The transition after hospital discharge involves communication and transfer of complex information at a time when patients and families are stressed. This communication may include diagnostic test results, discharge and medication instructions, and follow-up care.1 Failures in information transfer after hospital care are common. This may be caused by lack of communication between hospital-based and primary care physicians, incomplete or unavailable discharge summaries, and inaccuracies and confusion regarding discharge medication.2-4 These issues can harm care continuity and can lead to readmission, adverse events, and patient dissatisfaction with care.4-6
To improve patient outcomes and care coordination, solutions including alternative patient education methods, discharge planning, and disease management have been proposed.7-9 Although many of these interventions are home based, complex, and resource intensive, some of their effects may be achieved with postdischarge telephone calls.10,11 Most frequently, nurses or other clinical staff call patients to review information about medication administration, continued symptoms, or the postdischarge care plan.10,11
Postdischarge telephone calls have generally been found to be feasible and acceptable to patients1,12; furthermore, patients may be unlikely to initiate calls themselves when they have questions after discharge.13 This hesitation underscores the need for hospital personnel to initiate contact. A recent study found that postdischarge telephone calls significantly increased patient satisfaction, compliance with medication instructions, and the perception of health improvement.14 A trend toward lower readmissions was also reported.14 Evidence regarding the effects of postdischarge telephone calls on health outcomes (such as readmission) is conflicting. This is in part because of the variety of methods and samples used in studies.1 Calls have been substantiated to improve patients' overall satisfaction with care.10 In addition to improving the care experience for patients, postdischarge telephone calls can help hospital staff through notification of adverse outcomes.15 Postdischarge calls are especially effective in addressing both patient satisfaction and clinical outcomes when combined with other interventions. Implementing nurse-leader rounding and discharge telephone calls simultaneously has produced positive results.16
This project describes the Baylor Health Care System (Baylor) standardized process for placing postdischarge telephone calls to patients after a visit to the emergency department (ED). In addition, the metrics and guidelines related to this process are identified. These include lessons learned as a result of implementing the new process; models of delivery related to postdischarge telephone calls, including integration with an electronic health record (EHR); the future state of the postdischarge telephone call process; and preliminary findings of the new process.
Baylor is a not-for-profit healthcare system in North Texas comprising 26 owned, operated, ventured, or affiliated hospitals; 21 ambulatory surgery centers; 7 short-stay surgical hospitals; 136 Health Texas provider network locations with 450 physicians; 41 satellite outpatient clinics; 4 senior health centers; 4575 medical staff members; more than 20,000 employees; and a research institute.
In January 2010, following the commitment of Baylor organizational leaders, including members of the system-wide patient-centered experience council, Baylor implemented a standardized process for placing postdischarge telephone calls in the EDs of 10 acute care hospitals. Before this, calls had been placed inconsistently and focused on both patient satisfaction and experience. The ED council at Baylor, comprising medical and nursing emergency service leaders from each site, was the first to develop and implement this process in a standardized way.
The discharge telephone call is introduced at the time of discharge, with the provider explaining that he/she would like to reach the patient within the next 24 to 48 hours. The patient is told that the call is made to ensure that the patient understands home care instructions and to answer any questions. The provider asks if there are alternate telephone numbers or supporting family members or friends to be contacted, and patient preference is noted.
Staff standardized the call using HIPAA-compliant directed conversation. The script asks patients whether they have questions about 4 topics: discharge instructions, medications, symptoms, and initiating a follow-up appointment with their physician. Steps for escalation and documentation of situations and issues to leadership or risk management have been identified after more than 300,000 calls have been made.
Baylor ED staff made 26,803 postdischarge telephone calls in August 2010 in 10 hospitals. The attempt rate (the number of attempts staff made to discharged patients) was 92.3% (Figure 1). The actual contact (staff spoke directly with the patient or patient's caregiver) was 34.5%, a solid representation of our patient population.
|Figure 1. Baylor Health Care System postdischarge calls attempt and contact rate. Average attempt rate for system: 92.3%. Average contact rate for system: 34.5%. Total calls made: 26,803.|
Of the 9,240 patient contacts, 1,041 interventions were provided. Some patients received more than 1 intervention. The interventions (Figure 2) included clarification of medication and/or home care instructions (n = 346, 34%) and reminders for follow-up appointments or assistance with referrals (n = 314, 30%). Patients who reported new or unresolved symptoms were instructed to return to the ED or their primary care physician (n = 346, 33%). There were 29 patients requiring immediate escalation of care. Regarding these patients, 348 estimated lives were touched when data were extrapolated to an annual figure (29 calls/month x 12 months). In the same fashion, extrapolating 1,041 times helped/month x 12 months shows an annual impact of 12,492 key communications provided by these calls. Additional analysis is required to examine the financial impact on avoidable complications as a result of early medical treatment or follow-up care. Further analysis is also needed to determine the avoidable cost of care related to readmissions as part of chronic disease management.
|Figure 2. Types of interventions provided to patients.|
There were 17,563 (65.5%) patients whom we were not able to contact during the same period; no relevant conclusions can be drawn from this result. However, given the findings of the contacted patients, Baylor will continue efforts to increase contact rates. Models for call delivery will be assessed for potential management of inbound opportunities.
Focusing on 1 facility (Baylor Medical Center Garland), Figure 3 shows the positive trends in patient satisfaction in "likelihood to recommend" over a 12-month period. Similarly, Figure 4 shows an upward trend in the percentage of patients who remembered receiving calls. Cursory analysis shows that patients who remembered receiving postdischarge calls tend to give higher scores in "likelihood to recommend."
|Figure 3. Baylor Medical Center Garland: likelihood to recommend-emergency department (mean score). Range, 0 to 100.|
|Figure 4. Baylor Medical Center Garland: percentage remembering receiving calls-emergency department (mean score). Range, 0 to 100.|
Beginning in November 2010, there was a trend toward a higher percentage of ED patients remembering the postdischarge calls, as indicated from patient experience survey questions (Figure 4). Trend lines for patients' likelihood to recommend these services closely paralleled patients' recall of receiving the postdischarge call. Future work will assess the readmission rates for any significant correlation to a postdischarge call relationship.
Always Leave a Message, Carefully
In the early stages of making calls, it was discovered that if a call was placed to the patient's home, caller ID was an issue. If no message was left, patients and, sometimes, families were concerned that the ED had called about something urgent. In response, patients and members of their families would attempt to call the ED to inquire about the nature of the call. The script now includes language regarding how to handle messages and return calls.
Avoid Duplication: More Care Is Not Always Better Care
For patients who may have experiences in multiple emergency or inpatient settings, the likelihood that they will receive more than 1 call increases. Each department that engages with the patient in a discrete visit should identify who has the primary area of responsibility for calls. At Baylor, fast-track patients may receive calls from both the ED and radiology; patients might be called in up to 3 separate telephone encounters. This becomes both an efficiency and credibility issue when patients experience a lack of coordination with their follow-up and plan of care. To ensure true process improvement, departments should meet and coordinate efforts and information exchange. Launching calls without clear accountability for ownership and coordination is inefficient and is a lesson learned.
A manager of one of the Baylor EDs recalls the point when she realized how important these calls are and that they were not going to happen if she did not take the lead. She became committed to showing her staff that calls are an important part of continuous care responsibilities. The manager began making calls herself, attempting to contact first 10%, then 20%, then 50% of the patients. Over time, she was successful in transitioning the calls to staff, with recommendations about sharing the work. Eventually, select staff members who enjoyed and became proficient at making the calls executed them for the department by exchanging duties with other colleagues.
From our experience, it is recommended that managers begin by making the calls for a few shifts to increase understanding of the process of the call; to learn about the strengths, weaknesses, opportunities, and threats of the new system; and to identify any systems issues associated with the process. Managers must have a firm knowledge of the complexity of the calls from their patient population and the expected time and skill set required.
Report Early and Often, in Small Bites
One manager began the practice of sending summaries from the previous week's calls to all staff and administrative and physician leaders via e-mail. These reports included the number of questions answered and patients helped, the number of patients who were asked to return for care, patient concerns, and positive affirmation statements. The new system of implementing telephone calls and the process of the calls have proven valuable to support staff engagement. Employees are able to see they are making a difference. They are also able to receive recognition for exceptional patient care. The telephone call process reinforces the belief that following a patient's transition to home is part of the continuity of care expected for patients treated at Baylor.
Issues With Access to Patients
Baylor's EDs handle approximately 443,000 visits per year. An important finding was the high number of inaccurate or nonworking telephone numbers documented for patients. On average, 5% to 10% of telephone numbers were consistently incorrect numbers. These were either inactive, invalid numbers or they did not belong to the patient or his/her designated family. The documentation of these nonworking telephone numbers and methods used to validate telephone numbers to increase contact rates are key opportunities for improvement in the future. Particularly in ED, where critical values and pending test results may require follow-up, patient contact is an important issue to address. A thematic perception was reported among staff that many inaccurate numbers from patients were given to ED staff with an attempt to thwart contact. Addressing this issue and either validating or disproving this perception is a part of future research.
Service Versus Quality
At the onset of this project, leaders sought to ensure that patients felt well served and had no outstanding questions. It was soon determined that unless this work is framed as a component of a quality endeavor, the task of a telephone call will be triaged off the list of priorities for clinical providers as a nonpriority.
The ED is the "front door" of the hospital for a majority of Baylor patients, and the pressures of patient experience results can be perceived to affect only customer loyalty. "Loyalty" may be deemed a lower priority consideration in emergency settings where urgency is high and choice is more limited. It is difficult to measure customer lifetime value implications for healthcare if loyalty is lost in the ED. There are economic and regulatory pressures on ED and inpatient leaders to eliminate readmission and avoidable events. Helping patients transition home safely is a way to reinforce care. Early in the implementation process, the service call was translated to a "quality" call, focusing on patients' needs related to medication education, discharge information, signs and symptoms of recurrence, and physician follow-up concerns. Clinical oversight of the postdischarge calls is critical to success. Tracking outcomes is easier when the documentation is recorded into an EHR. Electronic continuity provides daily work lists to leaders, identifies patients discharged to home versus other facilities, and allows for multiple serial contacts in a live documentation setting so that work flow is not duplicated or untimely. Moreover, entering the findings from each call builds the longitudinal data set for assessment of general trends as well as patient-specific behaviors and risks. Finally, the timeliness of documentation enhances the ability of the follow-up physician to assess the postdischarge concerns.
Placing postdischarge telephone calls to every patient who is discharged home from the ED is a resource-intensive process. Standardized documentation of the calls, reporting the value of the calls, and auditing to ensure HIPAA compliance all required thoughtful implementation and some additional resources.
The 3 models of calls for the pilot included a centralized model for the entire hospital (specific staff members make all postdischarge calls), a unit-based model involving the distribution of calls across units or departments (someone who works on the discharge unit makes the call), and outsourcing of the postdischarge calls to a third-party vendor. Plans call for an analysis of the financial resources, the escalation of issues, turnaround time to resolve issues, and the need for data management in optimizing care for patients as they transition home. The ability for double-loop learning, the process of examining not only alternative means but also possibly alternative ends,17 for discharging units will be an important component to consider, as will the availability of timely discharge information in the EHR before physician follow-up.
The Reporting of Global Results
Implementing a system of calls to patients without having a reporting mechanism and format defined, drafted, and set in a template resulted in frustration for involved leaders. Upon this realization, the ED council and office of patient centeredness developed a report representing activities supporting the initiative (Figure 1). In defining the elements for the report, it was quickly discovered how disparate the process was across Baylor. It was challenging for the council members to have to regroup, redefine, and retrain all staff to perform these calls in a standardized manner. Although the system's patient-centered experience council had reviewed these exercises to carefully define the process, the original message was diluted and lost in communication within local cultures. The process could not be successfully adjusted until this same work of defining process elements was accomplished specifically with the teams who were making the calls and submitting the data for reporting. Restandardizing of the process took approximately 8 months. A concerted effort brought about the eventual success of standardization of calls and reporting of results ( Figure 1 and 2), and at this point, the data became meaningful for trending and learning.
The implementation of standardized postdischarge telephone calls has provided important lessons about the need for interpretive services for patients for whom English is not the primary language. The US Department of Health and Human Services and The Joint Commission have identified the implementation of language access services as an important priority for healthcare organizations.18,19 In considering that a quality call helps ensure safe transition to home, the postdischarge calls must therefore have access to an interpreter who is trained in medical terminology.20 Planning for this aspect of service requires reporting the frequency of the use of these services and the budgetary resources for access to them. At Baylor, interpretive services are used when a patient requires a translator.
Baylor Medical Center Garland (hospital 1) has been the first Baylor hospital to document postdischarge calls in conjunction with EHR documentation. Other Baylor hospitals will transition from paper-based to EHR-based documentation of calls.21 The features needed for this documentation were not available in the EHR software, and customization was required to advance this work on postdischarge calls, thus slowing the spread of a successful practice.
A second trend affecting postdischarge telephone calls is the growing importance of continuity of care, particularly in light of the aging US population and increasing prevalence of chronic illnesses associated with ED visits and readmissions.22,23 Condition-specific care coordination programs (ie, multidisciplinary interactions between patients and providers that focus on environmental assessments, education, communication, and discharge planning) have been shown to be effective in improving a variety of patient outcomes, including mortality, rehospitalization rates, and quality of life.7,24 These interventions are usually home based and therefore have important limitations, including cost and compliance. Hospital-based interventions such as postdischarge telephone calls may improve patient outcomes while also being cost-effective and practical.
With public reporting of the patient experience now poised to affect financial reimbursement through value-based purchasing, postdischarge calls are key for seamless transition to home. Patient feedback scores addressing communication, medications, and discharge instructions align well with this new process from the Baylor experience.
Postdischarge telephone calls can improve patient satisfaction, outcomes, and care continuity. Baylor has implemented a standardized process for placing ED discharge calls to patients to ascertain whether they have questions about their discharge instructions, medications, symptoms, or the initiation of follow-up appointments with their physician. Leadership support has played a crucial role in this process. Important lessons learned include the need for auditing to ensure standardization in placing and documenting calls, the varying budget and staffing needs required to support the calls depending on the model chosen, the need to optimize contact rates and reduce the number of nonworking telephone numbers, the need for interpretive services for patients for whom English is not the primary language, and the opportunities to ensure understanding at the time of discharge with double-loop learning for managers and staff. The future of postdischarge telephone calls at Baylor will be influenced by several trends, including the widespread adoption of EHRs and the growing importance of care continuity and chronic-disease management. Because of the number of patients who require information after returning home and the initial success of these telephone calls in its EDs, Baylor plans to standardize these calls in the inpatient and ambulatory settings in the future and intends to implement these calls in a wider variety of care settings. Baylor will implement the calls in conjunction with outpatient care after a review of the financial and contact implications of the various models.
The authors thank Kanan Garg, MS, for harvesting the data for this project; Kathleen M. Richter, MS, MFA, ELS, Kelli R. Trungale, MLS, ELS, and Linda Tjiong, RN, MSN, DBA, for writing and editorial assistance; and especially the Baylor Emergency Department Council for their steadfast care and service to our patients and their families.