Financial Sustainability

Sepsis poses a cost-containment challenge in the face of the COVID-19 pandemic

July 31, 2020 1:05 am

From left to right: David Kulick, MPH, founder and managing director, The Focus Group Consulting; Erin Bina, global product marketing manager, GE Healthcare; Shailja Vaghela, MPH, healthcare consultant, The Focus Group Consulting.

The average full marginal loss for a U.S. hospital’s treatment of sepsis amounts to as much as $34 million annually, making it one of the highest cost-containment challenges in the nation hospitals.

On a macro level, the economic toll of sepsis is well-known. Each year in the United States, sepsis accounts for 270,000 deaths and costs nearly $27 billion — making it one of the largest cost drivers for U.S. hospitals.[a]

Amid the COVID-19 pandemic, concerns about the cost of sepsis for individual hospitals across the nation have grown to unprecedented levels, given that sepsis is a frequently observed complication of patients with the disease.[b] The need for hospital leaders to understand economic factors impacting margins at their institutions has hit home due to the uncertain financial outlook for hospitals as a result of having to postpone elective procedures and delay necessary care, all the while incurring high costs for treatment of patients with COVID. This point is underscored by the financial impact of sepsis on the nation’s hospitals attributed COVID-19  as of June 30 was estimated to have surpassed $200 billion.[c]

Now, more than ever, hospitals can benefit from economic modeling that tracks the broad range of factors associated with managing sepsis. Here is one approach to economic modeling that provides insight into the marginal impacts of sepsis under ordinary conditions, even before accounting for the increased economic impact of COVID-19. (See the sidebar below, “The sepsis economic modeling methodology,” for details on the methodology used to develop this economic model.)

A new sepsis-focused economic model

When sepsis care is evaluated using a condition-based economic model, the financial implications come into clear focus: The annual marginal loss for large hospitals with more than 500 beds, on average, amounts to about $34 million, while for small hospitals with less than 200 beds, it averages about $9.9 million, as shown in the exhibit below.

Margin impact of sepsis: present on admission compared with hospital acquired


Large Hospitals

Small Hospitals

Present on admission

Average total payment

$18.80 million

$5.64 million

Minimum-maximum payment

$14.6 million – $29.1 million

$4.4 million – $8.7 million

Net margin loss (cost to payment difference)

$19.01 million

$5.60 million

Cost-to-payment ratio




Average total payment

$5.36 million

$1.61 million

Minimum-maximum payment

$4.2 million –  $8.3 million

$1.27 million – $2.49 million

Net Marginal Loss (cost to payment difference)

$14.83 million

$4.40 million

Cost-to-payment ratio




Total payment

$24.16 million

$7.25 million

Minimum-maximum payment

$18.8 million – $37.4 million

$5.67 million – $11.2 million

Net margin loss (cost-to-payment difference)

$33.90 million

$9.90 million

Source: GE Healthcare and Roche Diagnostics, 2020

Traditionally, hospitals evaluate costs and revenue across a service line, unit or facility. As patients utilize additional services, their costs become increasingly diffuse and difficult to track. Indirect costs for patient care further complicate matters, making transparency difficult.

In contrast to a cost-center model described above, the economic model takes a condition-centered approach to examining the total costs and revenue associated with sepsis treatment. The result is a detailed picture of the economic impact of multi-disciplinary sepsis care that delivers startling insights regarding the relationship between payment and cost.

Cost and payment realities

In the sepsis-focused economic model, costs were evaluated across three categories: direct costs, indirect costs and readmission costs. The COVID-19 pandemic will likely have an exaggerated impact on these three cost categories. Hospitals are facing sharp increases in costs as they prepare for and manage COVID-19 patients: extra equipment (PPE, ventilators and disease surveillance), remodeling of rooms to support high-acuity care (also negative pressure) and temporary structures for containment and drive-through clinics.

In the model, these costs were applied against each of the nine permutations of sepsis severity and occurrence. Readmission costs were calculated using a percentage cost of the initial index admission. After quantifying the average cost per sepsis case and the total cost of sepsis management for a hospital, the cost model was compared against a standard cost model for non-sepsis care to derive the incremental cost per hospital for sepsis against non-sepsis management.

Direct costs

Direct costs reviewed under the economic model included average length of stay (LOS) and personnel time, as well as other variable costs.

Average LOS. Sepsis patients have an average LOS that is 75% longer than for patients hospitalized for other conditions.[d] This longer LOS is a significant cost driver that correlates with condition severity, both in the ICU and for the overall hospital stay. Of note, ICU costs are two to three times higher than non-ICU cases, mainly because of the costs associated with personnel and therapeutics. ICU departments require higher ratios of healthcare professionals to patients, including more skilled and costly personnel.[e]

With COVID-19, another suspected increase in LOS comes not only from COVID-19 recovery time but also from the inability to discharge to a lower-acuity area because of lack of COVID-19 testing. Hospitals report being unable to discharge as they await test results.

Personnel time. The personnel burden is the main cost driver across the fixed-cost category. In this model, personnel captured specifically for sepsis-related care were regarded as a fixed cost at the institution level, rather than a variable expense across a budget unit. This approach allowed the partial time spent per case to be identified at a granular level not shown in traditional cost accounting. Again, this analysis was designed to understand cost by diagnosis. Therefore, because staff are not hired to manage a condition, they are considered to be fixed as a managed expense outside the variable impact of the sepsis case.

Once personnel time spent per case was determined from an institution level, the model used data on practitioner-to-patient ratios combined with daily cost rates in both ICU and non-ICU settings to determine the total personnel burden across an individual sepsis case. This approach is valuable in that it captures the specific time spent by personnel on sepsis-related care versus simply allotting full-time employees to the cost model.

Now, against the backdrop of the COVID-19 pandemic, staffing may become increasingly challenging. In a recent survey conducted by the U.S. Department of Health and Human Services (HHS), hospitals reported that they were already struggling with staffing limitations prior to the pandemic and emphasized a particular need for specialized staff, such as infectious disease providers, respiratory therapists, and physicians and nurses who can provide intensive and critical care.[f] In the same HHS survey, multiple hospitals reported concerns that prices of equipment, particularly masks, had increased significantly. And in an open letter, the Society for Healthcare Organization Procurement Professionals called-out increases in costs of about 1,000% due to volume procurement.

Variable costs. Such costs include accommodation charges by day, diagnostics, therapeutics (including cost of lab work, disposables, antibiotics, nutrition and blood products) and mechanical ventilation. Therapeutics is one of the largest variable cost drivers in sepsis management. This fact is not surprising given that most patients with severe conditions receive more frequent lab tests, blood and other nutrition fluids, and more doses of new generation antibiotics. Costs associated with mechanical ventilation also increase the total variable cost per case, averaging $1,552 per day.[g]

Indirect costs

Indirect costs are typically absorbed into hospital administration and seldom tied to specific conditions or disease management. These costs include litigation, disease monitoring and quality programs. Because several of these cost centers can play a major role in sepsis management, they were considered in this model. For example, sepsis surveillance programs are crucial for reducing mortality and morbidity, and they require significant staffing resources.

Another indirect cost factor is a hospital’s national quality ranking, which is a high-visibility component of brand value. Investing deeply in quality improvement resources may be needed to protect the hospital’s brand. If sepsis cases increase, a hospital can experience a costly negative impact on its good name and growth opportunities.

Finally, misdiagnosis and delayed treatment are common with sepsis, and each hour that a sepsis patient goes undiagnosed, the mortality rate increases by 8%, opening the door to liability and quality issues.[h]

Under the economic model, the indirect costs were calculated per case and then added correspondingly to direct variable costs by severity and presentation to calculate an overall variable cost per case.

COVID-19 also will have an impact on these costs: Throughout 2020, as hospitals apply higher rigor on COVID-19 treatment and containment, indirect costs stemming from surveillance monitoring and disease management will likely increase per person.

Total costs

To determine the cost per sepsis case, the full cost model (direct and indirect) for sepsis across severity and occurrence was compared with the total cost of managing adult individuals without sepsis. The model assumed non-sepsis cases were based on standard inpatient LOS and corresponding costs.

Comparing the non-sepsis case costs with total sepsis costs allowed for modeling of the total incremental cost of sepsis to hospitals. For large hospitals, the total cost averages are about $58 million compared with $21 million for an equal number of non-sepsis cases. Thus the incremental additional cost for sepsis cases is $37 million. For smaller hospitals, sepsis cases cost $17.2 million compared with $5.2 million for non-sepsis cases, for an incremental cost of $12 million.

Readmission costs

Readmissions can be a costly part of sepsis management because patients treated for the disease may return to the hospital after initial discharge. Such an event results not only in additional treatment expenditures, but also the risk of reduced payment and penalty under the CMS Hospital Readmission Reduction Program. Under the economic model, sepsis readmission cases were found to cost an additional $4.44 million in large hospitals and $1.32 million in small hospitals. This economic impact underlines the importance of timely, thorough treatment and the need for care transition strategies to reduce this adverse outcome. 

Payment and margin

A caveat to the model’s findings is that because the model assessed potential payment based on epidemiology of condition and age, payment could have been overestimated based on inaccurate hospital coding. In the foregoing exhibit, payment rates for DRGs 870, 871 and 872 were used as proxy estimations for severity of illness. It is important to note that sepsis, once diagnosed, takes the primary diagnosis position, resulting in only one of these three assigned DRGs. However, this fact does not account for miscoded cases where sepsis is not accurately identified. Therefore, hospitals may be absorbing even higher levels of cost for sepsis care than the model indicates.

The cost for hospital-acquired sepsis also is higher than for present-on-admission cases, resulting in a more significant marginal impact. The exhibit above shows the net marginal loss by DRG category for both types. Given payment ranges reported, the exhibit also shows the minimum-to-maximum payment ranges, which are then reflected in the total marginal loss (the cost-to-payment differences).        

All that said, the margin model presented could be conservative in light of COVID-19. Patients with COVID-19 have higher overall costs given staffing, variable costs and indirect costs, which means the a larger margin impact can be expected throughout the pandemic.

A margin-management imperative

The economic model described here not only underscores the high costs hospitals incur for treating sepsis but also the need for them to focus on managing thee costs. Clearly, again, the per person marginal impact of sepsis is likely to be greater in this current environment given higher variable cost pressures.


[a] CDC, 2019; and Hatfield, K.M., et al., “Assessing variability in hospital-level mortality among U.S. Medicare beneficiaries with hospitalizations for severe sepsis and septic shock,” Critical Care Medicine, November 2018.

[b] Zhou, F., et al., “Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: A Retrospective cohort study,” The Lancet, March 11, 2020.

[c] American Hospital Association, “Hospitals and health systems face unprecedented financial pressures due to COVID-19,” press release, May 2020.

[d] Hall, M.J., Williams, S.N., DeFrances, C.J. and Golosinskiy, A., “Inpatient Care for Septicemia or Sepsis: A Challenge for Patients and Hospitals,” U.S. Department of Health and Human Services, National Center for Health Statistics Brief, No. 62, June 2011.

[e] Halpern, N., Pastores, S. “Critical care medicine beds, use, occupancy and costs in the United States: a methodological review,” Critical Care Medicine, November 2015.

[f] Office of Inspector General, Hospital experiences responding to the pandemic: Results of a national pulse survey March 23-27,2020, U.S. Department of Health and Human Services, April 2020.

[g] Dasta, J.F., McLaughlin, T.P., Mody, S.H. and Tak Piech, C., “Daily cost of an intensive care unit day: the contribution of mechanical ventilation,” Critical Care Medicine, June 2005.

[h] Daniels, R., “Surviving the First Hours in Sepsis: Getting the Basics Right (An Intensivist’s Perspective),” Antimicrobial Chemotherapy, National Library of Medicine,, April 2011.

Key findings of the economic model at a glance

Direct costs

  • Length of stay, both in the ICU and inpatient floors, is the main driver of higher costs.
  • Evaluating personnel costs as a fixed expense delivers important granular detail on cost by diagnosis.
  • ICU personnel costs are two to three times higher than non-ICU staff.
  • Therapeutics and mechanical ventilation are leading variable costs.

Indirect costs

  • An effective economic model should assess the indirect costs of sepsis management such as quality programs and initiatives.
  • Factors such as brand management and litigation can have a significant economic impact.

Total costs

  • In a large hospital, sepsis cases cost $37 million more than an equal number of non-sepsis cases.
  • In a small hospital, sepsis cases have an incremental cost of $12 million.

Readmission costs

  • Readmission rates for sepsis are nearly 13%.
  • Sepsis readmission cases cost large hospitals more than $4 million annually.

Payment and margin

  • Hospitals absorb 1.9 to 3.8 times the cost against reimbursement for sepsis care.
  • Hospital-acquired sepsis at large hospitals has the highest cost to reimbursement ratio.

The sepsis economic modeling methodology

Sepsis is a devastating condition, as well as a costly one for individual hospitals and for the U.S. healthcare system as a whole. And its burden has become even more acute with the weight of the coronavirus pandemic. To better understand this burden, GE Healthcare and Roche Diagnostics constructed a precise, condition-based economic model to understand the impact of sepsis at the institution level, with the possible goal of helping to create a case for change, potentially leading to enhanced sepsis care and early identification.

Unique to this approach is assessment of both direct and indirect costs specific to the care and management of sepsis, which becomes increasingly relevant as it is frequently observed in COVID-19 patients. Further, this approach marks a departure from the traditional cost-accounting practices in most hospitals that examine departments, service lines and facilities through which patients move. The condition-based approach reflects a more holistic way to evaluate the precise impact that multi-disciplinary treatment models have on institutions and for the care of sepsis. It also provides an economic model that can help hospitals better understand the financial impact of caring for patients with other deterioration conditions.

The economic model used more than 70 sources of publicly available data and information pertaining to hospital costs and payment, including but not limited to peer-reviewed papers, industry research publications and government statistics.[a]

The authors examined market pressures that exist within hospitals to understand the current cost accounting across the continuum of care. They also interviewed eight U.S. hospital CFOs and other corporate staff to refine a search and review of secondary research.

The model reviewed and analyzed sepsis costs and payment across three levels of severity: sepsis, severe sepsis and septic shock. The model included large hospitals (greater than 500 beds) and smaller (fewer than 200 beds) across the United States.

The modeling work followed a three-step approach:

  1. An epidemiology of the condition by severity and whether it was present at the time of admission or was hospital-acquired.
  2. A detailed cost breakdown of both direct and indirect costs by condition severity across the overarching patient management, including readmission costs.
  3. An analysis of payment by payer mix and age against CMS Medicare rates adjusted in turn to Medicaid and commercial rates to compare and derive insights on the cost of sepsis treatment against standardized costs for non-sepsis cases.


[a] Data on file 



googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text1' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text2' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text3' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text4' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text5' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text6' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text7' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-leaderboard' ); } );