2000 Cost-effectiveness modelling

The randomised controlled clinical trial is a key tool for providing unbiased assessment of one or more experimental treatments, but a trial in isolation does not provide sufficient information to guide clinical management of patients with a particular disease. Cost-effectiveness modelling combines evidence from other trials of the same and competing therapies, on the disease incidence and severity, the risks of associated events and longer-term outcomes, and the resources and costs required to treat patients. The Unit has contributed to the early development of robust cost-effectiveness modelling for NICE and other policy decisions by providing exemplars in diverse clinical areas such as the assessment of cardiac imaging strategies, mechanical devices to support the failing heart and staging of lung cancer.

References

1. Thompson SG, Barber JA. How should cost data in pragmatic randomized trials be analysed? British Medical Journal 2000; 320: 1197-1200.
2. Nixon RM, Thompson SG. Parametric modelling of cost data in medical studies. Statistics in Medicine 2004; 23: 1311-1331.
3. Sharples L D, Buxton M, Caine N, Cafferty F, Demiris N, Dyer M, Freeman C. Evaluation of ventricular assist device programme in the UK Health Technology Assessment 2006; 10:1-138.
4. Sharples L D, Hughes V, Crean A, Dyer M, Buxton M, Goldsmith K, Stone D for the CECaT study group. Cost-effectiveness of functional cardiac testing in the diagnosis and management of coronary artery disease: an RCT of stress echocardiography, cardiac magnetic resonance imaging or SPECT as the initial test compared with angiography (CECaT). Health Technology Assessment 2007;11:1-136.
5. Jackson C H, Sharples L D, Thompson S G. Structural and parameter uncertainty in Bayesian cost-effectiveness models. Journal of the Royal Statistical Society Series C (Applied Statistics) 2010; 59: 233-253.
6. Jackson C H, Bojke L, Thompson S G, Claxton K, Sharples L D. Addressing structural uncertainty in decision models. Medical Decision Making 2011; 31: 662-674.
7. Sharples L D, Jackson C, Wheaton E, Griffith G, Annema J, Dooms C, Tournoy K, Deschepper E, Hughes V, Magee L, Buxton M, Rintoul R. Effectiveness and cost-effectiveness of Endobronchial and Endoscopic Ultrasound relative to Surgical Staging in Potentially Resectable Lung Cancer: results from the ASTER Randomised Controlled Trial. Journal of Health Technology Assessment 2012; 16: 1-82.

 

2000 The statistics of ageing

Since the late 1980s, the MRC Biostatistics Unit had undertaken the methodological development and study design for two large ageing studies: the MRC Cognitive Function and Ageing (CFAS) studies I and II (which include CFAS Wales) and the Cambridge City over 75 Cohort study (CC75C). Both studies have provided a rich source of papers for statistical development as well as providing great insight into population health of the ageing.

Most recently, and with much press coverage, Matthews et al. published a 2-decade comparison of the prevalence of dementia in individuals aged 65 years and older.

However, the Unit’s earliest publication on the biostatistics of senility was in 1939!

References

1. MRC CFA Study (Our contributors C. Gill, A.L. Johnson, D. Matthewson, M.A. McGee, N. Walker and N.E. Day). Cognitive function and dementia in six areas of England and Wales: The distribution of MMSE and prevalence of GMS organicity level in the MRC CFA Study. Psychological Medicine 1998; 28: 319-335.
2. Clayton DG, Spiegelhalter DJ, Dunn G, Pickles A. Analysis of longitudinal binary data from multiphase sampling (with Discussion). Journal of the Royal Statistical Society Series B 1998; 60: 71-87.
3. Brayne C, Spiegelhalter DJ, Dufouil C, Chi LY, Dening TR, Paykel ES, O’Connor DW, Ahmed A, McGee MA, Huppert FA. Estimating the true extent of cognitive decline in the old old. Journal of the American Geriatrics Society 1999; 47: 1283-1288.
4. The Neuropathology Group MRC CFAS (Writing committee: Esiri M, Matthews FE, and Brayne C; Ince P referee). Pathological correlates of late-onset dementia in a multicentre, community-based population in England and Wales. Lancet 2001; 357: 169-175.
5. Matthews FE, Brayne C. The incidence of dementia in England and Wales: findings from the five identical sites of the MRC CFA Study. PLOS Medicine 2005; 2: e193.
6. Matthews FE, Arthur A, Barnes LE, Bond J, Jagger C, Robinson L, Brayne C on behalf of the medical Research Council Cognitive Function and Ageing Collaboration. A two-decade comparison of prevalence of dementia in individuals aged 65 years and older from three geographical areas of England: results of the Cognitive Function and Ageing Study I and II. Lancet 2013; 382: 1405-1412.
7. Greenwood M, Irwin JO. The bisotatistics of senility. Human Biology 1939; 11: 1.

 

2002 Higgins and Thompson’s I2: quantifying heterogeneity in a meta-analysis

Julian Higgins and Simon Thompson proposed several measures for quantifying between-study heterogeneity and inconsistency in meta-analysis, including the I2 measure. By the end of 2013, this work had been cited more than 8000 times (BMJ paper) and more than 5000 times (Statistics in Medicine paper), making the two papers some of the most highly cited in their respective journals.

References

1. Higgins JPT, Thompson SG (2002). Quantifying heterogeneity in a meta-analysis. Statistics in Medicine 2002; 21: 1539-1558.
   2. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. British Medical Journal 2003; 327: 557-560.

 

2002 Multicentre Aneurysm Screening Study (MASS): randomized trial and cost-effectiveness

Abdominal aortic aneurysm (AAA) is a major cause of death in older men, in the UK and elsewhere. The large MASS randomised trial, led by the MRC Biostatistics Unit, evaluated the long-term benefits of ultrasound screening for AAA in men aged 65-74 years. It showed that mortality from AAA was halved, and that a screening programme would be highly cost-effective for the NHS. This provided the basis for the introduction of a UK national AAA screening programme in men aged 65; this was announced in 2008, initiated in 2009, and achieved full coverage of England in 2013.

References

1. Multicentre Aneurysm Screening Study Group (Scott RAP, Ashton HA, Buxton MJ, Day NE, Kim LG, Marteau TM, Thompson SG, Walker NM). The multicentre aneurysm screening study (MASS) into the effect of abdominal aortic aneurysm screening on mortality in men: a randomised controlled trial. Lancet 2002; 360: 1531-1539.
2. Multicentre Aneurysm Screening Study Group (Buxton MJ, Thompson SG, Campbell HE, Kim LG, Day NE, Marteau TM, Ashton HA, Scott RA). Multicentre aneurysm screening study (MASS): cost effectiveness analysis of screening for abdominal aortic aneurysms based on four year results from randomised controlled trial. British Medical Journal 2002; 325: 1135-1138.
3. Kim LG, Scott RAP, Ashton HA, Thompson SG. A sustained mortality benefit from screening for abdominal aortic aneurysm. Annals of Internal Medicine 2007; 146: 699-706.
4. Kim LG, Thompson SG, Briggs AH, Buxton MJ, Campbell HE. How cost-effective is screening for abdominal aortic aneurysms? Journal of Medical Screening 2007; 14: 46-52.
5. Thompson SG, Ashton HA, Gao L, Scott RAP. Screening men for abdominal aortic aneurysm: 10 year mortality and cost effectiveness results from the randomised Multicentre Aneurysm Screening Study. British Medical Journal 2009; 338: b2307.

 

2002 Bayesian measures of model complexity and fit

Model performance is typically assessed by taking a measure of model fit and penalising this with a measure of complexity. The Deviance Information Criterion (DIC), introduced by Speigelhalter, Best, Carlin and van der Linde, made use of a novel definition of complexity, the ‘effective’ number of parameters, which is applicable to Bayesian hierarchical models.
The DIC paper had over 1000 citations by the end of 2013 as DIC has become one of the most widely used tools for Bayesian model comparison. The paper and subsequent discussion have led to continued research into methods for Bayesian model criticism, including within the Unit.

References

1. Spiegelhalter DJ, Best NG, Carlin BR, van der Linde A (2002). Bayesian measures of model complexity and fit (with Discussion). Journal of the Royal Statistical Society Series B-Statistical Methodology 2002; 64: 583-639.
2. Marshall EC, Spiegelhalter DJ. Approximate cross-validatory predictive checks in disease mapping models. Statistics in Medicine 2003; 22: 1649-1660.
3. Demiris N, Sharples LD. Bayesian evidence synthesis to extrapolate survival estimates in cost-effectiveness studies. Statistics in Medicine 2006; 25: 1960-1975.
4. Sweeting MJ, De Angelis D, Hickman M, Ades, AE. Estimating hepatitis C prevalence in England and Wales by synthesizing evidence from multiple data sources. Assessing data conflict and model fit. Biostatistics 2008; 9: 715-734.
5. Presanis AM, Ohlssen D, Spiegelhalter DJ, De Angelis, D. Conflict diagnostics in directed acyclic graphs, with applications in Bayesian evidence synthesis. Statistical Science 2013; 28: 376-397.

 

2002 Multi-state Markov models for disease progression: first release of the MSM software

Multi-state Markov and hidden Markov models for modelling disease progression had been developed in the previous two decades, but there was no software for applying them easily until Christopher Jackson developed the msm package for R. The program is fully documented and flexible, allowing any multi-state model structure. The MSM program has been continually maintained and developed, and has enabled the wider use of these methods not only in medical applications but also in finance, ecology, social science and engineering.

References

1. Jackson CH, Sharples LD. Hidden Markov models for the onset and progression of bronchiolitis obliterans syndrome in lung transplant recipients. Statistics in Medicine 2002; 21: 113-128.
2. Jackson CH, Sharples LD, Thompson SG, Duffy SW, Couto, E. Multi-state Markov models for disease progression with misclassification error. The Statistician 2003; 52: 193-209.
3. Titman M, Sharples LD. Model diagnostics for multi-state models. Statistical Methods in Medical Research 2010; 19: 621-651.
4. Jackson CH. Multi-State Models for Panel Data: The msm Package for R. Journal of Statistical Software 2011; 38:8.

 

2004 Cochrane Reviewers’ Handbook, Issue 1.

Julian Higgins is co-editor of the Cochrane Handbook for Systematic Reviews of Interventions. A major revision was published in 2008, when it was published as a book by Wiley-Blackwell. This book had sold over 5000 copies by the end of 2013. The online version of the Handbook is regularly updated.

References

1. Alderson P, Green S, Higgins JPT (Editors). Cochrane Reviewers’ Handbook 4.2.2 [updated March 2004]. In: The Cochrane Library, Issue 1, 2004. Chichester, UK: John Wiley & Sons, Ltd.
2. Higgins JPT, Green S. (Editors). Cochrane Handbook for Systematic Reviews of Interventions. John Wiley & Sons, 2008

 

2004 The statistics of missing data: multiple imputation

Ian White worked with Patrick Royston (MRC Clinical Trials Unit) on the implementation of Multivariate Imputation by Chained Equations (MICE) in Stata. This software was widely used, and was supported by short courses run in BSU. Ian and Patrick subsequently advised Stata Corporation on its implementation of multiple imputation in 2009-2011.

References

1. Royston P. Multiple imputation of missing values. Stata Journal 2004; 4: 227–241.
2. Sterne JAC, White IR, Carlin JB, Spratt M, Royston P, Kenward MG, Wood AM, Carpenter JR. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. British Medical Journal 2009; 338: b2393.
3. White IR, Daniel R, Royston P. Avoiding bias due to perfect prediction in multiple imputation of incomplete categorical variables. Computational Statistics and Data Analysis 2010; 54: 2267–2275.
4. Royston P, White IR. Multiple imputation by chained equations (MICE): implementation in Stata. Journal of Statistical Software 2011; 45: 4.
5. White IR, Wood A, Royston P. Tutorial in biostatistics: Multiple imputation using chained equations: issues and guidance for practice. Statistics in Medicine 2011; 30: 377–399.