Nº de DOI: 10.34896/RSI.2026.12.91.002
AUTHORS
- José Vinicio Cóndor Chano. General Practitioner. Affiliated with Hospital Carlos Andrade Marín. Graduate of Universidad UTE. Based in Quito, Ecuador. https://orcid.org/0009-0000-0402-6847
- Eliana Jazmin Tobar Pozo. General Practitioner. Affiliated with Clínicas Privadas del Ecuador. Graduate of Pontificia Universidad Católica del Ecuador. Based in Ibarra, Ecuador. https://orcid.org/0009-0001-3553-9300
- Ernesto Alfredo Proaño Triviño. General Practitioner. Affiliated with Clínicas Privadas del Ecuador. Graduate of Universidad Técnica de Manabí. Based in Chone, Ecuador. https://orcid.org/0009-0003-4698-3655
- Victor Andrés Veloz Cárdenas. General Practitioner. Affiliated with Hospital General Latacunga. Graduate of Universidad Regional Autónoma de los Andes. Based in Latacunga, Ecuador. https://orcid.org/0009-0007-6809-7311
- Edgar Sebastián Arce Jácome. General Practitioner. Affiliated with Hospital Universitario Católico de Cuenca. Graduate of Universidad Católica de Cuenca. Based in Cuenca. https://orcid.org/0009-0005-2864-2974
ABSTRACT
Objective: To synthesize published evidence on the prevalence, time course, risk factors, functional consequences, screening and treatment of PTSD and post-traumatic stress symptoms (PTSS) after musculoskeletal injury.
Methods: A structured search of publicly indexed sources was carried out in June 2026. Peer-reviewed journal articles that reported PTSD or PTSS in people with orthopedic or musculoskeletal injuries, or in mixed injury cohorts containing relevant data, were eligible. Only articles whose authors, bibliographic details and digital object identifier (DOI) could be verified were retained. Twenty-three articles published between 2004 and 2026 were included and synthesized narratively because of marked clinical and methodological heterogeneity.
Results: Reported prevalence varied widely with population, instrument and timing. A pooled estimate of about one quarter of adults with acute orthopedic trauma, a pooled estimate of about one in seven children, and single-cohort figures ranging from roughly 10% to more than 50% were found. Symptoms were typically highest in the first weeks and declined in most patients, but a substantial minority showed persistent or delayed-onset symptoms. The most consistent correlates were uncontrolled pain, prior psychiatric history, early post-injury distress, road traffic or work-related mechanisms, and multiple injuries; objective injury severity was an inconsistent predictor. PTSD and PTSS were associated with more pain, pain catastrophizing, fear of movement, poorer function and quality of life, and reduced ability to return to work. Early, brief screening tools and stepped collaborative care were the most developed responses, with modest and still limited evidence of benefit.
Conclusions: PTSD after musculoskeletal injury is common enough, and consequential enough, to warrant routine attention within orthopedic and trauma care. Repeated screening during the first months, aggressive pain control, and referral pathways to collaborative or trauma-focused psychological care are reasonable steps, but high-quality trials in musculoskeletal populations are still needed.
KEY WORDS
Post-traumatic stress disorder, musculoskeletal injuries, orthopedic trauma, fractures, pain, screening, systematic review.
RESUMEN
Objetivo: Sintetizar la evidencia publicada sobre la prevalencia, evolución temporal, factores de riesgo, consecuencias funcionales, cribado y tratamiento del trastorno de estrés postraumático (TEPT) y de los síntomas de estrés postraumático (SEPT) tras una lesión musculoesquelética.
Métodos: En junio de 2026 se realizó una búsqueda estructurada en fuentes indexadas de acceso público. Se seleccionaron artículos de revistas revisadas por pares que informaban sobre TEPT o SEPT en personas con lesiones ortopédicas o musculoesqueléticas, o en cohortes de lesiones mixtas que contenían datos relevantes. Solo se incluyeron aquellos artículos cuyos autores, datos bibliográficos e identificador de objeto digital (DOI) pudieron verificarse. Se incluyeron veintitrés artículos publicados entre 2004 y 2026, sintetizados de forma narrativa debido a la marcada heterogeneidad clínica y metodológica.
Resultados: La prevalencia notificada varió considerablemente según la población, el instrumento utilizado y el momento de la evaluación. Se halló una estimación combinada de aproximadamente una cuarta parte de los adultos con traumatismo ortopédico agudo y de uno de cada siete niños, con cifras en cohortes individuales que oscilaban entre cerca del 10 % y más del 50 %. Los síntomas solían alcanzar su punto máximo en las primeras semanas y disminuían en la mayoría de los pacientes; sin embargo, una minoría considerable presentó síntomas persistentes o de aparición tardía. Los factores asociados más constantes fueron el dolor no controlado, los antecedentes psiquiátricos, el malestar emocional temprano tras la lesión, los mecanismos de lesión relacionados con accidentes de tráfico o laborales, y las lesiones múltiples; la gravedad objetiva de la lesión resultó ser un predictor inconsistente. El TEPT y los SEPT se asociaron con mayor dolor, catastrofización del dolor, miedo al movimiento, peor funcionalidad y calidad de vida, y una menor capacidad para reincorporarse al trabajo. Las herramientas de cribado temprano y breve, así como la atención colaborativa escalonada, fueron las respuestas más desarrolladas, aunque la evidencia sobre sus beneficios es modesta y aún limitada.
Conclusiones: El TEPT tras una lesión musculoesquelética es lo suficientemente frecuente y tiene consecuencias lo bastante importantes como para justificar su atención sistemática en el ámbito de la traumatología y la ortopedia. El cribado periódico durante los primeros meses, el control intensivo del dolor y las vías de derivación a atención psicológica colaborativa o centrada en el trauma constituyen medidas razonables; no obstante, sigue siendo necesaria la realización de ensayos de alta calidad en poblaciones con lesiones musculoesqueléticas.
PALABRAS CLAVE
Trastorno de estrés postraumático, lesiones musculoesqueléticas, traumatismo ortopédico, fracturas, dolor, cribado, revisión sistemática.
INTRODUCTION
Fractures, dislocations, ligament and nerve injuries and other musculoskeletal injuries are usually managed as physical problems. They are also sudden, painful and frightening events that can trigger post-traumatic stress disorder (PTSD), a condition that may go unrecognized in orthopedic settings and that can slow physical recovery.
OBJECTIVE
The objective of this systematic review was therefore to summarize the available journal evidence on PTSD and PTSS after musculoskeletal injury. The specific questions were:
1. What is the prevalence of PTSD or PTSS after musculoskeletal injury, and how does it change over time?
2. Which injury-related, psychological and sociodemographic factors are associated with PTSD or PTSS?
3. What are the consequences of PTSD or PTSS for pain, function, quality of life, work and use of health services?
4. How have screening and treatment approaches been evaluated, and what do they show?
5. Do children, older adults and people with nerve injuries differ from the general adult orthopedic trauma population?
METHODOLOGY
Design and reporting framework:
This was a systematic review with a narrative synthesis. The research question was defined at the outset, the eligibility criteria and data items were fixed before final selection, and the structure of the report follows the main principles of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement where they could be applied. The protocol was not registered. Deviations from a full PRISMA-compliant review are stated in this section and in the limitations paragraph of the Discussion, because they affect how far the findings can be generalized.
Eligibility criteria:
Articles were eligible if they met all of the following conditions: (1) published in a peer-reviewed scientific journal; (2) written in English; (3) reported PTSD, PTSS or a closely related post-traumatic stress outcome in people with a musculoskeletal or orthopedic injury, or in an injury cohort in which fractures or orthopedic patients were a substantial component, or synthesized such studies; (4) provided enough information to extract at least one of the following: prevalence, time course, risk factors, consequences, screening performance or treatment effect; and (5) had a DOI, with authors, journal, year, volume and page or article number that could be confirmed against a publisher, repository or indexing record.
The population of interest was people of any age with fractures, dislocations, limb trauma, pelvic injury, peripheral nerve injury of the limbs or other injuries managed by orthopedic or trauma teams. Injuries were included regardless of mechanism, so that road traffic collisions, falls, occupational accidents, violence and low-energy fragility fractures could all be examined. Burns, isolated traumatic brain injury and purely non-traumatic musculoskeletal conditions were outside the scope of the review.
Conference abstracts, trial registry entries, study protocols, news items, theses, guideline summaries without an identifiable journal publication, and documents without a DOI were excluded. Articles that addressed pain or psychological factors in musculoskeletal injury without reporting a post-traumatic stress outcome, such as experimental pain models, were also excluded. Articles whose bibliographic details or DOI could not be confirmed were excluded even when their abstracts were informative, because an unverifiable reference cannot be reported reliably.
Information sources and search strategy:
Searches were carried out in June 2026 through a general web search interface that indexes publisher sites, open-access repositories such as PubMed Central, journal directories and institutional repositories. Bibliographic databases such as MEDLINE, Embase, PsycINFO and the Cochrane Central Register of Controlled Trials were not searched directly. The search used combinations of the following terms: «posttraumatic stress disorder», «PTSD», «post-traumatic stress symptoms», «orthopedic trauma», «orthopaedic trauma», «musculoskeletal injury», «fracture», «limb injury», «hip fracture», «hand injury», «peripheral nerve injury» and «sport injury». Additional searches addressed screening («PTSD Checklist», «screening orthopedic trauma»), treatment («cognitive behavioral therapy», «collaborative care», «psychological intervention»), and functional outcomes («pain», «return to work», «disability»). No date limit was applied.
Because the search was iterative and conducted through a general search interface rather than a bibliographic database, an exact count of records identified and screened cannot be reported, and a PRISMA flow diagram with record-level numbers would not be accurate. The final set of included articles is reported in full in the Results section and the reference list.
Study selection and data extraction:
Titles and abstracts were screened against the eligibility criteria and full pages were consulted for candidate articles. For every included article the following data were extracted into a standard template: authors, year, journal, design, setting, population and sample size, injury type, PTSD instrument or diagnostic method, time points, prevalence or incidence, risk factors or predictors, associations with pain, function, quality of life, work or health service use, and intervention details and effects where relevant. Numerical values were transcribed as reported by the original authors. When a value could not be confirmed from the retrieved text, it was not used.
Appraisal of the evidence:
No formal risk-of-bias instrument was applied. Instead, each article was characterized descriptively by design (systematic review or meta-analysis, guideline, prospective cohort, cross-sectional or retrospective study, randomized trial), sample size, how PTSD was measured (diagnostic interview, validated self-report scale, or screening tool), length and completeness of follow-up, and the extent to which the population was specific to musculoskeletal injury. These characteristics were used to weigh the findings during synthesis. In general, more weight was given to meta-analytic and large prospective evidence and less to small or single-center samples, to cross-sectional studies when time course was discussed, and to retrospective comparisons when treatment effects were discussed.
Synthesis:
The included studies differed too much in populations, instruments, cut-off scores and time points to allow a new quantitative pooling. Findings were therefore grouped by question and described narratively. Pooled estimates that appear in the Results are those reported by the authors of the included meta-analytic or systematic reviews, and they are labelled as such. Prevalence figures from individual studies are reported with the instrument and time point wherever these were available, since the same patient group can appear to have very different prevalence depending on whether early acute stress reactions, probable PTSD or a clinician-confirmed diagnosis is being counted.
Terminology was handled as follows. «PTSD» is used when the authors reported a diagnosis or a validated threshold for probable PTSD. «PTSS» is used when the authors reported symptom scores or symptom-based positivity without a diagnosis. Where studies report PTSD in mixed injury cohorts, this is stated, so that readers do not mistake a general trauma finding for one that is specific to fractures or limb injury.
RESULTS
Characteristics of the included studies:
Twenty-three journal articles published between 2004 and 2026 met the eligibility criteria. They comprised two systematic reviews, one systematic review and meta-analysis, one narrative literature review, one clinical practice guideline summary, twelve cohort studies of adults or older adults, one randomized controlled trial, one retrospective comparative analysis of a psychological intervention, and four observational studies using cross-sectional, screening-comparison or short follow-up designs. Studies were conducted in North America, Europe, Australia and Asia3. Sample sizes of the primary studies ranged from 30 children in a single-center fracture series to more than 4,000 adults in a multicenter Dutch trauma cohort.
Eight of the cohort and observational studies were specific to orthopedic or musculoskeletal injury, such as fractures, orthopedic trauma admissions or limb injury. Others enrolled mixed injury populations in which musculoskeletal injury was common but not exclusive. Three studies focused on older adults with hip fracture. Table 1 summarizes each article.
Prevalence:
Prevalence estimates ranged from roughly 10% to more than 50%, and the range reflects methods more than biology. The highest figure came from a clinic-based questionnaire study in which 295 of 580 orthopedic trauma patients (51%) met the criterion for PTSD on the Revised Civilian Mississippi Scale, a self-report instrument2. The only meta-analytic estimate in adults, which pooled studies of acute orthopedic trauma to the appendicular skeleton or pelvis from searches run through June 2015, was 26.6% (95% CI 19.0% to 35.9%) for PTSD, and 16.8% (95% CI 9.0% to 29.4%) for PTSD and depression together in the six studies that measured both4.
Studies of hospitalized fracture patients and mixed trauma cohorts produced figures of a similar order. Among 204 inpatients with traumatic fractures in China, 30.39% screened positive for PTSS on a self-rating scale21. In an Australian cohort of 433 hospitalized trauma patients assessed 12 to 14 months after injury, 21.2% met the threshold for probable PTSD12. In a US national study of 2,931 injury survivors from 69 hospitals, about 23% had PTSD symptoms at 12 months9. In contrast, in a larger Dutch cohort of 4,239 adult trauma patients, PTSS were present in 10% at 12 months15. Differences in how the cohorts were assembled, the severity of injury, the instrument and the cut-off all plausibly contribute to this spread.
A small number of studies described how time of assessment changes the picture. A UK cohort of 125 adults admitted with acute musculoskeletal injuries found that almost all patients scored above the PTSS threshold at baseline (97.6%), while the proportion fell to 26.4% at 3 months and 17.6% at 6 months18. This pattern shows that early symptom scores in the first days after injury are very common, and probably reflect a normal acute stress response rather than established PTSD. In a screening comparison from a US Level I trauma center, a nine-item yes/no instrument given during admission (mean 7.1 days after injury) flagged 40.0% of 45 patients as being at risk, whereas the PTSD Checklist for DSM-5 given at follow-up (mean 154 days after injury) was positive in 18.9% of 285 patients; for gunshot wounds the corresponding figures were 80.0% and 39.1%22. The two samples were different, so these figures should not be read as a head-to-head test, but they illustrate how timing changes the number.
Estimates were lower in some special groups. Among 456 adults aged 60 or older undergoing surgical hip fracture repair, none met full PTSD criteria at 4 or 12 weeks, although partial PTSD was found in 7.4% at 12 weeks and fear of falling was reported by 58.5%11. In children, a systematic review of ten studies including 1,356 patients with a mean age of 10.2 years reported a pooled prevalence of 15% (95% CI -5.1% to 34.2%), roughly one in seven, with self-reported prevalence averaging 25% against 8% when reported by parents1. A literature review of mental health in orthopedic trauma reported PTSD in over half of victims in some of the studies it covered, which illustrates how widely the literature varies across populations and methods6.
Time course:
The available data suggest that post-traumatic stress after musculoskeletal injury is highest early and declines in most people, but not in all. In a UK multicenter cohort of 668 injured adults, post-traumatic distress peaked at one month and remained above baseline levels through 12 months13. A systematic review of upper limb peripheral nerve injury found evidence of PTSD at one month in three studies, with a decrease over time7. In a prospective cohort of 200 patients with musculoskeletal injuries, general psychological disturbance measured with the General Health Questionnaire increased from 11% before injury to 46% at 2 months and then fell to 22% at 6 months8. In a cohort of 113 patients with severe orthopedic trauma and 61 uninjured controls, however, ongoing disability, pain and PTSD symptoms were still evident at both 1 and 2 years, with little or no improvement between the two assessments10.
A two-year Korean cohort of 1,014 hospitalized injured patients separated early-onset from delayed-onset PTSD, defined as onset beyond six months. The two groups had different predictors, so the delayed form is not simply a more severe version of the early form17. In the clinic-based orthopedic series, a longer interval since injury was associated with a PTSD-range score, a cross-sectional finding that could reflect either true persistence or greater care-seeking among those with symptoms2.
Risk factors and correlates:
Psychological history and early response. A history of psychiatric disorder was among the most consistent correlates. In the UK cohort, previous psychiatric diagnoses predicted post-traumatic distress, and in the Korean cohort early-onset PTSD was predicted by prior psychiatric disorders, previous traumatic events, acute stress disorder or subthreshold acute stress disorder, and heightened anxiety13,17. In the Dutch cohort, preinjury psychological difficulties, physical frailty, unemployment, female sex and lower education were prognostic factors for poor mental health outcomes15. In the US national study, early post-injury emotional distress and physical pain, pre-injury depression, intensive care admission, benzodiazepine use and intentional injury were independently associated with PTSD symptoms at 12 months9. In older adults with hip fracture, baseline stress and depressive symptoms were significant predictors of later PTSD symptoms11.
Injury mechanism and severity. Mechanism of injury repeatedly mattered. Road traffic accidents significantly increased the risk of PTSS in the UK musculoskeletal cohort18. Traffic injuries, injuries from being struck by objects, and three or more injuries predicted distress in another UK cohort13. In the Dutch cohort, traffic and workplace accidents carried a higher risk than sports injuries15. Evidence for injury severity was mixed. Higher Injury Severity Scores and higher summed extremity Abbreviated Injury Scores were associated with PTSD in the clinic series, but neither variable discriminated well between patients with and without PTSD2. Greater injury severity predicted the delayed-onset form in the Korean cohort17. In the Australian orthopedic cohort, injury severity and type did not predict outcome, although lower limb fractures were linked to more pain and poorer physical outcomes than fractures elsewhere10. In children, post-traumatic stress reaction scores at one month were higher after lower than upper extremity fracture, but the difference was not statistically significant in a sample of 3023. In upper limb nerve injury, patients with combined nerve injuries had higher PTSD levels at one month than those with isolated injuries7.
Pain. Pain was the most consistent clinical correlate. Higher pain at discharge from a Level I trauma center was associated with PTSD at one year after adjustment for demographic factors and baseline psychological symptoms (OR 1.4, p = 0.03) 14. In Chinese fracture inpatients, increased pain was associated with positive PTSS (OR 3.34, 95% CI 1.82 to 6.11), whereas physical activity during the fracture was protective (OR 0.63, 95% CI 0.45 to 0.88) 21. In the Australian trauma cohort, PTSS were strongly associated with pain severity and pain-related disability12. Earlier work in an Australian orthopedic trauma cohort had already shown that pain, anxiety, depression and PTSD symptoms were the strongest predictors of outcome on most measures10.
Demographic and social factors. Female sex was a predictor in two cohorts13,15. Higher education was associated with delayed-onset PTSD in the Korean cohort, whereas lower education was a prognostic factor in the Dutch cohort, which suggests that education may act differently depending on the outcome definition, population or health system15,17. The US national study reported disparities between uninsured and insured white survivors9. In hip fracture patients aged 65 or older, frailty predicted depressive and anxious symptoms but did not predict PTSS16.
Consequences of post-traumatic stress:
Post-traumatic stress was repeatedly linked to poorer physical and social recovery. The clinical practice guideline summary states that PTSD is associated with increased pain, decreased functional outcomes, lower quality of life and a reduced ability to return to activity or work5. In the cohort of 200 patients with musculoskeletal injuries, posttraumatic disturbance correlated strongly with impaired function on the Short Form-36, the Sickness Impact Profile and the Musculoskeletal Function Assessment8. In the Australian cohort of orthopedic trauma patients, significant disability remained across all physical and mental health domains of the Short Form-36 two years after injury, and PTSD symptoms were among the strongest predictors of poor outcome10.
The mechanism most clearly described was the interaction of stress symptoms with pain. In the 433-patient cohort, PTSS were associated with pain catastrophizing and kinesiophobia (fear of movement), and depression, reduced self-efficacy and fear of movement partially mediated the relationship between PTSS and functional impairment12. In a systematic review of upper limb nerve injury, two of the included studies found a statistically significant correlation between early PTSD and reduced function at 12 months or more7.
Evidence on health service use and length of stay was sparse. In the Chinese fracture cohort, serious injury, critical illness during hospitalization and undergoing two surgeries were risk factors for a longer hospital stay, while the authors argued that strengthened care of both physical and mental health was needed to reduce PTSS21.
Special populations:
Older adults. Three studies examined hip fracture. One found no full PTSD and low rates of partial PTSD, but high rates of fear of falling11. In a Dutch cohort of 570 patients aged 65 or older, psychological distress, defined broadly to include anxiety, depression and posttraumatic stress, was present in 36% at one week and 31% at one year16. A retrospective analysis of 50 older patients with hip fracture and PTSD found that those who received structured perioperative psychological support, including counseling and relaxation training, had larger reductions in PTSD symptom severity at postoperative days 7 and 14 than controls20.
Children. The pediatric systematic review suggested that about one in seven children may develop PTSD after orthopedic trauma and that parents under-report symptoms compared with children’s own reports1.
Screening and treatment:
Screening. Timing and instrument influenced who was identified. The nine-item yes/no screen applied during admission identified a higher proportion of at-risk patients than the PTSD Checklist applied months later, and the authors argued that an early screen could allow intervention before the one-month point at which a formal DSM-5 diagnosis becomes possible22. A single item asking whether the emotional problems caused by the injury had been more difficult than the physical ones showed a fair ability to identify PTSD in the clinic series2. A six-month prediction model based on injury and psychological variables in the UK musculoskeletal cohort showed excellent discrimination (area under the curve 0.91)18. The clinical practice guideline summary recommends evaluating psychosocial factors including PTSD, with moderate strength of recommendation5.
Treatment. Treatment evidence in musculoskeletal populations was limited. A randomized trial of technology-enhanced stepped collaborative care in 121 injured trauma survivors who screened positive for PTSD risk reported modest reductions in PTSD symptoms over six months, with 45% of the intervention group achieving clinically meaningful improvement compared with 30% of controls, while using less provider time than earlier collaborative care models19. A literature review on mental health in orthopedic trauma identified cognitive behavioral therapy, virtual reality therapy and mindfulness-based approaches as interventions that reduced psychiatric symptoms and improved functional recovery, and called for collaborative care models involving surgeons, mental health specialists and social support6. The perioperative psychological intervention in older hip fracture patients is the only musculoskeletal-specific intervention analysis identified in the sample, and it was retrospective20.
DISCUSSION
This review brought together twenty-three journal articles on post-traumatic stress after musculoskeletal injury. Four findings stand out. First, PTSD is common: the only meta-analytic estimate in adults with acute orthopedic trauma was roughly one in four, a figure that is compatible with single-center results in fracture patients and with the children’s estimate of about one in seven1,4,21. Second, post-traumatic stress is dynamic. Almost all patients report some stress symptoms in the first days, most settle over weeks to months, and a minority keep or develop symptoms later10,13,17,18. Third, pain, prior psychiatric history, early distress and the circumstances of the accident are more consistent correlates than the anatomical severity of the injury9,13,14,17. Fourth, the consequences reach beyond mood: post-traumatic stress is tied to pain, catastrophizing, fear of movement, disability, lower quality of life and work outcomes5,10,12.
The spread between 10% and more than 50% should not be read as a sign that the underlying problem is poorly defined. It more probably reflects at least four methodological sources of variation. The first is the instrument. Self-report scales and screening tools are designed to be sensitive and tend to flag more people than diagnostic interviews. The highest prevalence in this review came from a clinic self-report scale, and an early screening tool flagged more patients than a later symptom checklist2,22. The second is timing. Because the formal DSM-5 diagnosis requires symptoms to last for at least one month, anything measured earlier is better described as acute stress, and the near-universal baseline positivity in one cohort illustrates this point18. The third is selection. Cohorts of patients attending orthopedic follow-up clinics are likely to be enriched for people who remain symptomatic, whereas cohorts recruited at admission include everyone who survived the injury, and loss to follow-up in either design can be related to mental health2,14. The fourth is the population. Studies of older adults with low-energy hip fractures found little or no full PTSD, whereas cohorts including road traffic collisions and violence found higher rates9,11,18.
For clinicians and for researchers, the practical lesson is that a single prevalence figure for «PTSD after fracture» is unlikely to be useful. A more informative statement is conditional: what proportion of which patients screen positive, on which tool, at what interval after injury.
Pain was the correlate that appeared most often across settings. Higher pain at hospital discharge was associated with PTSD one year later, even after accounting for baseline psychological symptoms14. In fracture inpatients, increased pain was associated with a more than threefold increase in the odds of positive PTSS21. In a large Australian trauma cohort, PTSS were strongly associated with pain severity, pain-related disability, catastrophizing and kinesiophobia, and depression, self-efficacy and fear of movement partially mediated the association between stress symptoms and functional impairment12.
These findings fit with the idea that pain and PTSD can maintain each other. Pain can serve as a continual reminder of the injury, and heightened arousal and avoidance in PTSD can increase pain perception and reduce activity. Observational designs cannot prove the direction of effect, and the available studies did not test causal pathways. The data are, however, consistent with the pathway described in the Australian cohort, in which fear of movement and reduced self-efficacy sit between psychological distress and disability12. Physical activity during the fracture was protective in the Chinese cohort, although this variable may partly reflect better health in those who stayed active21.
If pain and PTSD are linked in this way, there are two practical implications. Pain control during the acute period should be treated as part of mental health prevention and not only as an end in itself, and persistent pain at follow-up should prompt a question about traumatic stress symptoms and not only a search for mechanical causes.
A striking feature of the evidence is how weakly objective injury severity predicts PTSD. Among 580 orthopedic trauma patients, higher Injury Severity Scores were associated with PTSD, but none of the injury variables discriminated well between those with and without the disorder2. In a cohort of patients with orthopedic injuries, neither injury severity nor injury type predicted outcome, whereas pain, anxiety, depression and PTSD symptoms did10. Only the delayed-onset form of PTSD in a Korean cohort was tied to greater injury severity, while early-onset PTSD was predicted by previous psychiatric disorders, previous traumatic events and acute stress disorder17.
The more consistent injury-related predictors were not anatomical. Road traffic collisions, injuries from being struck, multiple injuries, intentional injury and intensive care admission were linked to higher risk9,13,15,18. These features are better indicators of perceived threat, loss of control and the sense of having come close to death than of fracture complexity. This interpretation is consistent with the finding in the first orthopedic series that the item contrasting emotional and physical difficulty was the best simple discriminator2. It also has a practical consequence: a patient with a simple fracture after a frightening road traffic collision may be at higher risk than a patient with a more complex fracture after a low-energy fall, and screening based on injury classification alone would miss this.
The time course data support a staged approach to assessment. Symptoms measured in the first days after injury are nearly universal and are therefore of little value for diagnosis, although they may have value for risk stratification18,22. Distress peaked at about one month in a UK cohort but remained above baseline through 12 months13. A subset develops symptoms only after six months, with different predictors17. In an Australian cohort, pain and PTSD symptoms were still evident two years after injury, with little change between years one and two10.
Taken together, these findings argue against a single screening occasion. A reasonable pattern would be an early brief risk screen during admission, followed by a structured symptom assessment at about one month, and a further assessment at three to six months, particularly when pain or functional recovery is not progressing as expected. The evidence in this review does not establish the optimum schedule. It does show that the answer will differ depending on when the patient is asked.
Older adults: Hip fracture patients differ from the typical trauma population. In older adults after surgical repair, full PTSD was not found and partial PTSD was uncommon, but fear of falling was highly prevalent, and a broader measure of psychological distress that included anxiety and depression was present in about a third of patients up to a year after fracture11,16. Frailty predicted depression and anxiety but not PTSS in the latter study16. One interpretation is that fear and anxiety, not classical re-experiencing of an event, may be the dominant emotional response to a fall in later life, and that standard PTSD tools may under-detect it. Alternatively, cognitive impairment, delirium, and attrition from follow-up among frail patients may reduce detection. The retrospective analysis in older hip fracture patients with PTSD suggests that those with symptoms exist and may benefit from structured perioperative psychological support, but the design does not allow a causal conclusion20.
Children: The pediatric review is a reminder that who reports matters. Children reported higher rates of PTSD symptoms than their parents did1. A clinician who asks only the parent about a child’s emotional response after a fracture may therefore underestimate the problem. The wide confidence interval around the pooled estimate and the small size of single studies such as the 30-patient fracture series mean that more research is needed, particularly on differences between upper and lower limb fractures23.
Nerve injury: In upper limb peripheral nerve injury, early PTSD was associated with later functional loss, and combined nerve injuries were associated with higher PTSD levels7. Nerve injuries often involve high-energy trauma, a long and uncertain recovery, and visible changes in the hand or arm, so psychological vulnerability is plausible. The authors of the review stressed the poor quality of the underlying studies, and the conclusions should be regarded as hypothesis-generating.
Several practical steps follow from the evidence, although none rests on a definitive trial in musculoskeletal patients.
The first step is to make post-traumatic stress part of the orthopedic assessment. Current practice guidance recommends evaluating psychosocial factors, including PTSD, anxiety, depression and premorbid psychiatric conditions, with a moderate strength of recommendation5. The data on risk factors suggest that clinicians should ask early about previous psychiatric history, previous traumatic events, mechanism of injury and the patient’s own account of how frightening the event was13,15,17.
The second step is to use a brief, structured tool rather than clinical impression. A nine-item yes/no screen applied during admission identified at-risk patients early in the hospital stay22. A prediction model that incorporated injury and psychological variables reached excellent discrimination at six months in one cohort, although such models need external validation before routine use18.
The third step is to treat pain aggressively and early, in recognition of its association with later PTSD and depression14,21.
The fourth step is to build referral pathways: A randomized trial of technology-enhanced stepped collaborative care in injured survivors who screened positive for PTSD risk reported modest symptom reductions19. The trial was not restricted to musculoskeletal injury and the effect was modest, but it shows that stepped, low-burden models are feasible in trauma settings. Cognitive behavioral therapy, virtual reality therapy and mindfulness-based approaches have been proposed in orthopedic trauma populations, although the evidence from musculoskeletal-specific randomized trials remains thin6.
Five gaps emerge. There are few randomized trials of PTSD prevention or treatment in populations selected for musculoskeletal injury; the only trial in this review enrolled mixed injury survivors19. The effect of treating PTSD on physical outcomes such as union, function and return to work has not been tested directly. Injury subtypes are rarely analyzed separately, so patients with hand injuries, amputations, pelvic fractures and sports injuries are poorly represented. Measurement is inconsistent, with different tools, cut-offs and follow-up schedules making cross-study comparison hard. Finally, most cohorts followed patients for 12 months or less, and the long-term course beyond two years is poorly described10,17.
The strengths of the review are a clear question, explicit eligibility criteria, verification of every reference, and numerical values taken as reported by the original authors.
CONCLUSIONS
Post-traumatic stress is a frequent and clinically important complication of musculoskeletal injury. Depending on the population, instrument and timing, between roughly one in ten and one in two patients report symptoms, with the best available adult pooled estimate at about one in four and the pediatric estimate at about one in seven. Symptoms are typically highest in the first weeks and decline, but a meaningful minority show persistent or delayed-onset symptoms that are linked to pain, catastrophizing, fear of movement, reduced function and poorer work outcomes.
The correlates that appear most consistently are uncontrolled pain, previous psychiatric history, early distress, and the circumstances of the accident, such as road traffic collisions, rather than anatomical severity alone. Children, older adults and people with nerve injury may present differently from young and middle-aged adults, and assessment may need to adapt to the patient and the reporter.
Orthopedic and trauma teams can act on this evidence now without waiting for definitive trials. Post-traumatic stress should be part of routine psychosocial assessment, brief screening should be considered during admission and repeated at about one month and at three to six months, pain should be managed as part of mental health prevention, and clear referral routes to collaborative or trauma-focused psychological care should exist. The evidence that such steps improve physical as well as psychological outcomes is limited and mostly indirect, and the review itself has important methodological limitations. Priority research needs are randomized trials in musculoskeletal-specific populations, standardized measurement and reporting, subgroup analyses by injury type, and longer follow-up.
REFERENCES
- Punski-Hoogervorst JL, Hoogervorst LA, Schoones JW, Avital A, de Witte PB. Prevalence of posttraumatic stress disorder in paediatric patients following orthopaedic trauma: a systematic review. J Child Orthop. 2026;20(2):103-10. doi:10.1177/18632521261419773
- Starr AJ, Smith WR, Frawley WH, Borer DS, Morgan SJ, Reinert CM, et al. Symptoms of posttraumatic stress disorder after orthopaedic trauma. J Bone Joint Surg Am. 2004;86(6):1115-21. doi:10.2106/00004623-200406000-00001
- Bryant R, O’Donnell M, Creamer M, McFarlane A, Clark C, Silove D. The psychiatric sequelae of traumatic injury. Am J Psychiatry. 2010;167(3):312-20. doi:10.1176/appi.ajp.2009.09050617
- Muscatelli S, Spurr SA, O’Hara NN, O’Hara LM, Slobogean GP. Prevalence of depression and posttraumatic stress disorder after acute orthopaedic trauma: a systematic review and meta-analysis. J Orthop Trauma. 2017;31(1):47-55. doi:10.1097/bot.0000000000000664
- Keizer BM, Wegener ST. Psychosocial risk factors after orthopaedic trauma. J Am Acad Orthop Surg. 2022;30(3):e307-12. doi:10.5435/JAAOS-D-21-00777
- Scott S, Brameier DT, Tryggedsson I, Suneja N, Stenquist DS, Weaver MJ, et al. Prevalence, resources, provider insights, and outcomes: a review of patient mental health in orthopaedic trauma. J Orthop Surg Res. 2024;19:538. doi:10.1186/s13018-024-04932-4
- Miller C, Peek AL, Power D, Heneghan NR. Psychological consequences of traumatic upper limb peripheral nerve injury: a systematic review. Hand Ther. 2017;22(1):35-45. doi:10.1177/1758998316679387
- Sutherland AG, Alexander DA, Hutchison JD. The mind does matter: psychological and physical recovery after musculoskeletal trauma. J Trauma. 2006;61(6):1408-14. doi:10.1097/01.ta.0000197562.34651.df
- Zatzick DF, Rivara FP, Nathens AB, Jurkovich GJ, Wang J, Fan MY, et al. A nationwide US study of post-traumatic stress after hospitalization for physical injury. Psychol Med. 2007;37(10):1469-80. doi:10.1017/S0033291707000943
- Ponsford J, Hill B, Karamitsios M, Bahar-Fuchs A. Factors influencing outcome after orthopedic trauma. J Trauma. 2008;64(4):1001-9. doi:10.1097/TA.0b013e31809fec16
- Kornfield SL, Lenze EJ, Rawson KS. Predictors of posttraumatic stress symptoms and association with fear of falling after hip fracture. J Am Geriatr Soc. 2017;65(6):1251-7. doi:10.1111/jgs.14771
- Giummarra MJ, Casey SL, Devlin A, Ioannou LJ, Gibson SJ, Georgiou-Karistianis N, et al. Co-occurrence of posttraumatic stress symptoms, pain, and disability 12 months after traumatic injury. Pain Rep. 2017;2(5):e622. doi:10.1097/PR9.0000000000000622
- Kendrick D, Baker R, Hill T, Beckett K, Coupland C, Kellezi B, et al. Early risk factors for depression, anxiety and post-traumatic distress after hospital admission for unintentional injury: multicentre cohort study. J Psychosom Res. 2018;112:15-24. doi:10.1016/j.jpsychores.2018.06.008
- Archer KR, Heins SE, Abraham CM, Obremskey WT, Wegener ST, Castillo RC. Clinical significance of pain at hospital discharge following traumatic orthopaedic injury: general health, depression, and PTSD outcomes at 1 year. Clin J Pain. 2016;32(3):196-202. doi:10.1097/AJP.0000000000000246
- de Munter L, Polinder S, Haagsma JA, Kruithof N, van de Ree CLP, Steyerberg EW, et al. Prevalence and prognostic factors for psychological distress after trauma. Arch Phys Med Rehabil. 2020;101(5):877-84. doi:10.1016/j.apmr.2019.10.196
- van de Ree CLP, de Munter L, Biesbroeck BHH, Kruithof N, Gosens T, de Jongh MAC. The prevalence and prognostic factors of psychological distress in older patients with a hip fracture: a longitudinal cohort study. Injury. 2020;51(11):2668-75. doi:10.1016/j.injury.2020.07.049
- Kang SG, Kim JW, Kang HJ, Jang H, Kim JC, Lee JY, et al. Differential predictors of early- and delayed-onset post-traumatic stress disorder following physical injury: a two-year longitudinal study. Front Psychiatry. 2024;15:1367661. doi:10.3389/fpsyt.2024.1367661
- Jadhakhan F, Falla D, Dallaway A. Predictors of post-traumatic stress symptoms after musculoskeletal trauma. PLoS One. 2026;21(5):e0348595. doi:10.1371/journal.pone.0348595
- Zatzick D, O’Connor SS, Russo J, Wang J, Bush N, Love J, et al. Technology-enhanced stepped collaborative care targeting posttraumatic stress disorder and comorbidity after injury: a randomized controlled trial. J Trauma Stress. 2015;28(5):391-400. doi:10.1002/jts.22041
- Liu YD, Ma JY, Wang JH, Sun T, Han ZY. Effect of perioperative psychological intervention on recovery in elderly hip fracture patients with comorbid post-traumatic stress disorder. World J Psychiatry. 2025;15(10):106705. doi:10.5498/wjp.v15.i10.106705
- Qi G, Dai X, Wang X, Yuan P, Li X, Qi M, et al. Epidemiological characteristics of post-traumatic stress symptoms and its influence on length of hospital stay in inpatients with traumatic fractures in Zunyi, China. AIMS Public Health. 2024;11(3):835-49. doi:10.3934/publichealth.2024042
- Nedder VJ, Breslin MA, Ho VP, Vallier HA. Prevalence of posttraumatic stress disorder in orthopedic trauma patients and a call to implement the Injured Trauma Survivor Screen as a prospective screening protocol in the United States. J Trauma Inj. 2024;37(1):67-73. doi:10.20408/jti.2023.0068
- Coskun HS, Usta MB, Say F, Yildirim AM. Factors affecting post-traumatic stress disorder in children with orthopedic injury. J Clin Investig Surg. 2020;5(1):24-8. doi:10.25083/2559.5555/5.1/24.28
APPENDICE
TABLE 1. Summary of the included articles (ordered by year of publication).
| Ref. | Design and population | Main post-traumatic stress findings |
| [2] | Questionnaire study, 580 orthopedic trauma patients | 51% met the PTSD-range criterion on the Revised Civilian Mississippi Scale; higher injury severity and longer time since injury were associated with PTSD but did not discriminate well |
| [8] | Prospective cohort, 200 patients with musculoskeletal injury | Psychological disturbance rose from 11% before injury to 46% at 2 months and 22% at 6 months, and correlated strongly with poorer function |
| [9] | Nationwide US cohort, 2,931 hospitalized injury survivors | About 23% reported PTSD symptoms at 12 months; early distress and pain were risk factors |
| [10] | Prospective cohort, 113 orthopedic trauma patients and 61 controls | Persistent disability, pain and PTSD symptoms at 1 and 2 years with little improvement |
| [3] | Longitudinal study of injury survivors | Described psychiatric sequelae of traumatic injury, including PTSD |
| [19] | Randomized trial, 121 injured survivors | Technology-enhanced stepped collaborative care gave modest PTSD symptom reduction |
| [14] | Prospective cohort, 213 orthopedic trauma patients | Higher pain at discharge was associated with PTSD at 1 year (odds ratio [OR] 1.4) and with depression (OR 3.3) |
| [4] | Systematic review and meta-analysis, adult acute orthopedic trauma | Pooled PTSD prevalence 26.6% (95% CI 19.0% to 35.9%); depression 32.6% |
| [11] | Prospective cohort, 456 adults aged 60 or older after hip fracture repair | No participant met full PTSD criteria at 4 or 12 weeks; partial PTSD 7.4% at 12 weeks |
| [12] | Cohort, 433 hospitalized trauma patients at 12 to 14 months | 21.2% had probable PTSD; PTSS were linked to pain, disability, catastrophizing and fear of movement |
| [7] | Systematic review, 6 studies, 245 patients with upper limb nerve injury | PTSD at 1 month that decreased over time; early PTSD correlated with reduced function at 12 months or more |
| [13] | Prospective cohort, 668 injured adults | Post-traumatic distress peaked at 1 month and stayed above baseline at 12 months |
| [15] | Multicenter cohort, 4,239 adult trauma patients | Acute traumatic stress in 13% at 1 week; PTSS in 10% at 12 months |
| [16] | Cohort, 570 patients aged 65 or older with hip fracture | Psychological distress in 36% at 1 week and 31% at 1 year; frailty did not predict PTSS |
| [23] | Observational study, 30 children with displaced extremity fracture | Higher PTSD reaction scores after lower than upper extremity fracture, not statistically significant |
| [5] | Clinical practice guideline summary | PTSD linked to more pain, worse function and quality of life, and reduced return to activity or work |
| [6] | Literature review of orthopedic trauma and mental health | Reported high prevalence of mental health disorders and discussed interventions and barriers |
| [22] | Screening comparison, 285 patients (PTSD Checklist for DSM-5) and 45 patients (Injured Trauma Survivor Screen) | Early screen flagged 40.0% as at risk, versus 18.9% positive on the later checklist |
| [21] | Observational study, 204 fracture inpatients | 30.39% had positive PTSS; pain increased and physical activity decreased the odds |
| [17] | Cohort, 1,014 hospitalized injured patients followed for 2 years | Early-onset and delayed-onset PTSD had different predictors |
| [20] | Retrospective analysis, 50 older hip fracture patients with PTSD | Perioperative psychological intervention was followed by larger early reductions in symptom severity |
| [18] | Cohort, 125 adults with acute musculoskeletal injury | PTSS present in 97.6% at baseline, 26.4% at 3 months and 17.6% at 6 months |
| [1] | Systematic review, 10 studies, 1,356 children | Pooled PTSD prevalence 15%; self-report higher than parent report |
Source: Prepared by the authors.
