Economic Aspects of Evaluation - Evaluation Methods in Biomedical Informatics - page 315

cost-effective options first, and strategies or programs added or funded in
order of their cost-effectiveness. In theory, this ensures that the resources
expended are purchasing the most “health” possible. So, when faced with a
series of choices between possible health-improving programs and a limited
budget, economic principle would dictate that you rank the possible options
in order of cost-effectiveness ratio (from the most cost-effective to the least
cost-effective) and purchase programs in decreasing cost-effectiveness
order until the budget limit is reached. In practice, this rank ordering of
options followed by spending the available health care resources is rarely
explicitly done. Often resources directed in one area (e.g., information ser-
vices) cannot be redirected to another area of public service (e.g., social
work). This is true at all levels of decision making, with each level contain-
ing a series of political, social, and organizational barriers to the strict appli-
cation of CEA. However, it remains useful to understand this as the
underlying concept in the intended use of CEA, which is designed to
achieve the highest quantity of the outcome (chosen and valued by the
investigator) for the least cost.
Graphically, the comparisons used in cost-effectiveness analysis are illus-
trated in Figure 11.2, which represents the cost-effectiveness plane. For any
new therapy or choice that is being made, the costs and benefits need to be
compared to the current strategy. For any new option, the new strategy can
be more effective, less effective, or equivalent to the current strategy, and can
be more expensive, less expensive, or equal in costs to the current strategy.
This divides the cost-effectiveness plane into four quadrants that have useful
interpretations. The lower right quadrant would represent strategies that are
both cheaper and more effective than the existing strategy: these programs
should simply be implemented. Similarly, those strategies that fall in the
upper left quadrant are both more expensive and less effective than the
current strategy: these should be avoided. It is only in the two remaining
quadrants (the upper right quadrant, where strategies are more expensive
and more effective , and the lower left, where strategies are less expensive but
not as effective ) that the use of CEA is appropriate. It is in these areas, where
there is a trade-off between costs and benefits, that CEA is most useful.
The Incremental Cost-Effectiveness Ratio
A crucial aspect of a CEA is that something cannot be cost-effective in and
of itself; it is only cost-effective (or not) compared to another alternative.
Consequently, the statement “This clinical information system is cost-
effective” is nonsensical; it must be accompanied by a description of what
the system is being compared to. In fact, CEA is most useful when the
strategies being examined represent a range of possibilities, each with dif-
ferent costs and effectiveness. Typical CEA studies provide the results of
each possible strategy, and compare each to the next least effective or
expensive.