BACKGROUND: Physical activity can support smoking cessation for smokers wanting to quit, but there have been no studies on supporting smokers wanting only to reduce. More broadly, the effect of motivational support for such smokers is unclear. OBJECTIVES: the objectives were to determine if motivational support to increase physical activity and reduce smoking for smokers not wanting to immediately quit helps reduce smoking and increase abstinence and physical activity, and to determine if this intervention is cost-effective. DESIGN: This was a multicentred, two-arm, parallel-group, randomised (1 : 1) controlled superiority trial with accompanying trial-based and model-based economic evaluations, and a process evaluation. SETTING AND PARTICIPANTS: Participants from health and other community settings in four English cities received either the intervention (n = 457) or usual support (n = 458). INTERVENTION: the intervention consisted of up to eight face-to-face or telephone behavioural support sessions to reduce smoking and increase physical activity. MAIN OUTCOME MEASURES: the main outcome measures were carbon monoxide-verified 6- and 12-month floating prolonged abstinence (primary outcome), self-reported number of cigarettes smoked per day, number of quit attempts and carbon monoxide-verified abstinence at 3 and 9 months. Furthermore, self-reported (3 and 9 months) and accelerometer-recorded (3 months) physical activity data were gathered. Process items, intervention costs and cost-effectiveness were also assessed. RESULTS: the average age of the sample was 49.8 years, and participants were predominantly from areas with socioeconomic deprivation and were moderately heavy smokers. The intervention was delivered with good fidelity. Few participants achieved carbon monoxide-verified 6-month prolonged abstinence [nine (2.0%) in the intervention group and four (0.9%) in the control group; adjusted odds ratio 2.30 (95% confidence interval 0.70 to 7.56)] or 12-month prolonged abstinence [six (1.3%) in the intervention group and one (0.2%) in the control group; adjusted odds ratio 6.33 (95% confidence interval 0.76 to 53.10)]. At 3 months, the intervention participants smoked fewer cigarettes than the control participants (21.1 vs. 26.8 per day). Intervention participants were more likely to reduce cigarettes by ≥ 50% by 3 months [18.9% vs. 10.5%; adjusted odds ratio 1.98 (95% confidence interval 1.35 to 2.90] and 9 months [14.4% vs. 10.0%; adjusted odds ratio 1.52 (95% confidence interval 1.01 to 2.29)], and reported more moderate-to-vigorous physical activity at 3 months [adjusted weekly mean difference of 81.61 minutes (95% confidence interval 28.75 to 134.47 minutes)], but not at 9 months. Increased physical activity did not mediate intervention effects on smoking. The intervention positively influenced most smoking and physical activity beliefs, with some intervention effects mediating changes in smoking and physical activity outcomes. The average intervention cost was estimated to be £239.18 per person, with an overall additional cost of £173.50 (95% confidence interval -£353.82 to £513.77) when considering intervention and health-care costs. The 1.1% absolute between-group difference in carbon monoxide-verified 6-month prolonged abstinence provided a small gain in lifetime quality-adjusted life-years (0.006), and a minimal saving in lifetime health-care costs (net saving £236). CONCLUSIONS: There was no evidence that behavioural support for smoking reduction and increased physical activity led to meaningful increases in prolonged abstinence among smokers with no immediate plans to quit smoking. The intervention is not cost-effective. LIMITATIONS: Prolonged abstinence rates were much lower than expected, meaning that the trial was underpowered to provide confidence that the intervention doubled prolonged abstinence. FUTURE WORK: Further research should explore the effects of the present intervention to support smokers who want to reduce prior to quitting, and/or extend the support available for prolonged reduction and abstinence. TRIAL REGISTRATION: This trial is registered as ISRCTN47776579. FUNDING: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 27, No. 4. See the NIHR Journals Library website for further project information.

OBJECTIVES: to develop a flexible droplet digital PCR (ddPCR) workflow to perform non-invasive prenatal diagnosis via relative mutation dosage (RMD) for maternal pathogenic variants with a range of inheritance patterns, and to compare the accuracy of multiple analytical approaches. METHODS: Cell free DNA (cfDNA) was tested from 124 archived maternal plasma samples: 88 cases for sickle cell disease and 36 for rare Mendelian conditions. Three analytical methods were compared: sequential probability ratio testing (SPRT), Bayesian and z-score analyses. RESULTS: the SPRT, Bayesian and z-score analyses performed similarly well with correct prediction rates of 96%, 97% and 98%, respectively. However, there were high rates of inconclusive results for each cohort, particularly for z-score analysis which was 31% overall. Two samples were incorrectly classified by all three analytical methods; a false negative result predicted for a fetus affected with sickle cell disease and a false positive result predicting the presence of an X-linked IDS variant in an unaffected fetus. CONCLUSIONS: ddPCR can be applied to RMD for diverse conditions and inheritance patterns, but all methods carry a small risk of erroneous results. Further evaluation is required both to reduce the rate of inconclusive results and explore discordant results in more detail.

. Background
. Mobility limitation in older age reduces quality of life, generates substantial health- and social-care costs, and increases mortality.
.
.
. Objective
. The REtirement in ACTion (REACT) trial aimed to establish whether or not a community-based active ageing intervention could prevent decline in physical functioning in older adults already at increased risk of mobility limitation.
.
.
. Design
. A multicentre, pragmatic, two-arm, parallel-group randomised controlled trial with parallel process and health economic evaluations.
.
.
. Setting
. Urban and semi-rural locations across three sites in England.
.
.
. Participants
. Physically frail or pre-frail older adults (aged ≥ 65 years; Short Physical Performance Battery score of 4–9). Recruitment was primarily via 35 primary care practices.
.
.
. Interventions
. Participants were randomly assigned to receive brief advice (three healthy ageing education sessions) or a 12-month, group-based, multimodal exercise and behavioural maintenance programme delivered in fitness and community centres. Randomisation was stratified by site and used a minimisation algorithm to balance age, sex and Short Physical Performance Battery score. Data collection and analyses were blinded.
.
.
. Main outcome measures
. The primary outcome was change in lower limb physical function (Short Physical Performance Battery score) at 24 months, analysed using an intention-to-treat analysis. The economic evaluation adopted the NHS and Personal Social Services perspective.
.
.
. Results
. Between June 2016 and October 2017, 777 participants (mean age 77.6 years, standard deviation 6.8 years; 66% female; mean Short Physical Performance Battery score 7.37, standard deviation 1.56) were randomised to the intervention arm (n = 410) or the control arm (n = 367). Data collection was completed in October 2019. Primary outcome data at 24 months were provided by 628 (80.8%) participants. At the 24-month follow-up, the Short Physical Performance Battery score was significantly greater in the intervention arm (mean 8.08, standard deviation 2.87) than in the control arm (mean 7.59, standard deviation 2.61), with an adjusted mean difference of 0.49 (95% confidence interval 0.06 to 0.92). The difference in lower limb function between intervention and control participants was clinically meaningful at both 12 and 24 months. Self-reported physical activity significantly increased in the intervention arm compared with the control arm, but this change was not observed in device-based physical activity data collected during the trial. One adverse event was related to the intervention. Attrition rates were low (19% at 24 months) and adherence was high. Engagement with the REACT intervention was associated with positive changes in exercise competence, relatedness and enjoyment and perceived physical, social and mental well-being benefits. The intervention plus usual care was cost-effective compared with care alone over the 2 years of REACT; the price year was 2019. In the base-case scenario, the intervention saved £103 per participant, with a quality-adjusted life-year gain of 0.04 (95% confidence interval 0.006 to 0.074) within the 2-year trial window. Lifetime horizon modelling estimated that further cost savings and quality-adjusted life-year gains were accrued up to 15 years post randomisation.
.
.
. Conclusion
. A relatively low-resource, 1-year multimodal exercise and behavioural maintenance intervention can help older adults to retain physical functioning over a 24-month period. The results indicate that the well-established trajectory of declining physical functioning in older age is modifiable.
.
.
. Limitations
. Participants were not blinded to study arm allocation. However, the primary outcome was independently assessed by blinded data collectors. The secondary outcome analyses were exploratory, with no adjustment for multiple testing, and should be interpreted accordingly.
.
.
. Future work
. Following refinements guided by the process evaluation findings, the REACT intervention is suitable for large-scale implementation. Further research will optimise implementation of REACT at scale.
.
.
. Trial registration
. This trial is registered as ISRCTN45627165.
.
.
. Funding
. This project was funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme and will be published in full in Public Health Research; Vol. 10, No. 14. See the NIHR Journals Library website for further project information.
.

BACKGROUND: Mobility limitations in older populations have a substantial impact on health outcomes, quality of life, and social care costs. The Retirement in Action (REACT) randomised controlled trial assessed a 12-month community-based group physical activity and behaviour maintenance intervention to help prevent decline in physical functioning in older adults at increased risk of mobility limitation. We aimed to do an economic evaluation of the REACT trial to investigate whether the intervention is cost-effective. METHODS: in this health economic evaluation, we did cost-effectiveness and cost-utility analyses of the REACT programme versus standard care on the basis of resource use, primary outcome, and health-related quality-of-life data measured in the REACT trial. We also developed a decision analytic Markov model that forecasts the mobility of recipients beyond the 24-month follow-up of the trial and translated this into future costs and potential benefit to health-related quality of life using the National Health Service and Personal Social Services perspective. Participants completed questionnaire booklets at baseline, and at 6, 12, and 24 months after randomisation, which included a resource use questionnaire and the EQ-5D-5L and 36-item short-form survey (SF-36) health-related quality-of-life instruments. The cost of delivering the intervention was estimated by identifying key resources, such as REACT session leader time, time of an individual to coordinate the programme, and venue hire. EQ-5D-5L and SF-36 responses were converted to preference-based utility values, which were used to estimate quality-adjusted life-years (QALYs) over the 24-month trial follow-up using the area-under-the-curve method. We used generalised linear models to examine the effect of the REACT programme on costs and QALYs and adjust for baseline covariates. Costs and QALYs beyond 12 months were discounted at 3·5% per year. This is a pre-planned analysis of the REACT trial; the trial itself is registered with ISRCTN (ISRCTN45627165). FINDINGS: the 12-month REACT programme was estimated to cost £622 per recipient to deliver. The most substantial cost components are the REACT session leader time (£309 per participant), venue hire (£109), and the REACT coordinator time (£80). The base-case analysis of the trial-based economic evaluation showed that reductions in health and social care usage due to the REACT programme could offset the REACT delivery costs (£3943 in the intervention group vs £4043 in the control group; difference: -£103 [95% CI -£695 to £489]) with a health benefit of 0·04 QALYs (0·009-0·071; 1·354 QALYs in the intervention group vs 1·314 QALYs in the control group) within the 24-month timeframe of the trial. INTERPRETATION: the REACT programme could be considered a cost-effective approach for improving the health-related quality of life of older adults at risk of mobility limitations. FUNDING: National Institute for Health Research Public Health Research Programme.

AbstractThe rapid evolution of novel, costly therapies for neuroendocrine neoplasia (NEN) warrants formal high-quality cost-effectiveness evaluation. Costs of individual investigations and therapies are high; and examples are presented. We aimed to review the last ten years of standalone health economic evaluations in NEN. Comparing to published standards, EMBASE, Cochrane library, Database of Abstracts of Reviews of Effects (DARE), NHS Economic Evaluation Database and the Health Technology Assessment (HTA) Database were searched for health economic evaluations (HEEs) in NEN published between 2010 and October 2019. of 12 economic evaluations, 11 considered exclusively pharmacological treatment (3 studies of SSAs, 7 studies of sunitinib, everolimus and/or 177Lu-DOTATATE and 1 study of telotristat ethyl) and 1 compared surgery with intraarterial therapy. 7 studies of pharmacological treatment had placebo or best supportive care as the only comparator. There remains a paucity of economic evaluations in NEN with the majority industry funded. Most HEEs reviewed did not meet published health economic criteria used to assess quality. Lack of cost data collected from patient populations remains a significant factor in HEEs where clinical expert opinion is still often substituted. Further research utilizing high-quality effectiveness data and rigorous applied health economic analysis is needed.

Abstract
Background
A systematic review of economic evaluations for lung cancer identified no economic models of the UK setting based on disease natural history. We first sought to develop a new model of natural history for population screening, then sought to explore the cost-effectiveness of multiple alternative potential programmes.

Methods
An individual patient model (ENaBL) was constructed in MS Excel® and calibrated against data from the US National Lung Screening Trial. Costs were taken from the UK Lung Cancer Screening Trial and took the perspective of the NHS and PSS. Simulants were current or former smokers aged between 55 and 80 years and so at a higher risk of lung cancer relative to the general population. Subgroups were defined by further restricting age and risk of lung cancer as predicted by patient self-questionnaire. Programme designs were single, triple, annual and biennial arrangements of LDCT screens, thereby examining number and interval length. Forty-eight distinct screening strategies were compared to the current practice of no screening. The primary outcome was incremental cost-effectiveness of strategies (additional cost per QALY gained).

Results
LDCT screening is predicted to bring forward the stage distribution at diagnosis and reduce lung cancer mortality, with decreases versus no screening ranging from 4.2 to 7.7% depending on screen frequency. Overall healthcare costs are predicted to increase; treatment cost savings from earlier detection are outweighed by the costs of over-diagnosis. Single-screen programmes for people 55–75 or 60–75 years with ≥ 3% predicted lung cancer risk may be cost-effective at the £30,000 per QALY threshold (respective ICERs of £28,784 and £28,169 per QALY gained). Annual and biennial screening programmes were not predicted to be cost-effective at any cost-effectiveness threshold.

Limitations
LDCT performance was unaffected by lung cancer type, stage or location and the impact of a national screening programme of smoking behaviour was not included.

Conclusion
Lung cancer screening may not be cost-effective at the threshold of £20,000 per QALY commonly used in the UK but may be cost-effective at the higher threshold of £30,000 per QALY.

BACKGROUND: Challenges of recruitment to randomized controlled trials (RCTs) and successful strategies to overcome them should be clearly reported to improve recruitment into future trials. REtirement in ACTion (REACT) is a United Kingdom-based multicenter RCT recruiting older adults at high risk of mobility disability to a 12-month group-based exercise and behavior maintenance program or to a minimal Healthy Aging control intervention. METHODS: the recruitment target was 768 adults, aged 65 years and older scoring 4-9 on the Short Physical Performance Battery (SPPB). Recruitment methods include the following: (a) invitations mailed by general practitioners (GPs); (b) invitations distributed via third-sector organizations; and (c) public relations (PR) campaign. Yields, efficiency, and costs were calculated. RESULTS: the study recruited 777 (33.9% men) community-dwelling, older adults (mean age 77.55 years (SD 6.79), mean SPPB score 7.37 (SD 1.56)), 95.11% white (n = 739) and broadly representative of UK quintiles of deprivation. Over a 20-month recruitment period, 25,559 invitations were issued. Eighty-eight percent of the participants were recruited via GP invitations, 5.4% via the PR campaign, 3% via word-of-mouth, and 2.5% via third-sector organizations. Mean recruitment cost per participant was £78.47, with an extra £26.54 per recruit paid to GPs to cover research costs. CONCLUSIONS: REACT successfully recruited to target. Response rates were lower than initially predicted and recruitment timescales required adjustment. Written invitations from GPs were the most efficient method for recruiting older adults at risk of mobility disability. Targeted efforts could achieve more ethnically diverse cohorts. All trials should be required to provide recruitment data to enable evidence-based planning of future trials.

Background. Health economic evaluations frequently include projections for lifetime costs and health effects using modeling frameworks such as Markov modeling or discrete event simulation (DES). Markov models typically cannot represent events whose risk is determined by the length of time spent in state (sojourn time) without the use of tunnel states. DES is very flexible but introduces Monte Carlo variation, which can significantly limit the complexity of model analyses. Methods. We present a new methodological framework for health economic modeling that is based on, and extends, the concept of moment-generating functions (MGFs) for time-to-event random variables. When future costs and health effects are discounted, MGFs can be used to very efficiently calculate the total discounted life-years spent in a series of health states. Competing risks are incorporated into the method. This method can also be used to calculate discounted costs and health effects when these payoffs are constant per unit time, one-off, or exponential with regard to time. MGFs are extended to additionally support costs and health effects which are polynomial with regard to time (as in a commonly used model of population norms for EQ-5D utility). Worked Example. A worked example is used to demonstrate the application of the new method in practice and to compare it with Markov modeling and DES. Results are compared in terms of convergence and accuracy, and computation times are compared. R code and an Excel workbook are provided. Conclusions. The MGF method can be applied to health economic evaluations in the place of Markov modeling or DES and has certain advantages over both.

Abstract
Background
Diagnosis of lung cancer frequently occurs in its later stages. Low-dose computed tomography (LDCT) could detect lung cancer early.

Methods
Our objective was to estimate the effect of LDCT lung cancer screening on mortality in high-risk populations. A systematic review of randomised controlled trials (RCTs) comparing LDCT screening programmes with usual care (no screening) or other imaging screening programme (such as chest X-ray (CXR)) was conducted. RCTs of CXR screening were additionally included in the network meta-analysis. Bibliographic sources including MEDLINE, Embase, Web of Science and the Cochrane Library were searched to January 2017. All key review steps were done by two persons. Quality assessment used the Cochrane Risk of Bias tool. Meta-analyses were performed.

Results
Four RCTs were included. More will provide data in the future. Meta-analysis demonstrated that LDCT screening with up to 9.80 years of follow-up was associated with a statistically non-significant decrease in lung cancer mortality (pooled relative risk (RR) 0.94, 95% confidence interval (CI) 0.74 to 1.19; p = 0.62). There was a statistically non-significant increase in all-cause mortality. Given the considerable heterogeneity for both outcomes, the results should be treated with caution.
Network meta-analysis including the four original RCTs plus two further RCTs assessed the relative effectiveness of LDCT, CXR and usual care. The results showed that in terms of lung cancer mortality reduction LDCT was ranked as the best screening strategy, CXR screening as the worst strategy and usual care intermediate.

Conclusions
LDCT screening may be effective in reducing lung cancer mortality but there is considerable uncertainty: the largest of the RCTs compared LDCT with CXR screening rather than no screening; there is imprecision of the estimates; and there is important heterogeneity between the included study results. The uncertainty about the effect on all-cause mortality is even greater. Maturing trials may resolve the uncertainty.

Background: Diagnosis of lung cancer frequently occurs in its later stages. Low-dose computed tomography (LDCT) could detect lung cancer early. Objectives: to estimate the clinical effectiveness and cost-effectiveness of LDCT lung cancer screening in high-risk populations. Data sources: Bibliographic sources included MEDLINE, EMBASE, Web of Science and the Cochrane Library. Methods: Clinical effectiveness - a systematic review of randomised controlled trials (RCTs) comparing LDCT screening programmes with usual care (no screening) or other imaging screening programmes [such as chest X-ray (CXR)] was conducted. Bibliographic sources included MEDLINE, EMBASE, Web of Science and the Cochrane Library. Meta-analyses, including network meta-analyses, were performed. Cost-effectiveness - an independent economic model employing discrete event simulation and using a natural history model calibrated to results from a large RCT was developed. There were 12 different population eligibility criteria and four intervention frequencies [(1) single screen, (2) triple screen, (3) annual screening and (4) biennial screening] and a no-screening control arm. Results: Clinical effectiveness - 12 RCTs were included, four of which currently contribute evidence on mortality. Meta-analysis of these demonstrated that LDCT, with ≤ 9.80 years of follow-up, was associated with a non-statistically significant decrease in lung cancer mortality (pooled relative risk 0.94, 95% confidence interval 0.74 to 1.19). The findings also showed that LDCT screening demonstrated a non-statistically significant increase in all-cause mortality. Given the considerable heterogeneity detected between studies for both outcomes, the results should be treated with caution. Network meta-analysis, including six RCTs, was performed to assess the relative clinical effectiveness of LDCT, CXR and usual care. The results showed that LDCT was ranked as the best screening strategy in terms of lung cancer mortality reduction. CXR had a 99.7% probability of being the worst intervention and usual care was ranked second. Cost-effectiveness - screening programmes are predicted to be more effective than no screening, reduce lung cancer mortality and result in more lung cancer diagnoses. Screening programmes also increase costs. Screening for lung cancer is unlikely to be cost-effective at a threshold of £20,000/quality-adjusted life-year (QALY), but may be cost-effective at a threshold of £30,000/QALY. The incremental cost-effectiveness ratio for a single screen in smokers aged 60-75 years with at least a 3% risk of lung cancer is £28,169 per QALY. Sensitivity and scenario analyses were conducted. Screening was only cost-effective at a threshold of £20,000/QALY in only a minority of analyses. Limitations: Clinical effectiveness - the largest of the included RCTs compared LDCT with CXR screening rather than no screening. Cost-effectiveness - a representative cost to the NHS of lung cancer has not been recently estimated according to key variables such as stage at diagnosis. Certain costs associated with running a screening programme have not been included. Conclusions: LDCT screening may be clinically effective in reducing lung cancer mortality, but there is considerable uncertainty. There is evidence that a single round of screening could be considered cost-effective at conventional thresholds, but there is significant uncertainty about the effect on costs and the magnitude of benefits.

Background: a systematic review was conducted to assess the diagnostic test accuracy of polymerase chain reaction (PCR)-based microsatellite instability (MSI) testing for identifying Lynch syndrome in patients with colorectal cancer (CRC). Unlike previous reviews, this was based on assessing MSI testing against best practice for the reference standard, and included CRC populations that were unselected, age-limited or high-risk for Lynch syndrome. Methods: Single- and two-gate diagnostic test accuracy studies, or similar, were identified, assessed for inclusion, data extracted and quality appraised by two reviewers according to a pre-specified protocol. Sensitivity of MSI testing was estimated for all included studies. Specificity, likelihood ratios and predictive values were estimated for studies that were not based on high-risk samples. Narrative synthesis was conducted. Results: Nine study samples were included. When MSI-Low results were considered to be negative, sensitivity estimates ranged from 67% (95% CI 47, 83) to 100% (95% CI 94, 100). Three studies contributed to estimates of both sensitivity and specificity, with specificity ranging from 61% (95% CI 57, 65), to 93% (95% CI 89, 95). Good sensitivity was achieved at the expense of specificity. When MSI-L was considered to be positive (effectively lowering the threshold for a positive index test result) sensitivity increased and specificity decreased. Between-study heterogeneity in both the MSI and reference standard testing, combined with the low number of studies contributing to both sensitivity and specificity estimates, precluded pooling by meta-analysis. Conclusions: MSI testing is an effective screening test for Lynch syndrome. However, there is significant uncertainty surrounding what balance of sensitivity and specificity will be achieved in clinical practice and how this relates to specific characteristics of the test (such as the panel of markers used or the thresholds used to denote a positive test).

The National Institute for Health and Care Excellence (NICE) invited the manufacturer of azacitidine (Celgene) to submit evidence for the clinical and cost effectiveness of this drug for the treatment of acute myeloid leukaemia with more than 30 % bone marrow blasts in adults who are not eligible for haematopoietic stem cell transplantation, as part of the NICE’s Single Technology Appraisal process. The Peninsula Technology Assessment Group was commissioned to act as the Evidence Review Group (ERG). The ERG produced a critical review of the evidence contained within the company’s submission to NICE. The clinical effectiveness data used in the company’s economic analysis were derived from a single randomised controlled trial, AZA-AML-001. It was an international, multicentre, controlled, phase III study with an open-label, parallel-group design conducted to determine the efficacy and safety of azacitidine against a conventional care regimen (CCR). The CCR was a composite comparator of acute myeloid leukaemia treatments currently available in the National Health Service: intensive chemotherapy followed by best supportive care (BSC) upon disease relapse or progression, non-intensive chemotherapy followed by BSC and BSC only. In AZA-AML-001, the primary endpoint was overall survival. Azacitidine appeared to be superior to the CCR, with median overall survival of 10.4 and 6.5 months, respectively. However, in the intention-to-treat analysis, the survival advantage associated with azacitidine was not statistically significant. The company submitted a de novo economic evaluation based on a partitioned survival model with four health states: “Remission”, “Non-remission”, “Relapse/Progressive disease” and “Death”. The model time horizon was 10 years. The perspective was the National Health Service and Personal Social Services. Costs and health effects were discounted at the rate of 3.5 % per year. The base-case incremental cost-effectiveness ratio (ICER) of azacitidine compared with the CCR was £20,648 per quality-adjusted life-year (QALY) gained. In the probabilistic sensitivity analysis, the mean ICER was £17,423 per QALY. At the willingness-to-pay of £20,000, £30,000 and £50,000 per QALY, the probability of azacitidine being cost effective was 0.699, 0.908 and 0.996, respectively. The ERG identified a number of errors in Celgene’s model and concluded that the results of the company’s economic evaluation could not be considered robust. After amendments to Celgene’s model, the base-case ICER was £273,308 per QALY gained. In the probabilistic sensitivity analysis, the mean ICER was £277,123 per QALY. At a willingness-to-pay of £100,000 per QALY, the probability of azacitidine being cost effective was less than 5 %. In all exploratory analyses conducted by the ERG, the ICER exceeded the NICE’s cost-effectiveness threshold range of £20,000–30,000 per QALY. Given the evidence provided in the submission, azacitidine did not fulfil NICE’s end-of-life criteria. After considering the analyses performed by the ERG and submissions from clinician and patient experts, the NICE Appraisal Committee did not recommend azacitidine for this indication.

Background. Immunosuppression is required in kidney transplantation to prevent rejection and prolong graft survival. We conducted an economic evaluation to support England's National Institute for Health and Care Excellence in developing updated guidance on the use of immunosuppression, incorporating new immunosuppressive agents, and addressing changes in pricing and the evidence base. Methods. A discrete-time state transition model was developed to simulate adult kidney transplant patients over their lifetime. A total of 16 different regimens were modelled to assess the cost-effectiveness of basiliximab and rabbit antithymocyte globulin (rabbit ATG) as induction agents (with no antibody induction as a comparator) and immediate-release tacrolimus, prolonged-release tacrolimus, mycophenolate mofetil, mycophenolate sodium, sirolimus, everolimus and belatacept as maintenance agents (with ciclosporin and azathioprine as comparators). Graft survival was extrapolated from acute rejection rates, graft function and post-transplant diabetes rates, all estimated at 12months post-transplantation. National Health Service (NHS) and personal social services costs were included. Cost-effectiveness thresholds of £20 000 and £30 000 per quality-adjusted life year were used. Results. Basiliximab was predicted to be more effective and less costly than rabbit ATG and induction without antibodies. Immediate-release tacrolimus and mycophenolate mofetil were cost-effective as maintenance therapies. Other therapies were either more expensive and less effective or would only be costeffective if a threshold in excess of £100 000 per quality-adjusted life year were used. Conclusions. A regimen comprising induction with basiliximab, followed by maintenance therapy with immediate-release tacrolimus and mycophenolate mofetil, is likely to be effective for uncomplicated adult kidney transplant patients and a costeffective use of NHS resources.

Background: Inherited mutations in deoxyribonucleic acid (DNA) mismatch repair (MMR) genes lead to an increased risk of colorectal cancer (CRC), gynaecological cancers and other cancers, known as Lynch syndrome (LS). Risk-reducing interventions can be offered to individuals with known LS-causing mutations. The mutations can be identified by comprehensive testing of the MMR genes, but this would be prohibitively expensive in the general population. Tumour-based tests - microsatellite instability (MSI) and MMR immunohistochemistry (IHC) - are used in CRC patients to identify individuals at high risk of LS for genetic testing. MLH1 (MutL homologue 1) promoter methylation and BRAF V600E testing can be conducted on tumour material to rule out certain sporadic cancers. Objectives: to investigate whether testing for LS in CRC patients using MSI or IHC (with or without MLH1 promoter methylation testing and BRAF V600E testing) is clinically effective (in terms of identifying Lynch syndrome and improving outcomes for patients) and represents a cost-effective use of NHS resources. Review methods: Systematic reviews were conducted of the published literature on diagnostic test accuracy studies of MSI and/or IHC testing for LS, end-to-end studies of screening for LS in CRC patients and economic evaluations of screening for LS in CRC patients. A model-based economic evaluation was conducted to extrapolate long-term outcomes from the results of the diagnostic test accuracy review. The model was extended from a model previously developed by the authors. Results: Ten studies were identified that evaluated the diagnostic test accuracy of MSI and/or IHC testing for identifying LS in CRC patients. For MSI testing, sensitivity ranged from 66.7% to 100.0% and specificity ranged from 61.1% to 92.5%. For IHC, sensitivity ranged from 80.8% to 100.0% and specificity ranged from 80.5% to 91.9%. When tumours showing low levels of MSI were treated as a positive result, the sensitivity of MSI testing increased but specificity fell. No end-to-end studies of screening for LS in CRC patients were identified. Nine economic evaluations of screening for LS in CRC were identified. None of the included studies fully matched the decision problem and hence a new economic evaluation was required. The base-case results in the economic evaluation suggest that screening for LS in CRC patients using IHC, BRAF V600E and MLH1 promoter methylation testing would be cost-effective at a threshold of £20,000 per quality-adjusted life-year (QALY). The incremental cost-effectiveness ratio for this strategy was £11,008 per QALY compared with no screening. Screening without tumour tests is not predicted to be cost-effective. Limitations: Most of the diagnostic test accuracy studies identified were rated as having a risk of bias or were conducted in unrepresentative samples. There was no direct evidence that screening improves long-term outcomes. No probabilistic sensitivity analysis was conducted. Conclusions: Systematic review evidence suggests that MSI- and IHC-based testing can be used to identify LS in CRC patients, although there was heterogeneity in the methods used in the studies identified and the results of the studies. There was no high-quality empirical evidence that screening improves long-term outcomes and so an evidence linkage approach using modelling was necessary. Key determinants of whether or not screening is cost-effective are the accuracy of tumour-based tests, CRC risk without surveillance, the number of relatives identified for cascade testing, colonoscopic surveillance effectiveness and the acceptance of genetic testing. Future work should investigate screening for more causes of hereditary CRC and screening for LS in endometrial cancer patients.

© Queen’s Printer and Controller of HMSO 2017. Background: Colorectal cancer is the fourth most commonly diagnosed cancer in the UK after breast, lung and prostate cancer. People with metastatic disease who are sufficiently fit are usually treated with active chemotherapy as first-or second-line therapy. Targeted agents are available, including the antiepidermal growth factor receptor (EGFR) agents cetuximab (Erbitux®, Merck Serono UK Ltd, Feltham, UK) and panitumumab (Vecitibix®, Amgen UK Ltd, Cambridge, UK). Objective: to investigate the clinical effectiveness and cost-effectiveness of panitumumab in combination with chemotherapy and cetuximab in combination with chemotherapy for rat sarcoma (RAS) wild-type (WT) patients for the first-line treatment of metastatic colorectal cancer. Data sources: the assessment included a systematic review of clinical effectiveness and cost-effectiveness studies, a review and critique of manufacturer submissions, and a de novo cohort-based economic analysis. For the assessment of effectiveness, a literature search was conducted up to 27 April 2015 in a range of electronic databases, including MEDLINE, EMBASE and the Cochrane Library. Review methods: Studies were included if they were randomised controlled trials (RCTs) or systematic reviews of RCTs of cetuximab or panitumumab in participants with previously untreated metastatic colorectal cancer with RAS WT status. All steps in the review were performed by one reviewer and checked independently by a second. Narrative synthesis and network meta-analyses (NMAs) were conducted for outcomes of interest. An economic model was developed focusing on first-line treatment and using a 30-year time horizon to capture costs and benefits. Costs and benefits were discounted at 3.5% per annum. Scenario analyses and probabilistic and univariate deterministic sensitivity analyses were performed. Results: the searches identified 2811 titles and abstracts, of which five clinical trials were included. Additional data from these trials were provided by the manufacturers. No data were available for panitumumab plus irinotecan-based chemotherapy (folinic acid + 5-fluorouracil + irinotecan) (FOLFIRI) in previously untreated patients. Studies reported results for RAS WT subgroups. First-line treatment with anti-EGFR therapies in combination with chemotherapy appeared to have statistically significant benefits for patients who are RAS WT. For the independent economic evaluation, the base-case incremental cost-effectiveness ratio (ICER) for RAS WT patients for cetuximab plus oxaliplatin-based chemotherapy (folinic acid + 5-fluorouracil + oxaliplatin) (FOLFOX) compared with FOLFOX was £104,205 per quality-adjusted life-year (QALY) gained; for panitumumab plus FOLFOX compared with FOLFOX was £204,103 per QALY gained; and for cetuximab plus FOLFIRI compared with FOLFIRI was £122,554 per QALY gained. The ICERs were sensitive to treatment duration, progression-free survival, overall survival (resected patients only) and resection rates. Limitations: the trials included RAS WT populations only as subgroups. No evidence was available for panitumumab plus FOLFIRI. Two networks were used for the NMA and model, based on the different chemotherapies (FOLFOX and FOLFIRI), as insufficient evidence was available to the assessment group to connect these networks. Conclusions: Although cetuximab and panitumumab in combination with chemotherapy appear to be clinically beneficial for RAS WT patients compared with chemotherapy alone, they are likely to represent poor value for money when judged by cost-effectiveness criteria currently used in the UK. It would be useful to conduct a RCT in patients with RAS WT.

Background: End-stage renal disease is a long-term irreversible decline in kidney function requiring renal replacement therapy: Kidney transplantation, haemodialysis or peritoneal dialysis. The preferred option is kidney transplantation, followed by immunosuppressive therapy (induction and maintenance therapy) to reduce the risk of kidney rejection and prolong graft survival. Objectives: to review and update the evidence for the clinical effectiveness and cost-effectiveness of basiliximab (BAS) (Simulect®, Novartis Pharmaceuticals UK Ltd) and rabbit anti-human thymocyte immunoglobulin (rATG) (Thymoglobulin®, Sanofi) as induction therapy, and immediate-release tacrolimus (TAC) (Adoport®, Sandoz; Capexion®, Mylan; Modigraf®, Astellas Pharma; Perixis®, Accord Healthcare; Prograf®, Astellas Pharma; Tacni®, Teva; Vivadex®, Dexcel Pharma), prolonged-release tacrolimus (Advagraf® Astellas Pharma), belatacept (BEL) (Nulojix®, Bristol-Myers Squibb), mycophenolate mofetil (MMF) (Arzip®, Zentiva; CellCept®, Roche Products; Myfenax®, Teva), mycophenolate sodium (MPS) (Myfortic®, Novartis Pharmaceuticals UK Ltd), sirolimus (SRL) (Rapamune®, Pfizer) and everolimus (EVL) (Certican®, Novartis) as maintenance therapy in adult renal transplantation. Methods: Clinical effectiveness searches were conducted until 18 November 2014 in MEDLINE (via Ovid), EMBASE (via Ovid), Cochrane Central Register of Controlled Trials (via Wiley Online Library) and Web of Science (via ISI), Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects and Health Technology Assessment (The Cochrane Library via Wiley Online Library) and Health Management Information Consortium (via Ovid). Cost-effectiveness searches were conducted until 18 November 2014 using a costs or economic literature search filter in MEDLINE (via Ovid), EMBASE (via Ovid), NHS Economic Evaluation Database (via Wiley Online Library), Web of Science (via ISI), Health Economic Evaluations Database (via Wiley Online Library) and the American Economic Association’s electronic bibliography (via EconLit, EBSCOhost). Included studies were selected according to predefined methods and criteria. A random-effects model was used to analyse clinical effectiveness data (odds ratios for binary data and mean differences for continuous data). Network meta-analyses were undertaken within a Bayesian framework. A new discrete time–state transition economic model (semi-Markov) was developed, with acute rejection, graft function (GRF) and new-onset diabetes mellitus used to extrapolate graft survival. Recipients were assumed to be in one of three health states: Functioning graft, graft loss or death. Results: Eighty-nine randomised controlled trials (RCTs), of variable quality, were included. For induction therapy, no treatment appeared more effective than another in reducing graft loss or mortality. Compared with placebo/no induction, rATG and BAS appeared more effective in reducing biopsy-proven acute rejection (BPAR) and BAS appeared more effective at improving GRF. For maintenance therapy, no treatment was better for all outcomes and no treatment appeared most effective at reducing graft loss. BEL + MMF appeared more effective than TAC + MMF and SRL + MMF at reducing mortality. MMF + CSA (ciclosporin), TAC + MMF, SRL + TAC, TAC + AZA (azathioprine) and EVL + CSA appeared more effective than CSA + AZA and EVL + MPS at reducing BPAR. SRL + AZA, TAC + AZA, TAC + MMF and BEL + MMF appeared to improve GRF compared with CSA + AZA and MMF + CSA. In the base-case deterministic and probabilistic analyses, BAS, MMF and TAC were predicted to be cost-effective at £20,000 and £30,000 per quality-adjusted life-year (QALY). When comparing all regimens, only BAS + TAC + MMF was cost-effective at £20,000 and £30,000 per QALY. Limitations: for included trials, there was substantial methodological heterogeneity, few trials reported follow-up beyond 1 year, and there were insufficient data to perform subgroup analysis. Treatment discontinuation and switching were not modelled. Future work: High-quality, better-reported, longer-term RCTs are needed. Ideally, these would be sufficiently powered for subgroup analysis and include health-related quality of life as an outcome. Conclusion: Only a regimen of BAS induction followed by maintenance with TAC and MMF is likely to be cost-effective at £20,000–30,000 per QALY.

Background: End-stage renal disease is a long-term irreversible decline in kidney function requiring kidney transplantation, haemodialysis or peritoneal dialysis. The preferred option is kidney transplantation followed by induction and maintenance immunosuppressive therapy to reduce the risk of kidney rejection and prolong graft survival. Objectives: to systematically review and update the evidence for the clinical effectiveness and cost-effectiveness of basiliximab (BAS) (Simulect,® Novartis Pharmaceuticals) and rabbit antihuman thymocyte immunoglobulin (Thymoglobuline,® Sanofi) as induction therapy and immediate-release tacrolimus [Adoport® (Sandoz); Capexion® (Mylan); Modigraf® (Astellas Pharma); Perixis® (Accord Healthcare); Prograf® (Astellas Pharma); Tacni® (Teva); Vivadex® (Dexcel Pharma)], prolonged-release tacrolimus (Advagraf,® Astellas Pharma); belatacept (BEL) (Nulojix,® Bristol-Myers Squibb), mycophenolate mofetil (MMF) [Arzip® (Zentiva), CellCept® (Roche Products), Myfenax® (Teva), generic MMF is manufactured by Accord Healthcare, Actavis, Arrow Pharmaceuticals, Dr Reddy’s Laboratories, Mylan, Sandoz and Wockhardt], mycophenolate sodium, sirolimus (Rapamune,® Pfizer) and everolimus (Certican,® Novartis Pharmaceuticals) as maintenance therapy in children and adolescents undergoing renal transplantation. Data sources: Clinical effectiveness searches were conducted to 7 January 2015 in MEDLINE (via Ovid), EMBASE (via Ovid), Cochrane Central Register of Controlled Trials (via Wiley Online Library) and Web of Science [via Institute for Scientific Information (ISI)], Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects and Health Technology Assessment (HTA) (The Cochrane Library via Wiley Online Library) and Health Management Information Consortium (via Ovid). Cost-effectiveness searches were conducted to 15 January 2015 using a costs or economic literature search filter in MEDLINE (via Ovid), EMBASE (via Ovid), NHS Economic Evaluation Databases (via Wiley Online Library), Web of Science (via ISI), Health Economic Evaluations Database (via Wiley Online Library) and EconLit (via EBSCOhost). Review methods: Titles and abstracts were screened according to predefined inclusion criteria, as were full texts of identified studies. Included studies were extracted and quality appraised. Data were meta-analysed when appropriate. A new discrete time state transition economic model (semi-Markov) was developed; graft function, and incidences of acute rejection and new-onset diabetes mellitus were used to extrapolate graft survival. Recipients were assumed to be in one of three health states: functioning graft, graft loss or death. Results: Three randomised controlled trials (RCTs) and four non-RCTs were included. The RCTs only evaluated BAS and tacrolimus (TAC). No statistically significant differences in key outcomes were found between BAS and placebo/no induction. Statistically significantly higher graft function (p < 0.01) and less biopsy-proven acute rejection (odds ratio 0.29, 95% confidence interval 0.15 to 0.57) was found between TAC and ciclosporin (CSA). Only one cost-effectiveness study was identified, which informed NICE guidance TA99. BAS [with TAC and azathioprine (AZA)] was predicted to be cost-effective at £20,000-30,000 per quality-adjusted life year (QALY) versus no induction (BAS was dominant). BAS (with CSA and MMF) was not predicted to be cost-effective at £20,000-30,000 per QALY versus no induction (BAS was dominated). TAC (with AZA) was predicted to be cost-effective at £20,000-30,000 per QALY versus CSA (TAC was dominant). A model based on adult evidence suggests that at a cost-effectiveness threshold of £20,000-30,000 per QALY, BAS and TAC are cost-effective in all considered combinations; MMF was also cost-effective with CSA but not TAC. Limitations: the RCT evidence is very limited; analyses comparing all interventions need to rely on adult evidence. Conclusions: TAC is likely to be cost-effective (vs. CSA, in combination with AZA) at £20,000-30,000 per QALY. Analysis based on one RCT found BAS to be dominant, but analysis based on another RCT found BAS to be dominated. BAS plus TAC and AZA was predicted to be cost-effective at £20,000-30,000 per QALY when all regimens were compared using extrapolated adult evidence. High-quality primary effectiveness research is needed. The UK Renal Registry could form the basis for a prospective primary study.

BACKGROUND: Lynch syndrome is an autosomal dominant cancer predisposition syndrome caused by mutations in the DNA mismatch repair genes MLH1, MSH2, MSH6 and PMS2. Individuals with Lynch syndrome have an increased risk of colorectal cancer, endometrial cancer, ovarian and other cancers. Lynch syndrome remains underdiagnosed in the UK. Reflex testing for Lynch syndrome in early-onset colorectal cancer patients is proposed as a method to identify more families affected by Lynch syndrome and offer surveillance to reduce cancer risks, although cost-effectiveness is viewed as a barrier to implementation. The objective of this project was to estimate the cost-utility of strategies to identify Lynch syndrome in individuals with early-onset colorectal cancer in the NHS. METHODS: a decision analytic model was developed which simulated diagnostic and long-term outcomes over a lifetime horizon for colorectal cancer patients with and without Lynch syndrome and for relatives of those patients. Nine diagnostic strategies were modelled which included microsatellite instability (MSI) testing, immunohistochemistry (IHC), BRAF mutation testing (methylation testing in a scenario analysis), diagnostic mutation testing and Amsterdam II criteria. Biennial colonoscopic surveillance was included for individuals diagnosed with Lynch syndrome and accepting surveillance. Prophylactic hysterectomy with bilateral salpingo-oophorectomy (H-BSO) was similarly included for women diagnosed with Lynch syndrome. Costs from NHS and Personal Social Services perspective and quality-adjusted life years (QALYs) were estimated and discounted at 3.5% per annum. RESULTS: all strategies included for the identification of Lynch syndrome were cost-effective versus no testing. The strategy with the greatest net health benefit was MSI followed by BRAF followed by diagnostic genetic testing, costing £5,491 per QALY gained over no testing. The effect of prophylactic H-BSO on health-related quality of life (HRQoL) is uncertain and could outweigh the health benefits of testing, resulting in overall QALY loss. CONCLUSIONS: Reflex testing for Lynch syndrome in early-onset colorectal cancer patients is predicted to be a cost-effective use of limited financial resources in England and Wales. Research is recommended into the cost-effectiveness of reflex testing for Lynch syndrome in other associated cancers and into the impact of prophylactic H-BSO on HRQoL.

BACKGROUND: in breast cancer patients, sentinel lymph node biopsy is carried out at the same time as the removal of the primary tumour to postoperatively test with histopathology for regional metastases in the sentinel lymph node. Those patients with positive test results are then operated on 2-4 weeks after primary surgery to remove the lymph nodes from the axilla (axillary lymph node dissection, ALND). New molecular tests RD-100i [one-step nucleic acid amplification (OSNA); based on messenger RNA amplification to identify the cytokeratin-19 (CK19) gene marker] (Sysmex, Norderstedt, Germany) and Metasin (using the CK19 and mammaglobin gene markers) (Cellular Pathology, Princess Alexandra Hospital NHS Trust, Harlow, UK) are intended to provide an intraoperative diagnosis, thereby avoiding the need for postoperative histopathology and, in positive cases, a second operation for ALND. OBJECTIVE: to evaluate the clinical effectiveness and cost-effectiveness of using OSNA and Metasin in the NHS in England for the intraoperative diagnosis of sentinel lymph nodes metastases, compared with postoperative histopathology, the current standard. DATA SOURCES: Electronic databases including MEDLINE, MEDLINE In-Process & Other Non-Indexed Citations, EMBASE, the Cochrane Library and the Health Economic Evaluations Database as well as clinical trial registries, grey literature and conference proceedings were searched up to July 2012. REVIEW METHODS: a systematic review of the evidence was carried out using standard methods. Single-gate studies were used to estimate the accuracy of OSNA with histopathology as the reference standard. The cost-effectiveness analysis adapted an existing simulation model of the long-term costs and health implications of early breast cancer diagnostic outcomes. The model accounted for the costs of an extended first operation with intraoperative testing, the loss of health-related quality of life (disutility) from waiting for postoperative test results, disutility and costs of a second operation, and long-term costs and disutility from lymphoedema related to ALND, adjuvant therapy, locoregional recurrence and metastatic recurrence. RESULTS: a total of 724 references were identified in the searches, of which 17 studies assessing test accuracy were included in the review, 15 on OSNA and two on Metasin. Both Metasin studies were unpublished. OSNA sensitivity of 84.5% [95% confidence interval (CI) 74.7% to 91.0%] and specificity of 91.8% (95% CI 87.8% to 94.6%) for patient nodal status were estimated in a meta-analysis of five studies [unadjusted for tissue allocation bias (TAB)]. At these values and a 20% node-positive rate, OSNA resulted in lifetime discounted cost-savings of £498 and a quality-adjusted life-year (QALY) loss of 0.048 relative to histopathology, that is, £4324 saved per QALY lost. The most favourable plausible scenario for OSNA in terms of the node-positive rate (range 10-40%), diagnostic accuracy values (91.3% sensitivity and 94.2% specificity, from three reports that adjusted for TAB), the costs of histopathology, OSNA and second surgery, and long-term costs and utilities resulted in a maximum saving per QALY lost of £10,500; OSNA sensitivity and specificity would need to be ≥ 95% for this figure to be ≥ £20,000. LIMITATIONS: There is limited evidence on the diagnostic test accuracy of intraoperative tests. The quality of information on costs of resource utilisation during the diagnostic pathway is low and no evidence exists on the disutility of waiting for a second surgery. No comparative studies exist that report clinical outcomes of intraoperative diagnostic tests. These knowledge gaps have more influence on the decision than current uncertainty in the performance of postoperative histopathology in standard practice. CONCLUSIONS: One-step nucleic acid amplification is not cost-effective for the intraoperative diagnosis of sentinel lymph node metastases. OSNA is less accurate than histopathology and the consequent loss of health benefits in this patient group is not compensated for by health gains elsewhere in the health system that may be obtained with the cost-savings made. The evidence on Metasin is insufficient to evaluate its cost-effectiveness. STUDY REGISTRATION: This study is registered as PROSPERO CRD42012002889. FUNDING: the National Institute for Health Research Health Technology Assessment programme.

We present NOAM, an integrated platform for the monitoring and analysis of news media content. NOAM is the data management system behind various applications and scientific studies aiming at modelling the mediasphere. The system is also intended to address the need in the AI community for platforms where various AI technologies are integrated and deployed in the real world. It combines a relational database (DB) with state of the art AI technologies, including data mining, machine learning and natural language processing. These technologies are organised in a robust, distributed architecture of collaborating modules, that are used to populate and annotate the DB. NOAM manages tens of millions of news items in multiple languages, automatically annotating them in order to enable queries based on their semantic properties. The system also includes a unified user interface for interacting with its various modules. © 2011 Authors.

Inferring causal networks behind observed data is an active area of research with wide applicability to areas such as epidemiology, microbiology and social science. In particular recent research has focused on identifying how information propagates through the Internet. This research has so far only used temporal features of observations, and while reasonable results have been achieved, there is often further information which can be used. In this paper we show that additional features of the observed data can be used very effectively to improve an existing method. Our particular example is one of inferring an underlying network for how text is reused in the Internet, although the general approach is applicable to other inference methods and information sources. We develop a method to identify how a piece of text evolves as it moves through an underlying network and how substring information can be used to narrow down where in the evolutionary process a particular observation at a node lies. Hence we narrow down the number of ways the node could have acquired the infection. Text reuse is detected using a suffix tree which is also used to identify the substring relations between chunks of reused text. We then use a modification of the NetCover method to infer the underlying network. Experimental results - on both synthetic and real life data - show that using more information than just timing leads to greater accuracy in the inferred networks. Copyright 2011 ACM.

We present a demonstration of a newly developed text stream event detection method on over a million articles from the New York Times corpus. The event detection is designed to operate in a predominantly on-line fashion, reporting new events within a specified timeframe. The event detection is achieved by detecting significant changes in the statistical properties of the text where those properties are efficiently stored and updated in a suffix tree. This particular demonstration shows how our method is effective at discovering both short-and long-term events (which are often denoted topics), and how it automatically copes with topic drift on a corpus of 1 035 263 articles. © 2010 Springer-Verlag Berlin Heidelberg.

We address the task of detecting surprising patterns in large textual data streams. These can reveal events in the real world when the data streams are generated by online news media, emails, Twitter feeds, movie subtitles, scientific publications, and more. The volume of interest in such text streams often exceeds human capacity for analysis, such that automatic pattern recognition tools are indispensable. In particular, we are interested in surprising changes in the frequency of n-grams of words, or more generally of symbols from an unlimited alphabet size. Despite the exponentially large number of possible n-grams in the size of the alphabet (which is itself unbounded), we show how these can be detected efficiently. To this end, we rely on a data structure known as a generalised suffix tree, which is additionally annotated with a limited amount of statistical information. Crucially, we show how the generalised suffix tree as well as these statistical annotations can efficiently be updated in an on-line fashion. © 2010 IEEE.

We present a complete working system that gathers multilingual news items from the Web, translates them into English, categorises them by topic and geographic location and presents them to the final user in a uniform way. Currently, the system crawls 560 news outlets, in 22 different languages, from the 27 European Union countries. Data gathering is based on RSS crawlers, machine translation on Moses and the text categorisation on SVMs. The system also presents on a European map statistical information about the amount of attention devoted to the various topics in each of the 27 EU countries. The integration of Support Vector Machines, Statistical Machine Translation, Web Technologies and Computer Graphics delivers a complete system where modern Statistical Machine Learning is used at multiple levels and is a crucial enabling part of the resulting functionality. © 2009 Springer Berlin Heidelberg.