Pharmaceutical innovations: Are those insured by statutory health insurance willing to pay for new types of insulin? – Determination of preferences through a market experiment

In 2005, the Institute for Quality and Efficiency in Health Care (IQWiG) recommended excluding certain types of insulin (so-called short-acting insulin analogues) from the statutory health insurance (SHI) benefits catalog. The reason given was that there was no additional benefit for patients compared to treatment with human insulin. This decision led to an intense debate about which pharmaceutical advances should be reimbursed by the statutory health insurance (SHI).

>> From a health economics perspective, the preferences of insured persons and patients are always crucial: Do they value the differences in the product properties of "Insulindetemir" compared to human insulin? This economic benefit can be expressed in a willingness to pay for the product properties and the drug as a whole. The cost-benefit analysis suggests the following criterion: If the willingness to pay of even those who do not depend on insulin exceeds the additional costs of the new drug, there is nothing to prevent it from being reimbursed by the statutory health insurance system. If the willingness to pay of those who do not depend on insulin is lower than the additional costs, the willingness to pay of diabetes patients in the form of a co-payment may still be sufficient to cover the costs.
The average annual cost of the new drug is €226 per diabetic. The additional benefit is not quantified. To measure this benefit for the drug "insulin detemir" compared to human insulin, a market experiment was conducted in 2007. Respondents were asked to repeatedly choose between the status quo (human insulin) and an alternative with varying properties. The properties considered were the following:
• Risk of hypoglycemia: While the number of hypoglycemias with human insulin is estimated at an average of 1 to 2 per week, this risk is on average 30 percent lower with „insulin detemir“ (see e.g. Hermansen 2004, Home 2004 or Russel 2002).
• Weight gain during the first six months of therapy: According to the UK Prospective Diabetes Study Group (UKPDSG 1998), the greatest weight gain occurs at the start of therapy. During the first six months, it averages 2.5 kg for human insulin. This can generally be avoided with insulin detemir (see, for example, Hermansen 2007, Haak 2003, or Russel 2004).
• Flexibility in injection timing: For human insulin, 10 p.m. has become established for evening injections to meet insulin needs in the early morning hours. Insulin detemir has a later peak effect (see, for example, Kurtzhals 2007). This allows patients to choose their injection time more flexibly and inject the insulin before 10 p.m. in the evening.
• Preparation of the preparation: When using long-acting human insulin, the preparation must be swirled thoroughly before injection to ensure even distribution of the active ingredient throughout the solution. "Insulin detemir" requires no preparation; it can be injected immediately.
• Financing through co-payment: While no co-payment is required from patients for human insulin, co-payments of 50, 150 and 300 euros per year are part of the alternative.
• Financing through increased statutory health insurance contributions: In the status quo, the (employee) contribution to statutory health insurance remains unchanged; in the alternative, there are surcharges of 0.5, 1 and 2 percent.
The properties of the new drug do not always correspond to those of insulin detemir. By varying the characteristics (which do not necessarily have to be realistic, but are intended to encourage respondents to switch between the status quo and the alternative), it is possible to identify properties that provide benefits to respondents. Since the "price" (co-payment, statutory health insurance contribution) of the new drug also changes each time, willingness-to-pay in euros and cents can be determined for the product features and for the drug as a whole.
Figure 1 shows an example of a decision that had to be made eight times by 1,100 statutory health insurance (SHI) insured persons. 500 of them suffered from diabetes (200 type 1, 150 type 2 treated with insulin, and 150 type 2 not dependent on insulin).
Results of the market experiment:
Financing through co-payment

Table 1 shows the results of the market experiment with patient-financed co-payments (see also Figure 2). The willingness to pay for the advantages of insulin detemir compared to human insulin is given in euros per year. These are a 30 percent reduction in the risk of hypoglycemia, weight stability, increased flexibility regarding the timing of the evening injection, and the ability to inject insulin without preparation.
All willingness-to-pay values are positive and significantly different from zero, with a few exceptions (indicated by open bars in Figure 2). Table 1 distinguishes four groups: non-diabetics, type 1 diabetics, insulin-naive type 2 diabetics, and insulin-treated type 2 diabetics.1 With the exception of insulin-treated type 2 diabetics, the most valued benefit of insulin detemir is the prevention of weight gain.
Significant differences in preference emerge between the four groups. For example, the willingness to pay for a 30 percent reduction in the risk of hypoglycemia ranges between €28 (type 1 diabetics) and €44 (type 2 diabetics without insulin) per year, if the treatment is financed with a co-payment. Non-diabetics and both groups of type 2 diabetics are willing to pay significantly more than type 1 patients to prevent weight gain. This makes intuitive sense. According to Russell-Jones 2007, 80 percent of type 2 diabetics are overweight, and according to Häussler et al. 2005, their BMI is significantly higher than the German average. In contrast, type 1 diabetics are more often underweight and would therefore likely be more likely to welcome weight gain.
The willingness to pay for more flexibility in injection timing is not significantly different from zero for type 1 patients. They have apparently learned to plan their daily routine and see no additional benefit in this product feature. The willingness to pay for the other groups ranges between €25 and €50 per year. Finally, regarding reprocessing with swirling, non-insulin-dependent diabetics do not show a significant willingness to pay for this, in contrast to insulin-dependent type 2 patients.
In addition, Table 1 also shows a "tendency toward alternatives." This reflects those unconsidered factors that contribute to the decision for or against a drug. The value shown indicates how much respondents are willing to pay to move away from the status quo (human insulin). Directly affected type 1 and type 2 patients are willing to pay €245 and €318 per year for "Insulin Detemir" out of their own pocket, respectively. Of particular interest is the willingness to pay of those not directly affected. The values here are €790 and €453 per year, respectively. Apparently, even non-diabetics see an advantage in the product properties of "Insulin Detemir" in the event that they themselves develop diabetes at some point in their lives.
However, caution is advised when interpreting these estimates; the €790 per year for non-diabetics and €453 for type 2 patients who do not rely on insulin would likely not be confirmed in a real-life purchasing decision. This overestimation is due to the tendency toward alternatives (€597 and €287, respectively), which account for over 75 and 63 percent of the total, respectively.

Results of the market experiment: Statutory health insurance financing

Table 2 (see also Figure 3) shows the results of the market experiment with financing via the statutory health insurance contribution. In this case, the drug must be paid for in both illness and health. The willingness to pay of the non-insulin-treated respondents was therefore weighted by their own subjectively assessed probability of developing diabetes later in life.2
Figure 3 shows that (with the exception of type 1 diabetics), avoiding weight gain elicits the greatest willingness to pay. This, as well as use without reprocessing, are particularly highly valued by type 2 insulin-treated patients. However, the standard errors of these estimates are so large that they are not significantly different from zero (indicated by open bars). One reason for this variation (not observed for this group in Figure 2) could be that some affected individuals were perhaps unsure whether they would actually have access to a new diabetes medication if it were financed by the statutory health insurance. The three remaining groups exhibit systematic preference differences, with type 1 diabetics placing the highest value on the four product attributes. Their willingness to pay for a 30 percent reduction in the risk of hypoglycemia reaches 34 euros per year, while that of non-diabetics is only 11 euros per year.
Surprisingly, they also have the highest willingness to pay for the prevention of weight gain via their statutory health insurance contributions, even though they are more likely to be underweight than overweight. Those not directly affected show willingness to pay €7 and €19 per year, respectively, for greater flexibility in the timing of injections. Similar amounts apply to use without preparation. Finally, non-diabetics also have a strong preference for alternatives, which (even without weighting by subjective probability of developing the disease) at €176 even exceeds the €130 of type 1 diabetics.
The willingness to pay for the drug "insulin detemir" as a whole ranges from €182 (type 2 without insulin) to €299 per year (type 1); the value of €1,089 for the "type 2 with insulin" group cannot be clearly distinguished from zero. In general, the estimated willingness to pay is somewhat high and may not always be reflected in actual purchasing decisions.
Comparison of the two forms of financing
It is expected that the estimated willingness to pay of the four groups will depend on the financing method. Those not affected (non-diabetics and insulin-naive type 2 diabetics) are likely to prefer the "patient-payment" option. For those affected (type 1 diabetics and insulin-dependent type 2 diabetics), the opposite should be true: They should prefer financing through increased contributions to statutory health insurance, which spreads the burden among all members of the community.
These expectations are confirmed by the overall values presented in Tables 1 and 2. When financed through co-payments, the willingness to pay of non-diabetics is €790, while that of insulin-naive type 2 diabetics (who must expect to rely on insulin in the future) is €453 per year. When financed through increased statutory health insurance contributions, these values drop to €232 and €182, respectively. For type 1 and type 2 diabetics with insulin treatment, however, the willingness to pay is low at €248 and €318 per year, respectively, if they are expected to pay for the medication themselves. It increases to €299 and (insignificantly) €1,089, respectively, if financed through increased contributions. However, the crucial finding is that even non-diabetics are willing to pay a higher statutory health insurance contribution for the drug "insulin detemir," which covers the additional costs of approximately €226 per year. This is the case due to the estimated amount of 232 euros.
The reasons for this can be varied. On the one hand, it's conceivable that non-diabetics are also willing to sacrifice part of their income so that diabetics suffer less from their disease and can enjoy a somewhat higher quality of life. On the other hand, non-diabetics also have a risk of developing diabetes at some point in their lives. To ensure the best possible treatment in this case, they are willing to pay something for it now.
Conclusion
From a health economics perspective, the preferences of insured persons and patients are crucial in deciding which pharmaceutical product innovations should be reimbursed by statutory health insurance. In the case of the long-acting insulin analogue "insulin detemir," the question arises: Do insured persons and patients view the differences in the product properties of "insulin detemir" compared to human insulin positively?
This economic benefit can be expressed in a willingness to pay for the product's features and the drug as a whole. The cost-benefit analysis suggests the following criterion: If the willingness to pay, even among those who do not rely on insulin, exceeds the additional costs of the new drug, there is nothing to prevent it from being reimbursed by statutory health insurance. Otherwise, the patients' willingness to pay themselves might be high enough to cover the additional costs.
In a so-called market experiment, preferences were determined by surveying 1,100 statutory health insurance (GKV) patients (including 500 diabetics) in 2007. The results show that not only insulin-treated diabetics, but also those who only require an oral medication, as well as non-diabetics, derive additional benefit from insulin detemir compared to human insulin. While non-diabetics in particular prefer financing through co-payments, their willingness to pay a higher GKV contribution is so high that it exceeds the additional costs compared to treatment with human insulin. Therefore, based on this cost-benefit analysis, there is nothing to prevent the GKV from financing the drug insulin detemir.