Health economic impact of early versus delayed treatment of herpes simplex virus encephalitis in the UK
Institute of Infection, Veterinary and Ecological Sciences, University of Liverpool, 8 West Derby Street, Liverpool, UK
Tropical and Infectious Disease Unit, Liverpool University Hospitals NHS Foundation Trust, Liverpool
National Institute for Health Research (NIHR) Health Protection Research Unit (HPRU) in Emerging and Zoonotic Infections, Institute of Infection, Veterinary and Ecological Sciences, University of Liverpool, Liverpool, UK
Clinical Sciences, Liverpool School of Tropical Medicine, Liverpool
Encephalitis International, Malton North Yorkshire, UK
The Walton Centre NHS Foundation Trust, Liverpool
Alder Hey Hospital Children’s NHS Foundation Trust, Liverpool, UK
Public Health Economics Group, Exeter, UK
The Pandemic Institute, Liverpool, UK
#Corresponding author; email: tsolomon@liverpool.ac.ukAbstract
Objective
Thanks to the introduction of recent national guidelines for treating herpes simplex virus (HSV) encephalitis health outcomes have improved. This paper evaluates the costs and the health-related quality of life implications of these guidelines.
Design and setting
A sub-analysis of data from a prospective, multi-centre, observational cohort ENCEPH-UK study conducted across 29 hospitals in the UK from 2012 to 2015.
Study participants
Data for patients aged ≥16 years with a confirmed HSV encephalitis diagnosis admitted for treatment with aciclovir were collected at discharge, 3 and 12 months.
Primary and secondary outcome measures
Patient health outcomes were measured by the Glasgow outcome score (GOS), modified ranking score (mRS), and the EuroQoL; health care costs were estimated per patient at discharge from hospital and at 12 months follow-up. In addition, Quality Adjusted Life years (QALYs) were calculated from the EQ-5D utility scores. Cost-utility analysis was performed using the NHS and Social Scare perspective.
Results
A total of 49 patients were included, 35 treated within 48 hours “early” (median [IQR] 8.25 [3.7-20.5]) and 14 treated after 48 hours (median [IQR] 93.9 [66.7 - 100.1]). At discharge, 30 (86%) in the early treatment group had a good mRS outcome score (0–3) compared to 4 (29%) in the delayed group. EQ-5D-3L utility value at discharge was significantly higher for early treatment (0.609 vs 0.221, p<0.000). After adjusting for age and symptom duration at admission, early treatment incurred a lower average cost at discharge, £23,086 (95% CI: £15,186 to £30,987) vs £42,405 (95% CI: £25,457 to £59,354) [p<0.04]. A -£20,218 (95% CI: -£52,173 to £11,783) cost difference was observed at 12-month follow-up post discharge.
Conclusions
This study suggests that early treatment may be associated with better health outcomes and reduced patient healthcare costs, with a potential for savings to the NHS with faster treatment.
Article Summary
Strengths and limitations of this study
- - Admissions to acute hospitals with suspected encephalitis, using predetermined inclusion criteria were recruited across 29 hospitals in the UK within a 3-year period, giving the largest cohort of prospectively recruited HSV encephalitis cases in the UK to date.
- - Precise definitions to characterise those individuals with proven HSV encephalitis were applied thus ensuring accurate diagnoses.
- - Individuals were followed up systematically for 12 months after discharge for clinical, and quality of life data providing the first study to assess the effect of treatment delays on health care resources, costs and health related quality of life.
- - The analysis is limited by its relatively small sample size due to it being a rare disease, and the case record forms although thorough may not capture all health care costs incurred. This is particularly so for primary care and community care contact outside of the study hospitals.
Article notes
Competing Interest Statement
The authors have declared no competing interest.
Funding Statement
This work represents independent research funded by the National Institute for Health Research (NIHR) under its Programme Grants for Applied Research scheme (RP-PG-0108-10048).
Summary of Updates:
Introduction
Herpes simplex virus (HSV) encephalitis is a rare but severe brain infection, resulting in inflammation of the brain parenchyma, which causes significant morbidity and continues to have a mortality of 10%, even when treated with antiviral drugs (1,2). In the United Kingdom (UK), approximately 1 in 250,000 to 500,000 people are newly diagnosed with HSV encephalitis annually (3). Early symptoms of HSV encephalitis include flu like symptoms and lethargy which are also common to a variety of infections while the later symptoms such as speech problems and seizures tend to mimic more common brain conditions such as stroke (4). The non-specific range of symptoms have been associated with delayed diagnosis and treatment (5).
A UK study showed that the median time to treatment for suspected HSV encephalitis was 48 (range: 2-432) hours (6). Since the publication of the clinical guidelines for the management of suspected encephalitis in 2012, more recent studies have shown a reduced median time to treatment for suspected encephalitis to around 15 hours (7, 8)). But while a number of studies have reported on the favourable (modified Rankin Scale of 0–3) clinical outcomes associated with the early treatment (≤ 48 hours from admission) of HSV encephalitis (9–15), there is limited research into the potential impact on healthcare resource utilisation and costs.
We sought to evaluate the healthcare cost and health outcome implications of the time interval from hospital admission to HSV encephalitis treatment with aciclovir. We assessed the cost per quality adjusted life year and estimated the potential cost savings to the National Health Service (NHS) from improving treatment times in the UK. For the long-term impact, we hypothesise that treating HSV encephalitis patients within 48 hours would result in better neurological outcomes and lower healthcare costs.
Methods and analysis
Study design and setting
A sub-analysis was conducted as part a wider prospective, multi-centre, observational cohort ENCEPH-UK study on ‘Understanding and Improving the Outcome of Encephalitis’ conducted across 29 hospitals in the UK from 2012 to 2015 (16).
Study participants
This sub-analysis was restricted to patients with confirmed HSV encephalitis aged ≥ 16 years admitted for treatment with aciclovir 10mg/kg three times daily (Figure 1). Patients were stratified into early treatment for those treated within 48 hours from admission and delayed treatment for those treated ≥ 48 hours from admission. The 48 hour cut-off was chosen based, as earlier mentioned, on the evidence of significantly better neurological outcomes and health related quality of life for patients who receive treatment within this time interval from admission compared to patients with delayed treatment (6,9).
Data collection
Patient data were collected prospectively using the standardised Case Report Forms (CRF) as part of the wider ENCEPH-UK study (16).
Patient demographic and clinical data
Patient demographics – age and gender, clinical data – symptom duration, Charlson comorbidity index, immunocompetent status, and Glasgow Coma Score (GCS) severity were recorded on admission. Clinical outcomes including modified Rankin Scale (mRS) where 0 to 3 represented good outcomes whilst 4-6 represented poor outcomes and Glasgow Outcome Score (GOS) on a 1 (death) to 5 (good recovery) scale, were recorded at hospital discharge and at 3 and 12months follow-up.
Resource use data and unit costs
Resource use data were collected at hospital discharge, 3 and at 12 months follow-up. This included: (1) length of hospital stay disaggregated into general ward and intensive care unit (ICU) to accurately reflect time spent in the different wards; (2) diagnostic tests [Lumbar Puncture (LP), Electroencephalogram (EEG), Magnetic Resonance Imaging (MRI) and Computerised Tomography (CT)]; (3) hospital transfer for those patients that required specialised care; and (4) outpatient follow-up costs including follow-up diagnostics, readmission and rehabilitation. Specific unit costs for hospital stay in ICU and general neurological wards were used to accurately reflect resources used in the different wards. The location of clinic appointment during follow-up was recorded which allowed for separate costs to be applied dependent on the location/type of the appointment. Data on each patients’ location at discharge and follow-up time points were recorded, noted either as home, rehabilitation or other. An average cost per episode of neurological rehabilitation for patients with an acquired brain injury was used to derive the cost of rehabilitation. Individual patient resource use data were applied to the corresponding unit costs obtained from the NHS reference costs 2018/19, drugs and pharmaceutical electronic market information tool (eMIT), 2019 where appropriate (17,18). All costs are in Great British pounds (GBP£) and a detailed breakdown of the resources utilised and unit costs are available in Table 1.
Cost utility analysis
We conducted a cost-utility analysis of early treatment compared to delayed treatment of HSV encephalitis with aciclovir on the imputed data. The differences in health outcomes (QALYs) and costs between the two treatment groups at 12 months follow-up was estimated using generalised linear models (GLMs), controlling for baseline covariates. While non-parametric methods of evaluation are typically preferred, results are median estimates which are not appropriate for the decision maker, who would prefer mean estimates (19). A GLM with a gaussian family and log link was used to predict the mean QALY at 12 months follow-up (20).
A GLM model was fitted, controlling for age, gender and symptom duration at admission and treatment, to minimise their independent effect on the resource use and costs. A simpler model with only treatment group as a covariate was also fitted, however model comparison using the Akaike Information Criterion (AIC) and the Bayesian information Criterion (BIC) showed minimal differences in AIC and BIC estimates (Appendix 1) and as such, the model with age and symptom duration as covariates was selected to adjust for the potential confounding bias.
The mean costs and QALYs were predicted from the GLM model. The incremental cost effectiveness ratio (ICER) – a ratio of difference in predicted mean costs to difference in predicted mean QALYs were presented with their confidence intervals. To explore the uncertainty around the ICER, a non-parametric bootstrap technique with 1000 iterations was employed on the mean cost and QALY predictions. Results were presented with the bias corrected confidence intervals on the mean estimates as well as a cost effectiveness plane (CEP).
We also projected the estimated average annual savings to the NHS, based on annual incidence rate of 0.4 per 100,000 (21), if HSV encephalitis patients received treatment within 48 hours from admission. We assumed that the proportions of patients in each time to treatment group were representative of management of HSV encephalitis across the UK to scale up costs to a national level. The analysis was undertaken from the UK NHS perspective and all costs were expressed in pounds (£) for the year 2018/19. All statistical analyses were conducted in Stata/IC 14.1 for Windows.
Missing data
Missing data were reported in variables – GCS severity, mRS and GOS scores as well as EQ-5D utility scores. Missing values were explored to assess the type of missingness. GSC severity at admission and mRS and GOS at discharge and follow-up contained missing values that were determined to be missing at random, as their probability of being missing was independent of unobserved data. Multiple imputation using chained equations (MICE) with a predictive mean matching model was used to impute the missing data on GCS severity, mRS and GOS using 20 imputations. MICE technique allows the for the simultaneous imputation of multiple variables while the predictive mean matching model maintains the original distribution structure of the data (22, 23). The Fraction of Missing Information (FMI) test statistic of 35% validated the choice of 20 imputations as sufficient.
Statistical analyses
Patient demographics and clinical characteristics at baseline were presented as means ± standard deviations (SD), medians with interquartile ranges [IQR], and frequencies with percentages as appropriate. Group differences were assessed using the student’s t-test or Mann–Whitney U test and the Chi-squared or Fisher’s exact test for continuous and categorical variables, respectively. Group demographic clinical outcome differences were considered significant at p ≤0.05.
Results
Baseline demographic and clinical Characteristics
Of 341 suspected encephalitis patients recruited, 49 had HSV encephalitis and were stratified into two groups based on the timing of their first dose of aciclovir treatment (Table 2). 35 (71%) patients received early while 14 (29%) patients received delayed treatment, at a median of 8.3 [IQR: 3.7 to 20.5] and 93.90 [66.70 to 100.08] hours from admission, respectively. A comparison of the early versus delayed treatment groups revealed significant differences in mean age (54.3 [SD: 16.5] vs. 65.8 [SD: 16.3]; P=0.031) and median symptom duration before hospital admission of (4.0 [IQR 3-7] vs. 1 [IQR 0-5]; P=0.014) days, respectively.
Significant differences in clinical outcomes at discharge between patient treatment groups were reported for mRS (p<0.0002) and GOS score at discharge (p=0.024). 71% (10/14) of patients in the delayed treatment group had a poor mRS score at discharge compared to 14% (5/35)) of the early treatment group. Of the 14 patients whose treatment was delayed, 3 died while only 1 patient had good recovery at discharge. 10 of the 35 patients who received treatment within 48 hours of admission had good recovery, with no neurological impairment at discharge. Patients receiving early treatment reported a significantly higher median EQ-5D utility values at discharge compared to delayed treatment patients, 0.61 (0.5 – 0.8) vs. 0.35 (0 – 0.4): p<0.0001).
The results in Table 3 below show a significant difference in the median [IQR] length of hospital stay between the early and delayed treatment groups; 31 (22–64) versus 95 (29–157) days respectively, (p=0.046). There was no significant difference in the duration of treatment with aciclovir between the two groups (p=0.13)
Unadjusted mean healthcare costs
Results in Table 4 below show that the unadjusted mean costs incurred by the patient at discharge was lower for patients receiving early treatment £22,854 (95% CI: £14,180 to £31,528) compared to delayed treatment £42,902 (95% CI: £25,859 to £59,945). The length of stay in hospital was the main driver of initial hospital admission costs, accounting for 91% of all costs incurred by patients treated early and 95% in the delayed treatment group. The mean cost of hospital stay was higher in the delayed treatment group in both the general ward and the ICU.
Patients receiving early treatment incurred lower average follow-up costs of £1,216 (95% CI: £172 to £2351) compared to £1,473 (95% CI: £1166 to £4112) for those whose treatment was delayed. Similarly, the average patient cost at 12 months was lower for patients receiving early treatment in comparison to delayed treatment patients, £24,372 (95% CI: £15,007 to £33,738) versus £45,211 (95% CI: £23,014 to 67,409), respectively.
Adjusted mean healthcare costs
Results show a higher mean patient cost at discharge of £23,086 (95% CI: £15,186 to £30,987) vs £42,405 (95% CI: £25,457 to £59,353) for patients in the early and delayed treatment groups respectively after adjusting for age and symptom duration before admission. This resulted in a difference of -£19,319 (95% C.I: -£37,783 to -£854) (Table 5).
Follow-up data were available on patient’s clinical outcomes during the first 12 months post inpatient stay for 43/49 patients. Four of the six patients with data missing on follow-up appointments were in the delayed treatment group. 3 patients died prior to discharge, all of whom were in the delayed treatment group, and a further 2 during the 12-month follow-up period. After adjusting for age and symptom duration before admission, there was a difference of -£6,570 (95% CI: - £25,326 to £12,185) in mean patient follow-up costs. The average costs from admission to 12 months post-discharge for patients on early treatment were £38,359 (95% CI: 22,470 to £54,247) versus £58,576 (95% CI: 32,141 to £85,011) for those receiving delayed treatment, resulting in difference of -£20,217 (95% CI: -£52,173 to £11,738).
Cost utility analysis
Table 6 presents a summary of the bootstrapped estimates for mean costs and QALYs for patients in the early and delayed treatment groups. The average cost per patient was higher in the delayed treatment group [£76,071 (BC 95% CI: £58,037 - £105,743) compared to the early treatment group [£34,866 (BC 95% CI: £30,715 - £39,423)]. Similarly, patients receiving early treatment reported better health outcomes with a mean QALY at 12 months of 0.613 (BC 95% CI: 0.599 - 0.630) compared to patients receiving delayed treatment 0.492 (BC 95% CI: 0.474 - 0.515).
Figure 2 is the Cost Effectiveness Plane (CEP) which shows that early treatment is both less costly and more effective and therefore dominates receiving delayed treatment.
Discussion
Principle findings from the Study
HSV encephalitis can have devastating long-term consequences for patients, with many experiencing neurological impairments (15) and on-going health issues such as memory problems epilepsy, and behavioural changes (24). Several studies have reported on the association between time to treatment with aciclovir and the clinical outcomes of patients with HSV encephalitis (12, 15). These clinical outcomes are often measured using Glasgow Outcome Scale (GOS) or modified Rankin Scale (mRS), and a number of studies found a significant association between delayed treatment with aciclovir and poor health outcomes, with the majority of studies classifying delay as greater than either 1 or 2 days (9–14,25). Although these studies have differed in the settings, populations and some of the outcome measures they used, they all showed that there is an increased chance of poor outcome with a longer time to treatment. Findings of these studies are in alignment with our analysis which found that 71% of patients in the delayed treatment groups reported a ‘poor’ mRS score at discharge compared to only 14% in the early treatment group (significant at p<0.0002). Similarly, for the GOS score at discharge, good recovery was reported of only 7% of the patients whose treatment was delayed compared to 29% of patients being treated early. However, these outcome measurement tools are crude and not nuanced and detailed enough to demonstrate more subtle problems that may impact on activities of daily living and employment e.g., lack of concentration, poor memory and fatigue.
No previous study has looked at the effect of treatment delays on healthcare resource, costs and HRQoL, as well as estimating cost effectiveness of early treatment. After adjusting for confounding, the mean cost at 12 months follow-up was £34,886 (BC 95% CI: £30,715 - £39,423) and £76,071 (BC 95% CI: £58,037 - £105,743) for patients receiving early and delayed treatment, respectively. It is important to note that the length of hospital stay, the main driver of healthcare costs incurred by the patients, was somewhat significantly longer for patients in the delayed treatment group (94 [IQR: 29 - 157] days) compared to the early treatment group (31 [IQR: 22 - 58] days), (p<0.05). The resulting difference in the overall mean healthcare costs between the two treatment groups was - £39,960 (BC 95% CI: -£40,891 to -£38,815). QALYs were higher in the early treatment group compared to the delayed treatment group, 0.613 (BC 95% CI: 0.599 - 0.630) and 0.488 (BC 95% CI: 0.464 - 0.522) respectively. This resulted in a cost per QALY of -£334,068 (-£339,523 to -£329,856). However, statistical significance should be interpreted with caution due to the lack of power in the analysis associated with the small sample size as well as the potential selection bias which occurs when the differences in demographic and clinical characteristics prior to treatment, have an independent effect in the patient outcomes (26). These estimated healthcare costs are likely to be an underestimation of the true costs of treating HSV encephalitis as previous studies have shown patients often suffer from long term sequelae many years after initial presentation (27,28). Therefore, it is probable that additional costs to the NHS in terms of GP attendances, further diagnostic investigations and costs associated with ongoing care may have been underreported in the study. In addition, an earlier study by our group found that the sequelae after HSV encephalitis had significant impacts on activities of daily living with patients likely to incur costs through productivity losses due to difficulties returning to work either in their prior or in a changed role; there are additional costs for family/friends who become carers (24). However, given that the perspective of our analysis was that of the payer (NHS), these wider social costs were not included in the analysis.
We also projected the potential annual cost to the NHS of treating HSV encephalitis patients in the UK by assuming that the proportions of patients in each time to treatment group are representative of management of HSV encephalitis across the UK. Scaling up costs to a national level equated to average healthcare costs of £11.7 million annually, based on an annual incidence rate of 0.4 per 100,000 (21)
The differences in average healthcare costs between those treated more promptly and those delayed highlights the potential for savings for the NHS through reducing treatment delays. Significant improvements have already taken place in the UK regarding management of HSV encephalitis over the past 20 years. Our group led the development and publication of guidelines for the management of suspected viral encephalitis (21) facilitating improvements in recognition and diagnosis of the condition to reduce delays in treatment (29). Without the availability of time-series data following the same hospitals over time, it is difficult to measure the precise extent of any improvements. However, the earliest UK study (6) reported a median time to treatment of 48 hours, with 56% of patients experiencing delays of 48 hours or more, compared to 29% in this recent study. This is still too high. The national guidelines suggest all patients should be on treatment withing 6 hours of hospital admission. National level costs of a case of HSV encephalitis pre guideline publication can be estimated by assuming the proportions of patients treated early versus delayed in the earliest study (6) are representative of UK management during that time (44% treated early, 56% delayed) and applying the average costs for each treatment group estimated from this current study, adjusted for inflation using the consumer price index with 2015 as the base year (30). Based on average costs estimated in our study this would have equated to past (pre-guidelines) annual healthcare costs of £11.7 million in the UK. If the reduction in the proportion of patients experiencing delays through improvements in management over time are reflective of the UK population as a whole, this would equate to an annual saving of £720,000.
Symptoms of HSV encephalitis, particularly in the early stages of disease, have similarities with many other viral infections. However, the disease can rapidly develop over days, making prompt treatment with aciclovir crucial. There are a number of reasons that could result in treatment delays, which were not explored in this current study. Previous studies have shown that delays in undertaking CT scans can result in delays in the initiation of aciclovir (6,11), as can a failure to recognise HSV encephalitis as an initial diagnosis, due to the non-specific nature of the illness (31). This was also highlighted in a study conducted with HSV encephalitis survivors and their families, who described the difficulties in navigating health care systems during their illness trajectories, often having to develop their own care pathways in order to obtain recognition for their symptoms/ illness (3). That study emphasised the importance of involving and listening to patients and their significant others concerning symptom recognition and changes in usual behaviour to assist with the diagnosis and reduce treatment delays. In our current study, significant differences in age between the two groups were observed (p=0.031), with those with delayed treatment being older on average. Differences in age may influence whether a patient is treated/diagnosed quicker for a number of reasons. For example, it may be more difficult to recognise symptoms of encephalitis in older patients, such as confusion or dizziness, and often their symptoms are attributed to other conditions common in the elderly such as delirium or stroke. Previous literature has also shown older age to be associated with poor outcomes (12,25). The observation that those delayed were older on average highlights that further improvements are needed on the management of suspected encephalitis in elderly patients, with those already at an increased chance of poor outcome being further disadvantaged through treatment delays.
Limitations of the analysis
There are a number of limitations to this study. Firstly, as a result of the rarity of HSV encephalitis and missing data on some patient’s time of admission due to patients being transferred from other hospitals, this was a statistically small sample size. Small sample size could potentially under power the analysis and result in inadequate treatment effect estimates (32). Secondly, as is usually the case with non-randomised studies, selection bias, which was not adequately identified and adjusted for in this analysis meant that the unadjusted comparison of mean cost estimates are biased and should be interpreted with extreme caution as well as the extrapolations. Attempting to correct for selection bias with a matching (26) technique would lead to an even smaller sample size and the potential loss of informative data points. Thirdly, relying on case report forms to record patients resource use to estimate healthcare costs may not capture all the costs incurred. This is particularly true for follow-up data where primary care and community care contact outside the study hospitals will not have been captured. In addition, as is common during studies with follow-up periods, some patients had missing data in the 3- and 12-month CRFs, further reducing the sample size. There were a number of reasons for this missing data, including withdrawal due to the burden of follow-up, particularly when experiencing significant sequalae, and returning to relatives’ homes instead of their own address due to need for carers. Follow-up costs were only captured for the first 12 months post discharge and did not include primary and community care costs, which are likely to be substantial for those HSV encephalitis patients, particularly with recent studies showing the encephalitis patients often suffer from ongoing sequelae many years after initial presentation (24, 27, 28). Future studies should examine differences in time to treatment delays on longer term follow-up costs.
Conclusions and implications
HSV encephalitis can have devastating consequences for patients. This study has shown that those treated more promptly had significantly lower inpatient stay costs than those with delayed aciclovir treatment. This was also seen during the first-year after discharge. Alongside previous literature showing that delays in treatment results in poorer patient outcomes, the results of this study show that there are also significant costs to the NHS as a consequence of these delays. Work has been undertaken in the UK over recent years to improve the management of suspected encephalitis and encourage prompt treatment with aciclovir if the condition is suspected, which has led to reductions in average treatment times. Based on the findings of this study, reducing the number of patients experiencing delays and thus improving the management of HSV encephalitis is likely to have resulted in savings to the NHS. In addition, this study indicates the potential savings that could be made in the future with further improvements in the management of HSV encephalitis in the UK. Results of this analysis validate the importance of the encephalitis management guidelines that have boosted efforts towards timely diagnosis and instigation of treatment.
Data Availability
All data produced in the present study are available upon reasonable request to the authors
Acknowledgements
The authors would like to thank all the participants who gave their time to help with this study, and the research nurses and principal investigators at the hospital sites for their help with recruitment. We would also like to thank Rebecca Spencer, Mike Scully, Kieran Crabtree, Katy Smith and Chris Jecks for ensuring data collection was entered from the sites.
Principal Investigators
Gavin Barlow1, Nicholas J Beeching2, Thomas Blanchard3, Richard Body4, Gavin Boyd5, Lucia Cebria-Prejan6, David Chadwick7, Richard Cooke8 Pamela Crawford9, Brendan Davies10, Nicholas W S Davies11, Sam Douthwaite12, Hedley Emsley13, Simon Goldenberg12, Clive Graham14, Steve Green15, Clive Hawkins10, Dianne Irish16, Kate Jeffrey17, Matt Jones18, Liza Keating19, Jeff Keep20, Susan Larkin8, Maria Leita17, Derek Macallan21, Jane Minton22, Kavya Mohandas23, Ed Moran24, David Muir25, Monicka Pasztor26, Matthew Reed27, Tom Solomon28, Philip Stanley29, Julian Sutton30, Peter Thomas31, Guy Thwaites12, John Weir19, Mark Zuckerman20.
- Hull and East Yorkshire Hospitals NHS Trust, Hull.
- Liverpool University Hospitals NHS Foundation Trust Liverpool
- North Manchester General Hospital, Manchester
- Manchester Royal Infirmary, Manchester
- Calderdale and Huddersfield NHS Foundation Trust, Huddersfield
- Mid Yorkshire Hospitals NHS Trust, Pindersfield
- South Tees Hospitals NHS Foundation Trust, Middlesborough
- Aintree University Hospital NHS Foundation Trust, Liverpool
- York Teaching Hospital NHS Foundation Trust, York
- University Hospitals North Midlands, Stoke on Trent
- Chelsea and Westminster Hospital NHS Foundation Trust, London
- Guy’s and St Thomas’ NHS Foundation Trust, London
- Lancashire Teaching Hospitals NHS Foundation Trust, Royal Preston Hospital, Preston
- North Cumbria University Hospitals NHS Trust, Carlisle
- Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield
- Royal Free London NHS Foundation Trust, London
- Oxford University Hospitals NHS Foundation Trust, John Radcliffe Hospital, Oxford
- Salford Royal NHS Foundation trust, Salford
- Royal Berkshire NHS Foundation Trust, Reading
- Kings College Hospital NHS Foundation Trust, London
- St George’s University Hospitals NHS Foundation Trust, London
- Leeds Teaching Hospitals NHS Trust, St James’ University Hospital, Leeds
- Wirral University Teaching Hospital NHS Foundation Trust - Arrowepark Hospital, Upton
- University Hospitals Birmingham NHS Foundation Trust, Birmingham Heartlands Hospital, Birmingham
- Imperial College Healthcare NHS Trust, St Mary’s Hospital, London
- University Hospitals of Morecambe Bay NHS Foundation Trust, Lancaster
- NHS Lothian, Royal Infirmary Edinburgh, Edinburgh
- The Walton Centre NHS Foundation Trust, Liverpool
- Bradford Teaching Hospitals NHS Foundation Trust, Bradford
- University Hospital Southampton NHS Foundation Trust, Southampton
- Milton Keynes University Hospital NHS Foundation Trust, Milton Keynes
ENCEPH-UK contributors
Gavin Barlow, Nicholas J Beeching, Thomas Blanchard, Richard Body, Gavin Boyd, Lucia Cebria-Prejan, David Chadwick, Richard Cooke, Pamela Crawford, Brendan Davies, Nick Davies, Sam Douthwaite, Hedley Emsley, Simon Goldenberg, Clive Graham, Steve Green, Clive Hawkins, Dianne Irish, Kate Jeffrey, Matt Jones, Liza Keating, Jeff Keep, Susan Larkin, Maria Leita, Derek Macallan, Jane Minton, Kavya Mohandas, Ed Moran, David Muir, Monicka Pasztor, Matthew Reed, Tom Solomon, Philip Stanley, Julian Sutton, Peter Thomas, Guy Thwaites, John Weir, Mark Zuckerman. Ruth Backman, Gus Baker, Rachel Breen, David Brown, Chris Cheyne, Martin Eccles, Robbie Foy, Julia Granerod, Julia Griem, Alison Gummery, Lara Harris, Helen Hickey, Helen Hill, Ann Jacoby, Ciara Kierans, Michael Kopelman, Rachel Kneen, Gill Lancaster, Michael Levin, Rebecca McDonald, Esse Menson, Natalie Martin, Andrew Pennington, Andrew Pollard, Julie Riley, Manish Sadarangani, Maria Thornton.
Footnotes
Funding Statement
This work represents independent research funded by the National Institute for Health Research (NIHR) under its Programme Grants for Applied Research scheme (RP-PG-0108-10048).
Competing interests statement
None declared.
Patient consent
The study protocols were approved by participating sites and the National Research Ethics Service (now part of the Health Research Authority) East Midlands Nottingham 1 committee (reference 11/EM/0442). Written consent for entry into the study was obtained from patients or an accompanying relative. Standardised case record forms for clinical, laboratory and radiological data were recorded on a secure online database (OpenclinicaTM).
Data sharing statement
No additional data available.