1. Introduction
Moral injury (MI) is a profound psychological distress that arises from perceived violations of one’s ethical or moral code in complex, high-stakes situations. It has emerged as a significant concern among medical personnel and first responders, particularly in high-stress environments such as COVID-19 units and combat zones, where exposure to ethically challenging events is frequent [
1]. MI is characterized by symptoms including guilt, shame, anger, anxiety, and intrusive thoughts, and is often associated with outcomes such as depression, burnout, post-traumatic stress, and impaired professional functioning. The experience and impact of MI can be shaped by individual vulnerability, organizational factors, and broader cultural or national contexts, all of which influence both the risk and manifestation of moral injury among healthcare workers and emergency responders [
2,
3].
The prevalence of MI varies in most studies between a range of 30 and 50% in civilian settings [
4,
5,
6], while the prevalence of MI in military health professionals is not specifically detailed.
Moral injury can have profound consequences, manifesting as both psychological comorbidities and functional impairments. On the psychological front, it often co-occurs with depression, anxiety, post-traumatic stress disorder (PTSD), and burnout, reflecting the deep emotional distress and guilt associated with such experiences. Beyond these individual symptoms, MI also has significant functional implications [
4]. It can lead to reduced job satisfaction and impaired teamwork, affecting professional performance and collaboration. Additionally, personal relationships may suffer due to emotional withdrawal or increased irritability [
1,
7]. Organizational factors, such as inadequate resources and poor leadership, can exacerbate these effects, further straining personal and professional life. Overall, the impact of MI is multifaceted, influencing not just mental health but also interpersonal dynamics and professional efficacy [
2].
Despite the growing recognition of MI, there remains a need for a comprehensive understanding of its risk factors, manifestations, and effective interventions across different healthcare and emergency response settings. This systematic review seeks to consolidate current research on MI among medical personnel and first responders, emphasizing validated assessment tools and empirical findings to guide effective mitigation and support strategies. By examining the current state of knowledge on MI in these critical professions, this review seeks to contribute to the development of targeted interventions and systemic support structures necessary to address the pervasive impact of MI on frontline and military healthcare workers.
2. Methods
A total of 725 records were identified from three databases: Semantic Scholar corpus (n = 498), PubMed (n = 128), Google Scholar (n = 93), and the Cochrane Central Register of Controlled Trials (CENTRAL; n = 6). After removing 248 duplicate records, 477 unique records were screened. Of these, 389 were excluded by AI and 34 by human reviewers, leaving 54 reports for further investigation. Of these, 6 reports could not be retrieved, resulting in 48 reports assessed for eligibility. Seven reports were excluded due to insufficient methodological detail or results. Ultimately, 41 studies met the inclusion criteria and were incorporated into the final review. The total number of participants (14,500) was calculated by summing the sample sizes reported in each of the 41 included studies.
2.1. Review Design
Given the complexity of the topic and the heterogeneity of the available evidence, we elected to conduct a narrative review rather than a systematic review. The literature on moral injury among medical personnel and first responders encompasses a wide range of study designs, including qualitative research, observational studies, and conceptual analyses, which vary considerably in methodology, context, and outcomes measured. A narrative review approach is particularly well-suited for synthesizing such diverse sources, as it allows for a more flexible and interpretive integration of findings. This method enables us to critically appraise and contextualize the evidence, identify emerging themes, and highlight conceptual gaps that may not be readily captured through quantitative synthesis. By adopting a narrative approach, we aim to provide a comprehensive and nuanced understanding of moral injury in these populations, offering insights that can inform both future research and practical interventions.
2.2. Ethics
No ethics approval was required for this manuscript, as it did not involve any human participants or animal subjects.
2.3. Registration
The protocol for this review was registered in PROSPERO (PROSPERO CRD420251019492). One protocol amendment was made to extend the outcomes, providing a more holistic understanding of the disease by incorporating symptoms, comorbidities, and functional implications.
2.4. Search Strategy
Four electronic databases (PubMed, CENTRAL, Google Scholar, and Semantic Scholar corpus) were searched to identify papers that reported on the prevalence, risk factors, and psychological impacts of MI among medical personnel and first responders. The databases searched included Semantic Scholar, which provided access to over 126 million academic papers. The search terms were designed to capture the population of interest (healthcare workers and first responders) and study purpose (moral injury). Full search terms are available in
Supplementary Table S1. Primary searches were limited to papers published from 2010 to 2025 to identify assessments used in current practice and/or research.
2.5. Inclusion/Exclusion Criteria
2.5.1. Population
Included
This includes healthcare workers (such as nurses and physicians), first responders (such as paramedics and firefighters), and other medical personnel—individuals with a medical education who work in hospitals or other healthcare facilities.
Studies reporting baseline prevalence of MI or moral distress.
Studies focusing on individuals experiencing MI, moral distress, or related moral problems (e.g., guilt, shame).
Excluded
Non-healthcare or non-emergency response populations (e.g., military-only samples).
General populations or non-healthcare/emergency response workers.
Studies focusing exclusively on burnout or occupational stress without addressing MI.
2.5.2. Intervention(s) or Exposure(s)
Included
Studies evaluating psychosocial (e.g., therapy, counseling), pharmacological, or combined interventions targeting MI.
Interventions addressing specific risk factors (e.g., shame, betrayal) or comorbid disorders like PTSD or depression.
Excluded
Interventions not specifically designed to address MI (e.g., general stress management programs).
Studies without detailed descriptions of interventions.
2.6. Screening
Papers were screened based on the following criteria.
Population Type: The study focused on healthcare workers (medical doctors, nurses, paramedics, EMTs, clinical staff) and/or first responders (police officers, firefighters, emergency response personnel) who provide direct patient care or emergency response services.
Direct Care/Response Role: Study participants were primarily involved in direct patient care or emergency response (excluding administrative, management, or support roles).
Moral Injury Focus: The study explicitly examined MI, or moral distress (excluding general work stress, burnout, or compassion fatigue).
Study Type: The study was a primary research study (quantitative, qualitative, or mixed-methods) with 5 or more participants.
Empirical Evidence: The study presented empirical data (excluding opinion pieces, editorials, or theoretical frameworks).
Outcomes: The study reported on prevalence rates, risk factors, psychological impacts, and mental health outcomes related to MI. These symptoms and functional outcomes—depression, anxiety, PTSD, burnout, reduced job satisfaction, impaired teamwork, and diminished professional performance—were selected because they are consistently reported as key consequences of moral injury in both the empirical literature and the studies included in this review [
7,
8,
9,
10,
11,
12,
13,
14].
We considered all screening questions together and made a holistic judgment about whether to screen in each paper.
2.7. Data Extraction
Data were extracted using a combined approach. Both a human reviewer and the Elicit AI research assistant independently extracted information from each included study. Extraction focused on the following subjects:
Study Design: Type of research (e.g., cross-sectional, longitudinal, cohort, case-control).
Data Collection Method: Method used (e.g., online survey, paper questionnaire, structured interview) and any tools or questionnaires applied (e.g., Moral Injury Symptoms Scale-Health Professional).
Sample Composition: Participant demographics, including total sample size, professional categories, gender distribution, age range or mean age, and geographic location.
COVID-19 Exposure: Details of participants’ exposure to COVID-19 (e.g., direct vs. non-direct care, roles, duration, and quantitative exposure measures).
Moral Injury Outcomes: Measurement tools used, prevalence, scoring methods, and key findings related to moral injury.
Mental Health Correlates: Associated mental health outcomes, such as depression, anxiety, and burnout.
After extraction, results from both the human reviewer and Elicit were compared and synthesized to ensure completeness and accuracy.
Data extraction was performed independently by both a human reviewer and the Elicit AI tool. Each extracted key variables from the included studies covering study design, data collection method, sample composition, COVID-19 exposure, moral injury outcomes, and mental health correlates. After extraction, results were compared and synthesized; any discrepancies were resolved through discussion with a second human reviewer, and if needed, by returning to the original study text for clarification. Quality control was ensured by having the human reviewer cross-check all AI-extracted data, addressing limitations such as potential AI misinterpretation or omission of nuanced information. This combined approach aimed to maximize both efficiency and accuracy in the review process.
Given the inclusion of various article types (e.g., critical reviews, rapid reviews, scoping reviews), we adopted steps to make sure we could compare them fairly. When studies were similar, we compared their findings directly. For studies that were very different, we focused on common themes to bring the evidence together.
2.8. Methodological Quality Assessment
Methodological quality was assessed for each included study by extracting available quality assessment scores, identifying key bias indicators (e.g., study design, sampling, measurement validity), summarizing limitations as reported by study authors, and evaluating generalizability. Given the heterogeneity of included studies, a domain-based approach was used rather than a single standardized tool. This information is summarized in the results tables and informs the interpretation of findings. Military relevance refers to any aspect of the study that pertains to military service, including populations with military backgrounds, settings within military institutions, or outcomes applicable to military personnel.
4. Discussion
This review highlights the significant and multifaceted impact of MI among medical personnel and first responders across diverse healthcare and emergency response settings. The findings demonstrate that MI is not only prevalent—affecting an estimated 30–50% of individuals in high-stress environments such as COVID-19 wards and combat zones—but also deeply consequential, with wide-ranging psychological and professional effects. Consistent with prior literature, individual risk factors such as female gender, younger age, and frontline roles (e.g., nurses, EMTs) were associated with increased vulnerability to MI, while pre-existing mental health conditions and high self-criticism further compounded risk [
3,
8,
13,
15,
24].
Organizational and situational factors emerged as particularly salient contributors. Chronic understaffing, inadequate resources (such as insufficient PPE), excessive workloads, poor ethical climate, and lack of leadership support were repeatedly identified as high-impact organizational drivers of moral injury. Situational exposures—including direct care of COVID-19 patients, repeated patient deaths, and ethically fraught decision-making—intensified the risk, especially during global health crises and in military contexts where dual roles and extreme conditions prevail [
10,
18,
19,
40,
41]. Addressing these organizational issues is essential for reducing both the prevalence and severity of MI in high-stress healthcare and emergency settings [
36,
39].
The consequences of MI extend beyond individual psychological distress, manifesting as depression, anxiety, PTSD, burnout, and even suicidal ideation. Professionally, MI undermines job satisfaction, work engagement, clinical decision-making, and teamwork, and is linked to increased intentions to leave the profession. These findings underscore the urgent need for systemic and organizational interventions, including leadership accountability, resource allocation, and the cultivation of a supportive ethical climate [
15,
40].
Although the current evidence for effective interventions targeting moral injury remains limited, several promising practices have emerged. Evidence-based resilience programs—including structured peer support, mindfulness-based interventions, and cognitive/behavioral strategies—have demonstrated potential in enhancing coping skills and reducing distress among healthcare workers and first responders. Organizational reforms, such as fostering a supportive ethical climate, transparent leadership communication, and participatory decision-making, are also critical in mitigating risk factors for moral injury [
8,
15,
40,
42,
43]. Peer support networks, regular ethics debriefings, and access to mental health services tailored to address guilt, shame, and betrayal are recommended as key components of comprehensive intervention strategies. Future intervention design should prioritize multi-level approaches that address both individual and systemic contributors to moral injury, utilize standardized outcome measures, and be adapted to diverse cultural and occupational contexts. Rigorous evaluation of these interventions is essential to establish their effectiveness and inform best practices moving forward [
8,
15,
40,
42].
Encouragingly, the review identifies adaptive coping strategies—such as social support, problem-solving, resilience training, and peer networks—as promising avenues for mitigation. However, the evidence base for effective, targeted interventions remains limited, highlighting a critical gap for future research. There is a pressing need for longitudinal studies and intervention trials to better understand the trajectory of MI and to develop evidence-based strategies for prevention and recovery.
This review has several limitations. Most included studies are cross-sectional, limiting causal interpretations and long-term outcome assessment. There is inconsistency in how MI is defined and measured, with varying tools (e.g., Moral Injury Symptom Scale vs. Moral Injury Events Scale) and overlap with constructs like burnout [
15,
21]. Research is predominantly focused on healthcare workers in high-income countries, reducing generalizability to other professions (e.g., first responders) and cultural contexts. Reliance on self-reported data and small sample sizes further limits robustness, and few studies adequately control for confounders. Additionally, there is a lack of standardized outcome measures and limited evidence for interventions, with most studies providing only preliminary findings. Notably, a substantial proportion of the included studies were conducted during the COVID-19 pandemic, which introduced unique stressors and circumstances that may not be directly comparable to pre-pandemic contexts [
5,
10,
19]
. This may limit the generalizability of findings across different time periods. To address these issues, we have added a subsection to the Methods detailing our structured approach for assessing methodological quality and synthesizing findings from diverse article types. Together, these findings highlight the multifaceted nature of MI and underscore the need for further research and intervention at multiple levels of the healthcare system.