus on the optimal numeric rating score cutoff for moderate–severe pain is lacking, but most studies use either 4 or 5 [31,32]. We defined moderate–severe pain as a score ≥4 because this is most consistent with VA clinical practice and policy . We had no direct measure of pain duration, so we defined persistent pain as three or more pain scores ≥4 recorded in at least three different months. This is consistent with common definitions of chronic pain that require at least 3–6 months duration.
ICD-9 codes from outpatient visits were then mapped to validated diagnostic groupings in order to determine the prevalence of PTSD and depression . A Veteran was considered to have PTSD or depression if codes occurred on two or more outpatient visits. This methodology has been used in the identification of psychiatric disorders in administrative data  and in identification of HIV in Medicaid data .
Chi-square tests were used to examine bivariate relationships between categorical variables and t-tests or nonparametric tests, as appropriate, were used for continuous variables. Because our study is a cohort study and the primary outcomes were common events, we used Poisson regression with a log link and robust variance estimates to calculate relative risks (RR) and 95% confidence intervals (CI). To control for potential confounding that could distort the risk estimates, all demographic variables that were significant in bivariate analyses were entered into the model as covariates. PTSD and depression diagnoses were not entered into the multivariable model given that the diagnoses may have occurred after the pain assessment.
The sample consisted of 18,481 female and 134,731 male Veterans with at least one eligible visit in the year after their last deployment (Table 1). Compared with males, female Veterans were younger (mean age 30.2 vs 32.6 P < 0.0001); more likely to be Black (30.3% vs 14.2% P < 0.0001); more likely to have a college education (28.7% vs 24.0%, P < 0.0001); less likely to be married (31.8% vs 50.1%, P < 0.0001); more likely to be officers (8.0% vs 7.0 %, P < 0.0001); and less likely to serve in the Marine Corps (2.9% vs 11.2%, P < 0.0001). PTSD was less frequently diagnosed in female Veterans (9.9% vs 11.3%, P < 0.0001); in contrast, depression was more common (12.2% vs 7.5%, P < 0.0001). Female Veterans also had a higher mean number of clinic visits (4.7 vs 4.1, P < 0.0001).
Table 1. Demographics, pain assessment, pain characteristics, by sex in OEF/OIF Veterans utilizing VA clinics in first year after end of deployment
Female n = 18,481
Male n = 134,731
Clinic visits include general medical, primary care, and women's clinics.
Defined as pain score ≥4, assessed only among those with any pain score ≥1.
Defined as ≥3 pain scores ≥4 in three different months, assessed among those with any pain score ≥1 and at least 3 available pain scores.
SD = standard deviation; PTSD = post-traumatic stress disorder.
Pain was assessed in 59.7% (n = 91,414) of Veterans in this sample. Among those assessed, 43.3% (n = 39,591) reported any pain. Among those reporting any pain, 63.2% (n = 25,028) reported moderate-severe pain. Of the 16,611 Veterans with at least three pain scores in three separate months, 20.6% (n = 3,427) met our definition of persistent pain (Figure 1).
In bivariate analysis, there was no significant difference in the probability of pain assessment by sex (59.6% vs 60.1%, P = 0.24). Female Veterans were significantly less likely to report any pain (38.1% vs 44.0%, P < 0001). Among Veterans with any pain, female Veterans were more likely to report moderate–severe pain (68.0% vs 62.6%, P < 0.0001) and less likely to report having persistent pain (18.0% vs 21.2%, P < 0.001).
Results were similar after adjusting for potentially confounding characteristics in multivariable analyses (Table 2). We found no significant difference in the probability of pain assessment by sex (RR = 0.98, 95% CI 0.96, 1.00). Female Veterans were less likely to report any pain (RR 0.89, 95% CI 0.86, 0.92). Among those with any pain, female Veterans were more likely to report moderate–severe pain (RR 1.05, 95% CI 1.01, 1.09) and less likely to have persistent pain (RR 0.90, 95% CI 0.81, 0.99) than male Veterans. Estimates for the full set of covariates are shown in Table 2.
Table 2. Results of Poisson regression models assessing the relative risk for pain assessment, any pain, moderate pain, and persistent pain
Relative risk (95%CI)
95% CI = 95% confidence interval; Ref. = reference group.
0.98 (0.96, 1.00)
0.89 (0.86, 0.92)
1.05 (1.01, 1.09)
0.90 (0.81, 0.99)
1.00 (1.00, 1.00)
1.00 (1.00, 1.00)
1.00 (1.00, 1.00)
1.00 (1.00, 1.01)
1.05 (1.02, 1.08)
1.00 (0.96, 1.04)
1.00 (0.95, 1.06)
0.99 (0.87, 1.12)
1.10 (1.08, 1.12)
0.88 (0.86, 0.90)
0.98 (0.95, 1.01)
0.80 (0.73, 0.87)
1.00 (0.98, 1.02)
0.97 (0.94, 1.00)
1.22 (1.18, 1.26)
1.10 (1.01, 1.20)
1.07 (1.05, 1.10)
0.92 (0.90, 0.95)
1.01 (0.97, 1.05)
0.68 (0.61, 0.76)
1.05 (1.01, 1.09)
0.97 (0.91, 1.03)
1.09 (1.01, 1.17)
0.70 (0.56, 0.89)
0.92 (0.88, 0.96)
0.97 (0.90, 1.03)
1.16 (1.07, 1.25)
1.17 (0.93, 1.46)
1.03 (1.02, 1.05)
0.92 (0.90, 0.95)
0.90 (0.87, 0.93)
0.75 (0.68, 0.82)
Less than high school
1.05 (0.99, 1.10)
0.98 (0.91, 1.07)
1.02 (0.93, 1.13)
0.89 (0.67, 1.17)
High school graduate
0.84 (0.82, 0.87)
0.88 (0.83, 0.92)
0.86 (0.80, 0.92)
0.78 (0.64, 0.95)
0.81 (0.76, 0.87)
0.89 (0.80, 0.99)
0.90 (0.78, 1.04)
0.73 (0.50, 1.07)
0.70 (0.68, 0.73)
0.92 (0.88, 0.97)
0.99 (0.93, 1.05)
0.86 (0.71, 1.05)
0.90 (0.88, 0.92)
1.02 (0.99, 1.06)
1.01 (0.97, 1.06)
0.96 (0.83, 1.11)
0.91 (0.88, 0.93)
0.94 (0.90, 0.98)
0.97 (0.92, 1.03)
0.95 (0.81, 1.11)
1.01 (1.01, 1.01)
1.02 (1.02, 1.02)
1.01 (1.00, 1.01)
1.03 (1.02, 1.03)
When stratified by depression or PTSD diagnosis, Veterans with depression or PTSD were more likely to be assessed for pain (P < 0.0001); however, there was no significant difference in the proportion of Veterans assessed for pain by sex in those with PTSD and depression (P > 0.05 in both cases) (Table 3).
Table 3. Homogeneity of odds for assessment, by PTSD and depression diagnoses
PTSD = post-traumatic stress disorder.
To our knowledge, this is the first study of the sex-specific prevalence of pain in OEF-OIF Veterans. We used pain scores from VA outpatient clinic visits to estimate pain prevalence in Veterans returning from the OEF/OIF conflicts. Importantly, our results documented a lower prevalence of pain in women compared with men in the first year post-deployment even after adjusting for other demographic and clinical factors. In this sub-population of Veterans with pain, we found that women were slightly more likely than men to have moderate–severe pain; however, women had a lower prevalence of moderate–severe pain overall than men. Persistent pain was also less common among women than among men.
The lower prevalence of pain in women Veterans is an unexpected finding that is contrary to studies conducted in civilian populations. Several hypotheses may be pertinent to pain in Veterans of OEF/OIF. First, the relatively low prevalence of pain in returning female soldiers compared with male soldiers may reflect differences in exposure to combat trauma and injury. While women are excluded from serving in direct combat, they do work in a variety of support positions and may come under direct fire, so the true risk of injury for female Veterans is difficult to assess.
If one assumes an equal burden of risk for injury, two other potential hypotheses can be proposed. The high intensity and persistent level of threat in OEF/OIF may act as an equalizer of risk for stress-associated conditions, including pain, so that risk is more strongly related to the intensity and frequency of combat experience than to gender . In addition, a “healthy warrior effect” (the disproportionate loss of psychologically unfit personnel early in training)  might even the playing field for deployed men and women, putting them at similar risk for pain syndromes.
The surprisingly lower rates of pain among women Veterans might also be a result of their reluctance to seek VA treatment due to either their gender or their higher likelihood of mental health disorders, especially depression. Women Veterans report more barriers to VA care and have used the VA less often than male Veterans in past eras [39–44]. Previous studies have also found that Veterans with mental health disorders are more likely to report barriers to seeking VA treatment , and mental health disorders, especially depression and PTSD , are often associated with chronic pain.
Our study has several limitations. Primarily, the accuracy of the pain numeric rating scale score as a screening test for pain has been questioned. Two studies that compared pain scores collected in routine clinical practice to the Brief Pain Inventory, a validated instrument that includes an assessment of pain related functional impairment, found that the pain score had only a modest accuracy for identifying patients with clinically important pain. In the first study, the pain numeric rating scale score missed nearly a third of patients with clinically important pain, and in the second study, pain numeric rating scale scores underestimated pain in 33% of cases and overestimated it in 12% [47,48]. This suggests that our estimates of pain prevalence based on pain scores may underestimate the true prevalence of pain in the OEF/OIF population; however, there is no evidence that misestimation would differ by sex. In addition, our analysis was limited to patients seen in clinics most likely to administer a pain score and thus, we are unable to evaluate OEF/OIF Veterans seen in other clinics or those not seeking VA care.
In conclusion, among OEF/OIF Veterans seen in VA outpatient clinics within 1 year of their last deployment, we found that women had a lower prevalence of overall pain, moderate-severe pain, and persistent pain then men. Examining whether sex differences in pain persist, increase or decrease over time is an important next step in VA pain research, and will help guide planning, resource allocation, and policy for women Veterans and OEF/OIF Veterans of both sexes.
Financial support: VA grants DHI 07-065-1 (Brandt PI, Haskell, Skanderson); VA HSR&D Research Enhancement Award Program (REAP) PRIME Project (Kerns PI; Goulet); CD207215-2 (Krebs PI).