Impact of a Transfer Center on Interhospital ... - Wiley Online Library

10 downloads 0 Views 137KB Size Report
Patricia A. Southard, RN, JD, Jerris R. Hedges, MD, John G. Hunter, MD,. Ross M. Ungerleider, MD. Abstract. Background: The partnership of faculty physicians ...
ACAD EMERG MED

d

July 2005, Vol. 12, No. 7

d

www.aemj.org

653

Impact of a Transfer Center on Interhospital Referrals and Transfers to a Tertiary Care Center Patricia A. Southard, RN, JD, Jerris R. Hedges, MD, John G. Hunter, MD, Ross M. Ungerleider, MD Abstract Background: The partnership of faculty physicians and senior clinical hospital administrators in the decision to accept interhospital transfers has not been fully studied. Transfers to academic medical centers on the basis of economics have been of particular concern. Objectives: To evaluate the impact of joint decision making on transfer acceptance, and to evaluate the basis for decisions to transfer patients to an academic medical center. Methods: This was a database study of requested adult interhospital transfers, excluding psychiatric transfers, occurring between January 1, 2003, and December 31, 2003, by using data from a computerized patient-tracking system. Where possible, comparisons with the prior calendar year (i.e., prior to implementation of the administrative review process) were made. Incidence of refusal to accept requested transfers and payer mix of transfer patients were the main outcomes of interest. Results: More than 90% of the adult patients were transferred for conditions that required tertiary care or met Emergency Medical Treatment and Labor Act (EMTALA)

requirements. The patient conditions that did not meet tertiary care needs included obstetric patients who did not have prenatal care, patients who had hand and facial trauma, and patients who weighed more than 300 pounds. The payer mix of transfer patients remained stable when using the administrator and physician team to determine acceptance of transfers. During the evaluation period, approximately 91,500 patients statewide lost some level of Medicaid coverage. Conclusions: The value of an administrator and physician team as partners in the interhospital transfer process was demonstrated. Active management of interhospital transfers supports transfer of patients who require tertiary care or who meet EMTALA criteria, thus conserving limited bed capacity and ensuring financial equity, while caring for the uninsured and underinsured patients throughout the state. Key words: transfer; interhospital; transfer center; administration; decision making. ACADEMIC EMERGENCY MEDICINE 2005; 12:653–657.

In states with a limited number of academic medical centers (AMCs), the issue of capacity constraint is universal. It is incumbent on the AMCs in such states to have strategies that coordinate interhospital transfers and ensure the best utilization of the limited bed resources. The AMCs have an obligation to the providers and patients of the state to have the capacity to accept those patients who require tertiary care. Another important phenomenon is the decline in the rates of reimbursement for Medicaid and Medicare patients. When coupled with an increase in the proportion of underinsured and uninsured patients, some providers and institutions may promote the interhospital transfer of these patients to AMCs for economic purposes, irrespective of the true need for tertiary care. Since AMCs already provide care for a

disproportionate number of underinsured patients, the interhospital transfer of underinsured patients not requiring tertiary care further compounds the economic stress on AMCs. Further, a recent change in the Emergency Medical Treatment and Labor Act (EMTALA) regulations mitigates the obligation of community hospitals to provide specialty physicians for around-the-clock emergency department (ED) consultations. The EMTALA clarification addresses the requirement for on-call coverage of specialists and subspecialists. In the clarifying language, hospitals are given discretion to have an on-call list that will best meet the needs of their patients. If on-call coverage is not consistently available, the hospital must have policies and procedures to be followed when on-call coverage is not available. This clarification to the regulation could encourage community hospital-based providers to transfer more patients to tertiary care facilities as a contingency for unavailable on-call specialist coverage.1 In this article, we report our experience with an AMC transfer center. The transfer center is physically located in the ED, and the center’s staff function under the direction of an emergency medicine medical director. The center was created to change the chaotic acceptance of patient transfer requests without

From the Department of Surgery (PAS, JGH, RMU) and the Department of Emergency Medicine (JRH), Oregon Health & Science University, Portland, OR. Received December 8, 2004; revision received February 28, 2005; accepted March 3, 2005. Address for correspondence and reprints: Patricia A. Southard, RN, JD, Department of Surgery, Oregon Health & Science University, Mail Code L 223, 3181 SW Sam Jackson Park Road, Portland, OR 97239. Fax: 503-494-5615; e-mail: [email protected]. doi:10.1197/j.aem.2005.03.515

654

Southard et al.

prioritization. Transfers were accepted through the center, capacity permitting, if the patient required tertiary care or otherwise met EMTALA criteria. In those instances when the hospital is on a no-bedavailable status, but the patient requiring transfer has an EMTALA condition that uniquely requires the services of an AMC, and the condition requires a procedure or surgery within two hours of admission, the faculty subspecialty physician in consultation with the emergency medicine faculty physician may accept the patient. The emergency medicine faculty physician assumes the interim care of the patient for the brief period of time prior to the definitive intervention. During the transfer period, extensive efforts are made to create bed capacity by the in-house hospital administrative nurse. We report administrative enhancements to the transfer center, including ongoing communication with hospitals throughout the state when a trend for economic-based requests for interhospital transfers was identified. Our analysis of the center’s operation focuses on maintenance of established providerto-provider referral patterns and compliance with EMTALA regulations.

METHODS Study Design. We performed an analysis of administrative data describing interhospital transfer requests screened at an AMC. The study met the criteria for expedited institutional review board review based on research involving data that have been collected for nonresearch purposes. Study Setting and Population. The transfer center is situated in the ED of an urban, Level 1 trauma center with approximately 42,000 annual visits. Emergency medical technician-level emergency dispatch operators staff the transfer center. Patients 17 years of age or older during the calendar years 2002 and 2003 were

d

IMPACT OF A TRANSFER CENTER

included. The data collection for the new process occurred prospectively from January 2003 through December 2003. The data for 2002 (January through December) were obtained via retrospective review. Psychiatric transfer patients were excluded from the study because the decision to accept transfer occurred outside the transfer center. Study Protocol. We aggregated patients by transferring hospital, payer status, reason for transfer, admitting service, and whether the transfer was for tertiary care and/or was an EMTALA-qualified patient who required acceptance. All requests for interhospital transfers were audio-recorded and stored for review in the transfer center. The EmSTAT (A4 Health Systems, Cary, NC) computerized patient-tracking system for the ED was customized to track all interhospital transfer requests from initial call to final outcome. Resident physicians were excluded from participation in the interhospital transfer request process. Beginning in 2003, a senior clinical administrator actively managed all adult interhospital transfers for the year. The administrator developed a transfer request review process and implemented post-tertiary care back-to-the-referring-hospital transfer agreements for patients requiring stabilizing interventions for life- or-limb threatening conditions. The administrator was proactively involved in discussions regarding whether to accept the requested transfer, and also coordinated feedback to community hospital administrators and referring providers. Data Analysis. Requests for interhospital transfers to the AMC were compared for the years before (2002) and during (2003) administrative active management. Patients accepted in transfer were compared on payer mix and transferring hospital bed size by means tests and chi-square tests where appropriate. A p-value of 0.05 was accepted as statistically significant.

Figure 1. Activity level for transfers and the transfer center. In 2002, 93.7% of transfers went through the transfer center; the refusal rate is unknown. In 2003, virtually 100% of transfers went through the transfer center, with 90.9% accepted.

ACAD EMERG MED

d

July 2005, Vol. 12, No. 7

d

655

www.aemj.org

Figure 2. Payer mix comparison for adult transfers. Medicaid/uninsured and other insurance proportions did not significantly change among the transferred patients. In 2003, approximately 91,500 Oregonians were removed from the Oregon Health Plan or received reduced coverage.

RESULTS There were 1,634 patients accepted for transfer in 2002, with 1,532 requests for transfer made through the transfer center. However, in 2003, following education of the community regarding the transfer center, virtually all of the 2,082 requests for transfer were made through the transfer center. Of these requests, 1,892 (90.9%) were accepted in transfer (Figure 1). Transfers were declined in the absence of capacity. Further, when capacity existed, transfers were generally declined if beds were limited and there was failure to meet either need for tertiary care or EMTALA criteria. In 2002, 47% of all transfers came from hospitals with 100 beds or less, versus 57% in 2003 (p , 0.01). The payer mix for 2002 and 2003 essentially remained stable, with a 1% decrease in Medicaid transfers and a 1% increase in Medicare transfers (p = NS). In 2003, 51% of all adult transfers were to a surgery service, 41% to a medicine service, and 8% to obstetrics (OB). In 2003, 55% of the transfers met tertiary care criteria, 38% met EMTALA criteria, and 7% (n = 132)

were not EMTALA or tertiary care transfers (i.e., these would be considered convenience transfers). The most common factor associated with transfer to our AMC was the presence of a complex illness or injury. In the study, more than 90% of the adult patients who were transferred in 2003 met EMTALA criteria and/or had conditions that required tertiary care. However, there were some conditions in which there was a disproportionately high Medicaid/uninsured payer type compared with other adult transfer patients. These patient groups were obstetric patients without prenatal care, hand and facial trauma patients, and patients who weighed more than 300 pounds. When combining the payer mix categories of uninsured and Medicaid patients, the payer mix of transferred patients remained stable in the two-year comparison (Figure 2). However, in 2003, there was a loss of support for the Oregon Medicaid population that moved 51,000 Oregon residents off state medical assistance programs. In April 2003, another 16,000 Oregon Health Plan beneficiaries were disqualified under new administrative policies for premium pay-

Figure 3. Comparison of two community health systems’ indigent payer mix transfer rates vs. the inpatient indigent rate.

656

Southard et al.

ments. Finally, in October 2003, there was a decrease of 24,500 enrollments in the Oregon Health Plan standard beneficiary enrollment. The cumulative impact was a loss of coverage or inability to enroll 91,500 Oregon residents.2 Maintaining the status quo in payer mix for interhospital transfers during such difficult times may be attributed to the senior clinical administrator’s working in partnership with the faculty physicians to review questionable transfer requests. Other studies that have looked at the payer mix issues related to interhospital transfers have found a relationship between insurance status and likelihood of transfer to a higher level of care.3–5 Active management of the transfer process was most effective with hospital systems in the Portland metropolitan area. For these community hospitals that identify themselves as providing tertiary care in the Portland metropolitan area, the combined number of transfers to the AMC declined by 47% with active management by the AMC, and the combined payer mix improved by 11% for two referring hospital systems. Nonetheless, in one of the two major health care systems in the Portland metropolitan area, the smaller hospitals in the system tended to transfer more Medicaid and uninsured cases to the AMC, as opposed to an affiliated tertiary hospital within their system. When compared with the proportion of uninsured/underinsured patients admitted to the same Portland metropolitan area hospital systems, the proportion of patients transferred to the AMC in 2002 who were uninsured/underinsured was significantly greater (p , 0.01). The proportion of patients transferred to the AMC in 2003 who were uninsured/ underinsured decreased for both systems. In only one of the hospital systems was the proportion of uninsured/underinsured status in transferred patients significantly greater than for patients admitted to the referring hospital system (Figure 3). Even with active management and a large increase in the declination of acceptance of transfer patients, no EMTALA complaint was filed during 2003.

DISCUSSION The enhancements to the AMC transfer center coupled with the faculty education and partnership with a senior clinical administrator worked to ensure appropriate transfers and a stable payer mix. It was found that, with a few exceptions, primarily in the Portland metropolitan area, patient transfer requests were predicated on complexity of care needs versus adverse payer mix. A more comprehensive review of transfer patterns from the two major community health care systems in the Portland metropolitan area revealed an adverse payer mix bias in the transfer patterns. Generally, the percentage of indigent inpatients was lower for the hospitals within each of the two Portland metropol-

d

IMPACT OF A TRANSFER CENTER

itan health care systems when compared with the percentage of uninsured/underinsured patient transfers to the AMC. In a similar study from Harborview Hospital in Seattle, the patient’s insurance status appeared to influence the decision to transfer to the AMC. Such a practice can lead to an inequitable distribution of the financial burden of uninsured patients. Contrary to the Harborview study, there did not appear to be an adverse payer mix transfer trigger for trauma patients in our data set, with the exception of patients transferred within the Portland metropolitan area from the other two major health care systems. There were 305 adult trauma patients who were transferred to our AMC, and the payer mix was 76% insured versus 24% Medicaid or uninsured. This compares well with insured versus Medicaid-uninsured rates of 72% versus 28% for the other transferred patients (p = NS). Studies conducted by the University of Alabama at Birmingham5 and Children’s Hospital of Philadelphia3 found that lack of insurance coverage or presence of Medicaid coverage resulted in financial transfers of pregnant women to higher-level institutions. These findings are congruent with our findings of maternal transfers related to insurance status. In 2003, there were 153 OB patients transferred and the payer mix was 69% Medicaid or uninsured, versus 31% insured (p , 0.01 vs. other transferred patients). Trauma, vascular, and neurosurgical patients who require emergency procedures or surgery are transferred without regard to insurance status. One could infer that patients requiring emergent interventions might be transferred in order to maintain the integrity of the elective operative cases in the community hospital. Patients who will require complex intensive care are also transferred irrespective of payer status. Given the resources available in a community hospital versus the resources of 24-hour physician availability in an AMC, community physicians are likely opting to transfer complex patients to the AMC. Further, the clarification in the EMTALA regulations has limited many community hospitals’ ability to provide on-call coverage for specific areas. These include hand trauma patients, facial trauma patients, patients who weigh more than 300 pounds, and nonsponsored OB patients without prenatal care. The fact that many orthopedic surgeons are removing hand trauma from their hospital privileges suggests that some surgeons are performing ‘‘economic credentialing’’ in recognition that many after-hours hand trauma patients are not insured or that delivering such service will impair the efficiency of scheduled surgeries. Obstetricians requesting transfer of pregnant patients who have not had prenatal care classify the patients as high risk and request the EMTALA-qualified transfer to a tertiary care center. The recognition of a large potential for litigation with these patients cannot be ignored. Finally, for the patients who weigh

ACAD EMERG MED

d

July 2005, Vol. 12, No. 7

d

657

www.aemj.org

more than 300 pounds, the common reason for the transfer is the unavailability of diagnostic equipment large enough to accommodate the patient. In summary, a review of all patient transfers during the study period demonstrated that more than 90% of the interhospital transfers were for medical reasons that required tertiary care. The transfers that appeared to be triggered by insurance status were limited to a narrow group of patient conditions, including OB transfers. Two administrative activities implemented at our AMC could prove to be of benefit to AMCs that have a significant volume of transfer patients. First, we implemented ‘‘back transfer’’ agreements during this project that were instrumental in freeing up bed capacity at the AMC. For patients who did not meet EMTALA criteria and who had the potential for an extended hospitalization past the need for tertiary care, a ’’back transfer’’ agreement was negotiated with the transferring facility prior to the transfer of the patient. These agreements allowed the AMC to return the patient to the originating hospital once tertiary care was no longer needed. The AMC legal counsel developed the agreements, which require the transferring hospital to identify the accepting physician for the return of the patient and obligate the transferring hospital to pay for transportation back to their hospital, if not covered by insurance. It should be emphasized that ‘‘back transfer’’ agreements can be used only for non-EMTALA-qualified transfers and, in fact, would most likely constitute a violation of EMTALA if an attempt was made to get this agreement prior to transfer acceptance. Second, we required correctional facilities to perform transfers under a contract for services. Initially, correctional facilities transferred high-risk OB patients to the AMC without a contractual agreement. They often removed the patient from custody for the hospitalization period and then re-arrested the patient upon hospital discharge. This process negated their responsibility to pay for any of the medical care. When this was discovered during the project, the Oregon Health & Science University Contracts Office worked with the carriers for the correctional services to obtain a contract that would prevent this type of activity and also improved the percentage of payment to both the hospital and the physician group. Since patient transfers from a correctional facility are not governed by EMTALA, the hospital is free to pursue a contract regarding the acceptance of transfer patients.

LIMITATIONS Because this was an observational study, the ability to attribute changes in transfer center activity, source of referrals, and payer mix is circumstantial at best. Nonetheless, we report valuable observations and methods that can be compared with operations and outcomes at other ED-based transfer centers. We pur-

posely excluded psychiatric patient transfers because of the unique nature of these cases in regard to EMTALA obligations and the virtual absence of a tertiary resource for such mental health needs in our community during the time of the study. We also excluded pediatric transfers because the transfer considerations for children are different and the alternatives are more limited. Other communities may find transfer centers to be helpful for coordinating transfers for these patient populations as well. Indeed, our transfer center uses a similar intake process for the transfer of children. Ideally, we would like to compare the clinical status, demographics, and socioeconomic status of ED patients evaluated at the referring hospitals who are admitted versus transferred to our facility both within the urban community of our hospital and in the entire region served. For those hospitals that are part of a hospital network, comparison of transfers to our hospital versus another hospital in the hospital network would help define the level of economic triage that many believe occurs. Such information might also help the Center for Medicare and Medicaid Services redefine EMTALA transfer expectations when the opportunity for economic triage from network hospitals exists.

CONCLUSIONS Generalization of our study observations, coupled with the growth of specialty hospitals, suggests that emergency surgery will soon be limited to a finite number of medical centers, predominantly AMCs. The designation of facilities providing full resources for emergency surgery, especially on patients with complex presentations or comorbid conditions, may be warranted. Policy makers should review the reimbursement system so that those facilities and physicians who are willing to care for these challenging patients receive additional and appropriate reimbursement in recognition of the more complex care required and the increased liability risk. Finally, our experience supports the participation of a senior clinical administrator in the triage process for the evaluation and processing of requests to transfer patients to the AMC. References 1. Emergency Medical Treatment and Labor Act (EMTALA). Fed Reg. 2003; 68(Sept 9):53222–64; (adding 42 C.F.R. § 489.24 (j)). 2. McConnell J, Wallace N. The impact of administrative changes, increased cost sharing, and benefit reductions on enrollment in the Oregon Health Plan. Portland, OR: Office of Oregon Health Policy and Research, Feb 2004. 3. Durbin DR, Giardino AP, Shaw KN. The effect of insurance status on likelihood of neonatal interhospital transfer. Pediatrics. 1997; 100(3):E8. 4. Nathens AB, Maier RV, Copass MK. Payer status: the unspoken criterion. J Trauma. 2001; 50:776–83. 5. Bronstein JM, Capilouto E, Carlo WA, Haywood JL, Goldenberg RL. Access to neonatal intensive care for low-birthweight infants: the role of maternal characteristics. Am J Public Health. 1995; 85:357–61.