Optimizing handover for family medicine outpatients using an electronic medical record–integrated tool

Best practices

To mitigate these risks, various handover tools have been developed and, when implemented, have successfully decreased medical errors and preventable adverse events.^3-5 Unfortunately, most of these tools have been developed for inpatient settings, with a paucity of tools specifically designed to tackle challenges unique to ambulatory settings, especially academic ones. In FMTUs, for example, the transfer of clinic patients from graduating to incoming residents occurs annually, increasing the likelihood that miscommunications compound over time.^6,7 Of importance, residents themselves acknowledge the need for some form of handover and believe the relevant documentation should be retained in the medical record.⁸

Our experience

At our FMTU, incoming residents take over the care of patients from outgoing residents at the beginning of their training. They are then responsible for their unique panel of patients for the duration of their residency, until the transfer process occurs again. Occasionally, graduating residents enter a “handover note” into the electronic record of patients they deem to be particularly complex or who would otherwise specifically benefit from a special communication. However, we have never had an official process or policy to support the transition of patient care between residents.

In an effort to address this so-called “revolving door of resident continuity practice,”⁷ we sought to develop an electronic medical record (EMR)–integrated handover tool to improve communication at the critical point of transition of care. Our goal was to improve patient safety by decreasing the number of missed results, investigations, consultations, and follow-ups, and to increase resident comfort, efficacy, and communication.

Tool development

To assess the perceived need for a handover tool in our teaching unit and to gain input from key stakeholders on its format, we distributed an anonymous qualitative needs assessment by e-mail to all family medicine residents and all faculty physicians and nurses at our site. In total, 20 residents and 18 faculty physicians and nurses responded to the survey. An exhaustive literature search on existing handover tools was also performed to guide the development of our own tool.

After we had synthesized data from the needs assessment and literature review, the following items were deemed necessary to include: action items, including investigations requiring follow-up, investigations to order, pending consultations, and patient complexity and the need for longer appointment times; and acknowledgment of handover—a feature to signify that a faculty physician at the FMTU had been notified that the handover tool was completed. The draft was reviewed and iteratively improved based on resident and faculty physician feedback, and then uploaded into the clinic EMR. The tool and instruction sheets for residents and preceptors are available from CFPlus.^*

Tool evaluation

Four charts belonging to patients from resident practices were selected based on their representativeness of resident practice patients. Handover tools were completed by the investigators for each of these patients. Twelve resident volunteers were then given 3 minutes to review each patient’s chart and list as many items as they could that required follow-up. Each resident was shown 2 charts with the embedded tool and 2 without. Residents were then asked to complete an anonymous postintervention survey, created by adapting domains from the Handoff CEX (clinical evaluation exercise), a validated handover assessment tool.⁹

A t test procedure was used to assess the difference between the number of items identified with and without the tool for each of the 4 patient charts and to assess the difference overall when combining all patient charts. Using a P value of .05, we found a significant difference between the number of items identified for follow-up with and without the handover tool. This difference was observed overall, as well as for each patient chart individually (mean [SD] 11.54 [3.63] vs 4.96 [1.81]; t₂₄ = 5.61; P < .05).

Qualitative data from the postintervention survey included Likert-type response scales. All 12 resident volunteers reported improved comfort when using the handover tool. All agreed that the handover tool contained essential content and would improve communication between residents, enhance organization and efficacy, and lead to improved patient safety and care.