An Analysis of 5-Level Version of EQ-5D Adjusting for Treatment Switching: The Case of Patients With Epidermal Growth Factor Receptor T790M-Positive Nonsmall Cell Lung Cancer Treated With Osimertinib.

OBJECTIVES: Treatment switching from control to treatment after disease progression is common in oncology trials. Analyses of survival data typically adjust for this bias, but such adjustments are rarely performed in analyses of patient-reported outcomes. This analysis aimed to examine the impact of adjusting for treatment switching on estimated treatment effects on 5-level version of EQ-5D (EQ-5D-5L) utilities and quality-adjusted life-years (QALYs). The AURA3 trial (NCT02151981) was a randomized controlled trial comparing osimertinib with platinum-based doublet chemotherapy (standard care) in patients with locally advanced or metastatic epidermal growth factor receptor mutant- and T790M-positive nonsmall cell lung cancer whose disease has progressed with previous epidermal growth factor receptor tyrosine kinase inhibitor therapy. METHODS: Descriptive analyses were used to compare treatment arms. The primary analysis used a 2-stage least squares instrumental variable regression to estimate treatment effect adjusting for treatment crossover. Time to deterioration, defined from baseline to minimally important deterioration in EQ-5D-5L utility, was assessed using a rank preserving structural failure time model. RESULTS: Intention-to-treat analysis of imputed data showed incremental QALYs for osimertinib of 0.23 at 60 weeks. Accounting for treatment switching increased this to 0.52 in the primary analysis and to 0.63 QALYs in sensitivity analysis at 150 weeks. Time to deterioration analysis showed longer health-related quality of life maintenance with osimertinib, of 12.76 weeks, although this was at the borderline of statistical significance (acceleration factor, ψ = -0.275; 95% confidence interval -0.50 to 0.00). CONCLUSIONS: This analysis demonstrates methods to adjust for treatment switching in the analysis of EQ-5D-5L from clinical trials. Failure to account for crossover substantially underestimated the QALY gain for osimertinib.

The cardiac output optimisation following liver transplant (COLT) trial: a feasibility randomised controlled trial.

BACKGROUND: Perioperative goal directed fluid therapy (GDFT) has been shown to reduce postoperative complications following major surgery; this intervention has not been formally evaluated in the setting of liver transplantation. METHODS: We conducted a prospective trial of GDFT following liver transplantation randomising patients with liver cirrhosis to either 12 h of GDFT using non-invasive cardiac output monitoring or standard care (SC). The primary outcome was feasibility. Secondary outcomes included survival, postoperative complications (Clavien-Dindo), quality of life (by EQ-5D-5L) and resource use. Trial specific follow up occurred at 90 and 180 days after surgery. RESULTS: The study was feasible. Of 224 eligible patients, 122 were approached, 114 consented to participate and 60 were enrolled into the trial. The mean (SD) volume of IV crystalloid administered to the GDFT group during the 12-h study period was 3968 (2073) ml for the GDFT group and 2510 (1026) ml for the SC group. As regards secondary outcomes there was no difference in survival or overall complication rates. There was no significant difference in quality of life scores and resource use between the groups. CONCLUSION: A randomised study of GDFT following liver transplantation is feasible. A post-trial stakeholder meeting supported proceeding with a full multi-centre trial.

Can Social Care Needs and Well-Being Be Explained by the EQ-5D? Analysis of the Health Survey for England.

BACKGROUND: The recent shift to an integrated approach to health and social care aims to provide cohesive support to those who are in need of care, but raises a challenge for resource allocation decision making, in particular for comparison of diverse benefits from different types of care across the two sectors. OBJECTIVE: To investigate the relationship of social care needs and well-being with a generic health status measure using multivariate regression. METHODS: We empirically compared responses to health and well-being measures and social care needs from a cross-sectional data set of the general population (the Health Survey for England). Multivariate regression analyses were conducted to examine whether social care needs measured by the Barthel index can be explained by health status as captured by the EuroQol five-dimensional questionnaire (EQ-5D) and two well-being measures-the Warwick Edinburgh Mental Wellbeing Scale (WEMWBS) and the General Health Questionnaire (GHQ-12). RESULTS: Our study found that poor overall scores for EuroQol visual analogue scale, EQ-5D index, GHQ-12, and WEMWBS indicated a need for social care. Investigation of the dimensions found that the EQ-5D dimensions self-care and pain/discomfort were statistically significantly associated with the need for social care. Two dimensions of the WEMWBS ("been feeling useful" and "had energy to spare") were statistically significantly associated with the Barthel index, but none of the GHQ-12 dimensions were. CONCLUSIONS: The results show that the need for social care, which is dependent on the ability to perform personal day-to-day activities, is more closely related to the EQ-5D dimensions than the well-being measures WEMWBS and GHQ-12.

COMPARISON OF METHODS FOR ESTIMATING CONTEMPORARY COSTS: AN APPLICATION TO LIVER TRANSPLANTATION IN THE UNITED KINGDOM.

OBJECTIVES: Our study addresses the important issue of estimating treatment costs from historical data. It is a problem frequently faced by health technology assessment analysts. We compared four approaches used to estimate current costs when good quality contemporary data are not available using liver transplantation as an example. METHODS: First, the total cost estimates extracted for patients from a cohort study, conducted in the 1990s, were inflated using a published inflation multiplier. Second, resource use estimates from the cohort study were extracted for hepatitis C patients and updated using current unit costs. Third, expert elicitation was carried out to identify changes in clinical practice over time and quantify current resource use. Fourth, routine data on resource use were obtained from National Health Service Blood and Transplant (NHSBT). RESULTS: The first two methods did not account for changes in clinical practice. Also the first was not specific to hepatitis patients. The use of experts confirmed significant changes in clinical practice. However, the quantification of resource use using experts is challenging as clinical specialists may not have a complete overview of clinical pathway. The NHSBT data are the most accurate reflection of transplantation and posttransplantation phase; however, data were not available for the whole pathway of care. The best estimate of total cost, combining NHSBT data and expert elicitation, is £121,211. DISCUSSION: Observational data from routine care are potentially the most reliable reflection of current resource use. Efforts should be made to make such data readily available and accessible to researchers. Expert elicitation provided reasonable estimates.

Exploring what lies behind public preferences for avoiding health losses caused by lapses in healthcare safety and patient lifestyle choices.

BACKGROUND: Although many studies have identified public preferences for prioritising health care interventions based on characteristics of recipient or care, very few of them have examined the reasons for the stated preferences. We conducted an on-line person trade-off (PTO) study (N=1030) to investigate whether the public attach a premium to the avoidance of ill health associated with alternative types of responsibilities: lapses in healthcare safety, those caused by individual action or lifestyle choice; or genetic conditions. We found that the public gave higher priority to prevention of harm in a hospital setting such as preventing hospital associated infections than genetic disorder but drug administration errors were valued similar to genetic disorders. Prevention of staff injuries, lifestyle diseases and sports injuries, were given lower priority. In this paper we aim to understand the reasoning behind the responses by analysing comments provided by respondents to the PTO questions. METHOD: A majority of the respondents who participated in the survey provided brief comments explaining preferences in free text responses following PTO questions. This qualitative data was transformed into explicit codes conveying similar meanings. An overall coding framework was developed and a reliability test was carried out. Recurrent patterns were identified in each preference group. Comments which challenged the assumptions of hypothetical scenarios were also investigated. RESULTS: NHS causation of illness and a duty of care were the most cited reasons to prioritise lapses in healthcare safety. Personal responsibility dominated responses for lifestyle related contexts, and many respondents mentioned that health loss was the result of the individual's choice to engage in risky behaviour. A small proportion of responses questioned the assumptions underlying the PTO questions. However excluding these from the main analysis did not affect the conclusions. CONCLUSION: Although some responses indicated misunderstanding or rejection of assumptions we put forward, the results were still robust. The reasons put forward for responses differed between comparisons but responsibility was the most frequently cited. Most preference elicitation studies only focus on eliciting numerical valuations but allowing for qualitative data can augment understanding of preferences as well as verifying results.

Does responsibility affect the public's valuation of health care interventions? A relative valuation approach to health care safety.

OBJECTIVE: Health services often spend more on safety interventions than seems cost-effective. This study investigates whether the public value safety-related health care improvements more highly than the same improvements in contexts where the health care system is not responsible. METHOD: An online survey was conducted to elicit the relative importance placed on preventing harms caused by 1) health care (hospital-acquired infections, drug administration errors, injuries to health care staff), 2) individuals (personal lifestyle choices, sports-related injuries), and 3) nature (genetic disorders). Direct valuations were obtained from members of the public by using a person trade-off or "matching" method. Participants were asked to choose between two preventative interventions of equal cost and equal health benefit per person for the same number of people, but differing in causation. If participants indicated a preference, their strength of preference was measured by using person trade-off. RESULTS: Responses were obtained from 1030 people, reflecting the sociodemographic mix of the UK population. Participants valued interventions preventing hospital-acquired infections (1.31) more highly than genetic disorders (1.0), although drug errors were valued similarly to genetic disorders (1.07), and interventions to prevent injury to health care staff were given less weight than genetic disorders (0.71). Less weight was also given to interventions related to lifestyle (0.65) and sports injuries (0.41). CONCLUSION: Our results suggest that people do not attach a simple fixed premium to "safety-related" interventions but that preferences depend more subtly on context. The use of the results of such public preference surveys to directly inform policy would therefore be premature.

Effect of post-operative goal-directed fluid therapy (GDFT) on organ function after orthotopic liver transplantation: Secondary outcome analysis of the COLT randomised control trial.

BACKGROUND: Goal-directed fluid therapy (GDFT) has been shown to reduce the complications following a variety of major surgical procedures, possibly mediated by improved organ perfusion and function. We have shown that it is feasible to randomise patients to GDFT or standard fluid management following liver transplant in the cardiac-output optimisation following liver transplantation (COLT) trial. The current study compares end organ function in patients from the COLT trial who received GDFT in comparison to those receiving standard care (SC) following liver transplant. METHODS: Adult patients with liver cirrhosis undergoing liver transplantation were randomised to GDFT or SC for the first 12 h following surgery as detailed in a published trial protocol. GDFT protocol was based on stroke volume (SV) optimisation using 250 ml crystalloid boluses. Total fluid administration and time to extubation were recorded. Hourly SV and cardiac output (CO) readings were recorded from the non-invasive cardiac output monitoring (NICOM) device in both groups. Pulmonary function was assessed by arterial blood gas (ABG) and ventilatory parameters. Lung injury was assessed using PaO2:FiO2 ratios and calculated pulmonary compliance. The KDIGO score was used for determining acute kidney injury. Renal and liver graft function were assessed during the post-operative period and at 3 months and 1-year. RESULTS: 60 patients were randomised to GDFT (n = 30) or SC (n = 30). All patients completed the 12 h intervention period. GDFT group received a significantly higher total volume of fluid during the 12 h trial intervention period (GDFT 5317 (2335) vs. SC 3807 (1345) ml, p = 0.003); in particular crystalloids (GDFT 3968 (2073) vs. SC 2510 (1027) ml, p = 0.002). There was no evidence of significant difference between the groups in SV or CO during the assessment periods. Time to extubation, PaO2: FIO2 ratios, pulmonary compliance, ventilatory or blood gas measurements were similar in both groups. There was a significant rise in serum creatinine from baseline (77 µmol/L) compared to first (87 µmol/L, p = 0.039) and second (107 µmol/L, p = 0.001) post-operative days. There was no difference between GDFT and SC in the highest KDIGO scores for the first 7 days post-LT. At 1-year follow-up, there was no difference in need for renal replacement therapy or graft function. CONCLUSIONS: In this randomised trial of fluid therapy post liver transplant, GDFT was associated with an increased volume of crystalloids administered but did not alter early post-operative pulmonary or renal function when compared with standard care.

Elicitation of Health-Related Utility in Perianal Fistula in Crohn's Disease.

BACKGROUND AND OBJECTIVE: Perianal fistulae are a common complication of Crohn's disease (CD) and pose a substantial burden on quality of life. Data capturing health-related utility associated with perianal fistulae in CD are scarce. The current study aims to value health states related to different stages of the disease to quantitatively evaluate the impact of complex perianal fistulae on CD patients' quality of life. METHODS: Eight health state descriptions associated with complex perianal fistulae in CD were developed following qualitative research with patients and validation by clinicians. Following pre-testing, a survey was administered online in two samples of UK respondents: the general population and patients with CD. A choice-based valuation technique, the time trade-off (TTO), was used for direct utility measurement. CD patients also valued their current health state using the TTO. Exclusion criteria for respondents displaying logical inconsistencies were applied. RESULTS: Usable responses were received from 835 respondents, reflective of the UK population in age and sex, in the general population survey and 162 CD patients in the patient survey. Non-remission states were valued much lower than the remission state by both samples, ranging from 0.20 for proctectomy with a negative outcome to 0.66 for chronic symptomatic fistulae with mild symptoms. Patients currently experiencing fistulae reported lower values for current health than those without fistulae. CONCLUSION: Low utility values were assigned to the non-remission health states for perianal fistulae in CD by the general public and patients with CD. This demonstrates the high humanistic burden of inadequately managed perianal fistula in CD.

Ward-based Goal-Directed Fluid Therapy (GDFT) in Acute Pancreatitis (GAP) trial: study protocol for a feasibility randomised controlled trial.

INTRODUCTION: Acute pancreatitis is an inflammatory disease of the pancreas with high risk of developing multiorgan failure and death. There are no effective pharmacological interventions used in current clinical practice. Maintaining fluid and electrolyte balance is the mainstay of supportive management. Goal-directed fluid therapy (GDFT) has been shown to decrease morbidity and mortality in surgical conditions with systemic inflammatory response. There is currently no randomised controlled trial (RCT) investigating the role of GDFT based on cardiac output parameters in patients with acute pancreatitis in the ward setting. A feasibility trial was designed to determine patient and clinician support for recruitment into an RCT of ward-based GDFT in acute pancreatitis, adherence to a GDFT protocol, safety, participant withdrawal, and to determine appropriate endpoints for a subsequent larger trial to evaluate efficacy. METHODS AND ANALYSIS: The GDFT in Acute Pancreatitis trial is a prospective two-centre feasibility RCT. Eligible adults admitted with new onset of acute pancreatitis will be enrolled and randomised into ward-based GDFT (n=25) or standard fluid therapy (n=25) within 6 hours from the diagnosis and continuing for the following 48 hours. Cardiac output parameters will be monitored with a non-invasive device (Cheetah NICOM; Cheetah Medical). The intervention group will consist of a protocolised GDFT approach consisting of stroke volume optimisation with crystalloid fluid boluses, while the control group will receive standard care fluid therapy as advised by the clinical team. The primary endpoint is feasibility. Secondary endpoints will include safety of the intervention, complications, mortality, admission to intensive care unit, cost and quality of life. ETHICS AND DISSEMINATION: Ethics approval was granted by the London Central Research Ethics Committee (17/LO/1235, project ID: 221872). The results of this trial will be presented to international conference with interest in general surgery and acute care and published in a peer-reviewed journal. TRIAL REGISTRATION NUMBER: ISRCTN36077283.

Amaze: a double-blind, multicentre randomised controlled trial to investigate the clinical effectiveness and cost-effectiveness of adding an ablation device-based maze procedure as an adjunct to routine cardiac surgery for patients with pre-existing atrial fibrillation.

BACKGROUND: Atrial fibrillation (AF) can be treated using a maze procedure during planned cardiac surgery, but the effect on clinical patient outcomes, and the cost-effectiveness compared with surgery alone, are uncertain. OBJECTIVES: To determine whether or not the maze procedure is safe, improves clinical and patient outcomes and is cost-effective for the NHS in patients with AF. DESIGN: Multicentre, Phase III, pragmatic, double-blind, parallel-arm randomised controlled trial. Patients were randomised on a 1 : 1 basis using random permuted blocks, stratified for surgeon and planned procedure. SETTING: Eleven acute NHS specialist cardiac surgical centres. PARTICIPANTS: Patients aged ≥ 18 years, scheduled for elective or in-house urgent cardiac surgery, with a documented history (> 3 months) of AF. INTERVENTIONS: Routine cardiac surgery with or without an adjunct maze procedure administered by an AF ablation device. MAIN OUTCOME MEASURES: The primary outcomes were return to sinus rhythm (SR) at 12 months and quality-adjusted life-years (QALYs) over 2 years after randomisation. Secondary outcomes included return to SR at 2 years, overall and stroke-free survival, drug use, quality of life (QoL), cost-effectiveness and safety. RESULTS: Between 25 February 2009 and 6 March 2014, 352 patients were randomised to the control (n = 176) or experimental (n = 176) arms. The odds ratio (OR) for return to SR at 12 months was 2.06 [95% confidence interval (CI) 1.20 to 3.54; p = 0.0091]. The mean difference (95% CI) in QALYs at 2 years between the two trial arms (maze/control) was -0.025 (95% CI 0.129 to 0.078; p = 0.6319). The OR for SR at 2 years was 3.24 (95% CI 1.76 to 5.96). The number of patients requiring anticoagulant drug use was significantly lower in the maze arm from 6 months after the procedure. There were no significant differences between the two arms in operative or overall survival, stroke-free survival, need for cardioversion or permanent pacemaker implants, New York Heart Association Functional Classification (for heart failure), EuroQol-5 Dimensions, three-level version score and Short Form questionnaire-36 items score at any time point. Sixty per cent of patients in each trial arm had a serious adverse event (p = 1.000); most events were mild, but 71 patients (42.5%) in the maze arm and 84 patients (45.5%) in the control arm had moderately severe events; 31 patients (18.6%) in the maze arm and 38 patients (20.5%) in the control arm had severe events. The mean additional cost of the maze procedure was £3533 (95% CI £1321 to £5746); the mean difference in QALYs was -0.022 (95% CI -0.1231 to 0.0791). The maze procedure was not cost-effective at £30,000 per QALY over 2 years in any analysis. In a small substudy, the active left atrial ejection fraction was smaller than that of the control patients (mean difference of -8.03, 95% CI -12.43 to -3.62), but within the predefined clinically equivalent range. LIMITATIONS: Low recruitment, early release of trial summaries and intermittent resource-use collection may have introduced bias and imprecise estimates. CONCLUSIONS: Ablation can be practised safely in routine NHS cardiac surgical settings and increases return to SR rates, but not survival or QoL up to 2 years after surgery. Lower anticoagulant drug use and recovery of left atrial function support anticoagulant drug withdrawal provided that good atrial function is confirmed. FURTHER WORK: Continued follow-up and long-term clinical effectiveness and cost-effectiveness analysis. Comparison of ablation methods. TRIAL REGISTRATION: Current Controlled Trials ISRCTN82731440. FUNDING: This project was funded by the NIHR Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 22, No. 19. See the NIHR Journals Library website for further project information.

Cardiac output Optimisation following Liver Transplant (COLT) trial: study protocol for a feasibility randomised controlled trial.

BACKGROUND: Patients with liver cirrhosis undergoing liver transplantation have a hyperdynamic circulation which persists into the early postoperative period making accurate assessment of fluid requirements challenging. Goal-directed fluid therapy (GDFT) has been shown to reduce morbidity and mortality in a number of surgery settings. The impact of GDFT in patients undergoing liver transplantation is unknown. A feasibility trial was designed to determine patient and clinician support for recruitment into a randomised controlled trial of GDFT following liver transplantation, adherence to a GDFT protocol, participant withdrawal, and to determine appropriate endpoints for a subsequent larger trial to evaluate the efficacy of GDFT in patients undergoing liver transplantation. METHODS: The Cardiac output Optimisation following Liver Transplant (COLT) trial is designed as a prospective, single-centre, randomised controlled study to assess the feasibility and safety of GDFT in liver transplantation for patients with cirrhosis. Consenting adults (aged between 18 and 80 years) with biopsy-proven liver cirrhosis who have been selected to undergo a first liver transplantation will be included in the trial and randomised into GDFT or standard care starting immediately after surgery and continuing for the first 12 h thereafter. Both groups will have cardiac output and stroke volume monitored using the FloTrac (EV1000) device. The intervention will consist of a protocolised GDFT approach to patient management, using stroke volume optimisation. The control group will receive standard care, without stroke volume and cardiac output measurement. After 12 h the patient's fluid management will revert to standard of care. The primary endpoint of this study is feasibility. Secondary endpoints will include a safety assessment of the intervention, graft and patient survival, liver function, postoperative complications graded by Clavien-Dindo criteria, length of intensive care and hospital stay and quality of life across the intervention and control groups. DISCUSSION: There is a growing body of evidence that the use of perioperative GDFT in surgical patients can improve outcomes; however, signals of harm have also been detected. Patients with liver cirrhosis undergoing liver transplantation have markedly different cardiovascular physiology than general surgical patients. If GDFT is proven to be feasible and safe in this patient group, then a multicentre trial to demonstrate efficacy and cost-effectiveness will be required. TRIAL REGISTRATION: International Standard Randomised Controlled Trial Registry, ID: ISRCTN10329248. Registered on 4 April 2016.

Mini-Stern Trial: A randomized trial comparing mini-sternotomy to full median sternotomy for aortic valve replacement.

OBJECTIVE: Aortic valve replacement (AVR) can be performed either through full median sternotomy (FS) or upper mini-sternotomy (MS). The Mini-Stern trial aimed to establish whether MS leads to quicker postoperative recovery and shorter hospital stay after first-time isolated AVR. METHODS: This pragmatic, open-label, parallel randomized controlled trial (RCT) compared MS with FS for first-time isolated AVR in 2 United Kingdom National Health Service hospitals. Primary endpoints were duration of postoperative hospital stay and the time to fitness for discharge from hospital after AVR, analyzed in the intent-to-treat population. RESULTS: In this RCT, 222 patients were recruited and randomized (n = 118 in the MS group; n = 104 in the FS group). Compared with the FS group, the MS group had a longer hospital length of stay (mean, 9.5 days vs 8.6 days) and took longer to achieve fitness for discharge home (mean, 8.5 days vs 7.5 days). Adjusting for valve type, sex, and surgeon, hazard ratios (HRs) from Cox models did not show a statistically significant effect of MS (relative to FS) on either hospital stay (HR, 0.874; 95% confidence interval [CI], 0.668-1.143; P = .3246) or time to fitness for discharge (HR, 0.907; 95% CI, 0.688-1.197; P value = .4914). During a mean follow-up of 760 days (745 days for the MS group and 777 days for the FS group), 12 patients (10%) in the MS group and 7 patients (7%) in the FS group died (HR, 1.871; 95% CI, 0.723-4.844; P = .1966). Average extra cost for MS was £1714 during the first 12 months after AVR. CONCLUSIONS: Compared with FS for AVR, MS did not result in shorter hospital stay, faster recovery, or improved survival and was not cost-effective. The MS approach is not superior to FS for performing AVR.

Amaze: a randomized controlled trial of adjunct surgery for atrial fibrillation.

OBJECTIVES: Atrial fibrillation (AF) reduces survival and quality of life (QoL). It can be treated at the time of major cardiac surgery using ablation procedures ranging from simple pulmonary vein isolation to a full maze procedure. The aim of this study is to evaluate the impact of adjunct AF surgery as currently performed on sinus rhythm (SR) restoration, survival, QoL and cost-effectiveness. METHODS: In a multicentre, Phase III, pragmatic, double-blinded, parallel-armed randomized controlled trial, 352 cardiac surgery patients with >3 months of documented AF were randomized to surgery with or without adjunct maze or similar AF ablation between 2009 and 2014. Primary outcomes were SR restoration at 1 year and quality-adjusted life years at 2 years. Secondary outcomes included SR at 2 years, overall and stroke-free survival, medication, QoL, cost-effectiveness and safety. RESULTS: More ablation patients were in SR at 1 year [odds ratio (OR) 2.06, 95% confidence interval (CI) 1.20-3.54; P = 0.009]. At 2 years, the OR increased to 3.24 (95% CI 1.76-5.96). Quality-adjusted life years were similar at 2 years (ablation - control -0.025, P = 0.6319). Significantly fewer ablation patients were anticoagulated from 6 months postoperatively. Stroke rates were 5.7% (ablation) and 9.1% (control) (P = 0.3083). There was no significant difference in stroke-free survival [hazard ratio (HR) = 0.99, 95% CI 0.64-1.53; P = 0.949] nor in serious adverse events, operative or overall survival, cardioversion, pacemaker implantation, New York Heart Association, EQ-5D-3L and SF-36. The mean additional ablation cost per patient was £3533 (95% CI £1321-£5746). Cost-effectiveness was not demonstrated at 2 years. CONCLUSIONS: Adjunct AF surgery is safe and increases SR restoration and costs but not survival or QoL up to 2 years. A continued follow-up will provide information on these outcomes in the longer term. Study registration: ISRCTN82731440 (project number 07/01/34).