Sharing the Burden: Could risk-sharing change the way we pay for healthcare?

SHARING THE BURDEN Could risk-sharing change the way we pay for healthcare? .
Table of contents

Journal Features
Risk-sharing agreements, present and future
Risk-sharing agreements, present and future
The economics and evidence on how cost- and risk-shifting may lower coverage and care

The need for an infrastructure to support information sharing is undisputed and urgent. The future of therapeutic innovation and increased pressure on health budgets will require alternative, more flexible models, personalized reimbursement models that allow alignment of medicines prices with the value they deliver in treating the several diseases.

Several mechanisms have been used, such as the introduction of co-payments for pharmaceutical dispensing, control of profit margins with medical and pharmaceutical products, price referral and performance of health technology assessments HTAs. At the same time, policymakers have been trying to find new ways to set the pricing and reimbursement of innovative medicines [ 1 , 2 ]. One of the instruments, whose use has been growing, is the risk-sharing agreements RSAs [ 2 , 3 ] between pharmaceutical companies and payers, aimed at guaranteeing access to innovation, but promoting the sustainability of the National Health Service NHS.

RSAs allow a mitigation of uncertainty about clinical outcomes and cost effectiveness of medicines by allowing conditional reimbursement of the drug, which is dependent on collecting additional evidence for final decision making, b basing reimbursement or payment on outcomes of the drug in the context of clinical practice, c restricting the use of the medicinal product to a sub-population of patients through selection or eligibility criteria where they are most effective and d managing the budgetary impact [ 4 ].

These agreements also allow the price of medicinal products to be aligned with the benefit they provide in a given therapeutic indication or in a particular combination of medicinal products, since different agreements may be implemented depending on the indication or association of medicinal products concerned. The aim of this study is to characterise the existing types of RSAs, identifying their advantages and disadvantages, characterise the existing experience in Portugal and also to sensitise the scientific community to the need to define new pricing modalities based on the benefit the medicinal product delivers, namely in the field of oncology, and the indispensability of creating registries that capture the data required for the implementation of RSAs.

The main focus was on the European context, but documentation has been reviewed with examples related to the United States, Australia and Canada. Only articles whose content was potentially relevant to the writing of this article were considered. A qualitative review of the articles was not carried out.

RSAs are divided into a financial agreements, where cost containment is defined merely on the basis of the price of the medicinal product or the cost of the treatment and b agreements based on clinical results, i. This type of RSA provides a great opportunity for the collection of data from real clinical practice real-world data [ 3 , 9 ]. Figure 1 outlines the various types of existing agreements [ 2 , 5 , 7 ].

There are a number of reasons that justify and encourage the implementation of RSAs, namely improving the sustainability of the health system without impeding access to medicinal products for treatment in therapeutic areas with therapeutic gaps, such as oncology [ 3 , 7 ].

However, like any other cost containment measure, there are also drawbacks. Table 1 summarises the advantages and disadvantages of RSAs [ 8 — 12 ]. Taxonomy for risk-sharing agreements. Advantages and disadvantages of risk-sharing agreements.

Journal Features

It is noted that countries have adopted different types of agreements according to their purpose and the type of health system [ 1 , 3 , 5 ]. As in many cost containment measures, theory is more attractive than practice, with great variability of outcomes between and within the various health systems. In addition, the information on their objectives, methodologies and monitoring systems is still scarce, as is information on the real impact of RSAs, both socially and economically [ 3 , 7 ].

However, according to a study carried out in , one of the few that evaluates the benefit of RSAs, the introduction of these agreements seems to have contributed substantially to an improvement in the access of cancer medicines to Italian patients. In fact, it was verified that the median time for authorisation of oncological drugs with an RSA was 84 days, while for authorisations without risk sharing the median time was days [ 1 , 3 , 9 ].

In addition, health professionals appear to be satisfied with the results of these agreements. Although the implementation of RSAs in various countries is still relatively recent and information is limited, their experience could undoubtedly contribute to the success of future RSAs. Figure 2 summarises some of these recommendations [ 2 , 4 , 5 , 11 ]. It is common for the same health technology to provide clinical benefits in multiple therapeutic indications and in different sub-groups of patients, namely, cancer drugs approved for various types of cancer and in several lines of treatment.

Ultimately, this practice could lead to the fact that some medicines may never be developed for a given indication, which could constitute a major technological breakthrough, simply because this would lead to a price lower than that set for another indication [ 14 , 15 ]. The drug may also be used off-label at a much higher price than due. The development of policies that allow the definition of price per indication PPI can contribute to a better alignment of the reimbursement with the value of the medicinal product [ 14 , 15 ] and the application of RSAs by indication may undoubtedly be an alternative as a way of implementing a PPI.

Three modalities to achieve the expected results of PPI are highlighted [ 15 ]:. Currently, the use and interest in PPI in Europe is still limited [ 14 ], unlike the USA where several discussions about this type of price definition are ongoing, to some extent limited by legal and regulatory barriers [ 14 , 16 ]. Nevertheless, countries such as France and Germany tend to use model b.

Others, such as Italy, have been using RSA model c , especially for cancer drugs, where these agreements allow the use of net prices, that is, prices actually paid, which vary between indications for the same drug under the RSAs. As in Italy, the United Kingdom and Sweden also showed interest in being able to follow this path [ 14 , 16 , 17 ]. Some operational challenges of this model are related to the availability of data and the capacity of managing agreements that involve different net selling prices per indication, requiring monitoring of the volume of use by indication and financial reconciliation ex post to ensure that the correct funds flow between the parties, both nationally and sub-nationally [ 14 ].

Risk-sharing agreements, present and future

At the same time, it will have to be possible to increase the price of a medicinal product if a higher value indication comes after one with a lower value and cost [ 3 ]. There are more and more drug combinations being tested and approved due to the increasing knowledge about the course of pathologies and their mechanisms of origin [ 18 ]. This reality is increasingly relevant in fields such as oncology, human immunodeficiency virus HIV , hepatitis C and diabetes.

When the various drugs can be combined in a single pharmaceutical form, a tablet for example, the exercise on the calculation of the price of the association appears to be relatively simple since the combination is treated as a single product with a different price. Goldman, Joyce, and Zheng find that cuts in plan generosity can lead to reduced compliance with drug therapies for chronic disease, and Buntin et al. Both Gruber and Hsu et al. Overall, the evidence shows that an optimal cost-sharing design may better serve consumers and the health care system when it takes into account all the considerations raised by different patient populations, therapies, and conditions.

Consumers simply do not have the necessary information or wherewithal to make many health decisions, and various factors may keep prices from accurately signaling quality or effectiveness. Patients in emergency situations are simply not able to assess hospital quality or direct their own treatment regimen. Patients, in both emergency and non-emergency situations, trust medical professionals to offer the best information and care, unlike sellers in the general marketplace. Efficient cost-sharing designs cannot be one-size-fits-all.

A universally applied excise tax on health benefits or reduced premium contributions to Medicare do not create the right incentives for the creation of the most efficient insurance policy; in fact, one might argue that they are blunt instruments that create no incentives except to purchase less expensive policies. In doing so, they shift costs onto workers, seniors, and their families, hitting those requiring high levels of medical care especially hard.

Many health policy experts have claimed that both taxing employer-sponsored health benefits and restructuring Medicare into a premium-support system could be powerful tools in restraining the overall growth of American health care costs without exposing Americans to much greater financial or health risks. Given that rapidly growing health costs exert real strains on both government budgets and family incomes, curbing them seems to be a worthy policy goal.

As it turns out, these policies have less reach in driving significant cost containment than is commonly recognized. Decreasing first-dollar coverage through higher deductibles and the like will probably miss many of the most expensive costs in the health system. As previously noted, the sickest 19 percent of the population in any given year accounts for 80 percent of total health spending. This includes people with chronic conditions, acute care needs, end-of-life care needs, etc. An increase in cost sharing among the big-ticket items such as transplants, major life-saving surgeries, or the management of chronic diseases such as diabetes has not been explicitly suggested and is universally recognized to be bad policy.

Swartz points out that it is often the health care providers and not the patients themselves who are the drivers of high health care spending. To the extent that moral hazard—induced overconsumption of health care is a significant problem, patients already active in the health care system e. At that point, patients exercise little control over the medical care they receive. The corollary is that those less active in the health care system may be more sensitive to prices, meaning they are more likely to forgo expensive care if they believe there is less of an immediate medical need for it.

To the extent that consumers do cut back on care in response to increased cost sharing, we noted before that they may well cut back on health-improving medical spending.

Second Bibliographic Guide to the History of Computing, Computers, and the Information Processing In.
!
.
The Great Netherton Tunnel Adventure!
Brennende Lust (German Edition).
;

But they may even cut back on medical spending that is cost effective in the long run. Proponents of increased cost sharing often implicitly suggest that consumers would only be forced to cut back on luxury items e. But a growing body of research indicates that this is not true; increased cost sharing does indeed often crowd out health-improving and cost-effective medical interventions. McWilliams, Zaslavsky, and Huskamp find that cuts in plan generosity can lead to higher overall medical spending.

In one related study, Goldman, Joyce, and Zheng find that higher cost sharing for pharmaceuticals is associated with an increased use of overall medical services, particularly for patients with greater needs e. Likewise, lower cost sharing is associated with a reduction in overall health spending, particularly for those with chronic diseases. For instance, Chernew et al. Furthermore, Muszbek et al.

Mahoney also finds that lowered cost sharing for diabetes patients reduces health costs per plan. The sum of this important research suggests that increased cost sharing in certain areas e. By not differentiating among medical goods and services based on effectiveness research, increased cost sharing stemming from overly blunt policies such as the excise tax or Medicare vouchers may be an ineffective and potentially harmful tool in making efficient cuts to health care utilization.

A careful examination of the growing value-based insurance design literature may produce a more effective policy response. To the extent that consumers do respond by cutting back on medical care, it becomes clear that any cost containment from these policies is driven by the reduced quantity of medical care consumed, not reduced prices.

That is, if increased cost sharing contains costs to any significant extent, it does so by encouraging people affected by it to buy less health care. To the extent this is true, policies that shift costs onto consumers are not likely to remedy this problem.

Besides problems in conception, specific policies that increase cost sharing also suffer significant problems in the way they are applied. The tax applies to the portion of premiums between the threshold and the total cost of the insurance policy. The premium thresholds are adjusted for workers in high-risk industries and for the age and gender of the workforce. In and beyond, the threshold above which premiums are taxed is indexed to the overall inflation rate plus 1 percentage point, not to the growth of medical costs, which is expected to be higher.

Employee contributions are excluded if the employee works at a firm with a cafeteria plan, a plan that allows employees to choose between taxable and nontaxable fringe benefits e. Subsidizing compensation paid in the form of health insurance encourages employers to offer health insurance, increasing the number of insured workers.

Nevertheless, some argue that limiting this tax exclusion would provide incentives for cost containment because it would make consumers more price sensitive, thereby leading to reduced health expenditures, and it would raise tax revenue that could be used in part to pay for coverage expansions.

In response to the excise tax, a Mercer survey finds that nearly two-thirds of employers plan to cut health benefits to avoid the tax and a full 7 percent would eliminate their health plan altogether. The Joint Committee on Taxation revenue estimates assume that only a small share of revenue would actually come directly from the excise tax as opposed to the large share of revenue from taxed wages , implying that employers and employees alike will shy away from the more expensive plans. Among workers at firms that drop insurance coverage altogether, some workers will become eligible for subsidized coverage in the state health exchanges established by the ACA.

However, expensive plans are not necessarily more generous plans. Many health plans are expensive because the employee population is older or sicker than average, and not because they provide more generous coverage. Small firms and firms with older workforces tend to have less bargaining power with insurance companies and face higher administrative costs.

All else equal, this leads to higher prices for insurance coverage, which may be no more comprehensive than lower-priced coverage for larger firms or those with younger workers. Using actuarial value as a proxy for plan generosity, Gabel et al. Other factors include type of plan e. Even after including these factors, much of the variation in premiums is left unexplained by plan features.

This reinforces that plan prices do not reflect plan value. To some extent, the health reform law acknowledges this reality and specifically raises the threshold of the excise tax for selected small groups of workers explicitly on the grounds that high cost is not synonymous with high value. For instance, it increases the threshold for health plans covering high-risk professions.

Dorn recognizes this problem and proposes an alternative solution to more clearly target the tax to high-value plans by using actuarial value to measure benefit generosity. One solution he outlines taxes plans above the 75th percentile of actuarial value among all enrollees in employer-sponsored insurance plans, and indexes this threshold to overall inflation over time. This policy prescription still ignores the fact that more comprehensive coverage benefits the least healthy the most. For instance, if overall inflation grows at an average rate of 2.

The result is that more and more insurance plans would be subject to the excise tax, leading more employers and workers to demand lower-priced and less comprehensive coverage. Additionally, proponents of taxing employer-sponsored health insurance benefits often note that if it encourages workers to take less compensation in the form of health insurance premiums, then this could raise other forms of compensation, especially cash wages.

Given that in the long run the excise tax will indeed likely lead to non-trivial cuts in employer contributions to insurance premiums, it is certainly possible that cash wages may rise as employer contributions to premiums fall. However, the large majority of these wage increases will simply be absorbed by higher out-of-pocket medical costs incurred with less generous coverage.

Given the large variation in annual health spending i. On average, after-tax, after—health care wages will rise much less than proponents often claim. Even if cash wages rise in response however doubtful in the current economy , these rising wages only come as other forms of compensation are falling. Further, because some compensation that was previously being subsidized through tax policy is now being taxed, the result is an unambiguous cut, not a raise, to total after-tax compensation.

However, McIntyre makes the additional point that, as a share of income, taxing ESI does not unambiguously increase the progressivity of the tax system across every segment of the income distribution. While many low-income households do not enjoy access to ESI, high-income households are required to contribute only a small share of their income to pay the tax because there is a limit to how much any household would want to take income in the form of health insurance.

For instance, compared with the home mortgage interest deduction, the tax advantage given to health insurance is more tightly distributed, as house prices vary more widely than do health premiums. McIntyre calculates the marginal federal income and payroll tax rate on converting a portion of tax-exempt wages in this case, health insurance into taxable wages. One of the instruments, whose use has been growing, is the risk-sharing agreements RSAs [ 2 , 3 ] between pharmaceutical companies and payers, aimed at guaranteeing access to innovation, but promoting the sustainability of the National Health Service NHS.

The main focus was on the European context, but documentation has been reviewed with examples related to the United States, Australia and Canada.

Only articles whose content was potentially relevant to the writing of this article were considered. A qualitative review of the articles was not carried out.

The concept of risk sharing is still relatively recent in the field of health policy; and therefore, remains a subject of dubious interpretation, both in terms of terminology and concept [ 1 , 3 , 5 ]. Several terms, definitions and taxonomies have appeared in the literature to classify these agreements [ 1 — 3 , 5 , 7 , 8 ]. RSAs are divided into a financial agreements, where cost containment is defined merely on the basis of the price of the medicinal product or the cost of the treatment and b agreements based on clinical results, i.

This type of RSA provides a great opportunity for the collection of data from real clinical practice real-world data [ 3 , 9 ]. Figure 1 outlines the various types of existing agreements [ 2 , 5 , 7 ]. There are a number of reasons that justify and encourage the implementation of RSAs, namely improving the sustainability of the health system without impeding access to medicinal products for treatment in therapeutic areas with therapeutic gaps, such as oncology [ 3 , 7 ]. However, like any other cost containment measure, there are also drawbacks.

Risk-sharing agreements, present and future

Table 1 summarises the advantages and disadvantages of RSAs [ 8 — 12 ]. It is noted that countries have adopted different types of agreements according to their purpose and the type of health system [ 1 , 3 , 5 ]. As in many cost containment measures, theory is more attractive than practice, with great variability of outcomes between and within the various health systems.

In addition, the information on their objectives, methodologies and monitoring systems is still scarce, as is information on the real impact of RSAs, both socially and economically [ 3 , 7 ]. However, according to a study carried out in , one of the few that evaluates the benefit of RSAs, the introduction of these agreements seems to have contributed substantially to an improvement in the access of cancer medicines to Italian patients. In fact, it was verified that the median time for authorisation of oncological drugs with an RSA was 84 days, while for authorisations without risk sharing the median time was days [ 1 , 3 , 9 ].

If the benefit of a medicinal product varies according to the indication and the line of treatment, it would be expected that, consequently, its price could also be variable [ 14 , 15 ]. However, in most healthcare systems today, the price of a drug is the same across indications except for different dosages and a single price for all indications may have negative consequences [ 16 ]. Ultimately, this practice could lead to the fact that some medicines may never be developed for a given indication, which could constitute a major technological breakthrough, simply because this would lead to a price lower than that set for another indication [ 14 , 15 ].

The drug may also be used off-label at a much higher price than due.

The economics and evidence on how cost- and risk-shifting may lower coverage and care

The development of policies that allow the definition of price per indication PPI can contribute to a better alignment of the reimbursement with the value of the medicinal product [ 14 , 15 ] and the application of RSAs by indication may undoubtedly be an alternative as a way of implementing a PPI. Three modalities to achieve the expected results of PPI are highlighted [ 15 ]:.

Currently, the use and interest in PPI in Europe is still limited [ 14 ], unlike the USA where several discussions about this type of price definition are ongoing, to some extent limited by legal and regulatory barriers [ 14 , 16 ]. Nevertheless, countries such as France and Germany tend to use model b. Others, such as Italy, have been using RSA model c , especially for cancer drugs, where these agreements allow the use of net prices, that is, prices actually paid, which vary between indications for the same drug under the RSAs.

As in Italy, the United Kingdom and Sweden also showed interest in being able to follow this path [ 14 , 16 , 17 ]. Some operational challenges of this model are related to the availability of data and the capacity of managing agreements that involve different net selling prices per indication, requiring monitoring of the volume of use by indication and financial reconciliation ex post to ensure that the correct funds flow between the parties, both nationally and sub-nationally [ 14 ].

It remains uncertain whether the implementation of PPI agreements will become standard practice, as there is interest from all stakeholders in its potential use, but some scepticism about the NHS ability to obtain good value from its use. At the same time, it will have to be possible to increase the price of a medicinal product if a higher value indication comes after one with a lower value and cost [ 3 ].

There are more and more drug combinations being tested and approved due to the increasing knowledge about the course of pathologies and their mechanisms of origin [ 18 ]. This reality is increasingly relevant in fields such as oncology, human immunodeficiency virus HIV , hepatitis C and diabetes. When the various drugs can be combined in a single pharmaceutical form, a tablet for example, the exercise on the calculation of the price of the association appears to be relatively simple since the combination is treated as a single product with a different price. However, in most cases, it is not possible to have the combination in a single tablet, or ampoule, and pricing becomes substantially more complex.

In addition, in the case of combinations of medicinal products from different companies, the challenges in defining the price and, consequently, in their reimbursement intensify. The fact that the pharmaceutical company owns a single or several components of the combination has implications at stakeholder level and at the variables to be discussed in a negotiation process. The reimbursement procedure for a drug combination will be delineated taking into account multiple factors, such as the number of drugs in the combination considered innovative, the duration of the patent for each of the components or the penetration rate of each drug.

It will also be crucial to deal with other, often interdependent and uncertain variables, such as the results of ongoing clinical trials with the combinations and their alternatives, as well as future pricing strategies for competing products [ 18 , 19 ]. It is also important to assess the impact that the combination will have in terms of duration of treatment compared to the alternatives; and in oncology, this is a central factor, because it is common for a new combination to bring greater survival and consequently a longer period of treatment.

RSAs can be of extreme importance in this context, because they allow setting discounts or limits on the costs of the combination, intra- or inter-pharmaceutical company, enabling access to them. Through these specifics, the system will be able to issue reports in real time, so that monitoring of the use of medicines and the clinical situation of patients can be accessed immediately and continuously.

The first step consists of defining a matrix containing the minimum required parameters to be collected, such as in an observational study or a clinical trial. This selection of variables must inevitably reflect the type of disease and take into account the nature of the various agreements that may be established, as well as the flow of information that potentially will be established between the various entities involved payer, provider and pharmaceutical company. Table 2 presents a suggestion of minimum data to be collected in order to implement an RSA for cancer drugs administered intravenously [ 20 ].

In this case, the fields displayed are grouped into four categories: