Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2007 September 29; 335(7621): 622–623.
Published online 2007 September 14. doi:  10.1136/bmj.39332.587581.BE
PMCID: PMC1995490

New methods of analysing cost effectiveness

Andrew H Briggs, Lindsay chair in health policy and economic evaluation

Value of information analyses must be integrated into the process of commissioning primary research

Interest in whether health interventions are value for money as well as effective has meant that the term cost effectiveness1 is commonly used (and sometimes misused) in the clinical literature. Consequently, methods for determining cost effectiveness have been refined, especially techniques for synthesising evidence and representing uncertainty in the results of such evaluations. Techniques such as multi-parameter evidence synthesis2 and value of information analysis3 are now routinely integrated into cost effectiveness studies, especially health technology appraisals (HTAs) conducted for the National Institute for Health and Clinical Excellence. But is there real value in the development and application of such techniques, or have these new methods emerged simply as a consequence of involving academics in the process of evaluation?

Colbourn and colleagues present a cost effectiveness and value of information analysis of strategies for preventing group B streptococcal and other bacterial infections in early infancy.4 This is a timely assessment of the potential cost effectiveness of various ways of organising a national screening programme for group B streptococci, which has the potential to influence UK policy on whether (and how) to implement such a screening programme. However, what do the sophisticated techniques used add to what we already know about the effectiveness and cost effectiveness of preventive strategies for this infection?

Firstly, the techniques of decision analysis combined with multi-parameter evidence synthesis allow a comprehensive assessment of all of the evidence that relates to the policy question, including the consideration of all possible strategies (something Colbourn and colleagues have taken to the extreme, with 341 strategies evaluated in this paper alone). This contrasts with the Cochrane review approach, which typically uses only randomised evidence to assess a single treatment comparison.

In addition, a probabilistic analysis of uncertainty in the parameters of the model allows a full assessment of the implication of the estimated uncertainty for the decision. This means the analysis can answer two fundamental questions relating to the choice between the strategies evaluated. Firstly, on the basis of the existing evidence, what is the preferred course of action? Secondly, should additional information be collected to better inform that decision?

The analysis by Colbourn and colleagues shows that, on the basis of existing evidence, it is likely that immediate changes to the organisation and delivery of services to prevent group B streptococcal infection would greatly benefit the health service. Furthermore, given the size of the population concerned, it highlights the value of further research, particularly into the potential use of an intervention (vaccination) that is not yet available in the United Kingdom. However, as the authors point out, even though further research may be valuable this does not mean that the proposed trial of screening for group B streptococci, at an estimated cost of £12m (€18m; $24m), is the correct way forward. Indeed, the analysis suggests that the screening strategies proposed as comparators in this trial are unlikely to be cost effective.

In the absence of an available vaccine, the value of additional evidence currently lies elsewhere—particularly in resolving the choice between intravenous and oral drugs for certain preterm infants. It is unfortunate that the two teams responsible for the synthesis of evidence5 and the design of a clinical trial, both funded by the HTA programme, seem to have worked independently and concurrently. The value of comprehensive evidence synthesis and value of information analysis is to inform the design of further research studies.

The Cooksey review called for an expansion of the National Health Service HTA programme “to enhance the evidence base informing decisions on the effectiveness and cost-effectiveness of technologies in the NHS,”6 while recognising the need from the outset to “develop a system of metrics that can accurately evaluate the impact of this expansion.”6 Value of information analysis, by seeking directly to investigate the potential returns to further investment in research, offers exactly this metric. However, for it to fulfil its full potential, it must become an integrated part of the process of commissioning primary research.


Competing interests: None declared.

Provenance and peer review: Commissioned; not externally peer reviewed.


1. Doubilet P, Weinstein MC, McNeil BJ. Use and misuse of the term “cost effective” in medicine. N Engl J Med 1986;314:253-6. [PubMed]
2. Ades AE, Cliffe S. Markov chain Monte Carlo estimation of a multiparameter decision model: consistency of evidence and the accurate assessment of uncertainty. Med Decis Making 2002;22:359-71. [PubMed]
3. Claxton K, Sculpher MJ, Drummond M. A rational framework for decision making by the National Institute for Clinical Excellence (NICE). Lancet 2002;360:711-5.
4. Colbourn T, Asseburg C, Bojke L, Philips Z, Welton NJ, Claxton K, et al. Preventive strategies for group B streptococcal and other bacterial infections in early infancy: cost effectiveness and value of information analyses. BMJ 2007 doi: 10.1136/bmj.39325.681806.AD
5. Colbourn TE, Asseburg C, Bojke L, Philips Z, Claxton K, Ades AE, et al. Prenatal screening and treatment strategies to prevent group B streptococcal and other bacterial infections in early infancy: cost effectiveness and expected value of information analysis. Health Technol Assess 2007;11(29).
6. Cooksey D. A review of UK health research funding London: Stationery Office, 2006

Articles from The BMJ are provided here courtesy of BMJ Publishing Group