The art of simulation

The increasing number of people with dementia worldwide,1GBD 2016 Dementia CollaboratorsGlobal, regional, and national burden of Alzheimer's disease and other dementias, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016.Lancet Neurol. 2019; 18: 88-106Summary Full Text Full Text PDF PubMed Scopus (1068) Google Scholar with estimated global costs of about 1 trillion US$,2Wimo A Guerchet M Ali GC et al.The worldwide costs of dementia 2015 and comparisons with 2010.Alzheimers Dement. 2017; 13: 1-7Summary Full Text Full Text PDF PubMed Scopus (609) Google Scholar has resulted in a focus on developing disease modifying treatments (although no such drug has yet been approved)3Winblad B Amouyel P Andrieu S et al.Defeating Alzheimer's disease and other dementias: a priority for European science and society.Lancet Neurol. 2016; 15: 455-532Summary Full Text Full Text PDF PubMed Scopus (970) Google Scholar and developing intervention programmes to target modifiable risk factors for prevention of dementia. The potential impact of several modifiable risk factors for dementia4Livingston G Sommerlad A Orgeta V et al.Dementia prevention, intervention, and care.Lancet. 2017; 390: 2673-2734Summary Full Text Full Text PDF PubMed Scopus (3064) Google Scholar and the outcome of the FINGER-trial5Ngandu T Lehtisalo J Solomon A et al.A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial.Lancet. 2015; 385: 2256-2263Summary Full Text Full Text PDF Scopus (1690) Google Scholar have resulted in the prevention of dementia becoming an area of great research interest. However, even if the prevention of dementia is clinically effective, is it cost-effective? Evidence regarding the cost-effectiveness of disease prevention is important to guide policy decisions regarding the implementation of such programmes. Economic analysis can help identify thresholds that define target groups for an intervention that can be expected to generate meaningful health benefits at acceptable costs. A prevention programme might not be cost-effective in an unselected population but could be cost-effective in selected target groups with sufficiently high baseline risk and ability to benefit from the intervention.6Mühlenbruch K Zhuo X Bardenheier B et al.Selecting the optimal risk threshold of diabetes risk scores to identify high-risk individuals for diabetes prevention: a cost-effectiveness analysis.Acta Diabetol. 2020; 57: 447-454Crossref Scopus (8) Google Scholar So far, the question whether dementia prevention is cost-effective has not been supported by empirical evidence for several reasons.7Handels R Wimo A Challenges and recommendations for the health-economic evaluation of primary prevention programmes for dementia.Aging Ment Health. 2019; 23: 53-59Crossref Scopus (10) Google Scholar The first issue is the discussion about causal links between results from observational studies on modifiable lifestyle and risk factors and long-term outcomes on dementia incidence and disability. The second issue is time. Generating empirical evidence is challenging, given the time it takes (between years to decades) to observe whether the manifestation of dementia is affected by the modification of a potential risk factor (and the possible bias in the likely drop-out rate during that period). One solution to the challenge of time is to focus on surrogate outcomes, indicating that a reduction in the risk factor will result in a reduction in incidence. Another solution could be economic simulations. In dementia, health economic models have been frequently used to evaluate pharmacological interventions, but rarely used for prevention.7Handels R Wimo A Challenges and recommendations for the health-economic evaluation of primary prevention programmes for dementia.Aging Ment Health. 2019; 23: 53-59Crossref Scopus (10) Google Scholar Most economic evaluations have an empirical core from randomised clinical trials of intervention effects on clinically relevant outcomes, but this has not been the case so far for prevention of dementia. The Article by Naaheed Mukadam and colleagues8Mukadam N Anderson R Knapp M et al.Effective interventions for potentially modifiable risk factors for late-onset dementia: a costs and cost-effectiveness modelling study.Lancet Healthy Longev. 2020; 1: e13-e20Scopus (16) Google Scholar in this issue of The Lancet Healthy Longevity must be seen in light of this situation. This paper is a valuable contribution to the scarce literature of the health economics of dementia prevention. A strength of the study is that it is based on a systematic review of modifiable risk factors in dementia.4Livingston G Sommerlad A Orgeta V et al.Dementia prevention, intervention, and care.Lancet. 2017; 390: 2673-2734Summary Full Text Full Text PDF PubMed Scopus (3064) Google Scholar In that review, nine risk factors with the potential to reduce dementia cases by 35% were analysed. Four of these nine factors (hypertension, smoking, hearing loss, and diabetes) were assessed in this paper. For hypertension, there is some support for a reduced dementia risk in randomised controlled trials, while for the other three risk factors the risk reduction is mainly based on observational studies. The study found that treatments for stopping smoking and provision of hearing aids reduced cost and that treatment of hypertension was cost-effective by reference to standard UK thresholds. The three interventions when fully implemented would save £1·863 billion annually in England and reduce dementia prevalence by 8·5%. There are different ways to integrate modification of several risk factors to reflect their adjusted effect. The use of weighted population-attributable fractions (used in Mukadem and colleagues' Article8Mukadam N Anderson R Knapp M et al.Effective interventions for potentially modifiable risk factors for late-onset dementia: a costs and cost-effectiveness modelling study.Lancet Healthy Longev. 2020; 1: e13-e20Scopus (16) Google Scholar) is one interesting approach. Another approach is to use a multivariate risk model such as the CAIDE index,9Kivipelto M Ngandu T Laatikainen T Winblad B Soininen H Tuomilehto J Risk score for the prediction of dementia risk in 20 years among middle aged people: a longitudinal, population-based study.Lancet Neurol. 2006; 5: 735-741Summary Full Text Full Text PDF PubMed Scopus (665) Google Scholar which was used in the FINGER project. Estimating the cost-effectiveness of risk factor intervention assumes knowledge about the potential effectiveness of the programme to achieve the intended change in routine care practice. The study by Mukadam and colleagues assumes a maximum effect on the potential excess risk reduction and can thus be seen as an upper bound on this effectiveness. In practice the challenges with implementation of prevention programmes that involve behavioural change are well documented—for example, in the context of reducing cardiovascular risk or the risk of cancer. An interesting question for further study is how specifically targeting the risk of dementia can be successful. Potentially communicating the risk of dementia could be a powerful motivator for behavioural change. Some of the risk factors for dementia considered by Mukadam and colleagues are also important as risk factors for other conditions, such as cardiovascular disease and lung cancer. Future economic evaluations could help illustrate whether including effects on dementia (over and above effects on other conditions) changes the thresholds at which these programmes become cost-effective and can thus be recommended, considering the overall health effects on all conditions. Perhaps these overall effects might be even stronger than the effects on dementia. Modelling is unavoidable when it comes to estimating lifetime outcomes.10Buxton MJ Drummond MF Van Hout BA et al.Modelling in economic evaluation: an unavoidable fact of life.Health Econ. 1997; 6: 217-227Crossref PubMed Scopus (480) Google Scholar The economic evaluation by Mukadam and colleagues is a methodological step forward, but clinicians, decision makers, and health economists are still waiting for better empirical inputs. Such evaluations should ideally be based both on randomised controlled trials with long-term evidence on the level of risk factor control and lifestyle interventions, and simulations of health outcomes and costs that cannot be directly observed in a prospective study trial. AW reports grants from MSD and the European Union, and personal fees from Biogen and WHO. RH reports consulting fees from Piramal, Roche and Eisai; public and private-public grants from international frameworks (H2020, JPND, IMI), national frameworks (Netherlands: ZonMw and Flutemetamol study; SNAC and SveDem), and patient organisations (Alzheimer Netherlands, European Brain Council, Alzheimer Research UK); among which a grant from Neurodegenerative Disease Research (JPND) for the project named MIND-AD aimed to identify (cost-)effective prevention strategies for dementia and Alzheimer's disease, outside of the submitted work. LJ reports personal fees from H Lundbeck A/S. Effective interventions for potentially modifiable risk factors for late-onset dementia: a costs and cost-effectiveness modelling studyThere is a strong case for implementing the three effective interventions on grounds of cost-effectiveness and quality-of-life gains, as well as for improvements in general health. The interventions have the potential to remain cost-saving or cost-effective even with variations in dementia incidence and costs and effectiveness of interventions.