Network Analysis of PTSD, Depression, and Anxiety Symptom Co-occurrence among US Veterans Seeking Treatment

Network Analysis of PTSD, Depression, and Anxiety Symptom Co-occurrence among US Veterans Seeking Treatment

How do PTSD, depression, and anxiety symptoms overlap in veterans — and why does it matter for treatment? A network analysis of 591 treatment-seeking U.S. veterans found that PTSD, depression, and generalized anxiety form five distinct but interconnected symptom communities. The overlap is concentrated in specific symptom clusters — not evenly distributed — which suggests that targeted and transdiagnostic treatment approaches both have a role, depending on which symptoms are driving distress.

TL;DR

  • PTSD symptoms do not behave as a single block — intrusion and avoidance cluster separately from hyperarousal, numbing, and negative cognition, each connecting differently to depression and anxiety.
  • Irritability, sleep disturbance, and detachment are the PTSD symptoms most likely to bridge into depression and anxiety symptom networks.
  • Depression and generalized anxiety formed their own separate communities in this veteran sample, suggesting they are related but not interchangeable conditions.
  • Providers managing co-occurring PTSD, depression, and anxiety in any high-exposure population should consider which symptom clusters are most active rather than treating a single diagnosis in isolation.

When a veteran — or any high-exposure professional — presents with overlapping symptoms of PTSD, depression, and anxiety, the clinical picture gets complicated fast. Providers frequently encounter patients who screen positive on multiple instruments, and the default approach is often to anchor treatment to a primary diagnosis. That creates a problem: if the conditions are not independent but are instead connected through shared symptoms, treating one in isolation may leave critical drivers untouched.

This is not a theoretical concern. In a sample of nearly 4.5 million veterans, 48.1% of those with PTSD also had co-occurring depression, and 13.9% had co-occurring anxiety disorders. Those numbers reflect a system-level reality: most veterans seeking treatment are not dealing with one condition. They are dealing with a symptom network — and the structure of that network determines which symptoms maintain distress over time.

A 2024 study published in the European Journal of Trauma & Dissociation used network analysis to map how PTSD, depression, and generalized anxiety symptoms co-occur in treatment-seeking veterans. The findings have direct implications for how providers in any high-exposure field — military, EMS, fire, law enforcement — should think about assessment and intervention.

What Did the Study Ask?

Researchers wanted to know how symptoms of PTSD, major depressive disorder (MDD), and generalized anxiety disorder (GAD) cluster together in veterans seeking treatment. Specifically, they asked whether symptom groupings align with traditional diagnostic categories from the DSM-5 — or whether symptoms reorganize across diagnoses into transdiagnostic communities.

The study also examined which individual symptoms act as bridges between conditions. These bridge symptoms are clinically significant because they may be the points where one condition activates or sustains another.

What Did the Researchers Find?

The analysis included 591 treatment-seeking veterans who completed validated self-report measures: the PCL-5 (PTSD), the PHQ-9 (depression), and the GAD-7 (generalized anxiety). The sample averaged 38.91 years of age, was 78.7% male, and 55.3% had deployed to a combat zone. Critically, 64.13% screened positive for probable PTSD, 66.67% met criteria for moderate to severe depression, and 65.14% met criteria for moderate to severe anxiety.

The network analysis identified five distinct symptom communities:

  1. Intrusion and Avoidance — PTSD Clusters B and C grouped together. These include intrusive thoughts, nightmares, flashbacks, and avoidance behaviors.
  2. Hyperarousal and Numbing — PTSD symptoms including detachment, restricted affect, irritable behavior, reckless behavior, hypervigilance, startle response, and sleep disturbance.
  3. Negative Alterations — PTSD Cluster D symptoms including amnesia, negative beliefs, self-blame, and persistent negative emotion.
  4. Depression Symptoms — PHQ-9 items formed their own community.
  5. Generalized Anxiety Symptoms — GAD-7 items formed their own community.

This five-community structure appeared in 50.8% of 1,000 bootstrapped replications, indicating relatively stable results with moderate uncertainty about the exact number of communities.

The strongest connection between diagnostic categories was between "easily annoyed" (GAD) and "irritable behavior" (PTSD), with an edge-weight of 0.37. Other notable cross-diagnostic links included anhedonia connecting to restricted affect, suicidal ideation connecting to reckless behavior, and concentration difficulties connecting to inability to relax.

Bridge Expected Influence (BEI) analysis revealed which symptoms most strongly connected different communities:

  • In the Intrusion & Avoidance community: physiological cue reactivity had the highest bridge influence (0.42)
  • In the Hyperarousal & Numbing community: sleep disturbance (0.58), detachment (0.52), and irritable behavior (0.52) had the strongest bridge influence
  • In the Negative Alterations community: persistent negative emotion (0.59) and negative beliefs (0.52) were the top bridge symptoms
  • In the Depression community: concentration difficulties (0.44) and worthlessness (0.43) were the strongest bridges
  • In the Anxiety community: easily annoyed (0.54) and inability to relax (0.43) had the highest influence across communities

How Does This Apply to First Responder Mental Health?

The study focused on veterans, but the relevance extends to any population with sustained occupational trauma exposure — EMS, fire, law enforcement. A prior study by Baker et al. (2023) using the same methodology in a first responder sample found a similar structure: intrusion and avoidance symptoms clustered with high specificity, while hyperarousal and negative cognition symptoms overlapped more with depression and anxiety.

This is where the system breaks down in practice. Standard screening catches the presence of PTSD, depression, or anxiety. It does not typically identify which symptoms are bridging across conditions and maintaining the broader distress pattern. A provider treating depression in a first responder may reduce PHQ-9 scores while leaving the hyperarousal and numbing symptoms — which connect back to both anxiety and depression networks — untouched.

The study's findings suggest that specific symptoms — irritability, sleep disturbance, detachment, negative emotion — are not just features of PTSD. They are functional connectors to depression and anxiety. In any high-exposure profession, these are also the symptoms most likely to affect operational performance, interpersonal relationships, and long-term career sustainability.

The researchers noted that transdiagnostic treatments like the Unified Protocol, which targets shared emotional processes rather than single diagnoses, showed promise for this type of co-occurring presentation. Mindfulness-based approaches also showed relevance among veterans with PTSD. The broader implication: when symptom networks overlap this heavily, single-diagnosis treatment models may be insufficient.

How Does Symptom Co-occurrence Present in Practice?

A typical case might involve a first responder or veteran presenting to a clinician with sleep disturbance, irritability, low energy, and difficulty concentrating. On screening instruments, the scores flag moderate depression and possible PTSD. Treatment begins with a depression-focused protocol.

What the network analysis indicates is that the sleep disturbance and irritability in this scenario may not be primarily depressive symptoms. They may be hyperarousal symptoms bridging into the depression network. If treatment addresses mood without addressing the arousal and numbing cluster, the bridge symptoms remain active — and recovery stalls.

This pattern is observable in practice across EMS, fire, and law enforcement settings where providers are managing ongoing exposure. Symptoms that look like depression or anxiety on a screening tool may actually be maintained by PTSD mechanisms that require a different intervention approach.

What Can Providers and Systems Do With This Information?

  • Screen across conditions, not just for a primary diagnosis. The co-occurrence rates in this sample — 64% PTSD, 67% depression, 65% anxiety — indicate that single-instrument screening is likely insufficient in high-exposure populations.
  • Identify bridge symptoms during assessment. Sleep disturbance, irritability, detachment, and negative emotion are the symptoms most likely to connect PTSD to depression and anxiety. These should be assessed explicitly and tracked over time.
  • Consider transdiagnostic treatment models. The study's findings support treatments that address shared emotional processes — negative affect, arousal dysregulation — rather than targeting diagnoses in isolation.
  • Do not assume intrusion and avoidance will respond to general distress treatment. These symptoms clustered uniquely in the network. The researchers noted that trauma-focused interventions directed at intrusions and avoidance may be most effective for core PTSD symptoms, consistent with APA and VA/DoD clinical practice guidelines.
  • Track irritability as a cross-diagnostic signal. The strongest between-category connection in the entire network was between GAD-related irritability and PTSD-related irritable behavior. In operational settings, irritability is often dismissed as personality or fatigue. This research suggests it may be a critical indicator of cross-condition symptom activation.

What Are the Limits of This Research?

The study has several important limitations that the researchers themselves acknowledged:

  • Cross-sectional design. The data represents a single time point. The researchers cannot establish causality — only that these symptom associations exist at the time of measurement. Longitudinal research is needed to determine whether these community structures persist through treatment.
  • Self-report measures only. Participants completed the PCL-5, PHQ-9, and GAD-7 without diagnostic interviews. Self-report instruments, while validated, do not replace clinical assessment.
  • Sample demographics. The sample was predominantly male (78.7%), White (61.7%), and Army-affiliated (43.1%). Generalizability to other demographic groups, branches, and non-veteran populations should be approached cautiously.
  • No trauma exposure or substance use measures. The study did not include specific assessments of trauma type or substance use, both of which can influence symptom network structure.
  • Moderate uncertainty in community number. The five-community solution appeared in 50.8% of bootstrapped replications. The 95% confidence interval ranged from 3.13 to 6.86 communities, indicating that while the structure is relatively stable, there is meaningful uncertainty about the exact groupings.
  • Mixed deployment history. The sample included both deployed (55.3%) and non-deployed veterans. Future analyses examining these groups independently would strengthen the findings.

Bottom Line

When managing PTSD, depression, and anxiety in high-exposure populations, providers should identify and prioritize the specific bridge symptoms — sleep disturbance, irritability, detachment, negative emotion — that maintain distress across conditions, rather than anchoring treatment to a single diagnosis.

References

Baker, L. D., Ponder, W. N., Carbajal, J., Norton, R., Price, M., Cassiello-Robbins, C., & Roberge, E. M. (2024). Network analysis of PTSD, depression, and anxiety symptom co-occurrence among U.S. veterans seeking treatment. European Journal of Trauma & Dissociation, 8, 100447. https://www.sciencedirect.com/science/article/pii/S246874992400070X

Baker, L. D., Ponder, W. N., Carbajal, J., Galusha, J. M., Hidalgo, J. E., & Price, M. (2023). Mapping PTSD, depression and anxiety: A network analysis of co-occurring symptoms in treatment-seeking first responders. Journal of Psychiatric Research, 168, 176–183. https://doi.org/10.1016/j.jpsychires.2023.10.038

Code 3 Academy is part of a broader effort with Emergency Services Outreach, Inc., focused on education, awareness, and long-term support for first responders.

Back to blog