Practice Economics

Nine Ways Hospitals Can Use Electronic Health Records to Reduce Readmissions


Editor’s note: This is the first of two articles from SHM’s Health Information Technology committee offering practical recommendations for improving electronic health records (EHRs) to reduce readmissions, along with practice-based vignettes to support the recommendations.

Despite limited support from the medical literature, hospital teams know that technology, specifically electronic health record (EHR) technology, can improve healthcare quality. Given 2009’s Health Information Technology for Economic and Clinical Health (HITECH) Act, the federal government is betting on this as well. These hospital teams, often led by hospitalists, are charged with creating workflows and EHR build that will affect measurable quality indicators. Hospital finances, tied to Medicare pay-for-performance incentives, hang in the balance.

As a hospitalist and chairperson of SHM’s IT Quality Subcommittee, I helped lead an effort to examine how technology and EHRs could be used to reduce readmissions. The subcommittee was composed of eight hospitalists from around the country and two mentors, Jerome Osheroff, MD, FACMI, of TMIT Consulting, and Kendall Rogers, MD, CPE, FACP, FHM, of the University of New Mexico. The goal of this effort was to create reproducible models of how EHR and technology in general could be leveraged to reduce readmissions.

Members of the committee initially were asked to evaluate all-cause, 30-day readmissions at their respective institutions. Any hospital with a readmissions rate less than 16% over the previous year was considered “high performing.” Members were asked to advocate for one technology/EHR intervention that had the most impact locally. Interventions were vetted within the committee and based on literature review.

Specific categories evaluated included:

  • Readmission risk assessment;
  • Communication with referring physicians;
  • Medication reconciliation;
  • Multidisciplinary rounds;
  • Patient education;
  • Discharge coaches;
  • Patient-centric discharge paperwork;
  • Post-discharge coordination of care; and
  • Medication compliance.

These site-specific experiences could be considered “springboards” for randomized trials of likely successful interventions.

Recommendation: Use readmission risk assessment to apply resources to most appropriate patients.

Ned Jaleel, DO, MMM, CPE, a hospitalist and informaticist for Meditech Corp., and Maruf Haider, MD, a hospitalist and informaticist for INOVA Healthcare, have mapped implemented processes for real-time assessment of readmission risk stratification and “measurevention” based on this data.

Augusta Health in Fishersville, Va., uses Meditech EHR to extract relevant data about risk assessment and display this data to case managers using the LACE model (length of stay, acuity, comorbidities, ER visits). The modified LACE model included medication information to create a readmission risk score. Case managers can then determine which patients require the most care and attention from the multidisciplinary team.

Dr. Haider has taken this process a step further by using the LACE score to determine the need for specific tiered intervention based on established risk. Average risk patients are simply set up with a follow-up appointment within seven days. Higher risk patients are set up with health coaching, home nursing, or more intense inpatient multidisciplinary rounds based on four tiers of risk stratification. Risk stratification is discussed on rounds, and providers are requested to order the additional services. Patients referred to transitional services had a 6.5% readmission rate compared to the hospitalist groups overall at 15.6%.

Recommendation: Use electronic communication to increase reliability of contact with primary care physicians.

At Lahey Health System in Burlington, Mass., I knew that hospitalists needed to improve communication with PCPs. Telephone communication was unreliable, and discharge summaries were not being delivered to referring physicians in a timely fashion.

The hospitalists already were using a homegrown “patient handoff report” to track currently admitted patients, along with clinical summaries. A decision was made that secure e-mail, driven by data in the handoff reports, could provide the solution. Because the system was between inpatient EHR vendors, it would need to be developed by in-house IT services. A specific challenge would be referring physicians with no attachment to the health system.

Using a secure messaging vendor (ZixCorp), we were able to create e-mail messages to referring physicians using data already in the handoff system to avoid duplication of data entry. The benefit to referring physicians was the brevity and timeliness of the data and the ability to ask questions directly via return e-mail. Fortunately, we were able to request enough e-mails to ensure that the majority of patients with non-system physicians would allow this type of communication. This and other interventions have allowed Lahey to reduce 30-day readmissions to less than 15%.

The benefit to referring physicians was the brevity and timeliness of the data and the ability to ask questions directly via return e-mail.

Recommendation: Use pharmacy resources to improve quality of medication reconciliation.

Rupesh Prasad, MD, MPH, of Aurora Healthcare in Milwaukee, Wisconsin, has spent part of his professional career optimizing medication reconciliation. The key has been incorporating EHR workflows that allow pharmacy to take an active role in medication reconciliation.

Initially, pharmacy technicians collect the home medication list using information from pharmacy, patient and family, and primary physician and enter into the system. This allows the admitting provider to perform the most accurate medication reconciliation possible. The EHR has allowed more accurate sharing of medication lists across inpatient and outpatient care areas and helped to prevent dosing errors and duplication via decision support. The discharge materials provide information in a patient-centric manner that helps reduce medication errors at home. These and other interventions have helped reduce readmission at Aurora to less than 16% at 30 days.

Recommendation: Use EHR resources to support BOOST rounds to improve collaboration.

Gaurav Chaturvedi, MD, of Northwestern Lake Forest Hospital in Lake Forest, Ill., has used his Cerner EHR in collaboration with SHM’s Project BOOST to reduce readmission in 2013 to a very impressive 11% at 30 days.

The key to success is daily multidisciplinary rounds at the bedside involving all members of the care team, including physicians, nurses, case managers, pharmacy, physical therapy, and social work. This ensures that all members of the care team, including the patient and family, are up to date on the care plan at the same time. The EHR has supported this process through creation of templates that pull in critical information for rounds such as ambulation, central lines, VTE prophylaxis, Foley, and medication reconciliation.

With all of the information readily available in the same template for rounds, the team can focus efficiently on the goals of care and discharge needs required to prevent readmissions.

Recommendation: Improve patient education by integrating with discharge workflows.

Dr. Chaturvedi also has experience integrating patient education into EHR workflows. His initial efforts involved heart failure education and resulted in reduction over 48 months in heart failure readmission rates to 8.3% from 27% at 30 days. Prior to discharge, heart failure patients received a guidebook with the medication summary, appointments, diet, and EHR-integrated educational materials from the Krames StayWell database. This highly successful, partially EHR-based intervention included a scale to promote daily weights.

After EHR implementation in 2012, Lake Forest Hospital needed to leverage similar successful functionality into their Cerner EHR. The hospital worked with Cerner to develop Mpages that allowed seamless multi-provider entry on discharge paperwork. This would include primary and secondary diagnoses and warning signs. These entries would “suggest” Krames’ patient-centric educational materials that would discuss diagnosis and treatment, along with warning signs specific to the diagnosis. TH

Dr. Finkel is a hospitalist at Lahey Hospital and Medical Center in Burlington, Mass. Email questions and comments to [email protected].

Nine recommendations for hospitals that are ready to use EHR technology to reduce readmissions

  1. Use readmission risk assessment to apply resources to most appropriate patients;
  2. Use electronic communication to increase reliability of communication with primary care physicians;
  3. Use pharmacy resources to improve quality of medication reconciliation;
  4. Use EHR resources to support BOOST rounds to improve collaboration;
  5. Improve patient education by integrating with discharge workflows;
  6. Use EHR workflows to support discharge coaches;
  7. Support EHR build that creates patient-centric multidisciplinary discharge paperwork;
  8. Support coordination of care with electronic means of scheduling post-discharge care prior to discharge; and
  9. Reduce technical and financial barriers to communication of medication list and medication compliance at home.

Next Article:

   Comments ()