6. Economic Analysis

Digital mammography and chest radiography are advanced technologies that provide medical benefits to patients, as well as economic benefits to patients and healthcare providers.

However, only some healthcare facilities and only some patients will have access to these benefits. The first cost of digital imaging equipment is substantially higher than the cost of conventional film-based X-ray units, and high cost is expected to limit the number of digital imaging systems placed in service.

The ATP-funded low-cost process will reduce the equipment cost of digital imaging, facilitate the deployment of additional units, and make it possible for many more Americans to participate in the above benefits. Figure 4 summarizes the flow of these benefits and distinguishes those that can be quantified from those where only a qualitative analysis is feasible.

METHODOLOGICAL APPROACH

The chief mechanism for delivering the healthcare and economic benefits of the LCM process is expected to be additional sales of Senographe 2000D and Revolution XR/d units beyond the unit sales that would occur without the LCM process.

Discussions with ATP’s industry partners, a review of independent market estimates, and broad-ranging discussions with healthcare industry participants, including healthcare and imaging center administrators, provided the basis for estimating additional unit sales that would occur following deployment of the ATP-funded LCM process. The following assumptions were developed following these discussions:

• Base Case: A projected year-to-year increase in sales 50 percent above the counterfactual case. This is equivalent to a six months’ increase in sales, over the counterfactual case.
• Step-Out Scenario: A projected year-to-year increase in sales 7/12 higher than for the counterfactual case. This is equivalent to an increase by seven months over the counterfactual case, or by one month over the base case.

Table 5 shows the projected number of additional units expected to be deployed as a result of the ATP-funded low-cost process based on these assumptions. Columns   and D present the counterfactual case, i.e., unit sales without the impact of the ATP-funded LCM process. Columns B and E present base-case estimates of additional unit sales associated with LCM cost reductions, and columns C and F present step-out scenario estimates of additional unit sales.

Table 5. Projected Unit Sales Enabled by ATP-Funded Low-Cost Process

Additional sales will be contingent on GE Medical Systems (GEMS) completing the remaining technical development tasks, implementing the low-cost process, and instituting price reductions to reflect decreased detector costs. Given intense competitive pressures in the medical equipment industry, there is a strong probability that GEMS will implement the low-cost process during the next two years, but not a 100 percent certainty. As a result, a 70 percent probability factor is used to adjust the expected value of combined cash flow benefits. The 70 percent probability estimate is based on three component factors expected to drive a potential business decision to implement the LCM process:

• Technical feasibility of the low-cost process.
• Growing market demand for digital imaging.
• Increasing competitive pressures to reduce component costs.

ATP AND INDUSTRIAL PARTNER INVESTMENTS

During the 1996–2000 period, ATP invested $1.575 million, and its industry partners, GE Global Research and PerkinElmer, invested $1.863 million in the LCM process. The ATP investment was approximately equivalent to the project’s direct costs. For purposes of the cash flow analyses, the ATP investment was normalized to 2002 dollars (using an average annual inflation rate of three percent) and assumed to occur in 1998, the midpoint of the four-year investment period.

The normalized ATP investment, expressed in 2002 dollars, was $1.773 million, and the normalized industry investment was $2.097 million. Before the low-cost process can be deployed for high volume detector fabrication, some additional GE and PKI investment will be required to complete the remaining technical tasks.

BASE-CASE ECONOMIC ANALYSIS

Additional unit sales of the Senographe 2000D and Revolution XR/d associated with the prospective implementation of the low-cost process will deliver substantial benefits to many Americans who would not otherwise be able to enjoy these benefits.

Of these benefits, only the following can be meaningfully quantified at this time:

• Higher productivity and reduced operating costs for healthcare providers.
• Avoided diagnostic costs from fewer mammography false-positive results.
• Avoided patient time and travel expenses resulting from fewer mammography false-positive results and fewer call backs for retakes.

Table 6 presents the estimated cash flow impact for the base case of these quantifiable benefits to healthcare providers and patients that will result from additional unit sales following deployment of the ATP-funded low-cost process. The cash flow estimates reflect the following assumptions:

Senographe Productivity Assumption: GE’s product financial pro formas specify that the Senographe 2000D has an operating capacity of 9,600 procedures per year, twice the procedure volume for one conventional mammography unit (General Electric, 2002).

GE’s Senographe 2000D financial pro formas further specify a retake rate (due to inadequate initial image quality) of 0.5 percent for the Senographe 2000D as compared to 2.5 percent for conventional units. While equipment costs for the Senographe 2000D are 2.5 times the cost of conventional units, operating costs for personnel and material are only 44 percent of the operating costs for conventional units. Combining these productivity/cost differences, the net cash flow advantage from deploying one additional Senographe 2000D unit instead of two conventional units is indicated below. At year one, the annual cash flow variance is $116,000 per unit. Beyond five years, the annual cash flow variance stabilizes at $169,000 per unit.

Table 6. Cash Flows from Additional Unit Sales, Assuming Implementation of ATPFunded Low-Cost Process in 2004 ($ Millions, in 2002 Dollars): Base Case

Assumption About Fewer False Positive Results: Senographe 2000D systems provide fewer false positive results and therefore require fewer patient recalls than conventional systems, by approximately 20 percent (Lewin et al., 2000). Based on 9,600 screening procedures per year, 10 percent, or 960 women, can expect to receive false-positive results from a conventional mammography unit (FDA, 1997). The Senographe 2000D reduces this number by 20 percent, or by 192 women, who will be able to avoid the medical risks, expenses, and anxiety associated with unnecessary follow-up procedures. The average medical cost of unnecessary procedures was conservatively estimated at $330 per woman, and the deployment of each additional unit is then associated with $63,360 of medical savings on behalf of 192 women.

Assumptions About Avoided Time Loss and Travel Expenses: Patient recalls and retakes involve loss of work time and travel costs. For every Senographe unit in use, 192 women avoid being recalled for unnecessary diagnostic procedures. In addition, 2 percent (of the 9,600 women screened), or 193 women, avoid being called back for retakes due to poor image quality. In total, 385 women will avoid unnecessary loss of work time and travel expenses. Assuming that 70 percent of 385 women work full time, earn $16.66 per hour (U.S. Department of Labor, 2001), and lose 3 hours of work to comply with recall and retake notices, and that travel expenses average $4.00 per incident, annual avoided expenses for 385 women are estimated at $15,003. Cash flow benefits from avoided time loss and travel expenses are calculated by multiplying the $15,003 savings by the cumulative number of additional Senographe units that are deployed as a result of the low-cost process.

Revolution XR/d Productivity Assumption: GE’s product financial pro formas specify that the system capacity of the Revolution XR/d is 24,000 Xray procedures per year, twice the annual volume of a screen film unit (General Electric, 2002a). The retake rate (due to inadequate initial image quality) for the Revolution XR/d is specified at 0.5 percent, as compared to 4 percent for screen film units. While Revolution XR/d equipment costs were 1.7 times the cost of conventional units, operating costs for personnel and material are only 23 percent. The net cash flow advantage of deploying one additional Revolution XR/d unit instead of two conventional units is indicated  below. Beyond five years, the annual cash flow variance stabilizes at $381,000 per unit.

Public returns on ATP’s investment in the low-cost process for the base case are summarized at the bottom of Table 6. The deployment of additional Senographe 2000D and Revolution XR/d units resulting from the implementation of the ATP-funded low-cost process is associated with a net present value of $219 million, an internal rate of return (public return) of 69 percent, and a benefit cost ratio of 125:1. These measures of return reflect benefits to the healthcare industry and to patients resulting from the ATP investment.

 Given clear GE representations that the high-risk LCM process would not have been undertaken without ATP funding, the analysis summarized in Table 6 attributes 100 percent of net cash flows to the ATP investment. To allow for uncertainties in predicting the future, net cash flows were adjusted by a 70 percent probability factor (based on technical and commercial discussions with GE, PKI, and other industry participants) as to the future deployment of the low cost process (see Methodological Approach above). This adjustment is applied in the last column of Table 6.

A component analysis of the net present value (Figure 5) points to the substantial impact of productivity gains from Revolution XR/d and Senographe 2000D technologies, accounting for 61 and 27 percent of net present value, espectively. Avoided treatment costs from reduced false-positive results account for 10 percent of the net present value, and avoided patient time and travel costs account for less than two percent.

STEP-OUT SCENARIO ECONOMIC ANALYSIS

Table 7 presents the estimated cash flow impact for the step-out scenario of quantifiable benefits resulting from additional unit sales and the ATP-funded low-cost rocess. Cash flow estimates reflect the following assumptions.

Senographe Productivity Assumptions: For the step-out scenario, the GE Senographe 2000D operating capacity was specified at 10,560 procedures per year, a 10 percent increase in the 9,600 procedures per year used in the base case and by the Senographe 2000D financial pro formas. The net cash flow productivity advantage of deploying one additional Senographe unit instead of two conventional units is indicated below. Beyond five years, the annual cash flow variance stabilizes at $275,000.

Table 7. Cash Flows from Additional Unit Sales, Assuming Implementation of ATPFunded Low-Cost Process in 2004 ($ Millions, in 2002 Dollars): Step-Out Scenario

Note: Appendix C provides a more detailed analysis.

Assumptions About Fewer False-Positive Results: Assuming 10,560 screening procedures per unit each year, 1,056 women can be expected to receive false-positive results with conventional equipment. The Senographe 2000D reduces that number by 20 percent, or 211 women. Further assuming for the step-out scenario that the per capita medical cost of unnecessary follow-up procedures increases by 9 percent over the base case to $360, the deployment of every additional Senographe unit deployed as a result of the ATP-funded LCM process can be associated with $75,960 of medical savings on behalf of 211 women.

Assumptions About Avoided Time Loss and Travel Expenses: For every Senographe 2000D unit, in addition to the 211 women who will avoid being recalled for unnecessary diagnostic procedures because of falsepositive results, 2 percent (of the 10,560 women screened), or 212 women, will avoid being called back for retakes due to poor image quality. In total, 423 women will avoid unnecessary loss of time from work and travel expenses. In the step-out scenario, travel expenses per each incident were increased to $5.00 from the $4.00 assumed in the base case.

Revolution XR/d Productivity Assumption: Annual system capacity was assumed to be 5 percent higher than for the base case, or 25,200 procedures per year. The net cash flow advantage of deploying one additional Revolution unit is indicated below for the first five years of operation. Beyond five years, the annual cash flow variance stabilizes at $436,000 per unit.

Public returns on ATP’s investment in the high-risk, low-cost process for the step-out scenario are summarized at the bottom of Table 7. The deployment of additional Senographe 2000D and Revolution XR/d units resulting from the implementation of the ATP-funded low-cost process is associated with a net present value of $339 million, an internal rate of return (public return) of 77 percent, and a benefit cost ratio of 193:1. Again, these measures of return reflect benefits to healthcare industry users and to patients resulting from the ATP investment.

ESTIMATED PRIVATE BENEFITS TO ATP INDUSTRIAL PARTNERS

Continued motivation to commercialize the ATP-funded technology on the part of General Electric Company and PerkinElmer is a necessary precondition for completing the remaining technical tasks and for implementing the low-cost process. Only then will the general public come to enjoy the associated medical and economic benefits from improved productivity, reduced false-positive rates, and avoided patient expenses.

To characterize the motivation of ATP’s industrial partners to implement the lowcost process, we estimated prospective worldwide sales revenues from additional Senographe 2000D and Revolution XR/d unit sales, expressed in 2002 dollars. Estimated sales revenues are shown in Table 8. Because GE operating cost projections were not available, future profit contributions from Senographe 2000D and Revolution XR/d units could not be estimated.

The magnitude of benefits, as reflected in additional revenues for GE and PKI from the low-cost process, appears to be consistent with ongoing motivation of the ATP industry partners to complete remaining technical tasks and to implement the ATP-funded low-cost process.

QUALITATIVE BENEFITS

In addition to the benefits quantified, many important qualitative benefits will be experienced by Americans who gain access to Senographe 2000D and Revolution XR/d equipment through additional units sold as a result of the ATP-funded LCM process technology.

Early Detection Benefits Associated with Analytical Continuity and Computer-Aided Detection

According to a 1997 FDA Report, breast cancers detected early (for example, through screening mammography) tend to be at significantly less advanced stages than cancers diagnosed for unscreened women. As Table 9 indicates, 73 percent of screened populations were diagnosed with in situ or stage 1 cancers, while only 54 percent of unscreened populations had similar outcomes, showing a substantial early detection advantage from periodic screening. Differential treatment costs for in situ and early stage breast cancers versus later stage cancers also point to significant economic benefits from early detection to be facilitated by Senographe 2000D technology and the ATP-funded low-cost process. “The average cost of treating early stage breast cancer is $11,000 and for late-stage lesions, $140,000” (Evans, 1999).

Table 8. Estimated Revenues to ATP Industry Partners from Future Implementation of ATP-Funded Low-Cost Process ($ Million, in 2002 Dollars)

Screening with the Senographe 2000D, instead of conventional X-ray units, can be expected to facilitate early detection of breast cancer by eliminating the loss of prior mammograms and by feeding digital data to CAD systems efficiently.

Eliminating the loss of mammograms ensures the availability of prior mammograms for continuity of interpretation. Continuity is considered to be an important analytical factor for accurate interpretation and early detection.

The Senographe 2000D generates direct digital images for CAD and facilitates the effective utilization of CAD systems. Studies indicate that CAD systems may improve breast cancer early detection rates by 20 percent compared to conventional mammography without CAD (Oncology News, 2001). While film mammograms can be digitized and fed into CAD systems, Senographe units will provide more efficient and timely digital input to optimize CAD system effectiveness.

Table 9. Percentage of Cancer Victims Diagnosed at Various Stages, Screened Versus Unscreened Populations

Radiation Exposure Reduction

Conventional X-ray units generally use a higher radiation dosage than digital mammography units to obtain an adequate image for dense breast tissue. Digital mammography is a more effective screening modality for dense breast tissue at lower dosage levels. The 20 percent reduction in false-positive results and associated avoidance of diagnostic procedures as well as the 80 percent reduction in retakes for unacceptable image quality further contribute to the reduction in radiation exposure from the use of the Senographe 2000D rather than conventional equipment.

Counteract Growing Shortage of Radiologists

Increased productivity and throughput from Senographe 2000D and Revolution XR/d will counteract the growing shortage of radiologists and mammography specialists in the United States. Increased throughput will also reduce patient waiting times, and potentially encourage more regular screening.

Facilitating Telemammography and Teleradiology

U.S. health disparities by income, race, and ethnicity have been well documented (NIH, 2000). More than 20 million women lack adequate access to high quality screening mammography services and probably twice that number of people (men, women, and children) lack adequate access to high-quality chest, throat, and other radiography services. Lack of adequate access results in underutilization of screening and other X-ray modalities and may be related, at least in part, to the higher percentages of advanced breast cancers among some minority populations (Lawson et al., 2000).

Telemammography and teleradiology represent evolving clusters of advanced network technologies, offering the promise of higher quality screening and imaging services for currently underserved populations.

Many diverse factors will have to line up to complete the development of viable telemammography and teleradiology networks without reliance on government subsidies. Such factors include additional technology advances, regulatory changes, novel institutional relationships, cost containment measures, market, and financial issues (Shtern, 1999). One important technical and economic requirement is direct digital image acquisition through inexpensive full-field digital imaging systems. The

ATP-funded low-cost process addresses that requirement directly and could significantly contribute to telemammography and teleradiology reaching their full potential and delivering improved and cost-effective medical services to currently underserved populations.

Cardiac Imaging

Conventional cardiac imaging units are priced around $1 million, and the incremental costs of digital over conventional technology are only 10 to 20 percent.

Because digital detector components represent a smaller part of the overall system price of digital cardiac systems, detector cost reductions are less likely to impact digital unit sales than is the case for mammography or radiography systems. Nevertheless, digital detector cost improvements may have some beneficial impact.

When this occurs, the ATP-funded low-cost process will be instrumental in delivering the spectrum of benefits of digital technology to cardiac patients who may not have otherwise had access to these benefits, including lower radiation dosage as well as improved fluoroscopic performance in imaging the higher heart beat rates of infants.

Knowledge Diffusion

As part of the GEGR/PKI joint-venture structure, proprietary GE detector technology, including the ATP-funded low-cost process, was transferred to PKI and may, in the future, be further transferred to sub-licensees for the potential development of nonmedical imaging applications, such as in industrial machine vision, nondestructive testing, and cargo inspection for airport security. Additional knowledge diffusion may result from two patents issued to General Electric Company: US5838054 for Contact Pads for Radiation Imagers, US5648296 for Post Fabrication Repair Method for Thin Film Imager Device, and a patent filed for Gated Diodes for Reduced Mask Imager Process. These patents and patent filings resulted from work on the ATPfunded low-cost process technology. 

Return to Table of Contents or go to Section 7. Conclusions.

Date created: April 25, 2003
Last updated: August 2, 2005

ATP website comments: webmaster-atp@nist.gov  / Technical ATP inquiries: InfoCoord.ATP@nist.gov.

NIST is an agency of the U.S. Commerce Department
Privacy policy / Security Notice / Accessibility Statement / Disclaimer / Freedom of Information Act (FOIA) /
No Fear Act Policy / NIST Information Quallity Standards / ExpectMore.gov (performance of federal programs)