ABSTRACT
Background and Objective
Delirium may occur in patients with traumatic brain injury (TBI), but this may be a marker for subsequent cognitive decline post‐TBI. We aimed to compare cognitive outcomes and all‐cause mortality between TBI patients with and without delirium.
Methods
We conducted a retrospective cohort study using the TriNetX global federated research network. We compared TBI patients who developed delirium within 14 days of TBI diagnosis with those who did not. The primary outcome was the 2‐year composite risk of dementia, vascular dementia and Alzheimer's disease. Secondary outcomes included dementia, vascular dementia, Alzheimer's disease and all‐cause mortality. Pre‐specified subgroup analyses were performed in female patients, older adults (≥ 75 years), and individuals with substance use.
Results
TBI patients with delirium (n = 2564; mean age, 66.6 ± 19.4 years; 59.4% male), when compared with TBI patients without delirium (n = 667,950; mean age, 48.0 ± 25.1 years; 60.3% male), had more baseline comorbidities, including psychiatric disease, cerebrovascular disease and ischemic heart disease. After propensity score matching, TBI patients with delirium, compared with those without delirium, had a higher 2‐year risk of the primary composite outcome (HR, 2.47; 95% CI, 1.84–3.31), as well as dementia (HR, 2.63; 95% CI, 1.92–3.60), vascular dementia (HR, 2.76; 95% CI, 1.45–5.24) and Alzheimer's disease (HR, 2.23; 95% CI, 1.21–4.09).
Conclusion
Delirium following TBI is associated with an increased risk of dementia and Alzheimer's disease. Delirium post‐TBI may be a marker or a contributing factor to long‐term cognitive decline in TBI patients.
Keywords: cognitive outcomes, delirium, traumatic brain injury
In this global federated TriNetX cohort, traumatic brain injury patients who developed delirium within 14 days were compared with those without delirium. After 1:1 propensity‐score matching, delirium was associated with higher two‐year risks of composite cognitive decline, dementia, vascular dementia, Alzheimer's disease and all‐cause death. The strongest association was observed for the composite cognitive outcome (HR 2.47; 95% CI 1.84–3.31). Delirium after traumatic brain injury may identify patients needing closer cognitive monitoring and follow‐up, particularly older adults at risk.
1. Introduction
Traumatic brain injury (TBI) results when external physical forces impact the head with sufficient intensity to cause damage to the brain [1]. TBI affects more than 50 million individuals annually and accounts for approximately 30%–40% of all injury‐related deaths worldwide [2, 3].
TBI is increasingly recognized as a chronic health condition with evolving neurological consequences that extend far beyond the acute injury period [4]. Emerging evidence demonstrates that TBI is associated with a substantially increased risk of neurodegenerative diseases, including Alzheimer's disease, Parkinson's disease, dementia with Lewy bodies and frontotemporal dementia [4, 5, 6, 7].
Delirium superimposed on dementia is highly prevalent among hospitalized older patients and is associated with significant adverse outcomes, including accelerated cognitive decline, increased risk of incident dementia, hospitalization and mortality [8]. Furthermore, cohort studies suggest that delirium itself is a strong predictor of subsequent dementia and death in older hospitalized populations [8, 9].
To the best of our knowledge, no prior study has directly compared cognitive outcomes between TBI patients with and without associated delirium. Therefore, we aimed to address this gap by comparing these two groups and evaluating the 2‐year risk of cognitive decline, including dementia, vascular dementia and Alzheimer's disease, as well as all‐cause mortality.
2. Methods
2.1. Study Design
This retrospective observational study was conducted using the TriNetX Research Network (TriNetX, Cambridge, MA, USA), a global federated health research network that provides access to anonymized electronic medical records from over 130 healthcare organizations, including academic medical centres, community hospitals and outpatient clinics. The network includes more than 300 million individuals, primarily in the United States (U.S.).
Available data include demographics, diagnoses coded using the International Classification of Diseases, Ninth and Tenth Revisions, Clinical Modification (ICD‐9‐CM and ICD‐10‐CM) codes and medications coded with Veteran Affairs (VA) codes or RxNorm classification systems, depending on the contributing institution. Further information about the network is available online (https://trinetx.com/data‐sets‐analytics/).
TriNetX operates in compliance with the Health Insurance Portability and Accountability Act (HIPAA) and US federal law. All patient data are de‐identified in accordance with HIPAA's Privacy Rule, and no patient‐level identifiers are accessible. Access to the data is governed through formal data‐sharing agreements with participating institutions. As a federated network using only de‐identified data, studies conducted within TriNetX do not require approval from institutional review boards or informed consent. Additional methodological details are provided in the Supporting Information.
2.2. Cohort Definition
The searches on the TriNetX online research platform were performed on March 23, 2026, using the Global Collaborative Network. At the time of data extraction, 168 healthcare organizations across the U.S. had contributed eligible patient data.
Based on recorded ICD‐10‐CM codes, two cohorts were constructed. The first cohort included patients with TBI who developed delirium within 14 days and had no prior history of delirium or dementia. This cohort comprised patients with a recorded diagnosis of TBI (codes S06.1–S06.9 and S06.A) between January 1, 2014, and January 1, 2024, along with a subsequent diagnosis of delirium (codes F05 or R41.0). The second cohort included patients with TBI who did not develop delirium within 14 days and had no prior history of delirium or dementia. This cohort comprised patients with a recorded diagnosis of TBI (codes S06.1–S06.9 and S06.A) between January 1, 2014, and January 1, 2024, with no recorded diagnosis of delirium (codes F05 or R41.0) during the study period. Patients with any documented history of delirium or dementia (codes F01–F03, G30, F05 or R41.0) prior to the TBI diagnosis were excluded. A detailed description of the ICD‐10‐CM codes, cohort eligibility criteria, exclusion criteria, and temporal definitions used for cohort construction is provided in Table S3.
2.3. Outcomes
The primary outcome was the two‐year risk of the composite outcome of cognitive decline (F01–F03 and G30). The secondary outcomes included the two‐year risk of dementia (F02, F03), vascular dementia (F01), Alzheimer's disease (G30) and all‐cause death. Follow‐up began on day 181 after the index event, with outcomes assessed from day 181 to day 730.
2.4. Statistical Analysis
Baseline characteristics of TBI patients with delirium versus TBI patients without delirium were compared and balanced using propensity‐score matching (PSM) at a 1:1 ratio. Matching was performed using the greedy nearest neighbour method, employing a caliper of 0.1.
The following covariates were included in the PSM model: age, sex, ethnicity, hypertension, cerebrovascular disease, diabetes mellitus, peripheral vascular disease, ischaemic heart disease, heart failure, mental and behavioural disorders due to psychoactive substance use, acute kidney failure, and chronic kidney disease, injuries to the head and surgical procedures on the skull, meninges and brain. The balance of demographic and clinical variables between groups was assessed using Absolute Standardized Mean Differences (ASD), with an ASD < 0.1 considered indicative of adequate balance.
Univariable Cox proportional hazards models, before and after PSM, were used to calculate hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the risk of adverse events in TBI patients with delirium compared to TBI patients without delirium. Aalen–Johansen cumulative incidence curves were used to estimate the incidence of all‐cause death and the composite cognitive outcome, accounting for competing risks between these events.
Pre‐specified subgroup analyses were performed to explore potential effect modification. Subgroups were defined according to: (i) patient characteristics, including age (patients aged ≥ 75 years) and sex (female); and (ii) history of substance use (F10, F11, and F17). For each subgroup, analyses were conducted for both primary comparisons (TBI patients with delirium vs. TBI patients without delirium). For each subgroup comparison, separate propensity‐score matching, and survival analyses were performed using the same matching criteria and statistical methods described for the primary analysis.
All statistical analyses were conducted using the TriNetX analytics platform, which incorporates R (v3.2–3) and Python 3.7 libraries (scikit‐learn for logistic regression; survival for Cox models). Missing values were not imputed. All tests were two‐tailed, with statistical significance defined as p < 0.05.
3. Results
The final cohorts included 2564 TBI patients with delirium (mean age 66.6 ± 19.4 years; 59.4% male) 667,950 TBI patients without delirium (mean age 48.0 ± 25.1 years; 60.3% male) (Table 1). The study flow‐chart is reported in Figure 1.
TABLE 1.
Baseline characteristics of the cohorts before and after propensity score matching.
| TBI patients with delirium versusTBI patients without delirium | |||||
|---|---|---|---|---|---|
| Before PSM | After PSM | ||||
| TBI patients with delirium (N = 2564) | TBI patients without delirium (N = 667,950) | TBI patients with delirium (N = 2564) | TBI patients without delirium (N = 2564) | ASD | |
| Age at Index | 66.6 ± 19.4 | 48.0 ± 25.1 | 66.6 ± 19.4 | 67.4 ± 19.1 | 0.042 |
| Male | 1523 (59.4%) | 402,637 (60.3%) | 1523 (59.4%) | 1527 (59.6%) | 0.003 |
| White | 1818 (70.9%) | 369,918 (55.4%) | 1818 (70.9%) | 1804 (70.4%) | 0.012 |
| Black or African American | 259 (10.1%) | 77,388 (11.6%) | 259 (10.1%) | 228 (8.9%) | 0.041 |
| Not Hispanic or Latino | 1882 (73.4%) | 391,714 (58.6%) | 1882 (73.4%) | 1877 (73.2%) | 0.004 |
| Hypertensive diseases | 1767 (68.9%) | 136,933 (20.5%) | 1767 (68.9%) | 1791 (69.9%) | 0.020 |
| Ischemic heart diseases | 826 (32.2%) | 52,552 (7.9%) | 826 (32.2%) | 828 (32.3%) | 0.002 |
| Diabetes mellitus | 752 (29.3%) | 60,392 (9.0%) | 752 (29.3%) | 750 (29.3%) | 0.002 |
| Peripheral artery disease | 656 (25.6%) | 37,313 (5.6%) | 656 (25.6%) | 631 (24.6%) | 0.022 |
| Heart failure | 543 (21.2%) | 27,977 (4.2%) | 543 (21.2%) | 512 (20.0%) | 0.030 |
| Cerebrovascular diseases | 1505 (58.7%) | 66,927 (10.0%) | 1505 (58.7%) | 1480 (57.7%) | 0.020 |
| Mental and behavioural disorders due to psychoactive substance use | 912 (35.6%) | 84,609 (12.7%) | 912 (35.6%) | 890 (34.7%) | 0.018 |
| Acute kidney failure or chronic kidney disease | 923 (36.0%) | 43,620 (6.5%) | 923 (36.0%) | 909 (35.5%) | 0.011 |
| Surgical procedures on the skull, meninges and brain | 237 (9.2%) | 8545 (1.3%) | 237 (9.2%) | 210 (8.2%) | 0.037 |
FIGURE 1.
Study flowchart.
Before PSM, TBI patients with delirium were older and exhibited a higher burden of comorbidities compared with TBI patients without delirium (Table 1). For instance, hypertension, diabetes mellitus, ischaemic heart disease and heart failure were all more common among TBI patients with delirium compared with their counterparts (Table 1). Before PSM, when compared to TBI without delirium, TBI patients with delirium showed a higher risk of all‐cause death, dementia, vascular dementia, Alzheimer's disease and composite outcomes of cognitive decline (HR 6.63, 95% CI 5.60–7.86; Table 2, Figure S1).
TABLE 2.
Cox Proportional‐Hazard regression of all‐cause death, composite outcome, dementia, vascular dementia and Alzheimer's disease in TBI patients with delirium versus TBI patients without delirium.
| Before PSM | After PSM | ||||||
|---|---|---|---|---|---|---|---|
| HR (95% CI) | Cohort I (N = 2564) | Cohort II (N = 667,950) | HR (95% CI) | Cohort I (N = 2564) | Cohort II (N = 2564) | ||
| TBI patients with delirium versus TBI patients without delirium | Composite outcome | 6.63 (5.60, 7.86) | 136 (4.7%) | 6343 (0.9%) | 2.47 (1.84, 3.31) | 124 (4.8%) | 71 (2.8%) |
| Dementia | 7.01 (5.85, 8.39) | 121 (4.2%) | 5337 (0.7%) | 2.63 (1.92, 3.60) | 111 (4.3%) | 60 (2.3%) | |
| Vascular Dementia | 9.14 (6.35, 13.14) | 30 (1.0%) | 988 (0.1%) | 2.76 (1.45, 5.24) | 28 (1.1%) | 14 (0.55%) | |
| Alzheimer's disease | 6.78 (4.69, 9.80) | 29 (1.0%) | 1289 (0.20%) | 2.23 (1.21, 4.09) | 27 (1.1%) | 17 (0.66%) | |
| All‐cause death | 3.65 (3.17, 4.20) | 198 (6.8%) | 16,518 (2.2%) | 1.50 (1.21, 1.85) | 181 (7.1%) | 170 (6.6%) | |
Abbreviations: HR, hazard ratio; TBI, traumatic brain injury.
After PSM, baseline characteristics, including demographics, cardiovascular comorbidities and surgical procedures, were well balanced across groups, with all ASDs below 0.1 (Table 1). After PSM, TBI patients with delirium compared to TBI patients without delirium had a higher risk of all‐cause death (HR 1.50, 95% CI 1.21–1.85), dementia (HR 2.63, 95% CI 1.92–3.60), vascular dementia (HR 2.76, 95% CI 1.45–5.24), Alzheimer's disease (HR 2.23, 95% CI 1.21–4.09) and composite outcomes of cognitive decline (HR 2.47, 95% CI 1.84–3.31) (Table 2, Figure S2).
3.1. Multivariable Cox Proportional‐Hazards Model
In the multivariable Cox proportional‐hazards model, age (HR 1.07, 95% CI 1.072–1.077), hypertensive disease (HR 1.12, 95% CI 1.06–1.20), diabetes mellitus (HR 1.29, 95% CI 1.20–1.38), major depressive disorder (HR 1.41, 95% CI 1.23–1.61), cerebral infarction (HR 1.58, 95% CI 1.45–1.72) and Parkinson's disease (HR 2.63, 95% CI 2.31–3.00) were significantly associated with the composite outcome (Table 3).
TABLE 3.
Multivariable Cox proportional hazards model for predictors of incident composite outcome.
| Predictor | HR (95% CI) |
|---|---|
| Age at index | 1.075 (1.072–1.077) |
| Male sex | 0.91 (0.86–0.96) |
| Hypertensive disease | 1.12 (1.06–1.20) |
| Diabetes mellitus | 1.29 (1.20–1.38) |
| Major depressive disorder | 1.41 (1.23–1.61) |
| Cerebral infarction | 1.58 (1.45–1.72) |
| Parkinson's disease | 2.63 (2.31–3.00) |
Abbreviations: CI, confidence interval; HR, hazard ratio.
3.2. Subgroup Analyses
After PSM, in the elderly age subgroup (≥ 75 years), the pattern of results mirrored the primary analysis. TBI patients with delirium showed a higher risk of developing the composite dementia outcome within 2 years compared with TBI patients without delirium (HR 1.70, 95% CI 1.17–2.47) (Figure 2).
FIGURE 2.
After PSM hazard ratios for subgroups comparing the 2‐year risk of composite cognitive outcome, including dementia, vascular dementia and Alzheimer's disease—between TBI patients with and without delirium.
3.3. Sensitivity Analyses
In a sensitivity analysis restricting the follow‐up period to 6–18 months, the results remained consistent with the primary analysis. Compared with TBI patients without delirium, TBI patients with delirium had a higher risk of the composite outcome both before propensity‐score matching (HR 5.06, 95% CI 3.90–6.58) and after matching (HR 1.80, 95% CI 1.19–2.72) (Table S1).
In a second sensitivity analysis excluding patients who died between 1 January 2014 and 1 January 2024, the association remained significant. TBI patients with delirium had a higher risk of the composite outcome compared with TBI patients without delirium, both before PSM (HR 7.12, 95% CI 5.82–8.69) and after PSM (HR 2.64, 95% CI 1.85–3.78; Table S2).
3.4. Cumulative Incidence of Outcomes
Among TBI patients with delirium, 174 patients (6.79%) died, with an estimated cumulative incidence of 11.39%, while 134 patients (5.23%) developed the composite dementia outcome, with an estimated cumulative incidence of 8.62% (Figure S3).
Among TBI patients without delirium, 15,119 patients (2.26%) died, with an estimated cumulative incidence of 3.51%, while 6228 patients (0.93%) developed the composite dementia outcome, with an estimated cumulative incidence of 1.44% (Figure S4).
4. Discussion
In this study, our principal findings are as follows: (i) comorbidities including hypertensive disease, cerebrovascular disease and DM were more prevalent among TBI patients with delirium compared with those without delirium; notably, hypertension and diabetes mellitus were also independent predictors of the composite cognitive decline outcome; (ii) compared with TBI patients without delirium, those with delirium had a higher risk of developing dementia, vascular dementia and Alzheimer's disease individually, as well as the composite cognitive outcome comprising these conditions; (iii) TBI patients with delirium had a higher risk of 2‐year all‐cause mortality compared with TBI patients without delirium; (iv) in the subgroup analysis, delirium was associated with an increased risk of the composite cognitive decline outcome among TBI patients aged > 75 years.
TBI patients with delirium, compared with TBI patients without delirium, in our cohort were older and carried a greater burden of comorbidities including hypertension and diabetes mellitus. Delirium is associated with increased inflammation and greater vulnerability to cerebral insults, including hypoperfusion and hypertension, during a hospital stay [10, 11, 12]. Also, a meta‐analysis of 52 studies showed that patients with diabetes had approximately a 78% higher odds of developing delirium compared to non‐diabetic patients [13]. Consistent with our findings that hypertension and diabetes were independent predictors of the composite outcome, other studies have similarly shown that both hypertension and diabetes mellitus are independently associated with a higher risk of dementia [14, 15, 16]. Taken together, these findings suggest that comorbidities are common among patients with delirium and that appropriate management may help reduce the risk of subsequent cognitive decline.
TBI patients with delirium showed a higher risk of developing dementia within the 2‐year follow‐up compared with the non‐delirium group. It is well established that patients with TBI are at a substantial risk of developing both delirium in the acute phase and subsequent dementia. Delirium is particularly common among critically ill TBI patients, especially in intensive care settings, and is associated with adverse short‐term outcomes [17]. Moreover, growing evidence suggests that TBI confers an increased long‐term risk of cognitive decline and dementia, highlighting its role as an important contributor to chronic neurocognitive impairment. Additionally, recent studies have demonstrated a link between delirium episodes and the future development of dementia [18, 19, 20]. Hence, the development of delirium after TBI could be considered a marker of increased dementia risk. Therefore, close monitoring and more structured follow‐up may play an important role in slowing dementia progression in these high‐risk patients.
In our study, the risk of vascular dementia over the 2‐year follow‐up was higher in TBI patients with delirium than in those without delirium. Both TBI and delirium are significantly associated with an increased risk of vascular dementia [21, 22]. Notably, a systematic assessment reported that beta‐blockers and vasodilator antihypertensives were associated with a reduced risk of major neurocognitive disorders, including vascular dementia [23]. As comorbidities such as hypertension were more common in the delirium group, close monitoring and appropriate management of hypertension may play an important role in reducing the risk of future vascular dementia by preserving cerebrovascular integrity and limiting cumulative brain injury [24, 25].
Similar to other forms of dementia, TBI patients with delirium showed a higher risk of developing Alzheimer's disease within 2 years compared with the TBI patients without delirium. TBI, including mild cases, is associated with a modestly increased risk of Alzheimer's disease (RR = 1.18, 95% CI 1.11–1.25) [26]. In addition, delirium episodes significantly increased the risk of developing Alzheimer's disease [9]. Notably, the use of a combination of angiotensin‐converting enzyme inhibitors (ACEIs) and statins after TBI has been associated with a lower risk of possible Alzheimer's disease compared with no treatment [27]. Taken together, screening TBI patients with delirium, ensuring close follow‐up, and managing underlying comorbidities may play an important role in reducing the future risk of Alzheimer's disease.
TBI patients with delirium also had a higher risk of all‐cause mortality compared with the TBI patients without delirium. One recent meta‐analysis of 253 studies reported that patients who experienced delirium had increased mortality across all follow‐up periods after hospital discharge compared with those without delirium [28]. These findings are further supported by evidence that persistent delirium is associated with increased mortality, with the risk rising for each additional day of delirium, highlighting the importance of early detection and management [29]. In line with our findings, monitoring and follow‐up of patients with delirium episodes may play an important role in improving long‐term survival.
After PSM, our subgroup analysis of patients aged over 75 years confirmed the main findings. Age was an independent predictor of cognitive decline. Similarly, two studies in populations aged over 65 and 85 years reported that delirium was associated with higher rates of cognitive decline [9, 30]. Notably, a recent population‐based cohort study including 132,113 matched pairs of community‐dwelling adults aged over 65 years reported that late‐life TBI was associated with a 69% higher risk of incident dementia within 5 years [31]. In line with these findings, greater emphasis should be placed on monitoring and closer follow‐up in older TBI patients who develop delirium.
Given our findings, several important clinical implications should be considered. Early identification of delirium in patients with TBI may help identify individuals at higher risk of subsequent cognitive decline and mortality. This highlights the importance of implementing routine screening strategies, optimizing the management of comorbidities, and ensuring close follow‐up, particularly in older patients. In addition, targeted interventions aimed at reducing delirium duration and severity may have the potential to improve both cognitive and survival outcomes. Further prospective studies are needed to determine whether early and structured management of delirium can modify long‐term neurocognitive trajectories in this high‐risk population.
5. Limitations
This study has several limitations that should be considered. First, the diagnoses of delirium, TBI and dementia were based on ICD‐10 coding, which may be subject to misclassification or coding inaccuracies. Second, due to the observational and retrospective nature of the study, causality cannot be established, and only associations can be inferred. Third, despite propensity‐score matching and adjustment for multiple covariates, residual confounding from unmeasured variables may still be present. Fourth, selection bias may exist due to the inclusion of patients captured within the TriNetX network, which may not be fully representative of the general population. Finally, the retrospective design limits the ability to account for changes in clinical status, severity of TBI, or delirium duration, which may influence the observed outcomes.
6. Conclusion
In this large real‐world cohort study, delirium following TBI is associated with an increased risk of dementia, vascular dementia and Alzheimer's disease, as well as higher all‐cause mortality. Delirium post‐TBI may be a marker or a contributing factor to long‐term cognitive decline in TBI patients. Careful monitoring of patients with TBI and delirium, particularly older adults, might have a role in facilitating improved screening and timely referral to specialist services for cognitive decline assessment and work‐up.
Author Contributions
Amirreza Sabahizadeh: visualization, writing – original draft, writing – review and editing; Amir Askarinejad: supervision, conceptualisation, writing – review and editing; Koohyar Ahmadzadeh: writing – original draft; Seyed Reza Mirjalili: review and editing; Khashayar Farshid: writing – original draft; Leandra Serio: formal analysis; Enrico Tartaglia: methodology; Riccardo Proietti: supervision; Rhys Davies: supervision; Gregory Y. H. Lip: conceptualisation, project administration, writing – review and editing and supervision.
Funding
The authors have nothing to report.
Ethics Statement
This study was conducted using the TriNetX Research Network, a global federated database of de‐identified electronic health records. TriNetX complies with the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule and applicable US federal regulations governing the use of de‐identified health data. As only anonymized, de‐identified data were used, institutional review board approval and informed consent were not required in accordance with applicable regulations and TriNetX policies.
Conflicts of Interest
G.Y.H.L. is a consultant and speaker for BMS/Pfizer, Boehringer Ingelheim, Daiichi‐Sankyo, Anthos, Huawei, Skylabs.
Supporting information
Table S1: Multivariable Cox proportional‐hazards regression of all‐cause death and composite cognitive outcomes with follow‐up limited to 6–18 months in TBI patients with delirium versus TBI patients without delirium.
Table S2: Multivariable Cox proportional‐hazards regression of composite cognitive outcomes after excluding deceased patients in TBI patients with delirium versus TBI patients without delirium.
Table S3: ICD‐10‐CM phenotype definitions, eligibility criteria and study timelines.
Figure S1: Hazard ratios comparing outcomes between TBI patients with delirium and TBI patients without delirium before PSM.
Figure S2: Hazard ratios comparing outcomes between TBI patients with delirium and TBI patients without delirium after PSM.
Figure S3: Aalen–Johansen cumulative incidence curves for TBI patients with delirium: purple indicates death, and green indicates the composite cognitive outcome.
Figure S4: Aalen–Johansen cumulative incidence curves for TBI patients without delirium: purple indicates death, and green indicates the composite cognitive outcome.
Acknowledgements
The authors have nothing to report.
Data Availability Statement
Data used in this research project will be available upon a reasonable request from the corresponding authors.
References
- 1. Rauchman S. H., Zubair A., Jacob B., et al., “Traumatic Brain Injury: Mechanisms, Manifestations, and Visual Sequelae,” Frontiers in Neuroscience 17 (2023): 1090672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Maas A. I., Menon D. K., Manley G. T., et al., “Traumatic Brain Injury: Progress and Challenges in Prevention, Clinical Care, and Research,” Lancet Neurology 21, no. 11 (2022): 1004–1060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Injury G , “Global, Regional, and National Burden of Traumatic Brain Injury and Spinal Cord Injury, 1990–2016: A Systematic Analysis for the Global Burden of Disease Study 2016,” Lancet Neurology 18, no. 1 (2019): 56–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Wilson L., Stewart W., Dams‐O'Connor K., et al., “The Chronic and Evolving Neurological Consequences of Traumatic Brain Injury,” Lancet Neurology 16, no. 10 (2017): 813–825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Zhang J., Zhang Y., Zou J., and Cao F., “A Meta‐Analysis of Cohort Studies: Traumatic Brain Injury and Risk of Alzheimer's Disease,” PLoS One 16, no. 6 (2021): e0253206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gardner R. C., Bahorik A., Kornblith E. S., Allen I. E., Plassman B. L., and Yaffe K., “Systematic Review, Meta‐Analysis, and Population Attributable Risk of Dementia Associated With Traumatic Brain Injury in Civilians and Veterans,” Journal of Neurotrauma 40, no. 7–8 (2023): 620–634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Brett B. L., Gardner R. C., Godbout J., Dams‐O'Connor K., and Keene C. D., “Traumatic Brain Injury and Risk of Neurodegenerative Disorder,” Biological Psychiatry 91, no. 5 (2022): 498–507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Han Q. Y. C., Rodrigues N. G., Klainin‐Yobas P., Haugan G., and Wu X. V., “Prevalence, Risk Factors, and Impact of Delirium on Hospitalized Older Adults With Dementia: A Systematic Review and Meta‐Analysis,” Journal of the American Medical Directors Association 23, no. 1 (2022): 23–32. [DOI] [PubMed] [Google Scholar]
- 9. Gordon E. H., Ward D. D., Xiong H., Berkovsky S., and Hubbard R. E., “Delirium and Incident Dementia in Hospital Patients in New South Wales, Australia: Retrospective Cohort Study,” BMJ 384 (2024): e077634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ormseth C. H., LaHue S. C., Oldham M. A., Josephson S. A., Whitaker E., and Douglas V. C., “Predisposing and Precipitating Factors Associated With Delirium: A Systematic Review,” JAMA Network Open 6, no. 1 (2023): e2249950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Mulkey M. A., Khan S., Perkins A., et al., “Relationship Between Angiotensin‐Converting Enzyme Inhibitors and Angiotensin Receptor Blockers Prescribing and Delirium in the ICU‐A Secondary Analysis,” Journal of the American Geriatrics Society 71, no. 6 (2023): 1873–1880. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Oliveira F. R., Oliveira V. H., Oliveira Í. M., et al., “Hypertension, Mitral Valve Disease, Atrial Fibrillation and Low Education Level Predict Delirium and Worst Outcome After Cardiac Surgery in Older Adults,” BMC Anesthesiology 18, no. 1 (2018): 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Komici K., Fantini C., Santulli G., et al., “The Role of Diabetes Mellitus on Delirium Onset: A Systematic Review and Meta‐Analysis,” Cardiovascular Diabetology 24, no. 1 (2025): 216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Lennon M. J., Lam B. C. P., Lipnicki D. M., et al., “Use of Antihypertensives, Blood Pressure, and Estimated Risk of Dementia in Late Life: An Individual Participant Data Meta‐Analysis,” JAMA Network Open 6, no. 9 (2023): e2333353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Jung H., Yang P.‐S., Kim D., et al., “Associations of Hypertension Burden on Subsequent Dementia: A Population‐Based Cohort Study,” Scientific Reports 11, no. 1 (2021): 12291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Cao F., Yang F., Li J., et al., “The Relationship Between Diabetes and the Dementia Risk: A Meta‐Analysis,” Diabetology & Metabolic Syndrome 16, no. 1 (2024): 101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wilson L. D., Maiga A. W., Lombardo S., et al., “Prevalence and Risk Factors for Intensive Care Unit Delirium After Traumatic Brain Injury: A Retrospective Cohort Study,” Neurocritical Care 38, no. 3 (2023): 752–760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Minami H., Toyoda K., Hata T., et al., “How Much Risk Does Delirium Represent for the Development of Dementia?: Retrospective Cohort Study From Over 260,000 Patients Record in a Solitary Institution,” Frontiers in Psychiatry 15 (2024): 1387615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Van Munster B. C., Strijkert F., Trzpis M., et al., “Delirium: A Modifiable Risk Factor for Dementia,” European Geriatric Medicine 17 (2025): 1–4. [DOI] [PubMed] [Google Scholar]
- 20. Leighton S. P., Herron J. W., Jackson E., Sheridan M., Deligianni F., and Cavanagh J., “Delirium and the Risk of Developing Dementia: A Cohort Study of 12 949 Patients,” Journal of Neurology, Neurosurgery & Psychiatry 93, no. 8 (2022): 822–827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Hwang G., Park C., Son S. J., et al., “Long‐Term Impact of Delirium on the Risk of Dementia in Hospitalized Older Patients: A Real‐World Multicenter Study,” American Journal of Geriatric Psychiatry 33, no. 2 (2025): 123–139. [DOI] [PubMed] [Google Scholar]
- 22. Simmonds E., Han J., Kirov G., Sharp D. J., Massey T. H., and Escott‐Price V., “Dementia Risk due to Traumatic Brain Injury in Subtypes of Dementia in the Welsh Population,” Neurology 105, no. 3 (2025): e213866. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Walker V. M., Davies N. M., Martin R. M., and Kehoe P. G., “Comparison of Antihypertensive Drug Classes for Dementia Prevention,” Epidemiology 31, no. 6 (2020): 852–859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Santisteban M. M., Iadecola C., and Carnevale D., “Hypertension, Neurovascular Dysfunction, and Cognitive Impairment,” Hypertension 80, no. 1 (2023): 22–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Ungvari Z., Toth P., Tarantini S., et al., “Hypertension‐Induced Cognitive Impairment: From Pathophysiology to Public Health,” Nature Reviews Nephrology 17, no. 10 (2021): 639–654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Graham A., Livingston G., Purnell L., and Huntley J., “Mild Traumatic Brain Injuries and Future Risk of Developing Alzheimer's Disease: Systematic Review and Meta‐Analysis,” Journal of Alzheimer's Disease 87, no. 3 (2022): 969–979. [DOI] [PubMed] [Google Scholar]
- 27. Li M., Reisman J., Morris‐Eppolito B., et al., “Beneficial Association of Angiotensin‐Converting Enzyme Inhibitors and Statins on the Occurrence of Possible Alzheimer's Disease After Traumatic Brain Injury,” Alzheimer's Research & Therapy 12, no. 1 (2020): 33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Tesfaye Y., Davis C. R., Hull M. J., et al., “Long‐Term Clinical Outcomes of Delirium After Hospital Discharge: A Systematic Review and Meta‐Analysis,” Age and Ageing 54, no. 7 (2025): afaf188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Bradley J., Tang F., Panzarella Z., et al., “Persistent Inpatient Delirium Associated With Increased Length of Stay and Mortality,” PLoS One 20, no. 9 (2025): e0331245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Davis D. H., Muniz Terrera G., Keage H., et al., “Delirium Is a Strong Risk Factor for Dementia in the Oldest‐Old: A Population‐Based Cohort Study,” Brain 135, no. 9 (2012): 2809–2816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Huang Y. Q., Vyas M. V., Bronskill S. E., et al., “Rate of Incident Dementia and Care Needs Among Older Adults With New Traumatic Brain Injury: A Population‐Based Cohort Study,” CMAJ: Canadian Medical Association Journal 197, no. 33 (2025): E1067–E1077. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Multivariable Cox proportional‐hazards regression of all‐cause death and composite cognitive outcomes with follow‐up limited to 6–18 months in TBI patients with delirium versus TBI patients without delirium.
Table S2: Multivariable Cox proportional‐hazards regression of composite cognitive outcomes after excluding deceased patients in TBI patients with delirium versus TBI patients without delirium.
Table S3: ICD‐10‐CM phenotype definitions, eligibility criteria and study timelines.
Figure S1: Hazard ratios comparing outcomes between TBI patients with delirium and TBI patients without delirium before PSM.
Figure S2: Hazard ratios comparing outcomes between TBI patients with delirium and TBI patients without delirium after PSM.
Figure S3: Aalen–Johansen cumulative incidence curves for TBI patients with delirium: purple indicates death, and green indicates the composite cognitive outcome.
Figure S4: Aalen–Johansen cumulative incidence curves for TBI patients without delirium: purple indicates death, and green indicates the composite cognitive outcome.
Data Availability Statement
Data used in this research project will be available upon a reasonable request from the corresponding authors.