CEO SUMMARY: Labs today are finding ways to send actionable information to referring physicians. The laboratory at MedCentral Health System in Ohio uses electronic alerts to inform clinicians about test results that may be significant. One alert helped the lab cut length of stay for patients by 1.5 days and saved $300,000. Another test identifies congestive heart failure patients who are most at risk of readmission and thus helps the hospitals avoid penalties by keeping its CHF readmission rate low.
ONE STEP ON THE PATH toward more consultative laboratory medicine at a Midwest health system involves the use of electronic alerts to inform physicians about laboratory test results that may be clinically significant
“Our lab’s goal is to develop effective ways to send actionable information to referring physicians in real time,” stated Eugenio H. Zabaleta, Ph.D. “The time is fast-approaching when it will no longer be adequate simply to send a lab test report back to the referring physician.”
Zabaleta is a clinical chemist in the laboratory at MedCentral Health System, a cancer and heart hospital in Mansfield, Ohio. MedCentral is using electronic alerts to inform physicians about test results that may be clinically significant.
“One alert is associated with lab testing done to identify patients with CHF,” he noted. “In this case, the lab runs tests to assess patients’ relative risk so the hospital can manage closely any patients likely to be readmitted.
“Another alert helps physicians make a more accurate and faster diagnosis of patients having a heart attack after they present in the emergency department (ED). When patients show up in the ED with chest pain, treating physicians need to rule out the possibility of acute myocardial infarction (AMI),” he continued. “The electronic alert for patients who may be having an AMI is based on the patient’s troponin levels in serial measurements.
“Troponin indicates a possible heart attack because it is released when heart muscle is damaged,” Zabaleta said. “This electronic alert was developed after ED physicians found we were not immediately identifying all patients who suffered a heart attack. In the lab, our goal was to provide them with an alert that would give them actionable information in real time.
“For certain patients in the ED, heart attack can be difficult to diagnose because they may have comorbidities or be on pain medications and not realize they are having an AMI,” he said. “Plus, they may have the symptoms of a heart attack (such as chest pain), but not actually be having an AMI.
“The problem was that we were missing some of these diagnoses initially,” Zabaleta said. “Each patient missed in this fashion required extra days in the hospital and thus increased costs. Our goal was to identify each AMI patient as soon as possible. Delayed diagnoses meant we were spending an extra $300,000 on Medicare patients each year.
“To identify these patients immediately, we record their troponin levels upon admission and again after 6 and 12 hours,” he noted. “By tracking the levels over time, we can see a sharp spike that would indicate a heart attack.
“Due to patient severity and comorbidities, some patients are admitted as inpatients,” he said. “But the ED test orders are continued, making the AMI diagnosis possible in the inpatient setting. The conti- nuity of patient care should be smooth and not disruptive, which is one goal of accountable care.
MIddleware Supports Alerts
“But it’s not enough simply to give the ED physicians the lab test results,” he continued. “We wanted to show the trend over time and develop an alert that would prompt the lab to call the provider when a patient’s troponin level spiked. To prompt the phone call, we added middleware to the laboratory information system (LIS).
“After installing this middleware from Pacific Knowledge Systems in Australia, we conducted a study,” said Zabaleta. “Patients in the control group had a length of stay (LOS) of 7.89 days. By contrast, patients in the study group had an LOS of 6.17 days, more than 1.5 days shorter than that of the control group.
“The only difference was the use of the troponin alert system for the study group, which meant we identified those patients in the study group who had an AMI earlier and they were treated quickly, reducing LOS,” Zabaleta explained.
“By cutting LOS for these patients, we saved Medicare about $300,000,” he said. “The bottom line is this project is improving timely communication between the lab and the provider, which leads to more effective and efficient care.”
Reducing Readmissions of Patients with CHF
RISK STRATIFICATION FOR PATIENTS with congestive heart failure helps MedCentral Health manage CHF patients aggressively.
Like the electronic alert for acute myocardial infarction (AMI), the alert for CHF patients uses middleware from Pacific Knowledge Systems. For this initiative, the hospital administrators wanted to reduce the readmission rate for CHF patients.
“Called Heart Success, this initiative is based on assessing the relative risk according to each patient’s troponin levels over time and on the patient’s BNP levels,” stated Clinical Chemist Eugenio H. Zabaleta, Ph.D. “The B- type natriuretic peptide (BNP) is measured in the blood. BNP rises when heart failure symptoms worsen and falls when heart failure symptoms become more stable.
“After performing these two lab tests, we can calculate a risk score for each patient,” he continued. “The higher the risk, the greater the chance the patient will be readmitted.
“We stratify patients according to their BNP and troponin results and base the electronic alert on this lab test data,” explained Zabaleta. “The alert automatically sends this information by email to nurses and clinicians in the Heart Success program. This helps clinical staff know which patients will require the most attention upon discharge.
“This alert program is quite successful, as shown by the numbers,” he said. “In 2011, CMS reported the national rate for readmission within 30 days of discharge for patients with heart failure is 24.8% and can be as high as 40%. At MedCentral, that rate for patients who are not enrolled in Heart Success is 26.7%. But for our enrolled patients, it is only 4%!”