Perspectives in Hospital Medicine

Patient‐centered communication is a strategy that is used to promote shared understanding of the plan of care among providers and patients.[1, 2, 3] Caring for hospitalized patients is a collaborative effort that requires seamless patient‐centered communication among a rapidly changing care team to safely progress a patient from admission through discharge. Yet, hospitals continue to struggle with improving the complex and increasingly electronic conversation patterns among care team members and patients to achieve effective patient‐centered communication.[4, 5] When members of the care team operate in this environment, patients often receive conflicting information regarding their plan of care, medications, and test results. Ineffective communication can lead to a suboptimal patient experience, additional costs, medical errors, and preventable adverse events.[6, 7, 8, 9, 10]

A critical first step to improving patient‐centered communication is identifying the care team.[11, 12] Accurate and reliable identification of all care team members is a pressing information need; it is fundamental to efficiently conveying information about the plan of care to those who know the patient the best, must make timely decisions, or will assume care once the patient leaves the hospital.[13] Furthermore, it has implications for engaging patients more meaningfully in their care.[14, 15, 16, 17] Ideally, the process of identifying an individual caring for the patient in a specific role is quickly and reliably determined from the electronic health record, the single source of truth where any provider can quickly identify other team members. This source of truth can be updated manually when individual members assign and remove themselves from the care team, or automatically when accessing the patient's record, writing a note, placing an order, or adding a patient to a coverage list. When providers correctly identify other team members in this way, hospital paging directories and secure messaging tools that link to the electronic health record become more effective at supporting care team communication.[18]

In general, the process of identifying care teams is difficult,[19] and maintaining role assignments in the electronic health record is equally challenging. Vawdrey et al. previously reported that care team lists are inaccurate and cannot be used to reliably identify other members at any given moment.[18] The inability to identify team members often leads to incorrectly routed pages, e‐mail messages, and phone calls.[20] Consequently, the potential to reliably manage the care team and improve electronic communication remains untapped, rendering team collaboration and care coordination less effective.[18, 21, 22]

In recent years, the trend toward restructuring inpatient teamsgeographical localization, structured communication interventions, teamwork training, and interdisciplinary roundswould seem to diminish the need for electronic care team identification, as those efforts have already made a positive impact with regard to interprofessional communication and collaboration, team satisfaction, and adverse events.[23, 24, 25, 26] Nonetheless, interdisciplinary teamwork, though critically important for patient‐centered communication, does not completely obviate the need for accurate and reliable care team identification.[26] Although care teams are statically located on units, the plan of care is dynamic; it evolves when the patient's status changes, when new information becomes available, and when key longitudinal providers (eg, primary care physician, subspecialty consultant) make recommendations. Thus, information conveyed as a team on rounds quickly becomes out of date, requiring additional forms of communication. Furthermore, due to frequent ad hoc coverage among team members, the identity of providers covering the patient at any given moment is often not clear.[27] This is particularly problematic for nonunit‐based providers who try to communicate with unit‐based care team members. These providers, in particular, have valuable knowledge and insight that can aid the primary team in decision making.[28, 29] However, they typically do not participate in rounds, often waste time identifying responsible providers,[20] and may communicate their recommendations directly with the patient without discussing with the primary team. These factors in part explain why geographic localization has shown limited improvement in shared understanding of the plan of care.[23]

From the perspective of patients and caregivers, identification of the providers entering and leaving their room is also challenging; only 11% to 51% of patients identify their providers correctly.[30] This adds to confusion regarding who is responsible for which aspects of the patient's care and can negatively affect the perception of the quality of care received.[31] Use of whiteboards has been shown to improve the proportion of patients who could identify key providers,[32] but these are not reliably updated and generally cannot accommodate all team members. When face cards are used, patients and caregivers report that they are more likely to identify their providers correctly.[14, 33, 34] However, potential confusion may ensue when another provider assumes care of the patient in the same role. Finally, use of technology to display team members at the bedside is typically a feature that patients like and can improve identification of care team members.[14, 15, 16] Yet, patient engagement technologies are not readily available in the hospital setting,[35] and ideally should be linked to the electronic health record, which again must be reliably updated.[11, 12, 15, 16]

If care team identification is so critical for delivering effective patient‐centered communication, why is maintaining role assignment problematic? At the individual level, reasons include discontinuity of the care team due to changing clinical rotations and intrateam coverage, shift‐based schedules, and lack of awareness and underutilization of functionality. Additionally, clinicians may have different ways to maintain lists of patients. At the institutional level, functionality to enforce role assignment when accessing patient records may be disabled (to avoid perceived burdens on clinical staff or nonclinical personnel who require access for administrative functions). Finally, electronic health record vendors currently have no incentive to adopt functionality that supports more effective care coordination across settings.[22]

However, more than technical solutions and policy changes are required; care team identification in the electronic health record requires a change in institutional culture. Maintaining an accurate relationship to each patient requires work without tangible benefitsthe benefits accrue only when everyone else identifies their role on the teama tragedy of the commons. This can be illustrated by our own experience. We conducted a quality improvement initiative (Table 1) as a part of 2 concurrent research initiatives that serve to promote patient‐centered communication:[12] PCORI (Patient‐Centered Outcomes Research Institute Transitions), the goal of which is to improve care transitions within the Partners' Pioneer Accountable Care Organization; and the PROSPECT (Promoting Respect and Ongoing Safety Through Patient‐Centeredness, Engagement, Communication, and Technology) project, an initiative funded by the Gordon and Betty Moore Foundation to eliminate preventable harms in the acute care environment.[29] Our goal was to electronically manage the care team with a high degree of fidelity. We enhanced a home‐grown application, which was developed to improve management of team lists for inpatient providers, accessible from our electronic health record, to facilitate role assignment. Specifically, we leveraged existing care processes (eg, nursing log‐on to the electronic medication administration system) to automatically assign certain providers to the care team at change of shift, added functionality to make it easy to assign a provider to all patients on a list for a defined period of time, and encouraged providers to assign their role by demonstrating benefits including quick access to patient‐specific group e‐mail and secure messaging tools (Table 1, Key Facets). The initiative was well‐received by most disciplines, but uptake was suboptimal. Our research assistants routinely assigned residents and others to the care team because our proactive attempts at advertising and reinforcing use of the application failed to reach a critical mass. Most did not see immediate benefits because it was an added step to their busy day, had other methods of managing team lists, and only saw benefit if everyone else participated. Key facets of our care team identification initiative, successes, and challenges are outlined in Table 1.

There were a few glimmers of hope, however. On several PROSPECT units, we displayed team members on a tablet‐based patient portal so that patients would recognize their providers.[11, 17, 36] Similar to recent work by O'Leary et al.,[14] patients on PROSPECT units were able to correctly identify several care team members, but regularly asked why other providers (eg, consulting fellow) were not listed. Those providers asked the same question, and some eventually learned to assign their role via the application. As part of PROSEPCT, we visited other institutions and learned of an effort to display team members on high‐definition televisions in the patient's room. Several providers, wondering why they were not listed, learned to assign their role and their picture then appeared. Social pressure was the driving force.

Coincidently, we recently implemented a new electronic health record at our institution. Anecdotally, although no formal policy was established, many providers (eg, attendings, first responders, nurses, care coordinators, and other unit‐based providers) appear to be assigning their roles. Other providers (eg, dieticians, physical therapists, residents) also assign their role, but often fail to end role assignments upon completing their rotation or when the patient transfers to another service. Finally, even when actively involved, most subspecialists still do not designate their role. Despite these gaps and inconsistencies, we have made progress toward improving care team identification. The reasons for this progress are straightforward; during required training for the new electronic health record, all inpatient providers were taught to assign their role on the treatment team when assuming care of patients and now have 1 option for managing team lists. However, most providers were not trained to end their role assignments, and many have learned that role assignment is not required to access the patient's record; functionality to enforce this was disabled. Based on lessons learned from our experience,[12] we offer several strategies that hospitalists can employ to improve care team identification in the electronic health record (Table 2).

In the future, care team identification in the electronic health record can be automated by integrating directly with electronic workflows, online scheduling applications, and provider directories. Hospitals could then leverage care team lists to facilitate patient‐centered communication via secure web‐based and mobile messaging applications configured to simultaneously update all team members (eg, group messaging apps, microblogs).[11, 37, 38] By synchronizing with the electronic health record, role assignments can be automatically updated via these applications, further increasing fidelity of care team identification.[12] Finally, as hospitals implement acute care patient portals, team lists can be leveraged to display all care team members correctly so that patients and caregivers can communicate more easily with providers.[17] The potential ramifications for patient‐centered communication are tremendous.

Patient‐centered communication is a strategy that is used to promote shared understanding of the plan of care among providers and patients.[1, 2, 3] Caring for hospitalized patients is a collaborative effort that requires seamless patient‐centered communication among a rapidly changing care team to safely progress a patient from admission through discharge. Yet, hospitals continue to struggle with improving the complex and increasingly electronic conversation patterns among care team members and patients to achieve effective patient‐centered communication.[4, 5] When members of the care team operate in this environment, patients often receive conflicting information regarding their plan of care, medications, and test results. Ineffective communication can lead to a suboptimal patient experience, additional costs, medical errors, and preventable adverse events.[6, 7, 8, 9, 10]

A critical first step to improving patient‐centered communication is identifying the care team.[11, 12] Accurate and reliable identification of all care team members is a pressing information need; it is fundamental to efficiently conveying information about the plan of care to those who know the patient the best, must make timely decisions, or will assume care once the patient leaves the hospital.[13] Furthermore, it has implications for engaging patients more meaningfully in their care.[14, 15, 16, 17] Ideally, the process of identifying an individual caring for the patient in a specific role is quickly and reliably determined from the electronic health record, the single source of truth where any provider can quickly identify other team members. This source of truth can be updated manually when individual members assign and remove themselves from the care team, or automatically when accessing the patient's record, writing a note, placing an order, or adding a patient to a coverage list. When providers correctly identify other team members in this way, hospital paging directories and secure messaging tools that link to the electronic health record become more effective at supporting care team communication.[18]

In general, the process of identifying care teams is difficult,[19] and maintaining role assignments in the electronic health record is equally challenging. Vawdrey et al. previously reported that care team lists are inaccurate and cannot be used to reliably identify other members at any given moment.[18] The inability to identify team members often leads to incorrectly routed pages, e‐mail messages, and phone calls.[20] Consequently, the potential to reliably manage the care team and improve electronic communication remains untapped, rendering team collaboration and care coordination less effective.[18, 21, 22]

In recent years, the trend toward restructuring inpatient teamsgeographical localization, structured communication interventions, teamwork training, and interdisciplinary roundswould seem to diminish the need for electronic care team identification, as those efforts have already made a positive impact with regard to interprofessional communication and collaboration, team satisfaction, and adverse events.[23, 24, 25, 26] Nonetheless, interdisciplinary teamwork, though critically important for patient‐centered communication, does not completely obviate the need for accurate and reliable care team identification.[26] Although care teams are statically located on units, the plan of care is dynamic; it evolves when the patient's status changes, when new information becomes available, and when key longitudinal providers (eg, primary care physician, subspecialty consultant) make recommendations. Thus, information conveyed as a team on rounds quickly becomes out of date, requiring additional forms of communication. Furthermore, due to frequent ad hoc coverage among team members, the identity of providers covering the patient at any given moment is often not clear.[27] This is particularly problematic for nonunit‐based providers who try to communicate with unit‐based care team members. These providers, in particular, have valuable knowledge and insight that can aid the primary team in decision making.[28, 29] However, they typically do not participate in rounds, often waste time identifying responsible providers,[20] and may communicate their recommendations directly with the patient without discussing with the primary team. These factors in part explain why geographic localization has shown limited improvement in shared understanding of the plan of care.[23]

From the perspective of patients and caregivers, identification of the providers entering and leaving their room is also challenging; only 11% to 51% of patients identify their providers correctly.[30] This adds to confusion regarding who is responsible for which aspects of the patient's care and can negatively affect the perception of the quality of care received.[31] Use of whiteboards has been shown to improve the proportion of patients who could identify key providers,[32] but these are not reliably updated and generally cannot accommodate all team members. When face cards are used, patients and caregivers report that they are more likely to identify their providers correctly.[14, 33, 34] However, potential confusion may ensue when another provider assumes care of the patient in the same role. Finally, use of technology to display team members at the bedside is typically a feature that patients like and can improve identification of care team members.[14, 15, 16] Yet, patient engagement technologies are not readily available in the hospital setting,[35] and ideally should be linked to the electronic health record, which again must be reliably updated.[11, 12, 15, 16]

If care team identification is so critical for delivering effective patient‐centered communication, why is maintaining role assignment problematic? At the individual level, reasons include discontinuity of the care team due to changing clinical rotations and intrateam coverage, shift‐based schedules, and lack of awareness and underutilization of functionality. Additionally, clinicians may have different ways to maintain lists of patients. At the institutional level, functionality to enforce role assignment when accessing patient records may be disabled (to avoid perceived burdens on clinical staff or nonclinical personnel who require access for administrative functions). Finally, electronic health record vendors currently have no incentive to adopt functionality that supports more effective care coordination across settings.[22]

However, more than technical solutions and policy changes are required; care team identification in the electronic health record requires a change in institutional culture. Maintaining an accurate relationship to each patient requires work without tangible benefitsthe benefits accrue only when everyone else identifies their role on the teama tragedy of the commons. This can be illustrated by our own experience. We conducted a quality improvement initiative (Table 1) as a part of 2 concurrent research initiatives that serve to promote patient‐centered communication:[12] PCORI (Patient‐Centered Outcomes Research Institute Transitions), the goal of which is to improve care transitions within the Partners' Pioneer Accountable Care Organization; and the PROSPECT (Promoting Respect and Ongoing Safety Through Patient‐Centeredness, Engagement, Communication, and Technology) project, an initiative funded by the Gordon and Betty Moore Foundation to eliminate preventable harms in the acute care environment.[29] Our goal was to electronically manage the care team with a high degree of fidelity. We enhanced a home‐grown application, which was developed to improve management of team lists for inpatient providers, accessible from our electronic health record, to facilitate role assignment. Specifically, we leveraged existing care processes (eg, nursing log‐on to the electronic medication administration system) to automatically assign certain providers to the care team at change of shift, added functionality to make it easy to assign a provider to all patients on a list for a defined period of time, and encouraged providers to assign their role by demonstrating benefits including quick access to patient‐specific group e‐mail and secure messaging tools (Table 1, Key Facets). The initiative was well‐received by most disciplines, but uptake was suboptimal. Our research assistants routinely assigned residents and others to the care team because our proactive attempts at advertising and reinforcing use of the application failed to reach a critical mass. Most did not see immediate benefits because it was an added step to their busy day, had other methods of managing team lists, and only saw benefit if everyone else participated. Key facets of our care team identification initiative, successes, and challenges are outlined in Table 1.

There were a few glimmers of hope, however. On several PROSPECT units, we displayed team members on a tablet‐based patient portal so that patients would recognize their providers.[11, 17, 36] Similar to recent work by O'Leary et al.,[14] patients on PROSPECT units were able to correctly identify several care team members, but regularly asked why other providers (eg, consulting fellow) were not listed. Those providers asked the same question, and some eventually learned to assign their role via the application. As part of PROSEPCT, we visited other institutions and learned of an effort to display team members on high‐definition televisions in the patient's room. Several providers, wondering why they were not listed, learned to assign their role and their picture then appeared. Social pressure was the driving force.

Coincidently, we recently implemented a new electronic health record at our institution. Anecdotally, although no formal policy was established, many providers (eg, attendings, first responders, nurses, care coordinators, and other unit‐based providers) appear to be assigning their roles. Other providers (eg, dieticians, physical therapists, residents) also assign their role, but often fail to end role assignments upon completing their rotation or when the patient transfers to another service. Finally, even when actively involved, most subspecialists still do not designate their role. Despite these gaps and inconsistencies, we have made progress toward improving care team identification. The reasons for this progress are straightforward; during required training for the new electronic health record, all inpatient providers were taught to assign their role on the treatment team when assuming care of patients and now have 1 option for managing team lists. However, most providers were not trained to end their role assignments, and many have learned that role assignment is not required to access the patient's record; functionality to enforce this was disabled. Based on lessons learned from our experience,[12] we offer several strategies that hospitalists can employ to improve care team identification in the electronic health record (Table 2).

In the future, care team identification in the electronic health record can be automated by integrating directly with electronic workflows, online scheduling applications, and provider directories. Hospitals could then leverage care team lists to facilitate patient‐centered communication via secure web‐based and mobile messaging applications configured to simultaneously update all team members (eg, group messaging apps, microblogs).[11, 37, 38] By synchronizing with the electronic health record, role assignments can be automatically updated via these applications, further increasing fidelity of care team identification.[12] Finally, as hospitals implement acute care patient portals, team lists can be leveraged to display all care team members correctly so that patients and caregivers can communicate more easily with providers.[17] The potential ramifications for patient‐centered communication are tremendous.

Patient‐centered communication is a strategy that is used to promote shared understanding of the plan of care among providers and patients.[1, 2, 3] Caring for hospitalized patients is a collaborative effort that requires seamless patient‐centered communication among a rapidly changing care team to safely progress a patient from admission through discharge. Yet, hospitals continue to struggle with improving the complex and increasingly electronic conversation patterns among care team members and patients to achieve effective patient‐centered communication.[4, 5] When members of the care team operate in this environment, patients often receive conflicting information regarding their plan of care, medications, and test results. Ineffective communication can lead to a suboptimal patient experience, additional costs, medical errors, and preventable adverse events.[6, 7, 8, 9, 10]

A critical first step to improving patient‐centered communication is identifying the care team.[11, 12] Accurate and reliable identification of all care team members is a pressing information need; it is fundamental to efficiently conveying information about the plan of care to those who know the patient the best, must make timely decisions, or will assume care once the patient leaves the hospital.[13] Furthermore, it has implications for engaging patients more meaningfully in their care.[14, 15, 16, 17] Ideally, the process of identifying an individual caring for the patient in a specific role is quickly and reliably determined from the electronic health record, the single source of truth where any provider can quickly identify other team members. This source of truth can be updated manually when individual members assign and remove themselves from the care team, or automatically when accessing the patient's record, writing a note, placing an order, or adding a patient to a coverage list. When providers correctly identify other team members in this way, hospital paging directories and secure messaging tools that link to the electronic health record become more effective at supporting care team communication.[18]

In general, the process of identifying care teams is difficult,[19] and maintaining role assignments in the electronic health record is equally challenging. Vawdrey et al. previously reported that care team lists are inaccurate and cannot be used to reliably identify other members at any given moment.[18] The inability to identify team members often leads to incorrectly routed pages, e‐mail messages, and phone calls.[20] Consequently, the potential to reliably manage the care team and improve electronic communication remains untapped, rendering team collaboration and care coordination less effective.[18, 21, 22]

In recent years, the trend toward restructuring inpatient teamsgeographical localization, structured communication interventions, teamwork training, and interdisciplinary roundswould seem to diminish the need for electronic care team identification, as those efforts have already made a positive impact with regard to interprofessional communication and collaboration, team satisfaction, and adverse events.[23, 24, 25, 26] Nonetheless, interdisciplinary teamwork, though critically important for patient‐centered communication, does not completely obviate the need for accurate and reliable care team identification.[26] Although care teams are statically located on units, the plan of care is dynamic; it evolves when the patient's status changes, when new information becomes available, and when key longitudinal providers (eg, primary care physician, subspecialty consultant) make recommendations. Thus, information conveyed as a team on rounds quickly becomes out of date, requiring additional forms of communication. Furthermore, due to frequent ad hoc coverage among team members, the identity of providers covering the patient at any given moment is often not clear.[27] This is particularly problematic for nonunit‐based providers who try to communicate with unit‐based care team members. These providers, in particular, have valuable knowledge and insight that can aid the primary team in decision making.[28, 29] However, they typically do not participate in rounds, often waste time identifying responsible providers,[20] and may communicate their recommendations directly with the patient without discussing with the primary team. These factors in part explain why geographic localization has shown limited improvement in shared understanding of the plan of care.[23]

From the perspective of patients and caregivers, identification of the providers entering and leaving their room is also challenging; only 11% to 51% of patients identify their providers correctly.[30] This adds to confusion regarding who is responsible for which aspects of the patient's care and can negatively affect the perception of the quality of care received.[31] Use of whiteboards has been shown to improve the proportion of patients who could identify key providers,[32] but these are not reliably updated and generally cannot accommodate all team members. When face cards are used, patients and caregivers report that they are more likely to identify their providers correctly.[14, 33, 34] However, potential confusion may ensue when another provider assumes care of the patient in the same role. Finally, use of technology to display team members at the bedside is typically a feature that patients like and can improve identification of care team members.[14, 15, 16] Yet, patient engagement technologies are not readily available in the hospital setting,[35] and ideally should be linked to the electronic health record, which again must be reliably updated.[11, 12, 15, 16]

If care team identification is so critical for delivering effective patient‐centered communication, why is maintaining role assignment problematic? At the individual level, reasons include discontinuity of the care team due to changing clinical rotations and intrateam coverage, shift‐based schedules, and lack of awareness and underutilization of functionality. Additionally, clinicians may have different ways to maintain lists of patients. At the institutional level, functionality to enforce role assignment when accessing patient records may be disabled (to avoid perceived burdens on clinical staff or nonclinical personnel who require access for administrative functions). Finally, electronic health record vendors currently have no incentive to adopt functionality that supports more effective care coordination across settings.[22]

However, more than technical solutions and policy changes are required; care team identification in the electronic health record requires a change in institutional culture. Maintaining an accurate relationship to each patient requires work without tangible benefitsthe benefits accrue only when everyone else identifies their role on the teama tragedy of the commons. This can be illustrated by our own experience. We conducted a quality improvement initiative (Table 1) as a part of 2 concurrent research initiatives that serve to promote patient‐centered communication:[12] PCORI (Patient‐Centered Outcomes Research Institute Transitions), the goal of which is to improve care transitions within the Partners' Pioneer Accountable Care Organization; and the PROSPECT (Promoting Respect and Ongoing Safety Through Patient‐Centeredness, Engagement, Communication, and Technology) project, an initiative funded by the Gordon and Betty Moore Foundation to eliminate preventable harms in the acute care environment.[29] Our goal was to electronically manage the care team with a high degree of fidelity. We enhanced a home‐grown application, which was developed to improve management of team lists for inpatient providers, accessible from our electronic health record, to facilitate role assignment. Specifically, we leveraged existing care processes (eg, nursing log‐on to the electronic medication administration system) to automatically assign certain providers to the care team at change of shift, added functionality to make it easy to assign a provider to all patients on a list for a defined period of time, and encouraged providers to assign their role by demonstrating benefits including quick access to patient‐specific group e‐mail and secure messaging tools (Table 1, Key Facets). The initiative was well‐received by most disciplines, but uptake was suboptimal. Our research assistants routinely assigned residents and others to the care team because our proactive attempts at advertising and reinforcing use of the application failed to reach a critical mass. Most did not see immediate benefits because it was an added step to their busy day, had other methods of managing team lists, and only saw benefit if everyone else participated. Key facets of our care team identification initiative, successes, and challenges are outlined in Table 1.

There were a few glimmers of hope, however. On several PROSPECT units, we displayed team members on a tablet‐based patient portal so that patients would recognize their providers.[11, 17, 36] Similar to recent work by O'Leary et al.,[14] patients on PROSPECT units were able to correctly identify several care team members, but regularly asked why other providers (eg, consulting fellow) were not listed. Those providers asked the same question, and some eventually learned to assign their role via the application. As part of PROSEPCT, we visited other institutions and learned of an effort to display team members on high‐definition televisions in the patient's room. Several providers, wondering why they were not listed, learned to assign their role and their picture then appeared. Social pressure was the driving force.

Coincidently, we recently implemented a new electronic health record at our institution. Anecdotally, although no formal policy was established, many providers (eg, attendings, first responders, nurses, care coordinators, and other unit‐based providers) appear to be assigning their roles. Other providers (eg, dieticians, physical therapists, residents) also assign their role, but often fail to end role assignments upon completing their rotation or when the patient transfers to another service. Finally, even when actively involved, most subspecialists still do not designate their role. Despite these gaps and inconsistencies, we have made progress toward improving care team identification. The reasons for this progress are straightforward; during required training for the new electronic health record, all inpatient providers were taught to assign their role on the treatment team when assuming care of patients and now have 1 option for managing team lists. However, most providers were not trained to end their role assignments, and many have learned that role assignment is not required to access the patient's record; functionality to enforce this was disabled. Based on lessons learned from our experience,[12] we offer several strategies that hospitalists can employ to improve care team identification in the electronic health record (Table 2).

In the future, care team identification in the electronic health record can be automated by integrating directly with electronic workflows, online scheduling applications, and provider directories. Hospitals could then leverage care team lists to facilitate patient‐centered communication via secure web‐based and mobile messaging applications configured to simultaneously update all team members (eg, group messaging apps, microblogs).[11, 37, 38] By synchronizing with the electronic health record, role assignments can be automatically updated via these applications, further increasing fidelity of care team identification.[12] Finally, as hospitals implement acute care patient portals, team lists can be leveraged to display all care team members correctly so that patients and caregivers can communicate more easily with providers.[17] The potential ramifications for patient‐centered communication are tremendous.

Vascular access devices (VADs), including peripherally inserted central venous catheters (PICCs) and traditional central venous catheters (CVCs), remain a cornerstone for the delivery of necessary therapy. VADs are used routinely to treat inpatients and increasingly outpatients too. PICCs possess characteristics that are often favorable in a variety of clinical settings when compared to traditional CVCs. However, a paucity of evidence regarding the indication, selection, application, duration, and risks associated with these devices exists. PICCs are often used in situations when peripheral venous catheters (PIVsincluding ultrasound‐guided peripheral intravenous catheters and midline catheters [midlines]) would meet patient needs and confer a lower risk of complications. An unmet need to define indications and promote utilization that conforms to optimal use currently exists. The purpose of this article was to highlight for hospitalists the methodology and subsequent key recommendations published recently[1] regarding appropriateness of PICCs as they pertain to other vascular access device use.

BACKGROUND

Greater utilization of PICCs to meet a variety of clinical needs has recently emerged in hospital‐based medicine.[2, 3] This phenomenon is likely a function of favorable characteristics when comparing PICCs with traditional CVCs. PICCs are often favored because of safety with insertion in the arm, compatibility with inpatient and outpatient therapies, ease of protocolization for insertion by vascular access nursing services, patient tolerability, and cost savings.[4, 5, 6, 7, 8] Yet limitations of PICCs exist and complications including malpositioning, dislodgement, and luminal occlusion[9, 10, 11] affect patient safety and outcomes. Most notably, PICCs are strongly associated with risk for thrombosis and infection, complications that are most frequent in hospitalized and critically ill patients.[12, 13, 14, 15, 16]

Vascular access devices and particularly PICCs pose a substantial risk for thrombosis.[16, 17, 18, 19, 20] PICCs represent the greatest risk factor for upper extremity deep vein thrombosis (DVT), and in one study, PICC‐associated DVT risk was double that with traditional CVCs.[17] Risk factors for the development of PICC‐associated DVT include ipsilateral paresis,[21] infection,[22] PICC diameter,[19, 20] and prolonged surgery (procedure duration >1 hour) with a PICC in place.[23] Recently, PICCs placed in the upper extremity have been described as a possible risk factor for lower extremity venous thrombosis as well.[24, 25]

Infection complicating CVCs is well described,[12, 15] and guidelines for the prevention of catheter‐associated blood stream infections exist.[26, 27] However, the magnitude of the risk of infection associated with PICCs compared with traditional CVCs remains uncertain. Some reports suggest a decrease risk for infection with the utilization of PICCs[28]; others suggest a similar risk.[29] Existing guidelines, however, do not recommend substituting PICCs for CVCs as a technique to reduce infection, especially in general medical patients.[30]

It is not surprising that variability in the clinical use of PICCs and inappropriate PICC utilization has been described[31, 32] given the heterogeneity of patients and clinical situations in which PICCs are used. Simple awareness of medical devices in place is central to optimizing care. Important to the hospitalist physician is a recent study that found that 1 in 5 physicians were unaware of a CVC being present in their patient.[33] Indeed, emphasis has been placed on optimizing the use of PICC lines nationally through the Choosing Wisely initiative.[34, 35]

A panel of experts was convened at the University of Michigan in an effort to further clarify the appropriate use of VADs. Panelists engaged in a RAND Corporation/University of California Los Angeles (RAND/UCLA) Appropriateness Methodology review[36] to provide guidance regarding VAD use. The RAND/UCLA methodology is a validated way to assess the appropriateness of medical and surgical resource utilization, and details of this methodology are published elsewhere.[1] In brief, each panelist was provided a series of clinical scenarios associated with the use of central venous catheters purposefully including areas of consensus, controversy, and ambiguity. Using a standardized method for rating appropriateness, whereby median ratings on opposite ends of a 1 to 9 scale were used to indicate preference of one device over another (for example 9 reflected appropriate and 13 reflected inappropriate), the methodology classified consensus results into three levels of appropriateness. These three levels are: appropriate when the panel median is between 7 and 9 and without disagreement, uncertain/neutral when the panel median is between 4 and 6 or disagreement exists regardless of the median, or inappropriate when the panel median is between 1 and 3 without disagreement.

RESULTS

Comprehensive results regarding appropriateness ratings are reported elsewhere.[1] Results especially key to hospital‐based practitioners are summarized below. Table 1 highlights common scenarios when PICC placement is considered appropriate and inappropriate.

CONCLUSIONS

In‐hospital healthcare providers are routinely confronted with dilemmas surrounding choice of VAD. The Michigan Appropriateness Guide for Intravenous Catheters (MAGIC) initiative is a multidisciplinary effort to clarify decision‐making related to VAD use. The systematic literature review and RAND/UCLA appropriateness method applied by the MAGIC panelists identifies areas of broad consensus surrounding the use of PICCs in relation to other VADs, and highlights uncertainties regarding the best practice to guide clinical care. Appropriateness statements facilitate standardization for the use, care, and discontinuation of VADs. These recommendations may be important to healthcare quality officers and payers as they allow for measurement of, and adherence to, standardized practice. In an era of electronic medical records and embedded clinical decision support, these recommendations may facilitate a just‐in‐time resource for optimal VAD management, outcomes measurement, and patient follow‐up. In addition to directing clinical care, these recommendations may serve as a lattice for the formation of future randomized clinical trials to further clarify important areas of the uncertainty surrounding VAD use.

Disclosures: Drs. Woller and Stevens disclose financial support paid to their institution of employment (Intermountain Medical Center) for conducting clinical research (with no financial support paid to either investigator). Dr. Woller discloses serving as an expert panelist for the Michigan Appropriateness Guide for Intravenous Catheters (MAGIC) initiative. The authors report no other conflicts of interest.

Vascular access devices (VADs), including peripherally inserted central venous catheters (PICCs) and traditional central venous catheters (CVCs), remain a cornerstone for the delivery of necessary therapy. VADs are used routinely to treat inpatients and increasingly outpatients too. PICCs possess characteristics that are often favorable in a variety of clinical settings when compared to traditional CVCs. However, a paucity of evidence regarding the indication, selection, application, duration, and risks associated with these devices exists. PICCs are often used in situations when peripheral venous catheters (PIVsincluding ultrasound‐guided peripheral intravenous catheters and midline catheters [midlines]) would meet patient needs and confer a lower risk of complications. An unmet need to define indications and promote utilization that conforms to optimal use currently exists. The purpose of this article was to highlight for hospitalists the methodology and subsequent key recommendations published recently[1] regarding appropriateness of PICCs as they pertain to other vascular access device use.

BACKGROUND

Greater utilization of PICCs to meet a variety of clinical needs has recently emerged in hospital‐based medicine.[2, 3] This phenomenon is likely a function of favorable characteristics when comparing PICCs with traditional CVCs. PICCs are often favored because of safety with insertion in the arm, compatibility with inpatient and outpatient therapies, ease of protocolization for insertion by vascular access nursing services, patient tolerability, and cost savings.[4, 5, 6, 7, 8] Yet limitations of PICCs exist and complications including malpositioning, dislodgement, and luminal occlusion[9, 10, 11] affect patient safety and outcomes. Most notably, PICCs are strongly associated with risk for thrombosis and infection, complications that are most frequent in hospitalized and critically ill patients.[12, 13, 14, 15, 16]

Vascular access devices and particularly PICCs pose a substantial risk for thrombosis.[16, 17, 18, 19, 20] PICCs represent the greatest risk factor for upper extremity deep vein thrombosis (DVT), and in one study, PICC‐associated DVT risk was double that with traditional CVCs.[17] Risk factors for the development of PICC‐associated DVT include ipsilateral paresis,[21] infection,[22] PICC diameter,[19, 20] and prolonged surgery (procedure duration >1 hour) with a PICC in place.[23] Recently, PICCs placed in the upper extremity have been described as a possible risk factor for lower extremity venous thrombosis as well.[24, 25]

Infection complicating CVCs is well described,[12, 15] and guidelines for the prevention of catheter‐associated blood stream infections exist.[26, 27] However, the magnitude of the risk of infection associated with PICCs compared with traditional CVCs remains uncertain. Some reports suggest a decrease risk for infection with the utilization of PICCs[28]; others suggest a similar risk.[29] Existing guidelines, however, do not recommend substituting PICCs for CVCs as a technique to reduce infection, especially in general medical patients.[30]

It is not surprising that variability in the clinical use of PICCs and inappropriate PICC utilization has been described[31, 32] given the heterogeneity of patients and clinical situations in which PICCs are used. Simple awareness of medical devices in place is central to optimizing care. Important to the hospitalist physician is a recent study that found that 1 in 5 physicians were unaware of a CVC being present in their patient.[33] Indeed, emphasis has been placed on optimizing the use of PICC lines nationally through the Choosing Wisely initiative.[34, 35]

A panel of experts was convened at the University of Michigan in an effort to further clarify the appropriate use of VADs. Panelists engaged in a RAND Corporation/University of California Los Angeles (RAND/UCLA) Appropriateness Methodology review[36] to provide guidance regarding VAD use. The RAND/UCLA methodology is a validated way to assess the appropriateness of medical and surgical resource utilization, and details of this methodology are published elsewhere.[1] In brief, each panelist was provided a series of clinical scenarios associated with the use of central venous catheters purposefully including areas of consensus, controversy, and ambiguity. Using a standardized method for rating appropriateness, whereby median ratings on opposite ends of a 1 to 9 scale were used to indicate preference of one device over another (for example 9 reflected appropriate and 13 reflected inappropriate), the methodology classified consensus results into three levels of appropriateness. These three levels are: appropriate when the panel median is between 7 and 9 and without disagreement, uncertain/neutral when the panel median is between 4 and 6 or disagreement exists regardless of the median, or inappropriate when the panel median is between 1 and 3 without disagreement.

RESULTS

Comprehensive results regarding appropriateness ratings are reported elsewhere.[1] Results especially key to hospital‐based practitioners are summarized below. Table 1 highlights common scenarios when PICC placement is considered appropriate and inappropriate.

CONCLUSIONS

In‐hospital healthcare providers are routinely confronted with dilemmas surrounding choice of VAD. The Michigan Appropriateness Guide for Intravenous Catheters (MAGIC) initiative is a multidisciplinary effort to clarify decision‐making related to VAD use. The systematic literature review and RAND/UCLA appropriateness method applied by the MAGIC panelists identifies areas of broad consensus surrounding the use of PICCs in relation to other VADs, and highlights uncertainties regarding the best practice to guide clinical care. Appropriateness statements facilitate standardization for the use, care, and discontinuation of VADs. These recommendations may be important to healthcare quality officers and payers as they allow for measurement of, and adherence to, standardized practice. In an era of electronic medical records and embedded clinical decision support, these recommendations may facilitate a just‐in‐time resource for optimal VAD management, outcomes measurement, and patient follow‐up. In addition to directing clinical care, these recommendations may serve as a lattice for the formation of future randomized clinical trials to further clarify important areas of the uncertainty surrounding VAD use.

Disclosures: Drs. Woller and Stevens disclose financial support paid to their institution of employment (Intermountain Medical Center) for conducting clinical research (with no financial support paid to either investigator). Dr. Woller discloses serving as an expert panelist for the Michigan Appropriateness Guide for Intravenous Catheters (MAGIC) initiative. The authors report no other conflicts of interest.

Vascular access devices (VADs), including peripherally inserted central venous catheters (PICCs) and traditional central venous catheters (CVCs), remain a cornerstone for the delivery of necessary therapy. VADs are used routinely to treat inpatients and increasingly outpatients too. PICCs possess characteristics that are often favorable in a variety of clinical settings when compared to traditional CVCs. However, a paucity of evidence regarding the indication, selection, application, duration, and risks associated with these devices exists. PICCs are often used in situations when peripheral venous catheters (PIVsincluding ultrasound‐guided peripheral intravenous catheters and midline catheters [midlines]) would meet patient needs and confer a lower risk of complications. An unmet need to define indications and promote utilization that conforms to optimal use currently exists. The purpose of this article was to highlight for hospitalists the methodology and subsequent key recommendations published recently[1] regarding appropriateness of PICCs as they pertain to other vascular access device use.

BACKGROUND

Greater utilization of PICCs to meet a variety of clinical needs has recently emerged in hospital‐based medicine.[2, 3] This phenomenon is likely a function of favorable characteristics when comparing PICCs with traditional CVCs. PICCs are often favored because of safety with insertion in the arm, compatibility with inpatient and outpatient therapies, ease of protocolization for insertion by vascular access nursing services, patient tolerability, and cost savings.[4, 5, 6, 7, 8] Yet limitations of PICCs exist and complications including malpositioning, dislodgement, and luminal occlusion[9, 10, 11] affect patient safety and outcomes. Most notably, PICCs are strongly associated with risk for thrombosis and infection, complications that are most frequent in hospitalized and critically ill patients.[12, 13, 14, 15, 16]

Vascular access devices and particularly PICCs pose a substantial risk for thrombosis.[16, 17, 18, 19, 20] PICCs represent the greatest risk factor for upper extremity deep vein thrombosis (DVT), and in one study, PICC‐associated DVT risk was double that with traditional CVCs.[17] Risk factors for the development of PICC‐associated DVT include ipsilateral paresis,[21] infection,[22] PICC diameter,[19, 20] and prolonged surgery (procedure duration >1 hour) with a PICC in place.[23] Recently, PICCs placed in the upper extremity have been described as a possible risk factor for lower extremity venous thrombosis as well.[24, 25]

Infection complicating CVCs is well described,[12, 15] and guidelines for the prevention of catheter‐associated blood stream infections exist.[26, 27] However, the magnitude of the risk of infection associated with PICCs compared with traditional CVCs remains uncertain. Some reports suggest a decrease risk for infection with the utilization of PICCs[28]; others suggest a similar risk.[29] Existing guidelines, however, do not recommend substituting PICCs for CVCs as a technique to reduce infection, especially in general medical patients.[30]

It is not surprising that variability in the clinical use of PICCs and inappropriate PICC utilization has been described[31, 32] given the heterogeneity of patients and clinical situations in which PICCs are used. Simple awareness of medical devices in place is central to optimizing care. Important to the hospitalist physician is a recent study that found that 1 in 5 physicians were unaware of a CVC being present in their patient.[33] Indeed, emphasis has been placed on optimizing the use of PICC lines nationally through the Choosing Wisely initiative.[34, 35]

A panel of experts was convened at the University of Michigan in an effort to further clarify the appropriate use of VADs. Panelists engaged in a RAND Corporation/University of California Los Angeles (RAND/UCLA) Appropriateness Methodology review[36] to provide guidance regarding VAD use. The RAND/UCLA methodology is a validated way to assess the appropriateness of medical and surgical resource utilization, and details of this methodology are published elsewhere.[1] In brief, each panelist was provided a series of clinical scenarios associated with the use of central venous catheters purposefully including areas of consensus, controversy, and ambiguity. Using a standardized method for rating appropriateness, whereby median ratings on opposite ends of a 1 to 9 scale were used to indicate preference of one device over another (for example 9 reflected appropriate and 13 reflected inappropriate), the methodology classified consensus results into three levels of appropriateness. These three levels are: appropriate when the panel median is between 7 and 9 and without disagreement, uncertain/neutral when the panel median is between 4 and 6 or disagreement exists regardless of the median, or inappropriate when the panel median is between 1 and 3 without disagreement.

RESULTS

Comprehensive results regarding appropriateness ratings are reported elsewhere.[1] Results especially key to hospital‐based practitioners are summarized below. Table 1 highlights common scenarios when PICC placement is considered appropriate and inappropriate.

CONCLUSIONS

In‐hospital healthcare providers are routinely confronted with dilemmas surrounding choice of VAD. The Michigan Appropriateness Guide for Intravenous Catheters (MAGIC) initiative is a multidisciplinary effort to clarify decision‐making related to VAD use. The systematic literature review and RAND/UCLA appropriateness method applied by the MAGIC panelists identifies areas of broad consensus surrounding the use of PICCs in relation to other VADs, and highlights uncertainties regarding the best practice to guide clinical care. Appropriateness statements facilitate standardization for the use, care, and discontinuation of VADs. These recommendations may be important to healthcare quality officers and payers as they allow for measurement of, and adherence to, standardized practice. In an era of electronic medical records and embedded clinical decision support, these recommendations may facilitate a just‐in‐time resource for optimal VAD management, outcomes measurement, and patient follow‐up. In addition to directing clinical care, these recommendations may serve as a lattice for the formation of future randomized clinical trials to further clarify important areas of the uncertainty surrounding VAD use.

Disclosures: Drs. Woller and Stevens disclose financial support paid to their institution of employment (Intermountain Medical Center) for conducting clinical research (with no financial support paid to either investigator). Dr. Woller discloses serving as an expert panelist for the Michigan Appropriateness Guide for Intravenous Catheters (MAGIC) initiative. The authors report no other conflicts of interest.

Increasing use of emergency departments (EDs) throughout the United States has become a focus of national healthcare policy and reform efforts. ED growth continues to outpace population growth, with the Institute of Medicine describing our ED systems as fragmented, overburdened, and at the breaking point.[1] Associations between ED crowding and patient dissatisfaction, delays in treatment, medical errors, and patient mortality speak to the urgency of systems improvements.[2] One major factor contributing to ED volumes is the growing number of hospital admissions that begin in EDs. From 1993 to 2006, the proportion of hospitalizations originating in EDs increased from 33.5% to 43.8%, with more than 17 million hospital admissions originating in EDs annually.[3, 4] Despite these challenges, discussions about alternative approaches to hospital admission remain at the periphery of healthcare policy conversations.

Direct admission to the hospital, defined as hospitalization without first receiving care in the hospital's ED, is an alternative approach to hospital admission, and may be a vehicle to both observation and inpatient hospital stays. Direct admissions account for 25% of all nonelective pediatric hospitalizations and 15% of nonelective adult hospitalizations in the United States.[5, 6] This admission approach was considerably more common in the past, facilitated by primary care providers (PCPs) or specialists who provided both outpatient and hospital‐based care for their patients.[4] However, as the number of hospitalists in the United States has grown over the last 30 years, the number of direct admissions has decreased concurrently. In fact, from 2003 to 2009, the number of direct admissions from clinics and physicians' offices decreased by a total of 1.6 million.[4] Although this decline is undoubtedly multifactorial, hospitalists may have contributed, both deliberately and inadvertently, to the shifting epidemiology of hospital admissions. Although many factors influence the source of hospital admissions and admission processes, direct admission has 2 important prerequisites: patients require timely access to outpatient providers for acute care, and hospitals, in partnership with outpatient‐based clinics and practices, require systems to safely and efficiently facilitate admissions without ED involvement. However, we know little about hospital admission systems, developed in the era of hospital medicine, to facilitate admissions independent of the ED.

Direct admission offers a number of potential benefits for both patients and healthcare delivery systems including reductions in the number of sites and providers of care, improved communication and coordination between outpatient and hospital‐based healthcare providers, greater patient and referring physician satisfaction, and reduced ED volumes and subsequent costs.[7] However, there are also risks and potential harms associated with direct admission, including potential delays in initial evaluation and management, inconsistent admission processes, and difficulties determining direct admission appropriateness, all of which could adversely impact patient safety and quality of care.[7, 8, 9] One study of adults with sepsis found that direct admission was associated with increased mortality compared to ED admission, which the authors speculated to be related to less timely care.[9] Similarly, a study of unscheduled adult hospitalizations found that patients admitted directly had higher mortality for time‐sensitive conditions such as acute myocardial infarction and sepsis than patients admitted through EDs, differences not observed among adults admitted with pneumonia, asthma, cellulitis, and several other common, yet frequently less emergent, reasons for hospitalization.[8] Among children with pneumonia, the most common reason for pediatric hospitalization, direct admission has been associated with significantly lower costs than admissions originating in the ED, with no significant differences in rates of transfer to the intensive care unit or hospital readmission.[10]

There is significant variation across both diagnoses and hospitals in rates of direct admission, raising questions about the contextual factors unique to hospital medicine programs that perform a substantial proportion of direct admissions.[5] This variation also highlights opportunities to identify the populations, conditions, and systems that facilitate safe and effective direct admissions. Certainly, direct admission is unlikely to be appropriate for all populations or conditions. Patients requiring emergent care or rapid diagnostic imaging are likely to receive more timely care in the ED; sepsis, acute myocardial infarction, and trauma are but a few examples of conditions for which rapid ED care decreases morbidity and mortality. Similarly, patients for whom the need for hospitalization is uncertainfor example, dehydration, asthmamay be more appropriate for initial ED management followed by re‐evaluation to inform the need for hospitalization. Finally, patients for whom the admitting diagnosis is uncertain and who require consultation for several subspecialists may be more efficiently evaluated in EDs. In our national survey of pediatric direct admission guidelines, less than one‐third of hospitals reported having formal criteria to assess the appropriateness of direct admissions, and respondents' perspectives regarding populations and diagnoses appropriate for this admission approach varied considerably.[7] These results point to the need for further research and quality‐improvement initiatives to inform the development of direct admission guidelines and protocols.

During the last decade, hospitals' discharge processes have been the focus of tremendous research, policy, and quality improvement efforts. The phrase transition of care is now widely understood to describe the changes in patient care that begin with discharge planning and conclude when patients' have established care at home or another healthcare facility. Transitions of care have been a focus of the Journal of Hospital Medicine since its inception, including publication of the Transitions of Care Consensus Policy Statement in 2009, as well as numerous other studies highlighting both risks associated with transitions of care as well as methods to address these.[11, 12, 13, 14, 15, 16] Similar to hospital discharge, hospital admission is an inherent feature of every hospitalization, and admission and discharge processes share many commonalities. Both involve transitions in sites of care, and handoffs between healthcare providers. Most involve changes in medical therapies, including both the addition of new medications and changes to existing treatments. Moreover, both are associated with significant stress to patients and their families. As a result, hospital admissions expose patients to many of same risks that have been the focus of hospital discharge reform: unstructured patient handoffs, poor communication between healthcare providers, and costly, inefficient care. The Society of Hospital Medicine has been a leader in articulating the importance of patient‐centered, clinically relevant medication reconciliation across the healthcare continuum.[17] However, with the exception of this important work, research and policy focused on understanding and improving transitions of care into the hospital have received disproportionately little attention.

To facilitate research and quality improvement efforts focused on hospital admission, we suggest that the transitions of care framework, typically discussed in the context of hospital discharge, be expanded to reflect the different origins of hospitalizations and multiple transitions that can be experienced by patients as they enter the hospital. A broadening of the transitions of care framework to incorporate hospital admissions brings numerous questions previously addressed in hospital‐to‐home transitions to the forefront. How do transitions into the hospital impact patients and healthcare systems? When is direct admission safe and effective, and how does this vary across conditions and hospital settings? What protocols and tools might optimize the associated transitions and reduce the risks of error and harm? There are numerous stakeholders who will undoubtedly bring diverse perspectives to these questionspatients and their families, hospital‐based healthcare providers, PCPs and specialists, ED physicians, and payors.

Increasing ED volumes, long wait times, and rising ED costs speak to the importance of better understanding hospital admission alternatives and the associated risks and benefits. Encouraging more direct admissions may be a solution, but evidence to guide best practices must precede this. The growing presence of round‐the‐clock pediatric and adult hospitalists across the country creates unique opportunities to transform hospital admission systems for the vast number of patients who do not require emergent care. The Affordable Care Act's expansion of insurance coverage and incentivized coordinated care within patient‐centered medical homes creates a unique opportunity for this broadened view of transitions of care. This suggests that the time is ripe for pursuing strategies that will both improve patients' transitions from outpatient to inpatient care and reduce stress on our overburdened emergency departments.

Disclosure: Dr. Lagu was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under award number K01HL114745. She has received consulting fees from the Institute for Healthcare Improvement, under contract to the Centers for Medicare and Medicaid Services (CMS), for her work on a project to help health systems achieve disability competence, and from the Island Peer Review Organization, under contract to CMS, for her work on development of episodes of care for CMS payment purposes (both unrelated to the current work). Dr. Leyenaar was supported by grant number K08HS024133 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality. The authors report no conflicts of interest.

Increasing use of emergency departments (EDs) throughout the United States has become a focus of national healthcare policy and reform efforts. ED growth continues to outpace population growth, with the Institute of Medicine describing our ED systems as fragmented, overburdened, and at the breaking point.[1] Associations between ED crowding and patient dissatisfaction, delays in treatment, medical errors, and patient mortality speak to the urgency of systems improvements.[2] One major factor contributing to ED volumes is the growing number of hospital admissions that begin in EDs. From 1993 to 2006, the proportion of hospitalizations originating in EDs increased from 33.5% to 43.8%, with more than 17 million hospital admissions originating in EDs annually.[3, 4] Despite these challenges, discussions about alternative approaches to hospital admission remain at the periphery of healthcare policy conversations.

Direct admission to the hospital, defined as hospitalization without first receiving care in the hospital's ED, is an alternative approach to hospital admission, and may be a vehicle to both observation and inpatient hospital stays. Direct admissions account for 25% of all nonelective pediatric hospitalizations and 15% of nonelective adult hospitalizations in the United States.[5, 6] This admission approach was considerably more common in the past, facilitated by primary care providers (PCPs) or specialists who provided both outpatient and hospital‐based care for their patients.[4] However, as the number of hospitalists in the United States has grown over the last 30 years, the number of direct admissions has decreased concurrently. In fact, from 2003 to 2009, the number of direct admissions from clinics and physicians' offices decreased by a total of 1.6 million.[4] Although this decline is undoubtedly multifactorial, hospitalists may have contributed, both deliberately and inadvertently, to the shifting epidemiology of hospital admissions. Although many factors influence the source of hospital admissions and admission processes, direct admission has 2 important prerequisites: patients require timely access to outpatient providers for acute care, and hospitals, in partnership with outpatient‐based clinics and practices, require systems to safely and efficiently facilitate admissions without ED involvement. However, we know little about hospital admission systems, developed in the era of hospital medicine, to facilitate admissions independent of the ED.

Direct admission offers a number of potential benefits for both patients and healthcare delivery systems including reductions in the number of sites and providers of care, improved communication and coordination between outpatient and hospital‐based healthcare providers, greater patient and referring physician satisfaction, and reduced ED volumes and subsequent costs.[7] However, there are also risks and potential harms associated with direct admission, including potential delays in initial evaluation and management, inconsistent admission processes, and difficulties determining direct admission appropriateness, all of which could adversely impact patient safety and quality of care.[7, 8, 9] One study of adults with sepsis found that direct admission was associated with increased mortality compared to ED admission, which the authors speculated to be related to less timely care.[9] Similarly, a study of unscheduled adult hospitalizations found that patients admitted directly had higher mortality for time‐sensitive conditions such as acute myocardial infarction and sepsis than patients admitted through EDs, differences not observed among adults admitted with pneumonia, asthma, cellulitis, and several other common, yet frequently less emergent, reasons for hospitalization.[8] Among children with pneumonia, the most common reason for pediatric hospitalization, direct admission has been associated with significantly lower costs than admissions originating in the ED, with no significant differences in rates of transfer to the intensive care unit or hospital readmission.[10]

There is significant variation across both diagnoses and hospitals in rates of direct admission, raising questions about the contextual factors unique to hospital medicine programs that perform a substantial proportion of direct admissions.[5] This variation also highlights opportunities to identify the populations, conditions, and systems that facilitate safe and effective direct admissions. Certainly, direct admission is unlikely to be appropriate for all populations or conditions. Patients requiring emergent care or rapid diagnostic imaging are likely to receive more timely care in the ED; sepsis, acute myocardial infarction, and trauma are but a few examples of conditions for which rapid ED care decreases morbidity and mortality. Similarly, patients for whom the need for hospitalization is uncertainfor example, dehydration, asthmamay be more appropriate for initial ED management followed by re‐evaluation to inform the need for hospitalization. Finally, patients for whom the admitting diagnosis is uncertain and who require consultation for several subspecialists may be more efficiently evaluated in EDs. In our national survey of pediatric direct admission guidelines, less than one‐third of hospitals reported having formal criteria to assess the appropriateness of direct admissions, and respondents' perspectives regarding populations and diagnoses appropriate for this admission approach varied considerably.[7] These results point to the need for further research and quality‐improvement initiatives to inform the development of direct admission guidelines and protocols.

During the last decade, hospitals' discharge processes have been the focus of tremendous research, policy, and quality improvement efforts. The phrase transition of care is now widely understood to describe the changes in patient care that begin with discharge planning and conclude when patients' have established care at home or another healthcare facility. Transitions of care have been a focus of the Journal of Hospital Medicine since its inception, including publication of the Transitions of Care Consensus Policy Statement in 2009, as well as numerous other studies highlighting both risks associated with transitions of care as well as methods to address these.[11, 12, 13, 14, 15, 16] Similar to hospital discharge, hospital admission is an inherent feature of every hospitalization, and admission and discharge processes share many commonalities. Both involve transitions in sites of care, and handoffs between healthcare providers. Most involve changes in medical therapies, including both the addition of new medications and changes to existing treatments. Moreover, both are associated with significant stress to patients and their families. As a result, hospital admissions expose patients to many of same risks that have been the focus of hospital discharge reform: unstructured patient handoffs, poor communication between healthcare providers, and costly, inefficient care. The Society of Hospital Medicine has been a leader in articulating the importance of patient‐centered, clinically relevant medication reconciliation across the healthcare continuum.[17] However, with the exception of this important work, research and policy focused on understanding and improving transitions of care into the hospital have received disproportionately little attention.

To facilitate research and quality improvement efforts focused on hospital admission, we suggest that the transitions of care framework, typically discussed in the context of hospital discharge, be expanded to reflect the different origins of hospitalizations and multiple transitions that can be experienced by patients as they enter the hospital. A broadening of the transitions of care framework to incorporate hospital admissions brings numerous questions previously addressed in hospital‐to‐home transitions to the forefront. How do transitions into the hospital impact patients and healthcare systems? When is direct admission safe and effective, and how does this vary across conditions and hospital settings? What protocols and tools might optimize the associated transitions and reduce the risks of error and harm? There are numerous stakeholders who will undoubtedly bring diverse perspectives to these questionspatients and their families, hospital‐based healthcare providers, PCPs and specialists, ED physicians, and payors.

Increasing ED volumes, long wait times, and rising ED costs speak to the importance of better understanding hospital admission alternatives and the associated risks and benefits. Encouraging more direct admissions may be a solution, but evidence to guide best practices must precede this. The growing presence of round‐the‐clock pediatric and adult hospitalists across the country creates unique opportunities to transform hospital admission systems for the vast number of patients who do not require emergent care. The Affordable Care Act's expansion of insurance coverage and incentivized coordinated care within patient‐centered medical homes creates a unique opportunity for this broadened view of transitions of care. This suggests that the time is ripe for pursuing strategies that will both improve patients' transitions from outpatient to inpatient care and reduce stress on our overburdened emergency departments.

Disclosure: Dr. Lagu was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under award number K01HL114745. She has received consulting fees from the Institute for Healthcare Improvement, under contract to the Centers for Medicare and Medicaid Services (CMS), for her work on a project to help health systems achieve disability competence, and from the Island Peer Review Organization, under contract to CMS, for her work on development of episodes of care for CMS payment purposes (both unrelated to the current work). Dr. Leyenaar was supported by grant number K08HS024133 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality. The authors report no conflicts of interest.

Increasing use of emergency departments (EDs) throughout the United States has become a focus of national healthcare policy and reform efforts. ED growth continues to outpace population growth, with the Institute of Medicine describing our ED systems as fragmented, overburdened, and at the breaking point.[1] Associations between ED crowding and patient dissatisfaction, delays in treatment, medical errors, and patient mortality speak to the urgency of systems improvements.[2] One major factor contributing to ED volumes is the growing number of hospital admissions that begin in EDs. From 1993 to 2006, the proportion of hospitalizations originating in EDs increased from 33.5% to 43.8%, with more than 17 million hospital admissions originating in EDs annually.[3, 4] Despite these challenges, discussions about alternative approaches to hospital admission remain at the periphery of healthcare policy conversations.

Direct admission to the hospital, defined as hospitalization without first receiving care in the hospital's ED, is an alternative approach to hospital admission, and may be a vehicle to both observation and inpatient hospital stays. Direct admissions account for 25% of all nonelective pediatric hospitalizations and 15% of nonelective adult hospitalizations in the United States.[5, 6] This admission approach was considerably more common in the past, facilitated by primary care providers (PCPs) or specialists who provided both outpatient and hospital‐based care for their patients.[4] However, as the number of hospitalists in the United States has grown over the last 30 years, the number of direct admissions has decreased concurrently. In fact, from 2003 to 2009, the number of direct admissions from clinics and physicians' offices decreased by a total of 1.6 million.[4] Although this decline is undoubtedly multifactorial, hospitalists may have contributed, both deliberately and inadvertently, to the shifting epidemiology of hospital admissions. Although many factors influence the source of hospital admissions and admission processes, direct admission has 2 important prerequisites: patients require timely access to outpatient providers for acute care, and hospitals, in partnership with outpatient‐based clinics and practices, require systems to safely and efficiently facilitate admissions without ED involvement. However, we know little about hospital admission systems, developed in the era of hospital medicine, to facilitate admissions independent of the ED.

Direct admission offers a number of potential benefits for both patients and healthcare delivery systems including reductions in the number of sites and providers of care, improved communication and coordination between outpatient and hospital‐based healthcare providers, greater patient and referring physician satisfaction, and reduced ED volumes and subsequent costs.[7] However, there are also risks and potential harms associated with direct admission, including potential delays in initial evaluation and management, inconsistent admission processes, and difficulties determining direct admission appropriateness, all of which could adversely impact patient safety and quality of care.[7, 8, 9] One study of adults with sepsis found that direct admission was associated with increased mortality compared to ED admission, which the authors speculated to be related to less timely care.[9] Similarly, a study of unscheduled adult hospitalizations found that patients admitted directly had higher mortality for time‐sensitive conditions such as acute myocardial infarction and sepsis than patients admitted through EDs, differences not observed among adults admitted with pneumonia, asthma, cellulitis, and several other common, yet frequently less emergent, reasons for hospitalization.[8] Among children with pneumonia, the most common reason for pediatric hospitalization, direct admission has been associated with significantly lower costs than admissions originating in the ED, with no significant differences in rates of transfer to the intensive care unit or hospital readmission.[10]

There is significant variation across both diagnoses and hospitals in rates of direct admission, raising questions about the contextual factors unique to hospital medicine programs that perform a substantial proportion of direct admissions.[5] This variation also highlights opportunities to identify the populations, conditions, and systems that facilitate safe and effective direct admissions. Certainly, direct admission is unlikely to be appropriate for all populations or conditions. Patients requiring emergent care or rapid diagnostic imaging are likely to receive more timely care in the ED; sepsis, acute myocardial infarction, and trauma are but a few examples of conditions for which rapid ED care decreases morbidity and mortality. Similarly, patients for whom the need for hospitalization is uncertainfor example, dehydration, asthmamay be more appropriate for initial ED management followed by re‐evaluation to inform the need for hospitalization. Finally, patients for whom the admitting diagnosis is uncertain and who require consultation for several subspecialists may be more efficiently evaluated in EDs. In our national survey of pediatric direct admission guidelines, less than one‐third of hospitals reported having formal criteria to assess the appropriateness of direct admissions, and respondents' perspectives regarding populations and diagnoses appropriate for this admission approach varied considerably.[7] These results point to the need for further research and quality‐improvement initiatives to inform the development of direct admission guidelines and protocols.

During the last decade, hospitals' discharge processes have been the focus of tremendous research, policy, and quality improvement efforts. The phrase transition of care is now widely understood to describe the changes in patient care that begin with discharge planning and conclude when patients' have established care at home or another healthcare facility. Transitions of care have been a focus of the Journal of Hospital Medicine since its inception, including publication of the Transitions of Care Consensus Policy Statement in 2009, as well as numerous other studies highlighting both risks associated with transitions of care as well as methods to address these.[11, 12, 13, 14, 15, 16] Similar to hospital discharge, hospital admission is an inherent feature of every hospitalization, and admission and discharge processes share many commonalities. Both involve transitions in sites of care, and handoffs between healthcare providers. Most involve changes in medical therapies, including both the addition of new medications and changes to existing treatments. Moreover, both are associated with significant stress to patients and their families. As a result, hospital admissions expose patients to many of same risks that have been the focus of hospital discharge reform: unstructured patient handoffs, poor communication between healthcare providers, and costly, inefficient care. The Society of Hospital Medicine has been a leader in articulating the importance of patient‐centered, clinically relevant medication reconciliation across the healthcare continuum.[17] However, with the exception of this important work, research and policy focused on understanding and improving transitions of care into the hospital have received disproportionately little attention.

To facilitate research and quality improvement efforts focused on hospital admission, we suggest that the transitions of care framework, typically discussed in the context of hospital discharge, be expanded to reflect the different origins of hospitalizations and multiple transitions that can be experienced by patients as they enter the hospital. A broadening of the transitions of care framework to incorporate hospital admissions brings numerous questions previously addressed in hospital‐to‐home transitions to the forefront. How do transitions into the hospital impact patients and healthcare systems? When is direct admission safe and effective, and how does this vary across conditions and hospital settings? What protocols and tools might optimize the associated transitions and reduce the risks of error and harm? There are numerous stakeholders who will undoubtedly bring diverse perspectives to these questionspatients and their families, hospital‐based healthcare providers, PCPs and specialists, ED physicians, and payors.

Increasing ED volumes, long wait times, and rising ED costs speak to the importance of better understanding hospital admission alternatives and the associated risks and benefits. Encouraging more direct admissions may be a solution, but evidence to guide best practices must precede this. The growing presence of round‐the‐clock pediatric and adult hospitalists across the country creates unique opportunities to transform hospital admission systems for the vast number of patients who do not require emergent care. The Affordable Care Act's expansion of insurance coverage and incentivized coordinated care within patient‐centered medical homes creates a unique opportunity for this broadened view of transitions of care. This suggests that the time is ripe for pursuing strategies that will both improve patients' transitions from outpatient to inpatient care and reduce stress on our overburdened emergency departments.

Disclosure: Dr. Lagu was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under award number K01HL114745. She has received consulting fees from the Institute for Healthcare Improvement, under contract to the Centers for Medicare and Medicaid Services (CMS), for her work on a project to help health systems achieve disability competence, and from the Island Peer Review Organization, under contract to CMS, for her work on development of episodes of care for CMS payment purposes (both unrelated to the current work). Dr. Leyenaar was supported by grant number K08HS024133 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality. The authors report no conflicts of interest.

A 71‐year‐old man with widely metastatic nonsmall cell lung cancer presented to an emergency department of a teaching hospital at 7 _pm with a chief complaint of severe chest pain relieved by sitting upright and leaning forward. A senior cardiologist, with expertise in echocardiography, assessed the patient and performed a bedside echocardiogram. He found a large pericardial effusion but concluded there was no cardiac tamponade. Given the patient's other medical problems, he referred him to internal medicine for admission to their service. The attending internist agreed to admit the patient, suggesting close cardiac monitoring and reevaluation with a formal echocardiogram in the morning. At 9 _am, the team and the cardiologist were urgently summoned to the echo lab by the technician who now diagnosed tamponade. After looking at the images, the cardiologist disagreed with the technician's interpretation and declared that there was no sign of tamponade.

After leaving the echo lab, the attending internist led a team discussion on the phenomenon of and reasons for interobserver variation. The residents initially focused on the difference in expertise between the cardiologist and technician. The attending, who felt this was unlikely because the technician was very experienced, introduced the possibility of a cognitive misstep. Having staked out an opinion on the lack of tamponade the night before and acting on that interpretation by declining admission to his service, the cardiologist was susceptible to anchoring bias, where adjustments to a preliminary diagnosis are insufficient because of the influence of the initial interpretation.[1] The following day, the cardiologist performed a pericardiocentesis and reported that the fluid came out under pressure. In the face of this definitive information, he concluded that his prior assessment was incorrect and that tamponade had been present from the start.

The origins of medical error reduction lie in the practice of using autopsies to determine the cause of death spearheaded by Karl Rokitansky at the Vienna Medical School in the 1800s.[2] Ernest Amory Codman expanded the effort through the linkage of treatment decisions to subsequent outcomes by following patients after hospital discharge.[3] The advent of modern imaging techniques coupled with interventional methods of obtaining pathological specimens has dramatically improved diagnostic accuracy over the past 40 years. As a result, the practice of using autopsies to improve clinical acumen and reduce diagnostic error has virtually disappeared, while the focus on medical error has actually increased. The forum for reducing error shifted to morbidity and mortality rounds (MMRs), which have been relabeled quality‐improvement rounds in many hospitals.

In these regularly scheduled meetings, interprofessional clinicians discuss errors and adverse outcomes. Because deaths are rarely unexpected and often occur outside of the acute care setting, the focus is usually on errors in the execution of complex clinical plans that combine the wide array of modern laboratory, imaging, pharmaceutical, interventional, surgical, and pathological tools available to clinicians today. In the era of patient safety and quality improvement, errors are mostly blamed on systems‐based issues that lead to hospital complications, despite evidence that cognitive factors play a large role.[4] Systems‐based analysis was popularized by the landmark report of the Institute of Medicine.[5] In our local institutions (the University of Toronto teaching hospitals), improving diagnostic accuracy is almost never on the agenda. We suspect the same is true elsewhere. Common themes include mistakes in medication administration and dosing, communication, and physician handover. The Swiss cheese model[6] is often invoked to diffuse blame across a number of individuals, processes, and even machines. However, as Wachter and Pronovost point out, reengineering of systems has limited capacity for solving all safety and quality improvement issues when people are involved; human error can still sabotage the effort.[7]

Discussions centered on a physician's raw thinking ability have become a third rail, even though clinical reasoning lies at the core of patient safety. Human error is rarely discussed, in part because it is mistakenly believed to be uncommon and felt to be the result of deficits in knowledge or incompetence. Furthermore, the fear of assigning blame to individuals in front of their peers may be counterproductive, discouraging identification of future errors. However, the fields of cognitive psychology and medical decision making have clearly established that cognitive errors occur predictably and often, especially at times of high cognitive load (eg, when many high stakes complex decisions need to be made in a short period of time). Errors do not usually result from a lack of knowledge (although they can), but rather because people rely on instincts that include common biases called heuristics.[8] Most of the time, heuristics are a helpful and necessary evolutionary adaptation of the human thought process, but by their inherent nature, they can lead to predictable and repeatable errors. Because the effects of cognitive biases are inherent to all decision makers, using this framework for discussing individual error may be a method of decreasing the second victim effect[9] and avoid demoralizing the individual.

MMRs thus represent fertile ground for introducing cognitive psychology into medical education and quality improvement. The existing format is useful for teaching cognitive psychology because it is an open forum where discussions center on errors of omission and commission, many of which are a result of both systems issues and decision making heuristics. Several studies have attempted to describe methods for improving MMRs[10, 11, 12]; however, none have incorporated concepts from cognitive psychology. This type of analysis has penetrated several cases in the WebM&M series created by the Agency of Healthcare Quality Research, which can be used as a model for hospital‐based MMRs.[13] For the vignette described above, a MMR that considers systems‐based approaches might discuss how a busy emergency room, limitations of capacity on the cardiology service, and closure of the echo lab at night, played a role in this story. However, although it is difficult to replay another person's mental processing, ignoring the possibility that the cardiologist in this case may have fallen prey to a common cognitive error would be a missed opportunity to learn how frequently heuristics can be faulty. A cognitive approach applied to this example would explore explanations such as anchoring, ego, and hassle biases. Front‐line clinicians in busy hospital settings will recognize the interaction between workload pressures and cognitive mistakes common to examples like this one.

Cognitive heuristics should first be introduced to MMRs by experienced clinicians, well respected for their clinical acumen, by telling specific personal stories where heuristics led to errors in their practices and why the shortcut in thinking occurred. Thereafter, the traditional MMR format can be used: presenting a case, describing how an experienced clinician might manage the case, and then asking the audience members for comment. Incorporating discussions of cognitive missteps, in medical and nonmedical contexts, would help normalize the understanding that even the most experienced and smartest people fall prey to them. The tone must be positive.

Attendees could be encouraged to review their own thought processes through diagnostic verification for cases where their initial diagnosis was incorrect. This would involve assessment for adequacy, ensuring that potential diagnoses account for all abnormal and normal clinical findings, and coherency, ensuring that the diagnoses are pathophysiologically consistent with all clinical findings. Another strategy may be to illustrate cognitive forcing strategies for particular biases.[14] For example, in the case of anchoring bias, trainees may be encouraged to replay the clinical scenario with a different priming stem and evaluate if they would come to the same clinical conclusion. A challenge for all MMRs is how best to select cases; given the difficulties in replaying one's cognitive processes, this problem may be magnified. Potential selection methods could utilize anonymous reporting systems or patient complaints; however, the optimal strategy is yet to be determined.

Graber et al. have summarized the limited research on attempts to improve cognitive processes through educational interventions and illustrate its mixed results.[15] The most positive study was a randomized control trial using combined pattern recognition and deliberative reasoning to improve diagnostic accuracy in the face of biasing information.[16] Despite positive results, others have suggested that cognitive biases are impossible to teach due to their subconscious nature.[17] They argue that training physicians to avoid heuristics will simply lead to over investigation. These polarizing views highlight the need for research to evaluate interventions like the cognitive autopsy suggested here.

Trainees recognize early that their knowledge base is limited. However, it takes more internal analysis to realize that their brains' decision‐making capacity is similarly limited. Utilizing these regularly scheduled clinical meetings in the manner described above may build improved metacognition, cognition about cognition or more colloquially thinking about thinking. Clinicians understand that bias can easily occur in research and accept mechanisms to protect studies from those potential threats to validity such as double blinding of outcome assessments. Supplementing MMRs with cognitive discussions represents an analogous intent to reduce biases, introducing metacognition as the next frontier in advancing clinical care. Errors are inevitable,[18] and recognition of our cognitive blind spots will provide physicians with an improved framework for analysis of these errors. Building metacognition is a difficult task; however, this is not a reason to stop trying. In the spirit of innovation begun by pioneers like Rokitansky and Codman, and renewed focus on diagnostic errors generated by the recent report of the National Academy of Sciences[19], it is time for the cognitive autopsy to be built into the quality improvement and patient safety map.

A 71‐year‐old man with widely metastatic nonsmall cell lung cancer presented to an emergency department of a teaching hospital at 7 _pm with a chief complaint of severe chest pain relieved by sitting upright and leaning forward. A senior cardiologist, with expertise in echocardiography, assessed the patient and performed a bedside echocardiogram. He found a large pericardial effusion but concluded there was no cardiac tamponade. Given the patient's other medical problems, he referred him to internal medicine for admission to their service. The attending internist agreed to admit the patient, suggesting close cardiac monitoring and reevaluation with a formal echocardiogram in the morning. At 9 _am, the team and the cardiologist were urgently summoned to the echo lab by the technician who now diagnosed tamponade. After looking at the images, the cardiologist disagreed with the technician's interpretation and declared that there was no sign of tamponade.

After leaving the echo lab, the attending internist led a team discussion on the phenomenon of and reasons for interobserver variation. The residents initially focused on the difference in expertise between the cardiologist and technician. The attending, who felt this was unlikely because the technician was very experienced, introduced the possibility of a cognitive misstep. Having staked out an opinion on the lack of tamponade the night before and acting on that interpretation by declining admission to his service, the cardiologist was susceptible to anchoring bias, where adjustments to a preliminary diagnosis are insufficient because of the influence of the initial interpretation.[1] The following day, the cardiologist performed a pericardiocentesis and reported that the fluid came out under pressure. In the face of this definitive information, he concluded that his prior assessment was incorrect and that tamponade had been present from the start.

The origins of medical error reduction lie in the practice of using autopsies to determine the cause of death spearheaded by Karl Rokitansky at the Vienna Medical School in the 1800s.[2] Ernest Amory Codman expanded the effort through the linkage of treatment decisions to subsequent outcomes by following patients after hospital discharge.[3] The advent of modern imaging techniques coupled with interventional methods of obtaining pathological specimens has dramatically improved diagnostic accuracy over the past 40 years. As a result, the practice of using autopsies to improve clinical acumen and reduce diagnostic error has virtually disappeared, while the focus on medical error has actually increased. The forum for reducing error shifted to morbidity and mortality rounds (MMRs), which have been relabeled quality‐improvement rounds in many hospitals.

In these regularly scheduled meetings, interprofessional clinicians discuss errors and adverse outcomes. Because deaths are rarely unexpected and often occur outside of the acute care setting, the focus is usually on errors in the execution of complex clinical plans that combine the wide array of modern laboratory, imaging, pharmaceutical, interventional, surgical, and pathological tools available to clinicians today. In the era of patient safety and quality improvement, errors are mostly blamed on systems‐based issues that lead to hospital complications, despite evidence that cognitive factors play a large role.[4] Systems‐based analysis was popularized by the landmark report of the Institute of Medicine.[5] In our local institutions (the University of Toronto teaching hospitals), improving diagnostic accuracy is almost never on the agenda. We suspect the same is true elsewhere. Common themes include mistakes in medication administration and dosing, communication, and physician handover. The Swiss cheese model[6] is often invoked to diffuse blame across a number of individuals, processes, and even machines. However, as Wachter and Pronovost point out, reengineering of systems has limited capacity for solving all safety and quality improvement issues when people are involved; human error can still sabotage the effort.[7]

Discussions centered on a physician's raw thinking ability have become a third rail, even though clinical reasoning lies at the core of patient safety. Human error is rarely discussed, in part because it is mistakenly believed to be uncommon and felt to be the result of deficits in knowledge or incompetence. Furthermore, the fear of assigning blame to individuals in front of their peers may be counterproductive, discouraging identification of future errors. However, the fields of cognitive psychology and medical decision making have clearly established that cognitive errors occur predictably and often, especially at times of high cognitive load (eg, when many high stakes complex decisions need to be made in a short period of time). Errors do not usually result from a lack of knowledge (although they can), but rather because people rely on instincts that include common biases called heuristics.[8] Most of the time, heuristics are a helpful and necessary evolutionary adaptation of the human thought process, but by their inherent nature, they can lead to predictable and repeatable errors. Because the effects of cognitive biases are inherent to all decision makers, using this framework for discussing individual error may be a method of decreasing the second victim effect[9] and avoid demoralizing the individual.

MMRs thus represent fertile ground for introducing cognitive psychology into medical education and quality improvement. The existing format is useful for teaching cognitive psychology because it is an open forum where discussions center on errors of omission and commission, many of which are a result of both systems issues and decision making heuristics. Several studies have attempted to describe methods for improving MMRs[10, 11, 12]; however, none have incorporated concepts from cognitive psychology. This type of analysis has penetrated several cases in the WebM&M series created by the Agency of Healthcare Quality Research, which can be used as a model for hospital‐based MMRs.[13] For the vignette described above, a MMR that considers systems‐based approaches might discuss how a busy emergency room, limitations of capacity on the cardiology service, and closure of the echo lab at night, played a role in this story. However, although it is difficult to replay another person's mental processing, ignoring the possibility that the cardiologist in this case may have fallen prey to a common cognitive error would be a missed opportunity to learn how frequently heuristics can be faulty. A cognitive approach applied to this example would explore explanations such as anchoring, ego, and hassle biases. Front‐line clinicians in busy hospital settings will recognize the interaction between workload pressures and cognitive mistakes common to examples like this one.

Cognitive heuristics should first be introduced to MMRs by experienced clinicians, well respected for their clinical acumen, by telling specific personal stories where heuristics led to errors in their practices and why the shortcut in thinking occurred. Thereafter, the traditional MMR format can be used: presenting a case, describing how an experienced clinician might manage the case, and then asking the audience members for comment. Incorporating discussions of cognitive missteps, in medical and nonmedical contexts, would help normalize the understanding that even the most experienced and smartest people fall prey to them. The tone must be positive.

Attendees could be encouraged to review their own thought processes through diagnostic verification for cases where their initial diagnosis was incorrect. This would involve assessment for adequacy, ensuring that potential diagnoses account for all abnormal and normal clinical findings, and coherency, ensuring that the diagnoses are pathophysiologically consistent with all clinical findings. Another strategy may be to illustrate cognitive forcing strategies for particular biases.[14] For example, in the case of anchoring bias, trainees may be encouraged to replay the clinical scenario with a different priming stem and evaluate if they would come to the same clinical conclusion. A challenge for all MMRs is how best to select cases; given the difficulties in replaying one's cognitive processes, this problem may be magnified. Potential selection methods could utilize anonymous reporting systems or patient complaints; however, the optimal strategy is yet to be determined.

Graber et al. have summarized the limited research on attempts to improve cognitive processes through educational interventions and illustrate its mixed results.[15] The most positive study was a randomized control trial using combined pattern recognition and deliberative reasoning to improve diagnostic accuracy in the face of biasing information.[16] Despite positive results, others have suggested that cognitive biases are impossible to teach due to their subconscious nature.[17] They argue that training physicians to avoid heuristics will simply lead to over investigation. These polarizing views highlight the need for research to evaluate interventions like the cognitive autopsy suggested here.

Trainees recognize early that their knowledge base is limited. However, it takes more internal analysis to realize that their brains' decision‐making capacity is similarly limited. Utilizing these regularly scheduled clinical meetings in the manner described above may build improved metacognition, cognition about cognition or more colloquially thinking about thinking. Clinicians understand that bias can easily occur in research and accept mechanisms to protect studies from those potential threats to validity such as double blinding of outcome assessments. Supplementing MMRs with cognitive discussions represents an analogous intent to reduce biases, introducing metacognition as the next frontier in advancing clinical care. Errors are inevitable,[18] and recognition of our cognitive blind spots will provide physicians with an improved framework for analysis of these errors. Building metacognition is a difficult task; however, this is not a reason to stop trying. In the spirit of innovation begun by pioneers like Rokitansky and Codman, and renewed focus on diagnostic errors generated by the recent report of the National Academy of Sciences[19], it is time for the cognitive autopsy to be built into the quality improvement and patient safety map.

A 71‐year‐old man with widely metastatic nonsmall cell lung cancer presented to an emergency department of a teaching hospital at 7 _pm with a chief complaint of severe chest pain relieved by sitting upright and leaning forward. A senior cardiologist, with expertise in echocardiography, assessed the patient and performed a bedside echocardiogram. He found a large pericardial effusion but concluded there was no cardiac tamponade. Given the patient's other medical problems, he referred him to internal medicine for admission to their service. The attending internist agreed to admit the patient, suggesting close cardiac monitoring and reevaluation with a formal echocardiogram in the morning. At 9 _am, the team and the cardiologist were urgently summoned to the echo lab by the technician who now diagnosed tamponade. After looking at the images, the cardiologist disagreed with the technician's interpretation and declared that there was no sign of tamponade.

After leaving the echo lab, the attending internist led a team discussion on the phenomenon of and reasons for interobserver variation. The residents initially focused on the difference in expertise between the cardiologist and technician. The attending, who felt this was unlikely because the technician was very experienced, introduced the possibility of a cognitive misstep. Having staked out an opinion on the lack of tamponade the night before and acting on that interpretation by declining admission to his service, the cardiologist was susceptible to anchoring bias, where adjustments to a preliminary diagnosis are insufficient because of the influence of the initial interpretation.[1] The following day, the cardiologist performed a pericardiocentesis and reported that the fluid came out under pressure. In the face of this definitive information, he concluded that his prior assessment was incorrect and that tamponade had been present from the start.

The origins of medical error reduction lie in the practice of using autopsies to determine the cause of death spearheaded by Karl Rokitansky at the Vienna Medical School in the 1800s.[2] Ernest Amory Codman expanded the effort through the linkage of treatment decisions to subsequent outcomes by following patients after hospital discharge.[3] The advent of modern imaging techniques coupled with interventional methods of obtaining pathological specimens has dramatically improved diagnostic accuracy over the past 40 years. As a result, the practice of using autopsies to improve clinical acumen and reduce diagnostic error has virtually disappeared, while the focus on medical error has actually increased. The forum for reducing error shifted to morbidity and mortality rounds (MMRs), which have been relabeled quality‐improvement rounds in many hospitals.

In these regularly scheduled meetings, interprofessional clinicians discuss errors and adverse outcomes. Because deaths are rarely unexpected and often occur outside of the acute care setting, the focus is usually on errors in the execution of complex clinical plans that combine the wide array of modern laboratory, imaging, pharmaceutical, interventional, surgical, and pathological tools available to clinicians today. In the era of patient safety and quality improvement, errors are mostly blamed on systems‐based issues that lead to hospital complications, despite evidence that cognitive factors play a large role.[4] Systems‐based analysis was popularized by the landmark report of the Institute of Medicine.[5] In our local institutions (the University of Toronto teaching hospitals), improving diagnostic accuracy is almost never on the agenda. We suspect the same is true elsewhere. Common themes include mistakes in medication administration and dosing, communication, and physician handover. The Swiss cheese model[6] is often invoked to diffuse blame across a number of individuals, processes, and even machines. However, as Wachter and Pronovost point out, reengineering of systems has limited capacity for solving all safety and quality improvement issues when people are involved; human error can still sabotage the effort.[7]

Discussions centered on a physician's raw thinking ability have become a third rail, even though clinical reasoning lies at the core of patient safety. Human error is rarely discussed, in part because it is mistakenly believed to be uncommon and felt to be the result of deficits in knowledge or incompetence. Furthermore, the fear of assigning blame to individuals in front of their peers may be counterproductive, discouraging identification of future errors. However, the fields of cognitive psychology and medical decision making have clearly established that cognitive errors occur predictably and often, especially at times of high cognitive load (eg, when many high stakes complex decisions need to be made in a short period of time). Errors do not usually result from a lack of knowledge (although they can), but rather because people rely on instincts that include common biases called heuristics.[8] Most of the time, heuristics are a helpful and necessary evolutionary adaptation of the human thought process, but by their inherent nature, they can lead to predictable and repeatable errors. Because the effects of cognitive biases are inherent to all decision makers, using this framework for discussing individual error may be a method of decreasing the second victim effect[9] and avoid demoralizing the individual.

MMRs thus represent fertile ground for introducing cognitive psychology into medical education and quality improvement. The existing format is useful for teaching cognitive psychology because it is an open forum where discussions center on errors of omission and commission, many of which are a result of both systems issues and decision making heuristics. Several studies have attempted to describe methods for improving MMRs[10, 11, 12]; however, none have incorporated concepts from cognitive psychology. This type of analysis has penetrated several cases in the WebM&M series created by the Agency of Healthcare Quality Research, which can be used as a model for hospital‐based MMRs.[13] For the vignette described above, a MMR that considers systems‐based approaches might discuss how a busy emergency room, limitations of capacity on the cardiology service, and closure of the echo lab at night, played a role in this story. However, although it is difficult to replay another person's mental processing, ignoring the possibility that the cardiologist in this case may have fallen prey to a common cognitive error would be a missed opportunity to learn how frequently heuristics can be faulty. A cognitive approach applied to this example would explore explanations such as anchoring, ego, and hassle biases. Front‐line clinicians in busy hospital settings will recognize the interaction between workload pressures and cognitive mistakes common to examples like this one.

Cognitive heuristics should first be introduced to MMRs by experienced clinicians, well respected for their clinical acumen, by telling specific personal stories where heuristics led to errors in their practices and why the shortcut in thinking occurred. Thereafter, the traditional MMR format can be used: presenting a case, describing how an experienced clinician might manage the case, and then asking the audience members for comment. Incorporating discussions of cognitive missteps, in medical and nonmedical contexts, would help normalize the understanding that even the most experienced and smartest people fall prey to them. The tone must be positive.

Attendees could be encouraged to review their own thought processes through diagnostic verification for cases where their initial diagnosis was incorrect. This would involve assessment for adequacy, ensuring that potential diagnoses account for all abnormal and normal clinical findings, and coherency, ensuring that the diagnoses are pathophysiologically consistent with all clinical findings. Another strategy may be to illustrate cognitive forcing strategies for particular biases.[14] For example, in the case of anchoring bias, trainees may be encouraged to replay the clinical scenario with a different priming stem and evaluate if they would come to the same clinical conclusion. A challenge for all MMRs is how best to select cases; given the difficulties in replaying one's cognitive processes, this problem may be magnified. Potential selection methods could utilize anonymous reporting systems or patient complaints; however, the optimal strategy is yet to be determined.

Graber et al. have summarized the limited research on attempts to improve cognitive processes through educational interventions and illustrate its mixed results.[15] The most positive study was a randomized control trial using combined pattern recognition and deliberative reasoning to improve diagnostic accuracy in the face of biasing information.[16] Despite positive results, others have suggested that cognitive biases are impossible to teach due to their subconscious nature.[17] They argue that training physicians to avoid heuristics will simply lead to over investigation. These polarizing views highlight the need for research to evaluate interventions like the cognitive autopsy suggested here.

Trainees recognize early that their knowledge base is limited. However, it takes more internal analysis to realize that their brains' decision‐making capacity is similarly limited. Utilizing these regularly scheduled clinical meetings in the manner described above may build improved metacognition, cognition about cognition or more colloquially thinking about thinking. Clinicians understand that bias can easily occur in research and accept mechanisms to protect studies from those potential threats to validity such as double blinding of outcome assessments. Supplementing MMRs with cognitive discussions represents an analogous intent to reduce biases, introducing metacognition as the next frontier in advancing clinical care. Errors are inevitable,[18] and recognition of our cognitive blind spots will provide physicians with an improved framework for analysis of these errors. Building metacognition is a difficult task; however, this is not a reason to stop trying. In the spirit of innovation begun by pioneers like Rokitansky and Codman, and renewed focus on diagnostic errors generated by the recent report of the National Academy of Sciences[19], it is time for the cognitive autopsy to be built into the quality improvement and patient safety map.

The United States is moving aggressively toward value‐based payment. The Department of Health and Human Services recently announced a goal to link 85% of Medicare's fee‐for‐service payments to quality or value by 2016.[1] Despite the inherent logic of paying providers for their results, evidence of the effectiveness of value‐based payment has been mixed and underwhelming. Recent reviews of pay‐for‐performancereflecting the emerging understanding of the complexities of designing successful programshave painted a more negative picture of their overall effectiveness.[2, 3] One study of over 6 million patients found that the Medicare Premier Hospital Quality Incentive Demonstration had no effect on long‐term patient outcomes including 30‐day mortality.[4] At the same time, research suggests that lower performing providers tend to have a disproportionate number of poor patients, many of whom are racial and ethnic minorities. Value‐based payment risks the dual failure of not improving health outcomes while exacerbating health inequities.

We have seen this movie before. In 2001, No Child Left Behind was enacted to improve quality and reduce inequities in K12 education in the United States. Much like healthcare, education suffers from uneven quality and wide socioeconomic disparities.[5] No Child Left Behind attempted to address these problems with new accountability measures. Based on the results from standardized tests, No Child Left Behind rewarded the highest performing schools with more funding while penalizing poor performing schools with reduced funding, and in some cases, forcing failing schools to cede control to outside operators.

In the aftermath of its implementation, however, it became clear that these incentives had not worked as intended. No Child Left Behind did not improve reading performance and was associated with improvements in math performance only for younger students.[6] These modest gains came at a high cost; consistent with teaching to the test, No Child Left Behind led to a shifting of instructional time toward math and reading and away from other subjects. It also led to widespread cheating, challenging the validity of observed performance improvements. Before No Child Left Behind was rolled out, the wealthiest school districts in the country spent as much as 10 times more than the poorest districts.[5] By penalizing the lowest performers, these gaps persisted. Schools were not given the support that they needed to improve performance.

The parallels to healthcare are striking (Table 1). Early results from Medicare's Hospital Value‐Based Purchasing and Readmission Reduction Program show that hospitals caring for more disadvantaged patients have been disproportionately penalized.[7] Similar reverse Robin Hood effects have been observed in incentive programs for physician practices.[8] Over time, financial incentive programs may substantially decrease operating revenue for hospitals and physicians caring for low‐income and minority communities. This could perpetuate the already large disparities in quality and health outcomes facing these populations. Although risk‐adjusting for socioeconomic status may alleviate these concerns in the short term, allowing low‐income or minority patients to have poorer health outcomes simply accepts that disparities exist rather than trying to reduce them.

How then is it possible to improve the quality of care at lower performing hospitals without simultaneously designing an incentive system that hurts them? Lessons from the education policy are again instructive. Every 3 years the Organization for Economic Cooperation and Development ranks countries by the performance of their 15‐year olds on a standardized test called the Program for International Student Assessment.[9] For the past 2 sets of rankings, Shanghai, China has topped the list. Like many attempts to generate international rankings, this one has its flaws, and Shanghai's top position has not been without controversy. For one, China is not ranked at the country‐level like other nations; yet, due to the city's status as a wealthy business and financial center, Shanghai certainly cannot be considered representative of the Chinese education system. Nevertheless, the story of how Shanghai reformed its education system and achieved its high position has important implications.

Prior to implementing reforms, Shanghai's rural outer districts struggled with less funding, high teacher turnover rates, and low test scores compared to wealthier urban districts. To reduce education disparities within the city's schools, the government of Shanghai enacted a number of policies aimed at bringing lower performers up to the same level as schools with the highest degree of student achievement.[10] The government gives schools a grade of A, B, C, or D based on the quality of their infrastructure and student performance. It then uses several programs to facilitate the exchange of staff and ideas among schools at different levels. One program pairs high‐performing districts with low‐performing districts to share education development plans, curricula, teaching materials, and best practices. Another strategycalled commissioned administrationinvolves temporary contracts between schools to exchange both administrators and experienced teachers. In addition to these approaches, the government sets a minimum level of spending for schools and transfers public funds to indigent districts to provide them with assistance to reach this level.

The notion that the very best can help the weak requires a sense of solidarity. This solidarity may falter in environments in which hospitals and physicians are in cutthroat competition. Though there will always be some tension between competition and collaboration, in most markets, competition between hospitals does not rule out collaboration. Policies can either relieve or reinforce the natural tension between competition and collaboration. This suggests that adopting reforms with the same intent as the Shanghai system is still possible in healthcare, especially through physician and other provider networks. The healthcare workforce has a rich history of cross‐organizational collaboration through mentorships, the publication of research, and participation in continuing medical education courses. The Centers for Medicare and Medicaid Services' Hospital Engagement Networks, a program in which leading organizations have helped to disseminate interventions to reduce hospital acquired conditions, are an example of this approach. Quality collaborativesgroups of providers who collaborate across institutions to identify problems and best practices for improvementhave similarly shown great promise.[11] Similar approaches have been used by the Institute for Healthcare Improvement in many of their quality improvement initiatives.

Such collaboration‐based programs could be harnessed and tied to financial incentives for quality improvement. For instance, top‐performing hospitals could be incentivized to participate in a venue where they share their best practices with the lower performers in their field. Low performers, in turn, could be provided with financial assistance to implement the appropriate changes. Over time, financial assistance could be made contingent on quality improvements. By providing physicians and other providers with examples of what success looks like and assisting them with garnering the resources to reach this level, improvement would not only be incentivized, it might also become more tangible.

Although some hospitals and physicians may welcome changes to incentive systems, implementation of collaboration‐based programs would not be possible without a facilitator that is willing to underwrite program costs, provide financial incentivizes to providers, and develop a platform for collaboration. Large insurers are the most likely group to have the financial resources and widespread network to develop such programs, but that does not mean that they would be willing to experiment with this approach. This may especially be the case if cost savings and measurable improvements in quality are not immediate. Even though the results of collaboration‐based efforts have been promising, the implementation of these programs has been limited, and adoption in different contexts may not yield the same results. Collaboration‐based programs that have already shown success can serve as models, but they may need significant adaptations to meet the needs of providers in a given area.

Despite its promise, collaboration‐based strategies alone will not be enough to improve certain aspects of quality and value. Although providing physicians with knowledge on how to reduce unnecessary care, for example, could help limit overutilization, it is not sufficient to overcome the incentives of fee‐for‐service payment. In this case, broader payment reform and population‐based accountability can be paired with programs to encourage collaboration. For instance, the Blue Cross and Blue Shield of Massachusetts' Alternative Contract has used a combination of technical assistance, shared savings, and large quality bonuses to improve quality and reduce medical spending growth.[12] Collaboration‐based strategies should be seen as a complement to these broad, thoughtful reforms and a substitute for narrow incentives that encourage myopia and destructive competition.

Evidence from education and healthcare shows that penalizing the worst and rewarding the best will not shift the bell curve of performance. Such approaches are more likely to entrench and expand disparities. Instead, policy should encourage and incentivize collaboration to expand best practices that improve patient outcomes. Lessons from education provide both cautionary tales and novel solutions that might improve healthcare.

Disclosure: Nothing to report.

The United States is moving aggressively toward value‐based payment. The Department of Health and Human Services recently announced a goal to link 85% of Medicare's fee‐for‐service payments to quality or value by 2016.[1] Despite the inherent logic of paying providers for their results, evidence of the effectiveness of value‐based payment has been mixed and underwhelming. Recent reviews of pay‐for‐performancereflecting the emerging understanding of the complexities of designing successful programshave painted a more negative picture of their overall effectiveness.[2, 3] One study of over 6 million patients found that the Medicare Premier Hospital Quality Incentive Demonstration had no effect on long‐term patient outcomes including 30‐day mortality.[4] At the same time, research suggests that lower performing providers tend to have a disproportionate number of poor patients, many of whom are racial and ethnic minorities. Value‐based payment risks the dual failure of not improving health outcomes while exacerbating health inequities.

We have seen this movie before. In 2001, No Child Left Behind was enacted to improve quality and reduce inequities in K12 education in the United States. Much like healthcare, education suffers from uneven quality and wide socioeconomic disparities.[5] No Child Left Behind attempted to address these problems with new accountability measures. Based on the results from standardized tests, No Child Left Behind rewarded the highest performing schools with more funding while penalizing poor performing schools with reduced funding, and in some cases, forcing failing schools to cede control to outside operators.

In the aftermath of its implementation, however, it became clear that these incentives had not worked as intended. No Child Left Behind did not improve reading performance and was associated with improvements in math performance only for younger students.[6] These modest gains came at a high cost; consistent with teaching to the test, No Child Left Behind led to a shifting of instructional time toward math and reading and away from other subjects. It also led to widespread cheating, challenging the validity of observed performance improvements. Before No Child Left Behind was rolled out, the wealthiest school districts in the country spent as much as 10 times more than the poorest districts.[5] By penalizing the lowest performers, these gaps persisted. Schools were not given the support that they needed to improve performance.

The parallels to healthcare are striking (Table 1). Early results from Medicare's Hospital Value‐Based Purchasing and Readmission Reduction Program show that hospitals caring for more disadvantaged patients have been disproportionately penalized.[7] Similar reverse Robin Hood effects have been observed in incentive programs for physician practices.[8] Over time, financial incentive programs may substantially decrease operating revenue for hospitals and physicians caring for low‐income and minority communities. This could perpetuate the already large disparities in quality and health outcomes facing these populations. Although risk‐adjusting for socioeconomic status may alleviate these concerns in the short term, allowing low‐income or minority patients to have poorer health outcomes simply accepts that disparities exist rather than trying to reduce them.

How then is it possible to improve the quality of care at lower performing hospitals without simultaneously designing an incentive system that hurts them? Lessons from the education policy are again instructive. Every 3 years the Organization for Economic Cooperation and Development ranks countries by the performance of their 15‐year olds on a standardized test called the Program for International Student Assessment.[9] For the past 2 sets of rankings, Shanghai, China has topped the list. Like many attempts to generate international rankings, this one has its flaws, and Shanghai's top position has not been without controversy. For one, China is not ranked at the country‐level like other nations; yet, due to the city's status as a wealthy business and financial center, Shanghai certainly cannot be considered representative of the Chinese education system. Nevertheless, the story of how Shanghai reformed its education system and achieved its high position has important implications.

Prior to implementing reforms, Shanghai's rural outer districts struggled with less funding, high teacher turnover rates, and low test scores compared to wealthier urban districts. To reduce education disparities within the city's schools, the government of Shanghai enacted a number of policies aimed at bringing lower performers up to the same level as schools with the highest degree of student achievement.[10] The government gives schools a grade of A, B, C, or D based on the quality of their infrastructure and student performance. It then uses several programs to facilitate the exchange of staff and ideas among schools at different levels. One program pairs high‐performing districts with low‐performing districts to share education development plans, curricula, teaching materials, and best practices. Another strategycalled commissioned administrationinvolves temporary contracts between schools to exchange both administrators and experienced teachers. In addition to these approaches, the government sets a minimum level of spending for schools and transfers public funds to indigent districts to provide them with assistance to reach this level.

The notion that the very best can help the weak requires a sense of solidarity. This solidarity may falter in environments in which hospitals and physicians are in cutthroat competition. Though there will always be some tension between competition and collaboration, in most markets, competition between hospitals does not rule out collaboration. Policies can either relieve or reinforce the natural tension between competition and collaboration. This suggests that adopting reforms with the same intent as the Shanghai system is still possible in healthcare, especially through physician and other provider networks. The healthcare workforce has a rich history of cross‐organizational collaboration through mentorships, the publication of research, and participation in continuing medical education courses. The Centers for Medicare and Medicaid Services' Hospital Engagement Networks, a program in which leading organizations have helped to disseminate interventions to reduce hospital acquired conditions, are an example of this approach. Quality collaborativesgroups of providers who collaborate across institutions to identify problems and best practices for improvementhave similarly shown great promise.[11] Similar approaches have been used by the Institute for Healthcare Improvement in many of their quality improvement initiatives.

Such collaboration‐based programs could be harnessed and tied to financial incentives for quality improvement. For instance, top‐performing hospitals could be incentivized to participate in a venue where they share their best practices with the lower performers in their field. Low performers, in turn, could be provided with financial assistance to implement the appropriate changes. Over time, financial assistance could be made contingent on quality improvements. By providing physicians and other providers with examples of what success looks like and assisting them with garnering the resources to reach this level, improvement would not only be incentivized, it might also become more tangible.

Although some hospitals and physicians may welcome changes to incentive systems, implementation of collaboration‐based programs would not be possible without a facilitator that is willing to underwrite program costs, provide financial incentivizes to providers, and develop a platform for collaboration. Large insurers are the most likely group to have the financial resources and widespread network to develop such programs, but that does not mean that they would be willing to experiment with this approach. This may especially be the case if cost savings and measurable improvements in quality are not immediate. Even though the results of collaboration‐based efforts have been promising, the implementation of these programs has been limited, and adoption in different contexts may not yield the same results. Collaboration‐based programs that have already shown success can serve as models, but they may need significant adaptations to meet the needs of providers in a given area.

Despite its promise, collaboration‐based strategies alone will not be enough to improve certain aspects of quality and value. Although providing physicians with knowledge on how to reduce unnecessary care, for example, could help limit overutilization, it is not sufficient to overcome the incentives of fee‐for‐service payment. In this case, broader payment reform and population‐based accountability can be paired with programs to encourage collaboration. For instance, the Blue Cross and Blue Shield of Massachusetts' Alternative Contract has used a combination of technical assistance, shared savings, and large quality bonuses to improve quality and reduce medical spending growth.[12] Collaboration‐based strategies should be seen as a complement to these broad, thoughtful reforms and a substitute for narrow incentives that encourage myopia and destructive competition.

Evidence from education and healthcare shows that penalizing the worst and rewarding the best will not shift the bell curve of performance. Such approaches are more likely to entrench and expand disparities. Instead, policy should encourage and incentivize collaboration to expand best practices that improve patient outcomes. Lessons from education provide both cautionary tales and novel solutions that might improve healthcare.

Disclosure: Nothing to report.

The United States is moving aggressively toward value‐based payment. The Department of Health and Human Services recently announced a goal to link 85% of Medicare's fee‐for‐service payments to quality or value by 2016.[1] Despite the inherent logic of paying providers for their results, evidence of the effectiveness of value‐based payment has been mixed and underwhelming. Recent reviews of pay‐for‐performancereflecting the emerging understanding of the complexities of designing successful programshave painted a more negative picture of their overall effectiveness.[2, 3] One study of over 6 million patients found that the Medicare Premier Hospital Quality Incentive Demonstration had no effect on long‐term patient outcomes including 30‐day mortality.[4] At the same time, research suggests that lower performing providers tend to have a disproportionate number of poor patients, many of whom are racial and ethnic minorities. Value‐based payment risks the dual failure of not improving health outcomes while exacerbating health inequities.

We have seen this movie before. In 2001, No Child Left Behind was enacted to improve quality and reduce inequities in K12 education in the United States. Much like healthcare, education suffers from uneven quality and wide socioeconomic disparities.[5] No Child Left Behind attempted to address these problems with new accountability measures. Based on the results from standardized tests, No Child Left Behind rewarded the highest performing schools with more funding while penalizing poor performing schools with reduced funding, and in some cases, forcing failing schools to cede control to outside operators.

In the aftermath of its implementation, however, it became clear that these incentives had not worked as intended. No Child Left Behind did not improve reading performance and was associated with improvements in math performance only for younger students.[6] These modest gains came at a high cost; consistent with teaching to the test, No Child Left Behind led to a shifting of instructional time toward math and reading and away from other subjects. It also led to widespread cheating, challenging the validity of observed performance improvements. Before No Child Left Behind was rolled out, the wealthiest school districts in the country spent as much as 10 times more than the poorest districts.[5] By penalizing the lowest performers, these gaps persisted. Schools were not given the support that they needed to improve performance.

The parallels to healthcare are striking (Table 1). Early results from Medicare's Hospital Value‐Based Purchasing and Readmission Reduction Program show that hospitals caring for more disadvantaged patients have been disproportionately penalized.[7] Similar reverse Robin Hood effects have been observed in incentive programs for physician practices.[8] Over time, financial incentive programs may substantially decrease operating revenue for hospitals and physicians caring for low‐income and minority communities. This could perpetuate the already large disparities in quality and health outcomes facing these populations. Although risk‐adjusting for socioeconomic status may alleviate these concerns in the short term, allowing low‐income or minority patients to have poorer health outcomes simply accepts that disparities exist rather than trying to reduce them.

How then is it possible to improve the quality of care at lower performing hospitals without simultaneously designing an incentive system that hurts them? Lessons from the education policy are again instructive. Every 3 years the Organization for Economic Cooperation and Development ranks countries by the performance of their 15‐year olds on a standardized test called the Program for International Student Assessment.[9] For the past 2 sets of rankings, Shanghai, China has topped the list. Like many attempts to generate international rankings, this one has its flaws, and Shanghai's top position has not been without controversy. For one, China is not ranked at the country‐level like other nations; yet, due to the city's status as a wealthy business and financial center, Shanghai certainly cannot be considered representative of the Chinese education system. Nevertheless, the story of how Shanghai reformed its education system and achieved its high position has important implications.

Prior to implementing reforms, Shanghai's rural outer districts struggled with less funding, high teacher turnover rates, and low test scores compared to wealthier urban districts. To reduce education disparities within the city's schools, the government of Shanghai enacted a number of policies aimed at bringing lower performers up to the same level as schools with the highest degree of student achievement.[10] The government gives schools a grade of A, B, C, or D based on the quality of their infrastructure and student performance. It then uses several programs to facilitate the exchange of staff and ideas among schools at different levels. One program pairs high‐performing districts with low‐performing districts to share education development plans, curricula, teaching materials, and best practices. Another strategycalled commissioned administrationinvolves temporary contracts between schools to exchange both administrators and experienced teachers. In addition to these approaches, the government sets a minimum level of spending for schools and transfers public funds to indigent districts to provide them with assistance to reach this level.

The notion that the very best can help the weak requires a sense of solidarity. This solidarity may falter in environments in which hospitals and physicians are in cutthroat competition. Though there will always be some tension between competition and collaboration, in most markets, competition between hospitals does not rule out collaboration. Policies can either relieve or reinforce the natural tension between competition and collaboration. This suggests that adopting reforms with the same intent as the Shanghai system is still possible in healthcare, especially through physician and other provider networks. The healthcare workforce has a rich history of cross‐organizational collaboration through mentorships, the publication of research, and participation in continuing medical education courses. The Centers for Medicare and Medicaid Services' Hospital Engagement Networks, a program in which leading organizations have helped to disseminate interventions to reduce hospital acquired conditions, are an example of this approach. Quality collaborativesgroups of providers who collaborate across institutions to identify problems and best practices for improvementhave similarly shown great promise.[11] Similar approaches have been used by the Institute for Healthcare Improvement in many of their quality improvement initiatives.

Such collaboration‐based programs could be harnessed and tied to financial incentives for quality improvement. For instance, top‐performing hospitals could be incentivized to participate in a venue where they share their best practices with the lower performers in their field. Low performers, in turn, could be provided with financial assistance to implement the appropriate changes. Over time, financial assistance could be made contingent on quality improvements. By providing physicians and other providers with examples of what success looks like and assisting them with garnering the resources to reach this level, improvement would not only be incentivized, it might also become more tangible.

Although some hospitals and physicians may welcome changes to incentive systems, implementation of collaboration‐based programs would not be possible without a facilitator that is willing to underwrite program costs, provide financial incentivizes to providers, and develop a platform for collaboration. Large insurers are the most likely group to have the financial resources and widespread network to develop such programs, but that does not mean that they would be willing to experiment with this approach. This may especially be the case if cost savings and measurable improvements in quality are not immediate. Even though the results of collaboration‐based efforts have been promising, the implementation of these programs has been limited, and adoption in different contexts may not yield the same results. Collaboration‐based programs that have already shown success can serve as models, but they may need significant adaptations to meet the needs of providers in a given area.

Despite its promise, collaboration‐based strategies alone will not be enough to improve certain aspects of quality and value. Although providing physicians with knowledge on how to reduce unnecessary care, for example, could help limit overutilization, it is not sufficient to overcome the incentives of fee‐for‐service payment. In this case, broader payment reform and population‐based accountability can be paired with programs to encourage collaboration. For instance, the Blue Cross and Blue Shield of Massachusetts' Alternative Contract has used a combination of technical assistance, shared savings, and large quality bonuses to improve quality and reduce medical spending growth.[12] Collaboration‐based strategies should be seen as a complement to these broad, thoughtful reforms and a substitute for narrow incentives that encourage myopia and destructive competition.

Evidence from education and healthcare shows that penalizing the worst and rewarding the best will not shift the bell curve of performance. Such approaches are more likely to entrench and expand disparities. Instead, policy should encourage and incentivize collaboration to expand best practices that improve patient outcomes. Lessons from education provide both cautionary tales and novel solutions that might improve healthcare.

Disclosure: Nothing to report.

We believe that patients, families, and other stakeholders can provide meaningful contributions throughout the research process. Involvement of a diverse group of stakeholders is also encouraged by the Patient Centered Outcomes Research Institute (PCORI), which emphasizes research focused on patient‐ and family‐centered outcomes.[1] Patient and family engagement in healthcare, however, has generally focused on children and adults with chronic conditions.[1, 2] Engagement of families of children with serious acute illnesses is infrequent, and no studies have documented the feasibility or acceptability of different methods of family engagement.[3] Furthermore, stakeholders, such as nurses, may participate in study execution but rarely receive opportunities to inform the research process. In this Perspective, we describe our experiences with family engagement using a novel approach of serial, focused, short‐term engagement of stakeholders.

PRESTUDY WORK

In 2012, our institution introduced a nurse‐led transitional home‐visit program, an approach associated with reduced healthcare utilization in adults.[4] Patients hospitalized for acute illness received a 1‐time transitional home visit 24 to 72 hours after hospital discharge. We formed a multidisciplinary team, consisting of physicians, nurse scientists, home healthcare (HHC) nursing staff, patient families, and research staff to design a mixed‐methods study of the transitional home visit, which was funded by PCORI in 2014. This study, the Hospital‐to‐Home Outcomes (H2O) study, has 3 aims: (1) identify barriers to successful transitions home and outcomes of such transitions that are meaningful to families, (2) optimize the transitional home visits to address family‐identified barriers and outcomes, and (3) determine the efficacy of transitional home visits through a randomized control trial.

Two parents joined the study team during study development. Both had children hospitalized for acute illnesses, received a transitional home visit, and participated in a pilot focus group to provide insight into barriers families encounter during care transitions. These parents made valuable contributions, including recommending strategies for patient enrollment and retention. They also committed to participating in regularly scheduled study meetings and ad hoc discussions. However, feedback from the pilot focus group also highlighted a potential research engagement challenge; specifically, once the acute illness resolved, many families were primarily focused on the return to their normal routine and may not be easily engaged in research.

Based on family input, we included several mechanisms to engage caregivers of children with acute illness in the study design of H2O. Each design element allowed families and other stakeholders to contribute via short‐term focused approaches (eg, focus groups, phone surveys). These short‐term interactions drove iterative changes in study processes and approaches, including how to measure outcomes important to families. Rather than a small group of stakeholders making a series of recommendations over a long period of time, we had dozens of individual stakeholders make a few recommendations apiece that were quickly implemented and subsequently tested via feedback from the next few stakeholders (Figure 1).

Figure 1

PATIENT AND STAKEHOLDER ENGAGEMENT IN THE H2O STUDY

Having the short‐term, focused engagement strategy built into the study proved beneficial, when the 2 parents who were part of the initial design team and had planned to participate longitudinally were no longer able to participate. Over time, their circumstances changed. One parent moved out of the area to pursue a professional opportunity, and the second parent became increasingly difficult to reach and unable to join planned study meetings, a situation anticipated by the pilot focus group participants. These 2 instances illustrate challenges with long‐term engagement of families in research when the potential primary driver of their engagement, their child's acute illness, has resolved.

Short‐Term Focused Engagement Via Phone Calls With Parents/Caregivers

To continuously improve study processes and the transitional home visit during the second aim of H2O, we relied on short‐term focused engagement from 2 stakeholder groups, families and field nurses. We completed 107 phone calls with families who received a transitional home visit during the visit optimization period. These calls, completed 3 to 7 days after the visit, assessed parental perceptions of the effect of recent visit modifications through a standardized survey documented in an electronic database. These data were utilized in plan‐do‐study‐act cycles,[5] every 1 to 2 weeks, to determine if additional modifications to the visits were necessary. A cycle ended when the calls no longer provided new information. The questions asked on the calls also changed over time as different interventions were tested.

As an example, in aim 1, families highlighted the lack of clarity of discharge instructions, particularly regarding when and why to return for medical care. Thus, we developed condition‐specific red flag reminder cards to be shared at transitional home visits to help families remember and recognize concerning signs and symptoms and understand when additional evaluation may be warranted (Figure 2). Families in postvisit calls endorsed the concept of red flags, but sometimes preferred electronic rather than paper versions of the red flag cards to facilitate sharing with family members. Thus, we tested and refined texting the red flag information to families. Subsequent calls strongly supported this practice, so we will continue to use it during the third aim, the randomized trial of the transitional home visit.

Figure 2

The remaining calls (N=72) were completed 14 days after the visit to mirror the time frame for follow‐up calls in the planned randomized trial. These calls allowed us to test measurement of family‐identified outcomes and determine their usability in the trial. We used family feedback to shorten the survey and reorder questions. We also used feedback from these calls to develop an optimal call‐back strategy to maximize family contacts.

ONGOING H2O WORK AND CONCLUSION

The third aim, with ongoing patient enrollment, involves a randomized trial to determine the efficacy of the revised transitional home visit compared with standard of care as measured by subsequent healthcare utilization and outcomes suggested in aim 1 and refined during aim 2, such as parental coping, stress, and confidence in care. We have engaged 1 parent to provide longitudinal feedback during regularly scheduled meetings.

We believe that our short‐term, focused engagement strategies have allowed integration of the invaluable perspective of families and other stakeholders into our research questions, intervention design, outcome measurement, and study execution. Our approach combined short‐term engagement from many stakeholders, blending qualitative techniques with rapid‐cycle implementation methods to quickly react to stakeholder input. Given the challenge of sustaining longitudinal engagement of families in research focused on acute care questions, and the tendency for many families interested in such engagement to be well versed in the care system due to chronic conditions, we propose this short‐term focused approach to include the unique viewpoints of families and patients whose care experience is confined to an acute period. Similarly, we propose that such an approach can efficiently include and rapidly react to input from other hard‐to‐engage key stakeholders such as field nurses.

We believe that patients, families, and other stakeholders can provide meaningful contributions throughout the research process. Involvement of a diverse group of stakeholders is also encouraged by the Patient Centered Outcomes Research Institute (PCORI), which emphasizes research focused on patient‐ and family‐centered outcomes.[1] Patient and family engagement in healthcare, however, has generally focused on children and adults with chronic conditions.[1, 2] Engagement of families of children with serious acute illnesses is infrequent, and no studies have documented the feasibility or acceptability of different methods of family engagement.[3] Furthermore, stakeholders, such as nurses, may participate in study execution but rarely receive opportunities to inform the research process. In this Perspective, we describe our experiences with family engagement using a novel approach of serial, focused, short‐term engagement of stakeholders.

PRESTUDY WORK

In 2012, our institution introduced a nurse‐led transitional home‐visit program, an approach associated with reduced healthcare utilization in adults.[4] Patients hospitalized for acute illness received a 1‐time transitional home visit 24 to 72 hours after hospital discharge. We formed a multidisciplinary team, consisting of physicians, nurse scientists, home healthcare (HHC) nursing staff, patient families, and research staff to design a mixed‐methods study of the transitional home visit, which was funded by PCORI in 2014. This study, the Hospital‐to‐Home Outcomes (H2O) study, has 3 aims: (1) identify barriers to successful transitions home and outcomes of such transitions that are meaningful to families, (2) optimize the transitional home visits to address family‐identified barriers and outcomes, and (3) determine the efficacy of transitional home visits through a randomized control trial.

Two parents joined the study team during study development. Both had children hospitalized for acute illnesses, received a transitional home visit, and participated in a pilot focus group to provide insight into barriers families encounter during care transitions. These parents made valuable contributions, including recommending strategies for patient enrollment and retention. They also committed to participating in regularly scheduled study meetings and ad hoc discussions. However, feedback from the pilot focus group also highlighted a potential research engagement challenge; specifically, once the acute illness resolved, many families were primarily focused on the return to their normal routine and may not be easily engaged in research.

Based on family input, we included several mechanisms to engage caregivers of children with acute illness in the study design of H2O. Each design element allowed families and other stakeholders to contribute via short‐term focused approaches (eg, focus groups, phone surveys). These short‐term interactions drove iterative changes in study processes and approaches, including how to measure outcomes important to families. Rather than a small group of stakeholders making a series of recommendations over a long period of time, we had dozens of individual stakeholders make a few recommendations apiece that were quickly implemented and subsequently tested via feedback from the next few stakeholders (Figure 1).

Figure 1

PATIENT AND STAKEHOLDER ENGAGEMENT IN THE H2O STUDY

Having the short‐term, focused engagement strategy built into the study proved beneficial, when the 2 parents who were part of the initial design team and had planned to participate longitudinally were no longer able to participate. Over time, their circumstances changed. One parent moved out of the area to pursue a professional opportunity, and the second parent became increasingly difficult to reach and unable to join planned study meetings, a situation anticipated by the pilot focus group participants. These 2 instances illustrate challenges with long‐term engagement of families in research when the potential primary driver of their engagement, their child's acute illness, has resolved.

Short‐Term Focused Engagement Via Phone Calls With Parents/Caregivers

To continuously improve study processes and the transitional home visit during the second aim of H2O, we relied on short‐term focused engagement from 2 stakeholder groups, families and field nurses. We completed 107 phone calls with families who received a transitional home visit during the visit optimization period. These calls, completed 3 to 7 days after the visit, assessed parental perceptions of the effect of recent visit modifications through a standardized survey documented in an electronic database. These data were utilized in plan‐do‐study‐act cycles,[5] every 1 to 2 weeks, to determine if additional modifications to the visits were necessary. A cycle ended when the calls no longer provided new information. The questions asked on the calls also changed over time as different interventions were tested.

As an example, in aim 1, families highlighted the lack of clarity of discharge instructions, particularly regarding when and why to return for medical care. Thus, we developed condition‐specific red flag reminder cards to be shared at transitional home visits to help families remember and recognize concerning signs and symptoms and understand when additional evaluation may be warranted (Figure 2). Families in postvisit calls endorsed the concept of red flags, but sometimes preferred electronic rather than paper versions of the red flag cards to facilitate sharing with family members. Thus, we tested and refined texting the red flag information to families. Subsequent calls strongly supported this practice, so we will continue to use it during the third aim, the randomized trial of the transitional home visit.

Figure 2

The remaining calls (N=72) were completed 14 days after the visit to mirror the time frame for follow‐up calls in the planned randomized trial. These calls allowed us to test measurement of family‐identified outcomes and determine their usability in the trial. We used family feedback to shorten the survey and reorder questions. We also used feedback from these calls to develop an optimal call‐back strategy to maximize family contacts.

ONGOING H2O WORK AND CONCLUSION

The third aim, with ongoing patient enrollment, involves a randomized trial to determine the efficacy of the revised transitional home visit compared with standard of care as measured by subsequent healthcare utilization and outcomes suggested in aim 1 and refined during aim 2, such as parental coping, stress, and confidence in care. We have engaged 1 parent to provide longitudinal feedback during regularly scheduled meetings.

We believe that our short‐term, focused engagement strategies have allowed integration of the invaluable perspective of families and other stakeholders into our research questions, intervention design, outcome measurement, and study execution. Our approach combined short‐term engagement from many stakeholders, blending qualitative techniques with rapid‐cycle implementation methods to quickly react to stakeholder input. Given the challenge of sustaining longitudinal engagement of families in research focused on acute care questions, and the tendency for many families interested in such engagement to be well versed in the care system due to chronic conditions, we propose this short‐term focused approach to include the unique viewpoints of families and patients whose care experience is confined to an acute period. Similarly, we propose that such an approach can efficiently include and rapidly react to input from other hard‐to‐engage key stakeholders such as field nurses.

We believe that patients, families, and other stakeholders can provide meaningful contributions throughout the research process. Involvement of a diverse group of stakeholders is also encouraged by the Patient Centered Outcomes Research Institute (PCORI), which emphasizes research focused on patient‐ and family‐centered outcomes.[1] Patient and family engagement in healthcare, however, has generally focused on children and adults with chronic conditions.[1, 2] Engagement of families of children with serious acute illnesses is infrequent, and no studies have documented the feasibility or acceptability of different methods of family engagement.[3] Furthermore, stakeholders, such as nurses, may participate in study execution but rarely receive opportunities to inform the research process. In this Perspective, we describe our experiences with family engagement using a novel approach of serial, focused, short‐term engagement of stakeholders.

PRESTUDY WORK

In 2012, our institution introduced a nurse‐led transitional home‐visit program, an approach associated with reduced healthcare utilization in adults.[4] Patients hospitalized for acute illness received a 1‐time transitional home visit 24 to 72 hours after hospital discharge. We formed a multidisciplinary team, consisting of physicians, nurse scientists, home healthcare (HHC) nursing staff, patient families, and research staff to design a mixed‐methods study of the transitional home visit, which was funded by PCORI in 2014. This study, the Hospital‐to‐Home Outcomes (H2O) study, has 3 aims: (1) identify barriers to successful transitions home and outcomes of such transitions that are meaningful to families, (2) optimize the transitional home visits to address family‐identified barriers and outcomes, and (3) determine the efficacy of transitional home visits through a randomized control trial.

Two parents joined the study team during study development. Both had children hospitalized for acute illnesses, received a transitional home visit, and participated in a pilot focus group to provide insight into barriers families encounter during care transitions. These parents made valuable contributions, including recommending strategies for patient enrollment and retention. They also committed to participating in regularly scheduled study meetings and ad hoc discussions. However, feedback from the pilot focus group also highlighted a potential research engagement challenge; specifically, once the acute illness resolved, many families were primarily focused on the return to their normal routine and may not be easily engaged in research.

Based on family input, we included several mechanisms to engage caregivers of children with acute illness in the study design of H2O. Each design element allowed families and other stakeholders to contribute via short‐term focused approaches (eg, focus groups, phone surveys). These short‐term interactions drove iterative changes in study processes and approaches, including how to measure outcomes important to families. Rather than a small group of stakeholders making a series of recommendations over a long period of time, we had dozens of individual stakeholders make a few recommendations apiece that were quickly implemented and subsequently tested via feedback from the next few stakeholders (Figure 1).

Figure 1

PATIENT AND STAKEHOLDER ENGAGEMENT IN THE H2O STUDY

Having the short‐term, focused engagement strategy built into the study proved beneficial, when the 2 parents who were part of the initial design team and had planned to participate longitudinally were no longer able to participate. Over time, their circumstances changed. One parent moved out of the area to pursue a professional opportunity, and the second parent became increasingly difficult to reach and unable to join planned study meetings, a situation anticipated by the pilot focus group participants. These 2 instances illustrate challenges with long‐term engagement of families in research when the potential primary driver of their engagement, their child's acute illness, has resolved.

Short‐Term Focused Engagement Via Phone Calls With Parents/Caregivers

To continuously improve study processes and the transitional home visit during the second aim of H2O, we relied on short‐term focused engagement from 2 stakeholder groups, families and field nurses. We completed 107 phone calls with families who received a transitional home visit during the visit optimization period. These calls, completed 3 to 7 days after the visit, assessed parental perceptions of the effect of recent visit modifications through a standardized survey documented in an electronic database. These data were utilized in plan‐do‐study‐act cycles,[5] every 1 to 2 weeks, to determine if additional modifications to the visits were necessary. A cycle ended when the calls no longer provided new information. The questions asked on the calls also changed over time as different interventions were tested.

As an example, in aim 1, families highlighted the lack of clarity of discharge instructions, particularly regarding when and why to return for medical care. Thus, we developed condition‐specific red flag reminder cards to be shared at transitional home visits to help families remember and recognize concerning signs and symptoms and understand when additional evaluation may be warranted (Figure 2). Families in postvisit calls endorsed the concept of red flags, but sometimes preferred electronic rather than paper versions of the red flag cards to facilitate sharing with family members. Thus, we tested and refined texting the red flag information to families. Subsequent calls strongly supported this practice, so we will continue to use it during the third aim, the randomized trial of the transitional home visit.

Figure 2

The remaining calls (N=72) were completed 14 days after the visit to mirror the time frame for follow‐up calls in the planned randomized trial. These calls allowed us to test measurement of family‐identified outcomes and determine their usability in the trial. We used family feedback to shorten the survey and reorder questions. We also used feedback from these calls to develop an optimal call‐back strategy to maximize family contacts.

ONGOING H2O WORK AND CONCLUSION

The third aim, with ongoing patient enrollment, involves a randomized trial to determine the efficacy of the revised transitional home visit compared with standard of care as measured by subsequent healthcare utilization and outcomes suggested in aim 1 and refined during aim 2, such as parental coping, stress, and confidence in care. We have engaged 1 parent to provide longitudinal feedback during regularly scheduled meetings.

We believe that our short‐term, focused engagement strategies have allowed integration of the invaluable perspective of families and other stakeholders into our research questions, intervention design, outcome measurement, and study execution. Our approach combined short‐term engagement from many stakeholders, blending qualitative techniques with rapid‐cycle implementation methods to quickly react to stakeholder input. Given the challenge of sustaining longitudinal engagement of families in research focused on acute care questions, and the tendency for many families interested in such engagement to be well versed in the care system due to chronic conditions, we propose this short‐term focused approach to include the unique viewpoints of families and patients whose care experience is confined to an acute period. Similarly, we propose that such an approach can efficiently include and rapidly react to input from other hard‐to‐engage key stakeholders such as field nurses.