Sexual Health Problems among Service Men: The Influence of Posttraumatic Stress Disorder

ABSTRACT

Military operational stressors, such as combat exposure, may increase the risk of sexual health problems. This study examined factors associated with sexual health problems, and tested the mediating effect of probable posttraumatic stress disorder (PTSD) on the association between stressors (i.e., combat deployment and sexual assault) and sexual health problems among U.S. service men. Using multivariable logistic regression (n = 16,603) and Cox proportional hazards models (n = 15,330), we estimated the risk of self-reported sexual health difficulties and sexual dysfunction medical encounters, respectively. Mediation analyses examined the effect of probable PTSD as an intermediate factor between high combat deployment and sexual assault on sexual health problems. Approximately 9% endorsed sexual health difficulties and 8% had a sexual dysfunction. Risk factors for these sexual health problems included older age, lower education level, enlisted rank, disabling injury, certain medical conditions, and higher body mass index. Probable PTSD significantly mediated the associations between high combat with sexual health problems and sexual assault with sexual dysfunction. Additionally, high combat was directly associated with sexual health difficulties. These findings indicate a relationship between these stressors and sexual health problems which suggests that treatment options should be expanded, especially to include psychogenic sexual dysfunctions.

Introduction

Sexual health encompasses physical, psychological, and social aspects of sexuality and sexual relationships and is not solely the lack of disease (World Health Organization, 2016). Sexual health influences relationship quality, quality of life, overall health, and well-being (Brotto et al., 2016; Goodwach, 2017). Conversely, sexual dysfunction encompasses a range of conditions that may prevent an individual from experiencing satisfaction from sexual activity and can negatively affect one’s life (World Health Organization, 2016). Moreover, sexual dysfunction has been shown to be associated with disability, mental disorders, physical health conditions, and suicide (Blais, Monteith et al., 2018; Khalifian et al., 2020; Schlichthorst et al., 2016; Tan et al., 2012).

Evidence suggests that military service members and veterans may be at increased risk for sexual health problems, sexual dysfunction (Breyer et al., 2014; Mulligan & Moss, 1991; Wilcox et al., 2014, 2015), and compulsive sexual behavior (Blais, 2020). Despite this, little empirical evidence has focused on the risk factors of sexual health problems among service men. According to Department of Veterans Affairs (VA) medical records, approximately 5% of male veterans returning from Iraq and Afghanistan have been diagnosed with sexual dysfunction, most commonly erectile dysfunction (ED) (Breyer et al., 2014; Hosain et al., 2013). Between 2004 and 2013, the annual incidence rate of ED among active duty service men more than doubled from 5.8 to 12.6 cases per 1,000 person years, with approximately half of diagnoses being for psychogenic ED (Armed Forces Health Surveillance Center, 2014). While the reason for this increase in ED incidence is not known, during this time there was an increase in other comorbid risk factors such as posttraumatic stress disorder (PTSD) (Cameron et al., 2019; Judkins et al., 2020), and extensive marketing campaigns to raise awareness of ED (Lexchin, 2006).

Certain experiences that can occur during military service, such as combat exposure (Badour et al., 2015; Mills & Turnbull, 2004) and sexual assault (Suvak et al., 2012; Turchik et al., 2012) may increase risk for sexual health problems, either directly or indirectly (Figure 1) (Rosebrock & Carroll, 2016). A recent comprehensive review by Rosebrock and Carroll (2016) discussed the link between individual and military-related factors with sexual functioning. Although this model was proposed among service women, there is evidence that similar relationships may exist between vulnerability factors (e.g., childhood trauma) and traumatic experiences (e.g., combat deployment and sexual assault) with sexual health and functioning of service men (Badour et al., 2015; Bigras et al., 2017; Turchik et al., 2012). The model proposes that traumatic experiences can affect sexual functioning either directly or indirectly through subsequent mental health conditions or other trauma sequelae. For instance, combat and sexual assault are strongly associated with PTSD (Kang et al., 2005; Smith et al., 2008), which is a risk factor for sexual dysfunction among service men (Breyer et al., 2016; Spivak et al., 2003). In line with the Rosebrock model (Rosebrock & Carroll, 2016), this suggests PTSD may be a mediator between these stressors and sexual health problems. There are multiple potential mechanisms for sexual health problems in individuals with PTSD, such as (1) emotional numbing (Bhalla et al., 2018; Nunnink et al., 2010), (2) changes in neuroendocrine markers (Lehrner et al., 2016; Yehuda et al., 2015), and (3) prescribed medication (Badour et al., 2015; Breyer et al., 2014). Yet, to our knowledge, no study has examined PTSD as a mediator between stressors such as combat and sexual assault with sexual dysfunction among service men.

Figure 1. Mediation pathways. Direct (C) Indirect (A•B) Total (A•B•C). PTSD = posttraumatic stress disorder.

Some evidence suggests that sexual assault is associated with sexual health problems among service men. Among a sample of male Marines (n = 286), childhood and adult sexual assault were predictive of different aspects of sexual functioning (Suvak et al., 2012). A large study that examined the Veterans Health Administration medical data also found that sexual dysfunctions were more prevalent among those with a history of sexual trauma while in service (Turchik et al., 2012). Although prior studies have identified sexual trauma as a predictor of sexual functioning problems among male veterans (Suvak et al., 2012; Turchik et al., 2012), these studies either could not or did not test for possible mediators. However, certain PTSD symptom clusters, such as emotional numbing and hyperarousal, have been found to mediate the relationship between sexual trauma and sexual health/satisfaction among female service members (Blais, Geiser et al., 2018). Since the effects of sexual trauma may differ by sex (O’Brien et al., 2008), it is important to examine the potential mediating role of PTSD in men as well.

Previous findings on the relationship of combat and sexual health problems have been less consistent, which may be related to small sample sizes, the use of a variety of combat measures, and failure to investigate possible indirect effects. In one study of male Marines (n = 286), combat experience was not associated with several domains of self-reported sexual health, although severity of recent combat was not assessed (Suvak et al., 2012). In another study of Army personnel with partners (n = 221), no association was found between combat experience and sexual anxiety (Bhalla et al., 2018). Conversely, among Operation Enduring Freedom and Operation Iraqi Freedom veterans (n = 150) seeking care for PTSD symptoms, researchers found an association between recent combat severity and sexual desire problems (Badour et al., 2015).

Lastly, some previous studies have found that certain demographic (e.g., age and education level) and military factors (e.g., Army, National Guard/Reserve members, and enlisted personnel) may increase risk for sexual dysfunction in male service members (Armed Forces Health Surveillance Center, 2014; Breyer et al., 2014; Hosain et al., 2013; Wilcox et al., 2014). However, these studies have been limited by cross-sectional designs, small samples of veterans, and/or exclusive use of VA administrative records (Badour et al., 2015; Bhalla et al., 2018; Breyer et al., 2014, 2016; Cosgrove et al., 2002; Helmer et al., 2013; Hirsch, 2009). Given the increasing rates of sexual health problems and their significant impact on well-being, there is a need to better understand specific mechanisms that may increase risk for sexual health problems among male service members and veterans.

The overarching aim of the current study was to extend previous research by prospectively investigating factors associated with sexual health outcomes among a large cohort of service men. Specifically, the main objective was to examine the association of sexual assault and combat exposure with sexual health outcomes, both directly and indirectly through PTSD. Based on the existing theory and literature, we hypothesized that the associations of sexual assault and combat exposure with sexual health problems would be mediated by PTSD. Our second objective was to identify other factors (demographic, military, historical psychological health, other stressful life experiences, and physical health) associated with sexual health outcomes. Due to the lack of previous research in this area among service men, this analysis was more exploratory in nature, rather than hypotheses testing.

Method

Study Population

The Millennium Cohort Study is an ongoing longitudinal study designed to investigate the effects of military service across the lifespan (Ryan et al., 2007). The Cohort enrolled over 200,000 participants in four phases between 2001 and 2013. Participants represent all service branches and components (i.e., active duty, Reserve, National Guard) of the military. Participants provided voluntary, informed consent and completed baseline surveys at the time of enrollment. Approximately every 3 to 5 years, participants complete follow-up surveys regardless of their military status. Surveys encompass a wide range of topics, including physical, mental, and behavioral health, as well as military and nonmilitary life experiences. Details of the Millennium Cohort Study have been published previously (Ryan et al., 2007). The study was approved by the Naval Health Research Center Institutional Review Board (protocol number NHRC.2000.0007).

Given that sexual health and dysfunction differ by sex, this study included only men, with a companionate study of sexual health among service women conducted separately. Participants eligible for these analyses must have completed the follow-up survey conducted between 2011 and 2013 (n = 61,320), referred to here as Time 1. To examine sexual health in service men, participants who were no longer serving in the military at Time 1 were excluded (n = 32,278), leaving 29,042 active duty, National Guard, or Reserve service men. For our first analysis examining self-reported sexual health, participants also were required to have completed a follow-up survey conducted between 2014 and 2016, referred to as Time 2 (n = 22,463 with Time 2 survey data). To account for potential differences in sexual health during deployments, men who were deployed or returned from a deployment within 30 days of either survey were excluded (n = 2,252). Of the 20,211 remaining participants eligible for the analysis of self-reported sexual health, those missing risk factor data (e.g., demographic factors, physical health, life experiences) were excluded (n = 3,608), resulting in a final study population of 16,603 individuals.

To conduct the second analysis on sexual dysfunction, archived electronic medical records were used to assess the outcome. This analysis was restricted to active duty service members at Time 1 (n = 17,922) because there is not full visibility of medical care for Reserve/National Guard members. Further, in order to examine new-onset sexual dysfunction, participants with a sexual dysfunction encounter prior to Time 1 (n = 1,421) were excluded. Of the 16,501 remaining eligible participants, those missing risk factor data (e.g., demographic factors, physical health, life experiences) were excluded (n = 1,171), resulting in a final study population of 15,330 active duty participants for the second analysis.

Outcome Measures

For the first analysis, self-reported sexual health was assessed at Time 1 and Time 2, based on two Patient Health Questionnaire (PHQ) items (Spitzer et al., 2000): sexual health difficulties were defined as responding “bothered a lot” to “pain or problems during sexual intercourse” or “little or no sexual desire or pleasure during sex” in the past four weeks. Sexual health at Time 2 was examined as the outcome, while sexual health at Time 1 was adjusted for in the sexual health difficulty models.

For the second analysis, new-onset sexual dysfunction was identified from electronic medical record data in the Military Health System Data Repository for active duty participants. Medical visits from U.S. military medical facilities and other reimbursable facilities (e.g., via TRICARE) were evaluated for select International Classification of Diseases, 9^th Revision, Clinical Modification (ICD-9-CM) codes. The accuracy of the U.S. military record system has previously been shown comparable with that of other large health-care systems (Meyer & Krakauer, 1998). Sexual dysfunction was determined by having at least one of the ICD-9-CM diagnostic codes that indicate sexual dysfunction: psychosexual dysfunction, unspecified (302.70), hypoactive sexual desire disorder (302.71), psychosexual dysfunction with inhibited sexual excitement (302.72), male orgasmic disorder (302.74), premature ejaculation (302.75), other psychosexual dysfunction (302.79), pathologic sexuality (302.9), painful erection (607.3), impotence of organic origin (607.84), retrograde ejaculation (608.87), painful ejaculation (608.89), or decreased libido (799.81) (Armed Forces Health Surveillance Center, 2014; Hosain et al., 2013; Wilcox et al., 2014). Diagnosis date corresponded to the earliest encounter after Time 1.

Deployment Experience

Deployment experience was assessed through administrative electronic data from the Defense Manpower Data Center (DMDC) in combination with self-reported combat experience(s). Recent deployment was determined by the presence of deployment dates in support of the operations in Iraq or Afghanistan within the 3 years prior to Time 1. Using the 12-item modified version of the Walter Reed Combat Experiences Scale (e.g., seeing dead bodies, being attacked, having an improvised explosive device explode near you), all combat experiences endorsed were summed to represent severity of combat experiences (range 0–12). Recent deployment and combat severity were combined into a four-level variable: no deployment, deployment without combat, deployment with low combat (score 1–3), and deployment with high combat (score ≥4) (Armenta et al., 2018; Hoge et al., 2004). Data from surveys completed prior to Time 1 (i.e., 2003, 2006, 2009) were used to classify historical deployment experience (never deployed, deployed without combat, and deployed with combat). This historical variable was based on a brief 5-item combat exposure measure (e.g., exposure to dead bodies, prisoners of war or refugees, or maimed soldiers or civilians) that was included on all Millennium Cohort surveys (Porter et al., 2018).

Sexual Assault

Sexual assault was assessed by endorsement of “forced sexual relations or sexual assault” on any Millennium Cohort survey between 2001 and Time 1. Therefore, sexual assault captured lifetime sexual assaults that may have occurred during or prior to military service.

Mental Health

Probable PTSD was measured using Time 1 data from the PTSD Checklist−Civilian Version, a validated instrument used to rate the severity of 17 PTSD symptoms (Blanchard et al., 1996) that has been shown to have good internal consistency in this cohort (T. C. Smith et al., 2007). Based on criteria from the Diagnostic and Statistical Manual of Mental Disorders, 4^th edition (DSM-IV), probable PTSD (no/yes) was indicated for those responding “moderate” or greater on at least one intrusion symptom, three avoidance symptoms, and two hyperarousal symptoms (American Psychiatric Association, 1994). Major depressive disorder (MDD; no/yes), assessed using the PHQ-8, was defined as responding “more than half the days” or more to at least five items, one of which was depressed mood or anhedonia. (Spitzer et al., 1999, 1994). This definition corresponds to DSM-IV (American Psychiatric Association, 1994) diagnostic criteria. Data from surveys prior to Time 1 (i.e., 2003, 2006, 2009) were used to classify having a history of PTSD (only), MDD (only), comorbid PTSD/MDD, or neither condition.

Risk Factors

Age, race/ethnicity, service branch, component (active duty, Reserve/National Guard), and pay grade were obtained from DMDC and reflect status at Time 1. Service component was only examined as a predictor of sexual health difficulties since new-onset sexual dysfunction was limited to active duty personnel. Marital status and education level were self-reported at Time 1. Sexual orientation and childhood trauma were only assessed at Time 2 and, therefore, only included in the sexual health difficulties analyses, while other predictors (e.g., disabling illness, medical conditions) were assessed at Time 1 and were included in both analyses. Sexual orientation was categorized as “heterosexual or straight,” “gay,” “bisexual” and “prefer not to answer.” Childhood trauma (yes or no) was assessed using four items from the Juvenile Victimization Questionnaire (Finkelhor et al., 2005; Hamby et al., 2005) (i.e., neglect, sexual abuse, verbal abuse, physical abuse before age 18). Recent (in the last 3 years) disabling illness was assessed with one item (Sale et al., 2012). Number of chronic medical conditions (diagnosis of sleep apnea, hypertension, or diabetes in the previous 3 years) (Budweiser et al., 2009; DeLamater & Karraker, 2009), were categorized as none, 1, or ≥2. Body mass index (BMI) was calculated as weight (kg)/height (m)² (Kolotkin et al., 2012).

Statistical Analysis

Descriptive statistics were performed to compare demographic, military, and behavioral factors by self-reported sexual health difficulties and new-onset sexual dysfunction. To investigate probable PTSD as a mediator between stressors and sexual health, causal mediation analyses were conducted using a SAS macro (Valeri & VanderWeele, 2015). Standard errors and confidence intervals were produced for the total, direct, and indirect effects when fully adjusting for available risk factors. Total effects were defined as the associations between stressors and outcomes when the mediator, probable PTSD, was not included in the model. The indirect effects estimated in our models were the impacts of stressors on sexual health mediated through probable PTSD. The remaining direct effects of each stressor on sexual health were estimated while holding the mediator, probable PTSD, constant. Finally, the proportion mediated was calculated, which is the ratio of the indirect effect over the total effect (VanderWeele, 2011).

Multivariable odds ratios (OR) and 95% confidence intervals (95% CI) were estimated with a logistic regression model. This multivariable logistic regression model was performed to determine which factors (i.e., race/ethnicity, age, marital status, education, sexual orientation, service branch, pay grade, service component, historic deployment experience, historic PTSD/depression, childhood trauma, disabling injury, history of sexual health difficulties, medical diagnoses, and BMI) were significantly associated with self-reported sexual health difficulties. In order to determine the influence of each factor above and beyond the other variables, all factors were included in the multivariable model regardless of statistical significance.

Similarly, hazard ratios (HR) and 95% CI were estimated with a Cox proportional hazards time-to-event model to determine which factors were associated with new-onset sexual dysfunction. Factors included were the same as the ones in the logistic regression model but excluded: sexual orientation, service component, history of sexual health difficulties, and childhood trauma. Factors were included in the multivariable model regardless of statistical significance. Person-days were calculated from date of Time 1 to the date of sexual dysfunction diagnosis, leaving active duty service, end of the follow-up period (January 1, 2016), or death, whichever occurred first. Multicollinearity was tested with a variance inflation factor threshold of ≥4, and the proportional hazards assumption was verified by plotting the cumulative sums of Martingale residuals for sexual assault and recent deployment experience against person-days. Data analysis was conducted using SAS version 9.4 (SAS Institute Inc., Cary, NC).

Results

Among the study populations, 8.9% of the 16,603 participants endorsed sexual health difficulties and 7.7% of the 15,330 active duty participants had a new-onset sexual dysfunction encounter (Table 1). While the sexual dysfunction population was restricted to active duty personnel, the two study populations were similar across most demographic and military characteristics. In both analyses, the majority of participants were white, non-Hispanic, married, less educated (<college degree), in the Army, and enlisted (Table 1).

Table 1. Baseline characteristics by sexual health outcomes

	Study sample n = 16,603		Sexual health difficulties^b n = 1,477 (8.90%)		Study sample n = 15,330		Sexual dysfunction n = 1,179 (7.69%)
Characteristic^a	n	Col %	n	Row %	n	Col%	n	Row %
Independent predictors
Recent deployment experience^c
Not deployed	10,475	63.09	910	8.69	7817	50.99	589	7.53
Deployed, no combat	1539	9.27	74	4.81	1888	12.32	124	6.57
Deployed, low combat	2109	12.70	169	8.01	2461	16.05	187	7.60
Deployed, high combat	2480	14.94	324	13.06	3164	20.64	279	8.82
Sexual assault^d
No	16,316	98.27	1422	8.72	15,093	98.45	1149	7.61
Yes	287	1.73	55	19.16	237	1.55	30	12.66
Mediator
PTSD^e
No	1566	94.36	1105	7.05	14,472	94.40	1063	7.35
Yes	937	5.64	372	39.70	858	5.60	116	13.52
Demographic factors
Race/ethnicity
White, non-Hispanic	12,638	76.12	1065	8.43	11,004	71.78	747	6.79
Black, non-Hispanic	1130	6.81	131	11.59	1278	8.34	152	11.89
Other	2835	17.08	281	9.91	3048	19.88	280	9.19
Age
Mean (standard deviation)	37.66 (8.42)		39.00 (8.44)		34.51 (6.85)		37.38 (7.08)
Marital status
Single	2018	12.15	163	8.08	1730	11.29	70	4.05
Married	12,904	77.72	1114	8.63	12,217	79.69	960	7.86
Divorced/separated	1681	10.12	200	11.90	1383	9.02	149	10.77
Education
Some college or less	8376	50.45	936	11.17	8973	58.53	698	7.78
College degree	4225	25.45	310	7.34	3131	20.42	235	7.51
Graduate degree	4002	24.10	231	5.77	3226	21.04	246	7.63
Sexual orientation
Heterosexual	16,035	96.58	1384	8.63	–
Gay	151	0.91	25	16.56
Bisexual	141	0.85	26	18.44
Prefer not to answer	276	1.66	42	15.22
Military factors
Service branch
Army	7087	42.69	824	11.63	5707	37.23	558	9.78
Navy/Coast Guard	2833	17.06	224	7.91	3070	20.03	221	7.20
Marine Corps	1230	7.41	124	10.08	1722	11.23	96	5.57
Air Force	5453	32.84	305	5.59	4831	31.51	304	6.29
Pay grade
Enlisted	10,292	61.99	1110	10.79	9931	64.78	785	7.90
Officer	6311	38.01	367	5.82	5399	35.22	394	7.30
Service component
Reserve/Guard	6788	40.88	613	9.03	–
Active duty	9815	59.12	864	8.80
Deployment experience 2001–2009^f
Not deployed	6774	40.80	518	7.65	5789	37.76	381	6.58
Deployed, no combat	3822	23.02	262	6.86	3673	23.96	288	7.84
Deployed with combat	6007	36.18	697	11.60	5868	38.28	510	8.69
Psychological health 2001–2009
PTSD/depression 2001–2009^e^,^g
Neither	15,543	93.62	1176	7.57	14,413	94.02	1069	7.42
Depression	198	1.19	45	22.73	201	1.31	27	13.43
PTSD	529	3.19	144	27.22	445	2.90	52	11.69
PTSD and depression	333	2.01	112	33.63	271	1.77	31	11.44
Life experiences
Childhood trauma^h
No	9821	59.15	612	6.23	–
Yes	6782	40.85	865	12.75
Disabling injury
No	15,683	94.46	1259	8.03	14,498	94.57	1082	7.46
Yes	920	5.54	218	23.70	832	5.43	97	11.66
Physical health
Sexual health difficulties^b
No	15,834	95.37	1075	6.79	–
Yes	769	4.63	402	52.28
Medical diagnoses^i
None	13,335	80.32	915	6.86	12,743	83.12	856	6.72
1	2758	16.61	422	15.30	2238	14.60	264	11.80
2–3	510	3.07	140	27.45	349	2.28	59	16.91
Body mass index
Normal/underweight (<25)	3998	24.08	211	5.28	3953	25.78	221	5.59
Overweight (25–29.9)	9642	58.07	811	8.41	9064	59.13	742	8.19
Obese (≥30)	2963	17.85	455	15.36	2313	15.09	216	9.34

^aAll were assessed at Time 1 except for sexual orientation and childhood trauma, which were assessed at Time 2.

^bSelf-reported sexual health difficulties was indicated if “bothered a lot” was endorsed to, “During the last 4 weeks, how much have you been bothered by any of the following problems? Little or no sexual desire or pleasure during sex? Pain or problems during sexual intercourse?”

^cDeployment, if deployment date was within 3 years of Time 1, was determined using administrative date information; combat was assessed by 12 combat-related items; participants with 0 endorsed items were categorized as no combat, 1–3 endorsed items were categorized as low combat, and 4 or more were categorized as high combat.

^dSexual assault was assessed by endorsement of “forced sexual relations or sexual assault” on any Millennium Cohort survey between 2001 and Time 1, which captured lifetime incidents of sexual assault.

^ePosttraumatic stress disorder (PTSD) was assessed using the PTSD Checklist–Civilian Version.

^fDeployment, if deployment date was between 2001–2009, was determined using administrative date information; combat was assessed by 5 combat-related items on surveys between 2001 and 2009.

^gDepression was assessed using the Patient Health Questionnaire-8.

^hChildhood trauma was assessed using 4 items from the Juvenile Victimization Questionnaire that ask about neglect, sexual abuse, verbal abuse, and physical abuse before age 18.

ⁱMedical diagnoses assessments included sleep apnea, hypertension, or diabetes in the previous 3 years.

Service men who deployed recently and experienced high combat, experienced sexual assault, and/or screened positive for probable PTSD were more likely to report both sexual health difficulties and new-onset sexual dysfunction (Table 1). Demographic and military factors associated with higher proportions of participants with sexual health difficulties and sexual dysfunction included black, non-Hispanic race/ethnicity, older age, divorced/separated marital status, lower education level, gay or bisexual sexual orientation, Army affiliation, enlisted status, and history of deployment experience (i.e., reported on 2001–2009 surveys). Those with a history of probable PTSD or depression, life stressors, and physical health conditions had higher levels of sexual health difficulties and sexual dysfunction.

Results from the mediation analyses are shown in Figure 2. Probable PTSD significantly mediated the impact of all pathways examined, except for the association between sexual assault and sexual health difficulties. The proportion mediated by probable PTSD ranged from 1.7% (sexual assault on sexual health difficulties) to 29.1% (high combat on sexual dysfunction). In addition, the direct effect of recent high combat exposure during deployment on sexual health difficulties was significant: those who recently deployed with high combat were at increased risk of sexual health difficulties compared with those who did not recently deploy beyond what could be accounted for by probable PTSD alone (direct effect adjusted odds ratio [AOR]: 1.24; 95% CI: 1.06–1.45).

Figure 2. Mediation analyses examining posttraumatic stress disorder (PTSD) between stressors and sexual health outcomes. The effect estimate (odds ratio [OR] for sexual health difficulties and hazards ratio [HR] for sexual dysfunction) and 95% confidence interval (CI) are reported. ^aModels are adjusted for race/ethnicity, age, marital status, education level, service branch, pay grade, previous deployment experience, historical PTSD/depression symptoms, disabling injury, number of medical diagnoses, and body mass index. Models i and ii are also adjusted for component, Time 1 sexual health, sexual orientation, and childhood trauma. Models i and iii are also adjusted for sexual assault; models ii and iv are also adjusted for recent deployment experiences. i. Mediation analysis of the effect of high combat on sexual health difficulties. Proportion mediation = 13.2%. ii. Mediation analysis of the effect of sexual assault on sexual health difficulties. Proportion mediation = 1.7%. iii. Mediation analysis of the effect of high combat on sexual dysfunction. Proportion mediation = 29.1%. iv. Mediation analysis of the effect of sexual assault on sexual dysfunction. Proportion mediation = 7.9%.

Table 2 shows the adjusted associations of risk factors for sexual health difficulties and sexual dysfunction. Factors associated with both sexual health difficulties and new-onset sexual dysfunction included: older age, less educational attainment, enlisted status, disabling injury, medical diagnoses, and higher BMI (i.e., overweight/obese). The four risk factors examined in the sexual health difficulties analysis only (i.e., sexual orientation, active duty status, experienced childhood trauma, history of sexual health difficulties) were significantly associated with the outcome. In addition, Army and Marine Corps personnel (compared with Air Force personnel) and those with a history of PTSD or MDD between 2001 and 2009 had increased odds for sexual health difficulties, but not for new-onset sexual dysfunction. Black, non-Hispanic, and current or formerly married men were at increased risk of new-onset sexual dysfunction, but not for sexual health difficulties.

Table 2. Adjusted associations of risk factors^a with sexual health outcomes

	Sexual health difficulties^b (N = 16,603)				Sexual dysfunction (N = 15,330)
	AOR^c	95% CI			AHR^d	95% CI
Demographic factors
Race/ethnicity (ref: White, non-Hispanic)
Black, non-Hispanic	0.99	0.79	–	1.23	1.43	1.19	–	1.71
Other	1.03	0.88	–	1.21	1.05	0.91	–	1.22
Age (5-year increase)	1.14	1.10	–	1.19	1.46	1.39	–	1.53
Marital status (ref: Single)
Married	0.94	0.77	–	1.15	1.35	1.05	–	1.72
No longer married	0.95	0.74	–	1.23	1.93	1.45	–	2.58
Education (ref: Graduate degree)
Some college or less	1.49	1.18	–	1.87	1.44	1.13	–	1.82
College degree	1.20	0.98	–	1.48	1.29	1.06	–	1.58
Sexual orientation (ref: Heterosexual)
Gay	1.95	1.18	–	3.21	–
Bisexual	1.68	1.03	–	2.73
Prefer not to answer	1.49	1.02	–	2.19
Military factors
Service branch (ref: Air Force)
Army	1.40	1.19	–	1.63	1.15	0.98	–	1.34
Navy/Coast Guard	1.16	0.96	–	1.42	0.94	0.78	–	1.12
Marine Corps	1.61	1.26	–	2.05	1.03	0.81	–	1.31
Pay grade (ref: Officer)
Enlisted	1.35	1.12	–	1.64	1.25	1.03	–	1.52
Service component (ref: Reserve/Guard)
Active duty	1.25	1.09	–	1.43	–
Deployment experience 2001–2009^e (ref: Not deployed)
Deployed, no combat	0.95	0.80	–	1.12	1.04	0.89	–	1.21
Deployed with combat	1.11	0.97	–	1.28	0.99	0.86	–	1.14
Psychological health 2001–2009
PTSD/depression 2001–2009 ^f,^g (ref: Neither)
Depression	1.78	1.20	–	2.63	1.38	0.94	–	2.04
PTSD	1.83	1.44	–	2.33	1.14	0.85	–	1.52
PTSD and depression	1.76	1.31	–	2.38	1.21	0.83	–	1.75
Life experiences
Childhood trauma^h (ref: No)
Yes	1.51	1.34	–	1.71	–
Disabling injury (ref: No)
Yes	1.72	1.42	–	2.09	1.53	1.24	–	1.90
Physical health
Sexual health difficulties^b (Time 1; ref: No)
Yes	6.80	5.70	–	8.11	–
Medical diagnoses^i (ref: None)
1	1.45	1.26	–	1.68	1.46	1.26	–	1.68
2–3	1.71	1.32	–	2.21	1.96	1.49	–	2.59
Body mass index (ref: Normal [<25])
Overweight (25–29.9)	1.36	1.15	–	1.60	1.25	1.08	–	1.46
Obese (≥30)	1.77	1.46	–	2.15	1.28	1.06	–	1.56

^aAll were assessed at Time 1 except for sexual orientation and childhood trauma, which were assessed at Time 2.

^bSelf-reported sexual health difficulties was indicated if “bothered a lot” was endorsed to, “During the last 4 weeks, how much have you been bothered by any of the following problems? Little or no sexual desire or pleasure during sex? Pain or problems during sexual intercourse?”

^cAdjusted odds ratio (AOR) adjusted for recent deployment experience, sexual assault, and posttraumatic stress disorder (PTSD).

^dAdjusted hazards ratio (AHR) adjusted for recent deployment experience, sexual assault, and PTSD. The survival model did not adjust for sexual orientation, service component, childhood trauma, and sexual health at Time 1.

^eDeployment, if deployment date was between 2001–2009, was determined using administrative date information; combat was assessed by 5 combat-related items on surveys between 2001 and 2009.

^fPosttraumatic stress disorder (PTSD) was assessed using the PTSD Checklist–Civilian Version.

^gDepression was assessed using the Patient Health Questionnaire-8.

^hChildhood trauma was assessed using 4 items from the Juvenile Victimization Questionnaire that ask about neglect, sexual abuse, verbal abuse, and physical abuse before age 18.

ⁱMedical diagnoses assessments included sleep apnea, hypertension, or diabetes in the previous 3 years.

Discussion

In this large prospective study of service men, experiencing high levels of combat during deployment was directly associated with subsequent sexual health difficulties. In addition, after adjustment for covariates, probable PTSD mediated a significant proportion of the impact of both high combat experience and sexual assault on sexual health problems. Other risk factors for sexual health difficulties identified in this study included older age, lower educational level, enlisted rank, disabling injury, certain medical conditions, and higher BMI.

Our results suggest that those who endorsed more types of combat experiences during recent deployment were more likely to report sexual health difficulties. We found a direct association between high combat and self-reported sexual health difficulties, including sexual desire and pain, but not with a new-onset sexual dysfunction. Similarly, previous results related to the direct association between combat and sexual health problems have been inconsistent (Badour et al., 2015; Bhalla et al., 2018; Suvak et al., 2012). The difference we found may be related to timing and help-seeking behavior. Individuals may report sexual health difficulties well before they receive a medical diagnosis; it takes time to decide to seek medical attention and obtain care; hence a post-deployment sexual dysfunction diagnosis may be delayed. Alternatively, the disparity may be related to the types of sexual health problems being assessed. The subjective sexual health measure asked more generally about sexual desire and pain in the past four weeks; new-onset sexual dysfunction diagnoses, on the other hand, represent specific disorders such as psychosexual dysfunction with inhibited sexual excitement and ED (which represented more than 68% of all cases). Future research using assessments that measure specific sexual health-related problems, and how they may be differentially associated with military risk factors, are warranted.

Indirect associations between high combat and sexual health problems, mediated through probable PTSD, were observed for both sexual health outcomes and supports the model proposed by Rosebrock and Carroll (2016). This expands upon previous findings in which PTSD was found to mediate the association of military stressors, such as warfare exposure, with family/relationship outcomes (Creech et al., 2016; B. N. Smith et al., 2017). Taken together, these results suggest that combat exposure increases the risk of PTSD, which, in turn, is associated with higher odds for relationship impairment and sexual health problems. This is important as healthy intimate relationships can act as a buffer against stress and depression (Hennessy et al., 2009).

Our findings suggest that the relation between lifetime sexual assault and sexual health is less consistent than the relation between recent combat experience and sexual health. Only the indirect effect of sexual assault with new-onset sexual dysfunction through probable PTSD was significant. These mixed results were unexpected and less aligned with prior studies that have identified an association of sexual assault and harassment during military service with sexual dysfunction among male service members (O’Brien et al., 2008; Suvak et al., 2012; Turchik et al., 2012). However, previous studies did not test for mediation, so it is not possible to know whether PTSD mediated the relationship between sexual trauma and sexual dysfunction in these prior samples. Furthermore, in the current study the self-reported sexual health outcome was based on sexual pain and lack of sexual desire, which is quite different from ED and other sexual health diagnoses. While there are no similar mediation studies to our knowledge, previous studies have identified associations between PTSD and adverse sexual health among male service members (Breyer et al., 2016; Spivak et al., 2003). Furthermore, sexual assault is known to be under-reported among service men likely due to fear, shame, and self-blame (Valente & Wight, 2007). In addition, acknowledging or reporting sexual assault may be stigmatizing and embarrassing for service men considering that military culture tends to promote heteronormativity and masculinity (O’Brien et al., 2015). It is possible that the under-reporting of lifetime sexual assault among service men may have impacted our ability to detect significant associations in these analyses.

Our most consistent result was the indirect effect of both stressors on sexual health mediated by probable PTSD. Although our results cannot delineate the specific ways PTSD may lead to sexual dysfunction, there are a number of possible mechanisms. Considerable research indicates that emotional numbing, one of the PTSD symptom clusters, has a negative impact on intimate relationship quality (Campbell & Renshaw, 2018), and it has been specifically associated with sexual health issues among service men (Badour et al., 2015; Bhalla et al., 2018; Nunnink et al., 2010). Alternatively, certain human hormones and neurotransmitters are involved with the different stages of sexual behavior (Bancroft, 1984; Iovino et al., 2019), and PTSD affects specific neuroendocrine markers (e.g., ambient cortisol and plasma dehydroepiandrosterone). It is plausible that these neuroendocrine changes may negatively impact sexual health (Lehrner et al., 2016). Additionally, the only class of medications that the U.S. Food and Drug Administration has approved to treat PTSD, selective serotonin reuptake inhibitors, are known to impair sexual health as a result of increases in serotonin (Breyer et al., 2014; Yehuda et al., 2015). Future research is needed to better understand the potential mechanisms that may explain how PTSD acts as a mediating factor between these stressors and sexual dysfunction.

In the final adjusted model, some military characteristics (i.e., Army/Marine Corps service members, enlisted pay grade, and active duty service component) were significantly associated with sexual health difficulties. Certain risk factors for sexual health difficulties (Institute of Medicine, 2011; Nnamani et al., 2019; Ponsford, 2003; Taylan et al., 2019), such as traumatic brain injury, spinal cord injury, and exposure to environmental toxins may be more common among these sub-groups of military personnel (Blair et al., 2012; Chin & Zeber, 2019; Schoenfeld et al., 2013; B. Smith et al., 2012). These factors may partially explain the higher odds of sexual health difficulties among these groups. Although we were unable to assess these specific risk factors, they should be examined in future research.

Consistent with previous research (Bigras et al., 2017; Breyer et al., 2014; Hosain et al., 2013; Kauth et al., 2014; Wilcox et al., 2014), this study identified factors such as older age, certain medial conditions, African American race, lower education level, childhood trauma, sexual minority status, and history of mental disorders as significantly associated with sexual health problems. The consistency of our findings supports the validity of our study and calls for future research to further examine these relationships among military populations. Although each of these groups make up a small proportion of the entire military, some factors, such as experiences of childhood trauma (Katon et al., 2015), are more prevalent among service members than civilians. Additionally, health disparities by sexual orientation and race/ethnicity have been understudied in the military. Gay and bisexual service men may have felt that they needed to conceal their sexual orientation, and both racial and sexual minorities may have experienced harassment, discrimination, and victimization because of their minority status (Burks, 2011; Moradi, 2009). These stressors may increase the risk for mental health problems (Cochran et al., 2013), which could lead to sexual health problems. The fact that certain minority groups in this all-military population were at a higher risk of sexual health difficulties suggests that differential risk goes beyond access to health care. Future research should further explore why these disparities may exist among male service members.

This research had many strengths. The Millennium Cohort is the largest cohort of military service members and veterans representing all service branches, components, and ranks. The longitudinal design of the study allowed for sexual health to be assessed approximately four years later. In addition to self-reported data, we were able to examine medical records for active duty personnel. This study also had limitations. First, self-reported sexual health was not assessed with a standardized measure, which limited the comparability of our results to prior research. In addition, self-reported sexual health difficulties experienced been underreported because of the sensitive nature of the questions or individuals experienced symptoms not assessed on the survey, such as issues related to orgasm, ease of arousal, and erections. Second, although the availability of objective medical diagnoses as a secondary outcome was advantageous, archived military medical data have limitations as well. Not all physicians may be adequately trained or comfortable in assessing and diagnosing sexual dysfunction, and patients may be unwilling to disclose these problems as well (Tsimtsiou et al., 2006). In addition, there is evidence sexual dysfunction diagnoses are not consistently documented in military health medical records (Helmer et al., 2013). Also, sexual assault represented lifetime experience and could not be classified as recent or military-related; service men may be less likely to report sexual assaults that occurred during military service. Participants were only categorized as experiencing sexual assault if they labeled their experience as such, which likely leads to lower endorsement (Ilies et al., 2003). Additionally, this study focused on the mediating role of PTSD for sexual dysfunction, but there are other possible mediators (e.g., alcohol misuse, tobacco use, social support) that should be examined in future research. Some of these are frequently comorbid with PTSD (Gros et al., 2016), making it challenging to parse confounding effects. Finally, probable PTSD was assessed using the PCL-C, which is a standardized self-report measure and does not specifically require the identification of a Criterion A event. While the PCL-C is a screening tool for and not diagnostic of PTSD, it has strong demonstrated validity (Wilkins et al., 2011).

Findings from this study have important implications for the well-being of active duty male service members, and, by extension, the health of the military as a whole. Currently, TRICARE/UnitedHealthcare, the medical health coverage for active duty service members, dependents, and retired personnel, does not cover sexual dysfunction therapy, although limited medication can be prescribed for those with ED of an organic origin (TRICARE, 2019). The effect of military experiences may persist after leaving service. The VA estimates that approximately 60% of OEF, OIF, and OND veterans utilize VA health care (Epidemiology Program, 2013) where there is a limit on the number of doses of ED medications that can be prescribed monthly (VA Pharmacy Benefits Management Services, 2014). A recent study of 297 veterans found that although many VA patients were not satisfied with their ED treatment, most (80%) planned to continue with treatment, possibly highlighting a lack of alternative treatments (Sussman et al., 2016). Private medical insurers in the U.S. do not consistently publicly publish coverage of advanced ED treatments, and when policies are publicly available insurers can place restrictions on covered treatments (Le et al., 2017). Civilian providers should ask patients about military service and experiences to be able to fully understand sexual health risks. Given the high prevalence of self-reported sexual health problems along with the strong and consistent relationship of military-related experiences with sexual health, treatment for sexual dysfunction should be expanded. In particular, insurance coverage should be extended to include a broader range of sexual dysfunction diagnoses, such as those that are determined to be psychogenic in nature. Moreover, sexual health problems were strongly linked to PTSD in this study. Reduction of PTSD symptoms has been shown to decrease sexual health problems (Schnurr et al., 2009), suggesting successful treatment of PTSD would improve long-term sexual health outcomes in service members and veterans.

In summary, long-term sexual health can be negatively affected by combat experiences and sexual assault among military service men. Our findings suggest that PTSD mediates the relationship between such stressors and subsequent sexual health. For some service men, sexual health problems appear to be linked to combat experience during military service, thus compelling more comprehensive treatment options, especially for psychogenic sexual health conditions. Clinicians can refer to the identified risk factors from this study when assessing patients, and they should be aware of the long-term adverse effects of these stressors and the significant role PTSD can have on sexual health problems. Effective treatment of PTSD may reduce the risk of future sexual health problems among service men.

Disclaimer

I am a military service member or employee of the U.S. Government. This work was prepared as part of my official duties. Title 17, U.S.C. §105 provides that copyright protection under this title is not available for any work of the U.S. Government. Title 17, U.S.C. §101 defines a U.S. Government work as work prepared by a military service member or employee of the U.S. Government as part of that person’s official duties. Report Number 20-88 was supported by the Military Operational Medicine Research Program, Defense Health Program, and Veterans Affairs under work unit no. 60,002. The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, nor the U.S. Government. The study protocol was approved by the Naval Health Research Center Institutional Review Board in compliance with all applicable Federal regulations governing the protection of human subjects. Research data were derived from an approved Naval Health Research Center Institutional Review Board protocol, number NHRC.2000.0007.

Declaration Of Conflicting Interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Acknowledgments

In addition to the authors, the Millennium Cohort Study Team includes Satbir Boparai, MBA; Felicia Carey, PhD; Toni Rose Geronimo, MPH; Isabel Jacobson, MPH; Rayna Matsuno, PhD; Deanne Millard; Chiping Nieh, PhD; Teresa Powell, MS; Ben Porter, PhD; Anna Rivera, MPH; Rudolph Rull, PhD; Beverly Sheppard; Daniel Trone, PhD; Jennifer Walstrom; and Steven Warner, MPH. The authors also appreciate contributions from the Deployment Health Research Department, Millennium Cohort Family Study Team, Birth and Infant Health Research Team, and Henry M. Jackson Foundation. We greatly appreciate the contributions of the Millennium Cohort Study participants.

Data Availability Statement

The datasets analyzed during the current study are not publicly available due to institutional regulations protecting service member survey responses but are available from the corresponding author on reasonable request (may require data use agreements to be developed). Additional information available at https://www.millenniumcohort.org/research/collaboration.

Additional information

Funding

The Millennium Cohort Study is funded through the Defense Health Program, U.S. Department of Veterans Affairs Office of Research and Development, and the U.S. Department of Veterans Affairs Office of Patient Care Services under work unit no. 60002. The funding agency had no part in the study design, collection of the data, analysis of the data, or writing of manuscript. No financial disclosures were reported by the authors of this paper.