In their 2012 HEPL article, ‘The evaluation of lifestyle interventions in the Netherlands’, Rappange and Brouwer problematized the structural underinvestment in preventive lifestyle interventions in the Netherlands (Rappange and Brouwer, Reference Rappange and Brouwer2012). They sought the explanation for this in the decision-making framework used in the Netherlands to delineate the basic benefits package of the National Health Insurance, a framework that seems ill-suited to include investments in lifestyle interventions in the basic benefits package. In their article, Rappange and Brouwer examined how the two most important criteria in the framework (‘necessity’ and ‘cost-effectiveness’) could or should be operationalized and applied in such a way that preventive lifestyle interventions will be included in the basic benefits package.
In this short essay, I will argue that it is true that the decision framework that they focused upon does not facilitate a social optimal choice for investments in the most effective health-promoting interventions, but that this has not so much to do with the decision framework as such, but with a more fundamental problem of social insurance schemes in the welfare state.
Let us start with textbook knowledge. There are two problems with respect to health-related risks that actuarial (private) insurance schemes cannot handle and that therefore require either public regulation or direct state provision. The first problem is that of adverse selection, meaning that it is always in the interest of private insurers to eliminate ‘bad risks’. As premiums are set to reflect expected loss, the strategy of adverse selection may lead to a spiral of escalating premiums, whereby more and more low-risk individuals drop out of the market until the principle of pooling is completely lost (Schut, Reference Schut1995; Barr, Reference Barr1998). The second problem is moral hazard, which is the difficulty that insurance companies would have in gathering the correct information when an involuntary injury, giving right to benefits, has occurred. As high-risk individuals have better information about their risk than the insurance company does, they may exploit this information surplus by buying insurance contracts at below what would be an appropriate price.
Given the fact that some form of compulsory contribution is necessary in order to pool risks adequately, this can only be provided by, or at least with the help of, the authoritarian power resources of the state. Hence, when the risks related to ill health and income loss came to be defined as social risks, the response has been the rise of social (compulsory) insurance schemes (or tax-funded systems) in order to pool these risks adequately (de Swaan, Reference de Swaan1988). Historical contingencies matter in this respect. Government’s responsibility for the health and well-being of its citizens began with improving the conditions of public health in order to prevent the emergence of health problems. This was done either by improving general living conditions, by promoting decent housing or investments in public goods such as sewerage systems, for example, or by informing citizens about the health risks associated with particular lifestyles or living conditions (prevention). As there were many determinants, the area of public health necessarily covered a wide range of issues and demanded all sorts of provisions (Helderman and van der Grinten, Reference Helderman and van der Grinten2007). When the most important determinants of mass epidemics were brought under control, political interest shifted from concern about public health towards concern about medical care and universal risk coverage. The rise of social health insurance schemes went hand in hand with great progress in medical science.
So far, so good. By applying the instruments of social insurance on behalf of an increasing number of citizens, covering an ever-greater variety of risks, society’s ability to treat its members more equally became a realistic aim. Once risks had been pooled, individuals no longer faced these risks alone, but as part of a collective (Baldwin, Reference Baldwin1990: 1–2). These social insurance systems, however, generated their own problems of moral hazard. As the costs of care for any individual were now spread across the large pool of insured individuals and prices were distorted, individual consumers and health care providers possessed an overwhelming tendency to consume (over-consume) health care, causing public spending on health care to spiral to unsustainable levels. The medicalization of health-related problems increased the burden of health care on the government’s health care budget substantially, whereas the marginal value of each dollar, euro or pound sterling spent on health care decreased. I guess that this is the more fundamental problem that Rappange and Brouwer wanted to address. Let me explain how the decision framework that they reflected upon in their article has been adjusted in the past decade in order to deal with these challenges.
The concept of ‘necessity’ already played an important role in the advice of the famous Dekker Committee dating from 1987, without it having been explicitly defined or operationalized. The concept cropped up again in the report of the Choices in health care Committee (the Dunning Committee): the so-called Dunning’s funnel. Dunning’s funnel made use of four criteria for answering the question of whether a medical treatment or drug should be included in the basic benefit package. The first criterion was that of necessity of care; the second criterion involved the question of whether this was effective, efficacious care; the third criterion was about the cost-effectiveness of the health care; and lastly, the fourth criterion was about the question of whether the care should, nevertheless, be at the patient’s own expense and responsibility (Brouwer and Rutten, Reference Brouwer and Rutten2004: 13). The responsibility for package management and for advising the government about what should be included in the basic benefit package and what not, was in the hands of the former Sickness Fund Council and its successor, the Health Insurance Council (CVZ).
For the CVZ, the ‘necessity’ criterion by far was the most complicated (non-discriminating) of the four criteria of Dunning’s funnel (Helderman et al., Reference Helderman, de Kruijf, Verhey and van Thiel2014). The problem, according to the CVZ, is that there are in fact few fields in medical care that are, in their totality ‘unnecessary’. Hence, there is no generic method in which this criterion can be operationalized because this depends on the discrete disorder (of a unique patient) and the discrete indication involved (CVZ, 2001: 15). With respect to the cost-effectiveness criterion, CVZ distinguished between cost-effectiveness upon entry via selective admission, and cost-effectiveness ‘from within’, that is, the way in which the care is given. Because of the lack of opportunities for limiting the basic package, the CVZ opted for the systematic promotion of cost-efficiency. By doing so, the CVZ nudged the criterion of ‘necessity’ in the direction of promoting the appropriate use or consumption of health care and the appropriate prescription of treatment and drugs. This was the only way in which a broadly compiled basic package could be sustainably maintained. Reference points for promoting cost-efficiency were to be found not only at the macro-level, but also at the meso-level via the purchasing policy of insurers and large institutions, and the micro-level via the individual authorization decisions of insurers and care needs assessments in consultation surgeries. In relation to the cost-effectiveness criterion, CVZ believed that the principles of evidence-based medicine (EBM) should guide these decisions (CVZ, 2007). The EBM criterion was internationally recognized and combined scientific knowledge about the efficacy of an intervention or medicine with its use. It was desirable to gather both content-based and practical knowledge of interventions from the various scientific associations in order to promote the quality of CVZ assessments based on the EBM criterion, but also to create a basis of support for using the EBM criterion in practical implementation (CVZ, 2007: 21). The recently established Health Care Institute, the successor of the CVZ, further develops its package and guideline policy along these lines. The criterion of ‘necessity’ is now about the question of whether a treatment or disease justifies a claim on solidarity, that is, whether there is a medical necessity for treatments and whether it is actually necessary to insure against the costs of the intervention. The efficacy criterion relates to the above-mentioned principle of EBM. In addition, CVZ increasingly involves the cost-effectiveness criterion in its assessments. Lastly, CVZ applies the feasibility criterion for mapping out which factors can hamper or promote the successful introduction of a package measure. For example, whether a sufficient basis of social support exists for a measure, what costs will be involved, whether tariffs will have to be established, etc. (CVZ, 2013). All these criteria are equally important, except when a medical intervention has proven to be ineffective, in which case the efficacy criterion is the reason for issuing a negative advice to the Minister. What we can learn from this very brief excursion is that the decision framework that Rappange and Brouwer have analysed has been adjusted in the past decade; however, in another direction, namely, to be able to deal more adequately with over-consumption of medical care by promoting the appropriate use of care and prescription of drugs.
This brings me to the more fundamental problem of preventive lifestyle interventions, which again, is based on textbook knowledge. First, insurance is not an option in the case of interdependent and catastrophic risks in which injuries hit very large parts of the population at the same time; and, second, insurance schemes are less suited to include preventive care investments or public health interventions, notwithstanding the fact these might have a substantial effect on the well-being of citizens (van de Ven and Schut, Reference van de Ven and Schut1994). If insurers do have an interest in reimbursing preventive treatments, it is typically the sort of prevention that mitigates the costs of medical treatments (e.g. fall prevention therapy for Parkinson patients). In other words, the decision framework that Rappange and Brouwer focused upon in their 2012 HEPL paper is about how to make rational (utilitarian) decisions about access to care, covered by social insurance, not for social investments and prevention. Its feasibility for dealing with investments in lifestyle interventions is weak, simply because these type of investments do not fit in with the actuarial foundations of any insurance scheme. This brings us then, finally, to a more fundamental aspect of modern welfare states. That is, we have increasingly come to conceive of all sorts of social problems in terms of actuarial risks, and once these risks were defined as social – actuarial – risks, we created social insurance schemes. This is not unique for health care, on the contrary, it is symptomatic and endemic for the way social solidarity has been taken care of in the welfare state in the past. Of course, this has come to the benefit of all citizens and it is a fundamental asset of universal inclusive welfare states; however, the investment in social insurance systems were developed at the costs of investments in ‘capacitating’ services that enable citizens to acquire the skills needed to make ‘healthy’ choices in various aspects of their lives and in all the domains in which their health is being affected (Sabel, Reference Sabel2012: 43). Moreover, given the increasing differentiation and heterogeneity of the population, these services need to be customized to the individual needs of citizens. The paper by Rappange and Brouwer essentially pointed to the limits of what social insurance systems can achieve when it comes to lifestyle interventions. Instead of medicalizing lifestyle interventions in order to bring them into the basic benefits package of a social health insurance scheme, health and investments in lifestyle interventions might be better served by promoting ‘health in all policies’. However, we need other frameworks to facilitate these type of decisions and another paper to elaborate on this argument.
In their 2012 HEPL article, ‘The evaluation of lifestyle interventions in the Netherlands’, Rappange and Brouwer problematized the structural underinvestment in preventive lifestyle interventions in the Netherlands (Rappange and Brouwer, Reference Rappange and Brouwer2012). They sought the explanation for this in the decision-making framework used in the Netherlands to delineate the basic benefits package of the National Health Insurance, a framework that seems ill-suited to include investments in lifestyle interventions in the basic benefits package. In their article, Rappange and Brouwer examined how the two most important criteria in the framework (‘necessity’ and ‘cost-effectiveness’) could or should be operationalized and applied in such a way that preventive lifestyle interventions will be included in the basic benefits package.
In this short essay, I will argue that it is true that the decision framework that they focused upon does not facilitate a social optimal choice for investments in the most effective health-promoting interventions, but that this has not so much to do with the decision framework as such, but with a more fundamental problem of social insurance schemes in the welfare state.
Let us start with textbook knowledge. There are two problems with respect to health-related risks that actuarial (private) insurance schemes cannot handle and that therefore require either public regulation or direct state provision. The first problem is that of adverse selection, meaning that it is always in the interest of private insurers to eliminate ‘bad risks’. As premiums are set to reflect expected loss, the strategy of adverse selection may lead to a spiral of escalating premiums, whereby more and more low-risk individuals drop out of the market until the principle of pooling is completely lost (Schut, Reference Schut1995; Barr, Reference Barr1998). The second problem is moral hazard, which is the difficulty that insurance companies would have in gathering the correct information when an involuntary injury, giving right to benefits, has occurred. As high-risk individuals have better information about their risk than the insurance company does, they may exploit this information surplus by buying insurance contracts at below what would be an appropriate price.
Given the fact that some form of compulsory contribution is necessary in order to pool risks adequately, this can only be provided by, or at least with the help of, the authoritarian power resources of the state. Hence, when the risks related to ill health and income loss came to be defined as social risks, the response has been the rise of social (compulsory) insurance schemes (or tax-funded systems) in order to pool these risks adequately (de Swaan, Reference de Swaan1988). Historical contingencies matter in this respect. Government’s responsibility for the health and well-being of its citizens began with improving the conditions of public health in order to prevent the emergence of health problems. This was done either by improving general living conditions, by promoting decent housing or investments in public goods such as sewerage systems, for example, or by informing citizens about the health risks associated with particular lifestyles or living conditions (prevention). As there were many determinants, the area of public health necessarily covered a wide range of issues and demanded all sorts of provisions (Helderman and van der Grinten, Reference Helderman and van der Grinten2007). When the most important determinants of mass epidemics were brought under control, political interest shifted from concern about public health towards concern about medical care and universal risk coverage. The rise of social health insurance schemes went hand in hand with great progress in medical science.
So far, so good. By applying the instruments of social insurance on behalf of an increasing number of citizens, covering an ever-greater variety of risks, society’s ability to treat its members more equally became a realistic aim. Once risks had been pooled, individuals no longer faced these risks alone, but as part of a collective (Baldwin, Reference Baldwin1990: 1–2). These social insurance systems, however, generated their own problems of moral hazard. As the costs of care for any individual were now spread across the large pool of insured individuals and prices were distorted, individual consumers and health care providers possessed an overwhelming tendency to consume (over-consume) health care, causing public spending on health care to spiral to unsustainable levels. The medicalization of health-related problems increased the burden of health care on the government’s health care budget substantially, whereas the marginal value of each dollar, euro or pound sterling spent on health care decreased. I guess that this is the more fundamental problem that Rappange and Brouwer wanted to address. Let me explain how the decision framework that they reflected upon in their article has been adjusted in the past decade in order to deal with these challenges.
The concept of ‘necessity’ already played an important role in the advice of the famous Dekker Committee dating from 1987, without it having been explicitly defined or operationalized. The concept cropped up again in the report of the Choices in health care Committee (the Dunning Committee): the so-called Dunning’s funnel. Dunning’s funnel made use of four criteria for answering the question of whether a medical treatment or drug should be included in the basic benefit package. The first criterion was that of necessity of care; the second criterion involved the question of whether this was effective, efficacious care; the third criterion was about the cost-effectiveness of the health care; and lastly, the fourth criterion was about the question of whether the care should, nevertheless, be at the patient’s own expense and responsibility (Brouwer and Rutten, Reference Brouwer and Rutten2004: 13). The responsibility for package management and for advising the government about what should be included in the basic benefit package and what not, was in the hands of the former Sickness Fund Council and its successor, the Health Insurance Council (CVZ).
For the CVZ, the ‘necessity’ criterion by far was the most complicated (non-discriminating) of the four criteria of Dunning’s funnel (Helderman et al., Reference Helderman, de Kruijf, Verhey and van Thiel2014). The problem, according to the CVZ, is that there are in fact few fields in medical care that are, in their totality ‘unnecessary’. Hence, there is no generic method in which this criterion can be operationalized because this depends on the discrete disorder (of a unique patient) and the discrete indication involved (CVZ, 2001: 15). With respect to the cost-effectiveness criterion, CVZ distinguished between cost-effectiveness upon entry via selective admission, and cost-effectiveness ‘from within’, that is, the way in which the care is given. Because of the lack of opportunities for limiting the basic package, the CVZ opted for the systematic promotion of cost-efficiency. By doing so, the CVZ nudged the criterion of ‘necessity’ in the direction of promoting the appropriate use or consumption of health care and the appropriate prescription of treatment and drugs. This was the only way in which a broadly compiled basic package could be sustainably maintained. Reference points for promoting cost-efficiency were to be found not only at the macro-level, but also at the meso-level via the purchasing policy of insurers and large institutions, and the micro-level via the individual authorization decisions of insurers and care needs assessments in consultation surgeries. In relation to the cost-effectiveness criterion, CVZ believed that the principles of evidence-based medicine (EBM) should guide these decisions (CVZ, 2007). The EBM criterion was internationally recognized and combined scientific knowledge about the efficacy of an intervention or medicine with its use. It was desirable to gather both content-based and practical knowledge of interventions from the various scientific associations in order to promote the quality of CVZ assessments based on the EBM criterion, but also to create a basis of support for using the EBM criterion in practical implementation (CVZ, 2007: 21). The recently established Health Care Institute, the successor of the CVZ, further develops its package and guideline policy along these lines. The criterion of ‘necessity’ is now about the question of whether a treatment or disease justifies a claim on solidarity, that is, whether there is a medical necessity for treatments and whether it is actually necessary to insure against the costs of the intervention. The efficacy criterion relates to the above-mentioned principle of EBM. In addition, CVZ increasingly involves the cost-effectiveness criterion in its assessments. Lastly, CVZ applies the feasibility criterion for mapping out which factors can hamper or promote the successful introduction of a package measure. For example, whether a sufficient basis of social support exists for a measure, what costs will be involved, whether tariffs will have to be established, etc. (CVZ, 2013). All these criteria are equally important, except when a medical intervention has proven to be ineffective, in which case the efficacy criterion is the reason for issuing a negative advice to the Minister. What we can learn from this very brief excursion is that the decision framework that Rappange and Brouwer have analysed has been adjusted in the past decade; however, in another direction, namely, to be able to deal more adequately with over-consumption of medical care by promoting the appropriate use of care and prescription of drugs.
This brings me to the more fundamental problem of preventive lifestyle interventions, which again, is based on textbook knowledge. First, insurance is not an option in the case of interdependent and catastrophic risks in which injuries hit very large parts of the population at the same time; and, second, insurance schemes are less suited to include preventive care investments or public health interventions, notwithstanding the fact these might have a substantial effect on the well-being of citizens (van de Ven and Schut, Reference van de Ven and Schut1994). If insurers do have an interest in reimbursing preventive treatments, it is typically the sort of prevention that mitigates the costs of medical treatments (e.g. fall prevention therapy for Parkinson patients). In other words, the decision framework that Rappange and Brouwer focused upon in their 2012 HEPL paper is about how to make rational (utilitarian) decisions about access to care, covered by social insurance, not for social investments and prevention. Its feasibility for dealing with investments in lifestyle interventions is weak, simply because these type of investments do not fit in with the actuarial foundations of any insurance scheme. This brings us then, finally, to a more fundamental aspect of modern welfare states. That is, we have increasingly come to conceive of all sorts of social problems in terms of actuarial risks, and once these risks were defined as social – actuarial – risks, we created social insurance schemes. This is not unique for health care, on the contrary, it is symptomatic and endemic for the way social solidarity has been taken care of in the welfare state in the past. Of course, this has come to the benefit of all citizens and it is a fundamental asset of universal inclusive welfare states; however, the investment in social insurance systems were developed at the costs of investments in ‘capacitating’ services that enable citizens to acquire the skills needed to make ‘healthy’ choices in various aspects of their lives and in all the domains in which their health is being affected (Sabel, Reference Sabel2012: 43). Moreover, given the increasing differentiation and heterogeneity of the population, these services need to be customized to the individual needs of citizens. The paper by Rappange and Brouwer essentially pointed to the limits of what social insurance systems can achieve when it comes to lifestyle interventions. Instead of medicalizing lifestyle interventions in order to bring them into the basic benefits package of a social health insurance scheme, health and investments in lifestyle interventions might be better served by promoting ‘health in all policies’. However, we need other frameworks to facilitate these type of decisions and another paper to elaborate on this argument.