Acquired Brain Injury Among Adolescents and Young Adults: A Nationwide Study of Labor Market Attachment

Purpose Young patients represent a particularly vulnerable group regarding vocational prognosis after an acquired brain injury (ABI). We aimed to investigate how sequelae and rehabilitation needs are associated with vocational prognosis up to 3 years after an ABI in 15-30-year-old patients. Methods An incidence cohort of 285 patients with ABI completed a questionnaire on sequelae and rehabilitation interventions and needs 3 months after the index hospital contact. They were followed-up for up to 3 years with respect to the primary outcome "stable return to education/work (sRTW)", which was defined using a national register of public transfer payments. Data were analyzed using cumulative incidence curves and cause-specific hazard ratios. Results Young individuals reported a high frequency of mainly pain-related (52%) and cognitive sequelae (46%) at 3 months. Motor problems were less frequent (18%), but negatively associated with sRTW within 3 years (adjusted HR 0.57, 95% CI 0.39-0.84). Rehabilitation interventions were received by 28% while 21% reported unmet rehabilitation needs, and both factors were negatively associated with sRTW (adjusted HR 0.66, 95% CI 0.48-0.91 and adjusted HR 0.72, 95% CI 0.51-1.01). Conclusions Young patients frequently experienced sequelae and rehabilitation needs 3 months post ABI, which was negatively associated with long-term labor market attachment. The low rate of sRTW among patients with sequelae and unmet rehabilitation needs indicates an untapped potential for ameliorated vocational and rehabilitating initiatives targeted at young patients.

Vocational/educational prognosis in adolescents and young adults with acquired brain injury: a nationwide cohort study

OBJECTIVE

To determine prognostic factors for work ability and employment/educational status among young patients referred to outpatient neurorehabilitation clinics after an acquired brain injury.

METHODS

A nationwide cohort study of 471 15-30-year-old patients who attended an interdisciplinary clinical assessment and provided questionnaire data at baseline and after one year. The outcomes were the Work Ability Score (WAS, 0-10 (best)) and employment/educational status after one year. Prognostic performance was analyzed using univariable regression and multivariable Ridge regression in a five-fold cross-validated procedure.

RESULTS

Preinjury, 86% of the patients were employed, while the percentage had decreased to 55% at baseline and 52% at follow-up. The model, which included clinical measures of function, showed moderate prognostic performance with respect to WAS (R2=0.29) and employment/educational status (area under the curve (AUC)=0.77). Glasgow Outcome Scale Extended (R2=0.15, AUC=0.68) and the cognitive subscale of the Functional Independence Measure (R2=0.09, AUC=0.64), along with fatigue measured with the Multidimensional Fatigue Inventory (R2=0.15, AUC=0.60) were the single predictors with the highest predictive performance.

CONCLUSION

Despite generally high scores in motor and cognitive tests, only about half of the patients were employed at baseline and this proportion remained stable. Global disability, cognitive sequelae and fatigue had the highest prognostic performance.

Home care provision and nursing home admission after myocardial infarction in relation to cardiogenic shock and out-of-hospital cardiac arrest status

Background

Autonomy is of great importance for quality of life. There is a paucity of data on autonomy for those who survive myocardial infarction (MI) with and without cardiogenic shock (CS) and out-of-hospital arrest (OHCA).

Purpose

To examine the association between CS, OHCA, and need for home care provision or nursing home admission as a proxy for impaired autonomy in a first-time MI population.

Methods

Danish nationwide registries were used to identify patients with first-time MI (2009–2019), who prior to the event were living at home without home care and discharged alive. The patients were stratified according to CS and OHCA status. We report 1-year cumulative incidence of a composite outcome of home care provision or nursing home admission with competing risk of death and as a secondary outcome all-cause mortality. Cause specific Cox regression models were used to estimate adjusted hazard ratios (HR) with patients without CS or OHCA as reference.

Results

We identified 61,451 patients in the period with MI (by groups: −OHCA/−CS: 59,316, −OHCA/+CS: 1,597, +OHCA/−CS: 913, and +OHCA/+CS: 669). The 1-year cumulative incidences of home care/nursing home were 6.9% for patients with −OHCA/−CS, 21.1% for −OHCA/+CS, 5.2% for +OHCA/−CS, and 8.1% for those with +OHCA/+CS. With the −OHCA/−CS as reference, the adjusted HRs for home care/nursing home were 3.12 (95% CI: 2.78–3.49) for patients with −OHCA/+CS, 1.27 (95% CI: 0.95–1.70) for +OHCA/−CS, and 2.31 (95% CI: 1.76–3.03) for +OHCA/+CS (Figure). The 1-year cumulative incidences of mortality were 4.8% for patients with −OHCA/−CS, 10.0% for −OHCA/+CS, 2.8% for +OHCA/−CS, and 3.7% for those with +OHCA/+CS (adjusted HRs: 2.81 (95% CI: 2.55–3.10), 1.09 (95% CI: 0.85–1.39) and 1.81 (95% CI: 1.42–2.30) (Figure 1).

Conclusion

In a selected cohort of patients with MI, without previous need for home care/nursing home and surviving until discharge date, patients with CS were independent of OHCA status associated with less autonomy after discharge with a more than two-fold higher 1-year incidence of home care provision or nursing home admission. Further, patients with CS were associated with a two-fold higher 1-year mortality compared with MI patients without CS independent of OHCA status.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): The work was supported by Rigshospitalets Research Foundation, Master cabinetmaker Sophus Jacobsen and Wife Astrid Jacobsen Foundation, and Director Jacob Madsen and Wife Olga Madsens Foundation. The funding source had no role in the design, conduct, analysis, or reporting of the study.

High risk of cardiovascular disease in curatively treated patients with oesophageal cancer: a Danish cohort study

Background

The cardiovascular disease (CVD) burden among patients with oesophageal cancer (EC) treated with curative intent is unclear.

Purpose

To determine CVD incidence and all-cause mortality in patients with EC.

Material and methods

Danish national health registries were used to identify patients diagnosed with primary EC between 2008 and 2018. Each EC patient was matched with ten individuals from the general population. The primary endpoint was a CVD hospital contact (CVD-HC), either admission or outpatient contact. Secondary endpoints were all-cause mortality and five specific CVD endpoints evaluated separately: atrial fibrillation, ischemic heart disease, heart failure, perimyocarditis, and venous thromboembolism. Using registries, all endpoints were assessed up to ten years following the EC diagnosis.

Results

The study included 1,525 patients with EC matched to 15,250 individuals from the general population. Patients with EC had a post-diagnosis one-year adjusted hazard ratio (HR) of CVD-HCs of 6.1 (95% CI: 5.6 to 6.8) compared with the general population. During the next nine years, the risk of CVD-HC was comparable between the two cohorts with an adjusted HR of 1.0 (95% CI: 0.9 to 1.3) (Figure 1). Patients with EC, particularly those with prevalent CVD (29%) had a high risk of ischemic heart disease with a one-year HR of 6.2 (95% CI: 3.7 to 10.4). The risk of venous thromboembolism and atrial fibrillation was 14-fold and four-fold elevated within the first year after EC diagnosis compared to the general population. After EC diagnosis, all-cause mortality was, as expected, increased in patients with EC compared with the general population. However, prevalent CVD among patients with EC did not appear to be associated with higher all-cause mortality.

Conclusion

CVD morbidity was transiently increased in the first year following EC diagnosis compared with the general population. All-cause mortality risks were high but did not appear to be affected by prevalent CVD. The very high risk of CVD in curatively treated patients with EC calls for healthcare initiatives to advance preventive and post-treatment strategies.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): The Danish Cancer SocietyCarpenter Jorgen Holm and Wife Elisa F. Hansen's Memorial Scholarship

Experiences, perspectives, and practice patterns of pulmonary embolism follow-up in Denmark. A mixed method study informing the development of A sTrucTurEd INtegrateD post PE care model

Background

Practice patterns of follow-up after pulmonary embolism (PE) varies both within and between countries, and high-quality evidence to provide clinical guidance is lacking [1,2]. This study is part of the trial – A sTrucTurEd INtegrateD post Pulmonary Embolism care model (Attend-PE) aimed at developing and testing the effectiveness of a structured, integrated and patient-centered follow-up care model for patients with PE.

Purpose

To explore experiences, perspectives, and practice patterns of PE follow-up. Ultimately identifying and describing the opportunity space for a structured integrated post PE care model.

Methods

We performed a convergent mixed method study based on data from patient journey mapping including surveys, observations, and semi-structured patient interviews across all main hospitals treating PE in Denmark. An integrated mixed methods interpretation was conducted using joint displays.

Results

Main results are presented in joint displays in Figure 1 and 2. Observations and surveys were conducted at 18 hospitals including 12 in-depth patient-interviews.

The structure of care varied greatly. Patients expressed a need for early follow up and specialist care but at 18% of the sites first follow-up visit was not until 3–6 months after PE and 24% of the sites only had in-person contact with patients when follow-up echocardiography was needed. At some sites it was also highlighted that not all patients were referred to the post PE follow-up. anticoagulation clinics. Further, transition of care to the general practitioner was identified as a particular challenge.

All sites considered Non-vitamin K antagonist oral anticoagulants the preferred anticoagulant treatment with low-molecular-weight heparins/Vitamin K antagonists as second choice. The possible need for extended treatment and reduced dose favored Rivaroxaban and Apixaban. The decision on treatment duration was most often taken at the 3–6-month follow-up visit, and patients described the decision on extended treatment as positive and reassuring.

Patient information (oral and written) varied greatly and was most often focused on anticoagulant treatment, while knowledge on PE in general, symptom management, and activity was scarce. Two sites offered structured group-based patient education, but unstructured patient education at the individual follow-up visits was most common.

Help and guidance on managing post-PE symptoms and returning to everyday life was perceived essential by patients, as it affected them both physically and mentally. However, none of the sites routinely offered or referred patients to rehabilitation.

Involvement of relatives was considered important but 33% of the hospital sites did not systematically encourage relatives to attend follow-up.

Conclusion

The considerable variation in structure of PE follow-up care demonstrated in this mixed method study highlights the need for a structured integrated post pulmonary embolism care model.

Funding Acknowledgement

Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): The Danish Heart AssociationThe Novo Nordisk Foundation

The comorbidity network in older adults with atrial fibrillation: a Danish nationwide cohort study

Introduction

Individuals affected by atrial fibrillation (AF) often present with multiple comorbidities that challenge their clinical management and worsen the prognosis. Characterizing the network structure of comorbid chronic conditions in AF patients may provide insights to better tailor medical and care management.

Purpose

To examine and compare the network structure of comorbidities in older people with and without AF.

Methods

Cross-sectional data derived from the Danish National Patient Register (period 2012–2017) were examined. Patients 60+ years old by 1. January 2017 with AF were selected and matched by age and sex to non-AF controls. Chronic conditions coded according to the International Classification of Diseases, 10th revision were retrieved and grouped into 60 clinically relevant disease categories for old age. Network analysis was applied to construct the disease networks and the centrality index of expected influence was measured to estimate the disease interconnectedness in each network. The difference in network structure and disease centrality between AF and non-AF patients was formally assessed through network comparison tests.

Results

A total of 96,117 AF patients (72 years old; 45% women) were identified and matched with 96,117 non-AF controls. The most prevalent chronic conditions in AF were hypertension (55.1%), large bowel diseases (36.4%) and ischemic heart disease (36.0%). A significant difference of global network structure was observed between AF and non-AF patients (p<0.001) (Figure1). Chronic obstructive pulmonary disease, depression, inflammatory arthropathy, chronic kidney disease and peripheral neuropathy had a higher connectivity with other diseases in the AF vs. non-AF patients (p<0.001). By contrast, hypertension, heart failure and stroke were more interconnected in the non-AF patients (p<0.001). Among AF patients, network differences were further observed between age categories (60–79 vs 80+ years) in male and female subgroups.

Conclusions

Older AF patients exhibited a complex network structure of chronic conditions that differed from age- and sex- matched non-AF patients. The network-based identification of highly co-morbid diseases in AF can improve our understanding of AF-related chronic conditions and potentially enhance prioritization and a personalized care for older patients with AF.

Funding Acknowledgement

Type of funding sources: Public grant(s) – EU funding. Main funding source(s): European Union's Horizon 2020 research and innovation programme

Predictors of disability in adolescents and young adults with acquired brain injury after the acute phase

AIM

To develop and validate a prediction model for disability among young patients with acquired brain injury (ABI) after the acute phase.

METHODS

Within a nationwide cohort of 446 15-30-year-old ABI-patients, we predicted disability in terms of Glasgow Outcome Scale - Extended (GOS-E) <7 12 months after baseline assessment in outpatient neurorehabilitation clinics. We studied 22 potential predictors covering demographic and medical factors, clinical tests, and self-reported fatigue and alcohol/drug consumption. The model was developed using multivariable logistic regression analysis and validated by 5-fold cross-validation and geographical validation. The model's performance was assessed by receiver operating characteristic curves and calibration plots.

RESULTS

Baseline assessment took place a median of 12 months post-ABI. Low GOS-E (range 1-8 (best)) and Functional Independence Measure (range 18-126 (best)) along with high mental fatigue (range 4-20 (worst)) predicted disability. The model showed high validity and performance with an area under the curve of 0.82 (95% confidence interval (CI) 0.77, 0.87) in the cross-validation and 0.81 (95% CI 0.73, 0.88) in the geographical validation.

CONCLUSION

We developed and validated a parsimonious model which effectively predicted disability. The model may be useful to guide decision-making in outpatient neurorehabilitation clinics treating young patients with ABI.

Lower perception of pharmacological treatment increases risk of non-adherence to medication

Funding Acknowledgements

Type of funding sources: Foundation. Main funding source(s): The Danish Heart Foundation

Background

A large proportion of patients fail to reach optimal adherence to medication following incident ischemic heart disease (IHD) despite amble evidence of the beneficial effect of medication. Non-adherence to medication increases risk of disease-related adverse outcomes but none has explored how perception about pharmacological treatment detail on non-adherence using register-based follow-up data.

Purpose

To investigate the association between patients’ perception of pharmacological treatment and risk of non-initiation and non-adherence to medication in a population with incident IHD.

Methods

This cohort study followed 871 patients until 365 days after incident IHD. The study combined patient-reported survey data on perception about pharmacological treatment (categorised by ‘To a high level’, ‘To some level’, and ‘To a lesser level’) with register-based data on reimbursed prescription of cardiovascular medication (antithrombotics, statins, ACE-inhibitors/angiotensin receptor blockers, and β-blockers). Non-initiation was defined as no pick-up of medication in the first 180 days following incident IHD and analysed by Poisson regression. Two different measures evaluated non-adherence in patients initiating treatment: 1) proportion of days covered (PDC) analysed by Poisson regression, and 2) risk of discontinuation analysed by Cox proportional hazard regression. All analyses were adjusted for confounding variables (age, sex, ethnicity, income, educational level, civil status, occupation, charlson comorbidity index, supportive relatives, and individual consultation in medication) identified by directed acyclic graph and obtained from national registers and the survey. Item non-response was handled by multiple imputation and item consistency was evaluated by McDonalds omega.

Results

Lower perceptions about pharmacological treatment was associated with increased risk of non-initiation and non-adherence to medication irrespectively of drug class and adherence measure in the multiple adjusted analyses (please see figure illustrating results on antithrombotics). A dose-response relationship was observed both at 180- and 365-days of follow-up, but the steepest decline in adherence differed when comparing the two adherence measures (results not shown). Moderate internal consistency was found for the summed measure of perception (McDonalds omega = 0.67).

Conclusion

Lower perception of pharmacological treatment was associated with subsequent non-initiation and non-adherence to medication, irrespectively of measurement method and drug class.

Abstract Figure. Figre: Multiple adjusted analyses

The impact of parental mental health and socioeconomic position on child preschool health

Background

Many children, approximately one out of four to five, live in families with parental mental health problems. Knowledge on the impact of other mental health problems than depression or anxiety is sparse as well as the impact of the father. Therefore, we aimed to examine the effect of maternal and paternal mental health on child respiratory illness.

Methods

A population-based birth cohort study was conducted including all Danish children born from 2000-2012 using the Danish nationwide registers. Two follow-up periods were created: From age 1-2 and age 3-5. Mental health was categorised in three: No mental health problems, minor mental health condition if handled in solely primary healthcare; and moderate-severe if handled in psychiatric healthcare settings. Child respiratory illness was identified by prescribed medication or hospital-based diagnoses. Incidence rate ratios (IRR) were calculated using Poisson regression analysis.

Results

The analyses included 810,243 children. 26% of the mothers and 17% of the fathers were classified with mental health conditions. Children of parents with mental health conditions were more likely to have respiratory illness (10-12% vs. 7.7%). Adjusted IRRs revealed higher risks for children of mothers with minor (IRR 1.32 (CI95% 1.30-1.34)) and moderate-severe mental health conditions (IRR 1.48 (CI95% 1.44-1.51)). For paternal mental health, the IRRs were 1.14 (CI95% 1.12-1.16) for minor and IRR 1.14 (CI95% 1.11-1.17) for moderate-severe mental health conditions. The IRRs were a little higher for the children aged 1-2 years compared to 3-5-years.

Conclusions

Children of parents with mental health conditions were at greater risk of respiratory illness. This was evident for the mental health of both parents, most pronounced for the mother. Not only moderate-severe, but also minor mental health conditions increased the risk of respiratory illness.

Ethnic differences in incidence and mortality of stroke in Denmark

Background

Stroke is a leading cause of disability and mortality worldwide. However, studies on incidence and mortality of stroke hereof among ethnic minorities compared with local born are still limited.

Methods

We conducted a Danish nationwide register-based cohort study between 2004 and 2018. All cases of first ever stroke aged 18-95 years were included. Country of birth was used to construct ethnic groups. Age standardized incidence rate ratio (IRR) of stroke stratified by country of birth and sex were estimated with the Danish born as a reference group. Ethnic minorities were grouped as Western and Non-western for mortality hazard ratio (HR) estimates.

Results

In overall, ethnic minorities had a higher risk of stroke compared with Danish born. Particularly, the IRR of all stroke was estimated to be 8.3 times higher among Polish men compared to Danish born men (IRR, 8.32; 95% CI, 6.89-10.05). Compared with Danish born women, Pakistan women had the highest risk of all stroke (IRR, 2.89; 95% CI, 2.46-3.39). By contrast, Swedish women had reduced risk of hemorrhagic stroke (IRR, 0.40; 95% CI, 0.18-0.89) and Norwegian women had reduced risk of ischemic stroke (IRR, 0.87; 95% CI, 0.72-1.06). Compared with Danish born men, all-cause 1-year mortality hazard for Non-western men was (HR, 1.61; 95% CI, 1.13-2.29) while for Western men was (HR, 1.07; 95% CI, 0.90-1.29) among ischemic stroke patients. Among hemorrhagic stroke patients, 1-year mortality hazard for Non-western men was (HR, 0.75; 95% CI, 0.39-1.47) whereas Western men had (HR, 1.48; 95% CI, 1.04-2.10). Among women, we observed reduced all cause 1-year mortality hazard in Non-western (HR, 0.32; 95% CI, 0.13-0.80) for hemorrhagic stroke whereas no difference in mortality hazard was observed for ischemic stroke.

Conclusions

Incidence and post stroke mortality appear to vary among ethnic minorities in comparison to Danish born. It may depend on the type of stroke and sex.

Key messages

The study contributes knowledge in migration and health. With good quality registers, we are in unique position to establish findings in Denmark, which has a growing migrant population.

Managing COVID-19 within and across health systems: why we need performance intelligence to coordinate a global response

Background

The COVID-19 pandemic is a complex global public health crisis presenting clinical, organisational and system-wide challenges. Different research perspectives on health are needed in order to manage and monitor this crisis. Performance intelligence is an approach that emphasises the need for different research perspectives in supporting health systems’ decision-makers to determine policies based on well-informed choices. In this paper, we present the viewpoint of the Innovative Training Network for Healthcare Performance Intelligence Professionals (HealthPros) on how performance intelligence can be used during and after the COVID-19 pandemic.

Discussion

A lack of standardised information, paired with limited discussion and alignment between countries contribute to uncertainty in decision-making in all countries. Consequently, a plethora of different non-data-driven and uncoordinated approaches to address the outbreak are noted worldwide. Comparative health system research is needed to help countries shape their response models in social care, public health, primary care, hospital care and long-term care through the different phases of the pandemic. There is a need in each phase to compare context-specific bundles of measures where the impact on health outcomes can be modelled using targeted data and advanced statistical methods. Performance intelligence can be pursued to compare data, construct indicators and identify optimal strategies. Embracing a system perspective will allow countries to take coordinated strategic decisions while mitigating the risk of system collapse.A framework for the development and implementation of performance intelligence has been outlined by the HealthPros Network and is of pertinence. Health systems need better and more timely data to govern through a pandemic-induced transition period where tensions between care needs, demand and capacity are exceptionally high worldwide. Health systems are challenged to ensure essential levels of healthcare towards all patients, including those who need routine assistance.

Conclusion

Performance intelligence plays an essential role as part of a broader public health strategy in guiding the decisions of health system actors on the implementation of contextualised measures to tackle COVID-19 or any future epidemic as well as their effect on the health system at large. This should be based on commonly agreed-upon standardised data and fit-for-purpose indicators, making optimal use of existing health information infrastructures. The HealthPros Network can make a meaningful contribution.

Parental mental vulnerability and use of healthcare services in infants

Background

Parental mental illness has been associated with a number of consequences for the health and use of healthcare services of the child. However, most research has focused on maternal depression. Research examining the impact of paternal mental vulnerability (MV) as well as different degrees of MV are needed to plan interventions. Therefore, the aim of this study was to examine the association between different categories of individual and combined parental MV and the child’s use of healthcare services the first year of life.

Methods

A population-based birth cohort study was conducted including all Danish children born from 2000-2016 using the Danish national registers. Exposure was parental MV of three categories according to the degree of MV: Group 1 “minor MV” with mental related contacts to primary healthcare and/or prescribed psychopharmaceuticals, group 2 “moderate MV” and group 3 “severe MV” both with contacts to psychiatric hospital. Outcome was contacts to GP the first year of life expressed as incidence-rate ratios (IRR) using Poission’s regression analyses.

Results

The analyses included 952,709 children. 21% of the mothers and 11% of the fathers were in the MV groups. Parental MV (any parent, any MV-group) was associated with an increased risk of GP contacts daytime and out-of-hour contacts. If both parents were classified as group 1 MV, IRR were 1.21 (CI95 1.20-1.22). IRR were 1.18 (1.17-1.18) resp. IRR 1.05 (1.04-1.06) if only the mother resp. father were in MV group 1. The same pattern were seen for out-of-hour contacts; IRR 1.28 (1.26-1.31) for both parents in group 1 and IRR 1.19 (1.18-1.20) resp. IRR 1.09 (1.08-1.11) for the mother resp. father.

Conclusions

Maternal and paternal MV were associated with an increased risk of GP contacts daytime and out-of-hour contacts although maternal MV had the highest risk. Even minor MV had an impact on healthcare contacts and the risk increased further if both parents were classified as minor MV.

Key messages

Both maternal and paternal mental vulnerability has an impact on the child’s healthcare contacts. Our results indicate the need for a focus also on minor mental vulnerability in the planning of interventions.

P631Socioeconomic status; how does it influence referral to cardiac rehabilitation after acute myocardial infarction?

Background

The number of patients with low socioeconomic status who are referred to cardiac rehabilitation (CR) has been documented to be relative lower than patients with high SES among all patients hospitalised with acute myocardial infarction (AMI).

Purpose

The aims of this study were to evaluate the referral process to CR and how it is influenced by socioeconomic variables.

Methods

In 2011–2014, 1229 patients were hospitalised with AMI at Department of Cardiology of our University Hospital, Denmark. All were evaluated for participation to CR. Socioeconomic status was measured by personal income, educational level, marital status, and employment and obtained from national registers. Multiple logistic regression assessed socioeconomic determinants in three phases of the referral process to CR: 1. information about CR, 2. wish to participate in CR, and 3. referral to specialiced- or municipality-based CR. All analyses were adjusted for sex, age, and comorbidities.

Results

A total of 1123 (91.4%) patients received information regarding CR. Of these, 854 (69.5%) patients wished to participate in the programme. Income was the most important socioeconomic variable when looking at who were informed about CR (OR 2.17, 95%-CI: 1.0- 4.64) and who wished to participate in CR (OR 1.55, 95%-CI: 1.02–2.35).

Characteristics of study participants Characteristics All participants STEMI NSTEMI UAP n=1229 n=402 n=711 n=116 Male (n, %) 907 (73.8) 322 (80.1) 503 (70.7) 82 (70.7) Age Group (yrs)   <65 591 (48.1) 227 (56.5) 308 (43.3) 56 (48.3)   65–74 371 (30.2) 116 (28.9) 215 (30.2) 40 (34.5)   ≥75 267 (21.7) 59 (14.7) 188 (26.4) 20 (17.2) Baseline Comorbidity   Hypertension 241 (19.6) 62 (15.4) 148 (20.8) 31 (26.7)   Diabetes 14 (1.1) <5 (<1) 8 (1.1) <5 (<1) Charlson Comorbidity Index   Low (0 points) 1088 (88.5) 358 (89.1) 630 (88.6) 100 (86.2)   Moderate/High (>0) 141 (11.5) 44 (10.9) 81 (11.4) 16 (13.8) Civil status (n, %)   Married/Partnership 793 (64.5) 253 (62.9) 449 (63.2) 91 (78.4)   Divorced/Unmarried/Widow 436 (35.5) 149 (37.1) 262 (36.8) 25 (21.6) Occupational status (n, %)   Employed 479 (39.0) 195 (48.5) 240 (33.8) 44 (37.9)   Unemployed/Retired 750 (61.0) 207 (51.5) 471 (66.2) 72 (62.1) Educational status (n, %)   Low 516 (42.0) 144 (35.8) 322 (45.3) 50 (43.1)   Medium 539 (43.9) 201 (50.0) 293 (41.2) 45 (38.8)   High 174 (14.2) 57 (14.2) 96 (13.5) 21 (18.1) Gross income, tertile (n, %)   Low 405 (33.0) 113 (28.1) 251 (35.3) 41 (35.3)   Medium 406 (33.0) 124 (30.8) 247 (34.7) 35 (30.2)   High 418 (34.0) 165 (41.0) 213 (30.0) 40 (34.5) STEMI: ST-elevated myocardial infarction; NSTEMI: non-ST-elevated myocardial infarction; UAP: unstable angina pectoris.

Conclusion

Two out of three patients received referral to CR. However, higher income was proportional with the likelihood of receiving information about CR and willingness to participate in the programme.

Acknowledgement/Funding

the Danish Heart Foundation

Preoperative antithrombotic therapy and risk of blood transfusion and mortality following hip fracture surgery: a Danish nationwide cohort study

Hip fracture surgery is associated with high risk of bleeding and mortality. The patients often have cardiovascular comorbidity, which requires antithrombotic treatment. This study found that preoperative use of oral anticoagulants was not associated with transfusion or mortality following hip fracture surgery, whereas increased risk may exist for antiplatelet drugs.

INTRODUCTION

Hip fracture surgery is associated with high bleeding risk and mortality; however, data on operative outcomes of hip fracture patients admitted while on antithrombotic therapy is sparse. We examined if preoperative antithrombotic treatment was associated with increased use of blood transfusion and 30-day mortality following hip fracture surgery.

METHODS

Using data from the Danish Multidisciplinary Hip Fracture Registry, we identified 74,791 hip fracture surgery patients aged ≥ 65 years during 2005-2016. Exposure was treatment with non-vitamin K antagonist oral anticoagulant (NOAC), vitamin K antagonists (VKA), or antiplatelet drugs at admission for hip fracture. Outcome was blood transfusion within 7 days postsurgery and death within 30 days.

RESULTS

A 45.3% of patients received blood transfusion and 10.6% died. Current NOAC use was associated with slightly increased risk of transfusion (adjusted relative risk (aRR) 1.07, 95% confidence interval (CI) 1.01-1.14), but similar mortality risk (adjusted hazard ratio (aHR) 0.88, 95% CI 0.75-1.03) compared with non-users. The pattern remained when restricting to patients with short surgical delay (< 24 h). VKA users did not have increased risk of transfusion or mortality. The risks of transfusion (aRR 1.15 95% CI 1.12-1.18) and 30-day mortality (aHR 1.18 95% CI 1.14-1.23) were increased among antiplatelet users compared with non-users.

CONCLUSIONS

In an observational setting, neither preoperative NOAC nor VKA treatments were associated with increased risk of 30-day postoperative mortality among hip fracture patients. NOAC was associated with slightly increased risk of transfusion. Preoperative use of antiplatelet drugs was associated with increased risk of transfusion and mortality.

Prescription and predictors of post-stroke antidepressant treatment: A population-based study

OBJECTIVES

Post-stroke depression and pathological crying are common and potentially serious complications after stroke and should be diagnosed and treated accordingly. Diagnosis and treatment probably rely on clinical experience and may pose certain challenges. We aimed to examine prescription and predictors of antidepressant treatment after ischemic stroke in a clinical setting.

MATERIALS AND METHODS

In this registry-based follow-up study, consecutive ischemic stroke patients were identified from the Danish Stroke Registry, holding information on antidepressant treatment during admission in Aarhus County from 2003 to 2010. Information on prescription after discharge was obtained from the Danish Prescription Database. Treatment initiation was analyzed using the cumulative incidence method including death as a competing risk. Multiple logistic regression was used to identify potential predictors of treatment.

RESULTS

Among 5070 consecutive first-ever ischemic stroke patients without prior antidepressant treatment, the cumulative incidence of antidepressant treatment and prescription over 6 months was 35.2% (95% CI: 33.8-36.6). Overall 16.5% (95% CI: 15.5-17.6) started treatment within 14 days corresponding to 48.1% (95% CI: 45.8-50.5) of all treated patients, and the most widely prescribed group of antidepressants was selective serotonin reuptake inhibitors (86%). Increasing stroke severity was associated with higher odds of initiating treatment.

CONCLUSION

Antidepressant treatment in this real-life clinical setting was common and initiated early, in almost half the treated patients within 14 days. Our results suggest that special focus should be given to the severe strokes as they may have a greater risk of requiring treatment.

Return to work predictors after traumatic brain injury in a welfare state

OBJECTIVE

To identify predictors of return to work (RTW) and stable labour market attachment (LMA) after severe traumatic brain injury (TBI) in Denmark.

MATERIALS & METHODS

Patients aged 18-64 years, admitted to highly specialized neurorehabilitation after severe TBI 2004-2012 were included and followed up for ≤6 years. Weekly LMA data were retrieved from a national register of public assistance benefits. Weeks without or with supplemental public assistance benefits were defined as LMA weeks. Time of RTW was defined as first week with LMA. Stable LMA was defined as weeks with LMA ≥75% first year after RTW. Multivariable regressions were used to identify predictors of RTW and stable LMA among preinjury characteristics, injury severity, functional ability and rehabilitation trajectories.

RESULTS

For the analyses of RTW and stable LMA, 651 and 336 patients were included, respectively. RTW was significantly associated with age (adjusted subhazard ratio 0.98, 95% CI 0.97-0.99), education (1.83, 95% CI 1.16-2.89), supplemental benefits (3.97, 95% CI 2.04-7.71), no benefits (4.86, 95% CI 2.90-8.17), length of stay in acute care (0.77, 95% CI 0.60-0.99) and time period of injury (1.56, 95% CI 1.15-2.10). The only significant predictor of stable LMA was age (adjusted odds ratio 0.97, 95% CI 0.95-0.99).

CONCLUSION

RTW after severe TBI was associated with several socio-economic factors, whereas maintaining LMA depended on age only. We suggest that RTW rates could be improved by extensive rehabilitation targeting people that are older and low-educated, as these were less likely to RTW.

Socioeconomic inequality in clinical outcome among hip fracture patients: a nationwide cohort study

The evidence is limited regarding the association between socioeconomic status and the clinical outcome among patients with hip fracture. In this nationwide, population-based cohort study, higher education and higher family income were associated with a substantially lower 30-day mortality and risk of unplanned readmission after hip fracture.

INTRODUCTION

We examined the association between socioeconomic status and 30-day mortality, acute readmission, quality of in-hospital care, time to surgery and length of hospital stay among patients with hip fracture.

METHODS

This is a nationwide, population-based cohort study using prospectively collected data from the Danish Multidisciplinary Hip Fracture Registry. We identified 25,354 patients ≥65 years admitted with a hip fracture between 2010 and 2013 at Danish hospitals. Individual-level socioeconomic status included highest obtained education, family mean income, cohabiting status and migrant status. We performed multilevel regression analysis, controlling for potential confounders.

RESULTS

Hip fracture patients with higher education had a lower 30-day mortality risk compared to patients with low education (7.3 vs 10.0% adjusted odds ratio (OR) = 0.74 (95% confidence interval (CI) (0.63-0.88)). The highest level of family income was also associated with lower 30-day mortality (11.9 vs 13.0% adjusted OR = 0.77, 95% CI 0.69-0.85). Cohabiting status and migrant status were not associated with 30-day mortality in the adjusted analysis. Furthermore, patients with both high education and high income had a lower risk of acute readmission (14.5 vs 16.9% adjusted OR = 0.94, 95% CI 0.91-0.97). Socioeconomic status was, however, not associated with quality of in-hospital care, time to surgery and length of hospital stay.

CONCLUSIONS

Higher education and higher family income were associated with substantially lower 30-day mortality and risk of readmission after hip fracture.

Ischemic stroke subtype is associated with outcome in thrombolyzed patients

OBJECTIVES

The impact of ischemic stroke subtype on clinical outcome in patients treated with intravenous tissue-type plasminogen activator (IV-tPA) is sparsely examined. We studied the association between stroke subtype and clinical outcome in magnetic resonance imaging (MRI)-evaluated patients treated with IV-tPA.

MATERIAL AND METHODS

We conducted a single-center retrospective analysis of MRI-selected stroke patients treated with IV-tPA between 2004 and 2010. The Trial of ORG 10172 in Acute Stroke Treatment criteria were used to establish the stroke subtype by 3 months. The outcomes of interest were a 3-month modified Rankin Scale score of 0-1 (favorable outcome), and early neurological improvement defined as complete remission of neurological deficit or improvement of ≥4 on the National Institute of Health Stroke Scale at 24 h. The outcomes among stroke subtypes were compared with multivariable logistic regression.

RESULTS

Among 557 patients, 202 (36%) had large vessel disease (LVD), 153 (27%) cardioembolic stroke (CE), 109 (20%) small vessel disease, and 93 (17%) were of other or undetermined etiology. Early neurological improvement was present in 313 (56.4%) patients, and 361 (64.8%) patients achieved a favorable outcome. Early neurological improvement and favorable outcome were more likely in CE patients compared with LVD patients (odds ratio (OR), 2.1 (95% confidence interval, 1.4-3.3), and 2.0 (95% confidence interval, 1.2-3.3), respectively).

CONCLUSIONS

Cardioembolic stroke patients were more likely to achieve early neurological improvement and favorable outcome compared with LVD stroke following MRI-based IV-tPA treatment. This finding may reflect a difference in the effect of IV-tPA among stroke subtypes.

Effectiveness of intensification therapies in Danes with Type 2 diabetes who use basal insulin: a population-based study

AIMS

To examine the usage and real-life effectiveness of intensification therapies in people with Type 2 diabetes treated with basal insulin.

METHODS

We used population-based healthcare databases in Denmark during 2000-2012 to identify all individuals with a first basal insulin prescription (with or without oral drugs), and evaluated subsequent intensification therapy with bolus insulin, premixed insulin or glucagon-like peptide-1 (GLP-1) receptor agonists. Poisson regression was used to compute the adjusted relative risks of reaching glycaemic control targets.

RESULTS

We included 7034 initiators of basal insulin (median age 64 years, diabetes duration 5.3 years, 84% with oral co-medication and median (interquartile range) pre-insulin HbA_1c level 77 (65-92) mmol/mol [9.2% (8.1-10.6%)]. Of these, 3076 (43.7%) received intensification therapy after a median of 11 months: 58.5% with premixed insulin, 29.0% with bolus insulin, 10.6% with GLP-1 receptor agonists, and 1.9% with more than one add-on. Overall, 22% had attained an HbA_1c level of < 53 mmol/mol (< 7%) by 3-6 months after intensification, while 38% attained an HbA_1c < 58 mmol/mol (< 7.5%). Compared with premixed insulin intensification, attainment of HbA_1c < 53 and < 58 mmol/mol was similar with bolus insulin add-on [adjusted relative risk 1.03 (95% CI 0.86-1.24) and 1.02 (95% CI 0.91-1.15), and higher for GLP-1 receptor agonist add-on [adjusted relative risk 1.56 (95% CI 1.27-1.92) and 1.27 (1.10-1.47)].

CONCLUSIONS

Among people with Type 2 diabetes, 22 and 38% reached a target HbA_1c < 53 mmol/mol (< 7%) or < 58 mmol/mol (< 7.5%), respectively, after intensification of their basal insulin therapy. Compared with premixed insulin, target attainment was similar with bolus insulin and higher with GLP-1 receptor agonists.

Patient-level predictors of achieving early glycaemic control in Type 2 diabetes mellitus: a population-based study

AIMS

To identify individual predictors of early glycaemic control in people with Type 2 diabetes mellitus after initiation of first glucose-lowering drug treatment in everyday clinical practice.

METHODS

Using medical registries, we identified a population-based cohort of people with a first-time glucose-lowering drug prescription in Northern Denmark in the period 2000-2012. We used Poisson regression analysis to examine patient-level predictors of success in reaching early glycaemic control [HbA_1c target of < 53 mmol/mol (7%)] < 6 months after treatment start.

RESULTS

Among the 38 418 people (median age 63 years), 27 545 (72%) achieved early glycaemic control. The strongest predictor of achieving early control was pre-treatment HbA_1c level; compared with a pre-treatment HbA_1c level of ≤ 58 mmol/mol (7.5%), the adjusted relative risks of attaining early control were 0.63 (95% CI 0.61-0.64) for baseline HbA_1c levels of > 58 and ≤ 75 mmol/mol (> 7.5 and ≤ 9%), and 0.58 (95% CI 0.57-0.59) for a baseline HbA_1c level of > 9% (> 75 mmol/mol). All other examined predictors were only weakly associated with the chance of achieving early control. After adjustment, the only characteristics that remained independently associated with early control (in addition to high baseline HbA_1c ) were being widowed (adjusted relative risk 0.95; 95% CI 0.93-0.97) and having a high Charlson comorbidity index score (score ≥ 3; adjusted relative risk 0.94; 95% CI 0.90-0.97).

CONCLUSIONS

In a real-world clinical setting, people with Type 2 diabetes mellitus initiating glucose-lowering medication had a similar likelihood of achieving glycaemic control, regardless of sex, age, comorbidities and other individual factors; the only strong and potentially modifiable predictor was HbA_1c before therapy start.

Can public health registry data improve Emergency Medical Dispatch?

BACKGROUND

Emergency Medical Dispatchers make decisions based on limited information. We aimed to investigate if adding demographic and hospitalization history information to the dispatch process improved precision.

METHODS

This 30-day follow-up study evaluated time-critical emergencies in contact with the emergency phone number 112 in Denmark during 18 months. 'Time-critical' was defined as suspected First Hour Quintet (FHQ) (cardiac arrest, chest pain, stroke, difficulty breathing, trauma). The association of age, sex, and hospitalization history with adverse outcomes was examined using logistic regression. The predictive ability was assessed via area under the curve (AUC) and Hosmer-Lemeshow tests.

RESULTS

Of 59,943 patients (median age 63 years, 45% female), 44-45.5% had at least one chronic condition, 3880 (6.47%) died the day or the day after (primary outcome) calling 112. Age 30-59 was associated with increased adjusted odds ratio (OR) of death on day 1 of 3.59 [2.88-4.47]. Male sex was associated with an increased adjusted OR of death on day 1 of 1.37 [1.28-1.47]. Previous hospitalization with nutritional deficiencies (adjusted OR 2.07 [1.47-2.92]) and severe chronic liver disease (adjusted OR 2.02 [1.57-2.59]) was associated with a higher risk of death. For trauma patients, the discriminative ability of the model showed an AUC of 0.74 for death on day 1.

CONCLUSION

Increasing age, male sex, and hospitalization history was associated with increased risk of death on day 1 for FHQ 112 callers. Additional efforts are warranted to clarify the role for risk prediction tools in emergency medical dispatch.

How can the research potential of the clinical quality databases be maximized? The Danish experience

In Denmark, the need for monitoring of clinical quality and patient safety with feedback to the clinical, administrative and political systems has resulted in the establishment of a network of more than 60 publicly financed nationwide clinical quality databases. Although primarily devoted to monitoring and improving quality of care, the potential of these databases as data sources in clinical research is increasingly being recognized. In this review, we describe these databases focusing on their use as data sources for clinical research, including their strengths and weaknesses as well as future concerns and opportunities. The research potential of the clinical quality databases is substantial but has so far only been explored to a limited extent. Efforts related to technical, legal and financial challenges are needed in order to take full advantage of this potential.

Early glycaemic control among patients with type 2 diabetes and initial glucose-lowering treatment: a 13-year population-based cohort study

AIM

To examine real-life time trends in early glycaemic control in patients with type 2 diabetes between 2000 and 2012.

METHODS

We used population-based medical databases to ascertain the association between achievement of glycaemic control with initial glucose-lowering treatment in patients with incident type 2 diabetes in Northern Denmark. Success in reaching glycated haemoglobin (HbA1c) goals within 3-6 months was examined using regression analysis.

RESULTS

Of 38 418 patients, 91% started with oral glucose-lowering drugs in monotherapy. Metformin initiation increased from 32% in 2000-2003 to 90% of all patients in 2010-2012. Pretreatment (interquartile range) HbA1c levels decreased from 8.9 (7.6-10.7)% in 2000-2003 to 7.0 (6.5-8.1)% in 2010-2012. More patients achieved an HbA1c target of <7% (<53 mmol/mol) in 2010-2012 than in 2000-2003 [80 vs 60%, adjusted relative risk (aRR) 1.10, 95% confidence interval (CI) 1.08-1.13], and more achieved an HbA1c target of <6.5% [(<48 mmol/mol) 53 vs 37%, aRR 1.07 95% CI 1.03-1.11)], with similar success rates observed among patients aged <65 years without comorbidities. Achieved HbA1c levels were similar for different initiation therapies, with reductions of 0.8% (from 7.3 to 6.5%) on metformin, 1.5% (from 8.1 to 6.6%) on sulphonylurea, 4.0% (from 10.4 to 6.4%) on non-insulin combination therapies, and 3.8% (from 10.3 to 6.5%) on insulin monotherapy.

CONCLUSIONS

Pretreatment HbA1c levels in patients with incident type 2 diabetes have decreased substantially, which is probably related to earlier detection and treatment in accordance with changing guidelines. Achievement of glycaemic control has improved, but 20% of patients still do not attain an HbA1c level of <7% within the first 6 months of initial treatment.

The risk of venous thromboembolism, myocardial infarction, stroke, major bleeding and death in patients undergoing total hip and knee replacement: a 15-year retrospective cohort study of routine clinical practice

We examined the risk of thrombotic and major bleeding events in patients undergoing total hip and knee replacement (THR and TKR) treated with thromboprophylaxis, using nationwide population-based databases. We identified 83 756 primary procedures performed between 1997 and 2011. The outcomes were symptomatic venous thromboembolism (VTE), myocardial infarction (MI), stroke, death and major bleeding requiring hospitalisation within 90 days of surgery. A total of 1114 (1.3%) and 483 (0.6%) patients experienced VTE and bleeding, respectively. The annual risk of VTE varied between 0.9% and 1.6%, and of bleeding between 0.4% and 0.8%. The risk of VTE and bleeding was unchanged over a 15-year period. A total of 0.7% of patients died within 90 days, with a decrease from 1% in 1997 to 0.6% in 2011 (p < 0.001). A high level of comorbidity and general anaesthesia were strong risk factors for both VTE and bleeding, with no difference between THR and TKR patients. The risk of both MI and stroke was 0.5%, which remained unchanged during the study period. In this cohort study of patients undergoing THR and TKR patients in routine clinical practice, approximately 3% experienced VTE, MI, stroke or bleeding. These risks did not decline during the 15-year study period, but the risk of dying fell substantially.

Statin use and rupture of abdominal aortic aneurysm

BACKGROUND

Ruptured abdominal aortic aneurysm (rAAA) is associated with high mortality. Research suggests that statins may reduce abdominal aortic aneurysm (AAA) growth and improve rAAA outcomes. However, the clinical impact of statins remains uncertain in relation to both the risk and prognosis of rAAA.

METHODS

This nationwide, population-based, combined case-control and follow-up study included all patients (aged at least 50 years) with a first-time hospital admission for rAAA and 1:1 matched AAA controls without rupture in Denmark from 1996 to 2008. Individual-level data on preadmission drug use, co-morbidities, socioeconomic markers, healthcare contacts and death were obtained from Danish nationwide registries.

RESULTS

The study included 3584 cases and 3584 matched controls. Current statin use was registered for 418 patients with rAAA (11.7 per cent) and 539 AAA controls (15.0 per cent), corresponding to an age- and sex-matched odds ratio (OR) of 0.70 (95 per cent confidence interval (c.i.) 0.60 to 0.81) for rAAA in current statin users versus never users. The decreased risk of rAAA remained after adjustment for potential confounding factors (adjusted OR 0.73, 0.61 to 0.86). The overall 30-day mortality rate from time of hospital admission among patients with rAAA was 46.1 per cent in current statin users compared with 59.3 per cent in never users (adjusted mortality rate ratio (MRR) 0.80, 95 per cent c.i. 0.68 to 0.95). Patients who had formerly used statins did not have reduced mortality (adjusted MRR 0.98, 0.78 to 1.22).

CONCLUSION

Statin use was associated with a reduced risk of rAAA and lower case fatality following rAAA. These results support current guidelines that recommend statin therapy in patients diagnosed with AAA.

Quality-of-care initiative in patients treated surgically for perforated peptic ulcer

BACKGROUND

Mortality and morbidity are considerable after treatment for perforated peptic ulcer (PPU). Since 2003, a Danish nationwide quality-of-care (QOC) improvement initiative has focused on reducing preoperative delay, and improving perioperative monitoring and care for patients with PPU. The present study reports the results of this initiative.

METHODS

This was a nationwide cohort study based on prospectively collected data, involving all hospitals caring for patients with PPU in Denmark. Details of patients treated surgically for PPU between September 2004 and August 2011 were reported to the Danish Clinical Register of Emergency Surgery. Changes in baseline patient characteristics and in seven QOC indicators are presented, including relative risks (RRs) for achievement of the indicators.

RESULTS

The study included 2989 patients. An increasing number fulfilled the following four QOC indicators in 2010-2011 compared with the first 2 years of monitoring: preoperative delay no more than 6 h (59·0 versus 54·0 per cent; P = 0·030), daily monitoring of bodyweight (48·0 versus 29·0 per cent; P < 0·001), daily monitoring of fluid balance (79·0 versus 74·0 per cent; P = 0·010) and daily monitoring of vital signs (80·0 versus 68·0 per cent; P < 0·001). A lower proportion of patients had discontinuation of routine prophylactic antibiotics (82·0 versus 90·0 per cent; P < 0·001). Adjusted 30-day mortality decreased non-significantly from 2005-2006 to 2010-2011 (adjusted RR 0·87, 95 per cent confidence interval 0·76 to 1·00), whereas the rate of reoperative surgery remained unchanged (adjusted RR 0·98, 0·78 to 1·23).

CONCLUSION

This nationwide quality improvement initiative was associated with reduced preoperative delay and improved perioperative monitoring in patients with PPU. A non-significant improvement was seen in 30-day mortality.

Increased one-year risk of symptomatic venous thromboembolism following total hip replacement

We examined the one-year risk of symptomatic venous thromboembolism (VTE) following primary total hip replacement (THR) among Danish patients and a comparison cohort from the general population. From the Danish Hip Arthroplasty Registry we identified all primary THRs performed in Denmark between 1995 and 2010 (n = 85 965). In all, 97% of patients undergoing THR received low-molecular-weight heparin products during hospitalisation. Through the Danish Civil Registration System we sampled a comparison cohort who had not undergone THR from the general population (n = 257 895). Among the patients undergoing THR, the risk of symptomatic VTE was 0.79% between 0 and 90 days after surgery and 0.29% between 91 and 365 days after surgery. In the comparison cohort the corresponding risks were 0.05% and 0.12%, respectively. The adjusted relative risks of symptomatic VTE among patients undergoing THR were 15.84 (95% confidence interval (CI) 13.12 to 19.12) during the first 90 days after surgery and 2.41 (95% CI 2.04 to 2.85) during 91 to 365 days after surgery, compared with the comparison cohort. The relative risk of VTE was elevated irrespective of the gender, age and level of comorbidity at the time of THR.

We concluded that THR was associated with an increased risk of symptomatic VTE up to one year after surgery compared with the general population, although the absolute risk is small.

Post-stroke fatigue and return to work: a 2-year follow-up

BACKGROUND

Post-stroke fatigue may affect the ability to return to work but quantitative studies are lacking.

METHOD

We included 83 first-ever stroke patients <60 years and employed either full-time (n = 77) or part-time (n = 6) at baseline. The patients were recruited from stroke units at Aarhus University Hospital between 2003 and 2005 and were followed for 2 years. Fatigue was assessed by the Multidimensional Fatigue Inventory. Pathological fatigue was defined as a score ≥12 on the General Fatigue dimension. Return to paid work was defined as working at least 10 h per week. Data were analyzed using multivariable logistic regression.

RESULTS

A total of 58% of patients had returned to paid work after 2 years. The adjusted Odds Ratio (OR) for returning to paid work was 0.39 (95% confidence interval (CI) 0.16-1.08) for patients with a General Fatigue score ≥12 at baseline. Persisting pathological fatigue after 2 years of follow-up was associated with a lower chance of returning to paid work [adjusted OR 0.29 (95% CI 0.11-0.74)]. Higher scores of General Fatigue at follow-up also correlated negatively with the chance of returning to paid work when analyzing fatigue on a continuous scale (adjusted OR 0.87, 95% CI 0.80-0.94 for each point increase in General Fatigue).

CONCLUSIONS

Post-stroke fatigue appears to be an independent determinant of not being able to resume paid work following stroke.

Low risk of vascular events following urgent treatment of transient ischaemic attack: the Aarhus TIA study

BACKGROUND AND PURPOSE

Patients with TIA have a high short-time risk of stroke and an increased long-term risk of ischaemic vascular events compared with the general population. Urgent intervention may reduce short-time stroke risk, but little is known about the effect beyond 3 months. We examined 1-year outcome and risk factor management in patients with TIA after urgent intervention.

METHODS

All patients with TIA referred to Aarhus University Hospital 1 March 2007-28 February 2008 were seen by an acute TIA team (ATT), integrating outpatient care and stroke unit facilities. Preventive treatment was initiated immediately, including fast-track surgery for carotid stenosis. Follow-up including nurse-conducted health counseling was carried out after 7, 90, and 365 days.

RESULTS

A total of 306 patients were included. Stroke, myocardial infarction, or vascular death occurred in 5.2% during 1 year of follow-up. The cumulated stroke rate was 1.6%, 2.0%, and 4.4% after 7, 90, and 365 days, respectively, compared to expected 4.5% [relative risk (RR) 0.36, 95% CI 0.13-0.98] and 7.5% (RR 0.26, 95% CI 0.11-0.63) after 7 and 90 days using ABCD(2) criteria. Recurrent TIA occurred in 10.2% (n = 32). Secondary prevention targets were attained in 47.6% after 1 year. Carotid surgery was performed in 8.1%; median time to operation was 11 days after contact with the ATT.

CONCLUSION

Urgent intervention after TIA by an ATT covering outpatient and stroke unit facilities combined with nurse-conducted health counseling is associated with a low 1-year risk of new vascular events and may improve risk factor control.

Paediatric arterial ischaemic stroke and cerebral sinovenous thrombosis in Denmark 1994-2006: a nationwide population-based study

AIM

To assess the incidence rates (IR), clinical characteristics, risk factors, treatment and outcomes of paediatric arterial ischaemic stroke (AIS) and cerebral sinovenous thrombosis (CSVT).

METHODS

Using population-based, nationwide medical registries, we identified all patients aged 0-18 years at the time of hospitalization with first-ever AIS and/or CSVT in Denmark between 1994 and 2006. Medical records were retrieved and reviewed.

RESULTS

We identified 211 patients with AIS and 40 patients with CSVT corresponding to IRs of 1.33 (95% CI 1.16-1.52) and 0.25 (95% CI 0.19-0.34) per 100,000 person-years, respectively. The IRs peaked in infancy (<1 year) for both AIS and CSVT with an additional peak among adolescents (15-18 years) for CSVT. The IR of AIS increased 3.9% per year (p=0.036), whereas no changes were found for CSVT. In total, 48.2% of the patients received antithrombotic treatment; no major complications were observed. All-cause and thrombosis-related 30-day case fatality ratios were 3.6% and 2.4%, respectively; neurological sequelae were found in 56.2% of patients.

CONCLUSION

The IR of AIS was highest in infants and had increased with 3.9% annually during the observation period. The IR of CSVT had an additional peak in adolescence and remained unchanged over time.

Variation in use of peroperative inotropic support therapy in cardiac surgery: time for reflection?

BACKGROUND

There is no well-established evidence-based clinical guidelines on the most appropriate use of peroperative inotropic support in cardiac surgery. We aimed to identify patient- and procedure-related factors associated with the use of peroperative inotropic support and to estimate physician-level variation.

METHODS

A population-based study using data from the Western Denmark Heart Registry on 3585 consecutive cardiac surgery cases from three university hospitals. Inotropic support was defined as infusion of inotropic drugs or nor epinephrine at the separation from cardiopulmonary bypass. Poisson's regression modelling was used to determine predictors of inotropic support and to compare use of high-dose inotropic support among experienced cardiac anaesthesiologists.

RESULTS

We identified a range of factors that were independently associated with an increased use of inotropic support therapy including pre-operative left ventricular dysfunction, pre-operative renal dysfunction, complex procedures, prior cardiac surgery, emergency surgery, pre-operative pulmonary hypertension, critical pre-operative state, extended extra corporal circulation-time and female gender. Further, we found substantial variation in use of inotropic support both at hospital- and at physician-level. The adjusted odds ratio of high-intensity inotropic support varied significantly at physician level from 2.3 [95% confidence interval (CI) 1.83-2.71] to 0.3 (95% CI 0.15-0.61) when the individual physicians were compared with the rest.

CONCLUSION

The use of inotropic support during cardiac surgery is associated with the pre-operative state of the patient, as well as type of surgery. However, the present study indicates that use of peroperative inotropic support is also highly dependent on physician's preferences, indicating the need for an evidence-based approach when initiating inotropic therapy in cardiac surgery.

Short- and long-term mortality following primary total hip replacement for osteoarthritis

We evaluated the short-term of 0 to 90 days and the longer term, up to 12.7 years, mortality for patients undergoing primary total hip replacement (THR) in Denmark in comparison to the general population. Through the Danish Hip Arthroplasty Registry we identified all primary THRs undertaken for osteoarthritis between 1 January 1995 and 31 December 2006. Each patient (n = 44 558) was matched at the time of surgery with three people from the general population (n = 133 674). We estimated mortality rates and mortality rate ratios with 95% confidence intervals for THR patients compared with the general population. There was a one-month period of increased mortality immediately after surgery among THR patients, but overall short-term mortality (0 to 90 days) was significantly lower (mortality rate ratio 0.8; 95% confidence interval 0.7 to 0.9). However, THR surgery was associated with increased short-term mortality in subjects under 60 years old, and among THR patients without comorbidity. Long-term mortality was lower among THR patients than in controls (mortality rate ratio 0.7; 95% confidence interval 0.7 to 0.7).

Overall, THR was associated with lower short- and long-term mortality among patients with osteoarthritis. Younger patients and patients without comorbidity before surgery may also experience increased mortality after THR surgery, although the absolute risk of death is small.

Risk factors for venous thromboembolism in patients undergoing total hip replacement and receiving routine thromboprophylaxis

BACKGROUND

Data on the risk factors for venous thromboembolism among patients undergoing total hip replacement and receiving pharmacological thromboprophylaxis are limited. The purpose of this study was to examine potential patient-related risk factors for venous thromboembolism following total hip replacement in a nationwide follow-up study.

METHODS

Using medical databases, we identified all patients who underwent primary total hip replacement and received pharmacological thromboprophylaxis in Denmark from 1995 to 2006. The outcome measure was hospitalization with venous thromboembolism within ninety days of surgery. We considered age, sex, indication for primary total hip replacement, calendar year of surgery, and comorbidity history as potential risk factors.

RESULTS

The overall rate of hospitalization for venous thromboembolism within ninety days following a primary total hip replacement was 1.02% (686 hospitalizations after 67,469 procedures) at a median of twenty-two days. The incidence of symptomatic deep venous thrombosis and of nonfatal pulmonary embolism was 0.7% (499 of 67,469) and 0.3% (205 of 67,469), respectively. The incidence of death due to venous thromboembolism or from all causes was 0.05% (thirty-eight patients) and 1.0% (678 patients), respectively. Patients with rheumatoid arthritis had a reduced relative risk for venous thromboembolism compared with patients with primary osteoarthritis (adjusted relative risk = 0.47; 95% confidence interval, 0.25 to 0.90). Patients with a high score on the Charlson comorbidity index had an increased relative risk for venous thromboembolism compared with patients with a low score (adjusted relative risk = 1.45; 95% confidence interval, 1.02 to 2.05). Patients with a history of cardiovascular disease (relative risk = 1.40; 95% confidence interval, 1.15 to 1.70) or prior venous thromboembolism (relative risk = 8.09; 95% confidence interval, 6.07 to 10.77) had an increased risk for venous thromboembolism compared with patients without that history.

CONCLUSIONS

The cumulative incidence of a venous thromboembolism within ninety days of surgery among patients with total hip replacement receiving pharmacological thromboprophylaxis was 1%. This information on the associated risk factors could be used to better anticipate the risk of venous thromboembolism for an individual patient.

Risk of revision of a total hip replacement in patients with diabetes mellitus: a population-based follow up study

We have evaluated the extent to which diabetes affects the revision rate following total hip replacement (THR). Through the Danish Hip Arthroplasty Registry we identified all patients undergoing a primary THR (n = 57 575) between 1 January 1996 and 31 December 2005, of whom 3278 had diabetes. The presence of diabetes among these patients was identified through the Danish National Registry of Patients and the Danish National Drug Prescription Database. We estimated the relative risk for revision and the 95% confidence intervals for patients with diabetes compared to those without, adjusting for the confounding factors. Diabetes is associated with an increased risk of revision due to deep infection (relative risk = 1.45 (95% confidence interval 1.00 to 2.09), particularly in those with type 2 diabetes (relative risk = 1.49 (95% confidence interval 1.02 to 2.18)), those with diabetes for less than five years prior to THR (relative risk = 1.69 (95% confidence interval 1.24 to 2.32)), those with complications due to diabetes (relative risk = 2.11 (95% confidence interval 1.41 to 3.17)), and those with cardiovascular comorbidities prior to surgery (relative risk = 2.35 (95% confidence interval 1.39 to 3.98)). Patients and surgeons should be aware of the relatively elevated risk of revision due to deep infection following THR in diabetes particularly in those with insufficient control of their glucose level.

Enforcing centralization for gastric cancer in Denmark

BACKGROUND

Population-based data on the early postoperative outcome after surgery for gastric cancer are very sparse. We examined the development in the quality of surgery and early postoperative outcomes in Denmark following centralization of gastric cancer surgery and implementation of national clinical guidelines.

METHODS

All patients in Denmark who underwent resection with curative intent for gastric cancer between 1st July 2003 and 31st December 2008 in one of five university hospitals were registered in a national database. Data on surgical quality and mortality were obtained from the database and compared with the results from the period before centralization (1999-2003).

RESULTS

A total of 416 patients underwent resection in the study period. The risk of anastomotic leakages for the whole period was 5.0% (95%CI; 3.2-7.7) compared to 6.1% (95%CI; 4.3-8.6) before centralization, whereas the 30-days hospital mortality was 2.4% (95%CI; 1.2-4.4) compared to 8.2% (95%CI; 6.0-10.4) before centralization. In addition, the percentage of patients with at least 15 lymph nodes removed increased during the study period from 19 in 2003 to 76 in 2008.

CONCLUSIONS

Centralization of gastric cancer surgery in Denmark and implementation of national clinical guidelines monitored by a national database was associated with improvements in surgical quality and substantially lower in-hospital mortality.

Smoking and venous thromboembolism: a Danish follow-up study

BACKGROUND

Large-scale prospective studies are needed to assess whether smoking is associated with venous thromboembolism (VTE) (i.e. deep venous thrombosis and pulmonary embolism) independently of established risk factors.

OBJECTIVE

To investigate the association between smoking and the risk of VTE among middle-aged men and women.

METHODS

From 1993 to 1997, 27,178 men and 29,875 women, aged 50-64 years and born in Denmark, were recruited into the Danish prospective study 'Diet, Cancer and Health'. During follow-up, VTE cases were identified in the Danish National Patient Registry. Medical records were reviewed and only verified VTE cases were included in the study. Baseline data on smoking and potential confounders were included in gender stratified Cox proportional hazard models to asses the association between smoking and the risk of VTE. The analyses were adjusted for alcohol intake, body mass index, physical activity, and in women also for use of hormone replacement therapy.

RESULTS

During follow-up, 641 incident cases of VTE were verified. We found a positive association between current smoking and VTE, with a hazard ratio of 1.52 (95% CI, 1.15-2.00) for smoking women and 1.32 (95% CI, 1.00-1.74) for smoking men, and a positive dose-response relationship. Former smokers had the same hazard as never smokers.

CONCLUSIONS

Smoking was an independent risk factor for VTE among middle-aged men and women. Former smokers have the same risk of VTE as never smokers, indicating acute effects of smoking, and underscoring the potential benefits of smoking cessation.

Arterial cardiovascular events, statins, low-dose aspirin and subsequent risk of venous thromboembolism: a population-based case-control study

BACKGROUND

Atherosclerotic disease has been associated with the risk of venous thromboembolism, but the available data are conflicting. There are similar confusions regarding the association of the use of aspirin and statins with venous thromboembolism.

OBJECTIVES

To determine whether arterial cardiovascular events, use of statins and low-dose aspirin were associated with the risk of venous thromboembolism.

PATIENTS AND METHODS

In this population-based case-control study, we identified 5824 patients with venous thromboembolism and 58 240 population controls with a complete hospital and prescription history. We used logistic regression to estimate the relative risk of venous thromboembolism, adjusted for potentially confounding factors.

RESULTS

Patients with a history of arterial cardiovascular events had a clearly increased relative risk. An event within 3 months before the index date conferred large increases in risk [relative risk 4.22 (95% confidence interval (CI), 2.33-7.64) after myocardial infarction, 4.41 (95% CI, 2.92-6.65) after stroke]. Myocardial infarction more than 3 months before the index date was not significantly associated with risk, although there was a relative risk of 1.29 (95% CI, 1.05-1.57) for myocardial infarction more than 60 months previously. A history of stroke was associated with small increases in risk after 3 months. Current use of statins was associated with a reduced risk of venous thromboembolism [relative risk=0.74 (95% CI, 0.63-0.85)]. Aspirin use was not associated with risk.

CONCLUSIONS

Patients with cardiovascular events are at a short-term increased risk of venous thromboembolism. Statins might prevent venous thromboembolism but aspirin does not. However, as the study is non-randomized residual confounding cannot be excluded.