Trauma exposure correlates among patients receiving care in federally qualified health centers

Abstract

Over 80% of adults in the general population experience trauma. Rates of patients with posttraumatic stress disorder (PTSD) are high in primary care settings and are likely to be even higher in federally qualified health centers (FQHCs). Trauma exposure has been linked to psychiatric symptoms and physical health comorbidities, though little research has focused on FQHC patients. This study addresses this by examining clinical and sociodemographic correlates of specific trauma types among FQHC patients. We analyzed secondary data from patients who screened positive for PTSD and were receiving health care in FQHCs in a clinical trial (N = 978). Individuals who did versus did not experience a specific trauma type were compared using between-group tests. In the sample, 91.3% of participants were exposed to a DSM-5 Criterion A traumatic event, with 79.6% experiencing two or more trauma types. Witnessing a life-threatening event (57.3%) and physical assault (55.7%) were the most common traumatic experiences. Physical health comorbidities and worse physical health functioning were associated with a higher likelihood of exposure to all trauma types, with effect sizes larger than PTSD, ds = 0.78–1.35. Depressive and anxiety symptoms were also associated with a higher likelihood of experiencing nearly all trauma types to a lesser magnitude. People of color, OR = 2.45, and individuals experiencing financial inequities, OR = 1.73, had higher odds of experiencing serious accidents as well as other trauma types. The findings highlight the need for trauma-informed care, including routine trauma and PTSD screening, for FQHC patients.

Trauma exposure is prevalent in the general population, with over 80% of individuals worldwide reporting trauma exposure that meets posttraumatic stress disorder (PTSD) Criterion A per the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association [APA], 2013; Benjet et al., 2016). Individuals with more trauma exposure are more likely to present for physical health concerns than mental health complaints (Center for Substance Abuse Treatment, 2014). This may be particularly important for patients served by federally qualified health centers (FQHCs), which are safety-net primary care clinics.

In one study, 65% of female FQHC patients in the sample reported experiencing interpersonal trauma (Lathan et al., 2021). A study in which almost 5,000 FQHC patients were screened for PTSD demonstrated that 30% screened positive and 21% met the PTSD diagnostic criteria (Meredith et al., 2014). This PTSD prevalence rate is 4 times higher than the rate found in the general population. However, limited research has focused on the prevalence of exposure to specific trauma types among FQHC patients, and no study to our knowledge has examined clinical and demographic correlates of trauma exposure among FQHC patients. Meredith et al. (2014) suggest that knowing which type(s) of trauma an individual has experienced would allow providers to offer contextualized care that addresses both trauma type (e.g., interpersonal) and the associated clinical presentation.

Trauma exposure is associated with several mental health conditions, including PTSD, depression, anxiety, and substance use disorders (Degenhart et al., 2022; Forman-Hoffman et al., 2016). Previous research has also found that individuals exposed to one or more traumatic events in their lifetime are more likely to have specific medical conditions (e.g., asthma, hypertension, sinusitis) than those without trauma exposure. Further, Rosenberg et al. (2000) found a positive correlation between the number of traumatic events to which an individual had been exposed and higher medical utilization but did not find an association between PTSD diagnosis and medical utilization (Rosenberg et al., 2000).

FQHCs are federally funded, community-based nonprofit health centers that provide comprehensive services to medically underserved areas and populations (Health Resources and Services Administration, 2022). FQHCs play a unique role in the health care system and are in an ideal position to prevent, detect, and treat mental and physical health concerns associated with trauma exposure (Lathan et al., 2021). Because FQHCs serve populations disproportionally affected by trauma (Nath et al., 2016), quantifying the types of trauma exposure FQHC patients experience and identifying correlates for trauma types may help providers better identify and treat trauma-exposed individuals; however, limited research exists on trauma exposure characteristics among FQHC patients.

Given the scarcity of research examining specific types of trauma exposure and correlates in FQHCs, this study aimed to fill this gap in the literature. The purpose of this study was to describe types of trauma exposure among patients seen in an FQHC who screened positive for PTSD. A second goal was to examine the clinical and sociodemographic correlates of exposure to specific trauma types among FQHC patients. Understanding correlates of trauma types is clinically relevant, as it may increase the identification of patients who have experienced trauma and allow for the creation of contextualized care plans that are unique to the person seeking care.

METHOD

Participants

Participants (N = 1,004) were patients enrolled in a pragmatic comparative effectiveness trial of two primary care–based approaches for delivering evidence-based treatment to individuals with PTSD and bipolar disorder in 24 clinics representing 12 FQHCs located in Arkansas, Michigan, and Washington (see Fortney et al., 2020, 2021). Individuals were eligible if they screened positive for either PTSD, assessed using the six-item PTSD Checklist [PCL-6; Lang & Stein, 2005]) or bipolar disorder. In this secondary data analysis, we used baseline survey data from participants who screened positive for PTSD (n = 978). Surveys were conducted via the internet (n = 516) or phone (n = 462). All procedures and materials for the clinical trial were approved by the Institutional Review Boards of the University of Arkansas for Medical Sciences, University of Michigan, and University of Washington. The trial was designed and conducted in close collaboration with consumer and policy advisory boards. Written informed consent was obtained for all study participants in the trial.

Measures

Demographic characteristics

Sociodemographic information was obtained from a self-report survey. We asked participants for their gender, racial and ethnic identities, age, sexual orientation, partner status, veteran status, socioeconomic status, location, and employment status.

Trauma exposure

The Brief Trauma Questionnaire (BTQ; Schnurr et al., 1999) is a self-report survey that is used to assess exposure to specific types of traumatic events, consistent with the definition of PTSD Criterion A in the DSM-5 (APA, 2013). Following the question, “Has this ever happened to you?,” participants are asked to indicate (“yes” or “no”) whether they have been exposed to 10 different traumatic events (i.e., war zone exposure, serious accident, disaster, life-threatening illness, physical assault, childhood physical abuse, sexual assault, other life-threatening event, violent death of loved one, and witnessing life-threatening event). If an item is endorsed, certain trauma types have up to two “yes” or “no” follow-up questions: “Did you think your life was in danger or you might be seriously injured?” and “Were you seriously injured?” Previous research has demonstrated evidence for criterion validity for the Brief Trauma Interview, from which the BTQ is derived, with the number of trauma types associated with PTSD severity (Schnurr et al., 2002).

Alcohol use severity

The Alcohol Use Disorders Identification Test (AUDIT; Babor et al., 2001) is a 10-item measure that is used to assess typical alcohol use frequency and quantity, the frequency of binge drinking, and symptoms of alcohol use disorder. AUDIT items have varying ordinal response options ranging from 0 to 4 for Items 1–8, and Items 9 and 10 have a three-option response scale. Previous AUDIT psychometric studies have demonstrated evidence of good internal consistency (Cronbach’s α M = .80; Meneses-Gaya et al., 2009). In the present sample, Cronbach’s alpha was .91, and McDonald’s omega was .90.

Drug-related consequences

The 10-item Drug Abuse Screening Test (DAST-10; Skinner, 1982) was used to assess drug use and its associated negative consequences (e.g., blackouts, interpersonal issues). Item response options are binary (0 = “no,” 1 = “yes”), with higher scores indicating more drug-related consequences. DAST-10 scores have demonstrated evidence of adequate concurrent validity, test–retest reliability (r = .71), and internal consistency reliability (Cronbach’s α = .86 −.84; Yudko et al., 2007. In the present sample, Cronbach’s alpha was .73, and McDonald’s omega was .79.

Depressive symptom severity

The Hopkins Symptoms Checklist Depression Scale-20 item version (HSCL-20; Derogatis et al., 1974) is a 20-item measure used to assess symptoms of major depressive disorder. The HSCL-20 is composed of the 13-item HSC Depression Scale and seven additional items from the larger HSCL-90 Revised scale. Item response options range from 0 (not at all) to 4 (extremely), and scores are averaged to create a mean total score. Higher scores indicate more severe depressive symptoms. The HSCL-20 has demonstrated evidence of convergent validity and good internal consistency reliability (Cronbach’s α = .85; Johns et al., 2013. In the present sample, Cronbach’s alpha was .88, and McDonald’s omega was .88.

PTSD symptom severity

PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2013) is a 20-item measure that is used to assess DSM-5 PTSD symptom severity, with response options ranging from 0 (not at all) to 4 (extremely). Scores are summed, with higher scores indicative of higher levels of symptom severity. The PCL-5 has demonstrated evidence of convergent and discriminant validity, test–retest reliability (r = .82), and internal consistency reliability (Cronbach’s α = .94; Blevins et al., 2015). In the present sample, Cronbach’s alpha was .93, and McDonald’s omega was .93.

Anxiety symptom severity

The 7-item Generalized Anxiety Disorder scale (GAD-7; Spitzer et al., 2006). was used to assess symptoms of generalized anxiety disorder. Respondents were asked to rate items on a scale of 0 (not at all) to 3 (nearly every day). GAD-7 scores have demonstrated evidence of convergent and criterion validity, good test–retest reliability (intraclass correlation = .83), and internal consistency (Cronbach’s α = .92; Spitzer et al., 2006). In the present sample, Cronbach’s alpha was .87, and McDonald’s omega was .87.

Physical health comorbidities

The Depression Outcomes Module comorbidity checklist (DOM; Smith et al., 2000), which lists 20 different physical health comorbidities (e.g., stroke, tuberculosis, asthma, cancer, migraine headaches), was used to assess comorbid physical problems. Response options are binary (0 = “no”, 1 = “yes”), and the number of endorsed comorbidities is summed, with higher counts indicative of more physical health comorbidities. The DOM has demonstrated evidence of content, construct, and face validity.

Physical health functioning

The Veterans 12-Item Short-Form Healthy Survey (VR-12) is a subset of the Veterans 36-Item Short-Form Health Survey (SF-36; Jones et al., 2001). The VR-12 consists of five different response options, including binary (“yes” or “no”), trichotomized, and 5-to-6–point Likert-type scales ranging from 0 (none of the time) to 5 (all of the time). Scores from these 12 items can be used to derive a Physical Component Summary (PCS) score. We used the PCS t test t-score to quantify physical health function. PCS scores have demonstrated evidence of content, construct, and face validity, as well as internal consistency reliability (Cronbach’s α = .90; Selim et al., 2022). In the present sample, Cronbach’s alpha was .83, and McDonald’s omega was .82.

Data analysis

Coding and missing data

Trauma types were dichotomized (i.e., experienced = 1, not experienced = 0). The reference group consisted of participants who reported not experiencing any type of DSM-5 Criterion A trauma (n = 85). Correlates assessed include age, gender (woman = 1, man = 0), race and ethnicity (minoritized racial and ethnic identities = 1, White = 0), partner status (single = 1, partnered = 0), veteran status (veteran = 1, nonveteran = 0), financial status (below poverty threshold = 1, above poverty threshold = 0), rurality (rural = 1, urban = 0), employment status (not working [laid off, on strike, unemployed, disabled] = 1, working = 0), and sexual orientation (lesbian, gay, bisexual, or another sexual orientation [LGB+] = 1, straight = 0). Rurality was determined from zip codes using Rural–Urban Commuting Area (RUCA) codes. Supplementary Table S1 provides more nuanced examinations of gender, race, and sexual orientation.

For normally distributed continuous variables, differences in the proportions of participants who experienced a specific trauma type were tested using independent samples t tests, and effect sizes were calculated using Cohen’s d. Nonnormally distributed continuous variables (i.e., AUDIT, DAST-10) were tested using Wilcoxon two-sample tests, and Cohen’s d was calculated with the point biserial correlation. Categorical variables were dichotomized and evaluated using chi-square tests or Fisher’s exact tests and odds ratios (ORs) to determine effect size. Cohen’s d effect sizes were interpreted using conventional standards (i.e., 0.2 = small, 0.5 = medium, 0.8 = large; Cohen, 1988), and small, medium, and large odds ratios were defined as 1.5, 3.5, and 9.0, respectively.

To adjust for multiple tests, we applied the Benjamini–Hochberg (B-H; 2000) procedure with a false discovery rate (Q) of 10% to each trauma type analysis for the 16 primary and 13 supplementary statistical tests. We assessed the strength of the associations between trauma types using chi-square tests. Phi coefficients greater than .25 were interpreted as very strong, greater than .15 as strong, greater than .10 as moderate, greater than .05 as weak, and less than .05 as very weak or indicative of no association. We then conducted a series of multivariable regressions with trauma type predicting clinical outcome scores. Pairwise deletion was used for missing data. All statistical analyses were performed in SAS (Version 9.4).

Although common in research, we refrain from labeling correlates as “risk factors” and “protective factors.” Describing sociodemographic correlates as using these terms suggests that someone with a particular identity is at risk of trauma exposure rather than someone with a particular identity having a higher likelihood of trauma exposure due to systemic inequities. For example, racism—not race—is a risk factor for several negative outcomes, which cannot be attributed to genetic differences (Chokshi et al., 2022; Yudell et al., 2020). In addition, we did not assess trauma timing relative to the onset of clinical symptoms.

RESULTS

Sample description

The analytic sample (N = 978) consisted of 70.1% women and 42.2% people from minoritized racial/ethnic backgrounds, with a mean participant age of 39.3 (SD = 12.7). Most participants endorsed a DSM-5 Criterion A event (91.3%). See Table 1 for additional sociodemographic descriptions. The most common trauma types were witnessing a life-threatening event (57.3%), physical assault (55.7%), the violent death of a loved one (46.1%), and experiencing a serious accident (45.3%). Most participants (79.6%) reported experiencing two or more trauma types. Descriptive statistics and endorsement frequency for each trauma type, as well as number the number of trauma types endorsed, see Table 2.

TABLE 1.

Sample descriptive statistics

Sociodemographic characteristic	M	SD
Age (years)a	39.33	12.86
	n	%
Genderb
Identifies as a woman	683	70.1
Identifies as a man	275	28.2
Identifies as transgender, nonbinary	16	1.6
Race
African American/Black	116	12.0
American Indian/Alaska Native	36	3.7
Another race	70	7.2
Arab/Middle Eastern	2	0.2
Asian American/Pacific Islander	3	0.3
European American/White	679	70.1
Multiracial	62	6.4
Identifies ethnicity as Hispanic, Latino/a/e, or of Spanish origin	121	12.4
Sexual orientationc
Heterosexual/straight	459	80.8
Lesbian/gay	19	3.3
Bisexual	59	10.4
Another sexual orientation	31	5.5
Marital status
Single	642	65.7
Partnered	335	34.3
Veteran status: Yes, but not currently on active duty or in the reserves	52	5.3
Socioeconomic status below poverty threshold	606	65.7
RUCA categorization (location)
Urban	482	49.3
Rural	495	50.7
Work status
Employed, retired, or student	431	45.4
Laid off, on strike, unemployed, or disabled	519	54.6

Note: RUCA = Rural–Urban Commuting Area.

^an = 974.

^bIndividuals who identified as transgender or nonbinary were excluded from the primary gender comparison but included in the supplementary analyses.

^cSexual orientation data were collected at 12-month follow-up.

TABLE 2.

Clinical descriptive statistics

Variable	N	%	M	SD
Clinical characteristic
Alcohol use (AUDIT)	967		4.13	6.97
Drug use symptoms (DAST-10)	963		1.18	2.14
Depressive symptoms (HSCL-20)	977		2.44	0.70
PTSD symptoms (PCL-5)	971		47.95	17.68
Generalized anxiety symptoms (GAD-7)	968		14.79	5.32
Physical health comorbidities (DOM Comorbidity Checklist)	978		3.99	2.67
Physical functioning (VR-12 PCS)	978		42.15	13.36
Trauma type	973
DSM-5 Criterion A event	888	91.3
War zone exposure	70	7.2
Serious accident	443	45.3
Disaster	231	23.6
Life-threatening illness	134	13.7
Childhood physical abuse	392	40.1
Physical assault	545	55.7
Sexual assault	413	42.2
Other life-threatening event	207	21.2
Violent death of a loved one	451	46.1
Witnessing life-threatening event	560	57.3
Number of types of trauma experienced	973
0	85	8.7
1	114	11.7
2	156	16.0
3	158	16.2
4	134	13.8
5	124	12.7
6	100	10.3
7	49	5.0
8	34	3.5
9	15	1.5
10	4	0.4

Note: AUDIT = Alcohol Use Disorder Identification Test; DAST-10 = Drug Abuse Screening Test–10; PCL-5 = Posttraumatic Stress Disorder Checklist for DSM-5; GAD-7 = Generalized Anxiety Disorder–7; DOM = Depression Outcome Module Comorbidity Checklist; VR-12 PCS = Veterans RAND 12-Item Health Survey Physical Component Summary score; DSM-5 = Diagnostic and Statistical Manual of Mental Disorders (5th ed.).

Clinical correlates

PTSD symptom severity, physical health comorbidities, and poor physical health functioning exhibited medium-to-large effects across all trauma types assessed, ds = 0.61–1.35. More severe depressive and anxiety symptoms were associated with a higher likelihood of experiencing all trauma types except life-threatening illness and war zone exposure, with small-to-medium effect sizes, ds = 0.26–0.41. Consequences associated with drug use demonstrated a small association with experiencing childhood physical abuse, d = 0.18; another (i.e., nonspecified) life-threatening event, d = 0.28; and witnessing a life-threatening event, d = 0.16. For all effect sizes, see Table 3.

TABLE 3.

Trauma exposure type correlates

Correlate	War zone exposure d	Serious accident d	Disaster d	Life- threatening illness d	CH physical abuse d	Physical assault d	Sexual assault d	Other life- threatening event d	Violent death of loved one d	Witness life- threatening event d
Alcohol (AUDIT)	0.29	0.08	0.03	0.08	0.01	0.07	0.00	0.12	0.05	0.05
Drugs (DAST-10)	0.27	0.18	0.13	0.15	0.13	0.18*	0.15	0.28*	0.15	0.16*
Depression (HSCL-20)	0.35*	0.28*	0.32*	0.29a	0.34*	0.29*	0.41*	0.36*	0.29*	0.32*
PTSD (PCL-5)	1.03***	0.67***	0.86***	0.61***	0.89***	0.81***	0.99***	0.90***	0.70***	0.77***
Anxiety (GAD-7)	0.29	0.28*	0.26*	0.29a	0.34**	0.28*	0.39**	0.39**	0.30*	0.30*
Physical health comorbidities (DOM)	0.97***	0.84***	1.00***	1.21***	0.83***	0.78***	0.85***	1.04***	0.83***	0.79***
Physical health functioning (VR-12 PCS)	1.35***	0.92***	1.16***	1.24***	0.82***	0.90***	0.91***	1.10***	0.88***	0.82***
Age	0.52**	0.52***	0.73***	0.82***	0.46***	0.47***	0.42***	0.60***	0.44***	0.39***
	OR	OR	OR	OR	OR	OR	OR	OR	OR	OR
Gender (identifies as a woman)	0.27***	0.63	0.86	0.92	0.86	0.92	3.15	0.88	0.83	0.73
Race and ethnicity (minoritized identity)	2.43**	2.45**	1.39	1.12	1.42	1.40	1.42	1.51	1.47	1.34
Marital status (single)	1.23	1.55	1.85*	1.22	1.45	1.75*	1.52	1.65	1.80*	1.34
Veteran status (veteran)	29.08***	5.32	5.42	3.26	6.38	4.50	3.13	4.66	5.08	6.38*
Financial status (poverty)	1.88	1.73*	1.68	1.80	1.58	1.67*	1.91*	1.92*	1.72*	1.57
Location (rurality)	0.57	0.58*	0.54*	0.59	0.58*	0.50**	0.51**	0.60	0.52**	0.61*
Employment (unemployment)	1.01	1.40	1.30	2.18**	1.50	1.43	1.49	1.44	1.39	1.34
LGB+	0.40	0.68	0.47	0.84	0.88	0.72	0.84	0.54	0.73	0.71

Note. A Benjamini–Hochberg correction was applied for each set of analyses associated with a specific trauma type. AUDIT = Alcohol Use Disorder Identification Test; DAST-10 = Drug Abuse Screening Test −10; PCL-5 = Posttraumatic Stress Disorder Checklist for DSM-5; GAD-7 = Generalized Anxiety Disorder–7; DOM = Depression Outcome Module Comorbidity Checklist; VR-12 PCS = Veteran RAND 12-item Health Survey Physical Component Summary score; LGB+ = lesbian, gay, bisexual, or another minoritized sexual identity; OR = odds ratio; CH = childhood.

^aComparison no longer significant at p < .05 after applying a Benjamini–Hochberg correction.

^*p < .05.

^**p < .01.

^***p < .001

Sociodemographic correlates

Age was associated with a higher likelihood of trauma exposure across all trauma types, with medium-to-large effects, ds = 0.42–0.82. Compared to men, women had higher odds of experiencing sexual assault, with a medium effect, OR = 3.15, but lower odds of war zone exposure, OR = 0.27. Minoritized racial and ethnic identity was associated with higher odds of war zone exposure, OR = 2.43, and serious accidents, OR = 2.45. Being single was associated with higher odds of experiencing disaster, physical assault, and witnessing life-threatening events, with small effect sizes, ORs = 1.75–1.85. Living below the poverty threshold was associated with higher odds of physical assault and witnessing a life-threatening event, as well as experiencing a serious accident, sexual assault, and another (i.e., nonspecified) life-threatening event, with small effect sizes, ORs = 1.67–1.92. Veterans had higher odds of experiencing war zone exposure, OR = 29.08, and witnessing a life-threatening event, OR = 6.38, compared to civilians, with large and medium effects, respectively. Not working was associated with higher odds of experiencing a life-threatening illness compared to being employed, retired, or a student, OR = 2.18. Living in a rural area was associated with lower odds of experiencing all trauma types except war zone exposure, a life-threatening illness, and another (i.e., nonspecified) life-threatening event compared to living in an urban area, ORs = 0.50–0.61. Refer to Table 3 for effect sizes and Supplementary Table S1 for the results of nuanced sociodemographic comparisons.

Associations between trauma types

Among the trauma types analyzed, most had a strong-to–very strong strength of association. Physical assault had the strongest association with multiple other trauma types, including childhood physical abuse, sexual assault, another (i.e., nonspecified) life-threatening event, and witnessing a life-threatening event, φs = .27–.33. A strong association was also found between childhood physical abuse and sexual assault, φ = .32. See Supplementary Table S2 for associations between all trauma types.

Trauma type as a predictor of clinical scores

See Supplementary Table S3 for regression results. When all trauma types were entered into the model, the following were statistically significant for the PCL-5: sexual assault, childhood physical abuse, another (nonspecified) life-threatening event, witnessing a life-threatening event, and physical assault. Experiencing a serious accident, natural disaster, life-threatening illness, and/or the violent death of a loved one were significant in the models predicting physical comorbidities (DOM) and physical functioning (VR-12). Positive estimates for comorbidities indicate more physical problems; negative functioning estimates indicate lower physical functioning. Sexual assault was a statistically significant predictor of comorbidities but not functioning, and experiencing another (i.e., nonspecified) life-threatening event was a significant predictor of functioning but not comorbidities. Only sexual assault was statistically significant in models predicting GAD-7 and HSCL-20 scores. Physical assault and witnessing life-threatening trauma were statistically significant predictors of DAST-10 scores, and only war exposure predicted AUDIT scores. Standardized beta coefficients from statistically significant trauma types ranged from .07 to .19.

DISCUSSION

The purpose of this study was to examine the prevalence of specific types of trauma exposure as well as the clinical and demographic correlates of trauma exposure. In this sample of FQHC patients who screened positive for PTSD, 91.3% of participants reported experiencing a Criterion A traumatic event, with the most common trauma types reported being witnessing a life-threatening event and physical assault. Aside from age, physical comorbidities and poorer physical health functioning were the most consistent correlates across trauma types. The odds of experiencing specific types of trauma exposure were higher among individuals with minoritized racial or ethnic identities, those facing financial inequity, and those in urban areas. The strength of most trauma type phi coefficients was strong to very strong, and experiencing multiple trauma types was the rule (i.e., approximately 80% of the sample reported experiencing two or more trauma types) rather than the exception (i.e., roughly 12% reported experiencing only one trauma type), with approximately 33% of sample reporting having experienced five or more trauma types. Ensuring that FQHCs are equipped to identify and treat patients who experience trauma is imperative.

Physical health comorbidities and poor physical health functioning were strongly associated across trauma types. Physical outcomes, including somatization, may be particularly relevant for certain sociodemographic groups that are more likely to receive health care at FQHCs. In some cultures, mental health conditions can present as somatic problems (Ryder et al., 2008). Further, Labash and Swartz (2021) found that Latina women were more likely to present with physical health complaints than mental health concerns in primary care settings, and Latina women who immigrated with two or more medical conditions had higher odds of trauma exposure than those with fewer medical conditions. Therefore, primary care providers are in an ideal position to recognize the enduring impact of trauma exposure on physical health.

The most consistent psychiatric correlate associated with all types of trauma exposure was PTSD, which is consistent with previous research demonstrating higher odds of experiencing PTSD among people who experienced war zone exposure (Magruder et al., 2005), interpersonal trauma (Kessler et al., 2017), and assault (Glover et al., 2010). We also found that interpersonal trauma (i.e., sexual assault, childhood physical abuse, and physical abuse) was associated with higher PCL-5 scores among all trauma types, which is consistent with previous work (Kessler et al., 2017). Depressive symptoms were associated with all trauma types except experiencing a life-threatening illness, and generalized anxiety was associated with all trauma types except experiencing a life-threatening illness and war zone exposure. Given the high comorbidity between PTSD, depression, and anxiety (Kessler et al., 2005), these findings are not surprising. However, we did not find consistent evidence for the well-established link between trauma exposure and drug-related consequence severity (Degenhardt et al., 2022; Schimmenti et al., 2023), and no trauma types were associated with hazardous alcohol use. All trauma types accounted for only 1% and 2% of the variance in alcohol and drug outcomes, respectively. More research is needed to better understand the associations between trauma and drug-related outcomes among FQHC patients.

Regarding physical comorbidities and functioning, trauma types that can lead to injury and/or illness (e.g., serious accidents, life-threatening illness, and natural disasters) were associated with worse outcomes. Unexpectedly, the violent death of a loved one was also a significant predictor of these physical outcomes. Although previous studies have reported an increased likelihood of psychiatric conditions following the unexpected death of a loved one (Keyes et al., 2014), the present study is the first study of which we are aware to examine the association between this trauma type and physical problems.

Regarding sociodemographic correlates of trauma exposure, age was significantly associated with experiencing all trauma types, which is likely due to more time for potential trauma exposure. Prior research has also found that individuals who are married are less likely to experience trauma (Benjet et al., 2016), and our work extends these findings by providing evidence that FQHC patients who are single are more likely to experience disasters, report physical assault, and witness a life-threatening event. The results indicate that women had a higher likelihood of experiencing sexual assault than men, consistent with previous research (Benjet et al., 2016). Tolin and Foa (2008) suggest that sexual violence is a particularly “toxic” kind of trauma exposure that is strongly associated with developing PTSD and other health concerns. In the present study, sexual assault was the only statistically significant trauma type associated with depressive and anxiety symptom scores, though it should be noted that the effect sizes were small. Given this consistent finding in the literature, FQHCs could screen for sexual trauma and offer patients information about local and national resources (e.g., the Rape, Abuse, and Incest National Network [RAINN]).

Within our sample, participants with a minoritized racial or ethnic identity had higher odds of experiencing specific trauma types, including serious accidents and war zone exposure. Our findings are consistent with another study with a sample of primarily Black women in primary care, which found accidents to be the most common type of trauma exposure within the sample (Gillespie et al., 2009). Furthermore, patients seeking asylum and those with refugee status, many of whom have experienced war-related trauma, often seek care at FQHCs (Bunn et al., 2023).

Individuals with incomes below the federal poverty line were more likely to experience several trauma types in this sample. Research has consistently shown that living below the poverty threshold results in a higher risk of trauma exposure (Gillespie et al., 2009). Our findings indicate that trauma exposure may be one mechanism that contributes to the “weathering” effect observed among people from communities that are marginalized and disadvantaged (Forde et al., 2019). The weathering hypothesis “encapsulates the ways in which social inequality may affect the health of population groups differentially and the ways in which these differences may be compounded with age” (Geronimus, 1992, p. 210). Previous work provides support the for the weathering hypothesis across numerous outcomes when comparing individuals from Black and White communities, as well as in comparisons of individuals with and without financial inequity (see Forde et al., 2019, for a meta-analysis).

Consistent with previous research (Lehavot et al., 2018), veterans within the sample had higher odds of experiencing war zone exposure and witnessing a life-threatening event than civilians. A national survey examining trauma exposure frequency found that the likelihood of experiencing war-related trauma was lower for individuals residing in rural settings compared to urban settings, whereas exposure to other trauma types did not differ based on the rural–urban continuum (McCall-Hosenfeld et al., 2014).

This study has several clinical implications. The findings shed light on the types of traumatic events FQHC patients experience and help clarify the correlates of these experiences. We found that among FQHC patients who screened positive for PTSD, over 90% reported experiencing a DSM-5 Criterion A event, and 82% had probable PTSD based on the PCL-5. Given the consistent associations between trauma exposure and physical comorbidities, poor physical functioning, depression, and anxiety, trauma-related disorders should be considered as a differential diagnosis during assessment when FQHC patients present with these conditions. Untreated PTSD is associated with an increased likelihood of psychiatric and physical comorbidities (Davidson, 2000), so it is essential to detect and treat PTSD as early as possible. Because an average of 12 years elapses between symptom onset and a diagnosis of PTSD (Wang et al., 2005), these findings could help increase recognition of trauma exposure correlates, leading to earlier PTSD diagnosis and intervention.

The strong associations between experiencing childhood physical abuse and sexual assault, as well as witnessing life-threatening events, suggest prevention and early intervention are needed in FHQC populations. Most patients experienced two or more trauma types, and one third experienced five or more. These numbers reflect the different types of trauma experienced rather than the number of traumatic events, which is likely much higher. Because research has demonstrated a higher likelihood of PTSD among people who experience trauma at an earlier age and those who have been exposed to multiple traumatic events (Koenen et al., 2002), early detection is imperative to improve patient outcomes.

Patients in primary care settings are infrequently screened for trauma exposure. However, findings from a primary care study demonstrated that implementing universal screening increased PTSD detection at a rate 5 times higher than when PTSD was identified through electronic health records (Murray‑Krezan et al., 2023). One recommended measure is the Primary Care PTSD Screen (PC-PTSD; Prins et al., 2003), which can be used to assess trauma exposure and PTSD symptoms and has demonstrated evidence of psychometric validity and clinical utility in primary care settings (Williamson et al., 2022). Although screening for PTSD in the primary care setting is recommended, more research is needed to identify optimal screening procedures (Raja et al., 2021). Screening should not be implemented in systems that cannot provide adequate follow-up, PTSD treatment, and/or referral to PTSD treatment (see Austin, 2021). There are several evidence-based treatments for PTSD, including cognitive processing therapy (Resick et al., 2017) and written exposure therapy (Sloan & Marx, 2019) that are feasible to deliver in primary care settings (e.g., Fortney et al., 2015). Further, a recent meta-analysis found psychological interventions for PTSD to be as effective for individuals with PSTD who experienced multiple traumatic events as those with PTSD who experienced one traumatic event (Hoppen et al., 2024).

Trauma-informed care (TIC) has emerged as an approach to address adverse childhood events, which are associated with poorer health (Harris & Fallot, 2001). TIC means adopting “universal precautions” in treating all patients as if they have experienced trauma (Racine et al., 2020). Because trauma exposure can make individuals sensitive to reminders of traumatic events, it can be a barrier to receiving preventive and routine medical care, disclosing personal information, and practicing health-promoting behaviors. TIC aims to instill a sense of safety in all health care interactions; build trustful treatment alliances; engage and empower patients; minimize the reactivation of traumatic experiences; promote resilience; and identify, diagnose, and treat trauma-related symptoms within a framework promoting equity, diversity, and inclusion (see Raja et al., 2015; Reeves, 2015; Roberts et al., 2019; Strahan, 2022).

When employing TIC among individuals with societally marginalized identities, it is imperative to consider the negative effects that both DSM-5 Criterion A events and more insidious forms of individual and intergenerational trauma stemming from systemic racism, discrimination, and economic inequities can have on physical and mental health (Brave Heart et al., 2020; Comas-Díaz et al., 2019). Although more research is needed to determine the impact of TIC on clinical outcomes, emerging evidence indicates that training staff in TIC improves knowledge and attitudes (Purtle, 2020).

Strengths of the current work include the use of data from a large sample of diverse patients from FQHCs across the United States. To our knowledge, this is the first study to provide a nuanced examination of trauma exposure and its correlates among FQHC patients. Limitations include limited generalizability beyond FQHC patients, a lack of data on the frequency of trauma exposure, an inability to infer causality due to cross-sectional data, the limited number of participants from specific minoritized backgrounds (e.g., people who identify as trans and nonbinary; n = 20), a lack of data on refugee or immigrant status, the dichotomization of race and ethnicity, and a lack of psychometric evaluations of the physical outcomes in primary care samples.

Future research directions include longitudinal examinations of trauma exposure and mental and physical health sequelae to establish the temporal sequence, as well as investigations of the impact of the frequency and combinations of trauma exposure on clinical and quality of life outcomes. Future research could examine whether people with specific intersectional identities are more likely to experience specific types of trauma to inform systems-level and community-specific prevention and intervention strategies. More work is also needed on the effectiveness of TIC.

In this sample of FQHC patients who screened positive for PTSD, 91.3% reported experiencing at least one DSM-5 Criterion A traumatic experience, and 79.6% reported experiencing two or more trauma types. Medical comorbidities, poorer physical functioning, depressive symptoms, anxiety symptoms, and age were associated with a higher likelihood of having experienced most trauma types, suggesting that FQHC patients who present with these problems should be screened for trauma exposure. More research is needed to identify the social and structural mechanisms responsible for the higher likelihood of trauma exposure among individuals with specific sociodemographic characteristics, including individuals with minoritized racial and ethnic identities and those experiencing financial inequities.