Developing a new care pathway for transient ischemic attack at a community medical center.
Review the steps needed to research, develop and put into action a new outpatient urgent care pathway for selected transient ischemic attack patients.
PHYSICIAN LEADERS ARE TASKED WITH finding new patient flow strategies that improve quality and patient experience while at the same time lowering costs.
Suspected transient ischemic attack is a common reason for hospitalization, based on the need for urgent evaluation and intervention of vascular risk factors to prevent impending stroke (See Box 1). However, many patients ultimately turn out to have alternate diagnoses, (1) although this may not be apparent until after the diagnostic evaluation has been completed. (2)
The diagnostic evaluation in the hospital is inefficient and usually takes two or more days to complete. (3) Medicare spends more than $350 million on TIA admissions annually, and along with other third-party payers is now conducting pre-payment reviews and denying many such cases as medically unnecessary for inpatient status. (3,4)
Clinical observation units have been effective in managing these patients without hospitalization, (5,6) but less than a third of U.S. community hospitals currently have such units, (7) and it is much less cost-effective to use observation status for patients comingled on regular medical floors. (8) TIA patients have been effectively managed by well-organized outpatient centers in other countries, (9,10,11) but this approach has been unpopular in the United States, where up to 70 percent of suspected TIA patients are still managed as hospital inpatients. (3)
We decided to pilot a daily open-access weekday outpatient TIA rapid evaluation center (TREC) for selected lower-risk patients presenting with suspected TIA at our community hospital as a viable alternative to hospitalization. Patients presenting to their primary physicians' offices or our emergency department with recent symptoms of TIA could be sent directly to the TREC on the same or next working day for an outpatient evaluation without an appointment.
Clinical stroke risk stratification scores, such as the ABCD2 scale, can be used to predict future stroke risk in patients presenting with suspected TIA, (12) and scores are lower in patients who ultimately turn out to have a non-vascular cause for their symptoms. (13,14) We suggested that referring physicians consider using the TREC as an alternative to hospitalization for suspected TIA patients with ABCD2 scores of 0-5, although ultimately the decision about hospitalization vs. outpatient TREC evaluation was left to the discretion of each individual attending physician.
All TREC patients underwent a diagnostic evaluation and consultation with a neurologist on the day of their visit, aimed at identifying and addressing stroke risk factors (See Figure 1). We knew that in order for the program to be successful, we had to take steps to secure patient referrals and cooperation from other hospital departments:
STEP 1: DATA REVIEW--We started with a retrospective chart review of all 398 patients admitted to our hospital with suspected TIA during the preceding four years (2008-2012). The mean length of stay was three days. No patients had recurrent cerebrovascular events or received thrombolytic therapy during their initial hospitalization. Sixty percent of patients were still felt to have had a TIA at the time of discharge.
Only 14 (5 percent) were readmitted to our institution within 90 days with further TIA and stroke, and this recurrent stroke risk correlated very strongly with a higher ABCD2 score of 6-7 (p=0.001). (15) This data demonstrated the low immediate stroke risk and predictive value of an ABCD2 score of 6-7 for future stroke risk at our institution, and was compelling evidence that outpatient next-day urgent care could be safe for selected patients.
[ILLUSTRATION OMITTED] BOX 1 MORE ABOUT TIA TIA is an important risk factor for impending stroke: As many as 10 percent of all TIA patients will develop a subsequent ischemic stroke within 3 months. (16,17) As many as 15 percent of all stroke patients report a recent prior TIA. (18) This clearly necessitates that all TIA patients undergo an urgent evaluation and intervention for underlying risk factors. This traditionally Is accomplished as a hospital inpatient because of the perceived Immediate stroke risk and lack of a timely organized outpatient alternative. Many patients admitted with suspected TIA end up with more benign alternate diagnoses once this evaluation Is completed. Third-party payers are increasingly denying hospital admission status for suspected TIA. BOX 2 WHAT WE DID Processes involved in the successful installation of the TREC pathway: * We secured buy-in from referring physicians using retrospective data review from our institution to confirm safety and need for an outpatient program. * We developed a consent/referral form to mitigate liability concerns from ED physicians. * We met with the leaders of our radiology and cardiology service lines and obtained their commitment to prioritize TREC patients so that testing could be accomplished in one day. * We kept TREC at the hospital to facilitate ED transfer in case of recurrent TIA or stroke. * We used existing physician and support staff already based at hospital to run the TREC program, without needing to hire additional staff.
[FIGURE 1 OMITTED]
STEP 2: REFERRAL FORM--We expected most referrals to come from the ED, so we started by presenting the TREC concept at its monthly staff meeting. The ED physicians voiced significant concerns about perceived liability from sending suspected TIA patients home for an outpatient evaluation. They also were concerned about losing patients to follow-up.
This led us to develop a three-page referral form (See Figure 2), which served multiple purposes:
* First, it could be used to calculate the ABCD2 score and quantify future stroke risk to present to the patient.
* Second, it facilitated communication with the patient, and included a signed "consent" statement.
* Finally, it included a referral form that could be faxed to our office whenever a patient was referred. The clinic was open-access and held daily, but receiving one page from the referral form allowed us to know which patients were expected each day, and we could contact no-shows directly to be sure no one fell through the cracks.
Figure 2 THREE-PAGE TREC REFERRAL FORM PAGE 1 PLEASE CONSIDER REFERRING PATIENTS: * Patients with one or more suspected TIAs within the last 2 weeks, complete resolution of their symptoms, and an ABCD2 score of 0-5. PLEASE DO NOT REFER THE FOLLOWING PATIENTS, WHO SHOULD BE SENT TO THE ED/ADMITTED: * Patients with acute stroke (i.e. residual neurological deficit) within 24 hours of onset. * Patients with recent TIA and high ABCD2 scores (>5), particularly if they cannot be seen within 24 hours because of weekend, holiday or some other scheduling conflict. * Patients with residual neurologic deficit (stroke), who will likely need inpatient rehabilitation services. INSTRUCTIONS: * Complete referral page, and fax it to XXX-XXX-XXXX * Please tell the patient to report to main registration by 8 a.m. the next working day * Complete Referral form (uses ADCD2 score and the risk chart below to calculate stroke risk). * Obtain patient signature on patient Information sheet (keep one copy for the chart and give one copy to the patient to take home as instructions) ABCD2 SCORE AND STROKE RISK: ABCD2 Score 2-day risk 7-day risk 90-day risk 0-3 1% 1.2% 3.1% 4-5 4.1% 5.9% 9.8% 6-7 8.1% 12% 18% ANY QUESTIONS OR CONCERNS: Call the stroke program coordinator at XXX-XXX-XXXX M-F 8 a.m. to 5 p.m., or the on-call stroke neurologist after hours. PAGE 2 REFERRING PHYSICIAN TO COMPLETE THIS FORM AND THEN FAX TO XXX-XXX-XXXX Patient name:-- Patient DOB:-- Best contact phone # for patient-- TIA or stoke & details:-- Date of event:-- Referral date: Symptoms:-- ABCD2 SCORE:-- ADCD2 Score Score AGE: Is the patient older than 60? 1 BLOOD PRESSURE: Systolic BP >140 and 1 or diastolic BP >90 CLINICAL FEATURES Unilateral weakness 2 Speech disturbance 1 without weakness Other 0 DURATION OF SYMPTOMS: >60mins 2 10-59 mins 1 <10mins 0 DIABETES 1 Note, if ABCD2 score is >5, consider admitting the patient, particularly if seen before a weekend or holiday Tests done in ED & results:  Brain Scan:  CT scan or  Brain MRI Date: Results:  Carotid Doppler Date: Results:  EKG (Time/Results) Date: Results:  INR (if taking Coumadin) Date: Results: Did the patient receive meds? Referring Physician Date & Time PAGE 3 TRANSIENT ISCHEMIC ATTACK (TIA) AND MINOR STROKE CENTER PATIENT INFORMATION To our Patient: You have been diagnosed with a Transient Ischemic Attack ("TIA"). We are providing you with the following information to assist you in making decisions which may seriously affect your short- and long-term health: 1. Diagnosis and Future Risk a. Based on your symptoms, your risk for future TiA or stroke without further interventions is:  within 48 hours  within one (1) week  within three (3) months 2. Intervention: Further testing/evaluation a. You need to undergo further testing as soon as possible to lower the risk of future events. b. These tests cannot be done in the Emergency Department or a doctor's office. c. Patients who prefer testing in the hospital, cannot be seen quickly enough as an outpatient because of weekend/holiday, or are not felt to be stable enough for discharge by the evaluating physician may be admitted to undergo further testing. d. If you prefer to arrange your own further testing by a neurologist or other physician, please notify their office that there cannot be any delay in scheduling the tests as time is of the essence. 3. Expedited testing: a. Expedited outpatient testing/evaluation is available through our TiA Rapid Evaluation Center for patients who do not want/need to be admitted. b. Services include any further diagnostic studies that need to be completed, counseling on stroke risk factors and arrangements for further treatment and follow-up. 4. If you choose our TiA Rapid Evaluation Center for your care: a. An appointment has been made for in the Stroke Center on-- b. Please report to Main Registration desk at 8 a.m. on that date where you will be directed to a member of our stroke team. c. Be sure to bring with you * Current list of all of your medications, including vitamins, supplements and over-the-counter drugs * Results from any recent blood work, CT or MRI studies * Other medical records you have at home d. Depending on the complexity of your case, you may need to spend several hours in the hospital, but you will be discharged before 4 p.m. If you have any questions about this information, please call our stroke services coordinator at XXX-XXX-XXXX, M-F 9 a.m. - 5 p.m. If you have any further symptoms or concerns before or after business hours, call 911 or return to the Emergency Department. I have read and understand this information: Patient Signature Date & Time
STEP 3: OPERATIONALIZATION--Before opening the clinic, we met with leadership from radiology and cardiovascular service lines to discuss prioritization of TREC patients so that all the necessary testing indeed could be accomplished in one day. We made the case that they would be performing this testing on all suspected TIA patients anyway, but that they would be easier to schedule and would secure higher reimbursement when done as outpatients.
We required a physician referral to the clinic and saw only patients with suspected TIA events within two weeks to avoid getting swamped with patients who would more appropriately be seen and evaluated in other more traditional venues. We decided to hold the program at the hospital to facilitate testing in one day and also keep patients close to the ED in case of further events. We identified a ward clerk at the hospital who was available to call for urgent precertifications from insurance companies when patients had commercial insurance. We used the on-call neurologist and the stroke coordinator to staff the TREC, since both clinicians were at the hospital anyway.
PILOT STUDY RESULTS--During the first year we saw 74 TREC patients within a mean of 1.25 days from referral, 25 percent referred directly from primary care offices and 75 percent from the ED. Only one patient required a reminder telephone call to appear for their TREC appointment. Only two patients were admitted to the hospital from the TREC--one of whom was ultimately diagnosed with multiple cardioembolic strokes from atrial fibrillation and the second with recurrent psychogenic spells. The remaining 97 percent were able to complete their diagnostic evaluation and management as outpatients.
During that same year, 88 patients still were admitted to the hospital with a primary admission diagnosis of TIA. Obviously, ABCD2 scores were lower for TREC than hospitalized patients, but because the referral decision was left to the discretion of each patient's attending physician, we did see some ABCD2 scores of 6-7 in the TREC and patients with lower scores in the hospitalized group.
TIA patients triaged to the TREC were younger and were less likely to have a final diagnosis of TIA after undergoing their neurologic evaluation than those who were admitted to the hospital. TREC patients were more likely to undergo a complete diagnostic evaluation that included carotid ultrasound and brain MRI than the hospitalized comparison groups (See table 1).
TIA patients were evaluated in the TREC at significant cost savings compared to the contemporaneous hospitalized cases. Charges were lower for TREC patients, but expenses also were much lower, and the hospital's contribution margin was positive, not negative. Mean charges and contribution margin for all TIA evaluations, including both TREC and hospitalized cases, were significantly more favorable for this year than for the preceding four years when all TIA patients were admitted to the hospital regardless of their stroke risk (See Table 2).
CONCLUSIONS--We were able to secure referrals of selected suspected TIA patients to our outpatient TREC program, avoiding hospitalization and achieving a more efficient diagnostic evaluation without any patient safety issues at significant overall cost savings.
Many suspected TIA patients referred to the TREC turned out to have alternate diagnoses. This does mean that some patients underwent unnecessary diagnostic testing. In many instances the non-TIA diagnosis only became apparent after the diagnosic evaluation was completed, and without the TREC these patients would probably have been admitted to the hospital for further testing at much greater expense anyway.
Securing physician referrals and support from radiology and cardiology service lines was critical for the success of this program.
Neil R. Holland, MBBS, MBA; Michael E. Chan, MD; Martin M. Armor, MD; Nagakrishnal Nachimuthu, MD; O. Steven Boyu; Alex Puma; Mohammed A. Sheta, MBBCh; Florence Armour, MS; Michael Y. Hwang; and Shirley S. Hwang, MS
Neil R. Holland, MBBS, MBA, is director of neurology at Geisinger Health System in Danville, Pennsylvania, and professor of clinical medicine at The Commonwealth Medical College in Scranton, Pennsylvania. firstname.lastname@example.org
Michael E. Chan, MD, is a hospitalist in internal medicine at St. James Health Care in Butte, Montana.
Martin M. Armor, MD, is a cardiovascular disease fellow at North Shore University Hospital in Manhasset, New York.
Nagakrishnal Nachimuthu, MD, is an infectious disease physician at CHI St Luke's Health Memorial Clinics in Livingston, Texas.
O. Steven Boyu is director of health care informatics at Monmouth Medical Center in Long Branch, New Jersey.
Alex Puma is an analyst at Monmouth Medical Center in Long Branch, New Jersey.
Mohammed A. Sheta, MBBCh, is a nephrologist at South Carolina Nephrology and Hypertension in Orangeburg, South Carolina.
Florence Armour, RNC, MSN, NPC, is an advanced practice nurse and care coordinator at the Brain Tumor Center at Monmouth Medical Center in Long Branch, New Jersey.
Michael Y. Hwang, MD, is an internal medicine resident at Mount Sinai Hospital in New York, New York.
Shirley S. Hwang, RN, MS, AOCNS, is vice president of business development and clinical integration at Monmouth Medical Center in Long Branch, New Jersey.
(1.) Prabhakaran S et al. Misdiagnosis of transient ischemic attacks in the emergency room. Cerebrovasc Dis, 26(6):630-5, 2008.
(2.) Nadarajan V et al. Transient ischemic attacks: mimics and chameleons. Pract Neurol, 14(1):23-31, Feb 2014.
(3.) Qureshi Al et al. Factors associated with length of hospitalization in patients admitted with transient ischemic attack in United States. Stroke, 44(6):16015, Jun 2013.
(4.) RAC Prepayment reviews move forward as their targets feel heat from other audits. Report on Medicare compliance, 22(5), 2015.
(5.) Nahab F et al. Impact of an emergency department observation unit transient ischemic attack protocol on length of stay and cost. J Stroke Cerbrovasc Dis, 21(8):673-8, Nov 2012.
(6.) Olivot J et al. Two aces: transient Ischemic attack work-up as outpatient assessment of clinical evaluation and safety, Stroke, 42(7):1839-43, Jul 2011.
(7.) Baugh C, Ventakesh A, and Bohan J, Emergency department observation units: A clinical and financial benefit to hospitals. Health Care Manage Rev, 36(1):28-37, Jan-Mar 2011.
(8.) Sheehy A et al. Hospitalized but not admitted, characteristics of patients with "observation status" at an academic medical center. JAMA Intern Med, 173(21):1991-8, Nov 25, 2013.
(9.) Luengo-Ferandez R, Gray A, Rothwell P. Effect of urgent treatment for transient ischemic attack and minor stroke on disability and hospital costs (EXPRESS study): a prospective population-based sequential comparison. Lancet Neurol, 8(3):235-43, Mar 2009.
(10.) Horer S, Schulte-Altedorneburg G, Habed R. Management of patients with transient ischemic attack is safe in an outpatient clinic based on rapid diagnosis and risk stratification. Cerebrovasc Dis, 32(5):504-10, 2011.
(11.) Martinez-Martinez M et al. Transient ischemic attacks clinics provide equivalent and more efficient care than early in-hospital assessment. Eur J Neurol, 20(2):338-43, Feb 2013.
(12.) Johnston S et al. Validation and refinement of scores to predict very early stroke risk after transient ischemic attack. Lancet, 369(9558):283-92, Jan 2007.
(13.) Quinn T et al. ABCD2 scores and prediction of non-cerebrovascular diagnoses in an outpatient population a case-control study. Stroke, 40(3):749-53, Mar 2009.
(14.) Josephson 5 et al. Higher ABCD2 score predicts patients most likely to have true transient ischemic attack. Stroke, 39(2):3096-8, Nov 2008.
(15.) Sheta M et al. TIA evaluation - is hospitalization really necessary? Stroke, 44, Abstract ATP402, 2013.
(16.) Johnston SC et al. Short-term prognosis after emergency department diagnosis of TIA, JAMA, 284(22):2901-6, Dec 13, 2000.
(17.) Wu CM et al. Early risk of stroke after transient ischemic attack: a systematic review and metaanalysis. Arch Intern Med, 167(22):2417--22, Dec 10, 2007.
(18.) Hankey GJ. Impact of treatment of people with transient ischaemic attack on stroke incidence and public health. Cerebrovasc Dis, 6(suppl 1):26-33, 1996.
Table 1 ABCD2 SCORES, FINAL DIAGNOSIS OF TIA, AND DIAGNOSTIC TESTING ACCOMPLISHED FOR TREC VS. CONTEMPORANEOUSLY HOSPITALIZED TIA PATIENTS. ABCD2 Scores Final diagnosis of TIA Mean 0-3 4-5 6-7 TREC 1.8 72% 24% 4% 19% In-pt 3.8 40% 48% 12% 77% P<0.001 P<0.0001 P<0.001 Diagnostic testing completed CT EKG ECHO Carotid MRI TREC 100% 100% 99% 99% 89% In-pt 100% 100% 91% 88% 68% P=0.03 P=0.001 P=0.001 Table 2 FINANCIAL ANALYSIS, AVERAGE S/CASE. Before TREC During EREC pilot ALL TREC IN-PT COMBINED Revenue 6509 2270 6232 3826 Total direct costs 3376 425 3287 1550 Contribution margin 3133 1845 2945 2277 Indirect costs 3255 241 3236 1417 Net income -122 1604 -292 859
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Patient Care|
|Author:||Holland, Neil R.; Chan, Michael E.; Armor, Martin M.; Nachimuthu, Nagakrishnal; Boyu, O. Steven; Pum|
|Publication:||Physician Leadership Journal|
|Date:||Nov 1, 2016|
|Previous Article:||The value proposition: a physician leader's guide.|
|Next Article:||Creating pediatric specialty ambulatory care dashboards.|