The Role of Price Variation in Economic Analyses for Cancer Screenings: A Rapid Review.

Key Points for Decision Makers

Introduction
As healthcare costs continue to rise, there is increasing pressure to study and deliver high value care [1]. Cost-effectiveness analysis (CEA) is one such tool to evaluate the economic value of healthcare services and technologies. In the USA, payers and professional organizations utilize CEAs to inform policies around insurance coverage, benefits, and technology adoption [2]. In many cases, the economic value of a service is incorporated into guidelines that influence clinical decision making [3]. Cost-effectiveness analyses are useful for a wide range of healthcare tests and treatments, from population-level prevention measures to treatments for rare diseases.
The two core components of CEAs—which measure value—are cost and quality. Most CEAs examine all direct costs of a healthcare service in the analysis, including the amount paid by the payer and the out-of-pocket amount paid by the patient. Since 2020, several policies in the USA improved price transparency for healthcare services, uncovering previously opaque pricing information, particularly for commercial insurers [4]. One notable finding that has emerged from the newly required publication of hospital pricing data is the high degree of price variation both within and between hospitals, as the price of a single service may vary up to ten-fold between different payers and providers [5–7].
The significance of massive price variation in healthcare is a topic of research and is highly relevant for economic analyses. Sensitivity analysis is used to test how changes in key assumptions impact cost effectiveness. In the fragmented payer system in the USA, certain healthcare services may be highly cost effective at one facility but not at another. Thus, the magnitude of healthcare price variation raises questions about the validity of CEAs and other measurements of value. Accurate modeling is essential in policy analysis and decision making, and it is unclear if current researchers and professional organizations consider this high degree of price variation in their recommendations.
Considering the ever-changing landscape of US healthcare, high value care remains a priority for public health. One subset of the United States Preventive Services Taskforce (USPSTF) recommendations that affect nearly every American is cancer screening. Cancer remains the 2nd leading cause of death in the USA, and screening costs an estimated $43 billion each year [8, 9]. Due to the high cost and frequency of testing across a population, cancer screening is an ideal subject for cost-effectiveness research and may offer valuable insights into the relationship between value and price variation. The objective of this study was to assess how cost-effectiveness analyses obtain price information for common cancer screenings and account for price variation.

Methods

Data Sources and Searches
To investigate these questions around data sources and price variation, we performed a review of CEAs for cancer screening in the USA for four services that are included in the USPSTF recommendations: (i) colonoscopy, (ii) mammogram, (iii) prostate-specific antigen (PSA), and (iv) low-dose computed tomography (LDCT) of the chest. These four services were analyzed because they represent about 80% of spending on cancer screening, and previous work has produced highly granular data regarding the price and price variation of these four services [5, 6, 9]. The analysis was performed with PubMed search terms “cost-effectiveness” or “economic evaluation”, “screening”, “United States,” and “breast cancer” or “lung cancer” or “prostate cancer” or “colon cancer.” Results were filtered from January 1, 2021 to November 21, 2024. This time period was chosen because price transparency rules went into effect in 2020, and researchers should have had access to price transparency data during this period. This review focuses on research methodology and did not meet the eligibility criteria for registration with PROSPERO.

Study Selection
Every study in English was included and underwent a preliminary abstract review by a single author, with review and oversight by a second author. Studies were included if they evaluated at least one of the four screening services. Studies were excluded if one of the four tests was not analyzed. Studies were also excluded if a CEA was not performed or had used data from another country.

Data Extraction and Quality Assessment
A single investigator extracted information from each article that was included. Table 1 shows the relevant information that was included: the diagnostic test evaluated in the study, the data source used in the CEA, the perspective of the CEA, the price or cost of service stated in each paper, sensitivity analysis methodology (if used), and the range of prices used in the sensitivity analysis (if applicable). A second investigator reviewed the data extraction for quality assessment. Any disagreements between investigators were addressed through structured discussion, and consensus was reached by application of the inclusion and exclusion criteria.

Data Synthesis and Analysis
Each article was evaluated for the methodology around pricing data and price variation. Price ranges that were used in CEAs were then compared to known prices for common cancer screening tests, which were obtained from price transparency data and health insurance claims transaction data [5, 6].

Results
The query resulted in 97 articles, 21 of which underwent full review, and 16 were included in the final analysis. Table 1 summarizes the key data in each article. Only 44% of articles incorporated any degree of price variation in their analysis for cancer screenings. In over half of those, the maximum price was less than two-fold the minimum price, a range that underestimates the true amount of price variation [6]. Thus, only 3 articles (19%) performed a CEA with an accurate degree of price variation, all three of which performed a probabilistic sensitivity analysis. Table 2 summarizes prior work from Alford-Holloway et al. regarding price variation of four common cancer screening tests, demonstrating real-world data that these diagnostic studies vary in price by > 25%.

Regarding data sources, 94% of articles cited CMS or the Medicare Fee Schedule as their data source for price, and the remaining articles did not list a source for price. Most articles performed a CEA from the perspective of the US healthcare sector (50%), while many others did not explicitly state a perspective (37.5%).

Discussion
Ignoring price variation appears to be a systemic flaw in CEA methodology for common cancer screenings. In this set of studies, over half did not perform a sensitivity analysis on price, and many of the remaining studies underestimated the degree of price variation. Sensitivity analyses help to strengthen the results of CEAs, and inadequate pricing information undermines the findings of these CEAs.
While this review only evaluated CEAs in cancer screenings, these findings may represent a broader misunderstanding of healthcare prices. Survey data reveal that most Americans grossly underestimate the degree of price variation between healthcare providers, as over 70% of Americans expect healthcare prices to vary less than two-fold between facilities for the same service [10]. Furthermore, price variation is rarely appreciated in important economic analyses in other medical subspecialty fields. For example, the ACC/AHA/HRS Guidelines for Patients with Ventricular Arrhythmias state that “the initial cost of an ICD device is similar regardless of the clinical indication, so variations in ICD cost effectiveness are driven primarily by potential differences in clinical effectiveness,” relying on several articles with data from before 2000, some of which use outdated cost-to-charge ratios [11–14]. More recent work has consistently shown wide price variation within cardiology and cardiac surgery, suggesting the initial cost of services is often significantly different even for the same clinical indication [7, 15, 16]. Other research that recently circulated in the mainstream press suggested that bariatric surgery was more cost effective than GLP-1 medications for weight loss, yet the authors underestimated the price variation of surgery [17–19]. In addition, Choosing Wisely is an organization with over 700 recommendations to improve value in healthcare, yet price variation is not reported or described as a major factor in the value of care [20].
Future research in value and cost effectiveness must accurately reflect the degree of price variation in healthcare. In this review, not a single study referenced price transparency data or claims data. Several studies estimated commercial rates as a multiple of CMS rates “as no data are available to inform commercial cost estimates,” which is incorrect with recent price transparency rules [21, 22]. In a widely cited special panel report, Sanders et al. recommend “conducting sensitivity analyses that allow for a reasonable range of rates” but provide no specific guidance on the range of possible prices [23]. We believe that price transparency data can improve the quality of research, and CEAs need to incorporate the pricing information that is now publicly available. In addition, research methods and practice guidelines need to be updated to reflect that value can change drastically depending on the payer and site of service.
Given that quality has been persistently difficult to measure in healthcare, price alone is perhaps the most important variable in the value equation for many services. In other words, the same service may be high or low value based on where the service takes place and who is paying.
This study is limited by the modest sample size of CEAs that were available for review. Several systemic factors make cost-effectiveness analysis difficult in the USA. Unlike in the UK, there are multiple payors, provider systems, and stakeholders that complicate healthcare in the USA. Thus, there is more heterogeneity in the research approach to economic analysis. Another limitation is that publicly available price transparency data are relatively new, and researchers are still becoming familiar with establishing best practices with the data. Because this brief review did not evaluate CEAs prior to 2020, the impact of the US price transparency initiative could not be assessed. This review did not evaluate search terms for “cost utility” studies, which may have resulted in a more limited evaluation. Finally, this work is limited in its scope to cancer screenings, and generalizability is limited for services beyond four common cancer screenings.
In summary, price variation is a crucial variable in healthcare and should be considered in all economic analyses and efforts to improve value in healthcare. Prices for healthcare services in the USA are known to vary widely, a central fact that has been systematically under-represented in cost-effectiveness analyses for common cancer screenings. Price transparency is an important tool to shed light on value in healthcare and might improve economic analyses moving forward.

Supplementary Information
Below is the link to the electronic supplementary material.