Jeffrey Helton, University of Colorado at Denver Business School

Contact:

Introduction

Hospitals in the United States are facing declining payments and reduced patient volumes that cause them to pursue cost reduction strategies, including laying off employees. While this has happened periodically since the advent of prospective payments to hospitals in the 1980s, there has been a more recent uptick in hospital layoffs since the end of the COVID-19 pandemic in 2021 [1, 2, 3, 4]. Historically, hospitals have targeted nursing and other direct caregiver positions as patient volumes declined with the growth of managed care programs that restricted hospital use [5, 6]. Growing revenues [7], adding new services, or addressing high-cost contract labor costs [8] have also been mentioned as strategies to address patient volume and revenue declines.

Common among strategies cited in the literature is the focus on revenue, volume of service, payment rates, or variable costs. Recent news of hospital layoffs has shown a change in hospital layoff targets. Recent workforce reductions have moved to the addition of non-patient care staff to the lists of employees subject to reductions in workforce, in particular managers and executive positions [1, 2, 3, 4, 5]. It seems notable that the fixed costs created by managerial and executive positions have come under broader cost-cutting scrutiny at the time of this writing. The extent of administrative costs in healthcare has been examined in the literature [9, 10, 11], but these studies based the expenses classified as “administrative” on the Centers for Medicare and Medicaid Services (“CMS”) Hospital Medicare Cost Report. Those expenses include not only administrative salaries, but other general administrative costs such as legal fees, accreditation fees, facility licensing costs, and even taxes for investor-owned organizations [9, 12]. Prior studies examined the proportion of administrative expense that encompassed both labor and non-labor components. None have addressed the question from the perspective of the number of administrators (and their associated fixed costs) with hospital profitability.

In their study of hospital administrative costs, McKay and colleagues [9] identified specific functions that generate those costs, including “management of human, financial, and facility resources,” as well as administrative compliance and billing/collection for services provided. Further, Holmes and colleagues [8] indicated that administrators devised and executed the strategies needed to improve financial performance in rural, financially challenged hospitals. Thus, it might seem perplexing why organizations might reduce the number of staff charged with the very functions that could improve a hospital’s resilience to environmental challenges such as reduction in patient volumes or payment rates.

In view of this seeming paradox, this paper seeks to examine this question: Is there a relationship between administrative staffing levels and profitability in hospitals? While other studies have evaluated the question of administrative costs, this preliminary study aims to be the first to explore the question from the perspective of the number of persons doing that administrative work. Since other costs beyond administrative labor were considered in prior studies, this work seeks to learn more about the influence of the administrative Full-Time Equivalent (FTE) on organizational profitability. This staffing approach may better inform hospital leadership on decisions to increase or reduce managerial positions in the organization.

There are multiple ways to define “profitability” in a hospital, including net income, return on assets, or operating margins. In current hospital markets, many hospitals – especially voluntary, non-profit hospitals – rely on non-operating income from investments to generate net income and achieve positive return on assets while losing money from hospital operations [13, 14]. While those non-operating sources are important for the long-term viability of a hospital business, the income from those sources is in many respects a function of the historical asset base for the business and less about the acumen of managers in operating the hospital’s core business: patient care [15].  In evaluating the administrative impact on profitability, this study will focus on income from patient care.  This is consistent with the work of Bai and Anderson [16].

Other studies of hospital profitability have given attention to environmental factors such as patient volumes, payer mix, market power, and insurer market penetration [6, 16].  These factors drive revenue and may not necessarily demonstrate management skill in operating a hospital business. Revenue performance certainly improves profitability and may show that administrators have the entrepreneurial skill needed to generate volumes of service and income streams. However, the operation of a hospital is complex and requires not only generating income streams, but also balancing the management of limited resources to keep operating expenses within those income streams while providing quality care to patients. The assessment of the impact of administrators on both revenue and expense measures, would be more informative for learning if this type of staffing can improve hospital profitability.

Bed size is a factor in hospital profitability. More beds could mean greater market presence, as well as an ability to spread fixed costs such as maintenance or depreciation across more activities and gain economies of scale [17]. For the purposes of this work, it may be more useful to consider revenues and expenses per unit of service rather than the overall size of the organization. Bai and Anderson used revenue and expense per adjusted discharge [16], and Lalani and colleagues [13] used revenues and expenses per patient day. The smaller unit of measure at the patient day level rather than discharges may be a better proxy for administrative effectiveness as managers must manage length of stay as well as operating expenses to maintain profitability [17].

Methods

Timeframe

This exploratory study uses a cross-sectional approach, incorporating publicly available secondary data from 2022, which represents the latest year with data for all United States hospitals. Given the preliminary nature of this work, one year of data is used to determine the general nature of the relationships posited here. Further, 2022 is the first year of more normal hospital operations following the COVID-19 pandemic of 2020 and 2021. Given the post-pandemic “new normal” of hospital operations, this research approach could better inform practice in coming years. There is only one year of complete post-pandemic data available for this work at the time of this writing. Also, the impact of non-operating revenues from federal support payments would distort the results of this analysis. Hence, this exploratory work is intended to evaluate only the general relationship of administrative staffing to measurable financial results in general, acute care hospitals in the United States in 2022. Veterans Administration (V.A.) hospitals, specialty, and children’s hospitals were omitted from consideration because their operating payment mechanisms generally differ from that of the general acute care hospital. Finally, any hospital that experienced a change in ownership or control status in 2022, was omitted under the premise that management decisions in those facilities would differ from normal operations either while the facility prepared for a change of ownership or assimilated a new control of operations [7,13].

Data Sources

Data for variables were extracted from hospital financial statement and operational data filed in annual Medicare cost reports filed with CMS each year [18].  Although the reports are titled “Medicare,” they include operating data for all aspects of the hospital and are appropriate for this type of analysis [9, 13, 16]. All revenue, expense, and staffing variables were adjusted to account for hospital size and volume, based on adjusted patient days. The adjusted patient day is a measure that considers both inpatient and outpatient service volumes in a hospital [17]. Case mix index data, which measures the overall relative severity of conditions treated in a hospital, were obtained from publicly available files from the CMS website [19].

The dependent variable in this analysis was patient service income per adjusted patient day. Patient service income is separately reported on the Medicare cost report and serves as a measure of operating income for a hospital. Revenue per adjusted patient day and expense per adjusted patient day were treated as dependent variables and examined in separate analyses to explore differences attributable to scale of operation and service mix. Since both revenue and operating expense are components of profitability, those two elements were also considered in this analysis. Net patient revenue and operating expenses were both divided by total adjusted patient days to be on the same unit of measure as profitability. Evaluation of financial performance metrics based on the adjusted patient day in the dependent variable, is consistent with work done by Karim et al and Epane’ et al [20, 21].

Labor Hours

To operationalize the level of administrative staffing in a hospital, the Medicare cost report identifies paid labor hours by hospital department. “Administration” and “Nursing Administration” departments were considered as “administrative” for purposes of this study. Administrative services provided by a multi-hospital system office or a management contract were also included in this variable. The administrative department in the cost report includes not only executives, but patient accounting, finance, and other management staff. The cost report does not segregate labor hours by administrative discipline. So, for purposes of this preliminary study, all administrative employee labor hours will be considered. The annual total paid labor hours for the administrative departments were then divided by 2,080 to arrive at an FTE value [17]. The total FTE value was then divided by the adjusted average daily census to arrive at an FTE per adjusted occupied bed.

Case Mix Index

The relative severity of illness in patients treated by hospitals are also a factor in hospital profitability performance, since more severe illness should result in higher overall revenue. In addition, if prices are set correctly by administrators, profits should increase as well. The common measure for the severity of patient illness in a hospital, known as the case mix index (CMI), were also included in this analysis, as was done in prior studies such as those by Bai and Anderson [16] and Lalani and colleagues [13]. Because rural critical access hospitals do not report case mix data since they are reimbursed on a cost basis, those facilities were omitted from the analysis. CMI data were obtained from the inpatient hospital payment resources page on the CMS website [19].

Hospital Characteristics

The analysis also included the following variables that controlled for hospital characteristics: rural or urban location, ownership (for profit, nonprofit, or public), system affiliation, teaching affiliation for physician residencies, operation of a distinct part inpatient psychiatric or inpatient rehab facility within the hospital. The latter two variables for the distinct part units were included because they are paid at different rates than medical/surgical services and could impact revenues for those hospitals [12].  Since the implementation or operation of such services is an administrative decision impacting profits from patient care, they were included in the study analysis. Rural/urban location was included as a binary variable, with rural location being the base case in operationalizing this variable in the analysis. Ownership was similarly converted to a dummy variable with public ownership being the base case in the analysis. All other variables for hospital characteristics were included as binary variables. Table 1 shows the variables used in this analysis and sources for those variables.

Table 1. Variables and Data Sources

Variable	Source
Income from Patient Services per Adjusted Occupied Bed	HCRIS, Worksheet G-3, Line 5, Column 1/ [(HCRIS, Worksheet S-3, Part I, Line 14, Column 8 / 365) x (HCRIS, Worksheet G-2, Line 28, Column 3/HCRIS, Worksheet G-2, Line 28, Column 1)]
Net Revenue per Adjusted Occupied Bed	Net Revenue / [(HCRIS, Worksheet S-3, Part I, Line 14, Column 8 / 365) x (HCRIS, Worksheet G-2, Line 28, Column 3/HCRIS, Worksheet G-2, Line 28, Column 1)]
Operating Expense per Adjusted Occupied Bed	Total Operating Expense / [(HCRIS, Worksheet S-3, Part I, Line 14, Column 8 / 365) x (HCRIS, Worksheet G-2, Line 28, Column 3/HCRIS, Worksheet G-2, Line 28, Column 1)]
Administrative FTE per Adjusted Occupied Bed	[(HCRIS, Worksheet S-3, Part 2, Line 27, Column 5 + HCRIS, Worksheet S-3, Part 2, Line 28, Column 5 + HCRIS, Worksheet S-3, Part 2, Line 38, Column 5)/2,080] / [(HCRIS, Worksheet S-3, Part I, Line 14, Column 8 / 365) x (HCRIS, Worksheet G-2, Line 28, Column 3/HCRIS, Worksheet G-2, Line 28, Column 1)]
Urban/rural status	HCRIS, Worksheet S-2, Part 1, Line 26, Column 1
Teaching status	HCRIS, Worksheet S-2, Part 1, Line 56, Column 1
Distinct Part Psychiatric Unit	HCRIS, Worksheet S-2, Part 1, Line 70, Column 1
Distinct Part LTACH Unit	HCRIS, Worksheet S-2, Part 1, Line 75, Column 1
Multi-Hospital Affiliation	HCRIS, Worksheet S-2, Part 1, Line 141, Column 1
Ownership/Control	HCRIS, Worksheet S-2, Part 1, Line 21, Column 1
Case Mix Index	CMS Case Mix Index Files from Hospital Inpatient PPS reimbursement files at https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS

Market factors beyond the control of administrators such as payer mix, population, or insurer market penetration were not included in this analysis.  Hospitals that underwent a change in ownership, did not report complete data for all study variables, or were critical access hospitals that were omitted from the study sample, leaving a total of 1,483 hospitals included in the study analysis. Descriptive statistics for all variables are listed in Table 2.

Table 2. Descriptive Statistics for Study Variables

Variable	Mean	Median	Std. Dev	Min.	Max.
Patient Service Income per Adjusted Occupied Bed	-164.36	-133.91	1,064.31	-9,005.15	17,909.86
Net Revenue per Adjusted Occupied Bed	3,370.93	2,918.23	2,150.06	429.96	31,654.64
Operating Expense per Adjusted Occupied Bed	3,535.29	3,014.06	2,121.49	603.34	29,191.11
Administrative FTE per Adjusted Occupied Bed	0.77	0.58	0.80	0.01	13.81
Case mix index	1.77	1.73	0.35	0.84	3.91

Counts for the categorical variables in the analysis are listed in Table 3 below.

Table 3. Counts for Study Categorical Variables

Variable	Count	% of Study Facilities*
Voluntary/non-profit control	950	64%
Proprietary control	296	20%
Public control	237	16%
Teaching status	497	34%
System affiliated	820	55%
Urban location	871	59%
Distinct part psychiatric unit	292	20%
Distinct part rehab unit	280	19%

 * Will not sum to 100% because some facilities may be represented in multiple categories

Regression Analyses

The research question was evaluated using multivariate linear regression analysis to determine the association between Income from Patient Services per Adjusted Occupied Bed and the independent variables with each study outcome and reported coefficients, betas, and statistical significance. The research question is premised on the idea that more administrators, exercising their entrepreneurial and management skills, will have a positive impact on profitability in a hospital. The analysis used three multivariable linear regression analysis models: one for profitability, another for net revenue per adjusted patient day, and a third for operating expense per adjusted patient day. The latter two used revenue per adjusted patient day and operating expenses per adjusted patient day as dependent variables in two separate additional regression analyses. The regressions with net revenue per adjusted patient day were performed to understand the relative impact of management on these component elements of overall profitability from patient care services. All regression analyses reported coefficients and betas for covariates in the models and figures illustrating marginal effects of fiscal year. A probability value of <0.05 (two-tail) was considered statistically significant for all tests.

Regression analyses in this study used a robust multivariate regression to control for potential serial correlation among variables. Patient service income per adjusted patient day was the dependent variable in one regression, net revenue per adjusted patient day in the other, and the third used for operating expense per adjusted patient day. Stata IC 17 was used to perform all statistical analyses.

Results

The first analysis was a regression using patient service income per adjusted patient day as the dependent variable. This first step was to establish the presence of any relationship between administrative staff levels and profitability. The results of this regression analysis are shown in Table 4.

Table 4. Regression Results for Patient Service Income per Adjusted Patient Day

Patient service income per adjusted patient day	Coefficient	Robust std. err.	P>|t|
Administrative FTE per Adjusted Occupied Bed	-695.44	385.17	.071
Voluntary/non-profit control	494.71	254.96	.053
Proprietary control	2,722.45	1,156.99	.019*
Teaching status	-349.11	242.58	.150
System affiliation	-975.75	631.32	.122
Urban location	307.39	326.24	.346
Case mix index	1,362.48	703.78	.053
Distinct part psychiatric unit	-406.12	244.95	.098
Distinct part rehab unit	-135.85	260.17	.602
Constant	-1,985.67	1,209.06	.101

 * – indicates significant relationship

Interestingly, only proprietary control is significant (p-value of 0.019) in this assessment of profitability per adjusted patient day. The level of administrative staffing approached statistical significance (p = 0.071), with the negative coefficient indicating reduction in profitability. Two other variables – voluntary/non-profit control and case mix index – also exhibited p-values just above the significance level of 0.05 (both p = 0.053). Teaching status, system affiliation, and the operation of a distinct part psychiatric unit, all showed very weak relationships to profitability in the analysis (p ranging from 0.098 to 0.150).

To better understand the potential drivers of profitability, the dependent variable of profitability per patient day was decomposed into its key component parts: revenue and expense. The analysis was then re-run using net revenue per adjusted patient day as the dependent variable in the regression model and then again with operating expense per adjusted patient day as the dependent variable. The results of analysis of administrative staffing on net revenue per patient day are shown in Table 5.

Table 5. Regression Results for Net Revenue per Adjusted Patient Day

Net revenue per adjusted patient day	Coefficient	Robust std. err.	P>|t|
Administrative FTE per Adjusted Occupied Bed	2,846.60	855.32	.001*
Voluntary/non-profit control	1,561.37	1,126.08	.166
Proprietary control	-639.89	533.03	.230
Teaching status	-585.66	369.42	.113
System affiliation	-199.64	801.39	.803
Urban location	378.85	453.07	.403
Case mix index	3,255.63	822.71	.000*
Distinct part psychiatric unit	-852.02	351.12	.015*
Distinct part rehab unit	663.83	500.54	.185
Constant	-4,241.41	1,348.93	.002*

* – indicates significant relationship

In this model, administrative FTE per adjusted occupied patient day was found to be significant (p =.001), with each additional FTE per adjusted occupied patient day associated with a $2,846.60 increase in net revenue per adjusted bed day. Also significant is the relationship between net revenue, and case mix index, and the operation of a distinct part psychiatric unit.

Continuing the decomposition of profitability to revenue and expense components in this analysis, the third regression using operating expense per adjusted patient day as the dependent variable yielded the results shown in Table 6.

Table 6. Regression Results for Operating Expense per Adjusted Patient Day

Operating expense per adjusted patient day	Coefficient	Robust std. err.	P>|t|
Administrative FTE per Adjusted Occupied Bed	3,542.05	690.61	.0001*
Voluntary/non-profit control	-666.37	248.21	.007*
Proprietary control	-145.18	430.64	.736
Teaching status	-236.55	222.65	.288
System affiliation	776.10	308.95	.012*
Urban location	71.45	233.86	.760
Case mix index	1,893.15	336.90	.000*
Distinct part psychiatric unit	-445.91	261.07	.088
Distinct part rehab unit	799.68	400.49	.046*
Constant	-2,750.46	651.39	.000*

* – indicates significant relationship

Administrative FTE per adjusted patient day was significant in this analysis (p = 0.0001), adding expense as staffing increased in this labor category. Also significant were variables for system affiliation, ownership/control, case mix index, and operation of a distinct part rehab unit.

Overall, these three regression analyses suggest that that some relationship between the level of administrative staffing in a hospital and its patient service income. Together, these findings represent fodder for further consideration.

Discussion

This study raises interesting perspectives on decisions made to set administrative staffing levels in contemporary hospital management. While the statistical relationship between administrative staffing and patient service income per adjusted patient day was not significant, that makes sense because the study purposely left out external market factors that may have some impact such as market share, payer mix, or market insurer factors. Despite that intended omission, the relationship between administrative staffing appears to have an overall net negative impact on a hospital’s profit performance, as noted in the patient service income per adjusted patient day regression.

Staffing, Net Revenue and Operating Expenses

Regressing administrative FTE on net revenue, there is a positive relationship between administrative labor and net revenue. The idea that administrators could have an impact on revenues is borne out in the observed relationship between administrative staffing and net revenue per patient day. This observation makes sense if we consider that administrators are making strategic decisions that expand business opportunities for a hospital, set prices, competitively, and in a way that exceed operating expenses, and structure favorable payment relationships with insurance plans. One possible explanation is that the entrepreneurial ability of administrators may improve revenues. However, while the administrative component of staffing includes executive management functions, it also includes labor hours attributable to patient accounting and finance functions. This may be important since those functions also have a positive impact on net revenues by virtue of improved collection for services rendered. Identifying a credible data source that segregates labor hours for those functions, would be a valuable next step in this line of research. As a result, it is not clear from this analysis the extent to which improved net revenue in hospitals comes from better collections versus better administrative decision-making. However, it stands to reason that the greater the net revenue level in a hospital, the greater the overall ability of that hospital to absorb the administrative overhead component.

More telling in the assessment of administrative impact on profitability is the relationship between administrative staffing and operating expenses, where the level of administrative staffing in a hospital increases operating expense per adjusted patient day. Ceteris paribus, adding administrative staff (including executive managers who oversee expense management) increases operating expense rather than controlling it at a lower level. It is not surprising that higher levels of administrative staffing are associated with increases in operating expenses for a hospital, considering that managers and executives are more highly paid than other front-line workers. The higher the proportion of these highly paid individuals, the greater the likelihood that operating expenses will increase overall.

Impact of Hospital Characteristics

In addition, the analysis found a strong relationship between multi-hospital system affiliation and operating expense as evidenced by the positive coefficient for that variable. This observation is a bit perplexing since one would expect the opposite effect based on a review of research from the American Hospital Association [21], suggesting that consolidation should reduce administrative costs. However, other non-labor administrative costs from a central office may be passed down to affiliate hospitals, increasing the expense burden to hospitals in this study. The potential lagged effects of diminished administrative savings from consolidation could not be considered in the one-year timeframe used in this study and merits further exploration in future work.

Interestingly, voluntary nonprofit hospitals appear to have a better grasp on operating expenses than do their proprietary competitors, noting the negative coefficient for that variable. This difference may also be attributable to the pass-through of corporate office expenses or the payment of taxes by proprietary hospitals. Those taxes would be recognized as administrative expenses in the cost report dataset.

This study is the first to examine the relationship between investments in administrative staff and profitability for hospitals. The findings are consistent with others who have examined administrative costs in total, and aligns with the idea that increased administrative expense has a limiting effect on hospital operating margins. The increase in administrative staffing for urban and for-profit facilities in this study aligns with the findings of McKay, et al [9] and Himmelstein et al [10]. The relationship between higher administrative staffing and diminished profitability is consistent with the observed inverse relationship between administrative expense and profitability in the work of Helton and colleagues [11].

Laying the Groundwork for Future Studies

One of the strengths of this study was its focus on administratively controllable factors like organization and investment in additional major lines of business, along with the reaction to managing the resources expended based on the severity of illnesses treated. It has also attempted to view ongoing operational norms in hospitals without the external influence or inclusion of one-time factors usually associated with the sale of a hospital to another party. The financial data used in the study is audited for submission to CMS and therefore has a higher degree of credibility than that from other sources, all the while encompassing a large number of facilities across the United States.

As with all research studies, this study also has limitations. The sampling criteria notably reduced the number of facilities ultimately included either by use of the case mix index variable that eliminated all critical access hospitals from the study sample or not allowing hospitals that had undergone a change in ownership or cost report filing status. Omission of specialty facilities such as surgical hospitals also reduced the sample size.  Importantly, the study uses only the latest year of data because of significant financial impacts in the preceding years arising from the COVID-19 pandemic [13].

This research study can be used as a foundation for future research studies. As more years of cost report data become available, studies examining these variables can employ a longitudinal design that can be even more informative to the field. Also, using credible data that segregates administrative labor disciplines, will allow a much finer analysis of administrative impacts that could be obscured by the fact that those disciplines are currently commingled with employees that perform billing and collection work. The strategic impact of administrators remains clouded by virtue of the available data.  Addition measures of market characteristics may be helpful to further delineate any potential benefit from administrative entrepreneurship and strategy development.

Conclusion

This is the first study to consider the relationship between the number of administrative labor hours and profitability in a hospital. Overall, the addition of non-revenue producing administrative overhead appears to burden profitability for hospitals already challenged by decreasing revenues and overall operating expense inflation. Hospital governing boards and corporate owners must be cognizant of the fixed costs that are generated by administrative staff, in view of currently declining payment rates that appear to reduce contribution margins needed to defray overhead costs.

 

References

[1]    Gooch K. 32 Hospitals, health systems cutting jobs. Becker’s Hospital Review. 2024 May 24.  Available from https://www.beckershospitalreview.com/finance/4-hospitals-health-systems-cutting-jobsjan19.html?origin=BHRSUN&utm_source=BHRSUN&utm_medium=email&utm_content=newsletter&oly_enc_id=3025H0689590I9X.

[2] Asplund, J. Carle Health to eliminate 600 jobs downstate after closing Health Alliance plans. Crain’s Chicago Business. 2025, June 6. Available from: https://www.chicagobusiness.com/health-care/carle-health-plans-lay-600-after-health-plan-closure

[3]    Gooch K.. California hospital lays off staff including executives. Becker’s Hospital CFO Report. 2023 November 21. Available from https://www.beckershospitalreview.com/finance/california-hospital-lays-of-staff-including-executives.html.

[4]    Muoio D. Forecast:  7 Immediate and Long-term Priorities for Hospital Leaders. Fierce Healthcare. 2022 December 21. Available at https://www.fiercehealthcare.com/providers/2023-forecast-7-immediate-and-long-term-priorities-hospital-leaders.

[5]    Cipriano P. Layoffs: Is the sky really falling? American Nurse.  2009 Mar.  Available from https://www.myamericannurse.com/layoffs-is-the-sky-really-falling/, March 20,2024.

[6]    Langabeer J, Lalani K, Champagne-Langabeer T, Helton J. Predicting Financial Distress in Acute Care Hospitals. Hospital Topics. 2018; 96(3): 75–79.

[7]    Langabeer J, Lalani K, Yusuf R, Helton J, Champagne-Langabeer T. Strategies of High-Performing Teaching Hospitals. Hospital Topics. 2018; 96(2):54-60.

[8]    Holmes G, Pink G. Adoption and Perceived Effectiveness of Financial Improvement Strategies in Critical Access Hospitals. Journal of Rural Health. 2011; 28(1):92-100.

[9]    McKay N, Lemak C, Lovett A, Wright R. Variations in hospital administrative costs. Journal of Healthcare Management. 2008; 53(3): 153.

[10]  Himmelstein D, Jun M, Busse R, Chevreul K, Geissler A, et al. A comparison of hospital administrative costs in eight nations: US costs exceed all others by far. Health Affairs. 2014; 33(9): 1586-94.

[11]  Helton J, Lalani K. Jennrich V, Stacey R. Don’t Let Overhead Overwhelm Your Organization. The Governance Institute’s e-Briefings. 2022; 19(4). Available at https://www.governanceinstitute.com/page/EBriefings_V19N4July2022

[12]  Centers for Medicare and Medicaid Services. Provider Reimbursement Manual, Part 2, Hospital and Hospital Health Care Complex Cost Report, Chapter 36. 2011. Available from https://www.cms.gov/regulations-and-guidance/guidance/manuals/downloads/p152_36.zip

[13]  Lalani K, Helton J, Vega F, Cardenas-Turanzas M, Champagne-Langabeer T,  Langabeer, J. The Impact of COVID-19 on Financial Performance of Largest Teaching Hospitals. Healthcare. 2023; 11.

[14]  Song non-op revenues

[15]  Kacik A. Philanthropy Growing in Importance for Nonprofit Hospitals. Modern Healthcare.  2022; 52(21): 8.

[16]  Bai G, Anderson G. A More Detailed Understanding of Factors Associated with Hospital Profitability. Health Affairs, 2016; 35(5): 889-897.

[17] Langabeer J, Helton J. Health Care Operations Management: A Systems Perspective 3rd ed. Burlington, MA:  Jones & Bartlett Learning; 2022.

[18]  Centers for Medicare and Medicaid Services.  Healthcare Cost Report Information System, Hospital 2552-10 Cost Report Data Files by Fiscal Year. 2024. Available https://www.cms.gov/httpswwwcmsgovresearch-statistics-data-and-systemsdownloadable-public-use-filescost-reportscost/hospital-2010-fy-2022.

[19]  Centers for Medicare and Medicaid Services. FY 2022 Case Mix Index files obtained from FY 24 Hospital Inpatient Prospective Payment System Proposed Rule files. 2023. Available from https://www.cms.gov/files/zip/fy2024-case-mix-index-file.zip.

[20]  Karim S, Pink G, Reiter K, Holmes G, Jones C, and Woodard E. The Effect of the Magnet Recognition[R] Signal on Hospital Reimbursement and Market Share. Nursing Economics, 2018; 36(3).

[21]  American Hospital Association. Fact Sheet – Hospital Mergers and Acquisitions Can Expand and Preserve Access to Care. Available: https://www.aha.org/fact-sheets/2023-03-16-fact-sheet-hospital-mergers-and-acquisitions-can-expand-and-preserve-access-care

 

Declaration of Conflicting Interests: The author has no conflicting interests.
Funding: There was no extramural funding received for this study.
Ethical Statement: The study used publicly available secondary data and so was deemed exempt from human subjects protection review.

 

 

 

 

 

 

 

Innovation and Adversity: The Implementation of a Unified Electronic Health Record for U.S. Veterans

(Stanford Graduate School of Business Case SM-402, May 7, 2025; 24 pages)

Authors: Priti Lakhani, John Beckett, David Brancato, and Kevin A. Schulman

Introduction:In the predawn stillness of a crisp Washington morning, Dr. Priya LaManna confronted a digital deluge that threatened to overwhelm her nascent leadership role. As the newly appointed director of quality, safety, and value for the electronic health record modernization at the U.S. Department of Veterans Affairs (VA), she found herself at the epicenter of a complex organizational transformation that was evaluating the boundaries of technological implementation and institutional change management.

The electronic health record (EHR) “go-live” in Spokane, Washington, had ignited a firestorm of criticism that now populated her inbox with a torrent of increasingly hostile communications. Each email represented not just a complaint, but a potential inflection in a multibillion-dollar modernization initiative that carried profound implications for health care delivery for America’s veterans.

LaManna’s professional trajectory had uniquely positioned her for this pivotal moment. Her career had traversed the complex landscape of health care technology, including a significant tenure at Cerner, where she had led the Patient Safety Council, and a subsequent role as chief medical information officer at a prominent academic medical center. Her dual expertise in patient safety and medical informatics made her an ideal candidate to navigate the turbulent waters of this massive technological integration.

The initial EHR implementation at the VA had faced such a public outcry that Congress had called for a “reset” before progressing. LaManna took a deep breath, took a sip of her mint tea, and considered how to tackle the emails and never-ending calls regarding safety issues, workforce dissatisfaction, and congressional inquiries.

Download the case here. For inquiries, contact Kevin Schulman: kevin.schulman@stanford.edu

Lauren Bilbo, Elman Amador Medina, and Klara Klarowicz, Graduate School of Business, Stanford University; Dashiell Miner and David Scheinker, School of Medicine, Stanford University; Stefanos Zenios, Graduate School of Business, Stanford University; and Kevin Schulman, School of Medicine and Graduate School of Business, Stanford University

Contact: kevin.schulman@stanford.edu

Support: The Ludy Family Foundation, The Hirsch Family Foundation, The Mindshare Institute, and the Government, Business and Society of the Graduate School of Business, Stanford University.

Introduction

There is an administrative-cost crisis in U.S. healthcare. Administrative costs are estimated to be at about 34.2% of total U.S. healthcare spending [1]. The United States spends about 10 times more on average on healthcare administrative expenses than any other OECD country (2). The result is that the U.S. spends more to manage the healthcare system than France, Germany and Italy combined spend on healthcare services. (3) In our multi-payer system, an estimated two-thirds of administrative spending is likely related to transactions or billing costs [4,5]. Administrative costs have been estimated to include $374 billion in waste in 2025 spending, although the data are not accurately tracked. (6,7,8)

The primary challenge in the market is that numerous components of an analog transaction system have been digitized separately, leaving the U.S. health sector without a uniform digital transaction platform. Each health insurance carrier must develop a contract with each in-network provider to describe the terms of how transactions will be documented and paid for by each patient covered through their plan.

With over 1,000 health insurance carriers offering 318,000 distinct health plan products in the U.S. market, this process is fraught with duplication. Each carrier develops its own contract terms such as payment schedules, documentation standards, and prior authorization processes. They can market multiple types of products, HMOs, PPOs, high-deductible health plans, and each plan can have their own set of in-network providers. The complexity of this process contributes to the high billing-related costs in the U.S. market (3,9-11). A large fraction of the transaction spend is due to transaction friction: interpreting heterogeneous payer-provider contracts, manual prior authorization (PA), serial claim denials/resubmissions, and reconciliation of remittance advice idiosyncrasies (8,9). Beyond direct transaction costs, both payers and providers must maintain substantial IT infrastructure and financial staff to support bespoke claims processing software, with ongoing overhead for tracking and implementing frequent changes to contract terms and coverage policies (12).

One pathway to reduce these administrative costs is to develop a true digital transaction standard for the market. (4,12,13) Standardizing the adjudication process is one important step towards reducing administrative costs in the healthcare industry. This would start with the development of digital contracts which could provide computable transactions for payers and providers. Such an effort would greatly simplify the transaction process. For payers and providers who seek novel transaction models, computable contracts would allow for such complexity but would use an algorithmic model for these transactions. By converting contract terms into executable logic, the computable framework inherently enables validation and systematic updating of coverage rules through structured testing against real-world claims data, reducing ancillary support costs. Financial modeling based on empirical data and mathematical simulation suggests that standardization has the potential to save over 60% of administrative costs – over $1 trillion dollars in 2025 and more than the potential savings of a single-payer system such as Medicare For All. (12,13)

Computable contracts would have specific contract features reduced to computable terms. Making contract terms computable would require identifying standards for moving the contract language into an algorithm. For example, SMART on FHIR combines a standards-based data layer with the FHIR API. (14) This architecture is based on open standards including HL7’s FHIR, OAuth2, and OpenID Connect. (14)

One computable contract framework has been publicly reported, the open-source PATIENTS framework. (15) This effort has developed standard modular contracts, an open payment system, a computable contract library, universal clinical coverage, standard plan mapping, and open transaction rails. (15)

We sought to extend this work to real-world payer-provider contracts. The PATIENTS framework proposes an idealized computable contract structure for optimal market conditions. Our work differs by testing whether existing data standards can accommodate actual commercial payer-provider contract language used in the market, and by addressing the incentive structures that have impeded adoption of standardized transaction models. Using the PATIENTS framework as a reference and contract language from a large healthcare system, we assessed the data standards available to develop a computable contract examining each term in the contract. This work stands as an assessment of the feasibility of this approach.

Semantic Ambiguity in Source Artifacts

Three partially overlapping artifacts govern most commercial medical claims today:

1. The executed agreement / contract (legal and financial terms)
2. The provider manual (procedural, operational, and utilization policy guidance, often narrative and line-of-business specific)
3. The X12 companion guide (transaction formatting deviations, situational usage rules)

Critical adjudication logic (e.g., PA triggers, frequency limits, “lesser of” pricing precedence, quality modifier eligibility) frequently resides only as prose scattered across these documents. Each payer’s unique blend of a legal contract, provider manual, and X12 companion guide creates a bespoke semantic layer that revenue-cycle staff, clearinghouses, and claims systems must continually re-interpret. Furthermore, frequent changes in the governing artifacts necessitate frequent, nuanced reinterpretation of coverage rules.

The absence of a machine-readable, authoritative layer means every actor in the market rebuilds the same logic (clearinghouse edits, internal rules engines, third‑party analytic overlays), propagating duplicated effort and inconsistency.

Divergent Incentives Lead to Divergent Technical Paradigms

Financial Incentives

At the point of service, providers generally maximize margin and operational efficiency when payment outcomes are predictable: given a set of clinical facts and billing codes, they want to know ex ante whether a claim will be paid, for what amount, and on what timeline. Determinism lets providers accurately staff clinical services, reduces working capital tied up in AR, and avoids costly billing rework cycles.

Payers, by contrast, benefit from ambiguity, vague coverage language, and discretionary prior‑authorization criteria. Multi‑stage denial rationales prolong decision windows, enable selective enforcement of payment rules, and increase the probability that some claims are abandoned or down‑coded. Opacity acts as a frictional filter on total paid claims expense, making it possible for payers to utilize administrative complexity as a deliberate tool to support financial management of the benefit pool, both in terms of amount and timing of payments. In this environment, procedural friction functions as implicit policy, determining financial outcomes through attrition and exhaustion rather than clinical or contractual merit. (16)

Resulting Preference Split

These underlying incentives manifest as two preferred technical archetypes:

• Provider‑preferred architecture: Transparent, rules‑based determinism. Explicit, human‑readable clauses compiled into executable logic (e.g., CQL/FHIRPath/DSL). Each decision produces a traceable proof: inputs → rule evaluations → allowance formula → outcome. Variance is eliminated; disagreements become focused on the adequacy of data, not on hidden logic.
• Payer‑preferred (status quo) architecture: Opaque, probabilistic discretion. Black‑box AI / heuristic scoring layers that classify claims or PA requests into “pay,” “pend,” or “deny” cohorts based on patterns in historical data rather than strictly enumerated contract rules. The internal model weights are not disclosed; justification can be post‑hoc, preserving flexibility to modulate denial intensity.

Technical Implications

The divergence in incentives between providers and payers results in diverging technical goals, preferences, and approaches, as shown in the table below.

Dimension	Rules-Based, Deterministic (Provider-preferred)	AI / ML, Probabilistic (Payer-preferred)
Primary Objective	Predictable cash flow & minimized rework	Reduced aggregate payout & increased denial leverage
Representation	Declarative rule sets mapped to standard terminologies; versioned & diff‑able	Proprietary model artifacts (weights, feature sets for algorithms), periodic validation and retraining cycles
Explanation	Intrinsic: rule execution trace is the explanation	Extrinsic: generated rationale templates; limited transparency
Governance Load	Up-front modeling & normalization effort; low runtime ambiguity	Lower up-front specification; ongoing tuning to maintain denial yield
Error Surface	Missing or stale rule → identifiable gap	Drift, bias, adversarial gaming; harder to contest individual outputs, such as clinical/administrative decisions
Provider Appeal Path	Challenge specific rule or input datum	Demand model transparency; since model explainability is often infeasible, fall back on manual override channels

 

Comparison of Technical Approaches

Neither the provider approach nor the payer approach is inherently “correct”. Rather, each technical approach offers unique advantages and disadvantages, as described in the table below.

Dimension	Rules-Based, Deterministic	AI / ML, Probabilistic
Primary Use	Coverage, PA, pricing, denials (explicit logic)	Pattern anomaly detection (e.g. detection of potentially inappropriate care), document parsing, data extraction; used to inform clinical decision makers when denials are initially recommended
Strengths	Transparency, auditability, low variance	Adaptable to noisy inputs, learns from historical denials/fraud patterns
Weaknesses	Brittle if inputs missing; manual rule authoring	Opaqueness, bias risk, harder to make legally binding
Failure Mode	Silent gaps, unhandled edge cases	False positives/negatives; adversarial gaming
Governance Fit	High (directly traceable to contract)	Low for binding decisions (should be advisory); suitable only as a decision-support tool for clinicians and administrators

 

Systemic Consequence

The coexistence of these paradigms produces a structural stalemate: providers escalate investment in contract “translation” layers to re-impose determinism locally; payers escalate ML-driven utilization management to retain flexibility. Neither approach reduces inherent administrative complexity.

Regulators have begun to intervene in this stalemate. In January 2024, the Centers for Medicare & Medicaid Services (CMS) issued a final rule for Medicare Advantage requiring that medical necessity determinations be “based on the circumstances of the specific individual…as opposed to using an algorithm or software that doesn’t account for an individual’s circumstances.” (17,18) The rule mandates physician review of determinations and public disclosure of evidence supporting algorithmic criteria. These requirements effectively limit the black-box AI approaches favored in the payer-preferred architecture, while remaining compatible with the rule-based determinism of computable contracts. This regulatory shift creates both pressure and a pathway for industry transformation.

A computable modular contract explicitly realigns both sides on a deterministic core for enforceable payment logic while confining probabilistic methods to assistive roles (e.g., anomaly detection, document extraction) that do not unilaterally affect payment without a corresponding explicit rule. The computable contract core serves as the foundation for more effective AI/ML deployment and agentic AI workflows: rather than training models to reverse-engineer opaque contract language, machine learning can focus on higher-value tasks like fraud pattern detection, care pathway optimization, and predictive risk stratification. These are solutions that can be built from the structured data that result from computable transactions (some of these elements can be built into the transaction such as fraud detection, some will use the data to build new applications to support and evaluate clinical care). AI becomes dramatically more accurate, useful and economically efficient when operating on clean, structured, deterministic rule sets (a novel optimized process) rather than parsing inconsistent natural language across thousands of contract variations (an inefficient “legacy” process). (17)

This approach aligns with recent CMS requirements for Medicare Advantage that mandate physician review of coverage determinations and prohibit purely algorithmic denials (18,19). Further, it creates a pathway to reconcile incentive divergence: payers gain fraud and leakage controls plus faster legitimate adjudication; providers gain predictability, shifting competition toward efficiency and quality rather than exploitation of ambiguity. These results are included as an appendix to this paper.

Conclusion

Overall, we found that data standards exist for almost all contract terms in payer-provider contracts. We were able to develop a consistent structure and typography while preserving the original substance of the contract templates we identified. This effort consolidates requirements across data standards, rules computability, and operational workflows.

Computable contracts could provide a pathway to the development of true digital transaction standards in healthcare and agentic AI workflows, with significant financial benefits from improving transaction efficiency. Such a set of transaction standards could also open a pathway for entry of novel insurance solutions into the healthcare market. (20,21)

Appendix

Technical Requirements for a Computable Payer–Provider Contract Digital Adjudication Platform

 

References

1. David U. Himmelstein, Terry Campbell, Steffie Woolhandler.Health Care Administrative Costs in the United States and Canada, 2017. Ann Intern Med.2020;172:134-142.
2. Turner A, Miller G, Lowry E. High US health care spending: where is it all going? Published October 4, 2023. Accessed December 20, 2025. https://www.commonwealthfund.org/publications/issuebriefs/2023/oct/high-us-health-care-spendingwhere-is-it-all-going
3. Schulman KA, Nielsen PK Jr, Patel K. AI Alone Will Not Reduce the Administrative Burden of Health Care. JAMA. 2023 Dec 12;330(22):2159-2160.
4. Istvan B, Schulman KA, Zenios S. Addressing Health Care’s Administrative Cost Crisis. JAMA. 2025 Mar 4;333(9):749-750.
5. Health Affairs. The Role Of Administrative Waste In Excess US Health Spending. October 6, 2022. https://www.healthaffairs.org/content/briefs/role-administrative-waste-excess-us-health-spending
6. Sahni  NR, Carrus  B, Cutler  DM. Administrative simplification and the potential for saving a quarter-trillion dollars in health care. JAMA. 2021;326(17):1677–1678.
7. Shrank WH, Rogstad TL, Parekh N. Waste in the US Health Care System: Estimated Costs and Potential for Savings. JAMA.2019;322(15):1501–1509 (waste as a percent was updated to 2025 National Health Expenditures).
8. Sahni NR, Istvan B, Bello Thornhill H, Joynt-Maddox KE, Cutler D, Emanuel EJ. Availability of consistent, reliable, and actionable public data on US hospital administrative expenses. Health Aff Sch. 2025 May 22;3(5):qxaf069.
9. Tseng, P., Kaplan, R.S., Richman, B.D., Shah, M.A. and Schulman, K.A., 2018. Administrative costs associated with physician billing and insurance-related activities at an academic health care system. Jama, 319(7), pp.691-697.
10. National Association of Insurance Commissioners. U.S. Health Insurance Industry: 2022 Annual Results. Published 2023. Accessed February 15, 2026. https://content.naic.org/sites/default/files/industry-analysis-report-2022-health.pdf
11. Sahni NR, Istvan B, Stafford C, Cutler D. Perceptions of prior authorization burden and solutions. Health Aff Sch. 2024 Aug 6;2(9):qxae096.
12. David Scheinker, Kevin Schulman, and Stefanos Zenios. It’s Time To Bring Health Care Systems Into the Digital Age | Opinion. Newsweek April 09, 2025. https://www.newsweek.com/its-time-bring-health-care-systems-digital-age-opinion-2056429
13. Scheinker, D., Richman, B.D., Milstein, A. and Schulman, K.A., 2021. Reducing administrative costs in US health care: Assessing single payer and its alternatives. Health Services Management Research, 34(3), pp.153-158
14. https://smarthealthit.org/smart-on-fhir-api/
15. https://open.turquoise.health/docs/getting-started/introduction/
16. Herd, P., & Moynihan, D. P. (2018). Administrative burden: Policymaking by other means. Russell Sage Foundation.
17. Oliver M., Faris R. A Blueprint for Enterprise-Wide Agentic AI Transformation. HBR. February 12, 2026. https://hbr.org/sponsored/2026/02/a-blueprint-for-enterprise-wide-agentic-ai-transformation. Accessed Feb 17, 2026.
18. Mello MM, Rose S. Denial—Artificial Intelligence Tools and Health Insurance Coverage Decisions. JAMA Health Forum. 2024;5(3):e240622.
19. CMS. CMS Interoperability and Prior Authorization Final Rule (CMS-0057-F). January 17, 2024. https://www.cms.gov/cms-interoperability-and-prior-authorization-final-rule-cms-0057-f
20. Brooke Istvan, Perry Nielsen Jr, Megan Eluhu, Bryan Kozin, Walt Winslow, David Scheinker, Kavita Patel, Kenneth Favaro, Stefanos Zenios, Kevin Schulman. 2024. Applying Precedents Thinking to the Intractable Problem of Transaction Costs in Healthcare. Health Management, Policy and Innovation (www.HMPI.org). Volume 9, Issue 3.
21. Cutler  DM. Reducing administrative costs in US health care. The Hamilton Project. Accessed July 14, 2024. https://www.hamiltonproject.org/publication/policy-proposal/reducing-administrative-costs-in-u-s-health-care/

 

Karoline Mortensen, Steven G. Ullmann, Richard Westlund, University of Miami Herbert Business School

Contact: sullmann@bus.miami.edu

With the U.S. healthcare system facing a wide range of challenges, “Navigating Change” was the focus of the 15^th annual Business of Health Care Conference hosted by Miami Herbert Business School of the University of Miami. More than 850 health care professionals, business executives, and students registered for the February 6, 2026 conference, which was held on the Coral Gables campus and live-streamed globally.

In a wide-ranging panel discussion, eight leaders in key healthcare sectors focused on the changing national political climate, the explosion of artificial intelligence (AI) tools, and the formidable cost and affordability challenges facing payers, providers and patients.

Brian D. Pieninck, president and CEO, GuideWell Mutual Holding Corporation, moderated the panel discussion with Anders Gilberg, SVP, Government Affairs, Medical Group Management Association; C. Ann Jordan, J.D., president and CEO, Healthcare Financial Management Association (HFMA); Lisa Kidder Hrobsky, SVP for advocacy, federal relations, and political affairs, American Hospital Association (AHA); Jennifer Mensik Kennedy, R.N., president, American Nurses Association (ANA); Roger A. Mitchell, Jr., M.D., president, National Medical Association (NMA); Bruce A. Scott, M.D., immediate past president, American Medical Association (AMA); Mike Tuffin, president and CEO, America’s Health Insurance Plans (AHIP); and Stephen J. Ubl, president and CEO, Pharmaceutical Research and Manufacturers of America (PhRMA).

Pieninck began the discussion by noting the “incredible need and incredible opportunity” to build a common sense of purpose in the $5.3 trillion healthcare industry. “We are spending the most on both an aggregate and per-capita basis, but we are not the healthiest nation in the world,” he said, adding that demand for care will continue to grow due to the nation’s aging population. “We need to find ways to create efficiencies, such as the introduction of AI.”

Advancing AI

Because the business of healthcare begins with human interactions, Pieninck asked Kennedy how frontline nurses and clinicians feel about AI. “People are excited and wary,” she said. “We have seen a lot of technology come to nurses without their involvement, so we need to have nurses and clinicians helping to design these tools,” she said. For instance, nurses are working with Columbia University on having an AI model read nursing documentation for key trends in a patient’s health.

Physicians are also excited and scared about AI, according to the AMA’s Scott, who called it “augmented intelligence” to emphasize the human component. “Electronic medical record (EMR) systems were developed without physician input, so this time we need to be sure it works in our environment,” he added. “The AMA has a new center for digital health, and we need to get AI right so it doesn’t become a problem.”

The insurance sector is optimistic about AI’s ability to boost productivity, said Tuffin, AHIP’s president. “We see AI as the opportunity to slay the administrative dragon,” he said. “So much of our healthcare system runs on manual processes or systems that don’t talk to each other, adding cost and complexity. Let’s ban the fax machines and connect to an AI-driven system.”

From a pharmaceutical perspective, AI is “super exciting” because it can reduce the risks associated with drug development, and accelerate companies’ ability to bring new therapies to market, said Ubl, PhRMA’s president. “We will have more and better medicines that are more affordable, managing chronic health risks, and keeping patients out of the hospital systems.”

AI challenges

For the healthcare industry to unleash the power of AI, a number of challenges must be addressed, said several panelists. “AI can be a wonderful enabler, but we still need to resolve fundamental technology issues, such as transmitting clinical information from one practice to another,” said Gilberg with the Medical Group Management Association. He added that AI is now embedded in basic business functions, such as phone and video applications, but more standards are needed to address higher-level healthcare functions.

Mitchell, president of the National Medical Association, which represents Black physicians, added that AI tools are only as good as their training on large learning modules (LLMs). If the underlying data or the training process don’t reflect the nation’s varied demographic communities, then the results will be biased or “monolithic” rather than personalized to individual patients. As an example, he cited the “old trope” about Black-White racial differences in kidney functions due to faulty assumptions about muscle mass and creatinine production. He also commented that AI tools must be equitably distributed across all communities in the U.S.

Kidder Hrobsky with the AHA agreed about the importance of a level playing field in AI deployments, noting that many small and rural hospitals lack the financial resources or expertise to implement these new tools. “Currently, hospitals are paying more for labor, while being underpaid by the federal government,” she said. “They are simply trying to stay afloat.”

Defining and measuring the quality of care is another challenging AI issue, said Scott. He noted that the AMA’s specialist societies are working on potential solutions, and that the insurance industry favors better metrics for quality care. However, the American “culture of freedom” means that, while the medical profession may be working to attain improved health outcomes, individuals still smoke, turn down vaccines, and ride motorcycles without helmets, increasing their personal health risks.

Asked about the challenges that AI can’t solve, Jordan with the HFMA cited the revenue cycle for providers, noting the fragmented nature of the healthcare system. On a different note, the AHA’s Kidder Hrobsky said AI won’t change ingrained patient behavior, adding, “Some people will still go to the emergency room, regardless of what AI says.”

People need to be able to override AI output, making clinical judgments, said Kennedy. She added it would be a mistake to replace nurses with AI tools, simply to save labor costs. “Far better for AI to allow nurses to spend more time with patients and families, building that human connection,” she said.

Technology investments in health care are driven by the profit motive, so money to fund AI tools will need to come from somewhere in the system, said Scott.  Following up on that comment, Tuffin, with AHIP, said there is potential to divert some resources now going to chronic disease care “downstream” to improve the social determinants of health, such as nutrition, fitness, and environmental conditions.

Currently, there is no tracking of the different forces in the healthcare system, said Jordan, the HFMA president. Even with technology advances, it is difficult to understand the variables that affect costs and outcomes. “Applying AI to the old system won’t do us any favors,” she said. “Rather than apply new tools to old problems, we need to build something better.”

Along the same lines, Kennedy added, “We are tinkering on the edges of the system. I think the tech companies may create a different system that gives consumers what they want, while we are still figuring things out.”

Cost concerns

Lowering healthcare inflation is a massive financial challenge, said Pieninck, who emphasized the importance of collaboration in finding solutions. One opportunity would be moving away from fee-for-service models and aligning financial incentives around patient outcomes, said several panelists. However, that would require a large-scale systemic change.

The HFMA’s Gilberg noted that many vendors in the private sector exist only because of inefficiencies in the system and constitute a significant part of the industry. Hospitals are institutions that play essential roles in their communities as employers and providers of acute and emergency services, said Kidder Hrobsky. In the ongoing transition to outpatient care, ways need to be found to keep these hospitals financially solvent.

Pharmacy costs average about 14 percent of the health care dollar, according to Ubl.  But generic medications and low copays typically reduce prices for consumers. “We get 90 percent of the political debate, but are still a modest part of the system, and the only segment where prices go down,” he said.

While there is no greater return on investment than preventive care, it is consistently the most underfunded segment of the industry, added Scott, the AMA’s past-president.

Several panelists said that reducing the administrative burden for providers can lead to cost savings. Tuffin estimated that about 25 percent of the $5.3 trillion overall cost is wasteful or harmful. He cited the time-consuming system of obtaining prior authorization from insurers for a patient’s treatment or prescription medication. While patient safeguards are needed, the process should be faster and simpler. One step forward would be to use consistent terminology and approval-rejection decisions, added Jordan. “We need to look at incremental changes to prior authorization and see what we can learn from other industries, such as financial services, to get out of the way of consumers.”

Price transparency

Price transparency is a worthy goal, but difficult to achieve because of the many providers who deliver patient care, including hospitals, physicians, therapists, and ancillary services. “It’s difficult for patients to understand their copays and responsibilities, and I don’t think anyone makes choices solely on cost alone,” said Kennedy.

Mitchell, the president of the NMA, added that variability in hospital costs and reimbursement sources is a big issue affecting patient bills. However, Tuffin noted that insurers often provide cost comparison tools for consumers as prices for services, such as imaging, can vary greatly in the same community.

Price transparency is also an issue for providers, and other participants. For instance, Ubl noted that inefficiencies in the pharma supply chain can lead to wide variations in consumer pricing. He said the industry is launching a consumer platform similar to the TrumpRX site, which contains pricing information for certain drugs.

Incentivizing patients

One key to reducing costs and improving outcomes would be to provide more incentives for patients to improve their own health. Mitchell called for the states to use incoming federal dollars to prioritize an “upstream” approach to care by helping individuals address the social determinants of health, while Kennedy said shifting more nurses from hospitals into primary care clinics could help individuals and families enhance their personal health.

However, the AMA’s Scott cited the increased flow of misinformation on vaccines and other kinds of preventive care as a public health challenge. Mitchell agreed, noting that many Americans are distrustful or fearful of seeing a doctor. “We need to address that issue on a personal level, and teach those interactions in medical and nursing schools,” he added. “The journey to health is a partnership between the physician and patient.”

One Big Bill

The panelists had different perspectives on the One, Big, Beautiful Bill Act signed by President Trump on July 4, 2025.  Nearly all the panelists noted its deep impact on the Medicaid population, such as increased work requirements, as well as the expiration of the enhanced premium tax credits for Affordable Care Act subsidies that made it affordable for many individuals to purchase health insurance.

While work requirements are popular politically, many Medicaid recipients are already working, said Scott. It would be a better policy to help individuals from falling through the administrative cracks in the program. Tuffin agreed, adding that the insurance industry would be working on the state level to help recipients maintain their coverage.

Both Jordan and Kidder Hrobsky said the Act will make it more difficult for rural hospitals to survive the expected loss of reimbursements, while other panelists pointed to restrictions on educational loans that could lead to fewer nurses, therapists and physicians in the future.

Several panelists pointed to the need to educate policymakers in Washington and state capitols about how the healthcare industry works. “Civic engagement is so important,” said Mitchell. “As a forensic pathologist, I see failed policies every day on the autopsy table.”

The next pandemic?

While government and industry stakeholders worked intensively together to develop effective COVID vaccines in 2020, a repeat of that collaboration in the face of a future pandemic is unlikely, according to participants. Trust in science and medical professionals has been eroded, and the federal government is questioning the value of current vaccines.

“We have left the World Health Organization, so global surveillance has been lost,” said Mitchell. “And we are facing other epidemics right now, like adolescent homicides, maternal deaths, and breast and prostate cancer. We are going down a path that will lead to more disease and deaths in the future.”

Reasons for hope

Despite the many challenges facing the industry, the panelists offered several reasons for hope. “We are entering the golden years of medicine with new curative therapies as long as we continue to nurture and invest in our ecosystem,” said Ubl, the PhRMA president.

While nothing will replace the healing touch of a doctor or nurse, AI will offer caregivers more tools to help their patients, added Scott.  “We need to reach out and personally engage with others in our communities to solve these issues.”

 

Anna-Lena Brecher and Lena Kraft, Hannover Medical School; Linea Schmidt, University of Potsdam, Icahn School of Medicine at Mount Sinai, and German Society of Digital Medicine; Anne Sophie Platzbecker, University of Potsdam, German Society of Digital Medicine, Dresden University of Technology, Medical University Lausitz – Carl Thiem; Ariel D. Stern, University of Potsdam and Icahn School of Medicine at Mount Sinai; and Volker E. Amelung, Hannover Medical School and Private Institute for Applied Health Services Research

Contact: amelung.volker@mh-hannover.de

Digitizing Health – Between Policy Vision and System Reality

The U.S, Food and Drug Administration (FDA) was among the first regulatory authorities to systematically address the evaluation and approval of digital health applications. Initial guidance for assessing mobile medical applications was published in 2011, with a finalized version released in 2013. This regulatory framework set the stage for the approval of innovative digital therapies that, in terms of efficacy, are comparable to conventional medicines [1]. Such tools are now commonly referred to as “Digital Therapeutics,” or “DTx,” and are defined as “evidence-based therapeutic interventions driven by software to prevent, manage, or treat a medical disorder or disease” [2]. A notable example is the reSET app from Pear Therapeutics, cleared by the FDA in 2017 as the first DTx using cognitive behavioral therapy for the treatment of substance use disorder [3].

In 2020, Germany introduced its own comprehensive strategy for digital health applications  (Digitale Gesundheitsanwendungen, or “DiGA”) following the German Parliament’s passage of the Digital Healthcare Act on November 7, 2019. This legislation established Germany as the first country worldwide to formally reimburse physician-prescribed digital health applications under its mandatory, statutory health insurance framework.

Germany’s equivalent of the FDA, the Federal Institute for Drugs and Medical Devices ((Bundesinstitut für Arzneimittel und Medizinprodukte, or BfArM), does not have the same influence on the medical device market as its U.S. counterpart. The BfArM does not determine which applications are approved for distribution in the medical device sector or which applications are validated as medical devices; these decisions are made by a decentralized group of European Notified Bodies, independent organizations responsible for product review.

Furthermore, unlike in the United States, where regulatory and reimbursement decisions are based on separate processes, Germany’s digital health applications, or DiGA, are reimbursed by health insurers. Jens Spahn, Federal Minister of Health from 2018 to 2021, emphasized the importance of a systematic testing process to ensure the quality of digital applications, underlining that the system should reimburse only those tools demonstrating evidence of clinical value. This approach was designed to bolster Germany’s digital healthcare market while ensuring the quality of care. Indeed, the guiding philosophy behind the launch and subsequent growth of DiGA was to support both the digitalization of the German healthcare system and the improvement of care. The enactment of the Digital Healthcare Act provided a functioning regulatory framework within a matter of months, establishing a fast-track process for evaluating DiGA, a collaboration between the application manufacturer and BfArM; provisional approval to facilitate a trial evaluation, and a pathway to price negotiations (see Figure 1).

Moreover, the BfArM emphasizes the importance of a systematic approach based on rigorous evidence for identifying applications that are fit for incorporation into standard care. The BfArM’s fast-track process tests applications for safety, functional suitability, data protection, and proven positive effects on care [4]. Beyond regulatory approval and pricing, the true test for reimbursable digital health applications lies in their real-world applicability within Germany’s multifaceted care structures. A core component of the DiGA system is its seamless integration into existing care infrastructure: applications are not intended to be standalone solutions, but to complement and enhance existing treatment processes.

At first glance, around six years after the passage of the DVG and over five years after the launch of the first DiGA, it may seem like this system is working. While other countries such as the United States are still searching for suitable authorization and reimbursement mechanisms for digital health applications, Germany has established an innovative, legally enshrined procedure that systematically evaluates and integrates digital applications into the standard of care. Indeed, the French PECAN system, which was launched in March of 2024, borrows much of its design from the DiGA fast-track process. Other European countries like Belgium are in the process of testing similar digital therapeutic reimbursement pathways.

However, the reality is much more complex. On the one hand, the provisional authorization of digital health applications for an initial period of one year continues to represent a logical approach: it allows rapid access to therapy and provides manufacturers time to provide the necessary evidence. In practice, however, many manufacturers, particularly those with limited financial resources, are burdened by the “double regulation” and additional bureaucracy associated with submissions for both provisional approval and then permanent approval of a new application. This raises fundamental questions about the practicality and sustainability of a process that intends to be start-up-friendly.

Concurrently, the DiGA experience to date establishes an important foundation for system improvements, among them: supporting manufacturers with more specialized assistance in generating real-world evidence outside of traditional clinical trials and using data such as medical claims, electronic health records, and app metadata; establishing a more robust connection to care pathways; and adopting a more timely approach toward realistic pricing models. The fast-track process could be expanded further, for example through graduated approval models currently used for classifying medical devices. These steps would improve the regulatory fit between application risk, evidence, and healthcare relevance.

To maintain momentum and avoid stagnation, the framework for reimbursable digital health applications needs to transition from a product-centric focus to that of an evolving system. The next phase, “DiGA 2.0,” must therefore concentrate on enhancing problem-solving capabilities, ensuring usability in low-threshold care settings, and providing transitional support until the product is integrated into the market – all while ensuring that the German insurance system continues to reimburse only those tools that truly create value for patients and the healthcare system overall. This article examines how reimbursable digital health applications can be further developed to align with the overarching goal of patient-centered care.

Legal Framework and Political Course Setting

DiGA are risk class I, IIa or IIb medical devices (as defined by the EU Medical Device Regulation) that are primarily based on digital technology and which carry the European Conformity safety certification. Review and listing follows the fast-track process, the only option for risk class I or IIa medical devices, and takes three months from the time manufacturers submit their application. Following the same evidence requirements, risk class IIb medical devices are eligible only for permanent DiGA listing, without the fast-track process. In addition to information such as data protection or product characteristics, the fast-track process (see Figure 1) requires the respective manufacturers to submit quantitative comparative studies that provide evidence of the DiGA’s positive health effect.

Figure 1: The DiGA Fast-Track Process

Note: This figure only holds for medical devices of risk class I and IIa. Source: Own illustration based on [4]

 

Following receipt of a complete application, the BfArM will verify within three months whether the digital health application meets the required criteria. If the evidence is incomplete, the application can be provisionally listed in the DiGA directory for up to 12 months – with the possibility for a trial period extension of up to two years – while the manufacturer provides the necessary studies. If the manufacturer can demonstrate so-called positive care effects within this period, the DiGA will be permanently listed. If the required evidence is not provided within this period, it will be removed from the DiGA directory’s approved list [4, 5].

While this approach offers DiGA a structured entry into the healthcare system, it also exposes tensions between early-market access and insufficient medium-term evidence. As market access is often the main concern for digital and other products (rather than market exit in the event of failure), systematically removing applications that cannot survive on the market or do not meet requirements after the review process has been completed, represents a practical challenge. Yet doing so is of the utmost importance to the integrity of the DiGA system, ensuring that no applications remain on the market if they lack evidence of clear benefit.

The DiGA fast-track process has been hailed as a breakthrough both nationally and internationally, particularly because of the accelerated approach and the low-risk classification specific to DiGA [6]. The determination and negotiation of DiGA pricing follows a clearly delineated process (see Figure 2). During the first year that an application is included in the DiGA directory, the provisional manufacturer’s price applies. The renegotiated reimbursement amount takes effect starting the 13th month. The subsequent permanent price is renegotiated between the manufacturers and the National Association of Statutory Health Insurance Funds (GKV-Spitzenverband). The BfArM informs the GKV-Spitzenverband about the price negotiation to be included in the directory [4].

Figure 2: Process of price determination and negotiation for DiGA

Source: Own illustration based on [7]

From Theory to Practice

Germany’s DiGA framework is entering its sixth year, allowing for an initial assessment of progress to date and starting points for discussion about the system’s further development. Some regulations were updated recently and are not reflected in the past data [8].

Figure 3: Evolution of the number of listed DiGA, as of October 31, 2025

Source: Own illustration based on [8]

 

From the initial inclusion of the first digital health application in the directory in September 2020 to December 31, 2024, a total of 861,000 DiGA prescriptions were filled by insured persons at least once. The number of activation codes redeemed per quarter (a common measure of utilization employed in the German healthcare system) has risen steadily. However, there are considerable DiGA utilization disparities. Concurrently, the number of listed DiGA has also increased steadily over time, growing from 10 DiGA in the fourth quarter of 2020 to 56 DiGA at the beginning of 2024. Since then, the number of DiGA in the directory has plateaued (see Fig. 3).

The majority of listed DiGA (51%) relate to mental and behavioral health (see Fig. 4). In second place, with around 12% of approved DiGA, are applications for musculoskeletal disorders, followed by metabolic diseases with around 11%. Seven other indication areas are also covered [8].

Figure 4: Applications listed in the DiGA-directory by indication area, as of October 31, 2025

Note: Assignment to multiple categories possible. Source: Own illustration based on [9]

 

A total of 73 digital health applications have been listed in the DiGA directory since its inception. Of these, 47 applications (64%) were permanently included after the manufacturers demonstrated positive care effects; 16 digital health apps have been removed from the directory due to an inability to prove benefit. The remaining 10 digital health applications are currently listed provisionally while trials are conducted to prove their benefit to patients and care processes. According to the GKV-Spitzenverband, only every second DiGA that was provisionally included in the health insurance catalogue for testing, achieved the promised benefit over the period of up to 24 months [8].

In the first year after launch, the price of a DiGA is set by its manufacturer, considering various costs associated with R&D such as development or clinical trials, and are capped by a maximum amount regulation. The manufacturers’ quarterly prices (90-day course of treatment) currently range from 119 to 952 euros at launch (approx. US $140 to US $1,121). In 2020, the average price was still 411 euros (approx. US $484), by 2024 it had increased to nearly 500 euros (approx. US $589). Following the permanent inclusion of DiGA in the directory, its manufacturer and the GKV-Spitzenverband negotiate the price that will apply beginning with the 13th month. The average negotiated price per 90-day course of treatment is 266 euros (approx. US $313), and negotiated prices range from 189 to 248 euros per quarter (approx. US $223 to US $292) [10].

Barriers and Bottlenecks

Despite Germany’s clear framework for the authorization and reimbursement of evidence-based digital health applications, stakeholder acceptance has been a challenge. In recent years, numerous studies have shown that some physicians are skeptical about digital therapeutics and this new approach to healthcare delivery. Primary areas of concern are the scientific evidence for effectiveness, the potential DiGA-related increase in physician workload, uncertainties regarding legal conditions and data protection, and the view that the remuneration of clinician services is insufficient compared to the perceived high costs of digital health applications [10, 11].

Figure 5: Pathways for patients to access a DiGA

Source: Own illustration based on [12]

 

Potentially the biggest barrier, even among doctors who generally agree with the DiGA concept, is a lack of knowledge about which DiGA are available, how they can be prescribed, and for which patients they are suitable. In many respects, this should not surprise us: Germany has the oldest physician population in Europe, many of whom are not digital natives [13]. Furthermore, the absence of physician digital education during medical training and its ongoing absence as a requirement in continuing medical education, points to a persistent digital literacy gap. In practice, only a small proportion of doctors regularly prescribe DiGA and plan to do so in the future [11, 14–16].

There are two different pathways to accessing and using DiGA. In addition to a physician prescribing digital health applications, patients may research and select suitable DiGA for their conditions and apply for coverage from their insurance provider directly (see Fig. 5). However, this access option is only used in around 10% of cases, highlighting the importance of doctors, especially GPs, in recommending DiGA to patients and being knowledgeable and supportive of digital therapeutics [8]. Patients who request an activation code from their health insurance company to gain access to the DiGA also encounter bureaucratic hurdles. Often, too much time passes, and ultimately, 17% do not redeem their code at all [17].

Insurers have also not wholeheartedly accepted DiGA. Above all, they have criticized what are perceived as high prices, particularly those set by the manufacturers during the first 12 months. That criticism has at times been linked to the view that reimbursement of provisionally approved DiGA equates to a subsidy for manufacturers. This is particularly salient for insurers in cases where the manufacturer fails to demonstrate positive effects through a study conducted within the trial period, resulting in the application’s removal from the DiGA directory. As the number of activated prescriptions continues to increase (+ 85% in 2024), insurers’ DiGA expenditures will grow proportionally. Germany’s insurance industry association therefore regularly questions the extent to which DiGA expenditure actually contributes to an improvement in healthcare [8].

Conversely, DiGA manufacturers have expressed discontent with the high level of bureaucracy involved in applying to the BfArM for inclusion in the DiGA directory, citing numerous regulatory and certification requirements as contributing factors to high (and growing) development costs. When it comes to generating evidence of the benefits associated with a DiGA, the question arises whether randomized controlled trials should always be the method of choice. Although the DiGA fast-track process explicitly allows for evidence generation beyond traditional randomized controlled trials, in practice, most manufacturers still rely on these trials due to perceived regulatory uncertainty and a lack of successful precedent for real-world evidence [18].

Lessons Learned and the Way Forward

Germany was the first country worldwide to implement a structured process for the assessment and reimbursement of digital health applications. In doing so, it positioned itself as a pioneer in this field. Other countries, such as France, Austria, and Belgium, have since followed, establishing their own frameworks to integrate digital therapeutics into standard care [17]. Germany’s DiGA pathway and the first half decade of experience have shown that the concept of reimbursable digital health applications could be improved and accelerated with targeted changes. For countries at the early stages of developing similar pathways, the German experience offers both inspiration and critical lessons. The core idea of selecting high-quality, evidence-based digital tools and embedding them into routine care has great potential. However, certain aspects of the system’s operationalization and execution could be improved to enhance acceptance among healthcare providers, reduce the perceived financial burden on the public healthcare system, stimulate innovation, and improve health outcomes for patients.

Improved Integration into Patient Journeys

All stakeholders in the healthcare system, especially in a system as strongly characterized by self-administration as in Germany, should be involved in the implementation of new processes early on. Together with insurers and healthcare providers, key therapeutic needs should be identified where digital health applications can offer real added value. Mental and behavioral health, for example, can accommodate easily digitizable approaches (such as cognitive behavioral therapy) and experience a shortage of healthcare providers.

This also involves considering how such health apps can be better integrated into existing care processes so that they are not perceived as foreign bodies within the system, but rather as an integral part of the patient’s journey. A recent example could be the complementarities between DiGA and GLP-1 agonist drugs like Semaglutide or Tirzepatide, which require a comprehensive therapeutic approach to achieve sustainable positive outcomes, and digital applications that support patients in monitoring, education, and implementing necessary lifestyle changes. By embedding such digital tools into the care pathway alongside pharmacological treatments, the overall effectiveness of therapy can be enhanced, adherence improved, and patients empowered to take an active role in managing their health. This kind of integration not only strengthens the continuity of care, but also underscores the added value of digital health solutions, not just as a product, but a real component of more holistic and patient-centered care and solutions [19].

Adjustment of Remuneration and Pricing

To ensure that physicians are motivated to familiarize themselves with prescribable health apps and provide targeted guidance to patients – including onboarding and ongoing support – it is essential to establish appropriate remuneration for these services. At the same time, the pricing of reimbursable digital health applications must be carefully balanced. Pricing should provide manufacturers with sufficient incentives to pursue the fast-track process, while avoiding undue strain on insurers and preserving stakeholder acceptance. Currently, all DiGA in Germany are reimbursed at a flat rate per quarter, regardless of usage or impact. In the future, more flexible, value-based reimbursement models should be considered to better reflect the actual benefit delivered. Additionally, introducing moderate patient co-payments – similar to those for pharmaceuticals – could contribute to the financial sustainability of these applications without creating significant access barriers [19].

From a health system perspective, the sustainability of DiGA reimbursement will critically depend on whether they can demonstrate value in terms of outcomes achieved relative to costs. The planned integration of post-market performance assessment (in Germany known as “Anwendungsbegleitende Erfolgsmessung”) is one step into this direction. Value-based healthcare concepts should guide this next phase of DiGA development, pricing and evaluation. What has been explicitly discussed is an outcome-based remuneration model with 20% reimbursement based on performance, which potentially improves the incentives alignment between manufacturers, payers, and providers. For that, it will be vital to select appropriate success measures that enable comparisons across different digital applications while also accounting for differences in indication areas, targeted outcomes, and intervention characteristics. Transparent post-market performance assessment is also relevant for research to better understand digital therapeutics as a care model, contributing to a broader availability of real-world evidence.

Flexible Approaches for Evaluating Digital Health Applications

To demonstrate the benefits of digital health applications, different evaluation methods are required than those used in clinical trials for new drugs. A range of additional study design aspects must also be considered to adequately reflect the characteristics of digital interventions. This includes the definition and establishment of novel clinical and non-clinical endpoints that may diverge from conventional measures. In addition to objective clinical outcomes, patient-centered dimensions such as quality of life, symptom perception, mental well-being, empowerment, self-management, and treatment adherence should be incorporated.

Further, many experts believe it will be key to adapt Patient-Reported Outcome Measures (PROMs) and Patient-Reported Experience Measures (PREMs) to digital contexts and establishing methodological guidance for their integration into digital platforms. These capture dimensions such as quality of life, symptom perception, mental well-being, and empowerment in self-management – all factors explicitly the DiGA regulation recognizes as “positive care effects.” DTx are particularly suited to continuously collect such outcomes through in-app surveys, wearable integration, or sensor-based monitoring. These types of endpoints not only allow for real-time care adjustment, but also provide real-world evidence that is highly relevant for patients themselves. Embedding PROs and PREMs into routine DiGA evaluations would strengthen patient-centeredness and create an evidence base that goes beyond short-term randomized clinical trials, reflecting real-world effectiveness, patient-centric endpoints, and adherence [18]. In line with this shift toward pragmatic evidence generation, the FDA announced in December 2025 that it would adopt a more flexible approach toward real-world evidence, allowing certain medical device submissions without requiring identifiable individual patient-level data [20].

In addition, opportunities to use meta-data from within the DiGA themselves, like usage duration and frequency, may create additional opportunities to understand, predict, and support digital therapeutics adherence. This data processing and potential secondary use for research purposes is subject to the data-protection and IT-security requirements set by Germany’s Federal Office for Information Security, outlined in § 4 DiGAV.

Effective Communication and Education

Above all, effective communication is essential. Germany’s case clearly shows that meaningful and well-designed innovations do not diffuse on their own, especially given the limited time and resources of everyday medical care. What is needed are persuasive communication strategies, transparent and accessible information, and targeted efforts to enhance digital health literacy among both healthcare providers and patients.

A critical yet often underestimated success factor is continuing education. To address DiGA knowledge gaps among physicians and nurses, structured training formats should be integrated into medical and nursing curricula as well as continuing education programs. In particular, new roles such as Advanced Practice Nurses or Health Community Nurses could play a pivotal role in onboarding patients, conducting home visits, and managing telemonitoring processes. By equipping healthcare professionals with digital competencies and creating new career paths that explicitly include digital health responsibilities, DiGA use can be embedded more effectively and sustainably into routine care.

Data Interoperability and Equity

A major limitation for scaling DiGA remains the lack of interoperability with existing electronic health record systems and practice software. Without seamless data exchange, physicians are unlikely to integrate digital applications into their workflows. National initiatives for standardized APIs and interoperable data infrastructures therefore represent a vital prerequisite for “DiGA 2.0.” At the same time, equity concerns must be addressed. Current uptake shows strong variation by age, socioeconomic status, and region. Especially in rural areas or among older patients, limited digital literacy and infrastructure barriers pose exclusion risks. Targeted support measures, simple user interfaces, and low-threshold onboarding services are essential to ensure that digital therapeutics contribute to reducing, rather than exacerbating, healthcare inequalities.

Conclusion

Germany’s experience with the DiGA fast-track process illustrates the feasibility and challenges of integrating DTx into statutory healthcare. Critical enablers for success include meaningful integration into patient journeys, appropriate remuneration and pricing mechanisms, flexible and patient-centered evaluation approaches, effective communication and education, and robust data interoperability. Addressing these dimensions is essential to strengthening acceptance among healthcare professionals, payers, and patients and unlocking the full potential of digital health applications to improve health outcomes and system efficiency.

References

1. Food and Drug Administration. Device Software Functions Including Mobile Medical Applications 2022. Available at: https://www.fda.gov/medical-devices/digital-health-center-excellence/device-software-functions-including-mobile-medical-applications?utm Accessed May 26, 2025.
2. Kampouraki D. Digital Therapeutics (DTx). Available at: https://www.edps.europa.eu/press-publications/publications/techsonar/digital-therapeutics-dtx_en Accessed November 03, 2025.
3. Zagorski N. FDA Clears First Prescription-Based ‘Digital Therapeutic’. Psychiatric News 2017;52(23):1.
4. Bundesinstitut für Arzneimittel und Medizinprodukte. Das Fast-Track-Verfahren für digitale Gesundheitsanwendungen (DiGA) nach § 139e SGB V: Ein Leitfaden für Hersteller, Leistungserbringer und Anwender. Bonn 2020.
5. Bertelsmann Stiftung. Digitale Gesundheitsanwendungen und Patient-Reported Outcome Measures: Markt- und Potenzialanalyse 2022.
6. Stern AD, Matthies H, Hagen J, et al. Want to See the Future of Digital Health Tools? Look to Germany. 2020. Available at: https://hbr.org/2020/12/want-to-see-the-future-of-digital-health-tools-look-to-germany Accessed November 03, 2025.
7. Techniker Krankenkasse. DiGA-Report 2022 2022.
8. GKV-Spitzenverband. DiGA-Bericht des GKV-Spitzenverbandes – 2024: Bericht über die Inanspruchnahme und Entwicklung der Versorgung mit Digitalen Gesundheitsanwendungen 2025.
9. Bundesinstitut für Arzneimittel und Medizinprodukte. DiGA-Verzeichnis 2025. Available at: https://diga.bfarm.de/de/verzeichnis Accessed May 01, 2025.
10. Weik L, Fehring L, Mortsiefer A, et al. Understanding inherent influencing factors to digital health adoption in general practices through a mixed-methods NPJ Digit Med 2024;7(1):47. doi:10.1038/s41746-024-01049-0 [published Online First: 27 February 2024].
11. Dahlhausen F, Zinner M, Bieske L, et al. Physicians’ Attitudes Toward Prescribable mHealth Apps and Implications for Adoption in Germany: Mixed Methods Study. JMIR Mhealth Uhealth 2021;9(11):e33012. doi:10.2196/33012 [published Online First: 23 November 2021].
12. Bundesinstitut für Arzneimittel und Medizinprodukte. Informationen für Leistungserbringende o.J. Available at: https://diga.bfarm.de/de/leistungserbringende Accessed May 09, 2025.
13. Medical workforce in the EU: an ageing profession 2015. Available at: https://ec.europa.eu/eurostat/web/products-eurostat-news/w/ddn-20250116-2 Accessed November 03, 2025.
14. Brecher A-L, Herrgen J, Mueller T. DiGA in Primary Care and the Influence of Patient-Related Factors – Results of a Survey of Family Doctors. Stud Health Technol Inform 2024;313:55–61.
15. Cirkel L, Lechner F, Schlicker N, et al. Adoption and perception of prescribable digital health applications (DiGA) and the advancing digitalization among German internal medicine physicians: a cross-sectional survey study. BMC Health Serv Res 2024;24(1):1353. doi:10.1186/s12913-024-11807-1 [published Online First: 6 November 2024].
16. Dahlhausen F, Zinner M, Bieske L, et al. There’s an app for that, but nobody’s using it: Insights on improving patient access and adherence to digital therapeutics in Germany. Digit Health 2022;8:20552076221104672. doi:10.1177/20552076221104672 [published Online First: 3 July 2022].
17. Spitzenverband Digitale Gesundheitsversorgung e.V. DiGA-Report 2024: Marktentwicklung digitaler Gesundheitsanwendungen 2025.
18. Stern AD, Brönneke J, Debatin JF, et al. Advancing digital health applications: priorities for innovation in real-world evidence generation. Lancet Digit Health 2022;4(3):e200-e206.
19. Spreafico AMC, Tarricone R, Stern AD. Comprehensive policies for scaling systemic and equitable integration of digital health technologies. NPJ Digit Med 2025;8(1):404. doi:10.1038/s41746-025-01820-x [published Online First: 4 July 2025].
20. Food and Drug Administration. FDA Eliminates Major Barrier to Using Real-World Evidence in Drug and Device Application Reviews 2025. Available at: https://www.fda.gov/news-events/press-announcements/fda-eliminates-major-barrier-using-real-world-evidence-drug-and-device-application-reviews Accessed January 5, 2026.

Tanner Dane Pulice, Stanford University, and Kevin Schulman, Division of Hospital Medicine, Stanford University School of Medicine and Graduate School of Business, Stanford University

Contact: kevin.schulman@stanford.edu

Introduction

Chimeric antigen receptor (CAR) T-cell therapy is a novel form of cancer immunotherapy that engineers a patient’s own T cells to recognize and kill cancer cells. It is primarily used when standard treatments have failed, particularly in advanced blood cancers such as lymphomas, leukemia, and multiple myeloma.^1,2 The first CAR T-cell therapy, tisagenlecleucel, was approved by the FDA in 2017 for the treatment of B-cell acute lymphoblastic leukemia (B-ALL).³ Upon its release, the therapy was priced at $475,000 (Table 1), significantly exceeding the already elevated median launch price of oncology drugs in the United States, which stood at approximately $155,000 USD.⁴ Although CAR-T therapies have demonstrated substantial clinical benefit,^5,6 the aggressive pricing strategy for this category of products has received limited evaluation. This study investigates trends in wholesale acquisition cost (WAC) of six FDA approved CAR-T therapies over time with the goal of assessing how pricing dynamics may be evolving for this category. Our objective is to evaluate the financial burden on payers, and we posit that supply-side moral hazard plays a central role in the year-over-year escalation of CAR-T therapy prices.

Methods

Dataset

We analyzed six FDA-approved CAR T-cell therapies: Yescarta, Tecartus, Carvykti, Abecma, Breyanzi, and Kymriah. Wholesale acquisition cost (WAC) data were obtained from multiple sources. The most recent data (2025) were drawn from Micromedex Red Book, while earlier values were compiled from publicly available pharmaceutical pricing databases, manufacturer disclosures, and peer-reviewed literature.^7-13 WAC reflects the list price in the United States and excludes discounts, rebates, or outcome-based payment adjustments. Pricing data were collected at launch and annually from 2022 through 2025. Global market size was sourced from a March 2025 CAR-T market report published by BCC Research.¹⁴ Market size was initially defined as total global sales revenue (in millions of U.S. dollars) for each product in 2022, 2023, and 2024. We converted these revenue values into estimated patient counts by dividing annual global revenue by the corresponding WAC for that year (i.e., market size = revenue ÷ price). To estimate market size at launch, we applied a 30% compound annual growth rate (CAGR) to back-calculate from observed 2024 patient counts, adjusting for the number of years since each therapy’s launch.^15,16,17 All monetary values were reported in nominal terms without inflation adjustment.

Statistical Analysis

To examine pricing trends in the CAR-T therapy market, we conducted two regression analyses. The first assessed whether a therapy’s initial launch price predicts its future pricing trajectory by regressing each product’s 2025 wholesale acquisition cost (WAC) on its original launch WAC. The second evaluated whether a product’s launch price is associated with its estimated global market size at launch. In addition, we calculated the compound annual growth rate (CAGR) for each therapy using the launch WAC and current 2025 WAC as endpoints.

Results

WAC prices are listed in Table 1 for the six products for the year of the product release and for the period 2022-2024. From these data, we can see that the launch price ranged from $373,000 to $475,000 for these products over the period of 2017-2022. If we consider the launch price of Kymriah as an outlier, other products were priced so that the newest product had a higher launch price than the existing products in the market.

Table 1. Pricing and market trends for six FDA-approved CAR-T therapies

Launch date, wholesale acquisition cost (WAC) from launch through 2025, compound annual growth rate (CAGR), and global market size from launch through 2024 are reported for each therapy. WAC values are listed in USD and reflect pricing trends over time. Market size refers to the estimated number of treated patients globally. CAGR is calculated based on WAC from 2022 to 2025.

Therapy	Generic Name	Market Date	WAC (Launch)	WAC (2022)	WAC (2023)	WAC (2024)	WAC (2025)	CAGR	Global Market Size (Launch)	Global Market Size 2022	Global Market Size 2023	Global Market Size 2024
Kymriah	Tisagenlecleucel	08/30/2017	$475,000	$475,000	$543,828	$582,000	$593,533	3.08%	135	1,128	934	849
Yescarta	Axicabtagene ciloleucel	10/18/2017	$373,000	$399,000	$424,000	$462,000	$503,580	6.32%	626	2,907	3,538	3,929
Tecartus	Brexucabtagene autoleucel	07/24/2020	$373,000	$399,000	$424,000	$462,000	$462,000	4.94%	494	749	873	1,410
Breyanzi	Lisocabtagene maraleucel	02/05/2021	$410,300	$410,300	$447,227	$487,477	$531,350	6.85%	476	444	814	1,045
Abecma	Idecabtagene vicleucel	03/26/2021	$419,500	$419,500	$457,255	$498,408	$528,312	6.31%	595	925	1,032	1,307
Carvykti	Ciltacabtagene autoleucel	02/28/2022	$465,000	$465,000	$465,000	$487,477	$555,310	3.30%	286	286	1,075	1,456

 

Furthermore, across all products, the CAGR for WAC ranged from approximately 3% to 7%. We find that WAC is a strong predictor of future pricing for CAR-T therapies. A linear regression of 2025 WAC on launch WAC yields a statistically significant relationship (p < 0.01) as shown in Figure 1.

Figure 1: Relationship between launch WAC and projected 2025 WAC for six FDA-approved CAR-T therapies

A scatterplot with linear regression line illustrates the association between launch wholesale acquisition cost (WAC) and 2025 WAC (USD) by brand. Each point represents a therapy, color-coded by brand. The trend line indicates that higher launch prices are associated with higher 2025 WAC.

P-value < 0.01

Furthermore, market size for these therapies ranged from 286 to 3,929 patients between 2022 and 2024. Using back-calculated estimates of market size at launch, our second regression showed that launch WAC is significantly negatively associated with estimated launch market size (p = 0.03), as illustrated in Figure 2. For example, Kymriah launched with the highest WAC ($475,000) and had the smallest estimated launch market size (135 patients), while Yescarta, priced lower at $373,000, had a substantially larger estimated launch market size (626 patients).

Figure 2: Estimated launch market size vs. launch WAC for six FDA-approved CAR-T therapies

Each point represents a therapy, with launch wholesale acquisition cost (WAC) on the x-axis and estimated number of patients treated at launch on the y-axis. The line shows a negative linear relationship, suggesting that therapies with higher launch prices tended to have smaller initial market sizes. WAC values are in USD; market size is estimated in number of patients.

P-value = 0.03

Discussion

Our study provides interesting descriptive and temporal data on the CAR-T market in the United States. We observe that sponsors adopted aggressive pricing strategies at launch, with later entrants generally setting higher launch prices than their predecessors. Notably, we find an interesting negative relationship between launch price and market size consistent with a hypothesis that launch price is associated with revenue expectation for a product.¹⁸

This descriptive analysis offers information that is not considered in economic evaluation. While some studies suggest that CAR-T therapies may be cost-effective under specific conditions,^18,19 a therapy offers better value for money at a lower price, even if it meets cost-effectiveness thresholds at a base case. Existing analyses already yield conflicting results, with estimated incremental cost-effectiveness ratios (ICERs) in adults ranging from $10,000 to over $4 million per quality-adjusted life year (QALY).²⁰ Moreover, cost-effectiveness is not an assessment of budget impact from therapies.²¹

WAC prices serve as the base acquisition price for hospitals, less any discounts provided by manufacturers (for hospital outpatient setting, the price after discounts is called the average sales price or ASP). WAC prices can also serve as the basis for pricing at the provider level. In 2022, hospitals received payments from private insurers averaging 254% of Medicare prices for hospital care and 281% of ASP for hospital outpatient infusion therapies.²²

Supply-side moral hazard refers to the influence of insurance coverage on a sponsor’s pricing strategy.²³ When patients are insulated from treatment costs by insurance, manufacturers such as those producing CAR-T therapies can adopt more aggressive launch pricing. In the current U.S. market environment, there is limited downward pressure on manufacturers setting these unprecedented prices.²⁴ In oncology, this dynamic is amplified. Patients facing life-threatening diagnoses are often emotionally vulnerable and price-insensitive, factors which further reduce market resistance to high treatment costs.²⁵ Overall, manufacturers and hospitals face perverse incentives in setting prices for these products. CAR-T therapies may now serve as a pricing precedent for future advanced therapeutics, raising broader questions about affordability, value, and sustainability in innovation.

References

1. Neelapu SS, Locke FL, Bartlett NL, et al. Axicabtagene Ciloleucel CAR T-Cell Therapy in Refractory Large B-Cell Lymphoma. New England Journal of Medicine. 2017;377(26):2531-2544. doi:https://doi.org/10.1056/nejmoa1707447
2. Hansen DK, Sidana S, Peres LC, et al. Idecabtagene Vicleucel for Relapsed/Refractory Multiple Myeloma: Real-World Experience From the Myeloma CAR T Consortium. Journal of Clinical Oncology. 2023;41(11). doi:https://doi.org/10.1200/jco.22.01365
3. Mullard A. FDA approves first CAR T therapy. Nature Reviews Drug Discovery. 2017;16(10):669. doi:https://doi.org/10.1038/nrd.2017.196
4. Rome BN, Egilman AC, Kesselheim AS. Trends in Prescription Drug Launch Prices, 2008-2021. JAMA. 2022;327(21):2145. doi:https://doi.org/10.1001/jama.2022.5542
5. Westin J, Sehn LH. CAR T cells as a second-line therapy for large B-cell lymphoma: a paradigm shift? Blood. 2022;139(18):2737-2746. doi:https://doi.org/10.1182/blood.2022015789
6. Melenhorst JJ, Chen GM, Wang M, et al. Decade-long leukaemia remissions with persistence of CD4+ CAR T cells. Nature. 2022;602(7897):503-509. doi:https://doi.org/10.1038/s41586-021-04390-6
7. Missouri Department of Social Services. SmartPA Criteria Proposal. Published June 16, 2022. https://dss.mo.gov/mhd/cs/advisory/drugpa/pdf/061622/CAR-TCell.pdf
8. Scheffer ER, Kelkar AH, Russler-Germain DA, et al. High Cost of Chimeric Antigen Receptor T-Cells: Challenges and Solutions. American Society of Clinical Oncology Educational Book. 2023;43(43). doi:https://doi.org/10.1200/edbk_397912
9. Drug Price Increase Reporting. Ny.gov. Published November 29, 2024. Accessed July 30, 2025. https://www.dfs.ny.gov/consumers/healthcare/drug-prices/reporting-202411
10. The Medical Letter, Inc. In Brief: Obecabtagene Autoleucel (Aucatzyl) – Another CAR-T Cell Immunotherapy for ALL (online only) | The Medical Letter Inc. Medicalletter.org. Published December 23, 2024. https://secure.medicalletter.org/TML-article-1718f
11. AcariaHealth. Pipeline Report: July 2024. Published July 2024. https://www.acariahealth.com/content/dam/centene/acariahealth/publications/pipeline—july-2024/AH%20Pipeline%20Q324.pdf
12. Department of Consumer and Business Services. Prescription Drug Price Transparency Results and Recommendations – 2020 . Published 2020. https://dfr.oregon.gov/drugtransparency/Documents/Prescription-Drug-Price-Transparency-Annual-Report-2020.pdf
13. Conduent Business Services, LLC. New Drug Fact Blast. Published 2021. https://dss.mo.gov/mhd/cs/advisory/rdac/pdf/abecma-idecabtagene-vicleucel-ndfb_mo.pdf
14. BCC Research. Current Research and Development Status of Chimeric Antigen Receptor (CAR) T-Cell Therapy Market. Bccresearch.com. Published March 25, 2025. Accessed July 30, 2025. https://academic.bccresearch.com/market-research/biotechnology/chimeric-antigen-receptor-car-t-cell-therapy-market-report.html
15. CAR-T Cell Therapy Market – Global Industry Analysis and Forecast (2025-2032). MAXIMIZE MARKET RESEARCH. Published April 16, 2025. Accessed July 30, 2025. https://www.maximizemarketresearch.com/market-report/global-car-t-cell-therapy-market/98045/
16. Research and Markets. Global CAR-T Therapy Market Report 2020: Market is Expected to Stabilize and Reach $3,150 Million in 2025 – COVID-19 Impact and Recovery Forecast to 2030. Prnewswire.com. Published February 2021. Accessed July 30, 2025. https://www.prnewswire.com/news-releases/global-car-t-therapy-market-report-2020-market-is-expected-to-stabilize-and-reach-3-150-million-in-2025—covid-19-impact-and-recovery-forecast-to-2030–301218802.html
17. Marketsandata. CAR T-cell Therapy Market Size, Share, Growth & Demand Forecast 2032. Markets and Data. Published 2018. Accessed July 30, 2025. https://www.marketsandata.com/industry-reports/car-t-cell-therapy-market
18. Trusheim MR, Berndt ER, Douglas FL. Stratified medicine: strategic and economic implications of combining drugs and clinical biomarkers. Nature Reviews Drug Discovery. 2007;6(4):287-293. doi:https://doi.org/10.1038/nrd2251
19. Petrou P. Is it a Chimera? A systematic review of the economic evaluations of CAR-T cell therapy. Expert Review of Pharmacoeconomics & Outcomes Research. 2019;19(5):529-536. doi:https://doi.org/10.1080/14737167.2019.1651646
20. Petrou P. Is it a chimera? A systematic review of the economic evaluations of CAR-T cell therapy – an update. Expert Review of Pharmacoeconomics & Outcomes Research. 2023;23(6):625-650. doi:https://doi.org/10.1080/14737167.2023.2214731
21. Thavorn K, Thompson ER, Kumar S, et al. Economic Evaluations of Chimeric Antigen Receptor T-Cell Therapies for Hematologic and Solid Malignancies: A Systematic Review. Value in Health: The Journal of the International Society for Pharmacoeconomics and Outcomes Research. Published online April 17, 2024:S1098-3015(24)023362. doi:https://doi.org/10.1016/j.jval.2024.04.004
22. Whaley CM, Kerber R, Wang D, Kofner A. Briscombe B. Prices Paid to Hospitals by Private Health Plans: Findings from Round 5.1 of an Employer-Led Transparency Initiative. Rand Research Report RR-A1144-2-v2. Published Dec 10, 2024. https://www.rand.org/pubs/research_reports/RRA1144-2-v2.html 
23. Mark DB, Schulman KA. PCSK9 Inhibitors and the Choice Between Innovation, Efficiency, and Affordability. JAMA. 2017;318(8):711. doi:https://doi.org/10.1001/jama.2017.8907
24. Schulman K. The Supply-Side Effects of Moral Hazard on Drug Prices. HMPI. Published October 14, 2018. https://hmpi.org/2018/10/14/the-supply-side-effects-of-moral-hazard-on-drug-prices/
25. 25. Patel MI, Riley A, Newcomer L, Schulman K. The Price Is NOT Right: Payers’ Roles in Addressing Financial Toxicity. JCO Oncology Practice. 2025;21(1):100-102. doi:https://doi.org/10

Stephen T. Parente, Jessica Haupt, and Molly Dube, University of Minnesota

Contact: paren010@umn.edu

Introduction

The University of Minnesota’s Carlson School of Management and its Medical Industry Leadership Institute (MILI) are recognized leaders in business education and research. MILI offers students innovative training, knowledge, and experience through industry-specific courses and unique hands-on learning[1].  In 2008, MILI started the Medical Valuation Laboratory, which has grown to be the signature program from the perspectives of business partners, inventors, and students.[2] The Medical Valuation Laboratory conducts rapid market assessments for new medical innovations. The students in the course analyze over 30 analyses per year, helping assess lifesaving ideas and streamlining the time-to-market for critical new products. Since 2008, more than 500 innovations have come through the Lab. These innovations have been sourced from organizations and individual inventors throughout Minnesota, across the nation, and around the globe. Each analysis concludes with a binary recommendation to the investor to invest or not as well as the inventor where students suggest whether they should continue to pursue the innovation.

In this paper, we survey the inventors examined by the Lab since 2008 to analyze any correlations between the Lab’s recommendations and the outcome of innovation in terms of market development. We find the Lab’s investor recommendations are suggestive of future success.  There is a strong correlation between positive recommendations and capital raised, reinforcing the importance of structured evaluation processes.

Background

The origins of the Medical Valuation Laboratory began with repeated calls from industry to MILI leadership to identify novel mechanisms to speed University assessments of medical innovations from clinical research to be translated into new products. Working in cooperation with the University of Minnesota’s Office of Technology Commercialization, MILI faculty sought to design a course where MBA students as well as graduate and professional students from different colleges and with different levels of expertise could provide ‘another set of eyes’ to relieve the backlog of projects at the University.  The experimental course went into ‘production mode’ in the fall of 2008, with a defined client/inventor intake process and outputs in the form of white paper and PowerPoint presentation to the inventor.

Clients seeking the Lab’s top-to-bottom analysis of their technology and its market prospects, are from the University of Minnesota, single entrepreneurs, hospitals and clinics, medical device manufacturers, healthcare startups, nonprofits, and more. Once a project is accepted into the Lab, students from nine different colleges at the University of Minnesota can enroll in the course to gain hands-on, real-world experience. Organized into interdisciplinary teams, students take five weeks to complete a full and thorough evaluation of the innovations by appling a business framework. They research market size/potential, competition, intellectual property, regulatory analysis, technical and user evaluations, and finance/reimbursement. Students then develop a recommendation advising both the inventor and potential investors on the feasibility of proceeding with the innovation.

Methods

We examined over 500 projects completed by the Valuation Lab since 2008. We sought to identify common barriers hindering inventors from transitioning their ideas into viable products available to patients. We used the data from the project reports and white papers as well as recommendations to discover if there are recurring themes influencing the process. Several questions drove our investigation:

• Would the data show us themes in the innovations that were successful?
• Are there common points of delay in projects that received a “do not to proceed” recommendation? If so, what are those gaps and how can we address and support them through education, training, and mentorship?

Providing targeted support to help inventors effectively navigate and mitigate roadblocks during the initial development stages, could accelerate the process of getting potentially lifesaving innovations to patients. By summarizing our research and offering recommendations for training, we envisioned a tangible impact on expediting the translation of groundbreaking medical innovation discoveries into life-saving interventions.

We then sent a survey to the inventors of the innovations examined by the Laboratory to assess whether our predictions of “do or do not invest” as well as “do proceed or do not proceed” were associated with the subsequent success of the innovation. Our response rate was 28%, not very robust for statistical inference. However, much of the contact information and innovation branding to achieve a higher response rate was either lost or not pursued.

Our methods include two main components. First, we identify common themes where recommendations advise against proceeding with commercialization efforts. By examining the research behind these recommendations, we sought to uncover recurring barriers such as market size concerns, regulatory hurdles, intellectual property challenges, or inadequate technical validation. Second, we examined the factors underpinning the success of innovations that receive positive recommendations. By recognizing patterns among these success stories, we can create actionable plans that can inform and guide future inventors toward more favorable outcomes. These insights may encompass aspects such as market demand, effective intellectual property strategies, strategic partnerships, or alignment with unmet clinical needs. In addition, we leveraged resources to correlate project data with Pitchbook or DUNS information to track project progress, funding acquisition, and employment growth. Overlaying this data with Valuation Lab recommendations enabled us to identify potential correlations and trends. Finally, we conducted surveys of Lab clients to compare our findings and recommendations with actual project outcomes.

Our data originate from a wide range of projects – from device to biologics to oncology and cardiac. They also come from a wide range of locations – from Minnesota to California, from Sweden to Cambridge in the United Kingdom. We therefore believe that our research and findings can be adaptable and potentially establish generalizable principles for accelerating translational science across the healthcare spectrum and around the globe. After identifying the barriers, and by providing a framework to acknowledge common translational obstacles, our project allows for inventors to be aware and apply our principles in ways that are most relevant and impactful.

In summary, our work aims to establish generalizable principles for accelerating translational science by systematically identifying common barriers, developing potential solutions to the barriers, and sharing the knowledge with others. Through these efforts, we seek to assist inventors across diverse settings and areas of healthcare to overcome translational challenges and expedite the translation of discoveries into real-world solutions that improve human health and well-being.

Results

Our results are organized into three sections.  First, we describe the relationship between the Lab’s recommendations and capital the inventors have raised for their innovations. Second, we decompose the differences by the types of technologies and success with capital acquired for the innovation. Third, we summarize the temporal trends we identified over that period with respect to investment decisions and capital invested.

Projects that received a positive Investor Recommendation (“Invest”) were strongly associated with higher capital raised. On average, projects recommended for investment attracted significantly more funding than those marked “Do Not Invest”. A positive correlation coefficient in technology categories (e.g., device, software) between the numeric investment recommendation score and capital raised, supports a moderate positive relationship and implies that expert evaluations influence or reflect the likelihood of successful capital acquisition.

Three additional insights were: 1) “Invest” recommendations had the highest mean capital raised and a wider distribution, suggesting a mix of both high-potential ventures and outliers; 2) projects labeled “Do Not Invest” received little to no funding, confirming alignment between investor skepticism and market reception, and 3) “Further Info Needed” cases showed more variable outcomes, reflecting some uncertainty in the evaluation process.

Projects in the Device category saw the highest average capital raised, indicating strong investor and market interest in tangible medical innovations. In contrast, categories like Digital Health and Service showed lower average capital, possibly reflecting either earlier-stage development or more competitive funding environments.

Projects at the Pre-Market stage surprisingly received more capital on average than those at Post-Market stage. This may reflect significant early-stage investments required for clinical development, prototyping, and regulatory approval before revenue generation.

With respect to temporal trends, capital raised has shown fluctuations across years, with notable spikes during select periods, possibly influenced by macroeconomic cycles, policy incentives, or lab focus shifts. The accompanying bar chart shows capital raised per year and highlights growth in certain years that may warrant further investigation.

Table 1 describes the potential return on investment (ROI) for the “Proceed” recommendation to the inventor.  The table compares two categories of technology investment decisions: those recommended “Do Not Proceed” versus those that were advanced under “Proceed.” In the “Do Not Proceed” group, 71 technologies were considered, with very modest mean investments of about $40,600 each and a total estimated opportunity ROI of only $8.1 million. In contrast, the “Proceed” group contained 236 technologies, each requiring a far greater mean investment of roughly $1.2 million but generating an enormous cumulative ROI potential of $662 million. This highlights a stark difference in both scale and outcomes between the two approaches.

Table 1: Return on Investment (ROI) Opportunity from Val Lab Proceed Recommendation

Although declining to proceed limited exposure to upfront costs and investment risk, the foregone opportunity value was significant. Advancing technologies required much larger financial commitments but ultimately yielded vastly higher returns, demonstrating the economic advantage of pursuing development. In essence, the “Proceed” strategy provided an ROI nearly 80 times greater than the alternative, indicating that, despite higher resource demands, the decision to proceed proved far more effective in generating financial value.

In contrast, Table 2 describes the potential (ROI) for the “Invest” recommendation to the inventor. The table compares outcomes under two recommendations: not to invest and to invest across various technology categories (biologics, devices, digital health, software, etc.). Under the “Do Not Invest” recommendation, the total opportunity ROI summed to approximately $358.3 million, while under the “Invest” recommendation the total was higher, at about $413.7 million. This shows that both strategies generated substantial returns, but the distribution of investments and their mean size differed. The “Do Not Invest” recommendation leaned on lower per-project investments with a few outsized returns, while the “Invest” recommendation concentrated larger investments across broader categories, including devices and software.

Table 2: Return on Investment (ROI) Opportunity from Val Lab Invest Recommendation

Interpreting the totals, the recommendation to invest produced a better overall ROI compared to not investing, with roughly a $55 million advantage. This suggests that taking on more projects and deploying capital more broadly yields higher aggregate returns. While not every individual investment category within the “Invest” set showed positive outcomes, the broader spread of investments captured enough high-performing opportunities (notably in devices and software) to outweigh weaker areas. Overall, the recommendation to invest was superior in ROI terms, demonstrating that a more aggressive capital deployment strategy across multiple technologies paid off better than holding back.

Lessons Learned for Future Translation

The journey of early-stage ventures provides valuable lessons for investors, particularly when reviewing a diverse set of founders and project outcomes. The following synthesis distills key insights from founders and innovators, highlighting the successes, setbacks, and pivots that shaped their paths. For investors, these reflections offer practical guidance on evaluating opportunities, anticipating risks, and understanding the lived realities of innovation.

One of the most common themes relates to intellectual property (IP) management. Several ventures pursued patent protection but ultimately abandoned applications after receiving feedback that weakened their prospects. For example, one founder noted, ‘A patent application was filed and also abandoned after feedback,’ while another concluded, ‘Final patent renewal due now, not renewing.’ These experiences illustrate that while patents can serve as important assets, they also require careful cost-benefit analysis. Investors should recognize that sunk costs in IP do not always justify continued investment if market validation or novelty is lacking.

Conversely, some ventures achieved progress with patents, including published applications that contributed to credibility and potential valuation. The lesson is that investors should carefully probe the IP strategy of startups—ensuring that patent pursuits align with broader commercialization goals rather than standing alone as markers of success.

Regulatory approval often defines the pace and scale of growth, especially in medical and health-related fields. Comments highlighted milestones such as, ‘Approved by FDA and market launched,’ and ‘FDA cleared on primary device, working on further claims.’ These achievements demonstrate both the potential value of regulatory clearance and the long, resource-intensive process leading up to it.

However, delays and shifting requirements can derail projects. Several founders referenced the time and expense tied to pursuing additional claims or extending regulatory coverage. Investors should anticipate these cycles and ensure portfolio companies are adequately resourced to navigate them.

A number of projects benefited from non-dilutive financing such as SBIR (Small Business Innovation Research) grants. One team shared, ‘Currently on our 3rd funded SBIR project from NIH.’ These examples highlight the importance of grant funding as both a validator of scientific merit and a mechanism to extend runway. Yet other comments revealed financial fragility. Ventures noted abandonment due to personal circumstances, COVID-19 disruptions, or lack of capital. Investors should weigh not only technological promise but also the team’s capacity to secure sustainable funding and weather external shocks.

Startups frequently recalibrated their business models. As one founder wrote, ‘We added B2B software for practitioners soon after,’ signaling a pivot from a direct-to-consumer focus toward a more viable business-to-business model. Others described adding features or altering their commercialization strategies in response to feedback. These pivots underscore the importance of agility. Investors should view pivots not as red flags, but as potential signals of responsiveness and market learning—provided that the changes are grounded in evidence rather than desperation.

The comments also capture how external shocks impact venture trajectories. The COVID-19 pandemic, for example, forced one team to halt progress entirely: ‘Due to Covid and personal issues, the project ended.’ Similarly, competition posed existential threats, as when students uncovered that another company had already developed a similar innovation.

For investors, these examples reinforce the importance of diversification and resilience planning. Even the most promising ideas can falter in the face of unexpected global or competitive dynamics. Partnerships with leading institutions surfaced as pivotal milestones. Examples include projects conducted ‘At Mayo Clinic and Mt. Sinai,’ signaling credibility and validation. Likewise, acquisitions provided tangible outcomes: ‘Acquired by international medical device maker.’ Such exits and collaborations illustrate that successful outcomes are not limited to independent commercialization. For investors, partnership-readiness and strategic alignment with established players can serve as leading indicators of future value.

The collective lessons highlight the complexities of early-stage innovation. Investors must balance optimism with diligence, recognizing the trade-offs in patent strategy, the unpredictability of regulatory pathways, the importance of non-dilutive funding, the inevitability of pivots, the risks of external shocks, and the value of partnerships. While not every venture achieves FDA approval or acquisition, each contributes insights that sharpen investor judgment. By internalizing these lessons, investors can more effectively evaluate opportunities, manage portfolio risk, and ultimately support the translation of innovation into impact.

Implications and Summary

There are three major takeaways from this analysis. First, the Lab’s recommendations are more often correct than not. There is a strong correlation between positive recommendations and capital raised, reinforcing the importance of structured evaluation processes.  This finding demonstrates that the comprehensive approach used by the Lab is effective at identifying early innovations that have the potential for significant return on investment.

Second, we find that early-stage capital is concentrated. A significant portion of funding goes into early-stage or pre-market projects, which underscores the need for robust pre-commercialization support and risk management. It is unsurprising that early investment in successful innovation tends to deliver significant multiplier returns for investors, compared to those who invest at later stages when the technology is more established and market competition has increased.

Finally, the innovation category and time matter. Medical device innovations dominate funding, and the timing of entry into the ecosystem (year and stage) significantly impacts capital outcomes. We suspect this finding is associated by the significant barriers to entry for medical devices that need to be overcome for regulatory and reimbursement approval. Other technologies, for example digital health innovations, often have multiple competitors and weak revenue potential (compared to medical devices or pharmaceuticals) and a much smaller multiplier for investment in and financial reward out for institutional investors and venture capital.

 

Acknowledgements: This research was supported by the National Institutes of Health’s National Center for Advancing Translational Sciences, grant UM1 TR004405. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health’s National Center for Advancing Translational Sciences.”

 

References

[1] MBA programmes that create leaders in global healthcare. Study International. 12 Nov 2020. https://studyinternational.com/news/mba-programmes-that-create-leaders-in-global-healthcare/ Accessed January 15, 2026.

[2] MILI Valuation Lab: Carlson School Lab Influences Med Tech. University of Minnesota. October 13, 2023. https://carlsonschool.umn.edu/news/mili-valuation-lab-carlson-school-lab-influences-med-tech Accessed January 15, 2026.

 

David Scheinker, Andrew Shin, and Rick Majzun, Lucile Packard Children’s Hospital, Stanford University

Contact: dscheink@stanford.edu 

Introduction

The United States healthcare system faces a persistent cost crisis. National health expenditures reached $4.9 trillion, representing 17.6% of GDP, with hospital care alone accounting for $1.5 trillion.[1] Over the past three decades, national expenditures on hospital care have increased at rates consistently exceeding GDP growth, straining households, businesses, and government budgets. Out-of-pocket expenditures alone totaled $506 billion in 2023. Spending growth is commonly attributed to factors inherent to healthcare: expensive new technologies, an aging population with a high burden of chronic disease, and high prices negotiated between hospitals and insurance companies.

The U.S. aviation industry—another complex, highly regulated sector—tells a strikingly different story. Despite facing similarly stringent regulatory oversight from the Federal Aviation Administration (FAA), complex operational challenges, unionized workforces, and substantial capital requirements, airlines have achieved dramatic productivity gains. Real air travel costs, measured as average domestic roundtrip fares in inflation-adjusted dollars, have declined steadily over the past 30 years despite rising fuel costs.

The similarities between aviation and hospital care were first highlighted in the Institute of Medicine’s seminal report, To Err is Human. [2] That work exposed the unacceptably high rates of patient harm and death associated with healthcare errors and described how focusing on system-level improvements — rather than  individual performance — could reduce them. The report underscored structural, operational, and regulatory similarities between aviation and healthcare, including high capital costs, extensive regulations, and large, interdependent workforces. It identified how airlines engineer safety into their processes, rather than rely on individual vigilance, and enumerated opportunities for hospitals to design similar system-based approaches to safety and quality.

To Err is Human made a compelling case that lessons from aviation are valuable for hospital leaders despite the profound differences between the industries. This work examines the divergent trajectories of labor productivity and national spending in U.S. hospitals and airlines. We identify the system-level strategies that have enabled airlines to sustain productivity growth and explore how similar principles could be adapted to improve efficiency, safety, and outcomes in hospitals and other healthcare systems.

Background

Industry Similarities and Differences

Hospitals and airlines share several key characteristics that make for appropriate comparisons of their productivity.

Regulatory

Hospitals and airlines operate under extensive regulatory oversight: hospitals by the Centers for Medicare & Medicaid Services (CMS), the Food and Drug Administration (FDA), and state agencies; and airlines by the FAA and Department of Transportation (DOT). In addition to setting numerous licensing, reporting, and administrative requirements, the regulators restrict capacity by, for example, limiting the number of licensed beds a hospital may operate and tightly controlling airspace where aircraft may fly.

Capital, Workforce, and Safety

Both industries demand substantial capital investments in infrastructure and equipment, rely on highly skilled unionized workforces, and operate continuously with complex demands to schedule and coordinate service. Each faces rigorous safety requirements, where failures carry severe consequences, and handles sensitive data that must be protected. Ultimately, both must balance operational efficiency with service quality in high-stakes environments where errors can be catastrophic. [2]

Contractual

In both industries, incentives are constrained by contractual relationships with non-customer counterparties. Hospitals negotiate with, and are primarily reimbursed by, public and private insurance companies. Airlines negotiate with, and pay landing fees, gate rental charges, and other usage fees, to publicly-owned airports.

Competitive environments

The arguments for why lessons from airlines are valuable for hospitals, despite their differences, are beyond the scope of this work. However, differences in their competitive environments warrant attention. The airline industry has a high level of price competition with extremely price-sensitive consumers and highly sophisticated pricing algorithms. In contrast, hospital payments are dominated by private and public third-party payers who negotiate long-term contracts at fixed prices, leading to relatively little consumer price-sensitivity and thus, price competition (in fact, prices are usually opaque to consumers and care providers). Weaker price competition weakens hospital incentives to reduce prices. However, hospitals still have strong incentives to increase profits by improving labor productivity and efficiency in order to reduce costs. [3]

Labor Productivity in Aviation

Algorithmic Contracting

Airlines negotiate multi-year use and lease agreements with airports that formalize incentives for airlines and airports to be efficient. These contracts use sophisticated rate-setting methodologies to encourage airlines and airports to bear financial risk for their volume of service and the efficiency of their operations, and are shaped by quantitative analyses of non-aeronautical revenue (shops, restaurants, parking lots and hotels) and aeronautical revenues (passenger and airline fees). [4,5] For airports, service-level agreements specify performance targets for turnaround times, facility availability, and passenger throughput, with penalties for underperformance and bonuses for exceeding benchmarks. Empirical data suggest that contract design effectively shapes organizational incentives, finding that well-designed contracts yield 18% to 23% higher operational efficiency than contracts where airlines guarantee airport revenues regardless of performance. [6]

Digital Scheduling

Airlines’ transition to self-service digital sales channels eliminated approximately $3 billion in annual costs associated with travel agent booking fees and call center labor expenses.[7] This shift improves customer convenience and allows airlines to implement real-time revenue management algorithms that adjust prices based on demand patterns and seat availability to maximize revenues and fleet efficiency. Algorithmic pricing systems, pioneered by airlines in the 1980s, generate revenue increases and improve the availability of desirable seats by, for example, offering price-sensitive travelers advance-purchase discounts and offering time-constrained travelers booking on short increased availability at higher prices. [8]

Integrated Mathematical Optimization

The airline industry has been a leading adopter of operations research and optimization techniques since the 1960s. [9-11] A well-known 1969 survey of crew scheduling optimization — one of the top cost drivers for airlines — detailed how airlines use various mathematical techniques to minimize crew scheduling costs while meeting safety regulations, union requirements, and company policies. [9] The following half century has seen steady progress in the use of optimization to schedule larger crews more efficiently. In 1991, American Airlines reported that their TRIP system saved $20 million annually through optimized crew pairing [12]. Today, sophisticated software solutions from vendors like Jeppesen (Boeing), Sabre, and IBS Software achieve, 3% to 15% reductions in crew-related costs through modern mathematical programming approaches [13]. More broadly, airlines use mathematical optimization to improve many aspects of their operations, from flight scheduling to aircraft routing, transforming operational efficiency across the industry. [14]

Labor Productivity in Healthcare

Ad-hoc Contracting

Most hospitals negotiate hundreds of separate payer contracts with varying reimbursement methodologies, quality metrics, and reporting requirements that create administrative burden without systematic incentives for operational efficiency. A significant fraction of hospital revenue comes from fee-for-service contracts negotiated as a fraction of hospital charge master list prices, meaning that providing more care leads to higher payments while improved efficiency can reduce reimbursements. [15] Even the move toward value-based contracting, intended to create incentives for more efficient care, has led to more changes in how patient care is coded than to improvements in how care is delivered.

Manual Scheduling

Most hospitals continue to operate labor-intensive scheduling call centers that preclude: algorithmic optimization of appointment scheduling, self-service patient portals, and dynamic pricing mechanisms to balance demand across time slots or incentivize patients to choose underutilized appointment times.[16]

Sequential Manual Optimization

Hospital and health system operations present numerous opportunities to improve efficiency through optimization technologies: scheduling patient hospital admissions or clinic visits; scheduling nurses to units or shifts; allocating operating room time and block schedules; sequencing elective and urgent surgical cases; and assigning hospital services to specific units. [16-19] Although these aspects of operations drive significant costs and revenues, hospitals have been slow to adopt optimization-based approaches to improve operational efficiency. Electronic Health Record (EHR) systems, now nearly universal, are primarily designed to support clinical documentation, billing, and regulatory compliance than to optimize operational efficiency [20]. A vast literature documents the unrealized potential of EHRs to improve quality and efficiency. [21,22]

Data and Methods

Productivity and Cost Data

Measuring productivity in service industries poses well-documented challenges. The Bureau of Labor Statistics (BLS) defines labor productivity as output per hour worked, but defining “output” for hospitals and airlines requires careful consideration described in industry-specific publications.[23,24] For hospitals, BLS developed a new methodology in 2015 based on counting courses of treatment (inpatient stays and outpatient visits) classified by diagnostic type, representing a substantial improvement over earlier deflated-revenue approaches.[23] For aviation, BLS measures labor productivity as an index of gross output, which is based in large part on passenger miles divided by the labor input index. [24]

Hospital labor and aviation labor productivity data were extracted from the BLS Industry Productivity Studies program. [25] Hospital expenditure data and the population used to calculate per-capita expenditures were extracted from the CMS National Health Expenditure data. [1] Average annual airfare cost data were extracted from the Bureau of Transportation Statistics.[26]

Counterfactual analysis

To estimate the potential savings associated with labor productivity improvements, we calculated the reduction in annual hospital expenditures each year from 1996 to 2019 as follows. From 1995 to 1996, per-capita hospital expenditures increased by 2.2% while average aviation prices declined by 7.8%. We estimated the counterfactual 1996 hospital per-capita spending by reducing the 1995 per-capita spending by half the corresponding reduction in aviation (3.9%) and extrapolating to the entire population to account for population growth. For 1997, we used the adjusted counterfactual 1996 per-capita expenditures as the baseline and adjusted it by half the 1996 to 1997 cost reduction in aviation. We repeated this for subsequent years.

Results

From 1995 to 2019, aviation labor productivity increased by 247% and the average price of a round trip declined by 28%, while hospital labor productivity declined by 1% and per-capita expenditures increased by 287% (Figure 1, Figure 2).

Figure 1: Aviation and hospital cumulative annual labor productivity growth

Figure 2: Aviation price and hospital per-capital expenditure cumulative annual growth

Counterfactual analysis

From 1995 to 2019, total hospital expenditures in 2023 dollar amounts were $18.24 trillion. A reduction in hospital expenditures corresponding to 50% of the corresponding annual price reduction in aviation costs, corresponds to $9.02 trillion, a savings of $9.22 trillion. From 1996 to 2019, hospital expenditures and cumulative savings in aviation increased, from a baseline expenditure of $350 billion in 1996, with a potential savings of $21 billion, to an expenditure of $1,193 billion in 2019, with a potential savings of $838 billion.

Discussion

In our analysis of 25 years of labor productivity, price, and expenditure data for U.S. aviation and hospital care, we found that the aviation industry achieved significant productivity gains and price reductions, while hospital labor productivity remained essentially unchanged and expenditures rose by nearly 300%. Our counterfactual analysis suggests that improving hospital labor productivity at a fraction of the rate achieved in aviation, would correspond to a reduction of expenditures by hundreds of billions of dollars, assuming those savings are passed on to consumers. These findings, along with the success of the safety practices hospitals adopted from airlines, suggest that hospitals should further investigate airline labor productivity practices.

Although limited rigorous research is available, available data suggest that hospitals spend far less on technology that improves productivity than do airlines. Interviews with the Chief Informatics Officers (CIOs) of nine hospital systems found that over 70% of the information technology budget is typically spent on maintenance versus 10% to 20% for innovation. One CIO reported that he, “spends 70% of his time thinking about transformation and 70% of his team’s resources on maintenance.” [27]

Our quantitative findings align with limited data contrasting how hospitals and airlines allocate spending on consulting services. McKinsey & Company, one of the largest healthcare consultants employing 1,700 healthcare-focused advisors, reported $20 million in earnings in a single year from one hospital chain. [28] Their focus, and the more general focus of healthcare consulting, tends to center on strategic planning, mergers and acquisitions, and regulatory compliance rather than mathematical optimization of operational efficiency.

In contrast, Boston Consulting Group, another large consultancy, reports helping one large U.S. airline reduce operational costs by 20% to 25% by optimizing crew staffing, improved maintenance scheduling, and higher airport utilization. This work, along with other consulting work in aviation, emphasizes the deployment of mathematical models to facilitate operational efficiency gains. Even more broadly, the use of cutting edge mathematical modeling is so established in aviation that students commonly pursue graduate degrees in operations research of industrial engineering with the intent of subsequently working in aviation.

This divergence in consulting focus likely both reflects and perpetuates the productivity gap between hospitals and airlines. Hospitals purchase consulting services for strategy and compliance; airlines purchase them for operational optimization. The consulting industry responds to client demand, creating a self-reinforcing cycle where hospitals receive relatively little consulting support for the operational improvements that drive airline productivity.

Implications for hospital managers and policy makers

If hospitals are to pursue the impressive productivity gains seen in aviation, they should follow a similar approach:

1. Negotiate contracts, ideally digital contracts with real-time monitoring, that create shared incentives for hospitals to improve the efficiency of hospital care.
2. Implement self-service digital scheduling for low-risk or routine clinic appointments to improve patient experience, reduce labor costs associated with scheduling, and deploy algorithms that incentivize patients to choose appointments that optimize hospital operational efficiency.
3. Invest in integrated operational optimization that simultaneously considers how surgical blocks are scheduled, how surgical cases are scheduled into blocks, how services are assigned to hospital units, how hospital units are staffed, etc.

Healthcare organizations should expand their current focus on serial or local optimization efforts, e.g., improving single-unit nurse scheduling efficiency, to integrated or global efforts that encompass the entire value stream. The consideration of the value stream should start at the level of contracting, where digital contracts can be designed to reward hospitals for improving efficiency and lowering costs. Increased contractual clarity, e.g., real-time prior authorization can allow the expansion of self-service digital scheduling, pushing the work to consumers and increasing satisfaction. As aviation aggregators such as Kayak have driven transparency and competition in aviation, policymakers and regulators should push hospitals to allow innovative aggregators to optimize transparency for patients and encourage efficiency-driven hospital competition.

Limitations

Our analysis has several limitations. Productivity measurement in both industries involves methodological challenges, and the BLS measures are not perfectly comparable across sectors. Similarly, measuring hospitalization expenditures presents challenges that we do not consider and is typically done at the level of specific services or specialties. However, our findings are sufficiently high-level so as not to depend on the choice of a particular definition of cost. Our counterfactual analysis assumes without justification that hospitals could achieve significant savings, comparable to 50% of those seen in aviation, and that those savings would be passed on to consumers, the government, and insurers in the form of lower prices. Numerous factors independent of labor costs, such as prices negotiated with private insurance companies, drive hospital expenditures. This estimate should be considered as a point of reference rather than a rigorous finding. Finally, we do not account for quality differences — BLS hospital productivity measures include some quality adjustment, but the relationship between operational efficiency and care quality requires further study.

Conclusion

The productivity gap between airlines and hospitals represents one of the most significant and under-recognized opportunities to target reduced hospital savings. From 1995 to 2019, aviation labor productivity increased by 247% and prices declined by 28% while hospital labor productivity declined by 1% and per-capita expenditures increased by 287%. Technology-driven productivity improvements in hospital care, with a focus on strategies successfully deployed by airlines but not yet pursued by hospitals, require additional research.

References

1. Centers for Medicare and Medicaid Services. National Health Expenditure Data. Accessed October, 2025. https://www.cms.gov/data-research/statistics-trends-and-reports/national-health-expenditure-data
2. Donaldson, M.S., Corrigan, J.M. and Kohn, L.T. eds., 2000. To err is human: building a safer health system.
3. Rosko, M., Al-Amin, M. and Tavakoli, M., 2020. Efficiency and profitability in US not-for-profit hospitals. International Journal of Health Economics and Management, 20(4), pp.359-379.
4. Sabel, J., 2004. Airline-airport facilities agreements: An overview. J. Air l. & Com., 69, p.769.
5. Fu, X., Homsombat, W. and Oum, T.H., 2011. Airport–airline vertical relationships, their effects and regulatory policy implications. Journal of Air Transport Management, 17(6), pp.347-353.
6. Karanki, F. and Lim, S.H., 2020. The effects of use agreements on airport efficiency. Journal of Air Transport Management, 84, p.101767.
7. Harison, E. and Boonstra, A., 2008. Reaching new altitudes in e-commerce: Assessing the performance of airline websites. Journal of Air Transport Management, 14(2), pp.92-98.
8. Williams, K., 2020. Dynamic airline pricing and seat availability.
9. Arabeyre, J. P., Fearnley, J., Steiger, F. C., & Teather, W. (1969). The airline crew scheduling problem: A survey. Transportation Science, 3(2), 140-163.
10. Barnhart, C., Belobaba, P., & Odoni, A. R. (2003). Applications of operations research in the air transport industry. Transportation Science, 37(4), 368-391.
11. Gopalakrishnan, B., & Johnson, E. L. (2005). Airline crew scheduling: State-of-the-art. Annals of Operations Research, 140(1), 305-337.
12. Anbil, R., Gelman, E., Patty, B., & Tanga, R. (1991). Recent advances in crew-pairing optimization at American Airlines. Interfaces, 21(1), 62-74.
13. Xu, Y., Wandelt, S. and Sun, X., 2024. Airline scheduling optimization: literature review and a discussion of modelling methodologies. Intelligent Transportation Infrastructure, 3, p.liad026.
14. Lohatepanont, M., & Barnhart, C. (2004). Airline schedule planning: Integrated models and algorithms for schedule design and fleet assignment. Transportation Science, 38(1), 19-32.
15. Koos, H., Scheinker, D., Schulman, K. and Baker, L., 2025. Who Contracts on Diagnosis Related Groups and How Are They Priced? Evidence From Hospital Price Transparency. Health Services Research, p.e70059.
16. Burke, E. K., De Causmaecker, P., Berghe, G. V., & Van Landeghem, H. (2004). The state of the art of nurse rostering. Journal of Scheduling, 7(6), 441-499.
17. Hulshof, P. J., Kortbeek, N., Boucherie, R. J., Hans, E. W., & Bakker, P. J. (2012). Taxonomic classification of planning decisions in health care: A structured review. Health Systems, 1(2), 129-175.
18. Dai, T. and Tayur, S., 2020. OM forum—Healthcare operations management: A snapshot of emerging research. Manufacturing & service operations management, 22(5), pp.869-887.
19. Al Amin, M., Baldacci, R. and Kayvanfar, V., 2025. A comprehensive review on operating room scheduling and optimization. Operational Research, 25(1), p.3.
20. Office of the National Coordinator for Health Information Technology. (2016). Hospital adoption of EHR technology. U.S. Department of Health and Human Services.
21. O’Malley, A.S., 2011. Tapping the unmet potential of health information technology. N Engl J Med, 364(12), pp.1090-1091.
22. Kellermann, A.L. and Jones, S.S., 2013. What it will take to achieve the as-yet-unfulfilled promises of health information technology. Health affairs, 32(1), pp.63-68.
23. Bureau of Labor Statistics. New measure of labor productivity for private community hospitals: 1993–2012. Accessed October, 2025. https://www.bls.gov/opub/mlr/2015/article/new-measure-of-labor-productivity-for-private-community-hospitals-1993-2012.htm
24. Bureau of Labor Statistics. Economic productivity in the air transportation industry: multifactor and labor productivity trends, 1990–2014. Accessed October, 2025. https://www.bls.gov/opub/mlr/2017/article/economic-productivity-in-the-air-transportation-industry.htm#edn2
25. Bureau of Labor Statistics. Productivity. Accessed October, 2025. https://www.bls.gov/productivity/
26. Bureau of Transportation Statistics. Air Fares. Accessed October, 2025. https://www.bts.gov/air-fares
27. Staudenmayer, K., Stewart, C., Purvis, C. and Schulman, K., 2021. Innovations in technology: perspectives of hospital CIOs. Health Management, Policy and Innovation, 6(1).
28. Government taps McKinsey to help manage $175 billion payout to hospitals. Accessed November, 2025. https://www.consulting.us/news/4410/government-taps-mckinsey-to-help-manage-175-billion-payout-to-hospitals
29. Boston Consulting Group. Airline Industry. Accessed November, 2025. https://www.bcg.com/industries/travel-tourism/airline-industry