• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Surgery. Author manuscript; available in PMC Aug 1, 2009.
Published in final edited form as:
PMCID: PMC2583342
NIHMSID: NIHMS63920

Real Money: Complications and Hospital Costs in Trauma Patients

Abstract

Background

Major postoperative complications are associated with a substantial increase in hospital costs. Trauma patients are known to have a higher rate of complications than the general surgery population. We sought to utilize the National Surgical Quality Improvement Program (NSQIP) methodology to evaluate hospital costs, length of stay, and payment associated with complications in trauma patients.

Study Design

Using NSQIP principles, patient data were collected on 512 adult patients admitted to the trauma service for > 24 hours at a Level 1 trauma center (2004–2005). Patients were placed in one of three groups: no complications (none), ≥ 1 minor complication (minor, e.g., urinary tract infection), or ≥ 1 major complication (major, e.g., pneumonia). Total hospital charges, costs, payment, and length of stay associated with each complication group were determined from a cost accounting database. Multiple regression was used to determine the costs of each type of complication after adjusting for differences in age, gender, new injury severity score (nISS), Glasgow coma scale score (GCS), maximum head abbreviated injury scale (AIS), and first emergency department systolic blood pressure.

Results

330 (64%) patients had no complications, 53 (10%) had ≥ 1 minor complication, and 129 (25%) had ≥ 1 major complication. Median hospital charges increased from $33,833 (none) to $81,936 (minor) and $150,885 (major). The mean contribution to margin per day was similar for the no complication and minor complication groups ($994 vs $1,115, p=0.7). Despite higher costs, the patients in the major complication group generated a higher mean contribution to margin per day when compared to the no complication group ($2,168, p<0.001). The attributable increase in median total hospital costs when adjusted for confounding variables was $19,915 for the minor complication group (p<0.001), and $40,555 for the major complication group (p<0.001).

Conclusion

Understanding the costs associated with traumatic injury provides a window for assessing the potential cost reductions associated with improved quality care. To optimize system benefits, payers and providers should develop integrated reimbursement methodologies which align incentives to provide quality care.

INTRODUCTION

Wide variations in the quality of hospital care are motivating changes in how hospitals and physicians are paid. A recent initiative from the Center for Medicare and Medicaid Services (CMS) has resulted in a new pay-for-performance fee structure for physicians.1 The Physician Quality Reporting Initiative (PQRI), which began July 1, 2007, links reporting of performance data to physician payment for Medicare claims. While voluntary, this program represents an important first step in linking quality measurement to reimbursement. Most recently, CMS announced a policy change whereby Medicare will no longer pay the extra costs of treating preventable errors, injuries, and infections that occur in hospitals. While targeted at not paying for “serious preventable events,” this new policy may have far reaching implications.2

Trauma patients often require expensive hospital-based health-care and are at significant risk for morbid events during their convalescence.3,4 Most mature trauma systems have active quality improvement programs in place to monitor performance improvement and patient safety.5 One difficulty is the inability to distinguish between the degree of risk for a morbid event that the burden of a patient’s traumatic injuries or comorbid conditions creates versus the relationship of the complication to potential provider lapses in quality patient care. Clinically, this situation manifests itself as the poly-trauma patient who sustains a femur fracture, multiple rib fractures, and a closed head injury who consequently develops pneumonia. Is this complication the result of non-adherence to an intensive care unit ventilator associated pneumonia prevention protocol or did the pneumonia develop because of unavoidable microaspiration at a time of biologic/immune risk due to the systemic inflammatory response syndrome triggered by the patient’s traumatic injuries?

Collection of prospective data for health-care quality performance measurement and improvement is expensive.6,7 However, it should be noted that quality assurance is a real line-item cost present in the production of manufactured goods and services in every other industry. While few in health-care debate the utility of the return associated with ensuring quality care, it remains unclear which stakeholder, payer or provider, should monetarily support quality initiatives. A previous study from the University of Michigan demonstrated that both hospitals and payers suffer financial consequences for patients who experience morbid events following an operation.8 However, the larger financial burden tended to fall on the health-care payer.

Because our previous work showed that trauma patients suffer morbid events at a much greater rate than general surgery patients,4 it is likely that trauma patients represent a population that could benefit both clinically and economically from quality improvement initiatives aimed at reducing the incidence of costly complications. To make a business case for a Trauma Quality Improvement Program (TQIP)9, it is necessary to examine the costs associated with complications in trauma patients and analyze the financial consequences borne by each stakeholder. To accomplish this, we utilized the data collection methodology of the NSQIP and applied it to trauma patients at our institution by coupling clinical data to financial data from the hospital cost accounting system. We judged the impact on the hospital by determining how costs and clinical margin change when complications occur. To establish the impact on health-care payers, the costs, payment, and length of stay associated with presence or absence of complication events were determined.

METHODS

Patient Data

From August 1, 2004 to July 31, 2005, 525 adult trauma patients 18 years of age or older with an injury severity score (ISS) ≥ 5 were admitted to the University of Michigan Trauma Service. Patients admitted directly to other services such as Orthopedics, Neurosurgery, or Internal Medicine were excluded. Patients admitted for less than 24 hours or with only burn injuries were also excluded. Data were collected on each patient using the NSQIP methodology and data definitions published for general surgery patients. Our data collector is a trauma service physician assistant who underwent NSQIP training at the West Roxbury Veterans Administration NSQIP training center in Boston, Massachusetts. Trauma registry data were abstracted from the American College of Surgeons National Trauma Registry System (NTRACS v3.4) for ISS, new ISS (nISS), Glasgow coma scale score (GCS), maximum head abbreviated injury scale (hAIS), and mechanism of injury.

Hospital Cost Data

Cost accounting data were obtained from the University of Michigan Health System Data Warehouse (HSDW). The cost accounting data from each patient encounter is entered into the HSDW using a detailed cost-accounting system, Transitions Systems Inc. (TSI). The TSI cost-accounting system tracks the use of all resources and assigns estimates of costs. These estimates are based on direct acquisition costs for supplies and time-and-motion studies for labor costs.10 Financial data obtained included hospital charges, net payment, direct costs, indirect costs, and total margin for the inpatient trauma encounter. We also obtained the primary payer data and placed them into appropriate categories for analysis.

Complications

The primary outcomes of interest were the hospital costs and margins associated with and without complications. All complications for the patients in the study were recorded into a TQIP database using standard definitions from NSQIP. Additional definitions were created for trauma specific non-NSQIP defined complications such as unplanned extubation, reintubation, tension pneumothorax, new onset arrhythmia, decline in GCS, diffuse intravascular coagulation, and shock > 12 hours. Patients were grouped according to complications. The first group of patients had no complications. The second group had one or more minor complications. The third group had one or more major complications. The severity of a complication determined whether it was categorized as major or minor.6 Major complications were those considered significant enough to result in increases to the length of stay or a need for substantial additional treatment interventions. A listing of the complications classification system is shown in Table 1.

Table 1
Aggregation of specific complications into minor and major complication groups.

Complication sub-groups were created for different complications within a similar organ system. The complication sub-groups produced included infectious or incisional, cardiovascular, neurologic, renal, respiratory, and thromboembolic or bleeding. Data were obtained as to what third party insurance provider was the primary payer for each patient. These payers were placed into categories of commercial or automobile, Blue Cross, Medicare, Medicaid, and M-Care HMO (University of Michigan Health Systems sponsored Health Maintenance Organization).

Statistical Analysis

Data were compared using both univariate and multivariate statistical measures. Patients who died in the emergency room or in the operating room prior to inpatient admission were excluded from the analysis. Continuous variables were analyzed using an unpaired two-tailed Student’s t-test for data with a normal distribution. Continuous data exhibiting a skewed distribution such as length of stay were analyzed using the Wilcoxon Rank Sum test. Discrete variables were compared using a Chi-square analysis. Multivariate analysis of costs was performed using multiple linear regression and adjusting for age, gender, nISS, total GCS, hAIS, and first emergency department or hospital systolic blood pressure (ED SBP). Prior to multivariate analysis continuous right-skewed data was natural log-transformed, the regression analysis was then conducted and the coefficient from the regression model was exponentiated to determine the percent increase in costs or length of stay associated with each variable. Database management and querying were performed using Microsoft Access software (Microsoft Corporation, Redmond, WA). All statistical analysis was performed using STATA SE 9.2 software (Stata Corporation, College Station, TX). Results are presented as mean values unless otherwise noted. Statistical significance was defined as a p-value < 0.05. Approval for this study was obtained from the University of Michigan Health System Institutional Review Board.

RESULTS

Patient demographics

A total of 525 adult trauma patients were studied during the one year data collection period. Thirteen patients were excluded who died in the emergency room or operating room prior to formal admission to a hospital bed leaving a total of 512 patients available for cost analysis. Patient characteristic data stratified by complication group (none, ≥ 1 minor, and ≥ 1 major) are listed in Table 2. The change in proportion of males to females in the ≥ 1 minor complication group can be attributed to a disproportionate amount of urinary tract infections occurring in females (21 females and 15 males with a urinary tract infection in the minor complication group). Table 3 demonstrates the rates of complications when ranked by different categories of nISS, GCS, and hAIS. Patients who experienced a major or minor complication were older and had a greater injury burden based on nISS and GCS data when compared to the no complication group in univariate analysis.

Table 2
Patient demographics.
Table 3
Injury Severity Score, Glasgow Coma Scale Score, and Head AIS categories.

Clinical and economic outcomes

The total number of patients who experienced each complication and the percent prevalence for the complication are listed in Table 4. Complications which occurred in greater than 4% of patients included pneumonia (15%), urinary tract infection (13%), deep venous thrombosis/thrombophlebitis (7%), sepsis (5%), and bleeding/transfusions (5%). Mortality was 3.6% in the no complication group, 0% in the ≥ 1 minor complication group and 14.7% in the ≥ 1 major complication group. Patients with a minor or major complication experienced a significant increase in both their median hospital and intensive care unit length of stay (Table 5).

Table 4
Complications.
Table 5
Outcomes and financials by complication group.

Presence of a minor complication increased hospital charges, net payment, and hospital costs when compared to patients with no complications. A minor complication did not significantly increase the total margin or contribution to margin per day. Patients who experienced a major complication had a substantial increase in hospital charges, net payment, hospital costs, total margin and contribution to margin per day when compared to the no complication group. For a major complication the median total hospital costs increased from $17,618 to $71,658 (p<0.001), and the mean total margin went up from $5,073 to $32,884 (p<0.001). Because patients with complications have a longer hospital length of stay, the total margin was divided by the hospital length of stay to generate a mean contribution to margin per day value that offered a normalized margin for comparison between patient groups. Patients with a major complication had an increase in their mean contribution to margin per day from $994 to $2,168 (p<0.001) when compared to patients who did not experience a complication.

To examine other variables which may contribute to costs besides complications, a multivariate analysis was performed. Financial and length of stay data were adjusted for differences in age, gender, nISS, GCS, hAIS, and ED SBP between the complication groups using multiple linear regression. Table 6 lists the attributable increase in financial measures and length of stay due to a minor or major complication when unadjusted or adjusted for differences in patient characteristics. When adjusted for confounding variables, the presence of a minor complication increased the median length of stay by 3.5 days (p<0.001), and a major complication resulted in an increase of 8.7 days (p<0.001) to the median hospital length of stay. Patients with a minor complication had an increase in mean contribution to margin per day of $170 attributable to the presence of the minor complication when compared to the group of patients with no complications. Whereas the unadjusted analysis attributed an increase of $1,173 (p<0.001) to the mean contribution to margin per day for a major complication, the adjusted analysis showed that this attributable increase due to a major complication was only $625 (p=0.1) which was not significant.

Table 6
Relationship of complications to unadjusted and adjusted financial measures and length of stay.

Clinical and economic outcomes by complication sub-group and payer type

When specific complication sub-group analysis was performed, the hospital charges, costs, and length of stay increased markedly for patients with a complication in each of the sub-group categories compared to those who did not experience a complication (Table 7). Respiratory complications resulted in the largest median increase in hospital length of stay, going from 5 days without to 20 days with a respiratory complication (p<0.001). The greatest increase in hospital charges was for presence of a thromboembolic or bleeding complication ($140,701) followed by respiratory ($122,149) and infectious or incisional complications ($120,196). Total hospital costs increased by $38,628 to $64,508 for patients with a complication in the organ system sub-groups. The largest increase in cost was for an infectious or incisional complication ($64,508). Respiratory complications also increased hospital costs by a similar amount ($62,890). However, only a thromboembolic/bleeding ($2,598 vs. $1,122, p<0.001) or neurologic ($2,811 vs. $1252, p<0.001) complication resulted in a statistically significant increase in the mean contribution to margin per day when compared to patients with no complications.

Table 7
Outcomes and financials by complication sub-group.

The State of Michigan operates under no-fault automobile insurance rules. Therefore, patients who are injured while traveling in an insured automobile have medical coverage regardless of their third party health insurance status. For analysis purposes, commercial medical insurance providers (e.g., Aetna) were lumped with automobile insurance providers (e.g. AAA) in the commercial or automobile category. Table 8 lists the distribution of insurance payers who were the primary coverage for medical care received at the University of Michigan for the trauma patients in this study.

Table 8
Payers.

The total margin and mean contribution to margin per day were positive for the commercial/automobile and Blue Cross insured patients whether they experienced a complication or not (Table 9). Medicare and Medicaid generated a negative total margin and mean contribution to margin per day if the patient did not develop a complication. Medicare patients who developed a complication had a positive total margin of $1,879 per patient which was not statistically significant when compared to those patients who did not have a complication.

Table 9
Financials by payer.

DISCUSSION

In this study we sought to determine the impact of complications on hospital costs, reimbursement, and length of stay for trauma patients. While some complications may have an iatrogenic component, it is also true that the biology of a patient responding to acute injury often sets the stage for many of the morbid events that occur in trauma patients. This is known as the two-event construct of postinjury multiple organ failure based on the premise that injury primes the host innate immune system in such a way that a second insult, during this vulnerable period, provokes a severe systemic inflammatory response resulting in end organ dysfunction.11 Of the top five complications present in this study, four are potentially influenced by the status of the systemic inflammatory system following injury (pneumonia, urinary tract infection, deep venous thrombosis, and sepsis). Likewise, those patients who were older and had a greater injury burden, as evidenced by nISS and GCS data, were at increased risk for minor or major complications. This is in contrast to the private sector NSQIP data in which patients who had a minor or major complication had a similar mean age when compared to the no complications group.6

Development of a minor or major complication in trauma patients increased hospital payments borne by third-party payers and length of stay in both univariate and multivariate analyses. A minor complication increased payment by $24,063 when adjusted for case mix, and a major complication increased payment by $47,128. Surprisingly, the increase in length of stay was not financially detrimental to the hospital as the mean contribution to margin per day did not decrease or increase in multivariate analysis when adjustments were made for confounding variables. Therefore, provided that hospital bed utilization is not at maximum capacity the bottom line did not change for the hospital if a patient stayed longer because of a complication.

The life blood of a health-care system is provision of high quality care with maintenance of a reasonable profit margin. Institutions must continually invest in quality initiatives in order to maintain their brand presence, while recruiting and retaining the best physicians and nurses within the region. Maintaining a market position related to optimal quality of care, provides for enhanced physician referrals, increased patient satisfaction and enhanced provider desirability with employers/payers. Conversely, the provision of low quality care will lead to poor patient outcomes, patient dissatisfaction, loss of future patient referrals, and potential medical liability. Providers have very direct and tangible incentives to delivering and reporting quality care. Therefore, both hospitals and health-care payers have strong incentives to work towards reducing complications through quality improvement initiatives.

One should not assume that the financial data from our institution is universally transferrable to all institutions or healthcare systems. Our study is limited by the fact that we treat a high proportion of trauma patients who have a blunt mechanism of injury and very few who suffer penetrating injuries. Also, we are a suburban rather than urban trauma center and our economics may not be applicable to higher volume intercity trauma centers. Morbid events have an increased prevalence in trauma patients who are older and have a greater burden of injury. This is evident from our data in this study. It should not be surmised that all or even a majority of these complications are preventable. However, even a small reduction in these complications could potentially result in a large hospital cost savings which was the basis for conducting our study. Lastly, the state of Michigan operates under no-fault automobile insurance rules and a patient injured in motor vehicle accident with an automobile insurance has unlimited medical benefits.

To better appreciate why it is difficult to directly demonstrate the financial returns related to enhanced quality inpatient care, one must first understand the cost structure of a typical hospital. Hospitals by their very nature are a high fixed cost business. Hospitals require huge ongoing investments in the physical plant, new technologies, skills training, and recruitment. As a result the bulk of a hospital’s costs are fixed. Financially, this means that the institution bears the majority of its costs regardless of the patient volume. By some reports a hospital’s overhead approximates 60–85% of the cost structure of the organization.12 While there is some accounting nuance to these numbers, the key understanding is appreciating that the main part of the institutional costs are determined prior to any patient arrival. Conversely, the variable costs associated with the delivery of care reflect only 15–40% of the total cost. It is in this variable cost allocation where the additional personnel whose focus is on quality initiatives rests. As a result of this variable cost allocation and the diffuse nature of quality work itself, a performance improvement initiative has a potentially limited opportunity for significant cost reduction.13 This makes it difficult to precisely ascertain the direct financial beneficiary of the incremental cost associated with quality improvement personnel. Current CMS quality measures, such as influenza vaccines, smoking cessation counseling, and beta-blocker and aspirin therapy, mainly improve patients outcomes after discharge from the hospital. Medicare realizes a financial savings from the patient’s improved long-term health, but the hospital bears the expense of implementing and continuously conducting these programs with no obvious cost benefit.13

Is there really a low opportunity cost to the hospital in having a patient’s length of stay increased by a complication? A look at other high fixed cost companies suggests that this may not be the case. Southwest Airlines is an example. Airplanes do not generate revenue when they are not in flight. Southwest boards its customers in groups without assigned seating and turns around its planes in 20 minutes, twice as fast as its competitors. If its turnaround times were 10 minutes longer, the airline would need 40 additional new aircraft to handle its current capacity at a cost of millions of dollars.13,14 Conceptualize an airplane as a 20 bed ICU and the reason for placing an emphasis on increased throughput begins to make sense as a way to improve quality. In its simplest terms, the hospitals incentive for quality improvement may be to focus on asset and capacity management (cash flow) to improve throughput by optimizing the utilization of fixed cost investments. This is in contrast to the traditional health-care strategy of performance improvement through cost-reduction initiatives.

Investment in quality improvement initiatives with the intent to reduce complications should lead to a decline in trauma patient length of stay. This reduction in length of stay could decrease cost/case as the substantial fixed costs are dispersed over a larger patient population, which is manifest by increased bed utilization. The end result is increased cash flow with incremental contribution to the margin. A reduction in complication events among trauma patients has the possibility to translate into substantial savings for payers who have an adjusted increase of 120% in payment for a minor complication and 235% for a major complication. Third party payer partnering with health-care systems to pay for continuous quality improvement initiatives for trauma patients is a potential win-win situation. This will become imperative as third party payers reduce or eliminate reimbursement for complications.

CONCLUSIONS

Development of a minor or major complication resulted in increased hospital charges, hospital costs, and net payment in trauma patients. Because the current third party payer system provides incremental reimbursement for complications, there is modest financial incentive for institutions to invest in proactive quality initiatives. To achieve long term success, hospitals and payers should partner to develop continuous quality improvement initiatives targeted at reducing the complications noted in this study.

Acknowledgments

Financial Support: Mark R. Hemmila was supported by the following grants: Central Surgical Association Foundation Enrichment Grant, American Association for the Surgery of Trauma Research and Education Foundation Scholarship, and by National Institutes of Health grant K08-GM078610 with joint support from the American College of Surgeons and the American Association for the Surgery of Trauma.

Abbreviations

hAIS
Head Abbreviated Injury Scale
CMS
Center for Medicare and Medicaid Services
GCS
Glasgow Coma Scale Score
nISS
New Injury Severity Score
NSQIP
National Surgical Quality Improvement Program
TQIP
Trauma Quality Improvement Program

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Presentation at the 3rd Annual Academic Surgical Congress Huntington Beach, California, February 12–15, 2008

Contributor Information

Mark R. Hemmila, Department of Surgery, University of Michigan Medical School, Ann Arbor, Michigan.

Jill L. Jakubus, Department of Surgery, University of Michigan Medical School, Ann Arbor, Michigan.

Paul M. Maggio, Department of Surgery, Stanford University School of Medicine, Stanford, California.

Wendy L. Wahl, Department of Surgery, University of Michigan Medical School, Ann Arbor, Michigan.

Justin B. Dimick, Department of Surgery, University of Michigan Medical School, Ann Arbor, Michigan.

Darrell A. Campbell, Jr, Department of Surgery, University of Michigan Medical School, Ann Arbor, Michigan.

Paul A. Taheri, Department of Surgery, University of Vermont College of Medicine, Burlington, Vermont.

REFERENCES

1. CMS Pay-for-Reporting Physician Quality Reporting Initiative (PQRI) American College of Surgeons Web site. [Accessed October 29, 2007]. Available at http://www.facs.org/ahp/pqri/index.html.
2. Medicare says it won’t cover hospital errors. New York Times Web site. [Accessed August 19, 2007]. Available at http://www.nytimes.com/2007/08/19/washington/19hospital.html.
3. O’Keefe GE, Maier RV, Diehr P, Grossman D, Jurkovich GJ, Conrad D. The complications of trauma and their associated costs in a level 1 trauma center. Arch Surg. 1997;132:920–924. [PubMed]
4. Hemmila MR, Jakubus JL, Wahl WL, Arbabi S, Henderson WG, Khuri SF, et al. Detecting the blind spot: Complications in the trauma registry and trauma quality improvement. Surgery. 2007;142:439–449. [PMC free article] [PubMed]
5. Performance improvement and patient safety. American College of Surgeons Committee on Trauma. 2006 ed. Chicago, IL: American College of Surgeons; 2006. Resources for the optimal care of the injured patient; pp. 101–110.
6. Dimick JB, Chen SL, Taheri PA, Henderson WG, Khuri SF, Campbell DA., Jr Hospital costs associated with surgical complications: a report from the private-sector national surgical quality improvement program. J Am Coll Surg. 2004;199:531–537. [PubMed]
7. Khuri SF. Quality, advocacy, healthcare policy, and the surgeon. Ann Thorac Surg. 2002;74:641–649. [PubMed]
8. Dimick JB, Weeks WB, Karia RJ, Das S, Campbell DA., Jr Who pays for poor surgical quality? Building a business case for quality improvement. J Am Coll Surg. 2006;202:933–937. [PubMed]
9. Pasquale MD. Outcomes for trauma: Is there an end (result) in sight? J Trauma. 2008;64:60–65. [PubMed]
10. Pronovost P, Angus DC. Cost reduction and quality improvement: it takes two to tango. Crit Care Med. 2000;28:581–583. [PubMed]
11. Moore EE, Moore FA, Harken AH, Johnson JL, Ciesla D, Banerjee A. The two-event construct of postinjury multiple organ failure. Shock. 2005;24 Suppl. 1:71–74. [PubMed]
12. Roberts RR, Frutos PW, Ciavarella GG, Gussow LM, Mensah EK, Kampe LM, Straus HE, Gnanaraj J, Rydman RJ. Distribution of variable vs fixed costs of hospital care. JAMA. 1999;281:644–649. [PubMed]
13. Ward WJ, Jr, Spragens L, Smithson K. Building the business case for clinical quality. Healthcare Financial Management. 2006;60:92–98. [PubMed]
14. Lewis C, Lieber R. Testing the latest boarding procedures. The Wall Street Journal Nov 2. 2005
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...