Why Randomized Controlled Trials are best way for Economic Evaluations in Healthcare and Elsewhere

By: Bikal Dhungel 

Randomized Controlled Trials ( RCT )is  one of the simplest but most powerful tools of research. In essence, the randomized controlled trial is a study in which people are allocated at random to receive one of several clinical interventions (Jadad, 1998). RCTs are used to examine the effect of interventions on particular outcome such as death or the recurrence of disease. RCTs give structured data which could be drilled efficiently for research purposes.

Cost effectiveness analysis is an important tool in economic evaluation that helps in decision making for example which care should be provided under what condition. It is one among four types of Economic Evaluation (Others being CUA, CMA, CBA). The National Institute for Health and Clinical Excellence (NICE) defines it as “an economic study design in which consequences of different interventions are measured using a single outcome, usually in natural units for example life-years gained, deaths avoided, cases detected etc. Alternative interventions are then compared in terms of cost per unit of effectiveness.”

Cost Effectiveness Analysis for Economic Evaluation can be done by using two methods:

.     i)  Randomised Controlled Trial based: This is done to obtain primary data and to gain information in the first hand. However it also has limitations which we will discuss in this paper briefly. 


.     ii)  Decision Analytic Model based: Decision Analytic modelling are undertaken for the purpose of economic evaluation of health technologies. This involves the application of mathematical techniques to synthesize available information about healthcare processes and their implications. Hence, it provides an explicit two way bridge between primary data and the decisions they inform. These models embrace a variety of techniques, including decision-tree modelling1. 
The aim of this model is to gain understanding about the connections between incremental costs and its given consequences. A good decision analytic model should contain the following characteristics: 


            -  Internal Consistency 


            -  Transparency 


            -  Reproducibility 


            -  Interpretability 


            -  Exploration of Uncertainty etc 
RCT based economic evaluation can be the best method to calculate cost-effective analysis because of the following points: 


.     i)  Early/First chance to obtain use data: Economic evaluations conducted alongside RCT provide an early opportunity to produce reliable estimates of cost effectiveness at low marginal costs. Access to individual patient data also permits a wide range of statistical and econometric techniques. For example, to examine the relation between events of interest and health related quality of life or to explore subgroup differences2. Even though RCT based evaluation have limitations, they are likely to continue to have an important role in producing reliable estimates of cost effectiveness. 


.     ii)  Can collect data on effectiveness and cost for the same group of individuals. 


.     iii)  Opportunity to make decisions based on patient level data instead of aggregated data: Patient level data helps to improve decision making process because they are more focused. Patient level data is also more updated or at least permit updating. 
They have certain advantages over aggregated data. Some of them are3:

                        -  Address questions not addressed in original publication 


                        -  Permits data checking 


                        -  Permits checking of analyses 


                        -  Permits ready use of time-to-event data for estimating survival 


                        -  Ability to address long-term outcomes 


                        -  Facilitates exploration of heterogeneity at the patient level and subgroup analyses 
of patient level data. 
Having patient specific data on both costs and outcomes is potentially attractive for analysis ( Drummond et al ) 


.     iv)  Have high degree of internal validity: There is a wide recognition that pragmatic randomized trials are the best vehicle for economic evaluation. This is because trials 


provide the best chance of ensuring internal validity, not least through the rigorous prospective collection of patient-specific data. Similarly, it was found that the development of standard operating procedures for economic evaluations alongside pragmatic trials managed by a registered trials unit improve trial conduct and hence the validity and generalizability of results4. Internal validity is one of the most important parts of a research. Without internal validity, there can be no cause and effect, which means, the result of the research cannot be generalized. If it cannot be generalized, the whole research might be useless in some case. However, we should also keep in mind that even though the results cannot be generalized, the research will not be completely a waste.

.     v)  Collaboration between Health Economists and Trial professionals: It also requires a close collaboration of health economists and trial professionals but it will result in a quality outcome. However, in case of conflict or dissatisfaction between them, it can lead to the accumulation of flawed data or can even undermine the research study. 


.     vi)  Modest Cost: Collecting clinical data are costly. Especially it can incur large amount of fixed costs. ( Drummond et al ) claims that given high fixed costs, the marginal cost of collecting economic data may be modest. 


.     vii)  Advantages on Statistical Analysis of Patient-Level Data: Increased use of trial-based economic evaluation has led to a greater focus on issues of statistical analysis ( Drummond et al). This is particularly the case because such research can provide data about the use of resource, treatment effect and sub-samples of patients. Sample data in other hand will help to cope with uncertainties around the estimations of total costs and cost-effectiveness in patient population. In such case, Drummond et al claims further that the presence of sampled patience-level data on costs and effects offers the opportunity to move from ‘deterministic’ to ‘stochastic’ analysis for trial- based economic evaluation. 


Many researchers also claim that patient level pragmatic trials provide an opportunity to execute the economic evaluation under real world conditions. However, many aspects should be considered before doing a RCT based economic evaluation. The instruments or procedures that are going to be used should be tested for its easiness, efficiency or clarity.

In terms of use in longer term data-drilling, RCT based CEA are important because CEA has a capability to accommodate various types of models like survival analysis models which can often

be used to estimate various health related information for example life expectancy even without an intervention. The given ability of RCT based CEA to construct such model has caused increase in its use.

Limitations :there is no right answer why RCT based CEA is the best way of economic evaluation. The methods should be suitable for the given research and it can also be other methods like decision modelling. Especially if the RCT was done in multiple location or even multi-nationally, it should be checked in advance if the data can be used for economic evaluation. Here, other factors like purchasing power parity (PPP), currency etc should also be calculated in a common unit. For the same reason, RCT based economic evaluation may be too costly and may present some methodological challenges. Either the RCTs provide all the information required to do economic evaluation is also another issue. Since Healthcare decision making is a sensitive issue, wrong or biased data can lead to negative consequences.

While talking about limitations of RCT based CEA, it is important to mention why and in which case decision analytic models can be better. Drummond et al claim, especially when there is a specific resource-allocation decision to be taken, decision modelling instead of RCT is useful and it offers a means of synthesizing available evidence from a range of sources than relying on a single study. Additionally, it provides a framework within which the limitations of randomized trials as a vehicle for economic evaluation can be addressed. It also helps to identify optimal interventions under uncertainty and contributes to the process of setting research priorities. ( Drummond et al )

Concluding everything, RCT the economic evaluation. Because they could be integrated so well, they are also called ‘The piggyback economic evaluation’. Since there is a possibility to have an access on individual patients level data, economic evaluations based on RCTs are gaining importance and are an important tool in research methods. They have been used to inform the healthcare decision makers and government institutions. (Glick et al) also argues that trial-based cost-effective analyses should adopt an intention to treat design. Our aim of this essay was to understand in detail about possible advantages of RCT based CEA and we have discussed several factors. Addition to that, it should also include issues of sampling uncertainty as well as the tests

of homogeneity of economic outcome. The whole aim of economic evaluation is to inform the policy makers, so the relevance of RCT based CEA should be dealt in in detail and we should address the following questions like

            -  Why are they relevant and what are the important attributes of it? And what are the attributed that doesn’t make them relevant? 


            -  Can single-trial based economic evaluation be relevant? 


            -  Can they be sufficient to make a decision? 
Hence, considering their advantages and also limitations, RCT based EE can be the best method. 


This article primary focused on health sector but the RCTs are gaining popularity in other sectors of Economics as well especially Development Economics. The Abdul Jameel Latib Poverty Lab of MIT led by Esther Duflo and Abhijeet Banerjee runs controlled RCTs in Southasia trying several aspects of poverty and development. The whole book “Poor Economics” includes findings of such RCTs. The book presents a good picture of why theory does not work, under which conditions the poverty reduction policies will work. This is a good start of looking the different aspect of development policy.

References Used: 

Reference & Bibliography

1. 1-  Decision Analytic Modelling in the Economic Evaluation of Health Technologies, A Consensus Statement http://www.shef.ac.uk/polopoly_fs/1.43761!/file/MCCABE.PDF ( Retrieved 2nd March 2014 )
   
2. 2-  S Petrou, A Gray, Economic Evaluation Alongside RCT: Design Conduct, Analysis and Reporting. BMJ 2011;342:d1548 http://www.bmj.com/content/342/bmj.d1548#ref-53 ( Retrieved 9th March 2014 )
   
3. 3-  G Laymann, N Kuderer, The strengths and limitations of meta-analyses based on aggregate data. BMC Medical Research Methodology 2005, 5:14 http://www.biomedcentral.com/1471-2288/5/14/ ( Retrieved 9th March 2014 )
   
4. 4-  Edwards et al, Economic Evaluation Alongside pragmatic randomised trials: developing a standard operating procedures for clinical trials units. US National Institutes of Health. V 9 2008. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2615739/ ( Retrieved: 9th March 2014 )
   

Briggs A, Sculpher M and Claxton K. Decision Modelling for Health Economic Evaluation. Handbooks in Health Economic Evaluation, Gray A & Briggs A, editors. Vol.1. Oxford University Press. Oxford. 2006.

Drummond et al, “Methods for the Economic Evaluation of Health Care Programmes”, Chapter 8 and 9. Oxford University Press, 2005 3rd edition. ISBN: 978-019-852945-3

Glick et al, “ Economic Evaluation in Clinical Trials “ , Chapter 2, 8 and 11. Oxford University Press, 2007. ISBN: 978-0-19852997

Jadad AR. Randomised controlled trials: a user's guide. London, England: BMJ Books, 1998
Noble SM, Hollingworth W and Tilling A. Missing data in trial-based cost-effectiveness analysis: the
current state of play. Health Econ 2012. 21: 187–200

Petrou S, Gray A. Economic evaluation alongside randomised controlled trials: design, conduct,
analysis, and reporting. BMJ 2011; 342 doi: http://dx.doi.org/10.1136/bmj.d1548

Ramsey S, Willke R, Briggs A. Good Research Practices for Cost-Effectiveness Analysis Alongside Clinical Trials: The ISPOR RCT-CEA Task Force Report. Value in Health 2005: Vol.8, N.5, 521– 533.