Duke Orthopedic Journal

ORIGINAL ARTICLE
Year
: 2020  |  Volume : 10  |  Issue : 1  |  Page : 26--31

The effect of initiation of an orthopedic trauma service on patient outcomes and clinical enterprise financial performance


Carson L Sanders, William R Barfield, Kit N Simpson, Dudley Colhoun, Langdon A Hartsock 
 Department of Orthopaedics and Physical Medicine, Medical University of South Carolina, Charleston, South Carolina, USA

Correspondence Address:
Dr. William R Barfield
Department of Orthopaedics and Physical Medicine, Medical University of South Carolina, 96 Jonathan Lucas Street, Suite 708, Charleston, South Carolina 29425
USA

Abstract

Background: As the demands of patients and the health-care reimbursement system continue to change rapidly, hospitals and surgeons are adopting new methods of delivering high-quality care at a lower cost. To this end, the use of a daily, dedicated orthopedic trauma room has been increasing in popularity. Our institution, a major Level I academic trauma center, however, has been a late adopter, only starting a dedicated orthopedic trauma room on November 1, 2013. The purpose of this retrospective data review was to assess the clinical (intensive care unit [ICU] length of stay [LOS] and hospital LOS) and financial outcomes of a dedicated trauma room. Design: This is a retrospective analysis of pre- and post-intervention measures of patient outcomes and financial performance using archival data from a trauma registry linked to cost, charge, and payment records. Methods: Our trauma registry was used to identify admission of patients requiring surgery for orthopedic injuries involving the lower extremity from the hip to the ankle for 11 months preceding the adoption of a daily trauma room (pre) and 11 months following its implementation (post). We compared pre- and post-hospital LOS and ICU LOS, while controlling for the effect of race, age, gender, Injury Severity Score (ISS), and type of insurance. There were 243 eligible patients identified for the “post” group and 258 in the “pre” group. Results: We found no statistically significant difference between the groups in mean age, ISS, or in distribution of patient sex or race. The estimated mean LOS (controlling for patient characteristics) was 8.35 days in the “pre” group and 7.79 days in the “post” group. This represented a statistically significant reduction (P < 0.0223) in overall LOS by more than a half-day per admission. Moreover, the mean estimated number of days in the ICU (for patients with any ICU use) decreased from 8.18 in the “pre” period to 5.94, a 2.24-day reduction (P < 0.0001) in the months following the adoption of the trauma room. This improvement in LOS and ICU use was not reflected in the observed difference between the pre- and post-time period in mean hospital charges (P = 0.5524), hospital cost (P = 0.5590), hospital payments (P = 0.8350), provider charges (P = 0.1985), or provider payments (P = 0.6388). However, the reductions in mean LOS and ICU used were estimated to free up 680.4 bed days, which would allow an additional 109.1 admissions per year, resulting in $1,299,498 estimated additional revenue to the hospital. Conclusions: These findings represent a significant improvement over the previous system and have implications for overall patient outcomes and also financial outcomes. Other studies have suggested that decreased total hospital and ICU stay have both been associated with overall better patient outcomes. Furthermore, the decreased LOS, especially in the ICU, frees scarce capacity in an institution experiencing a chronic shortage of available ICU beds. Our study illustrates the importance of capturing both resource use (such as bed days) and cost when evaluating the effect of process improvements in large hospitals. Costs and revenue measures alone may not capture the true economic benefits of process improvements in institutions where resources that are freed up by one service may be used by other service lines.



How to cite this article:
Sanders CL, Barfield WR, Simpson KN, Colhoun D, Hartsock LA. The effect of initiation of an orthopedic trauma service on patient outcomes and clinical enterprise financial performance.Duke Orthop J 2020;10:26-31


How to cite this URL:
Sanders CL, Barfield WR, Simpson KN, Colhoun D, Hartsock LA. The effect of initiation of an orthopedic trauma service on patient outcomes and clinical enterprise financial performance. Duke Orthop J [serial online] 2020 [cited 2024 Mar 28 ];10:26-31
Available from: https://www.dukeorthojournal.com/text.asp?2020/10/1/26/316554


Full Text



 Introduction



The conventions of health-care delivery continue to change at an ever-increasing rate. Whether driven by new clinical research or by changing reimbursement practices, clinicians and hospitals are constantly attempting to find new methods of delivering better care in the most efficient way possible.

In this light, care delivery for the orthopedic trauma patient has been undergoing recent changes. Previously, patients with fractures requiring surgical intervention were placed on the “add-on list”, going to surgery once operating room (OR) time and surgeons were available, often late in the day or at night. These cases were often performed after an elective OR day by a surgeon other than a dedicated traumatologist.[1] Aside from being disruptive to surgeon's personal lives as well as the OR schedule, there are clinical implications as well. Nighttime surgery has been associated with increased operative time and increased 1-year and 2-year mortality for hip fractures, while more minor complications such as painful hardware have been associated with after-hours intramedullary fixation of femur fractures.[2],[3]

A dedicated orthopedic trauma room (DOTR) is a daily, daytime OR staffed by a fellowship-trained traumatologist and reserved for the treatment of nonemergent orthopedic trauma. This type of dedicated OR has been associated with an overall decrease in the number of after-hours orthopedic procedures performed.[4] Furthermore, this change has been reported to decrease the use of OR resources after-hours while improving and expanding the utilization of previously scheduled elective ORs.[5],[6] As a result, the implementation of DOTRs has gained broad popularity at hospitals across the nation as hospitals and physicians continue to attempt to find more efficient ways of delivering higher quality care.[7],[8]

Despite this growing practice, our institution, a large regional academic Level I trauma center, was a late-adopter of the daily orthopedic trauma room. To our knowledge, we were the last major trauma center in our region to do so. With the hiring of an additional fellowship-trained orthopedic traumatologist our department lead the effort to improve musculoskeletal care with the advent of the DOTR in late 2013. As a result, we have a unique perspective by which to evaluate the results of the DOTR on patient outcomes and hospital and physician financial outcomes. We were particularly interested in whether the trauma room helped with overall trauma throughput, that is, the ability for the trauma room to shorten the period of time that patients spent receiving care following trauma admission. In addition, we were interested to see if a DTOR would change hospital costs, charges, revenues, or physician revenue.

The objective of our study was to assess the effects of the implementation of a DOTR at our Level I trauma center on measures of patient outcomes and enterprise financial measures. Patient outcomes were measured using length of intensive care unit (ICU) stay and length of hospital stay for (length-of-stay [LOS]) a cohort of patients with lower extremity injuries who were admitted to our hospital trauma service. We hypothesized that a dedicated, daily OR staffed by a trained traumatologist would result in a reduction in overall LOS for our cohort of patients as well as decreased ICU occupancy, when compared to a similar cohort of patients treated prior to the implementation of the dedicated OR. In addition, we hypothesized that a DOTR would reduce hospital charges and costs, but would not change physician revenue.

 Methods



This study is a retrospective analysis of pre- and post-intervention measures of patient outcomes and financial performance using archival data from a trauma registry linked to cost, charge, and payment records.

On November 1, 2013, our institution adopted the use of a DOTR. This development marked the joint effort of the orthopedic department along with perioperative services and OR staff. The OR's implementation also marked the completion of our complete orthopedic trauma service with a fellowship-trained orthopedic traumatologist available to staff the OR each day. With few exceptions, the OR is utilized for the treatment of inpatients who require fixation of orthopedic injuries. At our institution, the vast majority of such patients are admitted to our general surgery trauma and acute care service with orthopedics as a consultant service.

Institutional review board approval was acquired prior to study initiation. Our institution's trauma registry was the main source for identifying a cohort of trauma patients admitted to the general surgery trauma service who had undergone operative fixation of lower extremity injuries from the hip to the distal tibia (femur and/or tibia). Using ICD-9 procedure codes, we queried the trauma registry and extracted data for all patients who had undergone an operative procedure involving the femur or tibia during the study time period. These codes included an array of procedures that generally require use of the OR such as applications of external fixators, open reduction and internal fixation, as well as open reduction of dislocations. The complete list of ICD-9 procedure codes identified is detailed in [Table 1]. We selected this wide subset of surgical patients to allow for a broad look at the effect of our DOTR. Data were extracted for admissions with associated ICD-9 procedural codes in a time period from 12 months prior to the DOTR implementation and 12 months post DOTR implementation. The “pre”-DOTR timeframe for admission was November 1, 2012, through October 31, 2013. The “post” DOTR timeframe was November 1, 2013, through October 31, 2014. Data on date of admission, Injury Severity Score (ISS), sex, race, age, insurance status, hospital LOS, and ICU LOS were extracted from our EPIC trauma registry. Patients' medical record numbers were used to extract hospital charge, cost, and payment records as well as provider relative value unit (RVU) and payment data from hospital and practice financial systems (EPIC). Charge and cost data were aggregated at the level of the patient and documented separately for hospital charges, cost and payments, and provider charges and payments. All patients in the data set had complete records of charges, hospital cost, and provider RVUs. Some patients had zero values for either hospital or provider payments (or both). This is a routine occurrence in a hospital that delivers a disproportional share of care to indigent, uninsured patients. To prevent this institutional characteristic from biasing our payment analysis, we estimated the expected payments using mean charge-to-payment ratios derived from all admissions with the recorded payments. Missing hospital payments were imputed as Charges × (payment/charge ratio for group with payments). Expected provider payments were calculated based on the charge-to-payment ratio per RVU for patients with the recorded payments. The data set construction and analysis was performed using SAS statistical software version 9.4 (Cary, NC, USA). The pre- and post-implementation groups were compared using Chi-square tests for categorical variables and t-test or Wilcoxon test as appropriate for continuous variables. Outcomes of interest were also compared using multivariable modeling controlling for patient characteristics. Poisson models were used for LOS data and log-linked models for cost data. Statistical significance was assumed to be P < 0.05.{Table 1}

 Results



A total of 501 complete patient records were extracted from the trauma registry and linked to the financial data. Two hundred and fifty-eight patients received care during the pre-DOTR period, whereas 243 were treated during the post-DOTR time period. The two patient cohorts demonstrated similar demographic characteristics [Table 1]. The mean ISS were similar across the pre- and post-period [Table 1]. A nonsignificant difference in the distribution of ISSs was observed (P > 0.05). Of the 258 patients in the “pre” period, 71 were considered severely injured (ISS > 15) and 54 were severely injured in the “post” group (P = 0.17).

Males made up the large majority of subjects in the groups, 64.3% of the “pre” population and 58.9% of the “post” cohort. In the “pre” cohort, 62.4% of the patients were of African-American/Black/minority race (P = 0.27). The mean age of the patients was 47.0 years and 47.5 years (P = 0.23), in the pre- and post-groups, respectively. The two groups were also compared on insurance status. The patients were categorized into one of the following categories: (1) Medicaid, (2) Medicare, (3) private insurance, (4) self-pay/unknown, and (5) “other.” Comparing the two groups, the distribution among each of these insurance categories did not statistically significantly differ (P = 0.10). The most commonly cited insurance type in the “pre” period was self-pay/unknown, whereas private insurance was most common in the “post” period. In both cohorts, Medicaid was the least common. Thus, even though we used data from every clinically eligible patient who was admitted in the pre- and post-periods, the populations were very similar, with no statistically significant difference (P > 0.05) in the demographic composition of the two groups [Table 1].

The main outcomes of interest measured in the study were LOS in hospital, use of ICU days, and financial differences measured as hospital and provider charges, cost, and payments. We examined and reported both observed mean values and estimated values that adjust for the potential effects of differences in patient characteristics. In the observed data comparison of LOS for the groups, we found a mean overall length of hospital stay of 9.26 ± 10.33 days in the pretrauma room group, compared to 8.77 ± 12.77) hospital days in the posttrauma room group. This represented a mean reduction in overall hospital LOS of approximately a half-day (0.49) per admission for the group of patients treated after the adoption of the DOTR (P = 0.77). This difference in overall LOS was statistically significant [Table 2] when the groups were compared using multivariable modeling controlled for effects of age, race, sex, ISS score, and insurance type (P < 0.02).{Table 2}

Furthermore, when comparing the length of time in the ICU for patients who were admitted to the ICU, we observed a large difference between the groups in the mean number of days spent in the ICU. For the “pre” group, the mean ICU LOS was 8.18 ± 7.63 days, whereas the “post” group was found to spend an average of 6.30 ± 7.58 days in the ICU. This is a reduction of 2.49 days (P = 0.006) per patient admitted to the ICU in the post time period. This difference was slightly attenuated to 2.24-day reduction in the multivariable model estimate when estimates were adjusted for any differences attributable to the effects of age, sex, race, ISS score, and insurance coverage (P < 0.0001).

The mean hospital charge prior to the DOTR was $107,938 and $105,824 after initiating the DOTR [Table 3]. Hospital costs decreased slightly after starting the DOTR from $33,950 to $32,828. This was not a statistically significant change (P > 0.05). Hospital payments were nearly identical at $26,580 pre-DOTR to $26,280 after starting the DOTR.{Table 3}

The observed mean physician charges were $10,528 prior to the DOTR and $11,293 after starting the DOTR for a difference of $765 [Table 4]. This was not a statistically significant difference (P > 0.05). Payments went down by $81 from $1950 to $1869 after starting the DOTR. The mean RVU per patient remained nearly the same at 28.7 pre DOTR to 29.5 post DOTR initiation.{Table 4}

Overall, the hospital collected 13 times the physician revenue during the study period. For the physicians, the average charge/RVU was $366.83 and the average payment/RVU was $67.94 prior to the DOTR and $63.36 after starting the DOTR.

 Discussion



Caring for orthopedic trauma presents a challenge for physicians and hospitals alike. To answer this challenge, dedicated orthopedic trauma ORs staffed by trained orthopedic traumatologists have gained significant popularity. These DOTRs have demonstrated an ability to decrease the amount of surgery performed at night as well as improve the rapidity with which patients can be surgically treated.[7],[9] These enhancements have potential benefit for patient care as well as improvement in overall hospital operation. The current study attempts to build upon these findings.

The results of our study suggest that DOTRs have an association with a decrease in overall hospital LOS as well as ICU LOS. The greatest difference was seen for ICU stay. Given the demographic and injury severity similarities of the two groups, our “post DOTR” cohort group with lower extremity injuries requiring surgery spent significantly less time receiving intensive care. On an average, they spent 2.2 days less in the ICU. This is an important finding indicating improved quality of care because decreased ICU LOS is usually associated with a quicker recovery, less time on a ventilator, and an overall improved hospital course. In addition, given the significant cost associated with ICU care, there is a potential of substantial cost savings to the hospital. In addition, at a major medical center, there is continuous demand for ICU beds. Thus, it is highly unlikely that the ICU days avoided will not be used to care for other patients. This point is important because it is possible that the process change effected by the DOTR may have unmeasured benefits accruing to nonorthopedic patients.

To a lesser though statistically significant degree, adoption of the DOTR was associated with decreased overall hospital stay. Patients treated in the post-DOTR era spend an average of a half-day less in the hospital. This result in fact contradicts the existing literature regarding overall LOS in the setting of a DOTR. Elder et al. studied the outcomes of a large number of patients treated for subcapital femoral neck fracture in both a nontrauma room and trauma room settings at two separate trauma centers, finding a 4-day increase in the overall stay for patients treated at the center that utilized a DOTR.[7],[9]

Taken together, these findings strongly support the beneficial role of DOTRs in the acute care of patients with lower extremity trauma and the favorable effects of dedicated OR time to allow for improved patient throughput in the hospital setting.

By studying the before-and-after data at a single trauma center, our study helps to eliminate potential differences among clinical practices at different institutions, which could affect the LOS. This is one of the main strengths of our study. Other strengths include the continuous, yet concise time period over which our patient population was evaluated. Improvement in ICU and overall LOS cannot be attributed to advancing trauma/critical care techniques over our time period. The late adoption of our DOTR and the accompanying hiring of an additional traumatologist gives us a unique opportunity to see the immediate effects of the trauma service.

Our study also has several weaknesses. In particular, patients requiring surgical care for their lower extremity injuries, but not admitted through the general surgery trauma service, and thus not entered into the trauma registry, were excluded from our patient cohort. Although these represent a much smaller number of patients, the reported benefits of the trauma room could be affected by this omission. Similarly, by narrowing our focus to injuries involving just the femur and tibia, we did not include a large array of other orthopedic injuries that would certainly require care by our trauma service. Finally, while we are unaware of any specific changes to our trauma service, other hospital variables could have certainly changed during the course of our study. While we believe that the DOTR at our institution played a large role in reducing length of ICU and hospital stay, the nonrandomized, before-and-after design of our study does not allow us to interpret our findings as causal. Thus, further studies are needed to confirm the benefit that we observed.

Finally, our findings raise important issues regarding the lack of correlation that we observed between improvements in LOS and financial measures. From the outset, we wanted to measure the financial impact of the DOTR from a cost and revenue perspective because it is important to examine the effects of process improvements on financial measures used by the enterprise. We were surprised to find no differences in charges, cost, or revenue measures for the two time periods, and puzzled why these measures did not correlate with the findings in LOS. The performance of this study helped us identify several areas where improvements are needed in our financial data system in order for us to be able to use charge, cost, and revenue data to inform process improvements in orthopedic surgery. We found that the current system worked well for charges and hospital costs, but that our high rate of charity cases and the cash-accounting approach used to monitor payments for professional services did not lend themselves to accurately capture provider payments. Furthermore, the usual approach of comparing population means of cost and charges is inadequate for identifying benefits that accrue to the enterprise because resources previously used by orthopedic patients are now freed up for use by other service lines. This is an essential finding for others to consider when attempting to measure the effects of improving service line processes.

 Conclusions



Our study supports the premise that the implementation of DOTR staffed by qualified traumatologists is of benefit to patients and hospitals. The benefit is best observed by assessing an improvement in ICU and overall hospital LOS. While this finding may have other contributing factors, the decrease in ICU and hospital LOS was facilitated by the increased availability of a daytime OR for orthopedic trauma. We found that this change in resource use was not reflected in billing and collections information for our cohort of patients. However, when we estimated the payments that would be expected from freeing up the resources, additional payments of $900,000 to $1.3 million, depending on financial assumptions [Table 5], would be expected from payments for care provided to patients using the freed bed capacity.{Table 5}

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1Bhattacharyya T, Vrahas MS, Morrson SM, Kim E, Wiklund RA, Smith RM, et al. The value of the dedicated orthopaedic trauma operating room. J Trauma 2006;6:1336-40.
2Ricci WM, Gallagher B, Brandt A, Schwappach J, Tucker M, Leighton R. Is after-hours orthopaedic surgery associated with adverse outcomes? A prospective comparison study. J Bone Joint Surg Am 2009;91:2067-72.
3Min W, Wolinsky PR. The dedicated orthopedic trauma operating room. J Trauma 2011;71:513-5.
4Wixted JJ, Reed M, Eskander MS, Millar B, Anderson RC, Bagchi K, et al. The effect of an orthopedic trauma room on after-hours surgery at a level one trauma center. J Orthop Trauma 2008;22:234-6.
5Heng M, Wright JG. Dedicated operating room for emergency surgery improves access and efficiency. Can J Surg 2013;56:167-74.
6Althausen PL, Davis L, Boyden E, O'Mara TJ, Uppal R, Bray TJ. Financial impact of a dedicated orthopaedic traumatologist on a private group practice. J Orthop Trauma 2010;24:350-4.
7Featherall J, Bhattacharyya T. The dedicated orthopaedic trauma room model: Adopting a new standard of care. J Bone Joint Surg Am 2019;101:e120.
8Steeby SF, Harvin WH, Worley JR, Della Rocca GJ, Volgas DA, Stannard JP, et al. Use of the dedicated orthopaedic trauma room for open tibia and femur fractures: Does it make a difference? J Orthop Trauma 2018;32:377-80.
9Elder GM, Harvey EJ, Vaidya R, Guy P, Meek RN, Aebi M. The effectiveness of orthopaedic trauma theatres in decreasing morbidity and mortality: A study of 701 displaced subcapital hip fractures in two trauma centres. Injury 2005;36:1060-6.