Key words
cost-effectiveness - health economics - healthcare policy
Abbreviations
GB-A:
German Federal Joint Committee
ICER:
Incremental cost-effectiveness ratio
IQWiG:
Institute for Quality and Efficiency in Healthcare
MRM:
MR mammography
MWA:
Microwave ablation
NICE:
National Institute for Health and Care Excellence
omCRC:
Oligometastatic colorectal carcinoma
PET/CT:
Positron emission tomography and computed tomography
QALY:
Quality-adjusted life year
QoL:
Quality of life
RFA:
Radiofrequency ablation
SIRT:
Selective internal radiotherapy
WTP:
Willingness-to-pay
Introduction
As with other areas of medicine, radiology is subject to increasing cost and resulting
justification pressure. The decision-making situation for diagnostic or interventional
radiological measures in particular can lead to a strong focus on costs incurred in
the short term. Radiology is an integral part of the clinical value chain. In terms
of economics, short- and long-term effects must be taken into account. Although the
long-term benefits of these measures are in many cases beyond question, it is often
difficult to assess the cost/benefit ratio in the clinical environment as well as
from the perspective of the healthcare system. While studies and literature on the
diagnostic accuracy and efficacy of radiological procedures are frequently available,
clinically-oriented studies on economic aspects are often lacking. Although individual
issues such as lung cancer screening using computed tomography have already been analyzed
with regard to their cost-effectiveness in the long term [1]
[2], in many clinical decision-making situations, there is a lack of radiologically
initiated, systematic evaluations. The aim of this article is therefore to present
the basics of an appropriate cost-effectiveness analysis.
Cost-effectiveness analysis is a method used in health economics to systematically
compare different medical strategies in diagnostics, therapy and prevention. The comparison
is based on the costs associated with each strategy and the related effectiveness.
Various parameters can define effectiveness here; in the specific case of quantifying
medical benefit, the term cost-benefit analysis may also apply in the literature (in
this review article, however, the terms are used as synonyms for the sake of simplicity).
The need for medical cost-effectiveness analysis arises, as in other areas, from scarcity
of resources. The budget of a health insurance program should lead to a high benefit
for the insured (high-value care) [3]. The objective is therefore to reduce therapies without relevant benefit (low-value
care) or to replace them with better procedures. However, medical cost-effectiveness
analysis as a tool is also limited to certain areas of application, particularly by
social and ethical aspects of medical activity.
There is, for example, a legitimate medical and ethical interest in comparing various
strategies for the management of high blood pressure or other common diseases in order
to maximize the benefits for the insured. In contrast, the comparison of therapies
with a preventive and curative approach, for example, is inappropriate. In such cases,
the allocation of resources is subject to multifactorial reasons [4]. This article therefore focuses on typical examples of the application of cost-effectiveness
analyses in the context of diagnostic and therapeutic procedures in radiology.
Methods
Different viewpoints can be chosen for a cost-effectiveness assessment including the
perspective of the healthcare system or society, the provider or carrier, the patient
or the employer. Depending on the perspective, different costs have to be considered,
such as direct costs including the cost of a treatment, personnel or material costs,
indirect costs including transport costs of the patient or costs due to incapacity
to work, as well as intangible costs, which also include non-monetary costs. Often,
the perspective of the healthcare system is chosen to evaluate medical services in
the context of allocation decisions, and only direct costs, i. e., reimbursed services,
are considered.
The current reference standard for quantifying benefits is the quality-adjusted life-year
(QALY) [1]. Here, the lifetime gained is not considered in absolute terms, but multiplied by
the quality-of-life (QoL) factor. QALY is an assessment of both the quality and quantity
of life lived. QoL is primarily assessed using a patient questionnaire. The distribution
of medical resources should thus not be based solely on life-prolonging effects, but
should also necessarily take into account the quality of life during the anticipated
time-frame. Healthcare economic evaluation is therefore based on the concept of incremental
cost-effectiveness ratio (ICER), which is the result of comparing a new method with
the established standard. For the calculation, the additional costs of the method
compared to the standard are related to the additional benefits:
The benefit of a diagnostic or therapeutic method is quantified in terms of quality-adjusted
life years, which is the product of quality of life and length of life. This allows
direct comparison of a wide variety of methods on the basis of a common reference
value.
Using ICER as a measure of cost-effectiveness can support healthcare decision-makers
as a basis for allocation decisions. Thus, a willingness-to-pay threshold can be defined
that ranks medical services in terms of reimbursability. In the UK, a threshold of
£20 000–£30 000 per QALY serves as the basis for decision-making [5]. For the United States, a threshold of $50 000-$200 000 per QALY has been discussed
[6]
[7]. The German Institute for Quality and Efficiency in Health Care (IQWiG) has so far
used indication-specific cost-benefit assessments without an absolute threshold based
on legal principles [8].
Incremental effectiveness and costs are calculated using healthcare economic modeling
and decision analysis. First, a decision tree is constructed that includes the diagnostic
or therapeutic methods to be compared as well as all feasible outcomes. To model the
long-term costs and benefits, a Markov model is constructed that simplistically defines
different health states but realistically represents the real variety of existing
states ([Fig. 1]). A simulated case is in a state of health in each cycle of the model and, if necessary,
changes this state according to predefined probabilities at the beginning of each
new cycle. The respective condition is characterized by a defined quality of life
as well as associated costs. If the duration of a cycle is multiplied by the quality
of life, the resulting benefit results in the form of quality-adjusted life years.
For example, the Markov model can be used to represent the progression of disease
through a disease stage, a recovery stage, to recurrence or death; each of the states
occurs with a given probability and results in ongoing costs, if applicable. The simulation
over a period of time allows determination of the cumulative mean costs and QALYs
for all strategies and calculation of the incremental cost-effectiveness rate. Comprehensive
sensitivity analyses examine the uncertainty of the various variables and their impact
on the model and the resulting ICER.
Fig. 1 General illustration of a Markov model simulating effectiveness and long-term costs.
The individual states are assigned qualities of life and, if applicable, ongoing costs.
In each cycle, patients can change between states according to predefined probabilities.
In a cost-effectiveness plane, several studies/interventions can be compared with
respect to their incremental costs and benefits ([Fig. 2]). If a strategy is cost-saving and generates more benefits than the standard strategy,
it is positioned in the lower right quadrant as the dominant strategy. If a strategy
is more costly than the standard and shows less benefit, the strategy is said to be
dominated (upper left quadrant). The cost-effectiveness ratio can be calculated if
a strategy costs more than the standard and generates more benefits. A straight line
through the zero point with a slope in cost/QALY represents the willingness-to-pay
threshold.
Fig. 2 Cost-effectiveness plane: Incremental benefits and incremental costs compared to
the standard. The shaded area below the willingness-to-pay threshold denotes the area
of cost-effective strategies.
Appropriate quality control recommendations are available for the preparation of cost-effectiveness
analyses [9]
[10] which are summarized in [Table 1].
Table 1
Checklist for cost-effectiveness analysis.
|
item
|
instruction
|
|
Title
|
Running title of the study and identification as cost-effectiveness analysis
|
|
Abstract
|
Structured summary containing objectives, material and methods, results and conclusions
|
|
Introduction
|
|
|
Background
|
Background of the study and contextual transition to the key question of the study
|
|
Key question
|
Aim of the analysis
|
|
Material and methods
|
|
|
Target population
|
Characteristics of target population
|
|
Comparators
|
Description of the compared diagnostic or interventional modalities
|
|
Period of time
|
Time span considering results and patients
|
|
Discount rate
|
Determination of discount rate for costs and results
|
|
Utility
|
Fixation of a health outcome value (QALY)
|
|
Input parameter
|
Determination of all input parameters used for model calculation
|
|
Model choice and description
|
Choice and description of the utilized model f. e. Markov model with its pathways
and state transitions
|
|
Measurement and evaluation of effectivity
|
Comparison of costs and effectiveness and description outcome value (ICER)
|
|
Costs and utilities
|
Determination of costs and utilities utilized in the study
|
|
Sources
|
Description of the sources of the utilized input parameters
|
|
Results
|
|
|
Model case results
|
Results of cost-effectiveness analysis, incremental costs and ICER
|
|
Stability and uncertainty
|
Results of deterministic and probabilistic sensitivity analyses
|
|
Figures
|
Graphics illustrating results of cost-effectiveness analysis and sensitivity analyses
|
|
Discussion
|
|
|
Context references
|
Clinical context of the results
|
|
Relevance of study results
|
Describing the relevance of results in context of health policy and health economics
|
|
Limitations
|
Limitation of the study and discussion of robustness and uncertainty
|
|
Ethical implications
|
Ethical implications of study results
|
|
Disclosure
|
Potential conflict of interest regarding a funding source or other sources of support
|
Cost-effectiveness analysis of diagnostic procedures using the example of MR mammography
Cost-effectiveness analysis of diagnostic procedures using the example of MR mammography
Cost-effectiveness analyses play a particularly important role for imaging techniques
that indisputably offer additional diagnostic benefits, but which are considered to
be more expensive, at least in the short term, compared with established imaging techniques.
Here, it is important to assess how great the additional benefit, the exact diagnostic
and prognostic differences, and accordingly the cut-off value (ICER) are with respect
to the cost-benefit ratio of the two comparative methods.
In current national breast cancer screening, X-ray-based conventional mammography
is used every two years in women between the ages of 50 and 70, regardless of the
individual patient’s breast density [11].
According to the literature, patients with dense breast tissue have an increased risk
of breast cancer, regardless of their genetic predisposition, while it is known that
the sensitivity of mammography in dense glandular tissue is sometimes less than 50 %
[12]. In this case, there may be a reasonable opportunity to involve alternative, more
sensitive procedures and include them in breast cancer screening, thereby increasing
diagnostic efficiency, i. e., cost-effectiveness.
MR mammography (MRM) is a much more sensitive method in this regard, but it also appears
to be more cost-intensive at first. Several multicenter studies have demonstrated
that, in purely diagnostic terms, even a combination of all conventional imaging modalities
does not outperform the diagnostic accuracy of MRM [13]
[14]. This method is therefore already used as standard in high-risk screening. Initial
cost-effectiveness analyses showed indications of cost-effective use with regard to
this application several years ago [15]
[16].
However, because data on MR mammography have been limited to use in the high-risk
segment, only sparse analyses have been available regarding MR mammography for women
at intermediate risk for breast cancer due to their increased breast density.
However, recent studies have shown that the use of MR mammography for screening women
with dense breasts significantly reduced interval cancer rates compared to conventional
imaging options [17]. At the same time, this new data provided the opportunity for initial cost-effectiveness
analyses in this hitherto new segment.
Using these data, decision models for cost-effectiveness analyses can be generated
and evaluated accordingly. [Fig. 3a] shows an example of a possible decision model for breast cancer screening in high-risk
women that allows comparison of multiple strategies. A Markov model, as shown in [Fig. 3b], allows modeling of costs and benefits over time (Table S1). For mammography, ultrasound, the combination of mammography and ultrasound, and
MR mammography, this model yields cumulative costs of $36 202, $36 668, $37 984, and
$39 051 over a 30-year period, and cumulative effects of 19.53, 19.53, 19.55, and
19.59 QALYs, respectively. MR mammography would be a cost-effective strategy at an
ICER of $45 374 per QALY compared with standard mammography.
Fig. 3 Illustration of a diagnostic decision model. a Decision model for screening patients for the presence of breast cancer. b Markov model for estimating long-term costs and long-term effectiveness.
For women at intermediate risk for breast cancer, it has been shown that examination
by MRM can prevent or reduce other costs in the medium and long term due to the often
high breast density, despite significantly higher initial examination costs (operational)
[18]
[19]. This is achieved through the collection of prognostically valuable, therapy-relevant
information. In these analyses, ICER values for MRM compared with mammography were
consistently found to be well below the willingness-to-pay values described for Western
industrialized countries. From this it can be concluded that MRM in these patient
cohorts is definitely a suitable imaging modality from an economic point of view in
addition to the above-mentioned medical arguments.
Cost-effectiveness consideration of interventional radiological treatments using ablation
of hepatic metastases as an illustration
Cost-effectiveness consideration of interventional radiological treatments using ablation
of hepatic metastases as an illustration
In addition to advances in diagnostic imaging, the clinical added value of interventional,
minimally invasive image-guided procedures can increasingly be demonstrated by a large
number of prospective studies. Since both microtherapeutic procedures such as prostate
embolization or selective internal radiotherapy (SIRT), vasodilation procedures, or
CT- and MRI-guided ablative procedures are sometimes associated with substantial initial
costs, it is crucial to also transparently present their economic added value with
respect to the entire treatment process. This will be illustrated using the example
of the application of ablative procedures in oligometastatic tumor disease of the
liver.
Oligometastatic colorectal carcinoma (omCRC) is a very common tumor entity associated
with tumor disease of the liver, characterized by the presence of 3 to 5 liver metastases,
which have spread from a colorectal carcinoma via the portal venous system [20]. Surgical therapy is sometimes viewed as the only curative option for treating omCRC.
Since the hepatic metastases are often too close to vital vessels, and sometimes both
liver lobes are affected, only about 25 % of all patients are ideally suited for an
operation. This makes the interventional radiological options of treatment with respect
to ablation all the more relevant to provide the patient with effective therapy, improved
quality of life, and possibly improved overall survival [21]. According to studies, ablative therapy such as radiofrequency or microwave ablation
in the treatment of non-operable omCRCs supports significantly improved overall survival,
which is why this therapeutic principle is also recommended in the ESMO guidelines
for the treatment of metastatic colorectal carcinoma – also in combination with other
procedures [22]
[23]. Here, particular attention must be paid to ensuring a tumor-free ablation margin
of at least > 5 mm by the interventional radiologist to effectively prevent post-ablation
tumor progression [24]. This treatment strategy therefore is not only within the guidelines, but can also
be recommended from an economic point of view when effective [25]
[26]. [Fig. 4a] provides an example of an appropriate decision model for comparing therapeutic strategies
when treating oligometastatic tumor disease. Associated long-term costs as well as
long-term cost-effectiveness can then be projected using a Markov model as in [Fig. 4b]. The corresponding input parameters for the model must be defined for the model.
Table S2 of the Supplement summarizes examples of input parameters selected from the literature.
Based on these figures, when calculated over the lifetime of patients for resection,
radiofrequency ablation (RFA), and microwave ablation (MWA), respectively, the cumulative
costs are $41 847.96; $36 936.90; and $35 234.26, with an effectiveness of 6.80, 6.30,
and 6.95 QALYs, respectively. Thus, in this case, MWA would be the dominant strategy
because it is associated with overall lower costs and better effectiveness than the
other two strategies. However, this result is only intended to illustrate an example
of the procedure and interpretation of the results of a cost-effectiveness analysis.
In this case, an additional sensitivity analysis is warranted to check the robustness
of the results.
Fig. 4 Illustration of a therapeutic decision model. a Decision modeling for interventional therapy of liver lesions for the treatment of
oligometastatic tumor disease of the liver. b Example of a simple Markov model for modeling patient-specific outcomes. The starting
state of the patients is based on the decision model (e. g., after incomplete resection,
starting in the “active hepatic metastases” state). c Monthly modeling of Markov states after complete microwave ablation.
After initial treatment, regular imaging therapy monitoring is crucial for the further
course of the disease. Here, investigations using 18F-FDG PET/CT can detect both incomplete
ablation and recurrent disease at the ablation margins. The strategy of follow-up
using 18F-FDG PET/CT provides a significant cost reduction compared to CT alone despite
initially higher financial expenditure, as the cost of overlooked disease is significantly
higher. This not only improves overall survival, but also effectively reduces the
general cost of treatment [27].
Healthcare policy aspects
Healthcare policy aspects
Decision-makers in healthcare systems are faced with the challenge of performing cost-effectiveness
analyses requiring consideration of multifarious factors in the overall policy context
[28]. The concept of cost-effectiveness analyses presented by the authors in this review
represents the most widely used methodology in the healthcare system in order to be
able to adequately distribute limited resources that can be made available in the
respective healthcare system within society ([Fig. 5]) [29]. Thus, in the macroeconomic context, any amount made available for the healthcare
system, for example, is no longer available for education. Overall, this harbors potential
for conflict, especially in economies with clearly limited resources [30]
[31]. In the United Kingdom, for example, the National Institute for Health and Care
Excellence (NICE) makes approval of the reimbursability of innovative treatments conditional,
among other things, on the availability of a corresponding cost-effectiveness analysis
taking into account the respective QALYs. Likewise, in Germany, the “Law to Strengthen
Competition in Statutory Health Insurance” (GKV-WSG) came into force on April 1, 2007,
whereby Section 35b SGB V was revised. The Federal Joint Committee (G-BA) was authorized
to commission the Institute for Quality and Efficiency in Health Care (IQWiG) in accordance
with Section 139b (1) SGB V to evaluate future services according to their costs and
benefits and not only regarding their potential benefits, as was formerly the case.
In principle, IQWiG is not bound by fixed criteria with regard to the use of certain
methods for evaluating cost-benefit ratios; however, it must be based on “international
standards of evidence-based medicine and health economics recognized in the respective
specialist groups” and must include these standards in its decision-making process.
Some critics of cost-effectiveness analyses express concern that considering only
QALYs and corresponding ICERs could lead to limitations in the potential treatment
options available to patients, thereby denying treatment options that are “too expensive”.
It should be noted here that cost-effectiveness analyses based on scientific evidence
can inform payers and providers in the health care system that the ultimate decision
regarding the reimbursability of necessary services must be viewed both in the context
of the individual patient case as well as the context of the performance of the individual
health care system and its infrastructure. The thresholds of $5000–$200 000 per QALY
presented for the USA, for example, should not be regarded as absolute limits, but
rather as guideline values that do not apply in Germany in particular, since IQWiG
does not define absolute thresholds. With respect to the healthcare policy debate
on the reimbursability of radiological services, it is important to discuss which
threshold values should be used that lead to a significantly improved benefit for
the patient when comparatively “more expensive” diagnostics are used. The discontinuation
of method evaluation procedures by the Joint Federal Committee for the diagnostic
combination of positron emission tomography and computed tomography (PET/CT) communicated
in November 2020 illustrates the importance of cost-effectiveness analysis to prove
the tangible benefit of supposedly “expensive” examination techniques [27]
[32]. It is worth mentioning here that IQWiG, which was commissioned by the G-BA, developed
its own two-stage procedure for Germany, in which in the first step only the benefit
is assessed and only in the case of an increased benefit compared to the standard
treatment, in the second step an assessment is performed of the benefit in comparison
to the costs, e. g. by demonstrating a cost-effectiveness analysis. Addressing the
relevant analyses is also the task of the respective professional associations. For
example, within the German X-ray Society, the Working Group on Health Policy Responsibility
is concerned with identifying appropriate innovative methods and promoting their implementation
in day-to-day care for the benefit of patients, e. g. by carrying out cost-effectiveness
analyses.
Fig. 5 Different levels of resource allocation within an economy and their associated entry
points for cost-effectiveness analyses along the value chain.
Outlook
Using cost-effectiveness analyses, it is possible to model the effect of diagnostic
and interventional radiology methods in the short and long term. In radiology in particular,
short-term costs are often offset by long-term gains in quality of life and longevity,
as well as potential savings through better therapy planning. Thus, this methodology
has enormous potential, especially for radiology, by demonstrating and communicating
the benefits of diagnostic methods and interventional therapies. As discussed above,
economic analyses, and cost-effectiveness considerations in particular, are explicit
bases of reimbursement eligibility decisions in many healthcare systems [33].
Radiological expertise is essential for the identification of relevant issues as well
as realistic modeling of the clinical value chain. It is therefore imperative that
corresponding analyses be performed either by radiologists with appropriate economic
qualifications or by interdisciplinary teams taking radiological expertise into account
to ensure the clinical significance and technical accuracy of the results. It would
therefore be advisable, for example, to set up appropriate working groups within national
and international radiological societies and to specifically promote targeted training
in relevant economic analysis. Also due to its model-like character, interdisciplinary
as well as cross-site collaboration lends itself to cost-effectiveness considerations.
