Absolute risk of acquired brain injury and rehabilitation needs in Denmark, 2014–2023: a population-based prospective cohort study Authors: Lene Odgaard, Jesper Fabricius, Simon Svanborg Kjeldsen, Uwe M. Pommerich, Lars Evald, Jørgen Feldbæk Nielsen, Carsten Bøcker Pedersen Published: 2026 ABSTRACT SummaryBackgroundAlthough frequency estimates for many types of acquired brain injuries are available, absolute risk estimates (percentage of individuals who develop the disease during their lifetime) remain sparse. We aimed to create comprehensive, absolute risk estimates of acquired brain injury and rehabilitation needs.MethodsIn this prospective cohort study, we followed all individuals residing in Denmark from January 1, 2014, to December 31, 2023. We used population-based registers covering the entire Danish population (N = 5.7 million). People with acquired brain injury were identified through the Danish National Patient Register, including stroke, traumatic brain injury, encephalopathy, brain infection, and brain tumour. We estimated lifetime risks, absolute risks, and incidence rates for each acquired brain injury. We estimated one-year prevalence of acquired brain injury, rehabilitation need, and mortality. Estimates were stratified by age and sex.FindingsThe lifetime risk of acquired brain injury was 35.62% (95% confidence interval [CI], 35.50–35.73). The incidence rate of acquired brain injury was 48.11 per 10,000 person-years (95% CI, 47.91–48.30). Stroke had the highest lifetime risk of 19.62% (95% CI, 19.52–19.72), while brain infections had lowest lifetime risk of 1.43% (95% CI, 1.41–1.46). The one-year prevalence of rehabilitation need among patients with acquired brain injury was 28.18% (95% CI, 27.73–28.63). The prevalence varied across diagnostic groups and ages; it was highest in older patients and in patients with stroke (43.22%) and lowest in patients with traumatic brain injury (10.03%).InterpretationOne in three individuals will experience an acquired brain injury during their lifetime, underscoring the substantial public health burden. Rehabilitation needs varied markedly by diagnosis and age, indicating genuinely different needs across diagnostic groups, but also highlights the need for further research to determine whether differences reflect unassessed needs.FundingHealth Research Foundation of Central Denmark Region. Research in contextEvidence before this studySeveral studies have investigated the frequency of acquired brain injury and the availability of rehabilitation services. However, existing studies are often limited in several ways.First, they focus on a specific subgroup or a single type of acquired brain injury (e.g., ischemic stroke only), rather than addressing the full spectrum of acquired brain injuries; second, they rarely report absolute risks (percentage of individuals who will develop the disease during their lifetime); third, most studies focus on the provision of rehabilitation rather than the underlying need for such services. We searched PubMed from 1990 to November 14, 2025, for studies reporting absolute risks across at least two types of acquired brain injury or frequencies of rehabilitation needs across at least two types of acquired brain injury. We found no studies reporting absolute risks across more than one types of acquired brain injury. We found one study from the Global Burden of Disease reporting estimates of the need for rehabilitation across more than one type of acquired brain injury.Added value of this studyThis study presents comprehensive, detailed frequency estimates of acquired brain injury and rehabilitation need. We examine any and specific types of acquired brain injuries, including stroke, traumatic brain injury, encephalopathy, brain infection, and brain tumour. We report complementary frequency measures of acquired brain injury and rehabilitation: lifetime risk, age-specific and sex-specific risks, incidence rates, one-year prevalence of acquired brain injury and rehabilitation need. We provide novel age-specific and sex-specific absolute risk estimates supporting individual-level prediction. This is the first study to report lifetime as well as age- and sex-specific absolute risks across a wide spectrum of acquired brain injuries.Our rehabilitation estimates complement the Global Burden of Disease’s estimates of rehabilitation need, offering potentially greater precision and context-specific validity.Implications of all the available evidenceThe novel age- and sex-specific absolute risks presented in our study extend and complement the incidence rates presented in previous studies. They allow for individualised risk prediction and can help identify life-course exposures for specific acquired brain injuries. Our study may additionally highlight key challenges in planning and optimising rehabilitation services for individuals with ABI. The comprehensive frequency estimates can inform resource allocation, prevention strategies, healthcare planning, and future research on individuals with acquired brain injury. Research in context Evidence before this study Several studies have investigated the frequency of acquired brain injury and the availability of rehabilitation services. However, existing studies are often limited in several ways. First, they focus on a specific subgroup or a single type of acquired brain injury (e.g., ischemic stroke only), rather than addressing the full spectrum of acquired brain injuries; second, they rarely report absolute risks (percentage of individuals who will develop the disease during their lifetime); third, most studies focus on the provision of rehabilitation rather than the underlying need for such services. We searched PubMed from 1990 to November 14, 2025, for studies reporting absolute risks across at least two types of acquired brain injury or frequencies of rehabilitation needs across at least two types of acquired brain injury. We found no studies reporting absolute risks across more than one types of acquired brain injury. We found one study from the Global Burden of Disease reporting estimates of the need for rehabilitation across more than one type of acquired brain injury. Added value of this study This study presents comprehensive, detailed frequency estimates of acquired brain injury and rehabilitation need. We examine any and specific types of acquired brain injuries, including stroke, traumatic brain injury, encephalopathy, brain infection, and brain tumour. We report complementary frequency measures of acquired brain injury and rehabilitation: lifetime risk, age-specific and sex-specific risks, incidence rates, one-year prevalence of acquired brain injury and rehabilitation need. We provide novel age-specific and sex-specific absolute risk estimates supporting individual-level prediction. This is the first study to report lifetime as well as age- and sex-specific absolute risks across a wide spectrum of acquired brain injuries. Our rehabilitation estimates complement the Global Burden of Disease’s estimates of rehabilitation need, offering potentially greater precision and context-specific validity. Implications of all the available evidence The novel age- and sex-specific absolute risks presented in our study extend and complement the incidence rates presented in previous studies. They allow for individualised risk prediction and can help identify life-course exposures for specific acquired brain injuries. Our study may additionally highlight key challenges in planning and optimising rehabilitation services for individuals with ABI. The comprehensive frequency estimates can inform resource allocation, prevention strategies, healthcare planning, and future research on individuals with acquired brain injury. Introduction Acquired brain injury (ABI), most commonly caused by stroke and traumatic brain injury (TBI), is a major global public health challenge with high societal costs and substantial loss of healthy life years for affected individuals and their families.1 Extensive research have reported incidence rates of different types of ABIs, including the Global Burden of Disease (GBD) studies.2, 3, 4, 5, 6, 7 For instance, one GBD study estimated that in the United States, 115 people per 100,000 person-years at risk experienced a stroke.3 However, while incidence rates are informative at the level of the population, absolute risks (also termed risk, cumulative incidence, cumulative incidence function, cumulative incidence proportion), provide information on the individual risk of developing an ABI and are easier to understand.8 To date, no studies have reported lifetime risks of ABI; however, a few studies have reported residual lifetime risks of its major subgroups, stroke9, 10, 11, 12, 13, 14, 15 and TBI.16 For example, a GBD study estimated that 25-year-olds have a 25% residual lifetime risk of stroke,14 meaning that among 25-year-olds, we expect that 25 out of 100 people will develop a stroke during their lifetime. However, there remains a need for a comprehensive resource of absolute risk estimates across the full spectrum of ABIs. Hence, our primary aim was to provide age- and sex specific absolute risk estimates for a broad range of ABIs. This is the first study to report lifetime as well as age- and sex-specific absolute risks across a wide spectrum of ABIs. The absolute number of individuals suffering from ABI has increased over the past three decades.17 At the same time, the percentage of individuals dying from ABI has generally decreased,17 potentially resulting in more people living with ABI sequelae. Rehabilitation improves independence and autonomy, and is increasingly recognised as essential in mitigating the consequences of ABI.18 Referring patients for rehabilitation is a clinical decision based on an assessed need for rehabilitation, which varies across healthcare systems.19 Research has investigated the need for rehabilitation,20,21 and the provision of rehabilitation,22, 23, 24, 25, 26, 27, 28 using different denominators (general population20 versus ABI patients21, 22, 23, 24, 25, 26, 27, 28), impeding comparisons. For example, one GBD study reported that 1% of the global general population needed rehabilitation following stroke,20 while a study from South Korea found that 39% of stroke patients received rehabilitation.24 Few studies have presented age- and sex specific profiles related to ABI rehabilitation,20 and the potentially distinct rehabilitation needs associated with different ABI types.20 Overall, there is a need for comprehensive estimates on the need for rehabilitation after ABI across the full spectrum of ABIs. Our secondary aim was to estimate the prevalence of rehabilitation need following ABI. We used nationwide, individual-level data to generate the comprehensive epidemiological profiles across a spectrum of ABIs and associated rehabilitation and mortality. The ABIs considered were any ABI, stroke, TBI, encephalopathy, brain infection, and brain tumour. In summary, we aimed to create comprehensive, absolute risk estimates of acquired brain injury risk and rehabilitation needs. For comparability with previous studies, we also present incidence rates of ABI. Methods Data sources and study population All Danish citizens are eligible for free, tax-funded treatment and both hospital-based and municipal rehabilitation, leaving minimal demographic or socioeconomic differences in treatment seeking behaviours and care pathways compared to many other countries.29 We included the entire Danish population regardless of demographic, social, or health-related characteristics, by identifying all individuals who resided in Denmark for at least one day between January 1, 2014, and December 31, 2023.30 The Danish Civil Registration System, established in 1968, maintains longitudinal records of all individuals living in Denmark.31 It contains each person’s unique personal identification number (assigned at birth or upon immigration, which enables unambiguous individual level linkage to other national registries), date of birth, sex, and continuously updated information on emigration and vital status. For the present analyses, the personal identification number was used to link the study population from the Danish Civil Registration System to the Danish National Patient Register.32 The Danish National Patient Register was used to retrieve data on ABI diagnoses, hospital-based rehabilitation, and referral for municipal rehabilitation. The register has recorded data on all patients treated in somatic hospital departments since 1977 and has included information on outpatient visits, emergency department contacts, and inpatient rehabilitation since 1994. Diagnostic coding in the register followed the Danish version of the International Classification of Diseases, 8th revision (ICD-8) from 1969 to 1993, and the 10th revision (ICD-10) from 1994 onward. Since 2006, the Danish National Patient Register included information on referrals for municipal rehabilitation services. Classification of acquired brain injury and rehabilitation Individuals were classified as having an ABI if they had a diagnosis of stroke (ischemic, haemorrhagic, or subarachnoid haemorrhagic), TBI, encephalopathy (e.g., anoxic brain injury due to e.g., drowning, cardiac arrest; seizure-related encephalopathy; metabolic encephalopathy due to e.g., hypoglycaemia), brain infection, or brain tumour (including low- and high-grade primary brain tumours). Individuals were classified with a rehabilitation need if they within one year after the ABI had either a rehabilitation diagnosis or referral for municipal rehabilitation due to ABI (Appendix 2 and eTable S1, Supplement 1). Statistical analyses We estimated age- and sex-specific incidence rates and absolute risks of ABI using competing risk analyses.33 Lifetime risk of ABI was defined as the absolute risk of ABI at the 95th birthday. Individuals were followed from birth or January 1, 2014, whichever came last; and until the first diagnosis of the ABI, death, emigration from Denmark, December 31, 2023, or their 95th birthday, whichever came first. Individuals with a diagnosis of the specific ABI of interest or its sequela during a 37-year washout period (1977–2013) were excluded.34 We also calculated residual lifetime risks of ABI, i.e., among people alive at age e.g., 25 years, the probability of developing ABI during the rest of their lifetime. We estimated age-specific one-year prevalence of ABI. In addition, we estimated one-year prevalence of rehabilitation need and death. Prevalence was estimated using both the general population and ABI patients as denominator. To provide clinically relevant estimates, we also estimated TBI prevalence after excluding individuals with mild TBI. As a Supplementary analysis, incidence rates and prevalence of rehabilitation were standardised to the WHO world standard population35 (eTables S20–S32 in Supplement 1). Additionally, we calculated incidence rates by birth cohort and calendar period. All estimates were calculated separately for overall ABI and for each specific ABI subtype, i.e., stroke, TBI, encephalopathy, brain infection, and brain tumour. Data was curated using the SAS system version 9.4. We used the survival package in R 4.2.1 for estimation and ggplot2 for visualisation. Role of the funding source The funder had no roles in the study design, data collection, data analysis, interpretation, writing of the report, or the decision to submit this paper for publication. Ethics approval The study was approved by the Danish Data Protection Agency (j.nr: 2016-051-000001/2782) and data access was granted by the Danish Health Data Authority and Statistics Denmark. In accordance with Danish legislation, no informed consent was required, as the study relied solely on registry data. Results The entire Danish population of 5,652,673 individuals (49% male) were followed up for a total of 49,752,795 person-years at risk between 2014 and 2023. During this period 239,342 individuals suffered their first ABI (53% male), 347,191 individuals died (51% male), and 125,302 individuals emigrated from Denmark (54% male) (Table 1).Table 1Number of individuals, incidence rates and lifetime risks of acquired brain injury, Denmark, 2014–2023.DiseaseNo. of Prevalent Cases before follow-upNo. of Individuals at Risk at initiation of follow-upNo. of New Cases during follow-upNo. of Person-years at risk during follow-upIncidence rate of ABI per 10,000 person-years at riskaLifetime risk of ABI per 100 peoplebAny acquired brain injury546,5035,652,673239,34249,752,79548.11 (47.91–48.30)35.62 (35.50–35.73)Stroke152,0716,047,105124,22553,987,41323.01 (22.88–23.14)19.62 (19.52–19.72)Traumatic brain injury366,3785,832,798108,58451,616,27821.04 (20.91–21.16)16.56 (16.47–16.66)Encephalopathy21,8086,177,36813,64555,298,4252.47 (2.43–2.51)1.99 (1.96–2.03)Brain infection28,1766,171,000969155,221,3501.75 (1.72–1.79)1.43 (1.41–1.46)Brain tumour11,8686,187,30815,53955,383,3712.81 (2.76–2.85)2.24 (2.20–2.27)Numbers in parenthesis are 95% confidence intervals.aThe incidence rates measure the number of new cases per 10,000 person-years at risk.bThe lifetime risks (also termed the cumulative incidences or absolute risks) measure the probability of developing the acquired brain injury before the 95th birthday. Table 1Number of individuals, incidence rates and lifetime risks of acquired brain injury, Denmark, 2014–2023.DiseaseNo. of Prevalent Cases before follow-upNo. of Individuals at Risk at initiation of follow-upNo. of New Cases during follow-upNo. of Person-years at risk during follow-upIncidence rate of ABI per 10,000 person-years at riskaLifetime risk of ABI per 100 peoplebAny acquired brain injury546,5035,652,673239,34249,752,79548.11 (47.91–48.30)35.62 (35.50–35.73)Stroke152,0716,047,105124,22553,987,41323.01 (22.88–23.14)19.62 (19.52–19.72)Traumatic brain injury366,3785,832,798108,58451,616,27821.04 (20.91–21.16)16.56 (16.47–16.66)Encephalopathy21,8086,177,36813,64555,298,4252.47 (2.43–2.51)1.99 (1.96–2.03)Brain infection28,1766,171,000969155,221,3501.75 (1.72–1.79)1.43 (1.41–1.46)Brain tumour11,8686,187,30815,53955,383,3712.81 (2.76–2.85)2.24 (2.20–2.27)Numbers in parenthesis are 95% confidence intervals.aThe incidence rates measure the number of new cases per 10,000 person-years at risk.bThe lifetime risks (also termed the cumulative incidences or absolute risks) measure the probability of developing the acquired brain injury before the 95th birthday. Number of individuals, incidence rates and lifetime risks of acquired brain injury, Denmark, 2014–2023. Numbers in parenthesis are 95% confidence intervals. The incidence rates measure the number of new cases per 10,000 person-years at risk. The lifetime risks (also termed the cumulative incidences or absolute risks) measure the probability of developing the acquired brain injury before the 95th birthday. Risk of ABI The overall incidence rate for any ABI was 48.11 per 10,000 person-years (95% confidence interval [CI], 47.91–48.30), with a lifetime risk of 35.62% (95% CI, 35.50–35.73), meaning that 35 out of 100 people will experience an ABI during their lifetime. Stroke was the largest diagnostic group (Table 1). Generally, incidence rates and absolute risks were higher in males than in females and rates increased with age, although there were differences across the diagnostic groups (Fig. 1). The stroke incidence rate steadily increased from around age 40, while TBI rates showed minor peaks in young children and in adolescents, and rising rates particularly after age 70. Encephalopathy rates started to increase around age 45, particularly in males, and declined around age 85. Brain infection rates peaked in newborns, with additional smaller peaks around ages 30 and 80 years of age. Brain tumour was the only diagnostic group with highest risk in females, and rates increased from around age 40 and declined after age 75 (Fig. 1 and eTables S20–S25 in Supplement 1). Data showed no evidence of neither birth cohort nor calendar period effects, except for a suggestive minor decrease in TBI incidence following the Covid period (eFigs. S2 and S3 in Supplement 1).Fig. 1Risk of any acquired brain injury and five subtypes of acquired brain injury, Denmark, 2014–2023. (a) Any acquired brain injury. (b) Stroke. (c) Traumatic brain injury. (d) Encephalopathy. (e) Brain infection. (f) Brain tumour. Risk of any acquired brain injury and five subtypes of acquired brain injury, Denmark, 2014–2023. (a) Any acquired brain injury. (b) Stroke. (c) Traumatic brain injury. (d) Encephalopathy. (e) Brain infection. (f) Brain tumour. The lifetime risk of stroke and TBI in females was 18.44% (95% CI, 18.30–18.58) and 15.92%, (95% CI, 15.79–16.05) respectively, and in males it was 20.78% (95% CI, 20.63–20.92) and 17.20% (95% CI, 17.07–17.33), respectively (eTables S5 and S8). For stroke, the residual lifetime risk at age 50 years (eTable S7) was almost the same as the lifetime risk. Please see eTables S2–S19 in Supplement 1 for estimates of lifetime risks in the remaining ABI diagnoses. Prevalence of ABI rehabilitation needs The prevalence of rehabilitation need within a year post-ABI was 1.88 per 1000 people in the general population (95% CI, 1.84–1.91), i.e., approximately 2 per 1000 people in the general population needed ABI-related rehabilitation (Table 2). The prevalence of rehabilitation need had similar curve as the prevalence of ABI with higher prevalence of ABI corresponding to higher prevalence of rehabilitation need (eFig. S1).Table 2Rehabilitation need following acquired brain injury in the national Danish population (N = 5,795,652).aDiseaseNo. of patientsbNo. of patients with rehabilitation needa,cOne-year prevalence of rehabilitation need per 1000 peoplea,dOne-year prevalence of rehabilitation need per 100 patientsa,eAny acquired brain injury38,63610,8871.88 (1.84–1.91)28.18 (27.73–28.63)Stroke19,82685681.48 (1.45–1.51)43.22 (42.52–43.91)Traumatic brain injury12,18912230.21 (0.20–0.22)10.03 (9.51–10.58)Encephalopathy20453530.06 (0.05–0.07)17.26 (15.65–18.97)Brain infection11602450.04 (0.04–0.05)21.12 (18.80–23.58)Brain tumour42408040.14 (0.13–0.15)18.96 (17.79–20.18)Numbers in parenthesis are 95% confidence intervals.aFollowing hospital-based ABI treatment, rehabilitation need was evaluated by an interdisciplinary team of medical doctors, therapists and nurses (Appendix 2, Supplemental 1).bNumber of patients treated in year 2020.cNumber of patients treated in year 2020 with rehabilitation need within one year of injury.dThe one-year prevalences measure the proportion of people in the general population who were treated for the acquired brain injury in year 2020 with rehabilitation need within one year of injury.eThe one-year prevalences measure the proportion of patients treated for the acquired brain injury in year 2020 with rehabilitation need within one year of injury. Table 2Rehabilitation need following acquired brain injury in the national Danish population (N = 5,795,652).aDiseaseNo. of patientsbNo. of patients with rehabilitation needa,cOne-year prevalence of rehabilitation need per 1000 peoplea,dOne-year prevalence of rehabilitation need per 100 patientsa,eAny acquired brain injury38,63610,8871.88 (1.84–1.91)28.18 (27.73–28.63)Stroke19,82685681.48 (1.45–1.51)43.22 (42.52–43.91)Traumatic brain injury12,18912230.21 (0.20–0.22)10.03 (9.51–10.58)Encephalopathy20453530.06 (0.05–0.07)17.26 (15.65–18.97)Brain infection11602450.04 (0.04–0.05)21.12 (18.80–23.58)Brain tumour42408040.14 (0.13–0.15)18.96 (17.79–20.18)Numbers in parenthesis are 95% confidence intervals.aFollowing hospital-based ABI treatment, rehabilitation need was evaluated by an interdisciplinary team of medical doctors, therapists and nurses (Appendix 2, Supplemental 1).bNumber of patients treated in year 2020.cNumber of patients treated in year 2020 with rehabilitation need within one year of injury.dThe one-year prevalences measure the proportion of people in the general population who were treated for the acquired brain injury in year 2020 with rehabilitation need within one year of injury.eThe one-year prevalences measure the proportion of patients treated for the acquired brain injury in year 2020 with rehabilitation need within one year of injury. Rehabilitation need following acquired brain injury in the national Danish population (N = 5,795,652).a Numbers in parenthesis are 95% confidence intervals. Following hospital-based ABI treatment, rehabilitation need was evaluated by an interdisciplinary team of medical doctors, therapists and nurses (Appendix 2, Supplemental 1). Number of patients treated in year 2020. Number of patients treated in year 2020 with rehabilitation need within one year of injury. The one-year prevalences measure the proportion of people in the general population who were treated for the acquired brain injury in year 2020 with rehabilitation need within one year of injury. The one-year prevalences measure the proportion of patients treated for the acquired brain injury in year 2020 with rehabilitation need within one year of injury. The prevalence of rehabilitation need within a year post-ABI was 28.18% (95% CI, 27.73–28.63) in patients, i.e., 28 per 100 patients with ABI needed rehabilitation. The need was highest in stroke (43%) and lowest in TBI (10%) (Table 2). When individuals with mild TBI were excluded from the TBI population, the prevalence of rehabilitation need in TBI reached 29.09% (eTable S32). The rehabilitation need in encephalopathy, brain infections and brain tumours was lower (17–21%). The prevalence of rehabilitation needs in patients (Fig. 2) generally increased with age, with a slight decline after 90 years. The rehabilitation need in early adulthood was similar across diagnostic groups (9–22%, excluding mild TBI from the TBI population) (eTables S27 and 29–32). However, the need among older patients varied, ranging from 17% for encephalopathy to 53% for stroke. The one-year prevalence of death following rehabilitation increased with age, particularly among patients with stroke (Fig. 2).Fig. 2One−year prevalence of rehabilitation need and mortality per 100 patients with any acquired brain injury and five subtypes of acquired brain injury. (a) Any acquired brain injury. (b) Stroke. (c) Traumatic brain injury. (d) Encephalopathy. (e) Brain infection. (f) Brain tumour. One−year prevalence of rehabilitation need and mortality per 100 patients with any acquired brain injury and five subtypes of acquired brain injury. (a) Any acquired brain injury. (b) Stroke. (c) Traumatic brain injury. (d) Encephalopathy. (e) Brain infection. (f) Brain tumour. Comprehensive estimates of the prevalence of post-ABI rehabilitation need in the general population and in patients with ABI can be found in eTables S26–S38 in Supplement 1. Discussion This nationwide study is the first to report lifetime risk and age- and sex-specific absolute risk of ABI, providing novel frequency estimates useful for individual risk prediction. Additionally, this study presents prevalence of ABI, and the subsequent rehabilitation need and mortality. We found that approximately one-third of the Danish population will experience an ABI in their lifetime. Over a one-year period, 0.2% of the Danish population needed post-ABI rehabilitation corresponding to 28% of ABI patients. The estimated incidence rates differ from the rates reported in the GBD studies.2,4 While the GBD relies on modelled data and aggregated sources across included nations, our estimates are based on individual-level data with potentially greater precision, but representative of the Danish population only. When standardised to the WHO world standard population,35 our stroke incidence rates were similar in magnitude to those reported by the GBD,4 lending support to the validity of both methodological approaches. In contrast, the age standardised incidence rates for TBI were lower compared to those reported by the GBD,2 suggesting that prevention efforts in Denmark (such as traffic safety education) are effective.36 For brain tumours, we included low-grade brain tumours to capture the full spectrum of individuals who may require rehabilitation, resulting in higher standardised incidence rates compared to GBD estimates, which focus only on high-grade brain tumours.4 Consistent with previous studies,4 we found higher incidence rates of ABI in males than in females. However, in contrast to earlier findings,4 we observed higher incidence rates of brain tumours in females likely owing to inclusion of low-grade brain tumours.37 The estimated residual lifetime risks of stroke and TBI at ages 25 and 50 in our study were comparable to or lower than those reported in European countries, Japan, and the United States.9, 10, 11, 12, 13, 14, 15, 16 The lower risks may be attributed to prior studies’ inability to account for the competing risk of death, which are known to upwards bias cumulative risk estimates.33 To our knowledge, this is the first study to present lifetime risks and age- and sex-specific absolute risks for a broad range of ABI diagnoses while appropriately accounting for the competing risk of death.33 The reported age- and sex-specific absolute risks allow for individualised risk prediction and can help identify life-course exposures for specific ABIs. With their easier interpretation, these estimates extend and complement the incidence rates from previous studies, which have reported age- and sex-related variation in ABI incidence, e.g., from the Global Burden of Disease studies.5,6,17 For example, both absolute risks and incidence rates highlight potential prevention target areas: sports and traffic safety for children and young adults, fall and frailty prevention for the elderly to reduce the risk of TBI,38 and infant vaccination programs to address the high incidence of brain infections in early childhood.39 The estimated prevalence of rehabilitation needs following stroke and TBI in the general population was considerably lower than GBD estimates of rehabilitation needs20 for these conditions, also after standardisation to the WHO world standard population (eTables S27, S28, and S32). These lower estimates may reflect more effective acute care in Denmark, potentially resulting in fewer sequelae and reduced need for rehabilitation. The discrepancy may also indicate that GBD estimates overstate rehabilitation needs, as they infer need from average disability weights rather than from clinically assessed rehabilitation need. Our clinically grounded estimates may more accurately reflect the volume of rehabilitation services required for effective healthcare planning. Our estimates of stroke rehabilitation needs align with population-based studies reporting utilisation of rehabilitation services following stroke in South Korea and Taiwan.24,26 However, although both studies were population-based, one excluded individuals who died within six months after stroke,24 while the other included only incident cases,26 which may have led to an overestimation of rehabilitation utilisation following stroke in both studies. In contrast, our study estimates the need for stroke rehabilitation as assessed by clinicians regardless of prior stroke history, survival after hospital admission and whether rehabilitation was ultimately provided. These estimates are directly relevant for healthcare planning and service provision. Our estimates of rehabilitation needs following stroke may highlight a critical dilemma in planning rehabilitation services for individuals with stroke: a significant proportion of older stroke patients assessed as having rehabilitation needs died within one year of their diagnosis. This underscores the importance of taking advanced age into account when planning rehabilitation, as it raises the possibility of overtreatment in stroke rehabilitation and reflects broader concerns about overtreatment in healthcare more generally.40 Consistent with findings from the United States,23 the estimated rehabilitation need varied substantially across diagnostic groups. The particularly low prevalence of rehabilitation need among patients with encephalopathy, brain infections, and brain tumours reflect low clinical indication for rehabilitation in these groups. However, as rehabilitation pathways are less well defined for these diagnoses than for stroke (Appendix 2, Supplemental 1), this could also indicate unassessed rehabilitation needs and ultimately unmet rehabilitation needs in these groups. The extent of potentially unassessed and unmet rehabilitation needs in specific ABI subgroups remains an important area for future research, where the identification of unmet needs should be supported by functional and outcome-based data. Strengths of this study include the use of comprehensive, representative data encompassing all patients regardless of demographic or social factors.30 However, some limitations should be considered when interpreting our findings. First, the quality of data in the Danish National Patient Register depends on both the hospitalisation pattern of specific diseases and the accuracy of the diagnostic coding. Acute or severe diseases, such as stroke, brain infections and brain tumours, are likely to be recorded as they usually lead to a hospital contact. A validation study of stroke diagnoses (excluding subarachnoid haemorrhage) in the National Patient Register,41 reported high diagnostic accuracy with a positive predictive value of 93%. Subarachnoid haemorrhage has a lower diagnostic accuracy42 but accounts for only about 7% of all strokes in high-income countries,6 making its influence on the overall diagnostic accuracy of stroke relatively small. Other diseases such as mild TBI may be managed exclusively by the general practitioners and therefore not recorded in the register. Consequently, the incidence and prevalence of TBI may be underestimated. Second, the need for rehabilitation may be underestimated, as the need for some rehabilitation services, such as return-to-work programs, are assessed outside the healthcare system, and are therefore not recorded in the register. Similarly, rehabilitation needs, that evolve after discharge were not captured in our measure, which was based on assessments conducted during admission and at discharge from hospital. This is underscored by a study from Sweden (a country closely comparable to Denmark in terms of health care system and population composition) reporting that many patients experience unmet rehabilitation needs one year after stroke.21 Additionally, our measure does not capture the content or intensity of the rehabilitation need, as it was defined as a binary (yes/no) indication of need assessed by clinicians during the hospital stay. Finally, the rehabilitation estimates may not be generalisable to countries with differing acute care quality, as the level of acute care influences both the prevalence of ABI sequelae, the subsequent rehabilitation needs, and mortality. In conclusion, this nationwide study offers a detailed overview of the risk of ABI and rehabilitation need. It is the first study to report absolute risks and lifetime risks across a broad range of ABI diagnoses allowing for individualised risk prediction. The finding that one in three individuals will experience an ABI underscores its substantial public health impact. The finding that rehabilitation needs varied across diagnostic groups indicates genuinely different needs but also underscores the need for further research to examine whether the differences reflect unassessed needs. The comprehensive frequency estimates provide novel insights to inform resource allocation, prevention strategies, healthcare planning, and future research in individuals with ABI. Contributors CBP and LO had full access to the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. LO and CBP drafted the manuscript. All authors (LO, JF, SSK, UP, LE, JFN, CBP) were involved in the conception and design of the study and critically contributed to the paper for intellectual content. All authors (LO, JF, SSK, UP, LE, JFN, CBP) are responsible for the decision to submit this paper. Data sharing statement All analytical syntax and summary results are publicly available for download at Open Science Framework (osf.io/y8nh3). Access to individual-level Denmark data is governed by Danish authorities. These include the Danish Data Protection Agency, the Danish Health Data Authority and Statistics Denmark. Each scientific project must be approved before initiation, and approval is granted to a specific Danish research institution. Researchers at Danish research institutions may obtain the relevant approval and data. International researchers may gain data access if governed by a Danish research institution having needed approval and data access. Declaration of interests LO received funding from the Health Research Foundation of Central Denmark Region. The funder had no role in the design, conduct, or analysis of the study. All other authors declare no competing interests.