Pediatric Mental Health Boarding

Given the heterogeneity of published research, we applied a systematic scoping review approach following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines.¹⁵  We sought to answer 2 questions:

• 1. What is the epidemiology of pediatric mental health boarding in the United States with respect to frequency, setting, duration, and associated risk factors?

• 2. What patient or health system outcomes are associated with pediatric mental health boarding in the United States?

We searched Medline and PsycINFO in partnership with a reference librarian to identify all pediatric mental health boarding–relevant original research. We included research in which authors described ≥1 of the following: (1) frequency of pediatric mental health boarding; (2) duration of pediatric mental health boarding; (3) processes, outcomes, or risk factors associated with pediatric mental health boarding; and (4) child and adolescent experiences of care during mental health boarding. We defined mental health boarding using a definition adapted from The Joint Commission: the practice of holding patients in the ED, an inpatient medical unit, or another temporary location after the patient has been medically cleared and a decision to admit or transfer the patient for psychiatric care has been made.⁹  We included all studies that described this phenomenon, whether or not the term “boarding” was used. We did not exclude studies if authors used an alternative mental health boarding definition but documented study-specific boarding definitions.

Our final search was conducted on November 26, 2019 and included all studies indexed in Medline and PsycINFO with no date restrictions. We sought to identify all studies that were focused on mental health, behavioral health, or psychiatric boarding in children and adolescents ≤21 years of age in the United States. Studies were excluded if the authors reported no pediatric-specific results (operationalized as studies conducted at children’s hospitals, pediatric EDs, or presenting data specific to youth ≤21 years of age) or if they reported no results specific to mental health boarding (as distinct from boarding for other medical or surgical reasons). We limited our analyses to original research articles, excluding commentaries and non-English publications. Studies in which authors reported ED or inpatient length of stay only, without an associated mental health boarding definition, were excluded given our inability to distinguish between the length of stay spent evaluating and/or managing medical or psychiatric issues and time spent boarding (after the patient has been medically cleared and a decision to admit or transfer the patient for psychiatric care has been made). Quantitative, qualitative, and mixed-methods studies were retained. In addition, we reviewed all of the references cited by articles meeting our inclusion criteria to identify additional potential articles. Please refer to the Supplemental Information for our complete search strategies.

Two authors independently reviewed each abstract, and full-text articles were reviewed if ≥1 author(s) determined that the abstract met inclusion criteria. Two authors then independently reviewed each full-text article to determine if the article met inclusion criteria. Disagreements between reviewers were resolved through consensus after discussion and review with 3 authors. Reasons for exclusion were recorded.

We used a standardized electronic form to summarize information regarding study design, study population, setting, boarding definitions, boarding duration, patient and system factors associated with mental health boarding, processes of care during mental health boarding, and outcomes associated with mental health boarding. To minimize bias, information from each full-text article was extracted independently by 2 authors and differences in extraction were resolved through discussion and re-review. To critically appraise the quality of included publications, 3 authors independently applied a modified Newcastle-Ottawa Scale for the quality assessment of cohort studies, with disagreements resolved through discussion and re-review.^16–18  We then converted the Newcastle-Ottawa Scale numeric scores to the Agency for Healthcare Research and Quality classifications of “poor,” “fair,” or “good” by applying a previously published algorithm.^19,20  Risk of bias was not assessed for qualitative studies given the absence of comparable evaluation tools. After data extraction, we summarized key concepts relevant to our study objectives using an adapted version of the Webster and Watson²¹  concept matrix.

After duplicates were removed, 222 citations were identified by using the above-described search strategy. On the basis of review of the study title and abstract alone, 199 were excluded, leaving 23 full-text articles for review. Of these, 12 were excluded, including 9 that did not report any results specific to children or adolescents and 3 that did not report any frequencies, processes, outcomes, risk factors, and/or experiences of care related to mental health boarding. Correspondingly, 11 articles were included in this review (Fig 1).

Characteristics of each study’s design, target population, and research setting are summarized in Table 1. Of the 11 included studies, 10 were retrospective analyses, either chart reviews or analyses of administrative data, and 1 was a qualitative study of interviews with youth boarding at a children's hospital. Sample sizes ranged from 27 to 44 328 children and adolescents. Nine of the 11 studies were analyses of data from single hospitals and 2 were multisite studies. All of the single-site studies were conducted at children’s hospitals or general hospitals with dedicated pediatric EDs; all of these hospitals were located in urban or suburban settings. Of the 9 studies for which we could evaluate the risk of bias, 4 were of good quality and 5 were found to be of poor quality, all because of failure to control for, or specify approach to controlling for, confounding factors in the analysis (Supplemental Table 3).

The definitions of mental health boarding applied in included studies varied substantially (Table 2). In 3 of the 11 studies, authors defined boarding using time-based thresholds by examining time elapsed from ED arrival time until ED checkout. Each of these studies used different time thresholds to define boarding, ranging from 6 to 24 hours.^26,28,30  Wharff et al³¹  applied a hybrid definition that included both a time-based threshold (a length of stay of >12 hours in the ED) or documentation of patient boarding in the medical chart. In 5 of the remaining 7 studies, the authors defined boarding on the basis of transfer to an inpatient pediatric medical unit after medical clearance while awaiting an inpatient psychiatric bed.^{23,25,27,29,32}  Lastly, Campbell and Pierce²²  and Conrad et al²⁴  evaluated boarding in the ED only, defining boarding as the time from disposition decision or medical clearance in the ED until admission or transfer.

Patient-level pediatric mental health boarding times were reported in 7 studies. The average ED boarding times ranged from 5 to 41 hours, whereas median inpatient boarding times ranged from 2 to 3 days. Boarding in the ED was reported to occur in 23% to 58% of study-eligible youth with mental or behavioral health concerns, whereas inpatient boarding rates ranged from a low of 26% to a high of 49%.^23,26 

In 6 studies, authors reported associations between patients’ sociodemographic and clinical characteristics and mental health, with inconsistent findings.^{22,23,26–28,31}  Younger age was associated with boarding in 3 studies,^23,27,28  whereas payer type was associated with boarding in 2 studies.^26,27  With respect to clinical characteristics, the only factor associated with boarding in more than 1 study was presentation to hospital with suicidal and/or homicidal ideation.^26,27 

In 3 studies, authors examined systemic factors associated with pediatric mental health boarding.^26,27,31  Of these, 2 reported an increased likelihood of boarding among patients presenting on a weekend^27,31  and 2 reported increased risk of boarding during school (nonsummer) months.^26,27  Wharff et al³¹  reported that patients presenting overnight had 6.2 times the odds of boarding as those presenting from 8:00 am to 12:00 pm, whereas significant time-of-day effects were not found in the analysis by Hoffmann et al.²⁶ 

In 4 studies, authors reported processes or outcomes associated with pediatric mental health boarding beyond boarding duration and frequency.^23–25,32  Claudius et al²³  examined rates of counseling documentation and psychiatric medication receipt among their hospital's population of boarding youth, finding that only 6.1% of youth admitted for isolated psychiatric reasons had documented receipt of individual or family counseling, whereas 20.1% had received psychiatric medications (relative to 53.3% who had received psychiatric medications before hospitalization). Gallagher et al²⁵  reported considerably higher rates of receipt of psychosocial supports during inpatient boarding, including (1) psychoeducation (received by 91.1% of youth), (2) psychotherapy (received by 87.2% of youth), (3) psychotropic medication administration (received by 51.0% of youth), and (4) implementation of behavioral health plans (received by 27.5% of youth). Outcomes examined within the context of these process measures included (1) receipt of physical restraints or emergency intramuscular doses of psychotropic medication during admission (received by 6.4% of youth) and (2) improvement in a standardized measure of global functioning from hospital admission to discharge (the Clinical Global Impression scale), finding that 33% of youth had improvements in their functioning during the boarding period.²⁵ 

Beyond these clinical process and outcome measures, Claudius et al²³  examined inpatient hospital costs associated with boarding, estimating an average inpatient hospital cost of $4269 per boarded patient. Gallagher et al²⁵  examined readmission rates within their 1-year study period, finding that 21% of youth experienced ≥2 hospital readmissions and boarding episodes during their 1-year study period. In their analysis of pediatric mental health boarding in the ED, Conrad et al²⁴  examined the proportion of patients who left their ED without being seen (after check-in but before being placed in a bed to initiate medical care). They found that the median daily percentage of patients who left without being seen was significantly correlated with the daily number of mental health boarding hours and with the daily number of youth presenting with behavioral or mental health chief complaints.²⁴  In their qualitative analysis of adolescents’ perspectives and experiences, Worsley et al³²  described several themes regarding processes of care valued by adolescents. The remaining studies did not report processes or outcome measures beyond boarding time and/or duration.

We identified 11 studies in this comprehensive review of research examining mental health boarding of youth in the United States, the majority of which were retrospective analyses conducted at single hospitals. Of the single-center studies, all were conducted in urban or suburban children’s hospitals or hospitals with dedicated pediatric EDs. Notably absent were any national studies of pediatric mental health boarding prevalence or duration, studies conducted in general or community hospitals where the vast majority of children with behavioral and mental health concerns present for care,^33,34  and prospective studies implementing and evaluating programs or interventions to support youth during the mental health boarding period. The frequency of mental health boarding after medical stabilization in general pediatric or ICU, another factor associated with boarding in our clinical experience, was not quantified by the authors of any of the included studies.

Our ability to precisely quantify the national burden of pediatric mental health boarding is hampered by the use of heterogeneous boarding definitions in studies published to date and the absence of studies conducted outside of children’s hospitals and pediatric EDs. Despite this, it is clear that youth awaiting psychiatric admission spend considerable time in EDs, with a quarter to half of youth requiring inpatient mental healthcare boarding on inpatient medical units for an average of 2 to 3 days before transfer for psychiatric care. These numbers may underestimate the national prevalence of boarding given that several included studies were conducted a decade ago and rural and general hospitals are underrepresented. Despite these limitations, we can extrapolate from pediatric hospitalization data and our study results to estimate the national frequency of mental health boarding. Recognizing that >200 000 youth are admitted to acute care hospitals in the United States each year with primary behavioral or mental health concerns (excluding psychiatric facilities),³³  40 000 to 66 000 of these 200 000 hospital admissions may represent mental health boarding stays. The small body of literature and absence of diagnostic or billing codes for mental health boarding prevents us from making more precise estimates. Even so, this estimate would indicate that mental health boarding is one of the most common reasons for pediatric hospital admission in the United States.³⁴ 

The term “reverse triage” has been used to describe the prioritization of inpatient psychiatric care for patients deemed to have less emergent mental health problems, such that youth most in need of inpatient psychiatric care are least likely to receive it and most likely to board.²⁷  Although only 6 studies conducted to date have examined risk factors for pediatric boarding,^{22,23,26–28,31}  2 of these studies found that youth with suicidal or homicidal ideation were at greatest risk of mental health boarding. However, the small number of studies published to date indicate a need for more robust research on the subject. Notably absent from the literature are descriptions of boarding experiences for youth with other established risk factors for behavioral and mental health disorders, including rural residence³⁵ ; lesbian, gay, bisexual, and queer sexual orientation^36,37 ; poverty³⁸ ; andsubstance use.³⁹  Given the dearth of national and multisite studies, we are also unable to comment on systemic factors associated with mental health boarding beyond time of presentation to hospital, such as geographic region, hospital type (children’s versus nonchildren’s hospitals), or regional outpatient mental and behavioral health resources.

Several studies included in this review provide valuable insights into youth experiences of care during boarding and indicate that psychosocial service delivery is highly variable.^23–25,32  Correspondingly, no research published to date has been focused on the development, implementation, or evaluation of mental or behavioral health interventions for boarding youth. National recommendations to improve the care of youth experiencing mental or behavioral health crises have been largely focused on upstream solutions, such as increasing outpatient mental health services, addressing mental health provider shortages by using loan repayment programs, and expanding community crisis resources.⁴⁰  Although such interventions are clearly needed, there is a concurrent need for programs and resources to support the large number of youth who are currently boarding in EDs and inpatient units. A small number of studies have revealed that telepsychiatry services and dedicated mental health crisis teams are effective in reducing length of stay among children presenting to EDs with behavioral or mental health concerns.^41–44  The extent to which such services, or other novel programs, decrease rates of mental health boarding or improve patients’ experience of care while they board warrant study. In addition, cost-benefit analyses are needed to evaluate the relative benefits and costs of boarding alternatives and interventions and to inform health policy.

Our findings should be interpreted in light of this study’s strengths and limitations. First, although the term boarding has been defined by national organizations and described in national policy statements, it is possible that our search omitted studies that used other terms to describe this clinical phenomenon. Relatedly, the term boarding is not a medical subject heading in the National Library of Medicine databases. We aimed to mitigate this limitation by searching 2 national databases and manually reviewing the reference lists of all articles included in our review. Second, by including eligible articles that applied the term boarding and not restricting our study to articles that followed our Joint Commission–adapted definition, our findings reflect heterogeneous periods of clinical observation. However, given the small number of studies published on this topic area to date, we chose to include all such studies and demonstrate the variable application of the term boarding in published literature to date. Importantly, for those articles in which authors applied time-based definitions of mental health boarding, we are unable to differentiate between time spent awaiting care in the ED, receiving medical or psychiatric evaluations, or after medical clearance and the decision to admit the patient for psychiatric care. Related to this, our exclusion of articles that were focused specifically on ED length of stay (without reference to boarding) may have excluded thematically relevant work.^45,46  Finally, our ability to draw conclusions is limited by the relatively small number of studies published to date, the small sample sizes reflected therein, and their risk of bias as determined by using a standardized quality assessment tool.

Pediatric mental health boarding is a prevalent clinical phenomenon with a correspondingly small evidence base describing its frequency, contributing factors, and patient experiences of care. National initiatives representing diverse hospital types and geographic regions are needed to further quantify the frequency of pediatric mental health boarding, while programs and clinical interventions to improve patient experiences while they board warrant development, evaluation, and dissemination. Our findings may be used by clinicians, hospital administrators, policy makers, and funders to justify improvements in our systems of care for youth presenting to the hospital with mental and behavioral health crises.

ED

emergency department