The American Journal of Surgery (2014) 208, 511-516
Association of Women Surgeons
Protecting trauma patients from duplicated computed tomography scans: the relevance of integrated care systems Amy E. Liepert, M.D.a,*, Joseph Bledsoe, M.D.b, Mark H. Stevens, M.D.c, Amalia Cochran, M.D.d a
Department of Surgery, University of Wisconsin, Madison, WI, USA; bDepartment of Emergency Medicine and cDepartment of Surgery, Intermountain Medical Center, Murray, UT, USA; dDepartment of Surgery, University of Utah, Salt Lake City, UT, USA
KEYWORDS: Trauma care; Interfacility transfer; Duplicated CT scans
Abstract BACKGROUND: Duplicated computed tomography (CT) scans in transferred trauma patients have been described in university-based trauma systems. This study compares CT utilization between a university-based nonintegrated system (NIS) and a vertically integrated regional healthcare system (IS). METHODS: Trauma patients transferred to 2 Level I trauma centers were prospectively identified at the time of transfer. Imaging obtained before and subsequent to transfer and the reason for CT imaging at the Level I center were captured by real-time reporting. RESULTS: Four hundred eighty-one patients were reviewed (207 at NIS and 274 at IS). Ninety-nine patients (48%) at NIS and 45 (16%) at IS underwent duplicate scanning of at least one body region. Inadequate scan quality and incomplete imaging were the most common reason category reported at NIS (54%) and IS (78%). CONCLUSIONS: Fewer patients received duplicated scans within the vertically IS as compared with a traditional university-based referral system. Our findings suggest that the adoption of features of a vertically IS, particularly improved transferability of radiographic studies, may improve patient care in other system types. Ó 2014 Elsevier Inc. All rights reserved.
Amy E. Liepert had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. There were no relevant financial relationships or any sources of support in the form of grants, equipment, or drugs. The authors declare no conflicts of interest. * Corresponding author. Tel.: 11-608-265-6246; fax: 11-608-2529746. E-mail address: [email protected] Manuscript received January 15, 2014; revised manuscript May 19, 2014 0002-9610/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2014.05.014
Trauma systems in the United States have evolved as a series of regional networks that direct patients to designated trauma centers. Patients injured in rural areas are typically stabilized at critical access hospitals and those with significant traumatic injury are transferred to state or American College of Surgeons (ACS)-designated trauma centers. Triage of injured patients in rural medical centers is often variable depending on the facility and provider resources that are locally available. Several factors contribute to the performance of initial computed tomography (CT) scanning at the referring center, including fear of litigation and a desire to provide the receiving trauma center with a
512 definitive diagnosis.1 Nevertheless, many patients have imaging performed at the referring hospital, which is then duplicated at the receiving trauma center.2,3 Patients are also frequently subjected to additional or completion CT studies at the trauma center for a comprehensive trauma evaluation. This may result in inefficiency, redundancy, patient discomfort, increased radiation exposure, and increased cost. The majority of ACS-designated Level 1 trauma centers in the United States are at academic medical centers (AMCs).4 Integrated healthcare systems exist in which a single organization owns or administers multiple hospitals within a geographic region, including a Level 1 or 2 trauma center as their referral center. Vertically integrated systems (ISs) use the same electronic medical record and electronic radiology system throughout their facilities. This structure allows for access to patient data by the receiving trauma center when patients are transferred within the system. Prior studies have evaluated duplicated CT scanning in AMC-based tertiary referral systems but have not included data from a vertically integrated trauma system. The goal of this study is to compare the difference in CT utilization frequency and reasons for duplicated or additionally obtained CT scans at 2 regional Level 1 trauma centers, 1 AMC-based nonintegrated center, and the other within a vertically integrated regional healthcare system. Our hypothesis was that an IS would have a lower rate of duplicated CT scans when compared with a nonintegrated system (NIS).
Methods This is a prospective cohort study of patients referred to one of the 2 regional Level 1 adult trauma centers in Salt Lake City, UT, over a 6-month period. The University of Utah Medical Center is an urban ACS-verified Level 1 trauma center within an AMC with an annual average trauma admission volume of 1,100 patients. Intermountain Medical Center (IMC) is a university-affiliated teaching hospital with an ACS-verified Level 1 trauma center. It serves as the flagship hospital within the vertically integrated Intermountain Healthcare (IHC) system. The IHC network consists of 22 hospitals in 2 states with fully integrated electronic medical record and radiology picture archiving and communication system (PACS) systems. The average annual trauma admission volume at IMC is 1,800 patients per year. Each hospital has both fixed-wing and helicopter-based airmedical transport systems and standard ground emergency medical system availability. Each serves approximately the same 4-state catchment area. The University of Utah Medical Center is representative of a nonintegrated healthcare system for the purposes of this study and will be referred to as NIS, while IMC is representative of an integrated healthcare system and will be referred to as IS. Transferred trauma patients who received a CT scan at a referring hospital were prospectively identified at their time
The American Journal of Surgery, Vol 208, No 4, October 2014 of arrival at each of the Level 1 trauma centers. If a patient did not have any CT scans at either the referring or the receiving center, then they were excluded. The CT images were recorded by the body region(s) obtained and whether the images were transferred on disc or electronically. All CT images on CD discs were scanned into the trauma centers’ electronic radiology system, although this process was not always immediately available. The discs were collected for later review by the principle investigator at NIS and by a designated individual at IS. If the disc did not contain the CT studies completed before transfer, then the medical record, including referring hospital documentation, was examined for evidence of studies performed before the transfer. CT studies ordered at the receiving facility were included if they were ordered in the trauma bay by the trauma team leader or subspecialty physicians caring for the patient. CT scans were recorded in the database categorized by body region. The separate CT body regions included the following: head, cervical spine, thoracic spine, lumbar spine, chest, abdomen, pelvis, and other. Data collection forms were secured in the emergency department, and then were collected by a study designee. Duplicated scans were defined as any body region that had already been CT scanned at the referring hospital. Additional scans were defined as any body region that was not scanned before the trauma patient’s transfer to the tertiary trauma facility. A trauma team member completed a data collection sheet in real time that indicated the reason for duplicate or additionally obtained CT scans. The documented reasons were categorized into one of the 5 following categories: technical difficulty; change in clinical condition; inadequate, incomplete, or poor quality images; physician preference; or other. Technical difficulty included images not included on disc, inability to open images on disc, or no disc/images transferred with the patient. If a reason for CT imaging was not identified, the principle investigator contacted the trauma team to obtain the reason for imaging obtained at the receiving trauma center. If a reason was unable to be provided, it was documented as unknown. Data entry was completed within 72 hours of transfer. The study PI was responsible for reviewing all entered data. Statistical analysis was performed using STATA version 11.2 (College Station, TX). Patient demographics were compared using chi-square test. The Wilcoxon rank-sum test was used for comparison of reason for repeat CT scans. Fisher’s exact test and chi-square tests were used for further comparative analysis as appropriate. Institutional review board approval was obtained at both institutions.
Results Four hundred ninety-nine patients were evaluated from the 2 centers. The NIS cohort included 217 consecutive trauma transfers between December 1, 2009 and May 28, 2010. The IS evaluated 282 consecutive trauma transfers
A.E. Liepert et al. Table 1
Duplicated CT scans: system effects
513
Duplicated CT scans by body region Duplicated CT scans
Body region
NIS
Total Head Cervical spine Thoracic spine Lumbar spine Chest Abdomen Pelvis Other
178 60 61 12 12 7 9 11 5
P value (NS . .05)
IS (76%) (33.7%) (34.3%) (6.7%) (6.7%) (3.9%) (5.1%) (6.2%) (2.8%)
56 17 20 4 3 3 2 4 3
(24%) (30.3%) (35.7%) (7.1%) (5.4%) (5.4%) (3.6%) (7.1%) (5.4%)
,.001 NS NS NS NS NS NS NS NS
CT 5 computed tomography; IS 5 integrated system; NIS 5 nonintegrated system; NS 5 nonsignificant.
from February 22, 2010 to August 27, 2010. Nineteen patients were excluded from the study, 10 at NIS and 9 at IS, for no CT scans obtained at either the referring facility or receiving trauma center. Total included patients were 480, 207 at the NIS and 273 at the IS. Of the 273 patients included at the IS, 184 (68%) were transferred from within the IS, while 89 (32%) were referred from a hospital outside the IS. The overall median age of enrolled patients was 47 years, 44 years (interquartile range 32 to 70) at the NIS and 51 years (interquartile range 30 to 65 years) at the IS. Four hundred sixty-eight (97.5%) patients of the entire study population were blunt trauma patients, 198 (93.5%) at the NIS and 270 (97.7%) at the IS. The ISS at the NIS was 12.5, while at the IS it was 11 (P 5 .053). Duplicate CT scans were performed on a total of 144 patients, for a total of 234 duplicated body region scans. The NIS had 178 duplicated body region scans and the IS had 56 duplicated body region scans (Table 1). Of the 207 patients included at the NIS, 99 patients (48%) received duplicated CT imaging. Duplicate CT imaging was
performed on 45 (16%) of the 273 patients evaluated at the IS (P 5 .0001). However, of the 89 patients referred to the IS center from outside the IHC system, 31 (35%) underwent duplicated studies, while 14 (7.6%) of the 184 patients from within the IS underwent duplicated studies (Fig. 1). The majority of duplicate CT scans at both facilities were of the head and cervical spine. The NIS performed 60 (34%) duplicate head CTs, and 61 (34%) duplicate cervical spine CTs. The IS performed 17 (30%) duplicate head CTs and 20 (36%) duplicate cervical spine CTs. On a body region basis, the IS performed significantly fewer duplicated CT scans with the exception of chest CT scans (Table 1). Additional CT scans in total and on a body region basis were not obtained with any significant difference between the system types (Table 2). The most frequently cited reason for duplicate or additional CT scans at each facility was inadequate, incomplete, or poor quality of the CT scans from the referring center. This constituted 58% at the NIS versus 81% at the IS (P , .001). At the NIS, 12% of duplicate or additional scans were performed for change in clinical condition versus 5% at the IS (P 5 .027). Most category differences did not reach statistical significance (Table 3).
Comments
Figure 1
Patients who received duplicated CT scans.
This study shows that in a vertically integrated healthcare system the number of duplicated or additionally obtained CT scans performed on transferred trauma patients is significantly less than in a nonintegrated tertiary referral system. The NIS demonstrated a duplicate CT scan rate lower than (48%), but similar to, previously documented reviews of AMC-based trauma centers by Haley et al (53.2%) and Gupta et al (58%).2,3 However, the vertically integrated healthcare system had an overall duplicate CT scan rate of only 16% of transferred trauma patients. When further limited to patients only from within the vertical system, the rate dropped to 7.6%. Thirty-five percent of patients who were referred to the IS from outside the
514
The American Journal of Surgery, Vol 208, No 4, October 2014
Table 2
Additional CT scans by body region Additional CT scans
Body region
NIS
Total Head Cervical spine Thoracic spine Lumbar spine Chest Abdomen Pelvis Other
230 4 32 59 60 9 13 15 38
P value (NS . .05)
IS (42%) (2%) (14%) (26%) (26%) (4%) (5.6%) (6.5%) (17%)
317 17 53 76 72 21 31 30 17
(58%) (5%) (17%) (24%) (23%) (7%) (10%) (9%) (5%)
NS NS NS NS NS NS NS NS ,.001
CT 5 computed tomography; IS 5 integrated system; NIS 5 nonintegrated system; NS 5 nonsignificant.
IS underwent duplicated scans comparable with the NIS included in this study. Because the vertically IS revealed similar results when patients were transferred from outside of its integrated network, it appears that there is benefit to a vertically organized trauma system. This is likely attributable to the integrated medical record and PACS utilized throughout the vertical system, as well as other efficiencies of process afforded by integration. The potential benefits of an integrated healthcare system include improved patient care efficiency, cost reductions to both the patient and the system, and reduced radiation exposure to patients. However, measurement of integrated healthcare delivery is ambiguous at best. A recent systematic review showed 24 different methods of measurement described in the literature.5 However, the authors conclude that integrated healthcare systems have the ability to deliver higher quality health care at a lower cost while maintaining or improving patient health and satisfaction. The majority of publications available do not, however, give specific quality measures relevant to individual patients, such as radiation exposure.5,6 Integrated healthcare systems have the potential to reduce the number of duplicated CT scans in trauma patients, as demonstrated in this study. Comparing the patients with duplicate CT imaging at the NIS with the comparable cohort at the IS, those transferred from outside the IS, the rate was 13% lower at the IS (48% vs 35%; P 5 .2). The trend toward decreased duplicity of CT imaging may be related to an institutional mandate to upload all requested outside images to the institution’s
Table 3
PACS system. This allows consultants to use outside studies rather than duplicate them. In addition, the radiologists agreed to provide an over-read on outside CT images, a practice not widely accepted. The 2 trauma centers in this study share many geographic and referring pattern similarities. They are located in close proximity to each other and both share similar catchment areas made up of similar geographic distances and terrain, have mostly blunt mechanisms of injury, and a comparable ISS. Many of the referring hospitals to both trauma centers are located in remote locations and have limited resources. While it was beyond the scope of this study to characterize the specifics of each of these hospitals, they are likely to be representative of system challenges found throughout the United States, including variability in the type and generation of CT scanning equipment technology. The decreased rates of additional and duplicated studies in the IS may demonstrate some of the unmeasureable efficiencies afforded by this model. Both system types involved the same residents in patient care, albeit in different ways. The NIS had resident involvement in all traumas, but only the senior resident made imaging decisions, very frequently with attending input. The IS had limited resident involvement, but traumas were evaluated by physician assistants and nurse practitioners with attending surgeon consultation and limited resident involvement. When the data are broken down by body region, most patients receiving duplicated scans in both systems
Collected reasons for duplicated or additional CT scans
Category
NIS
IS
Total
P values
Inadequate, incomplete, poor quality Physician preference Technical difficulty or images did not transfer with the patient Change in clinical condition Other, unknown
97 (58%) 21 (13%) 21 (12%)
141 (81%) 15 (9%) 6 (3%)
238 (69.8%) 36 (10.4%) 27 (7.8%)
,.001 .251 .002
20 (12%) 10 (6%)
9 (5%) 4 (2%)
29 (8.4%) 14 (4.1%)
CT 5 computed tomography; IS 5 integrated system; NIS 5 nonintegrated system; NS 5 nonsignificant.
.027 .107
A.E. Liepert et al.
Duplicated CT scans: system effects
involved only 1 body region, 55 patients (56%) at the NIS and 36 (80%) at the IS. This discovery is most likely attributable to a close follow-up interval CT scan of the head to evaluate for evolution of intracranial hemorrhage and specialty-specific CT reformations for operative planning. The NIS had more patients with duplicated imaging than the IS. This may be related to techniques of how the initial scans were obtained and may have been limited by not having a direct reciprocity of information transfer between the centers that refer to the IS (Fig. 1). The most common body areas duplicated were of the spines. This may have occurred for a number of reasons including the following: incomplete imaging of the total spine that is necessary for full evaluation of the spinal column, preference of the consulting spine surgeons, and/or refusal of radiologists to over-read scans from another facility. Additional CT scans of body regions not previously imaged before patient transfer were performed at both the NIS and the IS with similar rates. This is likely because of appropriate discretion of the referring physicians in obtaining only imaging that will impact immediate diagnostics. Alternatively, it may represent newly identified injuries that had not been identified before transfer. It may also reflect a similar approach to comprehensive trauma patient evaluation at the 2 trauma centers. Although a different pool of trauma surgeons practice at the NIS and IS, they are all part of the same regional Advanced Trauma Life Support instructor cohort and therefore frequently interact in a trauma teaching environment. The issue of radiation exposure because of CT was best explored in 2 articles in the Archives of Internal Medicine.7,8 The data are mostly extrapolated from nuclear disasters, whether natural or man-made. Accepting this extrapolation, the greatest risk of radiation exposure is for individuals aged 40 and under. This is particularly salient to the trauma population as a large proportion of this population is young. Furthermore, the development of the trauma culture has dictated that injuries not be missed, often resulting in the ‘‘trauma scan’’ or ‘‘pan scan’’ in an attempt to reduce missed injuries. Evaluating the risk-benefit balance of this situation is difficult, but an evaluation of this practice would provide a meaningful contribution to the trauma literature. It is possible that patient or injury characteristics exist that may identify a subpopulation of trauma patients at risk for additional or duplicated imaging. This is beyond the scope of this study, but may benefit from additional investigation. This study was completed in a prospective manner with consecutive patient enrollment and is not without the inherent limitations of such a study. Selection bias may have occurred because of the lack of randomization. The clinicians were not blinded to the conduct of this study and therefore a Hawthorne effect may have influenced practice. However, if clinician practice changed as a result of being studied, it would likely have reduced the rate of duplicated CT scans, actually making our results an underestimate of the true rate. This study includes just one example of each type of center-integrated or traditional effects, leaving the possibility
515 of other regional effects. However, the close and overlapping referral and geographic base make for a closely matched comparison. The similarity of ISS between the 2 institutions in this study is important, but the relatively low severity of injury for injured patients transferred to both trauma centers is a potential limitation in extrapolating these data to other institutions. Data collection dates differed slightly between institutions for logistical reasons. Cost was unable to be determined in this study because of different cost versus charge structures between the 2 trauma centers but represents an important future avenue for research. This study categorizes the receiving trauma surgeons’ reasons for obtaining duplicated or additional imaging, something prior studies did not include. The study sample size resulted in relatively small subpopulations in the collected reasons for the CT imaging (Table 3). Most of the duplicated imaging was because of perceived quality issues of the initial scans before transfer. These data are subjective in nature, but attempt to categorize the impressions of front-line caregivers in a clinical real-time manner. This is an area with potential for quality improvement within trauma systems, whether as an increase in capital investment of inadequate CT scanning equipment, as improved communication, feedback from the trauma centers to advance techniques of CT scans obtained before transfer, or the development of specific trauma CT protocols at the referring hospitals. However, this is made difficult by the number of hospitals and personnel involved, as well as information technology restrictions and healthcare privacy law challenges. Determining the best approach may require additional study. Both the IS and NIS suffered from transfer CT quality issues; however, referral facilities within the IS may have an increased ability of leverage with referring facilities. While the rate of duplicated CT scanning at the NIS is lower than those previously documented at similar nonintegrated AMC-based trauma centers, documentation of the rate of duplicated CT scans at a trauma center within a vertically integrated healthcare system is a novel contribution to the literature. The remarkably reduced rate of study duplication in the IS demonstrates the benefit of vertical integration in healthcare systems in the direct delivery of patient care.
Conclusion Fewer patients received duplicated scans within a vertically IS as compared with a traditional universitybased nonintegrated referral system. Inadequate scan quality/incomplete imaging was the most common reason for duplicate or additional CT imaging in both types of systems. Technical and practical reasons for duplicate imaging are reduced within a vertically IS. Interoperability of medical records, and particularly radiographic studies, is characteristic of vertically ISs; our findings indicate that improvement of interoperability within nonintegrated healthcare systems is likely to benefit the processes of
516 trauma patient care. Further study of implementation and associated outcomes would be required.
Acknowledgments The authors thank Tom White, MD, Intermountain Medical Center, Jody Carter, APRN, Intermountain Medical Center, Sue Day, RN, Intermountain Medical Center, Janet Cortez, RN, University of Utah, and Iris Faraklas, RN, University of Utah.
References 1. Harrington DT, Connolly M, Biffl WL, et al. Transfer times to definitive care facilities are too long: a consequence of an immature trauma system. Ann Surg 2005;241:961–6; discussion, 966–8.
The American Journal of Surgery, Vol 208, No 4, October 2014 2. Haley T, Ghaemmaghami V, Loftus T, et al. Trauma: the impact of repeat imaging. Am J Surg 2009;198:858–62. 3. Gupta R, Greer SE, Martin ED. Inefficiencies in a rural trauma system: the burden of repeat imaging in interfacility transfers. J Trauma 2010; 69:253–5. 4. American College of Surgeons: Trauma Programs: Consultations/Verification Programs: Verified Trauma Centers. Available at: http://www. facs.org/trauma/verified.html. Accessed July 2, 2013. 5. Strandberg-Larsen M, Krasnik A. Measurement of integrated healthcare delivery: a systematic review of methods and future research directions. Int J Integr Care 2009;9:e01. 6. Wan TT, Wang BB. Integrated healthcare networks’ performance: a growth curve modeling approach. Health Care Manag Sci 2003;6: 117–24. 7. Berrington de Gonza´lez A, Mahesh M, Kim KP, et al. Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med 2009;169:2071–7. 8. Smith-Bindman R, Lipson J, Marcus R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med 2009;169:2078–86.
Association of Women Surgeons
Protecting trauma patients from duplicated computed tomography scans: the relevance of integrated care systems Amy E. Liepert, M.D.a,*, Joseph Bledsoe, M.D.b, Mark H. Stevens, M.D.c, Amalia Cochran, M.D.d a
Department of Surgery, University of Wisconsin, Madison, WI, USA; bDepartment of Emergency Medicine and cDepartment of Surgery, Intermountain Medical Center, Murray, UT, USA; dDepartment of Surgery, University of Utah, Salt Lake City, UT, USA
KEYWORDS: Trauma care; Interfacility transfer; Duplicated CT scans
Abstract BACKGROUND: Duplicated computed tomography (CT) scans in transferred trauma patients have been described in university-based trauma systems. This study compares CT utilization between a university-based nonintegrated system (NIS) and a vertically integrated regional healthcare system (IS). METHODS: Trauma patients transferred to 2 Level I trauma centers were prospectively identified at the time of transfer. Imaging obtained before and subsequent to transfer and the reason for CT imaging at the Level I center were captured by real-time reporting. RESULTS: Four hundred eighty-one patients were reviewed (207 at NIS and 274 at IS). Ninety-nine patients (48%) at NIS and 45 (16%) at IS underwent duplicate scanning of at least one body region. Inadequate scan quality and incomplete imaging were the most common reason category reported at NIS (54%) and IS (78%). CONCLUSIONS: Fewer patients received duplicated scans within the vertically IS as compared with a traditional university-based referral system. Our findings suggest that the adoption of features of a vertically IS, particularly improved transferability of radiographic studies, may improve patient care in other system types. Ó 2014 Elsevier Inc. All rights reserved.
Amy E. Liepert had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. There were no relevant financial relationships or any sources of support in the form of grants, equipment, or drugs. The authors declare no conflicts of interest. * Corresponding author. Tel.: 11-608-265-6246; fax: 11-608-2529746. E-mail address: [email protected] Manuscript received January 15, 2014; revised manuscript May 19, 2014 0002-9610/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2014.05.014
Trauma systems in the United States have evolved as a series of regional networks that direct patients to designated trauma centers. Patients injured in rural areas are typically stabilized at critical access hospitals and those with significant traumatic injury are transferred to state or American College of Surgeons (ACS)-designated trauma centers. Triage of injured patients in rural medical centers is often variable depending on the facility and provider resources that are locally available. Several factors contribute to the performance of initial computed tomography (CT) scanning at the referring center, including fear of litigation and a desire to provide the receiving trauma center with a
512 definitive diagnosis.1 Nevertheless, many patients have imaging performed at the referring hospital, which is then duplicated at the receiving trauma center.2,3 Patients are also frequently subjected to additional or completion CT studies at the trauma center for a comprehensive trauma evaluation. This may result in inefficiency, redundancy, patient discomfort, increased radiation exposure, and increased cost. The majority of ACS-designated Level 1 trauma centers in the United States are at academic medical centers (AMCs).4 Integrated healthcare systems exist in which a single organization owns or administers multiple hospitals within a geographic region, including a Level 1 or 2 trauma center as their referral center. Vertically integrated systems (ISs) use the same electronic medical record and electronic radiology system throughout their facilities. This structure allows for access to patient data by the receiving trauma center when patients are transferred within the system. Prior studies have evaluated duplicated CT scanning in AMC-based tertiary referral systems but have not included data from a vertically integrated trauma system. The goal of this study is to compare the difference in CT utilization frequency and reasons for duplicated or additionally obtained CT scans at 2 regional Level 1 trauma centers, 1 AMC-based nonintegrated center, and the other within a vertically integrated regional healthcare system. Our hypothesis was that an IS would have a lower rate of duplicated CT scans when compared with a nonintegrated system (NIS).
Methods This is a prospective cohort study of patients referred to one of the 2 regional Level 1 adult trauma centers in Salt Lake City, UT, over a 6-month period. The University of Utah Medical Center is an urban ACS-verified Level 1 trauma center within an AMC with an annual average trauma admission volume of 1,100 patients. Intermountain Medical Center (IMC) is a university-affiliated teaching hospital with an ACS-verified Level 1 trauma center. It serves as the flagship hospital within the vertically integrated Intermountain Healthcare (IHC) system. The IHC network consists of 22 hospitals in 2 states with fully integrated electronic medical record and radiology picture archiving and communication system (PACS) systems. The average annual trauma admission volume at IMC is 1,800 patients per year. Each hospital has both fixed-wing and helicopter-based airmedical transport systems and standard ground emergency medical system availability. Each serves approximately the same 4-state catchment area. The University of Utah Medical Center is representative of a nonintegrated healthcare system for the purposes of this study and will be referred to as NIS, while IMC is representative of an integrated healthcare system and will be referred to as IS. Transferred trauma patients who received a CT scan at a referring hospital were prospectively identified at their time
The American Journal of Surgery, Vol 208, No 4, October 2014 of arrival at each of the Level 1 trauma centers. If a patient did not have any CT scans at either the referring or the receiving center, then they were excluded. The CT images were recorded by the body region(s) obtained and whether the images were transferred on disc or electronically. All CT images on CD discs were scanned into the trauma centers’ electronic radiology system, although this process was not always immediately available. The discs were collected for later review by the principle investigator at NIS and by a designated individual at IS. If the disc did not contain the CT studies completed before transfer, then the medical record, including referring hospital documentation, was examined for evidence of studies performed before the transfer. CT studies ordered at the receiving facility were included if they were ordered in the trauma bay by the trauma team leader or subspecialty physicians caring for the patient. CT scans were recorded in the database categorized by body region. The separate CT body regions included the following: head, cervical spine, thoracic spine, lumbar spine, chest, abdomen, pelvis, and other. Data collection forms were secured in the emergency department, and then were collected by a study designee. Duplicated scans were defined as any body region that had already been CT scanned at the referring hospital. Additional scans were defined as any body region that was not scanned before the trauma patient’s transfer to the tertiary trauma facility. A trauma team member completed a data collection sheet in real time that indicated the reason for duplicate or additionally obtained CT scans. The documented reasons were categorized into one of the 5 following categories: technical difficulty; change in clinical condition; inadequate, incomplete, or poor quality images; physician preference; or other. Technical difficulty included images not included on disc, inability to open images on disc, or no disc/images transferred with the patient. If a reason for CT imaging was not identified, the principle investigator contacted the trauma team to obtain the reason for imaging obtained at the receiving trauma center. If a reason was unable to be provided, it was documented as unknown. Data entry was completed within 72 hours of transfer. The study PI was responsible for reviewing all entered data. Statistical analysis was performed using STATA version 11.2 (College Station, TX). Patient demographics were compared using chi-square test. The Wilcoxon rank-sum test was used for comparison of reason for repeat CT scans. Fisher’s exact test and chi-square tests were used for further comparative analysis as appropriate. Institutional review board approval was obtained at both institutions.
Results Four hundred ninety-nine patients were evaluated from the 2 centers. The NIS cohort included 217 consecutive trauma transfers between December 1, 2009 and May 28, 2010. The IS evaluated 282 consecutive trauma transfers
A.E. Liepert et al. Table 1
Duplicated CT scans: system effects
513
Duplicated CT scans by body region Duplicated CT scans
Body region
NIS
Total Head Cervical spine Thoracic spine Lumbar spine Chest Abdomen Pelvis Other
178 60 61 12 12 7 9 11 5
P value (NS . .05)
IS (76%) (33.7%) (34.3%) (6.7%) (6.7%) (3.9%) (5.1%) (6.2%) (2.8%)
56 17 20 4 3 3 2 4 3
(24%) (30.3%) (35.7%) (7.1%) (5.4%) (5.4%) (3.6%) (7.1%) (5.4%)
,.001 NS NS NS NS NS NS NS NS
CT 5 computed tomography; IS 5 integrated system; NIS 5 nonintegrated system; NS 5 nonsignificant.
from February 22, 2010 to August 27, 2010. Nineteen patients were excluded from the study, 10 at NIS and 9 at IS, for no CT scans obtained at either the referring facility or receiving trauma center. Total included patients were 480, 207 at the NIS and 273 at the IS. Of the 273 patients included at the IS, 184 (68%) were transferred from within the IS, while 89 (32%) were referred from a hospital outside the IS. The overall median age of enrolled patients was 47 years, 44 years (interquartile range 32 to 70) at the NIS and 51 years (interquartile range 30 to 65 years) at the IS. Four hundred sixty-eight (97.5%) patients of the entire study population were blunt trauma patients, 198 (93.5%) at the NIS and 270 (97.7%) at the IS. The ISS at the NIS was 12.5, while at the IS it was 11 (P 5 .053). Duplicate CT scans were performed on a total of 144 patients, for a total of 234 duplicated body region scans. The NIS had 178 duplicated body region scans and the IS had 56 duplicated body region scans (Table 1). Of the 207 patients included at the NIS, 99 patients (48%) received duplicated CT imaging. Duplicate CT imaging was
performed on 45 (16%) of the 273 patients evaluated at the IS (P 5 .0001). However, of the 89 patients referred to the IS center from outside the IHC system, 31 (35%) underwent duplicated studies, while 14 (7.6%) of the 184 patients from within the IS underwent duplicated studies (Fig. 1). The majority of duplicate CT scans at both facilities were of the head and cervical spine. The NIS performed 60 (34%) duplicate head CTs, and 61 (34%) duplicate cervical spine CTs. The IS performed 17 (30%) duplicate head CTs and 20 (36%) duplicate cervical spine CTs. On a body region basis, the IS performed significantly fewer duplicated CT scans with the exception of chest CT scans (Table 1). Additional CT scans in total and on a body region basis were not obtained with any significant difference between the system types (Table 2). The most frequently cited reason for duplicate or additional CT scans at each facility was inadequate, incomplete, or poor quality of the CT scans from the referring center. This constituted 58% at the NIS versus 81% at the IS (P , .001). At the NIS, 12% of duplicate or additional scans were performed for change in clinical condition versus 5% at the IS (P 5 .027). Most category differences did not reach statistical significance (Table 3).
Comments
Figure 1
Patients who received duplicated CT scans.
This study shows that in a vertically integrated healthcare system the number of duplicated or additionally obtained CT scans performed on transferred trauma patients is significantly less than in a nonintegrated tertiary referral system. The NIS demonstrated a duplicate CT scan rate lower than (48%), but similar to, previously documented reviews of AMC-based trauma centers by Haley et al (53.2%) and Gupta et al (58%).2,3 However, the vertically integrated healthcare system had an overall duplicate CT scan rate of only 16% of transferred trauma patients. When further limited to patients only from within the vertical system, the rate dropped to 7.6%. Thirty-five percent of patients who were referred to the IS from outside the
514
The American Journal of Surgery, Vol 208, No 4, October 2014
Table 2
Additional CT scans by body region Additional CT scans
Body region
NIS
Total Head Cervical spine Thoracic spine Lumbar spine Chest Abdomen Pelvis Other
230 4 32 59 60 9 13 15 38
P value (NS . .05)
IS (42%) (2%) (14%) (26%) (26%) (4%) (5.6%) (6.5%) (17%)
317 17 53 76 72 21 31 30 17
(58%) (5%) (17%) (24%) (23%) (7%) (10%) (9%) (5%)
NS NS NS NS NS NS NS NS ,.001
CT 5 computed tomography; IS 5 integrated system; NIS 5 nonintegrated system; NS 5 nonsignificant.
IS underwent duplicated scans comparable with the NIS included in this study. Because the vertically IS revealed similar results when patients were transferred from outside of its integrated network, it appears that there is benefit to a vertically organized trauma system. This is likely attributable to the integrated medical record and PACS utilized throughout the vertical system, as well as other efficiencies of process afforded by integration. The potential benefits of an integrated healthcare system include improved patient care efficiency, cost reductions to both the patient and the system, and reduced radiation exposure to patients. However, measurement of integrated healthcare delivery is ambiguous at best. A recent systematic review showed 24 different methods of measurement described in the literature.5 However, the authors conclude that integrated healthcare systems have the ability to deliver higher quality health care at a lower cost while maintaining or improving patient health and satisfaction. The majority of publications available do not, however, give specific quality measures relevant to individual patients, such as radiation exposure.5,6 Integrated healthcare systems have the potential to reduce the number of duplicated CT scans in trauma patients, as demonstrated in this study. Comparing the patients with duplicate CT imaging at the NIS with the comparable cohort at the IS, those transferred from outside the IS, the rate was 13% lower at the IS (48% vs 35%; P 5 .2). The trend toward decreased duplicity of CT imaging may be related to an institutional mandate to upload all requested outside images to the institution’s
Table 3
PACS system. This allows consultants to use outside studies rather than duplicate them. In addition, the radiologists agreed to provide an over-read on outside CT images, a practice not widely accepted. The 2 trauma centers in this study share many geographic and referring pattern similarities. They are located in close proximity to each other and both share similar catchment areas made up of similar geographic distances and terrain, have mostly blunt mechanisms of injury, and a comparable ISS. Many of the referring hospitals to both trauma centers are located in remote locations and have limited resources. While it was beyond the scope of this study to characterize the specifics of each of these hospitals, they are likely to be representative of system challenges found throughout the United States, including variability in the type and generation of CT scanning equipment technology. The decreased rates of additional and duplicated studies in the IS may demonstrate some of the unmeasureable efficiencies afforded by this model. Both system types involved the same residents in patient care, albeit in different ways. The NIS had resident involvement in all traumas, but only the senior resident made imaging decisions, very frequently with attending input. The IS had limited resident involvement, but traumas were evaluated by physician assistants and nurse practitioners with attending surgeon consultation and limited resident involvement. When the data are broken down by body region, most patients receiving duplicated scans in both systems
Collected reasons for duplicated or additional CT scans
Category
NIS
IS
Total
P values
Inadequate, incomplete, poor quality Physician preference Technical difficulty or images did not transfer with the patient Change in clinical condition Other, unknown
97 (58%) 21 (13%) 21 (12%)
141 (81%) 15 (9%) 6 (3%)
238 (69.8%) 36 (10.4%) 27 (7.8%)
,.001 .251 .002
20 (12%) 10 (6%)
9 (5%) 4 (2%)
29 (8.4%) 14 (4.1%)
CT 5 computed tomography; IS 5 integrated system; NIS 5 nonintegrated system; NS 5 nonsignificant.
.027 .107
A.E. Liepert et al.
Duplicated CT scans: system effects
involved only 1 body region, 55 patients (56%) at the NIS and 36 (80%) at the IS. This discovery is most likely attributable to a close follow-up interval CT scan of the head to evaluate for evolution of intracranial hemorrhage and specialty-specific CT reformations for operative planning. The NIS had more patients with duplicated imaging than the IS. This may be related to techniques of how the initial scans were obtained and may have been limited by not having a direct reciprocity of information transfer between the centers that refer to the IS (Fig. 1). The most common body areas duplicated were of the spines. This may have occurred for a number of reasons including the following: incomplete imaging of the total spine that is necessary for full evaluation of the spinal column, preference of the consulting spine surgeons, and/or refusal of radiologists to over-read scans from another facility. Additional CT scans of body regions not previously imaged before patient transfer were performed at both the NIS and the IS with similar rates. This is likely because of appropriate discretion of the referring physicians in obtaining only imaging that will impact immediate diagnostics. Alternatively, it may represent newly identified injuries that had not been identified before transfer. It may also reflect a similar approach to comprehensive trauma patient evaluation at the 2 trauma centers. Although a different pool of trauma surgeons practice at the NIS and IS, they are all part of the same regional Advanced Trauma Life Support instructor cohort and therefore frequently interact in a trauma teaching environment. The issue of radiation exposure because of CT was best explored in 2 articles in the Archives of Internal Medicine.7,8 The data are mostly extrapolated from nuclear disasters, whether natural or man-made. Accepting this extrapolation, the greatest risk of radiation exposure is for individuals aged 40 and under. This is particularly salient to the trauma population as a large proportion of this population is young. Furthermore, the development of the trauma culture has dictated that injuries not be missed, often resulting in the ‘‘trauma scan’’ or ‘‘pan scan’’ in an attempt to reduce missed injuries. Evaluating the risk-benefit balance of this situation is difficult, but an evaluation of this practice would provide a meaningful contribution to the trauma literature. It is possible that patient or injury characteristics exist that may identify a subpopulation of trauma patients at risk for additional or duplicated imaging. This is beyond the scope of this study, but may benefit from additional investigation. This study was completed in a prospective manner with consecutive patient enrollment and is not without the inherent limitations of such a study. Selection bias may have occurred because of the lack of randomization. The clinicians were not blinded to the conduct of this study and therefore a Hawthorne effect may have influenced practice. However, if clinician practice changed as a result of being studied, it would likely have reduced the rate of duplicated CT scans, actually making our results an underestimate of the true rate. This study includes just one example of each type of center-integrated or traditional effects, leaving the possibility
515 of other regional effects. However, the close and overlapping referral and geographic base make for a closely matched comparison. The similarity of ISS between the 2 institutions in this study is important, but the relatively low severity of injury for injured patients transferred to both trauma centers is a potential limitation in extrapolating these data to other institutions. Data collection dates differed slightly between institutions for logistical reasons. Cost was unable to be determined in this study because of different cost versus charge structures between the 2 trauma centers but represents an important future avenue for research. This study categorizes the receiving trauma surgeons’ reasons for obtaining duplicated or additional imaging, something prior studies did not include. The study sample size resulted in relatively small subpopulations in the collected reasons for the CT imaging (Table 3). Most of the duplicated imaging was because of perceived quality issues of the initial scans before transfer. These data are subjective in nature, but attempt to categorize the impressions of front-line caregivers in a clinical real-time manner. This is an area with potential for quality improvement within trauma systems, whether as an increase in capital investment of inadequate CT scanning equipment, as improved communication, feedback from the trauma centers to advance techniques of CT scans obtained before transfer, or the development of specific trauma CT protocols at the referring hospitals. However, this is made difficult by the number of hospitals and personnel involved, as well as information technology restrictions and healthcare privacy law challenges. Determining the best approach may require additional study. Both the IS and NIS suffered from transfer CT quality issues; however, referral facilities within the IS may have an increased ability of leverage with referring facilities. While the rate of duplicated CT scanning at the NIS is lower than those previously documented at similar nonintegrated AMC-based trauma centers, documentation of the rate of duplicated CT scans at a trauma center within a vertically integrated healthcare system is a novel contribution to the literature. The remarkably reduced rate of study duplication in the IS demonstrates the benefit of vertical integration in healthcare systems in the direct delivery of patient care.
Conclusion Fewer patients received duplicated scans within a vertically IS as compared with a traditional universitybased nonintegrated referral system. Inadequate scan quality/incomplete imaging was the most common reason for duplicate or additional CT imaging in both types of systems. Technical and practical reasons for duplicate imaging are reduced within a vertically IS. Interoperability of medical records, and particularly radiographic studies, is characteristic of vertically ISs; our findings indicate that improvement of interoperability within nonintegrated healthcare systems is likely to benefit the processes of
516 trauma patient care. Further study of implementation and associated outcomes would be required.
Acknowledgments The authors thank Tom White, MD, Intermountain Medical Center, Jody Carter, APRN, Intermountain Medical Center, Sue Day, RN, Intermountain Medical Center, Janet Cortez, RN, University of Utah, and Iris Faraklas, RN, University of Utah.
References 1. Harrington DT, Connolly M, Biffl WL, et al. Transfer times to definitive care facilities are too long: a consequence of an immature trauma system. Ann Surg 2005;241:961–6; discussion, 966–8.
The American Journal of Surgery, Vol 208, No 4, October 2014 2. Haley T, Ghaemmaghami V, Loftus T, et al. Trauma: the impact of repeat imaging. Am J Surg 2009;198:858–62. 3. Gupta R, Greer SE, Martin ED. Inefficiencies in a rural trauma system: the burden of repeat imaging in interfacility transfers. J Trauma 2010; 69:253–5. 4. American College of Surgeons: Trauma Programs: Consultations/Verification Programs: Verified Trauma Centers. Available at: http://www. facs.org/trauma/verified.html. Accessed July 2, 2013. 5. Strandberg-Larsen M, Krasnik A. Measurement of integrated healthcare delivery: a systematic review of methods and future research directions. Int J Integr Care 2009;9:e01. 6. Wan TT, Wang BB. Integrated healthcare networks’ performance: a growth curve modeling approach. Health Care Manag Sci 2003;6: 117–24. 7. Berrington de Gonza´lez A, Mahesh M, Kim KP, et al. Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med 2009;169:2071–7. 8. Smith-Bindman R, Lipson J, Marcus R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med 2009;169:2078–86.
