the current reimbursement systems for diagnostic tests (whether single or compan-
ion tests) are cost based rather than value based.
The (social) value of health achieved from a single health intervention can be
expressed as the difference between the discounted future expected value of health
gains created in the affected population, measured for each time period using an
aggregate population-level measure of health (i.e., quality-adjusted life year or
similar) and the total (social) costs of resources allocated to that health care
intervention. Towse and Garrison ( 2013 ) argue that prices that reflect this social
value would send positive signals for investment in personalized medicine and
investment in generating evidence to demonstrate its value and ultimately enable
value-for-money analysis (if so required and desired).^47 Currently, however, the
reimbursement systems for diagnostic tests are mainly cost based rather than value
based, i.e., reimbursement for companion diagnostics is usually based on the
expected cost of the laboratory process rather than the value that the test creates
for patients or the health care system. Garrison and Towse ( 2014 ) point out that the
implications of cost-based rather than value-based assessment of tests is that no
financial incentive is given for the costs of research or generating evidence in
support of test’s clinical utility. Paradoxically, this is precisely the type of infor-
mation that payers need in order to judge whether the test will generate value when
administered in patient populations. Better-adjusted pricing strategies could
involve increasing price flexibility for tests, until the evidence of their value is
developed for different groups of patients using large studies over time.^48
The value of health generated from pharmaceuticals is, in the realm of economic
evaluations, conventionally expressed in terms of quality-adjusted life years. How-
ever, the value of a complex technology such as personalized medicine products
may be difficult to describe using this metric. Generally speaking, QALYs are not
sensitive enough to capture all health benefits that accrue to patients and may be too
narrow in scope to capture all that patients’value about their health and the
processes within health care. The same applies to the use of the EQ-5D instrument
in this context. The issue of a too narrow focus of the traditional methods may be
especially important for companion diagnostics, for several reasons. Mainly, the
companion diagnostics do not produce direct outcomes in terms of health but
instead alter treatment strategies and generate only indirect effects on patient’s
health. Such benefits may be difficult to model and difficult to express in terms of
QALYs, while double counting may in this context pose a serious problem. Other
stated preference methods provide broader measures of value that can also be used
within the economic evaluation framework and that may account for the direct
benefits produced by companion diagnostics more adequately, such as willingness-
to-accept and willingness-to-pay methods or the well-being method. The well-
being valuation method and contingent valuation can be used to derive a monetary
(^47) For further discussion on the topic of value-based pricing for molecular diagnostics, also see
Garau et al. ( 2013 ), pp. 61–72.
(^48) Garrison and Towse ( 2014 ), pp. 484–490.
128 A. Bobinac and M. Vehovec