- Email: [email protected]
Surgeon reimbursement for trauma care
The American Journal of Surgery 188 (2004) 767–771
Scientific paper
Surgeon reimbursement for trauma care Mary F. Lumpkin, M.D., Daniel G. Judkins, R.N., M.P.H., John M. Porter, M.D., Mark D. Williams, M.D.* Department of Surgery, Section of Trauma and Critical Care, University of Arizona College of Medicine, 1501 North Campbell Avenue, PO Box 245063, Tucson, AZ 85724-5063, USA Manuscript received July 20, 2004; revised manuscript August 7, 2004 Presented at the 56th Annual Meeting of the Southwestern Surgical Congress, Monterey, California, April 18 –21, 2004
Abstract Background: Trauma care is a well-known financial burden for hospitals, yet reimbursement for the surgeon has not been reported. Methods: For 1999, the percent of the surgeons’ bills reimbursed for general surgery services (gPR) was compared with that for trauma services (tPR). Mean tPR for various groups were compared. Factors predictive of tPR lower than gPR were identified. Results: The gPR was 49%, and, for 371 trauma patients, tPR was 45% (P ⫽ 0.03). The mean tPR for injury severity score (ISS) ⱕ10 was 48%, and for ISS ⱖ11, 57% (P ⫽ 0.03). Patients transferred from outside facilities did not have a significantly lower mean tPR. Penetrating trauma (odds ratio 3.7, P ⫽ 0.008) was predictive of tPR lower than gPR. Conclusions: Surgeon reimbursements for trauma care was significantly, yet only slightly less than for all general surgery care. Surgeons should not be reluctant to take trauma call based on perceptions of low reimbursement. © 2004 Excerpta Medica Inc. All rights reserved. Keywords: Trauma; Surgeon; Reimbursement
Trauma care is well known to be a financial burden for most hospitals [1], yet reimbursement for the trauma surgeon has not been previously reported. A paucity of information is published regarding trauma surgeons’ compensation. In her address to the 1998 Annual Meeting of the American Association for the Surgery of Trauma, Dr Ledgerwood pointed out this fact, saying, “Despite multiple publications regarding the uncertain fiscal viability of trauma centers because of uncompensated and under funded care, few manuscripts divulge trauma surgeons’ salaries or reimbursements” [2]. In November, 2000, Fakhry et al [3] reported that the 1998 annual salaries of trauma surgeons ranged from $90,000 to $528,000. Although both these endeavors made important contributions to what is known regarding trauma surgeon compensation, both were based on survey results and neither addressed the percentage of surgeons’ bills reimbursed. We compared the percent of surgeon billing reimbursed for
* Corresponding author. Tel.: ⫹1-520-626-5056; fax: ⫹1-520-6265016. E-mail address: [email protected]
trauma services to the percentage reimbursement for all general surgery services at this institution in 1999.
Materials and Methods All patients at University Medical Center in Tucson, a level 1 trauma center associated with the University of Arizona, are billed for general and trauma surgeon services through a single professional corporation. For 1999, general surgeon and trauma surgeon charges, patient payments, and actual gross collection rates were examined and merged with the trauma register (Hospital Trauma Register: Version 4.3, Richard H. Cales) onto a Microsoft Excel 2000 spreadsheet (Microsoft Corp., Redmond, WA). The percent of the surgeons’ bills that were reimbursed for all general surgeon services (gPRs) was compared with the percent reimbursement for trauma surgeon services only (tPRs) using the chi-square calculation. Mean tPR for various groups were compared using Student t tests. With logistic regression analysis, demographic factors, mechanism of injury, and injuries predictive of the tPR being significantly lower than gPR were determined. Because the
0002-9610/04/$ – see front matter © 2004 Excerpta Medica Inc. All rights reserved. doi:10.1016/j.amjsurg.2004.08.038
768
M.F. Lumpkin et al. / The American Journal of Surgery 188 (2004) 767–771
Table 1 Surgeon billing, reimbursement, and percent reimbursed for trauma and general surgery services in 1999
Total Nontrauma Trauma
Billed
Reimbursed
Overall PR
$3,169,148 $2,894,143 $275,006
$1,543,273 $1,419,132 $124,141
.49 .49 .45
Blunt Penetrating
Number
Billed
Reimbused
Overall tPR
Mean tPR
292 72
$153,976 $115,375
$85,550 $34,047
.56 .30
.57 .37
tPR ⫽ trauma percent reimbursement.
PR ⫽ percent reimbursement.
outcome or dependent variable (yes the tPR was lower than the gPR or no it was not) is binary in nature, logistic regression, an excellent method for estimating the influence of independent variables, was used [4]. These multiple logistic regression analyses were done with Stata 6.0 software (StataCorp., College Station, TX). Data were transferred from Microsoft Excel and the trauma register into the Stata program using Stat/Transfer version 5 software (Circle Systems Inc., Seattle, WA). For categorical independent variables, the Stata software generated appropriate “dummy variables” automatically. A backward stepwise selection process was used, starting with the fully saturated model [5]. Independent variables were eliminated when they failed to meet the P ⫽ 0.05 level of statistical significance. Remaining independent variables were retested for significance. Results are expressed as odds ratios; the odds the tPR was lower than the gPR given the presence of an independent variable.
Results Total general surgeon charges in 1999 were $3,169,148, of which $1,543,273 was collected, yielding an overall percent reimbursement (gPR) of 49%. As shown in Table 1, for the 371 trauma patients (330 admissions), the overall percent reimbursement for trauma surgeon charges (tPR) was significantly lower at 45% (P ⫽ 0.03). The mean tPR for ISS ⱕ10, 48%, was significantly less than for ISS ⱖ11, 57% (P ⫽ 0.03) (Table 2). The mean tPR for penetrating trauma, 37%, was significantly lower than that for blunt trauma, 57% (P ⬍0.005) (Table 3). The mean tPR for assaults, 36%, was significantly lower than for motor vehicle collisions, 57% (P ⬍0.005) (Table 4). Patients transferred in from outside facilities (58 patients) did not have a significantly lower mean tPR than those who were not. Penetrating trauma (odds ratio 3.7, P ⫽ .008) was a signif-
icant predictor of the tPR being lower than the overall gPR of 49%. Admission (odds ratio 4.2) was not quite a significant predictor of tPR being lower than the gPR (P ⫽ 0.06). Surprisingly, suicide attempt (odds ratio 8.1, P ⫽ 0.03) was a significant predictor of tPR greater than the gPR.
Comments Dr Ledgerwood presented 1996 data from 99 centers responding to the American College of Surgeons Verification Program postreview questionnaire and commendably stated her own total wages for the previous year [2]. Her report included both amounts of trauma surgeon compensation and a wide range of reimbursement schemes. Hourly wages ranged from $30 to $75, whereas daily reimbursements were from $300 to $2,000. Some trauma surgeons received $200 per day plus $300 per resuscitation to $1,200 per day plus $100 per resuscitation, and one institution paid surgeons 70% of their charges [2]. In November 2000, Fakhry et al [3] reported 1998 annual salaries of trauma surgeons ranging from $90,000 to $528,000, with a mean of $229,142 ⫾ $78,045 per year. In this study, surgeon reimbursement for trauma care was surprisingly only slightly, yet still significantly less than for general surgery care. The finding of lower reimbursement in trauma is consistent with previously reported evidence that cognitive skills (ie, nonoperative management) reimburse less than operative management [6] and that trauma surgeons spend more time caring for yet bill less than subspecialists involved in trauma care [7]. In a 1991 survey, surgeons who preferred not to care for trauma patients held the perception that trauma care reimburses less than for nontrauma [8]. That survey also showed that reimbursement issues were less influential than other factors in the preference not to treat trauma patients or take trauma call. Perhaps attitudes toward the importance of reimbursement have
Table 4 Trauma etiology
Table 2 Injury Severity Score
ISS ⱕ10 ISS ⱖ11
Table 3 Blunt and penetrating trauma
Number
Billed
Reimbursed
Overall tPR
Mean tPR
219 152
$94,015 $180,991
$41,584 $82,557
.44 .46
.48 .57
tPR ⫽ trauma percent reimbursement; ISS ⫽ injury severity score.
Assault MVC
Number
Billed
Reimbursed
Overall tPR
Mean tPR
74 218
$108,664 $125,689
$31,323 $69,014
.29 .55
.36 .57
MVC ⫽ motor vehicle crash.
M.F. Lumpkin et al. / The American Journal of Surgery 188 (2004) 767–771
changed. Surgeons queried in 2000 ranked financial reward as the most important motivating factor [9]. Trauma surgeon reimbursement was higher for the more severely injured; motor vehicle collisions; and blunt trauma than for the less severely injured, assaults, and penetrating trauma. Surgeon reimbursement for these subgroups was consistent with our previous report on hospital reimbursement [1]. Unlike the hospital reimbursement data, this surgeon reimbursement data does not suggest any “dumping” of patients (the transfer of patients unlikely to reimburse). To our knowledge, this is the first study to actually document reimbursement for trauma care. Perhaps surgeons should not be reluctant to take trauma call based on perceptions of low reimbursement because this reimbursement is only slightly less than for general surgery physician services.
References [1] Lanzarotti S, Cook CS, Porter JM, Judkins DG, Williams MD. The cost of trauma. Am Surg 2003;69:766 –70. [2] Ledgerwood AM. The thrill of victory; at what price? J Trauma 1999;46:1– 8. [3] 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. [4] Al-Ghamdi AS. Using logistic regression to estimate the influence of accident factors on accident severity. Accid Anal Prev 2002;34: 729 – 41. [5] Hosmer DW, Lemeshow S. Applied Logistic Regression. New York: John Wiley & Sons, 1989. [6] Sutyak JP, D’Amelio LF, Chiu WC, Hammond JS. Cognitive trauma care is undervalued: adult splenic injury as a paradigm. Am Surg 1997;63:752–7. [7] Rogers FB, Osler T, Shackford SR, Healey MA, Wells SK. Charges and reimbursement at a rural level I trauma center: A disparity between effort and reward among professionals. J Trauma 2003;54:9 –15. [8] Esposito TJ, Maier RV, Rivara FP, Carrico CJ. Why surgeons prefer not to care for trauma patients. Arch Surg 1991;126:292–7. [9] Leitch KK, Walker PM. Surgeon compensation and motivation. Arch Surg 2000;135:708 –12.
Discussion Jeffery Saffle, M.D. (Salt Lake City, UT): Dr Lumpkin and her colleagues have debunked one of the frequent excuses given by surgeons for not taking trauma call, but their data need to be viewed in the wider context of the issue surrounding the future of trauma care in the United States. Dr Donald Trunkey has estimated that perhaps as few as 10% of practicing surgeons in the United States continue to care for trauma. The reasons cited are multiple, first of which is that trauma has become increasingly a nonoperative specialty as one of the papers today illustrates. The group at North Carolina has suggested that a surgery resident today will need to participate in the care of 500 patients to perform one splenectomy and one operative repair of a
769
liver injury. In the future, as the population of America ages, penetrating injuries will likely continue to decline and low energy trauma, such as ground level falls, will increase this trend. Trauma surgeons will have to have other practices to maintain their skills. Otherwise, they will find themselves resigned to the last thing they ever desired to be, cognitive specialists. Simultaneously, trauma centers have proven their effectiveness and prospered, but equally quickly many hospitals and surgeons have found it convenient to abdicate their role in providing this care. Competition for bed space and other resources, the allure of more lucrative and convenient specialty practices, and the increasingly dominating issue of lifestyle all make it so attractive for surgeons to get out of the trauma business. As a result, the remaining trauma centers are inundated with low acuity cases, contributing to trauma surgeon burnout and what JD Richardson has called de facto specialization. Taken against this background, Dr Lumpkin’s data contain both good and bad news. I am glad reimbursement is so good, but, on the other hand, 49% of a little is still a little. By my calculations, the patients in this series were billed an average of $741.00 each and paid an average of $334.00 each. The total reimbursement for trauma care, just over $124,000.00, is far too little to pay even a single surgeon’s salary, even Dr Anna Ledgerwood’s very small salary. Data from Vermont have confirmed what we already know. Specialists, including radiologists, bill five times as much as we do, even though they contribute relatively little ongoing care. Hospitals make money as well. Thus, surgeons who provide 24 hour a day call coverage, triage, resuscitation, workup, and follow-up care make money for everyone but themselves. This brings me to three questions for Dr Lumpkin. First, does your hospital pay surgeons to take trauma call and is this appropriate? Second, is trauma call at your hospital incorporated into general emergency call, so your surgeons can make a living with appendectomies while indulging their hobby of trauma care? And, finally, does your hospital maintain a surgeon-led, dedicated trauma service for the subsequent care of these patients after their orthopedic or neurosurgical procedures? I do have one final comment. I think it is critically important that the issue of surgeon involvement in trauma is discussed at meetings like this one, rather than only at subspecialty trauma meetings, so that we can continue to develop a trauma care system, that is as inclusive of as many surgeons as possible. And, as a last question, if you could let me know the hospital that pays $528,000.00 for trauma surgeons, I’d like to talk to them.
Answer Mary Lumpkin, M.D. (Tucson, AZ): First of all, the total amount billed represented in the article only represents a very small portion of the trauma patients. In the past at our facility, we have had some significant billing problems for trauma surgeon services that are in the
770
M.F. Lumpkin et al. / The American Journal of Surgery 188 (2004) 767–771
process of being improved, or, as I understand it, significantly updated. As a resident, it is not something that I am privy to. For instance, the number of admitted patients we looked at represented about one third of all the trauma admissions for that year. Second these data are from 1999. Subsequent to that, we became the only level one trauma center in Southern Arizona when Tucson Medical Center withdrew from trauma care. Do our trauma surgeons get paid to take call? I do not believe so, but, as a resident, it is not something that I would necessarily know. Another change that has occurred subsequent to this data being studied is that our trauma surgeons are now in-house and they are required to be in the trauma bay within a certain amount of time depending on the level of acuity of the trauma. As for your second question, no, trauma call is not separate from other general surgery emergencies. Our system is evolving. The trauma service takes care of the general surgery patient at night and operates if need be, but during the day other services take surgery consults and calls. Those general surgery patients operated at night by the trauma service are often postoperatively transferred off of the trauma service. As you might imagine, we have experienced a humungous burden by the doubling of our trauma population when the other trauma center closed. Do we have a dedicated trauma service? Yes, we do. We have a trauma service staffed by residents who are increasingly stretched thin with the resident work-hour restrictions and our doubling of trauma patients, which all happened on the same day at our facility. Question Ronald Stewart, M.D. (San Antonio, TX): For the residents in the audience, you can make a good living doing trauma surgery. Having said that, I have one really nagging question. I am not certain you used the right control. Your control was your university surgeons. Do you think you should have added another control of community surgeons in private practice?
imbursement is sometimes quite tenuous. The dominant influence between billing and reimbursements is the Federal Government. They have put together risk pools, and general surgeons (eg, vascular, oncology, and trauma) are all in the same risk pool. But sometimes, we have to bill outside of our risk category called CPEP. Don’t ask me what it stands for. For example, if I, as a general surgeon, read an EKG, I bill under the cardiologists’ codes. So my question is, do your trauma surgeons predominantly bill within general surgery coding (ie, as do-ologists) or do they “crosswalk” outside of their CPEP into somebody else’s reimbursement scheme and hence get reimbursed at the cognitive level (ie, as thinkoligist). I think the biggest predictor of reimbursement is how well you bill, and I do not think you address this issue. But did you get a sense general surgeons billed better and captured every billing opportunity or billed less well than trauma surgeons or vice versa?
Answer Mary Lumpkin, M.D. (Tucson, AZ): With respect to the billing codes and the CPEP, I have absolutely no idea. I would have to refer you to our statistician in our billing department. As I said previously, we have had some significant problems with trauma billing. We have a trauma director, Dr John Porter, who came to us in I believe 2001, and he has made a series of changes including requiring the attendings to take in-house call and I think that that is probably been very influential, because it is reflected in the documentation much more strongly. How well do we bill? Well, not very well at that time, but I think those things are being addressed One reason that I think this is a good place to do this study is because we have all services billed completely separately through a private corporation, University Physicians Incorporated, completely separate from the hospital.
Answer Question Mary Lumpkin, M.D. (Tucson, AZ): I am not sure how we would get access to their billing information first of all. It is an interesting concept given that our residency training program is a combination of university based and community based practice. With the closing of the other level one trauma center, we do not have any community based physicians taking trauma call now. Question David Easter, M.D. (San Diego, CA): I think you pointed out that the connection between billing and re-
Sara Hartsaw, M.D. (Gillette, WY): About 10 years ago now in my little hospital, the two general surgeons used this argument to connive, coerce, and cajole the hospital into paying us for taking trauma call. We did not have nearly as good a data as you had, but we were being reimbursed probably in the 70% range and we thought that was appalling, so I guess I am fortunate to be making a lot more money than the university types. Do you expect if you continue this study with the changes your director has made, that your reimbursements now exceeds what the general surgeons are making?
M.F. Lumpkin et al. / The American Journal of Surgery 188 (2004) 767–771
Answer Mary Lumpkin, M.D. (Tucson, AZ): I cannot really predict that without actually looking at more recent numbers, which we have not done.
771
Specifically, the motorcycle players historically have had the reputation of not paying up. Did you get a chance to do this kind of analysis of auto versus pedestrian, car accidents, motor vehicles, and motorcycle accidents?
Answer Question Clayton Shatney, M.D. (San Jose, CA): I realize your numbers are kind of small to break out into subgroup but the general impression among those of us that have historically been in trauma is that among the blunt trauma group there are some discrepancies based on the actual mechanism.
Mary Lumpkin, M.D. (Tucson, AZ): I agree with you. I thought that it was striking that the blunt, and the motor vehicle crash numbers almost all corresponded exactly. We did not really have enough numbers to feel that the other subgroups within blunt mechanisms were significant, such as motorcycle crashes.
Scientific paper
Surgeon reimbursement for trauma care Mary F. Lumpkin, M.D., Daniel G. Judkins, R.N., M.P.H., John M. Porter, M.D., Mark D. Williams, M.D.* Department of Surgery, Section of Trauma and Critical Care, University of Arizona College of Medicine, 1501 North Campbell Avenue, PO Box 245063, Tucson, AZ 85724-5063, USA Manuscript received July 20, 2004; revised manuscript August 7, 2004 Presented at the 56th Annual Meeting of the Southwestern Surgical Congress, Monterey, California, April 18 –21, 2004
Abstract Background: Trauma care is a well-known financial burden for hospitals, yet reimbursement for the surgeon has not been reported. Methods: For 1999, the percent of the surgeons’ bills reimbursed for general surgery services (gPR) was compared with that for trauma services (tPR). Mean tPR for various groups were compared. Factors predictive of tPR lower than gPR were identified. Results: The gPR was 49%, and, for 371 trauma patients, tPR was 45% (P ⫽ 0.03). The mean tPR for injury severity score (ISS) ⱕ10 was 48%, and for ISS ⱖ11, 57% (P ⫽ 0.03). Patients transferred from outside facilities did not have a significantly lower mean tPR. Penetrating trauma (odds ratio 3.7, P ⫽ 0.008) was predictive of tPR lower than gPR. Conclusions: Surgeon reimbursements for trauma care was significantly, yet only slightly less than for all general surgery care. Surgeons should not be reluctant to take trauma call based on perceptions of low reimbursement. © 2004 Excerpta Medica Inc. All rights reserved. Keywords: Trauma; Surgeon; Reimbursement
Trauma care is well known to be a financial burden for most hospitals [1], yet reimbursement for the trauma surgeon has not been previously reported. A paucity of information is published regarding trauma surgeons’ compensation. In her address to the 1998 Annual Meeting of the American Association for the Surgery of Trauma, Dr Ledgerwood pointed out this fact, saying, “Despite multiple publications regarding the uncertain fiscal viability of trauma centers because of uncompensated and under funded care, few manuscripts divulge trauma surgeons’ salaries or reimbursements” [2]. In November, 2000, Fakhry et al [3] reported that the 1998 annual salaries of trauma surgeons ranged from $90,000 to $528,000. Although both these endeavors made important contributions to what is known regarding trauma surgeon compensation, both were based on survey results and neither addressed the percentage of surgeons’ bills reimbursed. We compared the percent of surgeon billing reimbursed for
* Corresponding author. Tel.: ⫹1-520-626-5056; fax: ⫹1-520-6265016. E-mail address: [email protected]
trauma services to the percentage reimbursement for all general surgery services at this institution in 1999.
Materials and Methods All patients at University Medical Center in Tucson, a level 1 trauma center associated with the University of Arizona, are billed for general and trauma surgeon services through a single professional corporation. For 1999, general surgeon and trauma surgeon charges, patient payments, and actual gross collection rates were examined and merged with the trauma register (Hospital Trauma Register: Version 4.3, Richard H. Cales) onto a Microsoft Excel 2000 spreadsheet (Microsoft Corp., Redmond, WA). The percent of the surgeons’ bills that were reimbursed for all general surgeon services (gPRs) was compared with the percent reimbursement for trauma surgeon services only (tPRs) using the chi-square calculation. Mean tPR for various groups were compared using Student t tests. With logistic regression analysis, demographic factors, mechanism of injury, and injuries predictive of the tPR being significantly lower than gPR were determined. Because the
0002-9610/04/$ – see front matter © 2004 Excerpta Medica Inc. All rights reserved. doi:10.1016/j.amjsurg.2004.08.038
768
M.F. Lumpkin et al. / The American Journal of Surgery 188 (2004) 767–771
Table 1 Surgeon billing, reimbursement, and percent reimbursed for trauma and general surgery services in 1999
Total Nontrauma Trauma
Billed
Reimbursed
Overall PR
$3,169,148 $2,894,143 $275,006
$1,543,273 $1,419,132 $124,141
.49 .49 .45
Blunt Penetrating
Number
Billed
Reimbused
Overall tPR
Mean tPR
292 72
$153,976 $115,375
$85,550 $34,047
.56 .30
.57 .37
tPR ⫽ trauma percent reimbursement.
PR ⫽ percent reimbursement.
outcome or dependent variable (yes the tPR was lower than the gPR or no it was not) is binary in nature, logistic regression, an excellent method for estimating the influence of independent variables, was used [4]. These multiple logistic regression analyses were done with Stata 6.0 software (StataCorp., College Station, TX). Data were transferred from Microsoft Excel and the trauma register into the Stata program using Stat/Transfer version 5 software (Circle Systems Inc., Seattle, WA). For categorical independent variables, the Stata software generated appropriate “dummy variables” automatically. A backward stepwise selection process was used, starting with the fully saturated model [5]. Independent variables were eliminated when they failed to meet the P ⫽ 0.05 level of statistical significance. Remaining independent variables were retested for significance. Results are expressed as odds ratios; the odds the tPR was lower than the gPR given the presence of an independent variable.
Results Total general surgeon charges in 1999 were $3,169,148, of which $1,543,273 was collected, yielding an overall percent reimbursement (gPR) of 49%. As shown in Table 1, for the 371 trauma patients (330 admissions), the overall percent reimbursement for trauma surgeon charges (tPR) was significantly lower at 45% (P ⫽ 0.03). The mean tPR for ISS ⱕ10, 48%, was significantly less than for ISS ⱖ11, 57% (P ⫽ 0.03) (Table 2). The mean tPR for penetrating trauma, 37%, was significantly lower than that for blunt trauma, 57% (P ⬍0.005) (Table 3). The mean tPR for assaults, 36%, was significantly lower than for motor vehicle collisions, 57% (P ⬍0.005) (Table 4). Patients transferred in from outside facilities (58 patients) did not have a significantly lower mean tPR than those who were not. Penetrating trauma (odds ratio 3.7, P ⫽ .008) was a signif-
icant predictor of the tPR being lower than the overall gPR of 49%. Admission (odds ratio 4.2) was not quite a significant predictor of tPR being lower than the gPR (P ⫽ 0.06). Surprisingly, suicide attempt (odds ratio 8.1, P ⫽ 0.03) was a significant predictor of tPR greater than the gPR.
Comments Dr Ledgerwood presented 1996 data from 99 centers responding to the American College of Surgeons Verification Program postreview questionnaire and commendably stated her own total wages for the previous year [2]. Her report included both amounts of trauma surgeon compensation and a wide range of reimbursement schemes. Hourly wages ranged from $30 to $75, whereas daily reimbursements were from $300 to $2,000. Some trauma surgeons received $200 per day plus $300 per resuscitation to $1,200 per day plus $100 per resuscitation, and one institution paid surgeons 70% of their charges [2]. In November 2000, Fakhry et al [3] reported 1998 annual salaries of trauma surgeons ranging from $90,000 to $528,000, with a mean of $229,142 ⫾ $78,045 per year. In this study, surgeon reimbursement for trauma care was surprisingly only slightly, yet still significantly less than for general surgery care. The finding of lower reimbursement in trauma is consistent with previously reported evidence that cognitive skills (ie, nonoperative management) reimburse less than operative management [6] and that trauma surgeons spend more time caring for yet bill less than subspecialists involved in trauma care [7]. In a 1991 survey, surgeons who preferred not to care for trauma patients held the perception that trauma care reimburses less than for nontrauma [8]. That survey also showed that reimbursement issues were less influential than other factors in the preference not to treat trauma patients or take trauma call. Perhaps attitudes toward the importance of reimbursement have
Table 4 Trauma etiology
Table 2 Injury Severity Score
ISS ⱕ10 ISS ⱖ11
Table 3 Blunt and penetrating trauma
Number
Billed
Reimbursed
Overall tPR
Mean tPR
219 152
$94,015 $180,991
$41,584 $82,557
.44 .46
.48 .57
tPR ⫽ trauma percent reimbursement; ISS ⫽ injury severity score.
Assault MVC
Number
Billed
Reimbursed
Overall tPR
Mean tPR
74 218
$108,664 $125,689
$31,323 $69,014
.29 .55
.36 .57
MVC ⫽ motor vehicle crash.
M.F. Lumpkin et al. / The American Journal of Surgery 188 (2004) 767–771
changed. Surgeons queried in 2000 ranked financial reward as the most important motivating factor [9]. Trauma surgeon reimbursement was higher for the more severely injured; motor vehicle collisions; and blunt trauma than for the less severely injured, assaults, and penetrating trauma. Surgeon reimbursement for these subgroups was consistent with our previous report on hospital reimbursement [1]. Unlike the hospital reimbursement data, this surgeon reimbursement data does not suggest any “dumping” of patients (the transfer of patients unlikely to reimburse). To our knowledge, this is the first study to actually document reimbursement for trauma care. Perhaps surgeons should not be reluctant to take trauma call based on perceptions of low reimbursement because this reimbursement is only slightly less than for general surgery physician services.
References [1] Lanzarotti S, Cook CS, Porter JM, Judkins DG, Williams MD. The cost of trauma. Am Surg 2003;69:766 –70. [2] Ledgerwood AM. The thrill of victory; at what price? J Trauma 1999;46:1– 8. [3] 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. [4] Al-Ghamdi AS. Using logistic regression to estimate the influence of accident factors on accident severity. Accid Anal Prev 2002;34: 729 – 41. [5] Hosmer DW, Lemeshow S. Applied Logistic Regression. New York: John Wiley & Sons, 1989. [6] Sutyak JP, D’Amelio LF, Chiu WC, Hammond JS. Cognitive trauma care is undervalued: adult splenic injury as a paradigm. Am Surg 1997;63:752–7. [7] Rogers FB, Osler T, Shackford SR, Healey MA, Wells SK. Charges and reimbursement at a rural level I trauma center: A disparity between effort and reward among professionals. J Trauma 2003;54:9 –15. [8] Esposito TJ, Maier RV, Rivara FP, Carrico CJ. Why surgeons prefer not to care for trauma patients. Arch Surg 1991;126:292–7. [9] Leitch KK, Walker PM. Surgeon compensation and motivation. Arch Surg 2000;135:708 –12.
Discussion Jeffery Saffle, M.D. (Salt Lake City, UT): Dr Lumpkin and her colleagues have debunked one of the frequent excuses given by surgeons for not taking trauma call, but their data need to be viewed in the wider context of the issue surrounding the future of trauma care in the United States. Dr Donald Trunkey has estimated that perhaps as few as 10% of practicing surgeons in the United States continue to care for trauma. The reasons cited are multiple, first of which is that trauma has become increasingly a nonoperative specialty as one of the papers today illustrates. The group at North Carolina has suggested that a surgery resident today will need to participate in the care of 500 patients to perform one splenectomy and one operative repair of a
769
liver injury. In the future, as the population of America ages, penetrating injuries will likely continue to decline and low energy trauma, such as ground level falls, will increase this trend. Trauma surgeons will have to have other practices to maintain their skills. Otherwise, they will find themselves resigned to the last thing they ever desired to be, cognitive specialists. Simultaneously, trauma centers have proven their effectiveness and prospered, but equally quickly many hospitals and surgeons have found it convenient to abdicate their role in providing this care. Competition for bed space and other resources, the allure of more lucrative and convenient specialty practices, and the increasingly dominating issue of lifestyle all make it so attractive for surgeons to get out of the trauma business. As a result, the remaining trauma centers are inundated with low acuity cases, contributing to trauma surgeon burnout and what JD Richardson has called de facto specialization. Taken against this background, Dr Lumpkin’s data contain both good and bad news. I am glad reimbursement is so good, but, on the other hand, 49% of a little is still a little. By my calculations, the patients in this series were billed an average of $741.00 each and paid an average of $334.00 each. The total reimbursement for trauma care, just over $124,000.00, is far too little to pay even a single surgeon’s salary, even Dr Anna Ledgerwood’s very small salary. Data from Vermont have confirmed what we already know. Specialists, including radiologists, bill five times as much as we do, even though they contribute relatively little ongoing care. Hospitals make money as well. Thus, surgeons who provide 24 hour a day call coverage, triage, resuscitation, workup, and follow-up care make money for everyone but themselves. This brings me to three questions for Dr Lumpkin. First, does your hospital pay surgeons to take trauma call and is this appropriate? Second, is trauma call at your hospital incorporated into general emergency call, so your surgeons can make a living with appendectomies while indulging their hobby of trauma care? And, finally, does your hospital maintain a surgeon-led, dedicated trauma service for the subsequent care of these patients after their orthopedic or neurosurgical procedures? I do have one final comment. I think it is critically important that the issue of surgeon involvement in trauma is discussed at meetings like this one, rather than only at subspecialty trauma meetings, so that we can continue to develop a trauma care system, that is as inclusive of as many surgeons as possible. And, as a last question, if you could let me know the hospital that pays $528,000.00 for trauma surgeons, I’d like to talk to them.
Answer Mary Lumpkin, M.D. (Tucson, AZ): First of all, the total amount billed represented in the article only represents a very small portion of the trauma patients. In the past at our facility, we have had some significant billing problems for trauma surgeon services that are in the
770
M.F. Lumpkin et al. / The American Journal of Surgery 188 (2004) 767–771
process of being improved, or, as I understand it, significantly updated. As a resident, it is not something that I am privy to. For instance, the number of admitted patients we looked at represented about one third of all the trauma admissions for that year. Second these data are from 1999. Subsequent to that, we became the only level one trauma center in Southern Arizona when Tucson Medical Center withdrew from trauma care. Do our trauma surgeons get paid to take call? I do not believe so, but, as a resident, it is not something that I would necessarily know. Another change that has occurred subsequent to this data being studied is that our trauma surgeons are now in-house and they are required to be in the trauma bay within a certain amount of time depending on the level of acuity of the trauma. As for your second question, no, trauma call is not separate from other general surgery emergencies. Our system is evolving. The trauma service takes care of the general surgery patient at night and operates if need be, but during the day other services take surgery consults and calls. Those general surgery patients operated at night by the trauma service are often postoperatively transferred off of the trauma service. As you might imagine, we have experienced a humungous burden by the doubling of our trauma population when the other trauma center closed. Do we have a dedicated trauma service? Yes, we do. We have a trauma service staffed by residents who are increasingly stretched thin with the resident work-hour restrictions and our doubling of trauma patients, which all happened on the same day at our facility. Question Ronald Stewart, M.D. (San Antonio, TX): For the residents in the audience, you can make a good living doing trauma surgery. Having said that, I have one really nagging question. I am not certain you used the right control. Your control was your university surgeons. Do you think you should have added another control of community surgeons in private practice?
imbursement is sometimes quite tenuous. The dominant influence between billing and reimbursements is the Federal Government. They have put together risk pools, and general surgeons (eg, vascular, oncology, and trauma) are all in the same risk pool. But sometimes, we have to bill outside of our risk category called CPEP. Don’t ask me what it stands for. For example, if I, as a general surgeon, read an EKG, I bill under the cardiologists’ codes. So my question is, do your trauma surgeons predominantly bill within general surgery coding (ie, as do-ologists) or do they “crosswalk” outside of their CPEP into somebody else’s reimbursement scheme and hence get reimbursed at the cognitive level (ie, as thinkoligist). I think the biggest predictor of reimbursement is how well you bill, and I do not think you address this issue. But did you get a sense general surgeons billed better and captured every billing opportunity or billed less well than trauma surgeons or vice versa?
Answer Mary Lumpkin, M.D. (Tucson, AZ): With respect to the billing codes and the CPEP, I have absolutely no idea. I would have to refer you to our statistician in our billing department. As I said previously, we have had some significant problems with trauma billing. We have a trauma director, Dr John Porter, who came to us in I believe 2001, and he has made a series of changes including requiring the attendings to take in-house call and I think that that is probably been very influential, because it is reflected in the documentation much more strongly. How well do we bill? Well, not very well at that time, but I think those things are being addressed One reason that I think this is a good place to do this study is because we have all services billed completely separately through a private corporation, University Physicians Incorporated, completely separate from the hospital.
Answer Question Mary Lumpkin, M.D. (Tucson, AZ): I am not sure how we would get access to their billing information first of all. It is an interesting concept given that our residency training program is a combination of university based and community based practice. With the closing of the other level one trauma center, we do not have any community based physicians taking trauma call now. Question David Easter, M.D. (San Diego, CA): I think you pointed out that the connection between billing and re-
Sara Hartsaw, M.D. (Gillette, WY): About 10 years ago now in my little hospital, the two general surgeons used this argument to connive, coerce, and cajole the hospital into paying us for taking trauma call. We did not have nearly as good a data as you had, but we were being reimbursed probably in the 70% range and we thought that was appalling, so I guess I am fortunate to be making a lot more money than the university types. Do you expect if you continue this study with the changes your director has made, that your reimbursements now exceeds what the general surgeons are making?
M.F. Lumpkin et al. / The American Journal of Surgery 188 (2004) 767–771
Answer Mary Lumpkin, M.D. (Tucson, AZ): I cannot really predict that without actually looking at more recent numbers, which we have not done.
771
Specifically, the motorcycle players historically have had the reputation of not paying up. Did you get a chance to do this kind of analysis of auto versus pedestrian, car accidents, motor vehicles, and motorcycle accidents?
Answer Question Clayton Shatney, M.D. (San Jose, CA): I realize your numbers are kind of small to break out into subgroup but the general impression among those of us that have historically been in trauma is that among the blunt trauma group there are some discrepancies based on the actual mechanism.
Mary Lumpkin, M.D. (Tucson, AZ): I agree with you. I thought that it was striking that the blunt, and the motor vehicle crash numbers almost all corresponded exactly. We did not really have enough numbers to feel that the other subgroups within blunt mechanisms were significant, such as motorcycle crashes.