Hospitalists’ Perceptions of Pediatric Mental Health Boarding: Quality of Care and Moral Distress Free

In March 2021, we conducted a web-based survey of members of Pediatric Research in Inpatient Settings (PRIS), a voluntary pediatric hospital medicine research network that represents children’s and community hospitals in the United States and Canada.¹⁵  The survey was distributed via REDCap^16,17  to both PRIS site leads (primary contact persons at each hospital) and the PRIS membership (PRIS participants who are not site leads). Surveys were sent to US-based hospitalists only because of differences in Canadian federal regulations and oversight of mental health care. Up to 3 reminder e-mails (through April 2021) were sent to encourage responses. No participation incentives were provided.

Survey questions for this analysis (Appendix 1) focused on hospital and demographic characteristics, perceived mental health care quality, and moral distress pertaining to the care of boarding patients. A hospital-level analysis of boarding frequency, duration, and hospital resources has been previously published.¹²  Boarding was defined using the Joint Commission definition provided previously,¹¹  and respondents were asked to consider boarding in children and adolescents aged <18 years following medical clearance.

Quality of care and moral distress survey items included closed- and open-ended questions. Open-ended questions included: (1) in what ways does your hospital provide excellent quality of care to youth experiencing mental health boarding? and (2) in what ways is the quality of care for youth experiencing mental health boarding at your hospital suboptimal? Respondents were also asked if they tracked patient safety and/or quality of care measures among youth boarding at their hospital, and, if yes, to list the measures tracked.

Moral distress questions were selected from the previously validated Measure of Moral Distress for Health Care Professionals (MMD-HP),¹⁸  which asks respondents to rate both the frequency and intensity of 27 causes of moral distress, categorized as patient-, team-, and system-level factors. Recognizing the response burden incurred by asking participants to rate 27 items twice, we convened a 5-member stakeholder panel comprising a pediatric resident, chief resident, hospitalist, psychologist, and psychiatrist to select survey items most relevant to mental health boarding. Eleven items were prioritized for inclusion in our survey. Consistent with the original MMD-HP, participants were asked to rate, on a 5-point Likert scale (0 = never to 4 = very frequently), how frequently they had experienced each situation in caring for boarding youth. They were subsequently asked to rank how distressing the situations were for them (0 = never to 4 = very distressing). They were instructed to select the response of “0” (never) if they had never experienced a particular situation. Additionally, they were asked to describe, via free text response, other situations in which they felt moral distress caring for youth experiencing mental health boarding.

To optimize the survey before data collection, we conducted cognitive interviews with 9 hospitalists and chief residents at 3 geographically distinct hospitals (for diverse perspectives). This method involved administering draft survey questions and eliciting verbal feedback to ensure questions were interpreted as intended and produced comprehensive responses.^19,20  Cognitive interviews resulted in minor modifications to question wording and response options. Additionally, we asked 5 hospitalists to pilot test the survey in REDCap^16,17  to estimate the time required for survey completion and to ensure appropriate question flow and item clarity.

After ensuring deidentification of responses, open-ended survey data were analyzed in Dedoose 9.0.46 (Manhattan Beach, CA)²¹  using conventional (inductive) content analysis.^22,23  Our senior author (J.K.L.), who has led several previous peer-reviewed qualitative studies,^24–32  trained all team members in this analytic method (except S.D.F., who conducted the quantitative analysis). Applying this inductive approach, all team members read the raw data in full and devised initial open codes (reflecting content within the text perceived as relevant to our research question). We then met to develop the codebook, collectively reviewing open codes to define subconcepts/concepts emerging from the data and their associated definitions. Three authors (C.T.W., A.L.B., D.B..T) then independently conducted line-by-line coding of free text responses with all data coded by at least 2 authors; disagreements were resolved through subsequent discussion. Given the survey study design, we analyzed responses from all participants after data collection and elected not to discontinue analysis when data saturation was attained. After coding completion, team members agreed that data saturation was achieved. Three researchers (C.T.W., D.B.T., N.Y.P.) subsequently developed memos (written reflections) on the coded data. The research team convened to define emerging themes, defined as “patterned response or meaning” observed in the coded data^33,34 ; we used axial coding to group related concepts into themes. Additionally, the research team collectively summarized foci for quality improvement based on key concepts reflected within each theme.

Analysis of the MMD-HP data followed previously established procedures.¹⁸  For each of the 11 items, composite moral distress scores were calculated by multiplying the moral distress frequency score by the intensity score, resulting in a score range of 0 to 16 for each item. All quantitative analysis was conducted using R Statistical Software (v4.1.3).³⁵ 

This study was reviewed and deemed exempt by our institutional review board.

We received responses from 88 of 111 PRIS site leads or designates (79%) and 76 of 383 additional PRIS members (20%). Of these, 85 site leads (97%) and 65 PRIS members (86%) reported being familiar with their hospital’s mental health boarding processes and were included in this analysis. Survey respondents represented 85 hospitals, including 12 community hospitals (14%), 35 children’s hospitals nested within adult hospitals (41%), and 38 freestanding children’s hospitals (45%). The majority of respondents were female (71%), with a median age of 41 years (interquartile range, 37-45 years) and a median of 11 years of experience (interquartile range, 6-14) (Table 1).

A total of 89 respondents provided free text responses to open-ended questions about quality of care and moral distress. Perceptions of health care quality during boarding centered on 6 themes: (1) access to psychiatric services, (2) safety, (3) standardized workflows, (4) clinician training, (5) compassion/patient engagement, and (6) collaboration and disposition planning. Illustrative quotes summarizing high and suboptimal quality of care, and associated foci for quality improvement, are provided in Table 2. In the context of suboptimal health care quality, respondents described experiencing moral distress, particularly related to inadequate psychiatric services, prolonged boarding, and activity restrictions imposed to maintain patient safety (Table 3).

Respondents described provision of psychiatric services as a primary determinant of care quality, including psychotherapy, psychotropic medication management, behavioral de-escalation support, and safety/risk assessment. One respondent described the hospital’s positive experience, “We … have a psychiatrist who is solely dedicated to inpatient/[emergency department] consultation and has been a great point-person/champion for psychiatric care” (ID17001).

Unfortunately, respondents widely described inadequate, or no, availability of psychiatry/psychology services (including poor access/uptake of telepsychiatry), to provide boarding patients therapy or medication management. One respondent described, “We do not have psychiatrists or psychologists. We have overworked social workers who can barely see every new patient, not to mention boarding patient[s], daily” (ID28501). As a result, respondents described boarding hospitalizations as “wasted time” (ID14501) for patients presenting in “acute crisis” (ID10901), who ultimately “just sit in the hospital” (ID10205).

Several respondents expressed concern that boarding hospitalizations offered poor therapeutic benefit/value of care and invoked moral distress. They described sentiments of frustration and helplessness among inpatient clinicians, patients, and caregivers. For example, “It is difficult to have … [boarding patients] not receiving treatment for the reason they are admitted (i.e., suicidality, depression …). Parents are frustrated and patients are just waiting—not safe enough to go home but not receiving psychiatric care” (ID17503).

Pertaining to the theme of safety, respondents described “safe” patient rooms, 1:1 safety attendants, activity restrictions to prevent self-harm, and procedures to manage patient agitation.

Respondents described the value of patient rooms designed to minimize environmental safety hazards and facilitate monitoring; desired room capabilities included telemetry after ingestions and camera monitoring to give “some freedom yet continued safety” (ID19118). However, several respondents described their hospitals lacking appropriate rooms, such that “[it is] difficult to control … ligature risk” (ID14401).

Respondents described these restrictive policies as a source of moral distress: “Extreme lockdown situations are … morally distressing—children restricted to their rooms for prolonged periods (weeks, months) with no visitors and multiple guards” (ID16701). Additional sources of moral distress related to safety included staff injury by aggressive patients and physical/chemical restraint use.

Within this theme, respondents described standardized policies, protocols, and pathways for management of youth experiencing boarding. Respondents variably reported having standardized procedures for room preparation, policies outlining patient privileges, procedures for conducting suicide risk assessments, and standardized workflows for behavioral de-escalation.

Respondents particularly valued standardized processes for managing acute agitation. One respondent described need for “… a clear process for ensuring rapid and efficient assistance in the case of behavioral emergencies to both create behavioral plans when aggressive behavior is anticipated and to rapidly respond to aggressive behavior when staff are at risk” (ID17707). Further, respondents associated having a behavioral de-escalation workflow/team with “cut[ting] down on the number of medical/physical restraints” (ID17804). However, others reported lacking standardized policies and workflows, resulting in inconsistent care. One respondent described that “recommended treatments for behavioral escalation differ by psych provider and there is no routine pathway in place” (ID14301).

Just over one-quarter of PRIS site leaders (n = 21, 27%) reported tracking quality or safety measures related to mental health boarding. Reported measures are listed in Table 4.

The theme of clinician training included concepts related to previous mental health care experience and behavioral health training opportunities.

A few respondents applauded safety attendant and other staff training/experience in mental health. One respondent described the hospital’s “live 6-hour class for all behavioral health techs/1:1 staff/nurses on the care and management of the behavioral health patient in a non-psychiatric setting” (ID13001).

In contrast, respondents predominantly lamented a paucity of safety attendant and clinician behavioral health experience and/or [access to] training. One respondent stated, “The 1:1 sitters are patient care techs—they generally have not received any specific behavior/mental health training and can be generally supportive but don’t really provide a therapeutic benefit” (ID14401).

Moral distress responses pertaining to clinician training centered on respondents feeling underequipped to care for particularly escalated, mental health patients (“not knowing the best ways to handle psychotic or agitated patients” (ID16601)).

Within this theme, respondents described staff’s interactions with youth, efforts to provide wellness activities, caregiver communication, and provision of patient/family anticipatory guidance.

Respondents applauded multidisciplinary staff efforts to provide “a compassionate and supportive environment to the best of their ability” (ID14401); dedicated efforts to keep patients safe; and engagement of patients in wellness activities (e.g., “distraction or calming techniques, general entertainment” (ID10901)). One respondent described that “our sitters and child life are very involved and … provide good support to our boarders” (ID14201). In addition, respondents described dedicated efforts to provide anticipatory guidance and education for families regarding the current hospitalization, transfer logistics, inpatient psychiatry, and home safety planning.

Despite efforts to engage patients and families, several respondents described major limitations to allowable patient activities, resulting in patients “just sitting there” (ID25001), “locked in a room” (ID14716). Respondents described concerns that activity restrictions, lack of psychotherapy, and limited ability to engage patients worsened patients’ isolation and dysphoria. One respondent described, “Patients become very isolated as they are restricted from interacting with peers in a way that would not happen in such a routine way at a psychiatric hospital” (ID18801).

Moral distress relevant to this theme centered on frustration with limited opportunities for therapeutic engagement, resulting in “helplessness to improve these patients’ lives” (ID 16808), compounded by “inflexible policies that compromise care” (ID14732). One respondent recounted, “patients expressing that the care … is making them worse … due to the number of restrictions (no fresh air, minimal movement, etc.)” (ID28101).

Within this theme, respondents described interdisciplinary communication/collaboration and logistics around inpatient psychiatry bed placement.

Respondents particularly valued regular, timely communication with psychiatry services for acute behavioral concerns, daily check-ins, and disposition updates. Cited strategies included “multidisciplinary psychiatric rounds each morning” (ID12602) and “daily phone huddles” (ID18501).

Some respondents commended short, <24-hour stays and highlighted diligent efforts and “streamlined processes” (ID16801) to ensure “definitive placement ASAP” (ID14301). Strategies included “a dedicated [social worker] who coordinates inpatient placement” (ID17302) and “bed search … done multiple times a day (including weekends)” (ID17601).

Suboptimal care and moral distress for this theme centered on delays in placement (“in isolated rooms, sometimes for weeks” [ID13513]; “lack of clarity around discharge plans and [length of stay] is very difficult for families” [ID17804]). Particularly distressing to respondents were experiences of psychiatric facilities not accepting patients with comorbid medical conditions, resulting in prolonged boarding. Respondents lamented, “…even a [nasogastric] tube can delay care” (ID14501) and “autistic [children] that have nowhere to go” (ID15301).

Responses to the 11 MMD-HP items varied substantially, with values ranging from a low of 0 to a high of 4 for all items for both frequency and intensity (Fig 1). Composite scores (the product of frequency and intensity) were highest for system-level factors, followed by team- and patient-level factors. The highest frequency score (and highest composite score) was “experiencing compromised patient care due to lack of resources, equipment, and/or bed capacity,” whereas the most intense source was “participating in care that was perceived to cause unnecessary suffering or not adequately relieve pain or symptoms.”

Recommendations to improve care for youth experiencing mental health boarding have predominantly focused on upstream solutions, particularly related to expansion of outpatient mental health service funding and access.¹⁰  Although such services are essential, this study highlights opportunities for acute care hospitals to concurrently improve the care of youth experiencing mental health boarding.

Respondents in this study prioritized the need for expanded access to inpatient psychiatric services; however, given a national shortage of mental health professionals, innovative solutions may be needed to address quality of care gaps. Enhanced use of telepsychiatry and education of nonmental health clinicians represent 2 potential opportunities emerging from this study. Technology-based supports such as interfacility psychiatry consultation or video psychotherapy may increase availability of mental health services in acute care settings. Examples of larger institutions successfully providing telepsychiatry consultation to both affiliated and unaffiliated community/rural sites are published.^36–39  Video conferencing for multidisciplinary huddles/check-ins may also be an efficient means of providing psychiatric support to pediatric teams.⁴⁰ 

Mental health training needs prioritized by respondents included techniques to optimize patient-centered engagement and behavioral de-escalation skills. The former is supported by previous qualitative studies of boarding adolescents and their caregivers who desired more information regarding clinical care processes and emotional support resources.^41,42  Hospitals may consider offering de-escalation training; existing programs include those efforts offered by Crisis Prevention Institute,⁴³  Safety-Care,⁴⁴  and Pediatric Learning Solutions,⁴⁵  among others.⁴⁶  As suggested by Doupnik et al, hospitals can support training for clinician leaders by offering protected time, adjusted productivity targets, or continuing education credits to facilitate staff participation.⁴⁰ 

Hospitals may further improve care for youth experiencing boarding through the development of standardized policies, procedures, and clinical pathways, and by identifying and tracking quality and safety metrics. Such interventions are well established in hospital medicine and may improve health care quality without requiring additional personnel. In recent years, clinical pathways have been increasingly adopted in pediatric inpatient settings; studies of their use demonstrate improvements in clinical outcomes, reductions in practice variation, and reductions in hospital resource use.^47–53  A growing number of published mental health-related guideline implementation efforts and clinical pathways may serve as guides for managing these conditions in acute care settings. Published examples address suicide risk screening,^54,55  agitation/de-escalation,^56–58  somatic symptom and related disorders,^59,60  and disposition/care coordination.⁶¹  Collaboration between psychiatry and pediatrics, within or across hospitals, may support implementation of these tools.

Respondents in this study reported substantial moral distress, particularly related to lack of resources, personnel, and administrative support, illustrating an adverse impact on clinicians who perceive suboptimal quality of care. Although the MMD-HP does not have norms to define “high” moral distress, mean scores for 8 of the 11 items in this study exceeded the overall mean composite score (3.56) for physician respondents in the MMD-HP development and validation paper,¹⁸  as well as in a previous sample of US pediatric hospitalists (3.07).⁶²  Existing literature indicates a strong correlation between clinician moral distress, burnout, and provider turnover,^62–64  with high-quality health care provision a leading determinant of physician satisfaction.^65,66  More than 2 years into the COVID-19 pandemic, hospitals face emotional exhaustion among workers,⁶⁷  critical staff shortages,^68,69  and a national emergency in mental health.⁷⁰  It is possible that improving the quality of care for youth experiencing boarding would improve clinician experience as well as pediatric health outcomes.

Our results should be interpreted in the context of this study’s strengths and limitations. Although we received an excellent response from PRIS site leaders or designates, the PRIS member response rate was low and, thus, our results may be susceptible to nonresponse bias. Our study reflects the experiences of a specific group of participants, which may limit generalizability. Because we surveyed members of a hospital medicine research network, we did not elicit input from additional clinicians who also care for this population. In addition, respondents were predominantly academic hospitalists from children’s hospitals, with relatively low community hospital representation. Although we included open-ended survey questions to characterize the “why and how” of moral distress and health care quality during boarding, our survey design did not allow us to probe more deeply on these issues.⁷¹  Finally, given the response burden associated with completing the full MMD-HP, we used a multidisciplinary stakeholder prioritization approach to select survey items most applicable to mental health boarding. Correspondingly, the dimensions of moral distress evaluated quantitatively in this study are limited. Despite these limitations, our analysis provides insight into the quality of care provided during boarding across 85 geographically diverse hospitals in the United States, for which little published literature exists.

In conclusion, hospitalists described deficits in quality of care and high levels of moral distress in caring for youth experiencing mental health boarding, particularly related to health system factors. This study identifies several opportunities to support youth and potentially reduce moral distress among clinicians by investing in mental health personnel and/or technology, standardizing workflows, supporting interdisciplinary collaboration, and enhancing clinician training. Recognizing the national shortage of mental health professionals, dissemination of innovative solutions is needed.