Economic modeling comparing trauma and general surgery reimbursement

HEALTH SCI. LIB @

~

PAPERS FROM THE SOUTHWESTERN SURGICAL CONGRESS

DEC 1 4 2005 PAPERS PRESENTED Visceral Injuries in Nonaccidental Trauma: Spectrum of Injury and Outcomes

Roaten et at Patterns of Injury and Functional Outcome After Hanging Injury: Analysis of the National Trauma Data Bank

Martin et at Carotid Artery Stenting With Neuroprotection: Assessing the Learning f· Curve and Treatment Outcome

~

t\

Lin et at

~

Total Skin-Sparing Mastectomy Without m Preservation of the Nipple-Areola Complex

Margulies et at Economic Modeling Comparing Trauma and General Surgery Rermbursement

Aucar et at Impact of the SO-Hour Work Week on Resident Emergency Operative Experience

, Feanny et at Gender Differences in General Surgical Careers: Results of a Post-Residency Survey

Yutzie et at In-House Trauma Attendings: Is There a Difference? \

Durham et at

For Complete Table of Contents, Tum to Pages A3 through A6

Official Publication of: The Southwestern Surgical Congress The North Pacific Surgical Association The Association for Surgical Education The Association of Women Surgeons The American Society of Breast Surgeons

FUll-Text: http://www.americanjournalofsurgery.comj

The Association of VA Surgeons

Subm~t Manuscripts: http://ees.elsevier.com/AJS

The Midwest Surgical Association

~

Excerpta Medica

The American

E M

®

Journal of Surgery' The American Journal of Surgery 190 (2005) 932-940

Papers presented

Economic modeling comparing trauma and general

surgery rein1bursement

John A. Aucar, M.D., M.S.H.I. a ,*, Lanis L. Hicks, Ph.D. b

'Department of Surgery, University of Missouri-Columbia, MC 418, One Hospital Drive, Columbia, MO 65212, USA bDepartment of Health ManagemenT and Informatics, University of Missouri-Columbia, Columbia, MO, USA

Manuscript received April 8, 2005; revised manuscript August 8, 2005 Presented at the 57th Annual Meeting of the Southwestern Surgical Congress, San Antonio, Texas, April 10-12, 2005

Abstract Background: The viability of trauma care as a surgical subspecialty is continually challenged by economic pressures related to reimbursement and opportunity costs. Methods: The literature was examined for articles focused on economic implications of a trauma focused surgical practice. Economic forecasting techniques were applied using a recalculating spreadsheet to examine charge and revenue generation comparing the effects of numerous variables affecting a trauma or general surgical service. i Results: Elective general surgery practices derive the majority of revenues from procedural services, whereas trauma practices derive the majority of revenues from evaluation and management. Only centers with high admission volume can expect trauma surgeons to cover salary and expenses, predictably in association with high opportunity costs. Conclusion: The differences in time, effort, and patient volume required for a trauma surgeon to generate revenues comparable to an elective practice are dramatic. The current system creates disincentives for surgeons to participate in trauma care. © 2005 Excerpta Medica Inc. All rights reserved. Keywurds: Economic model; Compensation; Reimbursement; Trauma; Surgical practice

Trauma care is under constant economic and logistical pres­ sures that affect its status as a subspecialty of general surgery. Many institutions exist under constant threat of sacrificing trauma center designation, citing a lack of eco­ nomic viability of their trauma programs [1-3]. Institutions that maintain trauma services often cite difficulties in re­ cruiting and maintaining qualified surgeons with an interest in maintaining trauma as a clinical focus [4]. To hold trauma center designation, institutions must satisfy various regula­ tory requirements as set forth by individual states or by the American College of Surgeons [5] including the availability of specific clinical and ancillary services on a 24-hour basis. The availability requirements for designated trauma centers are much more stringent than for regular medical call coverage and preclude a surgeon from simultaneous elec­ tive general surgical practice. The emergent nature of trauma care precludes both urban and rural trauma centers

* Corresponding author. Tel.: + 1-573-882-8157; fax: E-mail address: [email protected]

+ 1-573-884-4611.

from financial screening of primary or transferred trauma patients. Unfortunately, a negative payer status seems to have a recognizable effect on the likelihood of transfer [6]. Lanzarotti et al [7] noted that patients transferred for trauma care from an initial receiving hospital were associated with a lower mean percentage reimbursement of hospital charges. Although the initial evaluation in nondesignated centers is often conducted by primary care physicians, once transferred, definitive trauma care is usually provided by a general surgeon. However, relative few victims of blunt trauma require operative interventions by a general surgeon. Only 15% of 58 patients reported by Grossman et al [8] required operation by a trauma surgeon, closely agreeing with the operative rates published by others [9]. Orthopedic and neurologic surgeons are generally not involved in the primary assessment and stabilization of trauma patients, although they frequently provide operative services. They also frequently defer the perioperative and subsequent acute care to the admitting trauma surgeon. Victims of penetrating trauma require operation by a general surgeon more fre­ quently; however, these injuries are most prevalent in indigent

JA Aucar et al. I The American Journal of Surgery 190 (2005) 932-940

populations treated at urban hospitals and are associated with uninsured status [7,10]. The regulatory requirements of im­ mediate availability, the high proportion of uninsured or underinsured patients, and the high ratio of nonoperative to operative service needs create a very high opportunity cost for the trauma surgeon compared with the pursuit of an elective surgical practice. In an effort to better understand the potential earning power of trauma surgeons, we use economic modeling techniques using a recalculating spread­ sheet (Excel XP; Microsoft Corporation, Redmon, CA) to examine hypothetical billings, collections, and net produc­ tivity based on varying degrees of trauma care involvement compared with an elective general surgical practice and discuss various practice models that have been considered to alleviate some of the inequities in trauma surgeons' compensation. Professional service charges are based on the Resource Based Relative Value System (RBRVS) and calculated by multiplying relative value units (RVUs) by a dollar per RVU factor. The dollar per RVU factor can vary according to type of service. One institution notes that evaluation and management (E&M) services were charged at only 60% of the rate used for surgical services and 54% of the rate used for radiology services [11]. This makes examination of charges a notoriously difficult method of assessing clinical activity. Also, using lower dollars per RVU charges for E&M services compared with procedures artificially in­ creases the apparent collection rate for E&M services. In Medicare's allowable fee schedule, payments are based on a fixed conversion factor multiplied by the defined total RVUs for a service. Ultimately, revenues will depend some­ what on payer mix. Accurate assessment of trauma com­ pensation systems requires examination beyond just collec­ tion percentage rates or salary surveys [12]. Our current system of classifying trauma as a surgical disease, even though relatively few cases require an opera­ tive intervention, produces a financial disincentive for sur­ geons to provide trauma care coverage rather than routine surgical services. Orossman et al [8], Rogers et al [11], and Sutyak et al [13] have suggested that the current compen­ sation system undervalues the cognitive services provided by trauma surgeons. Ssome of the problems with trauma compensation are caused by inherent deficiencies in the RBRVS system [14,15], which fails to distinguish between routine E&M services and the special circumstances sur­ rounding trauma care. This has led to the prediction of a "gloomy future" for trauma care [16]. The unwillingness of surgeons to participate in trauma care has been cited as a factor in the closure [17] or potential closure of trauma centers [4].

Methods A literature search was conducted using the Medline and Healthstar databases of the National Library of Medicine, using the search terms "physician compensation," "trauma," and "trauma center economics." The bibliographic citations

933

from selected articles were further examined for additional articles pertinent to the subject of reimbursement for sur­ geons in the context of trauma care. Common themes were identified regarding the economic aspects of reimbursement and compensation to surgeons for trauma care, in addition to the economic impact of trauma center designation on hos­ pitals. In this article, the term "reimbursement" is used in the context of payment for specific services made to a physician or practice plan and the term "compensation" is used to refer to collective income paid to surgeons as a result of billings, collections, and institutional support. Subsequently, an economic model using a recalculating spreadsheet (Excel XP) was developed that would allow the comparison of reimbursement patterns between trauma fo­ cused services and an elective general surgical practice under hypothetical but realistic practice conditions. The primary objective was to determine whether and how large of a disparity would exists in work load and reimbursement, between trauma surgery (TS) and elective general surgery (OS) services, under simulated circumstances. The spread­ sheet uses representative E&M and procedure codes (CPT 2004, American Medical Association) and the 2004 Medi­ care allowable fee schedule (Centers for Medicare and Med­ icaid Services) to calculate reimbursement income from focused (TS) and elective (OS) services. The current pro­ cedural terminology (CPT) manual contains approximately 8400 codes (Ingenix 2004). Table 1 shows the CPT codes selected to represent admission, subsequent care, proce­ dural, and discharge services, along with their Medicare­ allowable payment. In the model, TS and OS service activ­ ity is reflected as the sum of E&M services and procedural services based on assumptions described below. Input vari­ ables affecting TS activity include annualized trauma vol­ ume, average daily census (excluding new admissions and discharges), and the number of surgical full-time staffing equivalents (FIBs) required to maintain trauma service cov­ erage. Assumptions that can be altered within the spread­ sheet include the fraction of admissions requiring moderate acuity or high acuity services; the fraction of daily census patients qualifying for critical care services versus low-, mid-, or high-level subsequent care; and the fraction of patients expected to undergo a major surgical procedure or percutaneous gastrostomy and tracheostomy. It was as­ sumed that each patient admitted would require a discharge service. Admission and discharge services were calculated into the E&M component, but newly admitted and dis­ charged patients were not counted on the daily census list. Input variables affecting OS productivity included the total number of major cases performed each month and the fraction of outpatient consults required each month to gen­ erate those cases. Other variable assumptions include the distribution between moderate severity (99243) and high severity (99244) outpatient consults. It was assumed that admission, daily care, and discharge services were covered under the global procedure reimbursement for all payers. Seven representative procedures were selected as noted in

934

J.A. Aucar et al. / The American Journal of Surgery 190 (2005) 932-940

Table I Codes and fee schedule CPT code

Short description

Medicare fee sChect;;;;;"

Evaluation and Management Codes Admit 99221 lli"IT HOSP CARE-DAY E&M LOW SEVERITY 30 MIN 99222 I.l\lT HOSP CARE-DAY E&M MODERATE SEVERITY 50 MIN INIT HOSP CARE-DAY E&M HIGH SEVERITY 70 MIN 99223 Hospital Visit 99231 SUBSQT HOSP CARE-DAY E&M STABLEIRECOVER 15 MIN SUBSQT HOSP CARE-DAY E&M MINOR CMPL 25 MIN 99232 SUBSQT HOSP CARE-DAY E&M SIGNIFIC CMPL 35 MIN 99233 CRITICAL CARE E&M-CRIT ILLIINJUR; 1ST 30-74 MIN 99291 Discharge 99238 HOSPITAL D/C DAY MANAGEMENT; 30 MINUTESILESS HOSPITAL DISCHARGE DAY MANAGEMENT; > 30 MINUTES 99239 Trauma and Critical Care Procedures 31600 TRACHEOSTOMY PLANNED SEP PROC 43246 UGI ENDO; W/DIRECTED PLCMT PERQ GASTROSTOMY TUBE 38100 SPLENECTOMY; TOTAL-SEP PROC 47350 MGMT LIVER HEMORR; SIMPLE SUT LIVER WOUNDIINJURY

$63.54

$1,05.84

$147.45

$31.61

$52.23

$74.28

$193.75

$66.00 $90.02 $379.17 $22401 $781.91 $1,051.74

General Surgical Services 99243 99244 47563 49505 49521 49560 49566 44140 44145

Consults OFFICE CNSLT l'i'EWIESTAB MODERATE SEVERITY 40 MIN OFFICE CNSLT :t\'EWIESTAB MOD-HIGH SEVERITY 60 MIN Procedures LAPAROSCOPY SURGICAL; CHOLECT W/CHOLANGIO REPAIR INIT ING HERNIA AGE 5 YRiOVER; REDUCIBLE REPR RECUR ING HERN AGE; INCARCERAT/STRANGULAT REPAIR INITIAL INCIIVENTRAL HERNIA; REDUCIBLE REPR RECUR INCIIVENTRAL HERNIA; INCARCERISTRANGU COLECTOMY PARTIAL; WITH ANASTOMOSIS COLECTOMY PARTIAL; WITH COLOPROCTOSTOMY

Table 1. General surgery procedural services were calcu­ lated by defining a total number of cases per month and initially assuming a distribution of 20% each for cholecys­ tectomy, initial reducible inguinal hernia, and recurrent in­ guinal hernia, plus 10% each for the other procedures listed. Table 2 shows the relative distribution of E&M services and procedural services used for the described calculations, al­ though the model allows variation in each of these values across a wide representative range. To show realistic charges, the Medicare allowable fee is multiplied by a charge factor of 2.0 for E&M services and 2.5 for procedural services, consistent with common insti­ rutional practices. These charge factors can be manipulated within the spreadsheet model but are applied consistently for TS and GS services. To simulate the effect of a realistic payer mix, a collection ratio was applied to the gross charges. The model allows specification of independent collection ratios for TS and GS, although in most calculated scenarios equivalent collection ratios are used. To simulate a realistic compensation package, a practice overhead factor of 50% was initially assumed for TS and GS services, and the effect of altering that assumption was assessed. Initial assump­ tions used in the model are derived from a combination of published data and practice patterns observed by the author. This model calculates annualized total charges, collec­

$115.02 $162.59 $654.84 $427.12 $648.12 $629.77 $774.43 $1,131.76 $1,404.05

tions, and potential compensation after overhead based on the sum of E&M and procedural services provided by TS and GS under the circumstances specified. Gross revenues for each service component are calculated from the gross charges multiplied by a collection factor. Total available compensation for each of the services is calculated by mul­ tiplying gross collections by a factor of 1 minus the assumed practice overhead. Predicted compensations for each sce­ nario are compared with each other and also compared to the median annual compensation of $224,000 for Midwest­ ern academic surgeons of all ranks, derived from salary surveys published by the Medical Group Management As­ sociation (MGMA). Although the output for each scenario is considered an experimental result, no comparative statis­ tical analysis was performed because the outcomes are hy­ pothetical and could be manipulated to achieve any desired statistical result. Difference in predicted compensation are ex­ pressed in dollars and as FTE based on the MGMA median. Table 3 shows the major equations used by the model.

Results

Once the spreadsheet calculations were programmed, a volume and assumption comhinMinn W"~ rl~"';,,~r1 .~ o;~,,-

J.A. Aucar et al. / The American Journal of Surgery 190 (2005) 932-940

935

Table 2 Fractional distribution of services (initial assumptions, potentially variable) CPT code

Short description

Service fraction

Evaluation and Management Codes Admit INIT HOSP CARE-DAY E&M LOW SEVERITY 30 ML"l INIT HOSP CARE-DAY E&M MODERATE SEVERITY 50 MIN OOT HOSP CARE-DAY E&M HIGH SEVERITY 70 MIN Hospital Visit SUBSQT HOSP CARE-DAY E&M STABLEfRECOVER 15 MIN SUBSQT HOSP CARE-DAY E&M MINOR CMPL 25 MIN SUBSQT HOSP CARE-DAY E&M SIGNIFIC CMPL 35 MIN CRITICAL CARE E&M-CRIT ILLIINJUR; 1ST 30-74 MIN Discharge HOSPITAL D/C DAY MANAGEMENT; 30 MINUTESILESS HOSPITAL DISCHARGE DAY MANAGEMENT; > 30 MINUTES

99221 99222 99223 99231 99232 99233 99291 99238 99239 Trauma and Critical Care Procedures 31600 43246 38100 47350

TRACHEOSTOMY PLANNED SEP PROC UGI ENDO; WIDIRECTED PLCMT PERQ GASTROSTOMY TUBE SPLENECTOMY; TOTAL-SEP PROC MGMT LIVER HEMORR; SIMPLE SUT LIVER WOUNDIINJURY

0.0 0.7 0.3 0.4

0.0 0.4

0.2 0.7* 0.3* 0.1 t 0.1 t

O.U 0.05*

General Surgical Services Consults OFFICE CNSLT NEWIESTAB MODERATE SEVERITY 40 MIN OFFICE CNSLT NEWIESTAB MOD-HIGH SEVERITY 60 MIN Procedures LAPAROSCOPY SURGICAL; CHOLECT W/CHOLANGIO REPAIR INIT ING HERNIA AGE 5 YRiOVER; REDUCIBLE REPR RECUR ING HERN AGE; INCARCERAT/STRANGULAT REPAIR INITIAL INCI/VENTRAL HERNIA; REDUCIBLE REPR RECUR INCI/VENTRAL HERNIA; INCARCERISTRANGU COLECTOMY PARTIAL; WITH ANASTOMOSIS COLECTOMY PARTIAL; WITH COLOPROCTOSTOMY

99243 99244 47563 49505 49521 49560 49566 44140 44145

0.711 0.611 0.2 0.2 0.2 0.1 0.1 0.1

0.1

• Discharge acuity tied to admission acuity.

t 10 % of trauma patients receive both a tracheostomy and percutaneous gastrostomy.

* 15 % of trauma patients receive one major abdominal operation. II Approximately 1/3 of office consults are not operated.

late an elective general surgical practice that would generate the MGMA median compensation. This could be achieved by performing 30 major cases per month, with the distribu­ tion of procedures noted in Table 2. The E&M service component was based on the assignment of new outpatient consults totaling 130% of the number of major cases (30 X 1.3 = 39) with 21 of the consults assigned to CPT 99243 and 18 assigned to 99244 each month. This can also be

Table 3

~del equations and conversions

Gross charges Gross revenues Compensation Gap Percent gap

= (Medicare fee) X (charge factor)

=

(Gross charges) X (collection rate)

X (l - overhead)

GS compensation - TS compensation

(GS compensation - TS compensation)/(MGMA

reference income) (TS compensation)/(MGMA reference income) (GS compensation)/(MGMA reference income) 4 weeks 52 weeks

= (Gross revenues) = =

FTE for TS

=

FIE for GS I month 1 year

=

= =

viewed as operating on 77% of new office consults. Admis­ sions, discharges, and daily visits were not added to GS E&M services because these would presumably be included under global fee payments. Applying a 50% collection rate to gross charges yields E&M-derived revenues of $86,808 and procedure-derived revenues of $360,758. With a 50% practice overhead, assumed to include mal­ practice insurance and office expenses, the net E&M­ derived revenues would be $43,404 and net procedural revenues would be $180,379, for a total annualized com­ pensation of $223,783. With the distribution of services noted in Table 2, and including assumptions that 15% of trauma patients undergo major operations and 10% of patients have both a trache­ ostomy and percutaneous gastrostomy, it was found that the reference level of income for a single trauma surgeon could be achieved with 400 admissions per year, plus an average daily census of 10 patients. That volume, with a 50% col­ lection rate, would yield E&M-based gross revenues of $371,837 and procedural revenues of $76,438. Accounting for a 50% practice overhead, the net E&M and procedural

936

J.A. Aucar et al. / The American Journal of Surgery 190 (2005) 932-940

Table 4 The effect of decreasing the trauma collection rate to 40% is to create a gap equivalent to 0.2 FTE compated to MGMA reference, reducing compensation by more than $44,000 less than the general surgery equivalent

Trauma service E & M revenue Procedure revenue TS Total General surgery E & M revenue Procedure revenue GS Total GS-TS MGMAFTE GS TS For TS = 400, census 0.5.

Rev/y

Rev/y-OH

$297,470 $61,151 $358,621

$148,735 $30,575 $179,310

$86,808 $360,758 $447,566 $88,946

$43,404 $180,379 $223,783 $44,473

1.00 0.80

= 10, Collection = 0.4 vs GS

= 30, collection =

revenues would be $185,919 and $38,219, respectively. This would yield an annual compensation of $224, 138. This represents an excess of $355 compared to the GS compen­ sation or 0.2% over the MGMA target. However, if it is' assumed that the TS service achieves only a 40% collection rate, because of an unfavorable payer mix, the "gap" or GS ­ TS difference becomes $44,473 (less for TS) or 19.9% below the MGMA target. Table 4 indicates the effect of this reduction in collection rate on the E&M- and procedural­ derived revenues for both services and on the apparent FTE compared with the MGMA reference. The entire yearly trauma volume cannot be scheduled conveniently to avoid nights and weekends, and it is im­ practical to assume that a single surgeon will provide 24­ hour coverage, 365 days per year. If we examine the trauma volume required to cover 2 surgeons with full-time dedica­ tion to a trauma service, we find that an annual adrriission volume of 1,100 patients, with an average daily census of 15, assuming the same distribution of services and a 50% collection rate, will yield E&M-based gross revenues of $653,521 and procedural gross revenues of $210,206. With a 50% overhead, this would yield a per surgeon compensa­ tion of $215,932, with a gap of only $7,852 (less) per TS surgeon compared with the single GS surgeon or 3.5% below the MGMA target. If the effect of a 40% collection rate is examined under this scenario, a gap of $51,03 8 (less) per TS surgeon is created, relative to the GS surgeon's activity. This represents compensation that is 22.8% below the MGMA target and makes the TS surgeons appear to be working at 0.77 FTE, based on financial productivity. The effect of this difference in collection rate for a TS service with 1 and 2 surgeons is shown in Fig. 1. The model was used to examine the effect of assigning 3 dedicated FTE surgeons to a trauma service. At a volume of 1000 admissions with a daily census of 10, each surgeon would generate a gap of $110,808. The trauma volume would have to

increase to approximately 1600 admissions with an averaae daily census of 25 for the surgeons to exceed the MGMbA target compensation by 0.6%, if they are to generate the target compensation from clinical income. Table 5 contains the out­ put summaries obtained by increasing the yearly trauma vol­ ume in increments of 100 and gradually increasing the TS daily census until the gap disappears for a 3-surgeon TS service. A decrease in collections to 40% applied to a trauma volume of 1,600 yearly admissions and census of 25 patients, with the previous assumptions held constant produces a gap of $43,748 for each of the 3 surgeons $131,244 collectively. This represents a deficit of 19.2% below the MGMA target. In a scenario in which the GS service volume is increased to an average of 35 cases monthly, with a corresponding increase in consults to 45.5, so that the ratio of consults to operations and other assumptions remains the same, the GS productivity increases to $261,080 or 1.17 times the MGMA reference FI'E. This would create a GS-TS gap of $36,036 for each of 3 TS surgeons with 1,600 yearly admissions. This represents 16.1 % below the MGMA reference. At this volume, decreasing the TS collection rate to 40% would produce a gap of$81,045 per surgeon. For a TS service with 3 surgeons to gain an equivalent level of productivity with equivalent collection rates of 50%, the yearly volume would need to increase to 1800 with a daily census of 30. This translates into approximately 17 more trauma admissions per month to offset the effect of 5 additional cases per month. At this volume level, reducing the collection rate produces a gap of $51,160. Table 6 shows the spreadsheet summaries for these scenarios, and the effect of decreasing TS service collections to 40%. In addition, a time-based analysis was performed using pUblished data pertaining to laparoscopic cholecystectomy [18]. It can be assumed that an average of 150 minutes is adequate for evaluating, operating, and following up the laparoscopic cholecystectomy patient. This can be com-

0;

Gap: GS- TS

When trauma collections decrease by 10 %

$60,000.00

$50,000.00 $40,000.00 $30, DOD. 00

IilVolume'" 400, FTE 1

$20,000.00

.Volume '" 1100, FTE'" 2

$10,000.00

$0.00 -$10,000.00 TS collection rate vs 50 % for GS

Fig. 1. The effect of decreasing the collection rate to 40% (right), from 50% (left), for 1 and 2 trauma surgeons with the specified admission volume is to create a progressively larger deficit compared with the equivalent gen­ eral surgery activity.

J.A. Aucar et al. / The American Journal of Surgery 190 (2005) 932-940

tge 1A ~et

ut­ 01­

rs rs a of 'It,

or %

937

Table 5 Model summary showing the effect of increasing trauma volume on the compensation gap between GS and TS for a 3 surgeon trauma service (results graphed in Figure 2) (Bold indicates change in value relative to the immediately preceding column) Summary Trauma volume Trauma census Trauma FTE GS Cases GS Consults Percent GS operated TR collection rate GS collection rate Overhead GS-TS % of MGMA median

1,000 10 3 30 39 77% 0.5 0.5 0.5 $110,808 49.5%

1,100 15 3 30 39 77% 0.5 0.5 0.5 $79,829 35.6%

1,200 IS 3 30 39 77% 0.5 0.5 0.5 $73,452 32.8%

1,300 20 3 30 39 77% 0.5 0.5 0.5 $42,473 19.0%

1,400 20 3 30 39 77% 0.5 0.5 0.5 $36,095 16.1%

1,500 25 3 30 39 77% 0.5 0.5 0.5 $5,116 2.3%

1,600 25 3 30 39 77% 0.5 0.5 0.5 -$1,261 -0.6%

ed 'lg

to

,S A

,6 IS.

is ld th th ld is IS ~r

te ~t

g g y ~s

e 1-

pared with data by Grossman et al [8], who conducted a time-motion study to examine the activity of admitting 58 victims of blunt trauma at a level II trauma center, without residents. The average time consumed by patients on day of admission was 300 minutes for operated patents versus 148 minutes for nonoperated patients. The difference was not entirely accounted for by the operative time, implying that patients who required operation also required more nonop­ erative attention. Intoxicated patients required on average 193 minutes, significantly more than the 135 minutes aver­ age time consumed by nonintoxicated patients, independent of injury severity. The average time required for all patients was 171 ± 9 minutes. At an average of 171 minutes per admission, not counting operating, following, and discharg­ ing the patient, 400 yearly trauma admissions would occupy 68,400 minutes. Approximately 456 laparoscopic cholecystectomy cases could be performed with the same time commitment (68,400 minutes/150 minutes per case). This represents the equiva­ lent of 38 cases per month. If entered into the current model, the result would be a gap of over $59,000 in favor of the general surgeon. Increasing the trauma volume would only serve to increase the size of the gap, since for any given increase in trauma admissions, cholecystectectomy services could have been increased by a factor of 1.14 (171/150).

Comments A potential shortcoming of this model is the assumption of an equivalent practice overhead affecting trauma and general surgeons. Although acute trauma care is predomi­ nantly hospital based and elective referrals do not apply, clinic or office resources are generally required to provide postinjury surgical follow-up. TS services tend to require a greater degree of the ancillary services provided by ad­ vanced practice nurses and physician assistants, often sup­ ported by clinical depart~ents. Other services, such as social work, discharge planning, and trauma program coordination, are often supported by the hospital. Malpractice insurance, billing and collection costs, and secretarial support are as­ sumed to be similar between TS an~ GS services. Many academic and private administrative units share overhead expenses, so the assumption is held to be generally valid. The spreadsheet model allows the overhead expense factor to be manipulated, but those variations were not explored in the described scenarios. Although the model is simplified and does not account for the great variety of services provided in a typical GS practice or on a TS service, it is sufficiently robust to show the economic principles affecting the delivery of trauma and general surgical services. The diversity of E&M and proce-

Table 6 The effect of increasing GS monthly cases to 35. decreasing TS collections to 40%, increasing trauma volume, census and collections to compensate, and decreasing collection rate again Summary Trauma volume Trauma census Trauma FTE GS cases GS consults Percent GS operated TR collection rate GS collection rate Overhead GS-TS % of MGMA median

* Indicates

1,600 25 3 30 39 77% 0.5 0.5 0.5 -$1,260.88 -0.6%

1,600 25 3 *35 45.5 77% 0.5 0.5 0.5 $36,036.32 16.1 %

change in value relative to the immediately preceding column.

1,600 25 3 35 45.5 77% *0.4 0.5 0.5 $81,045.14 36.2%

*1,800 *30 3 35 45.5 77% *0.5 0.5 0.5 -$1,320.00 -0.6%

1,800 30 3 35 45.5 77% *0.4 0.5 0.5 $51,160.09 22.8%

938

J.A. Aucar et al. / The American Journal of Surgery 190 (2005) 932-940

dural services used in the model offset the unlikely scenario that a general surgeon could have a practice consisting of purely elective laparoscopic cholecystectomy, although comparison to a dedicated hemiology practice might yield different results. If the premise that E&M-based services require more time and reimburse less than procedural ser­ vices is accepted, any model that mixes the effort assign­ ment between TS and GS would be predicted to show that total compensation would decrease as the proportion of effort devoted to trauma increases. Sutyak et al [13] compared charges and costs among 45 patients with splenic injury who underwent splenectomy and 29 patients treated nonoperatively. Surgeon charges, hospital charges, and collections were significantly lower for patients undergoing nonoperative management. The In­ jury Severity Score did not correlate with economic vari­ ables, dispelling arguments that operated patients were sicker and harder to manage. I',"otably, radiologist's charges were not different between operated or nonoperated pa­ tients. Rogers et al [11] compared charges and reimburse­ ment between trauma surgeons and subspecialists for 344 patients at a rural level I trauma center receiving 98% blunt trauma. Trauma surgeons spent 52% of their weekly time in trauma care but generated only 16 % of their billings from trauma. The remaining 84% of charges were generated in the 48% of time spent practicing general sur­ gery. Subspecialists, such as orthopedics and neurosur­ geons, generated average charges 6 times greater than trauma surgeons, yet subspecialists also tend to resist in­ volvement with trauma in favor of elective surgical practice because of the inconvenience of emergency care. For the hospital, where average charges per patient were over 27 times greater than average trauma surgeon charges, net income calculated from full absorption costing, where in­ come or loss is identified by subtracting fixed and variable costs from reimbursement revenues, trauma services re­ sulted in approximately one third of their total profit. This and other articles [19,20] suggest that trauma care can be profitable for hospitals, even if not for the individual pro­ viders. However, this depends largely on a predominantly blunt injury population and a favorable payer mix [20,21]. For most urban trauma centers, a negative operating margin remains the norm [1,4]. The traditional disparity in reimbursement between E&M and procedural services in all specialties has been defended on the basis of extra training, professional risk, and extra services expected from clinicians in a procedure­ based specialty [14]. Despite the trend by congress to in­ crease payments for E&M services, with a compensatory decrease in procedural payments, a prominent difference in these 2 types of services remains. The current RBRVS does not account for differences in resource demands for providing trauma care services compared to routine E&M services [13]. These differences are based on the higher acuity and more complex nature of multidisciplinary care coordination asso­ ciated with trauma.

The resulting disparity in work effort and reimbursement between trauma care and general surgery is aggravated by the inherently emergent nature of trauma care, the need for continuous coverage, the inability to screen for payer status, and the propensity for centralizing care by transferring pa­ tients to busy centers that satisfy designation criteria. This has led to the perception of trauma care as a "poorly Com­ pensated, labor intensive, non-operative task that interferes with elective practice" [13]. Despite trauma being consid­ ered a surgical disease, a 1992 survey of surgical residents revealed that only 18% would wish for trauma care to be a regular or significant part of their practice [22]. Tradition­ ally, it has been left up to the individuals willing to partic­ ipate in trauma care to carry the burden of increased work relative to reimbursement. Some practice groups distribute that burden evenly so that no individual has a significant disadvantage. That approach eliminates the advantages of a dedicated trauma service, impairs quality of care, and in­ creases costs of care delivery [23]. Surgical departments or groups sometimes distribute the burden among a group of providers by shifting costs to accommodate salary guaran­ tees for trauma service providers. This has the effect of making those who provide trauma services unpopular among their peers because they are perceived as low financial produc­ ers, lowering the effective compensation of their associates who must "carry their weight." Increasingly, hospitals and health care systems (ll"e recognizing the secondary benefits accrued in the form of increased elective general and specialty care referrals because of a favorable public image associated with trauma center designation. This has come to be called the "halo effect" [4]. Hospitals benefit directly from trauma services through the increased revenues tied to trauma system activations and through the hospital component of specialty services, such as orthopedic and neurosurgical operations. A sufficiently robust accounting system is needed to distinguish the financial implications of trau­ ma-related services from elective surgery and from non­ trauma emergency services. Recognizing these benefits, many hospitals have begun providing extra pay for trauma service call coverage or support for surgeons' salaries either as base salary support or as income guar­ antees designed to offset the effects of inadequate pro­ cedural volume or unfavorable payer mix. This is some­ times criticized by other specialties as preferential treatment. Those criticisms generally do not take into account the underlying issues related to increased re­ source demands, opportunity costs, and other disincentives for surgeons to participate in trauma care or the secondary benefits that other services derive from maintaining a trauma program. Another strategy increasingly advocated for minimizing the financial disparities associated with trauma is the inclU­ sion of all emergency general surgical services with the trauma service [24,25]. This serves to help the surgeon maintain operative skills when he/she is otherwise relegated

a z s c g e

It n d g tc v­

ti tr rc C1 Vi

ti hi OJ

w in si se th Cl

e\

as se tr, be tn se pr pc pr, tra

c<

to en sy de in

, I

J.A. Aucar et al. / The American Journal of Surgery 190 (2005) 932-940

'! r

to a service with relatively little operative activity and potentially increases procedural revenues relative to E&M services. However, emergency general surgery services are still subject to the risks of unfavorable payer mix and an inconvenient lifestyle. High-trauma volume in this setting increases the risk that the attending surgeon will be operat­ ing on an emergency general surgical patient when a high acuity trauma patient presents, requiring either the mobili­ zation of additional surgical resources or a lowering of the standard of care below the guidelines set forth for trauma center designation. Surgical staffing for 2 types of emer­ gency services during potential peak demand can lead to an even greater revenue gap, particularly if actual demand is less then predicted. Another potential approach is to adopt the European model in which general trauma surgeon performs orthope­ dic surgical care. In that model, abdominal surgeons are generally consulted in cases in which internal bleeding leads to laparotomy [26]. This combines the bulk of patient care with the specialty most likely to perform surgical interven­ tions. This could also be achieved by training general trauma surgeons to perform common orthopedic and neu­ rosurgical procedures [27]. That model is quite distinct from current practice, is not likely to be readily accepted by the various specialties affected, and has unfavorable implica­ tions for risk management. Eventually, the disparities in trauma care economics will have to be addressed at a societal level. The value of organized trauma care, trauma centers, and trauma systems will need to be recognized and supported through public initiatives or the RBRVS will need to be adjusted to con­ sider the resource demands required for maintaining trauma services. Defining distinct E&M service codes or modifiers that account for specialty evaluation under emergency cir­ cumstances would help to offset current deficiencies. How­ ever, this is not likely to improve matters if it is approached as a zero sum maneuver that reduces compensation for other services. A "carve out" of trauma services from payer con­ tracts containing deep discounts for other services has also been proposed but is likely to be feasible only where single­ trauma centers have a relative monopoly in the regional service market. State and federal initiatives to fund im­ provement and regionalization of trauma systems are im­ portant moves forward but have generally overlooked the prevailing disincentives for surgeons to participate in trauma care.

Conclusion

Hypothetical economic modeling techniques are useful to identify the effects of specific practice variables on rev­ enues. In the context of trauma care, current reimbursement systems do not account for disparities between resource demands and payments, opportunity costs, and lifestyle dis­ incentives relative to elective surgical practice. The trend

939

for qualified and experienced trauma surgeons to abandon trauma in favor of elective surgical practice is based on deeply rooted administrative and economic practices that jeopardize the viability of trauma as a subspecialty of gen­ eral surgery. Alternative models for trauma care delivery should be considered.

References [1] Cornwell EE 3rd, Berne TV, Belzberg H, et al. Health care crisis from a trauma center perspective: The LA story. JAMA 1996;276: 940-4. [2] Selzer D, Gomez G, Jacobson L, et al. Public hospital-based level I trauma centers: financial survival in the new millennium. J Trauma 2001;51:301-7. [3] Bazzoli GJ, Meersman PJ, Chan C. Factors that enhance continued trauma center participation in trauma systems. J Trauma 1996;41: 876-85. [4] Eastman AB, Bishop GS, Walsh JC, et al. The economic status of trauma centers on the eve of health care reform. J Trauma 1994;36: 835-44. [5] American College of Surgeons Committee on Trauma. Resourcesfor the Optimal Care of the Trauma Patient. Chicago: American College of Surgeons; 1999. [6] Nathens AB, Maier RV, Copass MK, Jurkovich GJ. Payer status: the unspoken triage criterion. J Trauma 2001;50:776-83. [7] Lanzarotti S, Cook CS, Porter JM, et al. The cost of trauma. Am Surg 2003;69:766-70. [8] Grossman MD, Schwab CW, Chu-Rodgers S, Kestner M. Time and motion: a study of trauma surgeons' work at the bedside during the first 24 hours of blunt trauma care. J Trauma 1999;46:757-63. [9] Bulinski P, Bachulis B, Naylor DF Jr, et al. The changing face of trauma management and its impact on surgical resident training. J Trauma 2003;54:161-3. [10] Clancy TV, Misick LN, Covington D, et al. The financial impact of intentional violence on community hospitals [erratum appears in J Trauma 1994;37:704]. J Trauma 1994;37: 1-4. [11] Rogers FB, Osler T, Shackford SR, et al. Charges and reimbursement at a rural level I trauma center: a disparity between effort and reward among professionals. J Trauma 2003;54:9-14; discussion 14-5. [12] Fakhry SM, Watts DD. What's a trauma surgeon worth? A salary survey of the Eastern Association for the Surgery of Trauma. J Trauma 2000;49:833-8. [13] Sutyak JP, D'Amelio LF, Chiu WC, et al. Cognitive trauma care is undervalued: adult splenic injury as a paradigm. Am Surg 1997;63: 752-7. [14] Maloney IV Jr. A critical analysis of the resource-based relative value scale [see comment] [erratum appears in JAMA 1992;267:2896]. JAMA 1991;266:3453-8. [15] Levy JM, Borowitz M. The Medicare fee schedule unveiled: an account of physician payment reform. Health Policy 1992;22: 263-86. [16] Moore FA, Moore EE, Read RA, Ogunkeye JO. Medicare's resource­ based relative value scale fee schedule portends a gloomy future for trauma surgery. J Trauma 1993;34:900-6; discussion 906-7. [17] Dailey JT, Teter H, Cowley RA. Trauma center closures: a national assessment. J Trauma 1992;33:539-46; discussion 546-7. [18] Stahl JE, Rattner D, Wiklund R, et al. Reorganizing the system of care surrounding laparoscopic surgery: a cost-effectiveness anal­ ysis using discrete-event simulation. Med Decis Making 2004;24: 461-71. [19] Taheri PA, Butz DA, Greenfield LJ. Paying a premium: how patient complexity affects costs and profit margins. Ann Surg 1999;229:807­ 11; discussion 811-4.

940

JA Aucar et al. / The American Journal of Surgery 190 (2005) 932-940

[20] Rogers FB, Osler TM, Shackford SR, et al. Financial outcome of treating trauma in a rural environment. J Trauma 1997;43:65- dis­ cussion 72-3. [21] Taheri PA, Butz DA, Watts CM, et al. Trauma services: a profit center? J Am ColI Surg 1999; 188:349-54. [22] Richardson JD, Miller FE. Will future surgeons be interested in trauma care? Results of a resident survey. J Trauma 1992;32:229-33; discussion 233-5. [23] Abernathy JH 3rd, McGwin G Jr, Acker IE 3rd, Rue LW 3rd. Impact of a voluntary trauma system on mortality, length of stay, and cost at a level I trauma center. Am Surg 2002;68: 182-92.

i

l

[24] Pryor JP, Reilly PM, Schwab CW, et al. Integrating emergency general surgery with a trauma service: impact on the care of injured patients. J Trauma 2004;57:467-71; discussion 471-3. [25] Scherer LA, Battistella FD. Trauma and emergency surgery: an evolutionary direction for trauma surgeons. J Trauma 2004;56: 7-12. [26] AlIgower M. Trauma systems in Europe. Am J Surg 1991;161: 226-9. [27] Harris CH, Smith RS, Helmer SD, et al. Placement of intracra_ nial pressure monitors by non-neurosurgeons. Am Surg 2002;68: 787-90.