Cumulative trauma, neural circuits, and burnout: an integrative model of healthcare worker post-traumatic stress syndromes

ABSTRACT

Background: Frontline healthcare workers (HCWs) experience unique patterns of repeated, chronic, and unpredictable traumatic event exposure, coupled with physiologic stress in the setting of shift-work circadian rhythm disruption, contributing to high rates of post-traumatic stress syndromes (PTSS) and substantial workforce and economic burden. The neurobiology underlying HCW-specific risk remains incompletely understood.

Objective: To synthesise epidemiological, neuroimaging, physiological, and interventional evidence into a mechanistic model of HCW PTSS and to identify priorities for biomarker-guided prevention and care.

Method: Literature for this narrative review was identified through a comprehensive search of peer-reviewed articles in PubMed, PsycINFO, and Google Scholar up to May 2025. Studies were included if they addressed (1) the epidemiology of PTSD in healthcare settings, (2) risk and protective factors specific to occupational trauma exposure, (3) neural, physiological, or molecular mechanisms associated with stress-related disorders in trauma-exposed personnel, or (4) interventions targeting PTSD/PTSS in HCWs and first responder populations.

Results: Across studies, PTSS prevalence among HCWs is variable (≈15–74%). Repeated, chronic, and unpredictable occupational trauma, exacerbated by circadian disruption, appears to destabilise frontal-limbic circuits and systemic stress pathways, culminating in allostatic overload. Converging data suggest that multimodal biomarkers, including resting-state and task-evoked fMRI metrics, MR spectroscopy, heart rate variability, sleep architecture, cortisol and inflammatory indices can identify prodromal dysregulation and define risk stratification.

Conclusions: Longitudinal, multimodal cohort designs are critically needed to track trajectories and evaluate neuroscientifically-informed treatment modalities for PTSS in this population. Framing HCW PTSS as an occupational neurobiological injury highlights the need to identify and prevent functional decline. A biomarker-guided strategy that links brain-circuit measures with autonomic, sleep, and molecular indices may offer a path to earlier identification, precision interventions, and improved outcomes for a critically at-need population that is essential to our workforce.

HIGHLIGHTS

• The prevalence of post-traumatic stress syndromes has skyrocketed in frontline healthcare workers, with estimates ranging from 15 to 74% across studies.

• There is a large gap in the literature regarding unique neural mechanisms underlying the occupational stress response in healthcare workers

• Chronic occupational trauma appears to disrupt frontal-limbic circuits and drive allostatic overload, ultimately resulting in post-traumatic stress syndromes.

• Multimodal biomarkers (fMRI, biometrics, circadian disruption) may enable early risk stratification.

• Effective interventions for individuals with high occupational trauma exposure are urgently needed to sustain the healthcare workforce.

1. Introduction

Posttraumatic stress disorder (PTSD) is a psychiatric condition characterised by persistent re-experiencing of trauma, avoidance of trauma-related stimuli, negative alterations in cognition and mood, and heightened arousal and reactivity after experiencing or witnessing a traumatic event (American Psychiatric Publishing, 2013). The rate of PTSD among healthcare workers, especially emergency personnel, is alarmingly high due to their frequent exposure to traumatic events. Healthcare workers (HCWs) routinely encounter high-stress and traumatic situations – from emergency resuscitations and patient deaths to workplace violence – as part of their jobs; this level of occupational trauma can lead to serious mental health outcomes including post-traumatic stress disorder (PTSD), depression and anxiety, substance use disorders, and suicide (Havaei, 2021; Jain et al., 2024; Patel et al., 2023). On top of a high level of traumatic event exposure, HCWs have also been under increasing amounts of stress due to the unpredictable availability of staff and resources creating prolonged conditions of uncertainty and reduced sense of personal agency, factors that contribute further to burnout and stress-related psychopathology (Chen et al., 2024). Such exposure to extreme occupational settings can result in a prolonged stress response which in some cases may lead to diagnostic criteria for PTSD, but can also be an attempt to adapt to this severe environment, resulting in a broader and ongoing post-traumatic stress syndrome (PTSS). Although PTSS is not a defined entity in current diagnostic nomenclature, it comprises a broader set of post-trauma stress responses that may encompass subthreshold or partial PTSD and related sequelae that may overlap with but not completely fulfil current diagnostic criteria for PTSD; this allows for the fact that symptoms related to chronic trauma exposure occur along a spectrum (Haslam & Mallon, 2003; Sparks, 2018).

Understanding the underlying neurobiology and neural mechanisms of traumatic stress syndromes in healthcare workers is crucial for multiple reasons. First, insights into the neurobiological alterations associated with PTSD can enhance diagnostic precision, allowing for earlier identification and intervention in affected individuals. Such knowledge can uncover the pathophysiological pathways involved, including dysregulation of neural circuits related to fear conditioning, emotion regulation, and the stress response. Currently, there is a large gap in the literature regarding the unique neural mechanisms underlying occupational stress response in HCWs, who face repeated and unpredictable traumatic event exposures as well as unique cultural factors that play a role in their trauma response. While other groups share elevated lifetime trauma exposure and substantial PTSS burden (e.g. high-violence communities), the composition and context of exposure diverge in ways that warrant HCW-specific study. HCWs experience repeated direct and vicarious trauma under a duty-to-act, high-stakes performance context with frequent exposure to moral injury, and are additionally subject to shift work – related circadian misalignment and sleep loss  – all factors that modulate stress systems and neural circuits relevant to threat processing and regulation. Focusing on HCWs is not redundant with studying chronic trauma broadly; rather, it targets a population with distinctive exposure profiles (duty-bound, morally injurious, circadian-disrupted, and vicarious) and modifiable occupational determinants, supporting tailored mechanistic hypotheses and interventions.

PTSS/PTSD in HCWs should be framed within a stress neurobiology context, wherein chronic, uncontrollable demands may degrade prefrontal regulatory networks and affect salience and threat appraisal (Arnsten & Shanafelt, 2021). In physicians specifically, occupational distress and burnout have been linked to stress-induced alterations in prefrontal network function and cognitive control (Arnsten & Shanafelt, 2021). Although distinct phenomena, recent neurobiological perspectives on physician burnout and chronic work stress provide relevant mechanistic anchors for studying PTSD in HCW populations (Abe et al., 2022; Arnsten & Shanafelt, 2021). Arnsten and Shanafelt (2021) synthesise data from previous brain research studies of stress showing that uncontrollable stress rapidly affects prefrontal cortical (PFC) regulation via catecholaminergic dysregulation (Arnsten & Shanafelt, 2021). Under conditions of sustained uncontrollable stress, PFC synaptic and gray matter connections appear to weaken; top-down control over attention, decision making, and emotion regulation is lost; and behaviour shifts toward amygdala-based ‘habit’ circuitry (Arnsten & Shanafelt, 2021). These patterns map onto clinical manifestations of occupational distress such as impaired cognitive control, impaired emotion regulation, and decreased motivation (Arnsten & Shanafelt, 2021). Importantly, they argue that restoring physicians’ sense of control is a neurobiologically coherent intervention point, as PFC function can recover with stress relief (Arnsten & Shanafelt, 2021).

Complementing this framework, Abe et al. provide structural anchors in medical professionals with burnout: in 43 actively practicing nurses, higher emotional exhaustion correlated with lower gray matter volume in bilateral ventromedial PFC (vmPFC) and left insula, while depersonalisation correlated with lower volume in left vmPFC and thalamus (Abe et al., 2022). These limbic-prefrontal findings dovetail with stress-related impairments in threat appraisal and regulatory control, reinforcing a model in which chronic occupational stress erodes vmPFC-mediated regulation and insula-mediated affective integration, with thalamic involvement potentially indexing broader cortico-subcortical dysregulation (Abe et al., 2022; Arnsten & Shanafelt, 2021). Together, these studies provide mechanistic and anatomical substrates for framing PTSD in HCWs as emerging from chronic, uncontrollable demands that tax PFC regulatory networks and bias threat systems (Abe et al., 2022; Arnsten & Shanafelt, 2021).

‘Allostasis’ refers to the adaptive regulation of internal states to maintain homeostasis according to environmental demands, while allostatic load denotes the cumulative ‘wear and tear’ from repeated activation of stress mediators (catecholamines, glucocorticoids, and cytokines) across molecular, cellular, and systems-level (hypothalamic-pituitary axis, autonomic nervous system, and network connectivity) processes (McEwen, 1998; McEwen & Stellar, 1993). Beyond occupational exposure, chronic stress adaptations can accrue from structural and social adversity (e.g. racialized stress, discrimination), which contributes to elevated allostatic load and may interact with workplace trauma to shape PTSS risk (Forde et al., 2019). In HCWs, recurrent occupational trauma, circadian disruption, and lack of control accelerate allostatic load, thereby increasing vulnerability to PTSS. We use this framework throughout to connect endocrine, immune, mitochondrial, and circuit changes to symptom trajectories.

Elucidating these neurobiological mechanisms can first help inform the development of targeted, neurobiologically-informed interventions designed to mitigate or reverse the adverse effects of chronic trauma exposure, tailored to the unique stressors faced by this population. Second, enabling rapid, biomarker-guided detection and triage via objective neural signatures would allow teams to identify at-risk clinicians before full-blown PTSD manifests, shortening the latency to evidence-based interventions. Third, mechanistic insight can specify which circuits to target with pharmacotherapy, neuromodulation, or exposure-based psychotherapies, and provide criteria for determining when clinicians are neurologically ready to resume frontline duties. Precise neural markers can also flag early functional changes (e.g. impaired attention, affective dysregulation) that jeopardise clinical judgment and patient safety. Fourth, linking specific neural stress signatures to modifiable environmental triggers (e.g. night shift circadian disruption amplifying limbic reactivity) can equip medical directors to restructure schedules and debriefing protocols. Fifth, by preventing chronic neurobiological stress adaptations that underpin both the post-traumatic stress response and burnout, institutions can curb attrition and staff turnover, currently an existential issue to understaffed hospitals and EMS agencies. Finally, demonstrating that healthcare post-traumatic stress symptoms arise from quantifiable brain-circuit alterations reframes an often-stigmatized issue as a treatable occupational injury, which may encourage HCWs to seek help and facilitate reimbursement for care.

2. Literature search

This narrative review was conducted to synthesise current knowledge on the neurobiological mechanisms, clinical features, and interventional strategies relevant to PTSD/PTSS in frontline HCWs. Literature was identified through a comprehensive search of peer-reviewed articles in PubMed, PsycINFO, and Google Scholar up to May 2025. Search terms included combinations of ‘posttraumatic stress disorder,’ ‘PTSD,’ ‘healthcare workers,’ ‘first responders,’ ‘occupational trauma,’ ‘burnout,’ ‘neuroimaging,’ ‘neurobiology,’ ‘intervention,’ ‘biomarkers,’ and ‘emergency medicine.’ Both observational and interventional studies were considered, with a focus on studies involving HCWs, first responders, or closely related occupational trauma populations (e.g. firefighters, emergency personnel) when relevant.

Studies were included if they addressed (1) the prevalence or epidemiology of PTSD in healthcare settings, (2) risk and protective factors specific to occupational trauma exposure, (3) neural, physiological, or molecular mechanisms associated with PTSD or stress-related disorders in trauma-exposed personnel, or (4) interventions targeting PTSD or stress symptoms in HCWs or first responder populations. Reference lists of included articles and relevant systematic reviews were also examined to identify additional sources. Emphasis was placed on recent (post-2010) findings with mechanistic or translational relevance. Non-English language articles, case reports, and non-peer-reviewed sources were excluded. Included studies are summarised in Table 1.

Table 1.

Healthcare occupational trauma studies.

First author (Year)	Population	Design	Key findings
Mills (2005)	EM resident physicians	Observational study	30% reported PTSD symptoms; up to 21.6% with diagnosis
Brymer et al. (2006)	General PFA framework	Field operations guide	Outlined PFA principles for early trauma response
Berninger et al. (2010)	Firefighters	Longitudinal cohort study	15.5% probable PTSD; high delayed onset; associated with functional impairment
Levy-Gigi and Richter-Levin (2014)	Firefighters	RCT	Trauma-exposed groups showed impaired hippocampal learning
Mealer et al. (2014)	ICU nurses	Feasibility RCT	Resilience training reduced symptoms in both intervention and control groups
Jung et al. (2016)	Firefighters	Resting-state fMRI	EMDR improved disrupted functional network properties
Jeong et al. (2019)	Firefighters	Resting-state fMRI	Increased insula-amygdala connectivity associated with symptom severity
Chen et al. (2020)	Medical staff (China)	Observational report	High reluctance to seek help despite distress symptoms
Albott et al. (2020)	HCWs (COVID-19)	Pilot resilience programme	Operationalized support increased access to mental health services
Hooper et al. (2021)	First responders	Systematic review	PFA, EMDR, trauma risk management showed efficacy
Crowe et al. (2021)	Critical care nurses	Cross-sectional survey	74% reported PTSD symptoms during COVID-19; 38% significant symptoms
Greenberg et al. (2021)	ICU staff (UK)	Observational study	High rates of PTSD, depression, and suicidal ideation
Cairns et al. (2021)	HCWs during pandemics	Scoping review	13 papers, mostly low-quality, described well-being interventions
Taylor et al. (2022)	HCWs	Multisite RCT	Digital mindfulness app reduced depression and anxiety
LoMauro et al. (2022)	HCWs (Covid-19)	Observational study	Frontline HCWs had higher theta relative power, lower peak alpha frequency, and higher interhemispheric coherence of alpha/theta rhythms
Ramineni et al. (2023)	ICU staff	RCT	Immediate intervention group had fewer intrusive memories
Hoell et al. (2023)	Paramedics	Systematic review and meta-analysis	12-month work-related PTSD prevalence of 20%
Hermosilla et al. (2023)	General (PFA evaluation)	Systematic review	PFA shows promise but has methodological limitations
Kanstrup et al. (2024)	Frontline HCWs	RCT	ICTI reduced intrusive memories and PTSD symptoms up to 6 months
Pihlgren et al. (2024)	HCWs	Qualitative study	ICTI perceived as acceptable intervention
Roger et al. (2024)	ICU staff	Multi-centre cross-sectional study	PTSD prevalence ranged from 16.8% to 30.8%

Abbreviations: PFA (psychological first aid), HCWs (healthcare workers), RCT (randomised controlled trial), ICU (intensive care unit), ICTI (imagery competing task intervention), PTSD (post-traumatic stress disorder), EMDR (eye movement desensitisation reprocessing), fMRI (functional magnetic resonance imaging).

3. Epidemiology, risk factors, and societal cost

The prevalence of PTSD in healthcare workers has been estimated across a variety of studies, with high rates noted across diverse contexts. Several studies have focused on Intensive Care Unit (ICU) staff. In a study of 109 critical care nurses in Canada, 74% self-reported symptoms of PTSD during the COVID-19 pandemic (with 38% reporting significant symptoms [Crowe et al., 2021]), while a large international cross-sectional multi-centre study of ICU staff workers during the second year of the COVID-19 pandemic found an overall rate of 16.8% of PTSD, with some sites as high as 30.8% (Roger et al., 2024). Another study of ICU staff found a broader spectrum of probable clinical significance for PTSD (40%), severe depression (6%), severe anxiety (11%), and problem drinking (7%); 13% reported thoughts of self-harm or a passive death wish (Greenberg et al., 2021). Intense intrusion symptoms have been reported in ICU workers (Deltour et al., 2023), as well as higher rates of hyperarousal, avoidance, numbness, and sleep disturbance (Lee et al., 2018). Among US physicians, it is estimated that 400 die by suicide each year, and that 14% have considered suicide (Kishore, 2019). Taken together, these findings point to a pervasive mental-health crisis in frontline medicine.

The COVID-19 pandemic substantially amplified trauma exposure and psychiatric morbidity in HCWs. Meta-analytic estimates from over 90,000 HCWs indicate pooled PTSD symptom prevalence of approximately 21–22% during the pandemic, alongside marked elevations in depression and anxiety; physician-focused synthesis similarly reports 18.3% PTSD with higher risk among women, trainees, and frontline specialties (Kamra et al., 2024; Y. Li, Scherer, et al., 2021). ICU cohorts showed high rates of probable mental health disorders and self-harm ideation during initial surges, underscoring the severity of acute impact on frontline teams (Greenberg et al., 2021). Pandemic-specific stressors including high patient mortality, infection risk and quarantine, PPE scarcity, moral injury from resource rationing, stigma, and disrupted circadian rhythms were consistently associated with worse outcomes in reviews across viral outbreaks and in COVID-19–specific moral injury syntheses (Kisely et al., 2020; Rushton et al., 2022; Serrano-Ripoll et al., 2020). Emerging longitudinal data suggest symptoms worsened during peak waves and partially remitted in some settings as caseloads and restrictions abated, while residual burnout and moral injury persisted (Lavell et al., 2025).

First responders and emergency medical personnel also show high rates of PTSD. In a longitudinal study of firefighters exposed to the World Trade Center disaster, 15.5% reported probable PTSD post-9/11 (n = 5656) with associated significant functional impairment, and nearly half of these cases were delayed onset (Berninger et al., 2010). A recent systematic review of a large sample of paramedics found work-related 12-month PTSD prevalence to be 20% (Hoell et al., 2023). The prevalence of PTSD among emergency medicine physicians was estimated at 15.1% in 2016 (Delucia), while 30% of emergency medicine resident physicians have reported symptoms of PTSD, and 11.9–21.6% may carry a diagnosis of PTSD (Mills, 2005). Thus, these figures highlight a critical occupational health challenge, emphasising the need for early screening, adaptive stress training, and readily accessible mental-health interventions.

Other studies have looked at specific risk factors associated with PTSD amongst healthcare workers. Across studies, increased overall scores on PTSD scales were associated with unpredictability, uncertainty, confrontation with death, burnout, catastrophic patient outcomes, experiencing clinical situations that remind one of past traumatic situations, participating in failed resuscitations, being female, younger age, performing duties outside of perceived skills; injury, death, or serious illness of a coworker; feeling like one’s own life was in danger; coping behaviours that included denial, self-distraction, self-blaming, and behavioural disengagement; and providing futile care (Colville et al., 2017; Deltour et al., 2023; Dennis et al., 2023; McMeekin et al., 2023; Mealer et al., 2009; Schreiber et al., 2019). Protective factors have also been examined, and include higher self-report resilience, support at work, sufficient resources, counselling, higher mindful awareness, and lower distress intolerance (Deltour et al., 2023; P. Li, Kuang, et al., 2021; McDonald et al., 2022; Tam et al., 2004). Overall, these findings map a risk-and-resilience profile, highlighting the need for multifaceted workplace strategies that both minimise modifiable stressors and actively cultivate protective resources.

PTSD among HCWs not only affects individual well-being, it imposes significant financial burdens on healthcare systems and society at large. While precise economic evaluations specific to HCWs are limited, extrapolations from broader PTSD economic burden studies can offer insight into potential costs. A comprehensive analysis estimated the total excess economic burden of PTSD in the United States at $232.2 billion for 2018, equating to approximately $19,630 per affected individual. These costs encompass direct health care expenses, indirect costs such as lost productivity, and non–health care costs including disability (Davis et al., 2022). Unresolved chronic mental health problems are also associated with dropout from the healthcare workforce, which comes at a large cost to society. Replacement cost estimates for paramedics and EMTs range from $50,000 to $80,000 per employee (NAEMT, 2021); a recent review by the NSI Nursing Solutions found that average costs for replacing a bedside nurses range from $52,350 to $97,216 per nurse (NSI Nursing Solutions, 2023), and physician replacement typically ranges from $500,000 to $1,000,000 factoring in recruitment, credentialing, onboarding, and lost productivity (Merritt Hawkins, 2019). Consequently, comprehensive mental-health prevention and intervention programmes for HCWs are essential investments in the fiscal sustainability and resilience of the entire healthcare system.

4. Interventional studies

Interventional studies conducted in HCWs with high occupational trauma exposure are limited. A systematic review of interventions for first responders identified 12 studies that described 6 early psychological interventions: psychological first aid, EMDR, and trauma risk management all showed efficacy in at least two studies; resilience and coping for the healthcare community, Anticipate-Plan-Deter (APD), and ‘resilience at work’ programmes showed positive results in single studies (Hooper et al., 2021; Schreiber et al., 2019). A review of interventions to support the well-being of healthcare workers during pandemics or other crises identified 13 papers (6 prospective and 7 descriptive studies), none of which were deemed to be high-quality theoretically-informed interventions (Cairns et al., 2021).

A randomised controlled trial of a multimodal resilience training programme for ICU nurses (n = 27, single centre) was shown to be feasible and acceptable; although both the intervention and control groups showed significant decreases in PTSD symptoms (as measured by the Posttraumatic Diagnostic Scale), the effect size was slightly larger for the intervention group (Mealer et al., 2014). Attempts have been made to operationalise resilience interventions, with a focus on self-care, self-efficacy, social connection, and rapid access to mental health consultation, as in the Battle Buddies programme piloted at the University of Minnesota Medical Center during COVID-19 (Albott et al., 2020). Small positive effect sizes have been seen for depression and anxiety with an unguided digital mindfulness based self-help app in HCWs (Taylor et al., 2022). Research also points to reluctance to participate in structured psychological interventions for medical staff (Chen et al., 2020), where it was observed that staff refused to accept medical help and denied having any problems despite showing irritability, excitability, and signs of psychological distress.

Imagery competing task interventions (ICTIs; brief behavioural interventions that typically involve a brief trauma cue followed immediately by engagement in a high-load visuospatial task, such as playing the game Tetris) have shown early promise in reducing intrusive memories of work-related trauma in HCWs (Iyadurai et al., 2018). A randomised controlled trial utilised an ICTI (compared to an attention-based control task) in frontline HCWs with trauma exposure during the COVID-19 pandemic, and found a significant reduction in the number of intrusive memories as well as fewer post-traumatic stress-related symptoms at 1, 3, and 6 month follow-up (Kanstrup et al., 2024); use of the intervention was perceived as acceptable by HCWs (Pihlgren et al., 2024). A similar ICTI in ICU staff showed strong evidence for a positive treatment effect, with fewer intrusive memories reported when there was immediate (vs delayed) access to the intervention (Ramineni et al., 2023).

Psychological First Aid (PFA) is an approach designed to reduce initial distress caused by traumatic events and to foster short- and long-term adaptive functioning and coping (Ramineni et al., 2023). PFA emphasises immediate, supportive responses that include active listening, ensuring safety, facilitating connections to social support, and providing practical assistance and information. By incorporating PFA into institutional support systems, healthcare organisations have attempted to mitigate acute stress responses, thereby promoting resilience, maintaining workforce stability, and safeguarding the overall well-being of their personnel. However, as highlighted in a recent review, while studies suggest an overall positive impact of PFA, a high risk of bias, inconsistent interventions, and inadequate evaluation methods create challenges in accurately assessing PFA’s effectiveness (Hermosilla et al., 2023).

In sum, current intervention research for trauma-exposed healthcare workers is encouraging but fragmentary: small randomised and quasi-experimental studies suggest that early psychological first aid, trauma-risk management, resilience skills programmes, digital mindfulness, and imagery-competing tasks can all yield modest reductions in intrusive memories, depression, or anxiety, yet few trials are powered to detect clinically meaningful change, most lack rigorous theoretical grounding, and uptake often falters because busy clinicians hesitate to engage in structured mental-health programmes. To translate these promising signals into durable, system-wide benefit, the field now needs mechanism-based trials that (i) leverage neurobiological and cognitive markers to match interventions to risk profiles, (ii) embed low-friction, technology-enabled interventions into existing workflows, and (iii) examine long-term system-based outcomes alongside symptom trajectories. Such a coordinated, evidence-driven agenda is essential if we are to move from ad-hoc ‘wellness’ offerings to scalable, empirically validated interventions for the mental health, performance, and retention of the healthcare workforce.

5. Neuroscientific studies in frontline healthcare workers

Before detailing neuroscientific findings, we briefly orient readers to (i) the core brain circuits repeatedly implicated in PTSD/PTSS and (ii) the EEG outcomes used later in this review. PTSD/PTSS consistently involves (i) threat-detection and learning systems (amygdala; hippocampus for contextualisation and extinction memory), (ii) prefrontal regulatory circuitry (vmPFC and rostral anterior cingulate cortex (rACC) supporting fear inhibition/extinction recall), (iii) salience and interoceptive circuitry (insula and dorsal ACC (dACC) linked to hypervigilance and conflict monitoring), and (iv) alterations within and between large-scale networks, particularly the default mode network (DMN, comprised of medial prefrontal and posterior cingulate hubs, supporting self-referential thoughts) and salience network (anterior insula, dACC, detecting and filtering salient stimuli) (Hayes et al., 2012; Neria, 2021). Across task and resting paradigms, meta-analytic and network-level reviews converge on amygdala and insula hyperreactivity, vmPFC and rACC hypoactivity, dACC hyperactivity, hippocampal dysfunction, and DMN instability, with corresponding changes in prefrontal–limbic connectivity (Etkin & Wager, 2007; Hayes et al., 2012; Neria, 2021). We use ‘EEG outcome measures’ here to refer to event-related potentials (ERPs) indexing time-locked cognitive–affective processes which show alterations in PTSD (particularly to trauma- or emotion-laden stimuli) and resting-state spectral metrics (alpha and theta power) that capture tonic arousal and control processes (Butt et al., 2019; Javanbakht et al., 2011).

Despite the high prevalence of PTSD in frontline HCWs, there have been few studies on the neural mechanisms underlying PTSD symptomatology in this population. A pilot EEG study (recorded at rest for 8 min) with a limited sample of frontline COVID-19 HCWs compared to HCWs in COVID-free units showed higher theta relative power, lower peak alpha frequency, and higher interhemispheric coherence of alpha and theta rhythms, postulated to be related to high levels of stress and mental fatigue (LoMauro et al., 2022). A 2014 study utilised a hippocampal-dependent context reversal paradigm in active duty firefighters, CSI police, and civilian matched-controls with no trauma exposure, and found that firefighters struggled to learn that a previously negative context was later associated with a positive outcome (which previous studies using this paradigm have suggested reflect deficits in hippocampal-related structure and function); performance on this task did not correlate with levels of PTSD, anxiety, or depression, suggesting that trauma-exposed firefighters have impaired reversal of negative context even without a diagnosis of PTSD (Levy-Gigi et al., 2014; Levy-Gigi & Richter-Levin, 2014). A subsequent study sought to investigate the neural effects of repeated traumatic stress in firefighters, including 98 healthy firefighters with repeated traumatic experiences but no diagnosis of mental illness. In comparison to non-firefighters with no history of trauma, stronger functional connectivity was seen between insula and several regions associated with fear and stress-responding circuitry including the amygdala, hippocampus, and vmPFC. Greater insula-amygdala connectivity was also associated with more severe trauma-related symptoms, while greater insula-vmPFC connectivity was associated with milder symptoms, suggesting insula functional connectivity as a potential candidate pathway for differentiating risk for or resilience to repeated traumatic stress (Jeong et al., 2019). Further evidence for network-level neural changes comes from a study of partial PTSD (pPTSD  – subthreshold PTSD which does not meet full diagnostic criteria) in trauma-exposed firefighters with and without pPTSD (n = 9 and 8, respectively) and non-traumatized health controls (n = 11) using resting state fMRI. Firefighters with pPTSD showed disruption in both global and local network properties, which were ameliorated after eye movement desensitisation and reprocessing (EMDR) therapy (Jung et al., 2016). Although limited by a small sample size, this study does provide preliminary evidence that early intervention for subthreshold PTSD in this population can potentially reverse underlying functional neural changes. Taken together, these converging yet still sparse data implicate hippocampal-insula-prefrontal circuit disruptions in trauma-exposed responders and underscore the need for larger, longitudinal neuroimaging studies of frontline healthcare workers to pinpoint neurobiological markers of risk, functional adaptation, and treatment responsiveness.

6. A novel neuroscientific model of PTSD in frontline healthcare workers

As evidenced above, there is comparatively little high-quality PTSD research that examines healthcare workers as a distinct group. HCWs have unique patterns of repeated, cumulative trauma and often vicarious trauma, which may not identically mirror single-incident trauma biology. There is a large knowledge gap regarding this specific population; while some of the same neural and inflammatory changes observed in other PTSD cohorts may apply, it is crucial to identify differences in biomarkers, resilience factors, and efficacy of interventions.

We propose an integrated model in which repeated occupational trauma in HCWs leads to progressive dysregulation across neural circuits, stress response systems, and cognitive-emotional processes, ultimately resulting in PTSD. This model emphasises stressors unique to HCWs – namely, frequent exposure to trauma (both direct and secondary) and cultural pressures to suppress distress – and how these factors may cumulatively induce neurobiological changes (Figure 1).

Figure 1.

Repeated and cumulative trauma exposure in frontline HCWs coupled with workplace culture, emotional suppression, moral injury, primary appraisal (is this a threat, challenge, or irrelevant), and secondary appraisal (perceiving oneself as well-prepared and supported, or overwhelmed and ill-equipped) (Folkman et al., 1986) initiates a cascade of changes. Key neural circuits of fear and stress (amygdala, vmPFC, ACC, hippocampus, insula) are progressively dysregulated. Amygdala and insula become hyperactive, while the vmPFC/ACC’s capacity to regulate fear and extinguish trauma memories is reduced, and hippocampal context processing is impaired. ACC dysregulation is also implicated in aberrant self-referential processing that occurs in the setting of moral injury, exacerbating feelings of guilt and shame. Concurrently, systemic stress response systems are altered: HPA axis shows altered cortisol dynamics, the autonomic nervous system shifts toward sympathetic dominance (low HRV), and the immune system exhibits elevated inflammatory signalling. Neurochemical alterations (excess glutamate relative to GABA, monoamine imbalances) further promote a hyperexcitable and dysregulated brain state. These changes, compounded by high allostatic load, drive the development of chronic PTSD pathology in HCWs.

6.1. Key neural circuits implicated in occupational trauma-related PTSD

Repeated trauma exposure in HCWs engages the same core fear and stress circuits known in PTSD, notably the amygdala, medial prefrontal cortex (mPFC, including vmPFC and ACC), hippocampus, and insula. The amygdala, a central hub for threat detection and emotional memory, becomes hyperactive in PTSD, contributing to exaggerated fear responses and hypervigilance (Saccenti et al., 2024; Wang et al., 2016). Conversely, the vmPFC and ACC exhibit hypoactivity, reflecting impaired top-down regulation of fear; as a result, inhibitory control over the amygdala is weakened (Saccenti et al., 2024). This imbalance is a hallmark of PTSD neurocircuitry (Wang et al., 2016). The hippocampus, responsible for distinguishing safe versus dangerous contexts and contextualising memories, also shows reduced volume and activity in previous studies, contributing to context generalisation and intrusive flashbacks; while the anterior insula, which processes interoceptive cues and salience, is frequently hyperactive, reflecting heightened awareness of distressing sensations and anxiety in PTSD (Wang et al., 2016). The insula and amygdala also act in concert; when their increased activity is coupled with reduced prefrontal modulation, this yields the characteristic anxiety, emotional dysregulation, and re-experiencing symptoms of PTSD (Wang et al., 2016).

In frontline HCWs, these circuit dynamics are repeatedly taxed. Emergency physicians, nurses, and first responders face high-risk situations regularly, and this chronic repetition of trauma may induce sensitisation of fear circuitry. With each traumatic event, stress-related neuroplastic changes may occur: for instance, the amygdala-insula response may be re-triggered, while persistent activation of regulatory regions could lead to functional exhaustion of the vmPFC/ACC. HCWs also experience vicarious trauma from empathically sharing patients’ pain and fear on a near-daily basis, which may engage similar neural pathways. Previous research on empathy and vicarious stress has shown that witnessing others’ trauma can activate the observer’s own pain and threat circuitry (particularly the insula and ACC) sometimes nearly as strongly as direct trauma (Russell & Brickell, 2015). Over time, this vicarious activation likely compounds the direct effects of first-hand trauma exposure. Additionally, cultural pressure and professional necessity in healthcare to remain stoic in the moment encourages emotion suppression. HCWs often must cognitively inhibit their distress instead of processing it, potentially leading to insufficient fear extinction and persistent limbic activation. In HCWs, constant suppression of fear or grief on the job may blunt engagement of adaptive regulatory circuits (such as reappraisal via the dorsal PFC) and leave the amygdala-insula responses unchecked.

6.2. Progressive neural circuit alterations

We hypothesise that repeated occupational trauma produces progressive neuroplastic changes in these circuits via chronic stress mechanisms. Initially, frequent trauma exposure may cause changes in functional connectivity among key regions – as seen in the study of repeatedly trauma-exposed firefighters, which found increased connectivity between insula and amygdala, with greater insula-amygdala coupling linked to greater trauma-related symptoms, and stronger insula-vmPFC coupling associated with milder symptoms (Jeong et al., 2019). This suggests a potential early adaptation: if the insula couples more strongly with prefrontal regulatory regions, this may be a marker of resilience, but if increased insula-amygdala coupling is seen, pathological fear salience will dominate. With cumulative trauma, however, such compensatory insula-prefrontal connectivity may wane.

Chronically increased amygdala reactivity without prefrontal inhibition can lead to fear circuit potentiation over time (Guadagno et al., 2021), where the threshold for amygdala activation lowers, contributing to hypervigilance and startle at mild triggers. Simultaneously, vmPFC/ACC circuitry may undergo synaptic remodelling or dendritic retraction due to prolonged stress exposure (Chen et al., 2008; Zhang et al., 2019), further reducing its regulatory efficacy. The hippocampus is especially vulnerable to chronic stress; evidence from studies of PTSD and burnout indicates that ongoing stress can impair hippocampal neurogenesis and plasticity (Russell & Brickell, 2015). Repeated trauma exposure might thus erode hippocampal volume or connectivity, compromising contextual memory processing. Reduced hippocampal volume is often seen in PTSD, and one theory posits that individuals with chronic trauma exposure accumulate microdamage to the hippocampus (Yehuda, 2001), although other studies suggest that smaller hippocampi precede trauma and increase vulnerability (Lindgren et al., 2016). In either case, diminished hippocampal function over time could exacerbate the over-generalisation of fear, e.g. a clinician feeling anxiety in any hospital room after experiencing multiple codes there.

Chronic, repeated trauma is also expected to progressively degrade regulatory prefrontal circuitry through stress-mediated neuroplasticity. At a systems level, sustained glucocorticoid and catecholamine surges during repeated threat exposure impair mPFC and ACC network function and weaken top-down control of limbic structures, producing hypoconnectivity within prefrontal networks and reduced vmPFC–amygdala coupling, which is critical for fear inhibition and extinction recall (Arnsten, 2015; Milad & Quirk, 2012; Woo et al., 2021). In healthcare professionals with occupational burnout (a distinct but related phenotype of chronic work-related stress), studies have shown reduced gray matter in ACC and dorsolateral PFC, and altered amygdala connectivity associated with impaired regulation of negative affect (Abe et al., 2022; Golkar et al., 2014). Collectively, these findings support a model in which repeated trauma burdens prefrontal regulatory systems, thereby diminishing regulatory efficacy over amygdala-insula responses and increasing vulnerability to intrusive imagery and hyperarousal in frontline HCWs.

It is important to recognise that these neural changes do not occur in isolation; they evolve alongside stress-related physiological changes (e.g. endocrine, autonomic, immune), which both influence and reflect stress-related neural circuitry changes. Chronic activation of limbic circuitry drives ongoing output to the hypothalamic-pituitary-adrenal (HPA) axis downstream, and hormonal and inflammatory signals in turn feed back to the brain, creating further neural changes (Ulrich-Lai & Herman, 2009). In this manner, cumulative trauma creates a cycle whereby each traumatic incident reactivates an increasingly sensitised neural network, triggers outsized physiological stress responses, and leaves behind incremental biological scars that make subsequent stress responses bigger and more difficult to regulate (Yehuda & LeDoux, 2007).

6.3. HPA axis dysregulation

Frontline HCWs under constant stress often show dysregulation of the HPA axis (Ibar et al., 2021). Acute trauma typically provokes an increase in cortisol, but PTSD has been paradoxically associated with lower basal cortisol and exaggerated negative feedback sensitivity over time (Morris et al., 2012; Pan et al., 2018). Chronic occupational trauma may initially drive high cortisol outputs; however, as stress becomes persistent, the system may ‘overshoot,’ leading to blunted baseline cortisol levels due to hypersensitive glucocorticoid receptors and excessive feedback inhibition (Morris et al., 2012; Schumacher et al., 2019). Meta-analyses show that individuals with PTSD tend to have lower circulating cortisol than trauma-exposed controls, as well as higher levels of corticotropin-releasing factor (CRF) in the brain, suggesting a state of HPA axis exhaustion or sensitisation rather than simple over-activation (Bremner et al., 1997; Morris et al., 2012; Pan et al., 2018). In HCWs, a flattened diurnal cortisol rhythm may develop, especially as shift work and sleep deprivation further disrupt the neuroendocrine system (Burek et al., 2024; Li et al., 2018). Cortisol normally helps terminate the stress response and has anti-inflammatory effects, thus, hypocortisolism in chronic PTSD may remove a systemic brake on inflammation, leading to a pro-inflammatory, hyper-reactive physiological state (Heim, 2020).

Simultaneously, repeated epinephrine/norepinephrine surges during each crisis event can lead to sustained sympathetic nervous system activation, even at rest, as evidenced by elevated heart rate and blood pressure in individuals with PTSD (Schneider & Schwerdtfeger, 2020). Over time, the allostatic mechanisms that normally adapt to stress may fail – if continuously bombarded with high catecholamines, chronically high sympathetic tone and low parasympathetic tone persist, a profile reflected in measures like heart rate variability (HRV) (Schneider & Schwerdtfeger, 2020). HCW studies and PTSD research show significantly lower HRV in those with PTSD, indicating impaired parasympathetic regulation (Ge et al., 2020; Järvelin-Pasanen et al., 2018; Ulmer et al., 2018). Low HRV in turn correlates with hyperarousal, insomnia, and emotion dysregulation in PTSD (Ge et al., 2020; Schneider & Schwerdtfeger, 2020). Our model posits that cumulative trauma drives an autonomic imbalance characterised by an ‘always alert’ state, in which the body’s fight-or-flight system fails to ever fully down-regulate.

6.4. Neuroinflammatory changes

Chronic occupational stress and disrupted HPA axis signalling can also precipitate an immune inflammatory response. It has been previously demonstrated that some individuals with PTSD have elevated inflammatory biomarkers, including higher circulating C-reactive protein (CRP), interleukin-6 (IL-6), and tumour necrosis factor-alpha (TNF-a), and lifetime trauma burden appears to contribute to these inflammatory changes (Costello et al., 2019; Kim et al., 2020; Kuring et al., 2023). Peripheral inflammatory signals reach the brain via multiple pathways, including (i) humoral routes at the choroid plexus and cytokine-induced endothelial signalling, (ii) neural routes via sympathetic afferents, and (iii) stress-related increases in blood-brain barrier permeability; these inputs (particularly with repeated stress) collectively activate microglia and bias neuroimmune tone toward pro-inflammatory states (Dantzer, 2018; Dantzer et al., 2008; Nahum et al., 2022; Li, Liao, et al., 2021). In our model, HCWs with repeated trauma accrue an allostatic load that includes an inflammatory component; this inflammation feeds back into the brain, affecting neural circuits and behaviour. At the neurochemical level, inflammatory cascades upregulate indoleamine-2,3-dioxygenase, diverting tryptophan to kynurenine (resulting in reduced serotonin availability) and increase glutamatergic transmission via microglial and kynurenine pathway metabolites; both mechanisms impact synaptic plasticity in fear and emotion networks (Haroon et al., 2017; Miller et al., 2013; Miller & Raison, 2016). Elevated IL-6 has been linked to heightened amygdala reactivity to threat and increased anterior insula and subgenual ACC engagement, which may exacerbate anxiety and depressive symptoms and mirror regulatory circuit abnormalities in PTSD (Alvarez et al., 2020; Harrison et al., 2009; Inagaki et al., 2012; Kim et al., 2020; Muscatell et al., 2016). Thus, we incorporate a bidirectional loop, where chronic stress causes inflammation, which alters neurotransmission and functional connectivity in fear and emotion circuits, which worsens PTSD symptoms, which in turn can sustain stress (Costello et al., 2019; Kim et al., 2020; Kuring et al., 2023).

6.5. Neurochemical imbalance

Repeated trauma is also hypothesised to disrupt the brain’s glutamate/GABA neurotransmitter balance. Acute stress causes surges in glutamate (the primary excitatory neurotransmitter) which can lead to excitotoxic effects and neuronal damage over time, particularly in the hippocampus and PFC (Averill et al., 2022; Reid et al., 2024; Rosso et al., 2017). Chronic PTSD has been associated with elevated glutamate levels and markers of excitatory activity, alongside possible reductions in inhibitory GABAergic tone (Reid et al., 2024; Rosso et al., 2017). An imbalance that favours glutamate could produce hyperarousal and anxiety and impair the formation of new neurons (Reid et al., 2024). Meanwhile, reduced GABAergic function equates to less inhibition of overactive circuits, contributing to insomnia, hypervigilance, and exaggerated startle responses (Reid et al., 2024; Rosso et al., 2017; Rosso et al., 2022). Our model suggests that chronic stress in HCWs results in an excitatory/inhibitory imbalance in key brain regions, which could be detected in future MR spectroscopy studies.

Monoaminergic systems (serotonin, dopamine, norepinephrine) are also altered by chronic trauma (Krystal & Neumeister, 2009). PTSD is often characterised by high noradrenergic tone and by changes in serotonin signalling (which may contribute to the high rates of comorbid depression) (Krystal & Neumeister, 2009). Effects on dopamine may impact reward processing and anhedonia in HCWs in a state of burnout (Torrisi et al., 2019). These neurochemical changes are interlinked with HPA axis and inflammatory changes; prolonged cortisol elevation early in allostatic load can deplete monoamines, and inflammation can degrade tryptophan (a serotonin precursor), leading to serotonin deficits (Dziurkowska & Wesolowski, 2021; Tafet & Nemeroff, 2016; Tseilikman et al., 2020). Over time, depletion of neurotrophic factors (such as brain-derived neurotrophic factor (BDNF)) due to high stress and glutamate may occur, which hinders synaptic plasticity and recovery in areas like the hippocampus (Quigley et al., 2025). In sum, the neurochemical environment in a chronically trauma-exposed HCW brain shifts toward one that favours the development of chronic anxiety, depression, and cognitive function impairment.

6.6. Cognitive modulators and allostatic load

Cognitive appraisal and emotion regulation play a crucial role as modulators of these neurobiological processes. For HCWs, appraisal of trauma can be complex: traumatic events may be minimised or normalised as part of the job, or conversely internalised as personal failures. Negative appraisals (e.g. ‘I am helpless’ or ‘The world is completely unsafe’) are known to maintain PTSD symptoms by sustaining a sense of current threat. In the healthcare context, moral injury (feelings of having violated one’s moral code or witnessing gross injustice in care) can further colour these appraisals, contributing to anger, guilt, and shame that perpetuate stress activation.

Emotion regulation is another key mediator. As previously mentioned, a culture of stoicism may lead HCWs to rely on expressive suppression (inhibiting outward signs of emotion). While suppression may keep one functional in the moment, it does not decrease the internal physiological arousal. Previous neuroimaging studies show that emotion suppression often leaves amygdala and insula activity relatively high (even if facial expressions are controlled) and can increase sympathetic nervous system activation due to the stress of the effort required (Konrad et al., 2025). Over time, this pattern may strengthen limbic circuitry since emotions are not fully processed and increase allostatic load due to sustained autonomic arousal. Conversely, more adaptive strategies like cognitive reappraisal (a technique that involves regulating emotions by reinterpreting the meaning of an event [Wallace-Hadrill & Kamboj, 2016]) engage the dlPFC and vmPFC to modulate limbic activity (Tang et al., 2025). Our model predicts that individuals with better emotion regulation skills will show less dysfunction in prefrontal regulatory circuits and a slower accumulation of allostatic load.

In HCWs, allostatic load may manifest as a combination of neuroendocrine strain (altered cortisol, high catecholamines), cardiovascular strain (hypertension, elevated resting heart rate, reduced HRV), metabolic changes, and immune activation. PTSD has been described as a state in which normal stress adaption systems are maxed out (allostatic overload) (Carbone et al., 2022); different individuals have different thresholds for allostatic overload, which could explain why some HCWs develop PTSD after similar exposures while others remain resilient (Carbone et al., 2022). Factors like genetic predisposition, prior trauma history, and social support affect this threshold. Our model (Figure 1) posits that once allostatic load crosses a certain threshold, self-perpetuating pathological cycles begin (for example, chronic inflammation begets neurotoxicity, sleep disruption impairs emotion regulation). Cognitive and emotional factors modulate allostatic load by either accelerating it (when someone catastrophizes each event and never truly rests) or buffering it (when someone utilises effective coping and social support to discharge stress between events).

7. Future directions and experimental framework for model validation

In keeping with our translational model, we focus here on intervention domains that map onto the specific stress exposures and constraints of frontline HCWs, including rotating shifts and circadian disruption, high autonomic load during prolonged vigilance, sensory overload, and morally injurious or self-referential stressors, rather than attempting an exhaustive review of all PTSD treatments. The approaches below were selected because they target mechanistic nodes highlighted in our framework (allostatic load, regulatory circuit degradation, and mitochondrial/energetic strain) and are feasible to deploy within healthcare systems, making them particularly suitable for HCW-focused prevention and intervention.

7.1. Workplace and system-level injuries

It is important to first highlight that while we target biological processes, we should also fix the environment causing the injury. In a stressed healthcare system, HCWs frequently deal with a lack of resources, staffing, and appropriate organisational interventions. Future research should focus on organisational interventions as a form of prevention. Institutional policies that reduce chronic stress such as better staffing ratios, consistent or waterfall-scheduled shifts to avoid short turnaround times and excessive circadian rhythm disruption, and providing on-site psychological debriefing teams that staff are given the ability to see during their shifts are possible interventions to be evaluated. Future studies should also empirically test whether rotating staff out of high-trauma roles more frequently or mandating decompression breaks after difficult cases (which would require healthcare services to increase staffing to safely facilitate) measurably lowers biological markers of stress and results in lower rates of PTSD. This systems-level institutional approach will complement individual-focused treatments.

7.2. Continuous biometric monitoring

Emerging evidence suggests that disruptions in circadian rhythms, decreased physical activity, and autonomic dysregulation (frequently measured via HRV), are closely associated with PTSD. PTSD is characterised not only by emotional dysregulation but also by significant disturbances in physiological and behavioural homeostasis, including sleep and activity. Circadian rhythm disruption is frequently observed in PTSD, manifesting as delayed sleep onset, fragmented sleep patterns, altered melatonin secretion, and dysregulated cortisol rhythms (Agorastos & Olff, 2020; Germain, 2013). These disturbances exacerbate clinical symptoms, particularly impaired emotion regulation and intrusive memories, and predict poor treatment response (Germain, 2013; Hunt et al., 2023). Reduced physical activity levels are also consistently documented in PTSD populations. Lower activity contributes to symptom severity, negatively impacts emotional processing and stress resilience, and correlates with impaired functional connectivity in stress-related neural circuits (Dong & Lin, 2025; Hall et al., 2020). Conversely, increasing activity levels through targeted exercise interventions has shown promise in symptom reduction and improved stress regulation, suggesting a bidirectional relationship (Whitworth & Ciccolo, 2016). HRV, a robust marker of autonomic nervous system function, is significantly lower in individuals with PTSD, indicative of impaired parasympathetic regulation and heightened sympathetic tone (Minassian et al., 2014; Shaikh al arab et al., 2012). Lower HRV is associated with elevated symptom severity, hyperarousal, emotional dysregulation, and impaired fear extinction processes (Dennis et al., 2016; Gillie & Thayer, 2014). Importantly, improvements in HRV following interventions (e.g. biofeedback, cognitive-behavioural therapy, mindfulness) correlate with reduced PTSD symptoms and enhanced emotional resilience (Pyne et al., 2023; Tan et al., 2011).

For HCWs, the relevance of these metrics lies less in activity volume (which may be relatively high on many shifts) and more in the patterning and timing: irregular rotations, night shifts, and prolonged ‘on-call’ periods distort circadian phase, fragment sleep, and sustain sympathetic activation despite physical fatigue. Frequent transitions between intense psychomotor activity (codes, resuscitations) and prolonged vigilant monitoring (ICU, ED boarding) may produce characteristic signatures of misaligned and non-restorative activity, which are not captured by generic step counts but can be quantified with continuous monitoring.

These findings underscore the importance of integrating physiological and behavioural metrics in future studies – including monitoring circadian rhythm patterns, activity levels, and HRV – for the purpose of further elucidating PTSD pathophysiology, enhancing diagnostic accuracy, and informing personalised intervention approaches. One avenue by which these metrics can be integrated into future studies is by utilising wearable devices (such as the WHOOP band, Oura ring, Fitbit, etc.), which have been previously validated in research studies and allow for continuous non-invasive tracking of sleep, activity, heart rate, and HRV in real time (Khodr et al., 2024; Miller et al., 2020; Miller et al., 2021; Miller et al., 2022; Schyvens et al., 2024). In frontline HCWs, this could enable real-time workplace-integrated monitoring, where deviations in sleep regularity, HRV, or activity timing around high-intensity rotations flag individuals at elevated risk and trigger stepwise, occupation-tailored interventions (e.g. schedule adjustments, structured recovery periods, or targeted psychotherapy), providing a rationale for including this domain in HCW-focused research.

7.3. Neuromodulation therapies

Neuromodulatory therapies offer possibilities to directly target dysfunctional brain circuitry in PTSD. Studies of transcranial magnetic stimulation (TMS) in PTSD, particularly over the right dlPFC, have shown reductions in hyperarousal and improved executive function; work is ongoing to identify optimal targets and stimulation protocols (Siddiqi et al., 2024). Real time fMRI neurofeedback has also been utilised to help individuals learn to consciously modulate the activity of brain regions; this has been used to help PTSD patients learn to modulate amygdala activity (Gerin et al., 2016; Zhao et al., 2023), PCC (Lieberman et al., 2023) and ACC activity (Zweerings et al., 2018). To target deeper brain structures, invasive means were historically required for neurostimulation; low intensity focused ultrasound (LIFU) now appears to be a promising non-invasive new methodology for neuromodulation of deep structures (Arulpragasam et al., 2022); although not tested yet in PTSD, symptomatology may be able to be mitigated by LIFU modulation of areas such as the anterior and posterior cingulate, insula, and amygdala. Within the HCW context, neuromodulation is attractive for several reasons. First, our earlier framework emphasises vmPFC/ACC regulatory failure, dACC–insula hyper reactivity, and amygdala over-activity in HCWs exposed to uncontrollable and morally injurious events; these are precisely the circuits that TMS, neurofeedback, and LIFU can target. Second, many HCWs are reluctant to use psychotropic medication because of sedation, cognitive side effects, or stigma related to licensure; device-based, circuit-targeted interventions may therefore be more acceptable in this occupational group. Future trials should explicitly test whether HCW-specific exposure profiles (e.g. moral injury, vicarious trauma) moderate neuromodulation response and whether circuit normalisation predicts safe return to frontline duties.

7.4. Float therapy

Floatation therapy involves floating supine in a shallow pool of warm water saturated by Epsom salt, typically in an environment (open or pod-based) calibrated such that sensory stimulation is minimised. There are currently no neuroscientific studies of floatation-REST (Reduced Environmental Stimulation Therapy) in PTSD populations; however, a previous study of individuals with a broad spectrum of anxiety and stress-related disorders and depression showed reduced state anxiety, stress, depression, and negative affect after a 1-hour float session (Feinstein et al., 2018). Only one published study has combined neuroimaging and float-REST; this demonstrated resting state functional connectivity changes between and within hubs of the default mode network (DMN) as well as somatosensory cortices, possible reflecting reduced self-reflective processes with attention directed instead toward the current state of the body (Al Zoubi et al., 2021). The DMN, including medial prefrontal, posterior cingulate/precuneus, and medial temporal hubs, supports self-referential processing and autobiographical memory and is frequently disrupted in PTSD (Raichle, 2015). Previous studies have shown reduced DMN connectivity in individuals with PTSD, with greater reductions correlating with more severe symptoms (Lanius et al., 2020); these PTSD-associated alterations in DMN appear to be complex, with one study suggesting that this is limited to the medial temporal lobe subsystem (Miller et al., 2017). Given links between DMN instability and symptom severity in PTSD, we hypothesise that DMN-rebalancing via reduced exteroceptive load in float-REST could lower allostatic load in HCWs by dampening continuous salience/self-threat appraisal, a testable mechanistic pathway for future trials. Float-REST has the potential to be particularly effective for HCW PTSD due to targeting of self-referential DMN processes; negative self-appraisal (both personally and through the lens of group membership) plays a role in the development and maintenance of PTSD symptoms in HCWs, and is a key function that maps onto the DMN (Agathos et al., 2024; Delahoy et al., 2022). Additionally, the minimised sensory environment provided by float-REST is ideal for decreasing the sustained sympathetic nervous system activity and ‘always alert’ state previously described, and directly counteracts features of the frontline clinical environment including continuous alarms, crowded visual fields, time pressure, and high cognitive load by providing a brief, reproducible state of extreme sensory, social, and performance-demand reduction. Future interventions can test the underlying neural mechanisms of float-REST in this population, as well as associated physiological changes.

7.5. Mitochondrial bioenergetics

Psychological stress modulates mitochondrial dynamics, bioenergetics, and signalling; chronically, these adaptations constitute ‘mitochondrial allostatic load,’ linking neuroendocrine and immune mediators to cellular energy dysregulation with resulting behavioural consequences (Picard & McEwen, 2018). Emerging evidence implicates a breakdown in mitochondrial bioenergetics in PTSD, marked by (i) altered phosphocreatine (PCr) and creatine (Cr) dynamics and (ii) excitatory–inhibitory neurotransmitter imbalance (Lushchak et al., 2022). Phosphorous magnetic resonance spectroscopy (MRS) imaging provides a non-invasive window on cerebral energy flux (PCr, ATP), while proton MRS simultaneously quantifies glutamate/GABA in PTSD-relevant brain regions such as ACC and hippocampus. Studies have reported increased Cr in the amygdala of adults with PTSD and, more recently, higher ACC Cr predicting superior stress-recovery in trauma-exposed veterans, suggesting that robust high-energy phosphate buffering may confer resilience  (Amorim et al., 2025; Yancey et al., 2024). Incorporating MRS with mitochondrial health indices will directly test whether HCW stress trajectories map onto cerebral energy flux (PCr/ATP) and excitatory–inhibitory balance, and whether metabolic augmentation (e.g. creatine supplementation) normalises these indices alongside symptoms. Future work should utilise longitudinal MRS to trace bioenergetic trajectories in frontline HCWs, and experimentally test whether oral creatine monohydrate supplementation can normalise PCr/Cr ratios, dampen excess glutamate, and reduce symptom burden. Early open-label clinical observations and pre-clinical work demonstrate mood-stabilizing and neuroprotective effects of creatine, underscoring its safety and mechanistic plausibility as a low-cost metabolic adjunct in PTSD. For HCWs specifically, mitochondrial allostatic load is a plausible downstream integrator of chronic sleep/circadian disruption, repeated acute stress surges, and sustained cognitive–emotional demand inherent to frontline care. Longitudinal MRS in HCWs across high-intensity rotations could identify bioenergetic phenotypes associated with impaired recovery and inform stratification for metabolic augmentation strategies. Including this future area of study thus reflects our broader aim of linking occupationally patterned stress exposures to quantifiable bioenergetic targets that may be particularly salient in HCWs compared with other trauma-exposed groups.

8. Limitations

Similar to the general population, a proportion of frontline HCWs have previous trauma exposure not from their occupation: in a secondary analysis of a 2003 study of 176 HCWs, 68% reported at least one lifetime experience of trauma, abuse, or neglect, and 33% reported this adversity before age thirteen (Maunder et al., 2010). Amongst practicing physicians, 44.5% reported at least 1 adverse childhood event (ACE) (Stork et al., 2020), and amongst female RNs in the Nurses’ Health Study II, 57% reported childhood abuse (Boynton-Jarrett et al., 2013). Future research into HCW occupational trauma should collect data on lifetime traumatic events and ACEs, as this may play a role in both the neural correlates identified and response to intervention.

As demonstrated herein, relatively little high-quality neuroscientific research has been conducted that examines healthcare workers as a distinct group. Structural barriers (shift work, circadian misalignment, on-call demands) may discourage participation and introduce physiological confounds that are seldom measured or controlled for (Morris et al., 2016). Stigma and licensing concerns further reduce disclosure and research engagement, biasing the literature toward convenience, cross-sectional samples (Dyrbye et al., 2017). Exposure constructs are inconsistently defined and often conflated (PTSD, burnout, secondary traumatic stress, moral injury), creating measurement non-equivalence and construct overlap (Cieslak et al., 2014; Mantri et al., 2020).

Another important question is to what degree occupational trauma in HCWs can be separated from other trauma types to yield biomarkers that would uniquely inform intervention development. Meta-analytic work indicates shared PTSD circuitry across trauma modalities, implying some degree of neurobiological overlap (Ross & Cisler, 2020). What may be distinctive in HCWs is the composition of trauma exposure dimensions – chronic uncontrollability, vicarious threat and suffering, moral injury, and circadian disruption – which is more likely to shift network-level neural engagement than to produce single-region specificity. Thus, unique HCW biomarkers may more plausibly be multivariate patterns mapped to specific domains than categorical diagnoses. Design implications include rigorous separation of occupational (including moral injury and vicarious trauma) versus lifetime non-occupational trauma, modelling sleep/circadian and inflammation as time-varying covariates, and adequately powered, network-level analyses to inform target selection and treatment matching.

9. Conclusions

Frontline HCWs experience a uniquely hazardous blend of direct, cumulative, and vicarious trauma that places them at disproportionate risk for PTSD/PTSS. Epidemiological data reveal consistently elevated PTSD prevalence across emergency, critical care, and first responder settings, with substantial personal, institutional, and societal costs. Synthesising the available evidence, we propose that repeated occupational trauma progressively dysregulates frontal-limbic circuitry – marked by amygdala/insula hyper responsivity, impaired vmPFC–ACC regulation, and hippocampal dysfunction – while concurrently driving HPA-axis exhaustion, autonomic imbalance, neuro inflammation, and excitatory–inhibitory neurochemical shifts. Cognitive factors such as maladaptive appraisal, emotion suppression, and moral injury accelerate these biological cascades, culminating in an allostatic overload that manifests clinically as PTSD.

Translational opportunities arise from this model. Multimodal longitudinal studies that pair fMRI, MRS, and autonomic nervous system and sleep monitoring can identify early circuit and bioenergetic signatures of risk, while randomised trials of neuromodulation, supplement-based metabolic support, floatation-REST, and system-level staffing reforms can test causal links and preventive potential. Critically, individual-level interventions must be embedded within organisational strategies that reduce chronic stressors and foster a culture of psychological safety, thereby addressing both ‘the person’ and ‘the place’ of injury. By integrating neuroscience with occupational actualities, future research can move the field from descriptive epidemiology to evidence-based prevention and treatment.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.