Skip to main content
NCBI home page
VA Author Manuscripts logoLink to VA Author Manuscripts
. Author manuscript; available in PMC: 2026 May 7.
Published in final edited form as: Biopsychosoc Sci Med. 2025 May 7;87(6):355–361. doi: 10.1097/PSY.0000000000001397

Posttraumatic Stress Disorder, Obesity, and Epigenetic Aging: A Replication Study in 1,828 Veterans

Kyle J Bourassa 1,2, Melanie E Garrett 3; VA Mid-Atlantic MIRECC Workgroup1, Allison E Ashley-Koch 3, Jean C Beckham 1,4, Nathan A Kimbrel 1,4,5
PMCID: PMC12590413  NIHMSID: NIHMS2120167  PMID: 40524549

Abstract

Objective:

Posttraumatic stress disorder (PTSD) is associated with poor health, and prior research suggests that accelerated epigenetic aging could help explain this association. A recent study found that veterans with both PTSD and obesity had greater risk for accelerated epigenetic aging compared to those with either PTSD or obesity individually, or neither condition. The objective of this study was to conduct a replication and extension of this prior work.

Methods:

This study included models approximating the recent study’s analytic approach in a sample of 1,828 post-9/11 veterans. Our extension also included additional aging measures (PC-GrimAge and DunedinPACE), a more diverse sample, additional covariates (chronological age, smoking), and use of continuous measures of PTSD, obesity, and accelerated aging.

Results:

In contrast with the original report, we did not find evidence that obesity moderated the association of PTSD and aging, indicating that veterans with both conditions had greater risk for accelerated aging. Although several significant interactions were observed, they were in the opposite direction of the original study findings (i.e., PTSD was more strongly associated with aging scores among veterans with less body mass). Our results instead demonstrated that PTSD was associated with accelerated aging across all continuously measured aging scores (0.08 ≤ all βs ≤ 0.10), and that obesity was associated with faster DunedinPACE aging scores (β = 0.36, 95% CI [0.28, 0.44]).

Conclusions:

Our findings provide additional evidence that PTSD and obesity may be useful targets for interventions aiming to slow aging and improve health.

Keywords: PTSD, epigenetic aging, obesity, body mass, veterans

Introduction

People who develop posttraumatic stress disorder (PTSD) following the experience of trauma are at greater risk of poor health (1–5). This includes increased incidence of chronic diseases spanning multiple organ systems, including the cardiovascular (2–6), metabolic (7–8), and nervous systems (9–10). The wide range of health consequences observed in people with PTSD suggests that similarly non-specific physiological mechanisms would be helpful in characterizing how PTSD and trauma might result in poor health and premature mortality (11).

Recent studies have highlighted the role that accelerated biological aging could play in linking adversity, trauma, and PTSD to poor health (11). Faster rates of aging are theorized to increase risk for a wide range of chronic diseases typically associated with advancing chronological age (12–16). The development of new epigenetic markers of aging derived from DNA methylation (DNAm; 17) has enabled tests of whether trauma and PTSD are associated with individual differences in epigenetic aging (11). For example, recent studies have found veterans with more lifetime trauma burden (18) or current PTSD (18–20) have faster biological aging, as assessed by DNAm. Notably, accelerated epigenetic aging predicts the onset of both major chronic disease and premature mortality (21).

Continuing this line of important work, a recent study of 1,135 male, European-American U.S. military veterans (22) from the National Health and Resilience in Veterans Study (NHRVS; 23) found that obesity moderated the association between posttraumatic stress disorder (PTSD) and epigenetic aging, such that veterans with comorbid obesity and PTSD had the greatest risk of accelerated aging (22). However, there were limitations in the previous study’s analytic approach that warrant further study, particularly the dichotomization of PTSD symptoms, body mass, and the aging outcome GrimAge (22).

Present Study

In the current study, we set out to extend findings from Fischer and colleagues (22) using data from 1,828 veterans enrolled in the Post-Deployment Mental Health Study (PDMH; 23), a diverse cohort of post-9/11 veterans previously used to investigate associations between PTSD and epigenetic aging (i.e., DunedinPACE aging scores; 18). We first tested models that best approximated the methodological and statistical approaches taken in the prior study. We then extended these results by adding additional epigenetic aging outcomes, including the continuous version of the second-generation clock used in the original study (GrimAge), a newer version of GrimAge (PC-GrimAge), and a third-generation measure (DunedinPACE). We also report associations for continuously measured PTSD symptoms and body mass, in addition to the PTSD diagnosis and obesity measures derived from these values in each study.

Methods

Participants

Participants were drawn from the Post-Deployment Mental Health Study (23), a multi-site study that enrolled U.S. military veterans from the post-9/11 service era between 2005 and 2016. The study included 1,828 veterans (416 women and 1,412 men) who had DNAm, PTSD, and body mass data (see Supplemental Figure 1) and was comprised of 871 non-Hispanic Black veterans and 957 non-Hispanic White veterans with an average age of 37.6 (SD, 10.1). The Durham, Richmond, Hampton, and Salisbury Veteran Affairs’ (VA) Institutional Review Boards approved the study, and all participants provided informed consent. Data from the PDMH are part of a VA data repository and are available to researchers who request and are approved for access. Medical record data from the VA Corporate Data Warehouse are available to researchers who request and are approved for access through the Office of Research and Development Data Access Request Tracker system. Code and materials are available on request from the corresponding author or PDMH data repository (for materials).

Measures

Epigenetic aging.

Epigenetic aging was assessed using standard algorithms for GrimAge (24), PC-GrimAge (24,25), and DunedinPACE (26,27). As previously described (18), DNA methylation (DNAm) data was generated from peripheral whole blood using the Infinium HumanMethylation450 or MethylationEPIC Beadchip (Illumina Inc., San Diego, CA). Internal replicates were checked for consistency using single nucleotide polymorphisms on each array. Quality control was performed using the minfi and ChAMP R packages (28–29). Probe quality control and data normalization were performed within each batch using the R package watermelon (30). Raw beta values were normalized using the dasen approach and batch and chip adjustments were completed using ComBat in the R package sva (31). Principle component (PC) adjusted GrimAge was included, as it shows better reliability by accounting for technical variation in the original GrimAge measure (25). GrimAge and PC-GrimAge were residualized for chronological age to derived epigenetic age acceleration values. Continuous aging measures were standardized to a mean of 0 and SD of 1 to allow for comparisons between different measures. To provide comparisons with the models used by Fischer and colleagues (22), we also derived their dichotomized Accelerated GrimAge measure; individuals with GrimAge values greater than 5 (when residualized for age) were categorized as having accelerated aging. In total, 202 (10.9%) veterans had Accelerated GrimAge, whereas the remainder (n = 1,626, 89.1%) did not.

Posttraumatic stress disorder (PTSD) diagnosis and PTSD symptoms.

PTSD was assessed using self-reported PTSD symptoms scores, which were confirmed using clinical interviews (32) and diagnostic codes from VA health records. Self-reported PTSD symptoms were assessed using the Davidson Trauma Scale, a 17-item self-report measure assessing PTSD symptoms (33). Veterans with scores of 35 or above—a reliable and valid clinical cutoff with specificity of 0.95 and sensitivity of 0.91 (34)—were coded as having PTSD if they also met criteria for a PTSD diagnosis using either clinical interview or health records. Clinical interviews used the Diagnostic Interview Schedule (32) according to then current versions of DSM-IV. PTSD diagnoses were also ascertained from health records using ICD-9 (309.81) and ICD-10 (F43.10–12) codes present in Veteran’s Affairs (VA) electronic health records when veterans enrolled in the PDMH. Health record diagnoses were generated from VA outpatient, inpatient, and purchased care data (21). In total, 770 (41.5%) veterans met criteria for PTSD.

Body mass and obesity.

Body mass was calculated using the standard formula utilizing height and weight, which were assessed at baseline using an average of the three most recent valid measurements within a 2-year lookback period prior to the PDMH baseline (21). For 190 veterans who did not have valid height and weight scores during the electronic health record lookback period, we used height and weight assessed during the PDMH baseline (a small subset of veterans had body mass directly assessed in the PDMH). Veterans were coded as having obesity if they had a body mass of 30 or more, and 843 (45.5%) veterans met these criteria.

Demographic covariates.

Participants self-reported their age, sex, race, ethnicity, years of education, and smoking. Sex, race, and ethnicity were confirmed using genetic data. Smoking was coded as never smoker (0), past smoker (1), and current smoker (2).

DNAm technical covariates.

DNAm technical covariates included a dichotomous variable denoting methylation chip and estimated white blood cell proportions for T lymphocytes (CD4+ and CD8+), B cells (CD19+), monocytes (CD14+), NK cells (CD56+) and neutrophils using the FlowSorted.Blood.450k and FlowSorted.Blood.EPIC packages (35).

Data Analysis

We first tested whether obesity moderated the association of PTSD and Accelerated GrimAge using an approach that best approximated the prior study’s methods. To do so, we first tested the main effects of PTSD and obesity in predicting dichotomized Accelerated GrimAge, then included an interaction term to test the moderation. Next, we extended this work by specifying a number of additional models. We tested the prior model that included the interaction of PTSD and obesity using other aging outcome measures, specifically GrimAge, PC-GrimAge, and DunedinPACE (all measured continuously). We then specified models using the continuous predictors (PTSD symptoms and body mass) that were used to derive the dichotomous predictors (PTSD and obesity) in the earlier models. Finally, we tested these prior models when including only non-Hispanic White male veterans, to provide an additional set of models aiming to best match the prior study’s inclusion criteria. All associations were tested using generalized linear models. Models testing associations for Accelerated GrimAge used logistic regression, whereas models testing associations for GrimAge, PC-GrimAge, and DunedinPACE used linear regression. All models adjusted for demographic covariates (sex, race and ethnicity, education, smoking, and age) and DNAm technical covariates (methylation chip and estimated white blood cell proportions).

Results

Replication Models Predicting Dichotomized Accelerated GrimAge

Accelerated GrimAge was not associated with PTSD or obesity, nor did PTSD interact with obesity to predict Accelerated GrimAge (Table 1).

Table 1.

Associations between PTSD, Obesity, and Epigenetic Aging Scores for U.S. Military Veterans

Accelerated GrimAge GrimAge PC-GrimAge DunedinPACE
N = 1,828 OR 95% CI β 95% CI β 95% CI β 95% CI
Main associations
 PTSD 1.20 [0.83, 1.72] 0.08* [0.01, 0.15] 0.10** [0.04, 0.17] 0.09** [0.01, 0.17]
 Obesity 0.74 [0.51, 1.08] 0.05 [−0.02, 0.12] 0.01 [−0.06, 0.07] 0.36** [0.28, 0.44]
 PTSD × Obesity 0.77 [0.37, 1.60] −0.13 [−0.27, 0.01] −0.17* [−0.30, −0.04] −0.04 [−0.19, 0.11]
PTSD associations stratified by obesity status
 PTSD, with obesity 0.93 [0.51, 1.69] 0.02 [−0.08, 0.12] 0.03 [−0.06, 0.12] 0.09 [−0.02, 0.21]
 PTSD, without obesity 1.38 [0.86, 2.23] 0.14** [0.04, 0.24] 0.16** [0.07, 0.26] 0.11* [0.00, 0.22]
Associations for continuous predictors
 PTSD symptoms (Sx) 1.15 [0.96, 1.37] 0.06** [0.03, 0.10] 0.07** [0.04, 0.10] 0.07** [0.03, 0.11]
 Body mass 0.87 [0.72, 1.05] 0.05** [0.02, 0.09] 0.02 [−0.01, 0.05] 0.24** [0.20, 0.28]
 PTSD Sx × Body mass 1.04 [0.86, 1.25] −0.04* [−0.08, −0.01] −0.05* [−0.06, 0.01] −0.03 [−0.07, 0.01]
Open in a new tab

Note: Models controlled for demographic covariates (sex, race and ethnicity, education, smoking, and age for models using DunedinPACE) and technical covariates (chip type, white blood cell type proportions). All associations represent independent models, though models with interaction terms included main effects. Models for Accelerated GrimAge used logistic regression; models for GrimAge, PC-GrimAge and DunedinPACE used linear regression. Continuous predictors and outcome estimates were standardized to a mean of 0 and SD of 1. CI = confidence interval.

*

p < .05.

**

p < .01.

Continuous Aging Scores

Veterans with PTSD had faster aging across all three continuous epigenetic aging outcomes (GrimAge, PC-GrimAge and DunedinPACE). DunedinPACE was also positively associated with obesity, whereas GrimAge and PC-GrimAge were not (Table 1). Descriptively, associations between PTSD and faster rates of epigenetic aging were stronger among veterans who were not obese (Figure 1); however, only PC-GrimAge showed a significant interaction effect (Table 1; Supplemental Table 1 for full results).

Figure 1.

Figure 1.

Open in a new tab

Standardized aging scores (GrimAge and DunedinPACE) residualized for covariates and organized by PTSD and obesity statuses. Error bars represent 95% confidence intervals and ns represent veterans within each category.

Models Utilizing Continuous PTSD Symptoms and Body Mass

Veterans with more PTSD symptoms had faster epigenetic aging in terms of DunedinPACE (consistent with a published study from this cohort; 18), GrimAge, and PC-GrimAge. Body mass was also positively associated with GrimAge and DunedinPACE scores, but not PC-GrimAge (Table 1). Body mass moderated the association of PTSD symptoms with GrimAge and PC-GrimAge, such that associations between PTSD symptoms and epigenetic aging were weaker among veterans with more body mass. There were no significant associations for DunedinPACE or Accelerated GrimAge.

Analyses of Non-Hispanic White Male Veterans

Among the 812 non-Hispanic White male veterans in the PDMH (best matching inclusion criteria from the prior study), results were similar to those in the full sample (Supplemental Table 2) with some descriptively stronger associations for PTSD. Obesity did not moderate the association of PTSD with Accelerated GrimAge, GrimAge, PC-GrimAge, or DunedinPACE in the direction reported by Fischer and colleagues (22). Any moderations were in the opposite direction of the prior study—PTSD has weaker associations with epigenetic aging in veterans with obesity or more body mass.

Moderation by Age and Chronic Disease

Age and chronic disease burden (see 21 for details regarding assessment of chronic disease burden using the Charlson Comorbidity Index) did not moderate associations in any primary models, suggesting our results were consistent across the ages and health statuses represented in the PDMH (Supplemental Analysis 1).

Discussion

In this replication study using data from 1,828 post-9/11 veterans, we did not find evidence that obesity moderated the association of PTSD and accelerated aging such that veterans with comorbid obesity and PTSD had the greatest risk of accelerated aging. This is in contrast to a recent study (22). Instead, any significant interactions between PTSD and obesity were in the opposite direction, such that associations between PTSD and aging were attenuated among veterans with obesity or more body mass. We also did not observe significant associations between PTSD, obesity, and dichotomized Accelerated GrimAge. However, we did find that veterans with PTSD and more PTSD symptoms had faster rates of epigenetic aging when assessed continuously. Veterans with obesity or more body mass also had faster DunedinPACE aging scores and greater body mass was associated with faster GrimAge. These findings align well with prior evidence that PTSD (20) and higher body mass (36–37) are associated with accelerated aging, including associations with PTSD reported in the NHRVS (18) and from our team using PDMH data (19).

There were important differences between the two studies relevant to interpreting these results, which may have contributed to the discrepancy in findings between studies. There were notable methodological differences between the studies with respect to DNAm assessment, PTSD assessment, and sample composition. Fischer and colleagues used salivary samples to derive DNAm, whereas the PDMH DNAm data were generated using whole blood (22). Recent studies have found lower reliability when comparing salivary and blood-based DNAm data (38). It is not established to what extent one tissue might be better suited for testing associations relevant to this study; however, it is notable that the majority of current epigenetic aging measures—including those used in this study (24–27)—were trained on DNAm data generated from whole blood. There were also differences in the methods used to assess PTSD. Fischer and colleagues used a self-report DSM-5 measure of PTSD symptoms, whereas we used a self-report measure of DSM-IV PTSD symptoms that was confirmed by PTSD diagnosis via clinical interview or medical records.

In addition, the study samples differed in their composition. The NHRVS is a national sample covering multiple service eras and the prior study sample included 1,135 men of European ancestry from the NHRVS with an average age of 63.3. In comparison, the PDMH is comprised of post-9/11 veterans with more variation in sex, race, and ethnicity, as well as younger individuals, with an average age of 37.6. There were also differences in the predictors and outcomes in the studies that could have contributed to the observed results: 30.9% of the NHRVS veterans were obese, 10.0% had probable PTSD, and 15.7% had Accelerated GrimAge. In contrast, 45.5% of the PDMH sample were obese, 41.5% had PTSD, and 10.9% had Accelerated GrimAge. Notably, we did not find evidence of moderation when analyses were limited to 812 non-Hispanic White male veterans. Similarly, we did not find that age moderated the interactions of interest in this study, suggesting that the results did not differ within the age range present in the PDMH. That said, it remains challenging to interpret the extent to which age may have influenced the differences in the associations observed between the two samples.

We also examined the associations of interest in PDMH using methodological and statistical approaches that provide important additional context to our results. We conducted analyses using both continuous and dichotomized predictors and outcomes, controlled for additional important covariates, and included additional methods of assessing PTSD and aging. We believe one major strength of our study approach is that our study included models using the continuous measures underlying PTSD diagnosis, obesity status, and accelerated aging. Although using established clinical cutoffs in cases like PTSD diagnosis and obesity is defensible, it is also useful to provide sensitivity analyses to ensure that the dichotomizing of continuous variables is not responsible for observed associations. To our knowledge, however, there is not a validated clinical cutoff for GrimAge, PC-GrimAge, or DunedinPACE. Given our aim was to replicate the prior study, we approximated the previous approach by including a dichotomized measure of GrimAge, but we remain cautious about dichotomizing continuous outcomes (40–41) unless they are shown to have adequate sensitivity and specificity (42). This is particularly true given that epigenetic aging scores are largely continuous measures with normal distributions and dichotomized outcomes can produce biased results. Our study results present a useful example for this caution. In the non-Hispanic White men, dichotomizing GrimAge reversed the direction of the association with body mass—body mass had a negative association with Accelerated GrimAge, but a positive association with GrimAge scores when assessed continuously (Supplemental Table 2).

We acknowledge Fischer and colleagues’ justification for their approach, including: 1) non-normality of their GrimAge data due to a significant Kolmogorov-Smirnov D test; 2) a desire to focus on clinically-meaningful differences; and 3) to provide comparisons to prior studies (22). We note in large samples and real-world data, Kolmogorov-Smirnov D tests are almost certain to reach significance. Visual inspection can be a more useful approach to assess normality, and aging scores showed largely normal distributions in our cohort (Supplemental Figure 2). In terms of clinically meaningful differences and comparisons to prior studies, three of the studies cited by Fischer and colleagues (43–45) examined associations between mortality and “Δage” (DNAm predicted age minus chronological age) using first generation clocks (Hannum and Horvath). Associations were expressed as hazard ratios in 5-year increases in Δage, rather than clinical cut points. The subtraction method used in these studies produce similar (but not identical) values to residualization methods of deriving age acceleration scores. For example, they are correlated at r = .89 in the PDMH. Conflating Δage and residualization approaches can be problematic, beyond concerns with using different generations of DNAm clocks.

In terms of covariates, Fischer and colleagues tested a number of self-reported psychosocial variables (e.g., exercise, optimism, loneliness, etc.) as moderators, and included these measures as covariates in their models. We did not include such measures, but added chronological age and smoking as covariates in all our models. Although controlling for age can be redundant in cases when age is residualized out of aging outcomes, such as is the case for GrimAge and PC-GrimAge acceleration, dichotomization affects distributions and could make controlling for age more relevant. In addition, previous reports have suggested that age is a critical covariate to include in models assessing epigenetic aging when including other predictors in full models (39). In terms of smoking, it could be argued that adjusting for smoking is an over control, given that it is likely in the causal pathway linking PTSD to accelerated aging and poor health (11). However, GrimAge (24) and PC-GrimAge (25) include a methylation measure of tobacco exposure in their algorithm, making it particularly important to control for smoking when testing associations for these measures with exposures of interest (i.e., PTSD), whether presented in final models or sensitivity analyses. We note that a prior study using the NHRVS data presented associations between PTSD and Accelerated GrimAge both with and without adjustments for smoking (19), which we find to be a useful approach.

Finally, our study included some measures that helped address possible reporting bias. PTSD diagnoses were confirmed with clinical interview or health records and our measure of body mass was generated using health records, providing some protection from biased responding. Second, we included multiple epigenetic aging measures as outcomes, including PC-GrimAge and DunedinPACE. Although we are reticent to reflexively include a large number of epigenetic measures of aging in our studies, we also believe selective inclusion helps provide confidence that any observed associations are not better explained by variation in a single aging measure.

Our study limitations include an inability to generalize to races and ethnicities beyond non-Hispanic Black and non-Hispanic White veterans. In addition, veterans from our sample may or may not represent veterans from other service eras or geographic regions. Given differences in the two samples’ average ages, it is difficult to determine whether this may have been responsible for why results in the study by Fischer and colleagues that did not replicate in the PDMH. Future studies of PTSD and obesity would benefit from methodological approaches that might better support causal inference, such as twin studies or instrumental variable designs (11). In addition, other assessments of obesity beyond body mass, such as waist-to-hip ratio, would be useful measures to investigate.

Conclusions

This study used data from a cohort of 1,828 post-9/11 veterans to conduct a replication and extension of a recent study of PTSD, obesity, and epigenetic aging (22). In contrast to prior study results, we did not find evidence that obesity moderated the association of PTSD and accelerated aging such that veterans with obesity and PTSD had the fastest aging. Instead, significant interactions were in the opposite direction of this effect—PTSD was more strongly associated with aging among veterans without obesity. We also found that veterans with PTSD had accelerated aging across multiple epigenetic measures, and that a third-generation aging measure (DunedinPACE) was associated with obesity. Taken together, our findings suggest that PTSD and obesity could each represent modifiable targets for interventions aiming to slow aging and improve health.

Supplementary Material

Supplement

Acknowledgements

Funding/Support:

This research was supported by Award #IK2CX002694 to Dr. Bourassa from the Clinical Science Research and Development (CSR&D) Service of VA ORD, and Award #I01BX002577 to Dr. Beckham from the Biomedical Laboratory Research and Development (BLRD) Service. Drs. Kimbrel (#IK6BX006523) and Beckham (#lK6BX003777) were supported by VA Research Career Scientist Awards.

Role of the Funder/Sponsor and Disclaimer:

The funders/sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the VA, the U.S. government or any other affiliated institution.

Abbreviations:

PTSD

Posttraumatic stress disorder

DNAm

DNA methylation

PDMH

Post-Deployment Mental Health Study

NHRVS

National Health and Resilience in Veterans Study

MIRECC

Mental Illness Research, Education, and Clinical Center

VA

Veteran’s Affairs

Group information:

VA Mid-Atlantic MIRECC Workgroup contributors for this paper include: Patrick S. Calhoun, PhD, Eric Dedert, PhD, Eric B. Elbogen, PhD, Robin A. Hurley, MD, Jason D. Kilts, PhD, Angela Kirby, MS, Scott D. McDonald, PhD, Sarah L. Martindale, Ph.D, Christine E. Marx, MD, MS, Scott D. Moore, MD, PhD, Rajendra A. Morey, MD, MS, Jennifer C. Naylor, PhD, Jared A. Rowland, PhD, Robert D. Shura, PsyD, Cindy Swinkels, PhD, H. Ryan Wagner, PhD.

Footnotes

Conflicts of interest: None of the authors have conflicts of interest to report.

Author access to data: Dr. Kyle Bourassa had full access to the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. This work was not preregistered.

Data Sharing Statement:

Data from the Post Deployment Mental Health (PDMH) Study is part of a Veterans Affairs data repository and is available to researchers who request access through the VISN 6 MIRECC and follow appropriate data access protocols. Medical record data from the Veteran Affairs Corporate Data Warehouse are available to researchers who request and are approved for access through the Office of Research and Development (ORD) Data Access Request Tracker (DART).

References

  • 1.Pacella ML, Hruska B, Delahanty DL. The physical health consequences of PTSD and PTSD symptoms: a meta-analytic review. J Anxiety Disord. 2013;27(1):33–46. doi: 10.1016/j.janxdis.2012.08.004 [DOI] [PubMed] [Google Scholar]
  • 2.Kubzansky LD, Koenen KC, Spiro A 3rd, Vokonas PS, Sparrow D. Prospective study of posttraumatic stress disorder symptoms and coronary heart disease in the Normative Aging Study. Arch Gen Psychiatry. 2007;64(1):109–116. doi: 10.1001/archpsyc.64.1.109 [DOI] [PubMed] [Google Scholar]
  • 3.Ebrahimi R, Lynch KE, Beckham JC, Dennis PA, Viernes B, Tseng CH, et al. Association of posttraumatic stress disorder and incident ischemic heart disease in women veterans. JAMA Cardiol. 2021;6(6):642–651. doi: 10.1001/jamacardio.2021.0227 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Edmondson D, Kronish IM, Shaffer JA, Falzon L, Burg MM. Posttraumatic stress disorder and risk for coronary heart disease: a meta-analytic review. Am Heart J. 2013;166(5):806–814. doi: 10.1016/j.ahj.2013.07.031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Khan N, Iqra Tanveer Khan S, Joti S, Malik J, Faraz M, Ashraf A. Association of cardiovascular diseases with post-traumatic stress disorder: An updated review. Cardiol Rev. doi: 10.1097/CRD.0000000000000628 [DOI] [PubMed] [Google Scholar]
  • 6.Thurston RC, Jakubowski K, Chang Y, Wu M, Barinas Mitchell E, Aizenstein H,, et al. Posttraumatic stress disorder symptoms and cardiovascular and brain health in women. JAMA Netw Open. 2023;6(11):e2341388. doi: 10.1001/jamanetworkopen.2023.41388 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Rosenbaum S, Stubbs B, Ward PB, Steel Z, Lederman O, Vancampfort D. The prevalence and risk of metabolic syndrome and its components among people with posttraumatic stress disorder: A systematic review and meta-analysis. Metabolism. 2015;64(8):926–933. doi: 10.1016/j.metabol.2015.04.009 [DOI] [PubMed] [Google Scholar]
  • 8.Roberts AL, Agnew-Blais JC, Spiegelman D, Kubzansky LD, Mason SM, Galea S, et al. Posttraumatic stress disorder and incidence of type 2 diabetes mellitus in a sample of women: a 22-year longitudinal study. JAMA Psychiatry. 2015;72(3):203–210. doi: 10.1001/jamapsychiatry.2014.2632 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Günak MM, Billings J, Carratu E, Marchant NL, Favarato G, Orgeta V. Post-traumatic stress disorder as a risk factor for dementia: systematic review and meta-analysis Br J Psychiatry. 2020;217(5):600–608. doi: 10.1192/bjp.2020.150 [DOI] [PubMed] [Google Scholar]
  • 10.Yaffe K, Vittinghoff E, Lindquist K, Barnes D, Covinsky KE, Neylan T, et al. Posttraumatic stress disorder and risk of dementia among US veterans. Arch Gen Psychiatry. 2010;67(6):608–613. doi: 10.1001/archgenpsychiatry.2010.61 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Bourassa KJ, Sbarra DA. Trauma, adversity, and biological aging: behavioral mechanisms relevant to treatment and theory. Transl Psychiatry. 2024;14(1):285. doi: 10.1038/s41398-024-03004-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kaeberlein M. Longevity and aging. F1000Prime Rep. 2013;5:5. doi: 10.12703/P5-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153(6):1194–1217. doi: 10.1016/j.cell.2013.05.039 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Campisi J, Kapahi P, Lithgow GJ, Melov S, Newman JC, Verdin E. From discoveries in ageing research to therapeutics for healthy ageing. Nature. 2019;571(7764):183–192. doi: 10.1038/s41586-019-1365-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Barzilai N, Cuervo AM, Austad S. Aging as a Biological Target for Prevention and Therapy. JAMA. 2018;320(13):1321–1322. doi: 10.1001/jama.2018.9562 [DOI] [PubMed] [Google Scholar]
  • 16.Moffitt TE. Behavioral and Social Research to Accelerate the Geroscience Translation Agenda. Ageing Res Rev. 2020;63:101146. doi: 10.1016/j.arr.2020.101146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Rutledge J, Oh H, Wyss-Coray T. Measuring biological age using omics data. Nat Rev Genet. 2022;23(12):715–727. doi: 10.1038/s41576-022-00511-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Bourassa KJ, Garrett ME, Caspi A, Dennis M, Hall KS, Moffitt TE, et al. Posttraumatic stress disorder, trauma, and accelerated biological aging among post-9/11 veterans. Transl Psychiatry. 2024;14(1):4. doi: 10.1038/s41398-023-02704-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Na PJ, Montalvo-Ortiz JL, Nagamatsu ST, Southwick SM, Krystal JH, Gelernter J, et al. Association of Symptoms of Posttraumatic Stress Disorder and GrimAge, an Epigenetic Marker of Mortality Risk, in US Military Veterans. J Clin Psychiatry. 2022;83(4):21br14309. doi: 10.4088/JCP.21br14309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Wolf EJ, Logue MW, Stoop TB, Schichman SA, Stone A, Sadeh N, et al. Accelerated DNA Methylation Age: Associations With Posttraumatic Stress Disorder and Mortality. Psychosom Med. 2018;80(1):42–48. doi: 10.1097/PSY.0000000000000506 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Bourassa KJ, Anderson L, Woolson S, Dennis PA, Garrett ME, Hair L, et al. Accelerated epigenetic aging and prospective morbidity and mortality among U.S. veterans. Journal of Gerontology: Medical Sciences. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Fischer IC, Na PJ, Nagamatsu ST, Jeste DV, Cabrera-Mendoza B, Montalvo-Ortiz JL, et al. Posttraumatic Stress Disorder, Obesity, and Accelerated Epigenetic Aging Among US Military Veterans. JAMA Psychiatry, 2024;81(12):1276–1277.. doi: 10.1001/jamapsychiatry.2024.3403 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Brancu M, Wagner HR, Morey RA, Beckham JC, Calhoun PS, Tupler LA, et al. The Post-Deployment Mental Health (PDMH) study and repository: A multi-site study of US Afghanistan and Iraq era veterans. Int J Methods Psychiatr Res. 2017;26(3):e1570. doi: 10.1002/mpr.1570 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Lu AT, Quach A, Wilson JG, Reiner AP, Aviv A, Raj K, et al. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging (Albany NY). 2019;11(2):303–327. doi: 10.18632/aging.101684 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Higgins-Chen AT, Thrush KL, Wang Y, Minteer CJ, Kuo PL, Wang M, et al. A computational solution for bolstering reliability of epigenetic clocks: Implications for clinical trials and longitudinal tracking. Nat Aging. 2022;2(7):644–661. doi: 10.1038/s43587-022-00248-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Belsky DW, Caspi A, Corcoran DL, Sugden K, Poulton R, Arseneault L, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. Elife. 2022;11:e73420. Published 2022 Jan 14. doi: 10.7554/eLife.73420 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Belsky DW. DunedinPACE calculator. 2022. https://github.com/danbelsky/DunedinPACE
  • 28.Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics. 2014;30(10):1363–1369. doi: 10.1093/bioinformatics/btu049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Morris TJ, Butcher LM, Feber A, Teschendorff AE, Chakravarthy AR, Wojdacz TK, et al. ChAMP: 450k Chip Analysis Methylation Pipeline. Bioinformatics. 2014;30(3):428–430. doi: 10.1093/bioinformatics/btt684 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Pidsley R Y Wong CC, Volta M, Lunnon K, Mill J, Schalkwyk LC. A data-driven approach to preprocessing Illumina 450K methylation array data. BMC Genomics. 2013;14:293. doi: 10.1186/1471-2164-14-293 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Leek JT, Johnson WE, Parker HS, Jaffe AE, Storey JD. The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics. 2012;28(6):882–883. doi: 10.1093/bioinformatics/bts034 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Robins L, Cottler L, Bucholz K, Compton W. Diagnostic interview schedule for DSM-IV (DIS-IV). 1995. [DOI] [PubMed]
  • 33.Davidson JR, Tharwani HM, Connor KM. Davidson Trauma Scale (DTS): normative scores in the general population and effect sizes in placebo-controlled SSRI trials. Depress Anxiety. 2002;15(2):75–78. doi: 10.1002/da.10021 [DOI] [PubMed] [Google Scholar]
  • 34.McDonald SD, Beckham JC, Morey RA, Calhoun PS. The validity and diagnostic efficiency of the Davidson Trauma Scale in military veterans who have served since September 11th, 2001. J Anxiety Disord. 2009;23(2):247–255. doi: 10.1016/j.janxdis.2008.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Houseman EA, Accomando WP, Koestler DC, Christensen BC, Marsit CJ, Nelson HH, Wiencke JK, Kelsey KT. DNA methylation arrays as surrogate measures of cell mixture distribution. BMC Bioinformat. 2012;13(1):1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Li J, Wang W, Yang Z, et al. Causal association of obesity with epigenetic aging and telomere length: a bidirectional mendelian randomization study. Lipids Health Dis. 2024;23(1):78. Published 2024 Mar 12. doi: 10.1186/s12944-024-02042-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Lundgren S, Kuitunen S, Pietiläinen KH, Hurme M, Kähönen M, Männistö S, et al. BMI is positively associated with accelerated epigenetic aging in twin pairs discordant for body mass index. J Intern Med. 2022;292(4):627–640. doi: 10.1111/joim.13528 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Apsley AT, Ye Q, Caspi A, Chiaro C, Etzel L, Hastings WJ, et al. Cross-tissue comparison of epigenetic aging clocks in humans. Aging Cell. 2025:e14451. doi: 10.1111/acel.14451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Krieger N, Chen JT, Testa C, et al. Use of Correct and Incorrect Methods of Accounting for Age in Studies of Epigenetic Accelerated Aging: Implications and Recommendations for Best Practices. Am J Epidemiol. 2023;192(5):800–811. doi: 10.1093/aje/kwad025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Altman DG, Royston P. The cost of dichotomising continuous variables. BMJ. 2006;332(7549):1080. doi: 10.1136/bmj.332.7549.1080 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.MacCallum RC, Zhang S, Preacher KJ, Rucker DD. On the practice of dichotomization of quantitative variables. Psychol Methods. 2002;7(1):19–40. doi: 10.1037/1082-989x.7.1.19 [DOI] [PubMed] [Google Scholar]
  • 42.Hassanzad M, Hajian-Tilaki K. Methods of determining optimal cut-point of diagnostic biomarkers with application of clinical data in ROC analysis: an update review. BMC Med Res Methodol. 2024;24(1):84. doi: 10.1186/s12874-024-02198-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Christiansen L, Lenart A, Tan Q, Vaupel JW, Aviv A, McGue M, Christensen K. DNA methylation age is associated with mortality in a longitudinal Danish twin study. Aging Cell. 2016;15(1):149–54. doi: 10.1111/acel.12421 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Marioni RE, Shah S, McRae AF, Chen BH, Colicino E, Harris SE, et al. DNA methylation age of blood predicts all-cause mortality in later life. Genome Biol. 2015;16(1):25. doi: 10.1186/s13059-015-0584-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Perna L, Zhang Y, Mons U, Holleczek B, Saum KU, Brenner H. Epigenetic age acceleration predicts cancer, cardiovascular, and all-cause mortality in a German case cohort. Clin Epigenetics. 2016;;8:64. doi: 10.1186/s13148-016-0228-z [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement
NIHMS2120167-supplement-Supplement.docx (89.8KB, docx)

Data Availability Statement

Data from the Post Deployment Mental Health (PDMH) Study is part of a Veterans Affairs data repository and is available to researchers who request access through the VISN 6 MIRECC and follow appropriate data access protocols. Medical record data from the Veteran Affairs Corporate Data Warehouse are available to researchers who request and are approved for access through the Office of Research and Development (ORD) Data Access Request Tracker (DART).

RESOURCES