Table 4 Scenario 6: Influence of response shift on inclusion in healthcare package (macro level)

A cost-effectiveness (costs per QALY) study is carried out among patients with Crohn’s disease. Treatment A (standard care) is a colostomy, after which patients need to use stoma bags. Treatment B delays the need for a colostomy for 6 years, has no side-effects, and costs EUR 53,000. The total costs of stoma care for 6 years (group A) are estimated at EUR 7000. Treatment B thus costs EUR 46,000 more than standard care.

Utility is determined from the perspective of the general public. Based on a scenario describing aspects of life with a stoma, the general public estimates life with a stoma at a value of 0.8. [23] Treatment B would increase this utility from 0.8 to 1.0 for 6 years. Treatment B would thus yield 0.2 × 6 = 1.2 QALY, at incremental costs of EUR 46,000.

Thus, treatment B has a cost-effectiveness of EUR 38,000 per QALY, which is acceptable in most Western countries.

However, the health valuations by patients with colostomies are significantly higher, at 0.92, [23] probably partly due to response shift. If the patients’ own utility scores had been used, the incremental utility would only be 0.08 for 6 years, leading to a smaller incremental value of 0.08 × 6 = 0.48 QALYs. Combined with the incremental costs of EUR 46,000, the cost-effectiveness would be EUR 96,000 per QALY, which might not be acceptable in many Western countries. [24]

Thus, using the valuations of the general public, treatment B would be reimbursed, while it would not if patients were asked to value their own health states.

The situation is not problematic. Using utility tariffs derived from the general public instead of the patient group for cost-effectiveness analyses does not conflict with a utilitarian point of view. Utilitarianism does include the option to let society determine the desirability or undesirability of health states. In other words, it may be left to the general public to determine how ‘bad’ it considers certain health states to be, and the amount of money it is willing to spend to improve these health states.

The situation is problematic in the sense that only health benefits that improve functioning should be taken into account in decisions for reimbursement, instead of self-reported HRQoL (including influences of response shift and disposition) or valuations of health states (utility). Thus, using utility tariffs is always in conflict with the theory of fair equality of opportunity. The more utility tariffs (derived from the general public) differ from the actual health states of patients, the more problematic it becomes.

In this case, the patients’ valuations would be higher than the valuations derived from the general public, partly due to response shift. Therefore, using the valuations of the general public is less problematic than using patient valuations. In this particular example, using valuations of the general public leads to reimbursement of treatment B, with six extra years of functioning without having to use stoma bags. Thus, the patients’ range of capabilities and opportunities is optimally protected.