The key evidence for our product is a well-conducted non-inferiority trial. Why do you think this won’t support the reimbursement for our product we are looking for?

When comparing the effectiveness of two treatments (X and Y), a study may aim to show one of three things: X is better than Y (a superiority trial), X and Y are the ‘same’ (an equivalence trial), or X is not worse than Y (a non-inferiority trial).

Non-inferiority is different from equivalence, although these terms are sometimes used interchangeably. In an equivalence trial, the objective is to show that two products are the same or ‘not unacceptably different’ from each other. In a non-inferiority trial, the aim is to show that a new product is not unacceptably worse than an older one. Non-inferiority trials are often used in situations in which it is considered unethical to perform a trial against a placebo.

Showing that X is not unacceptably inferior to Y may offer a useful treatment option if there are non-quantified advantages such as fewer side effects, acceptability to patients, easier implantation procedure, or less need for maintenance.

In this case, the non-inferiority trial was sufficient for regulatory approval and the client was looking for a substantial price premium compared with standard treatment. Unfortunately, non-inferiority trials are not usually sufficient to support a higher cost than standard treatment. The conclusions payers draw from trials are shown in Figure 1 below. These follow from the use of the incremental cost effectiveness ratio (ICER), which compares the additional (incremental) benefit of X vs. Y compared with the additional (incremental) cost of X vs. Y. Generally the bigger the incremental benefit, the bigger the incremental cost can be while still remaining acceptable.

Figure 1

Trial type	Results	Implications for reimbursement
Superiority	X is better than Y	In principle, a premium can be paid for X. The size of the acceptable premium is determined by the added value
Equivalence	X is the same as Y	Normally, X can be reimbursed at the same rate as the standard treatment
Non-inferiority	X is not unacceptably worse than Y	Normally, if reimbursed, the cost of X will be lower than the cost of Y, the size of the negative premium being determined by how much ‘worse’ was defined in the trial design

 

Benefits include all positive non-financial advantages of adopting a treatment. Advantages such as fewer side effects or improved quality of life (perhaps because the device is simpler to use, or more acceptable socially) should, if the trial design is appropriate to the requirements of making a reimbursement case, be quantified and included in incremental benefits.

In general the amount of reimbursement you can aim for is limited by the size of the additional benefits offered by your product. In planning to collect data, you should consider how much the results will support a case for premium pricing, compared with current alternative treatments. Trial designs which are entirely acceptable for regulatory purposes do not necessarily produce results which support the case you will want to make to payers, who do not just want to know whether something is effective and safe, but how much more effective and safe a device is compared with another device (or sometimes a pharmaceutical or another intervention).

On the financial side, any cost-offsets will reduce the incremental cost, which either means that you can propose between value for money at a given price, or increase the price while maintaining an acceptable ICER.

If you are at an early stage in the product lifecycle and setting up a pivotal trial, think of how you will be able to use the results to support a business case to a payer which leads to reimbursement at the level you are aiming for. Payers in both public and private social insurance systems are increasingly looking for value for money, with convincing evidence from good quality trials to support proven quantified benefits which are meaningful to them. By implication, a non-inferiority trial cannot logically support a case for higher reimbursement than an alternative that may even be better than the new product. The best it can do for you, however well it works out, is to accept the product as an alternative option at equal cost for equal effectiveness.

One statistically legitimate strategy is to plan a superiority trial with an alternative prespecified analysis to show non-inferiority. That way, if the results are not as good as you hope, an alternative analysis is available to get you through the regulatory hoop.

If you are already committed to using data from a non-inferiority trial, the most cost effective how to develop implementation support tools: this might be a detailed implementation checklist, if the change is substantial—a switch from secondary to primary care—or clinical support, such as instruments which make procedures faster. Look for opportunities to reduce the cost of the pathway of care, making room for a corresponding increase in the reimbursement your product can command.