Family Outcomes After the Pediatric Intensive Care Unit: A Scoping Review

Background Intensivists are increasingly attuned to the postdischarge outcomes experienced by families because patient recovery and family outcomes are interdependent after childhood critical illness. In this scoping review of international contemporary literature, we describe the evidence of family effects and functioning postpediatric intensive care unit (PICU) as well as outcome measures used to identify strengths and weaknesses in the literature. Methods We reviewed all articles published between 1970 and 2017 in PubMed, Embase, PsycINFO, Cumulative Index of Nursing and Allied Health Literature (CINAHL), or the Cochrane Controlled Trials Registry. Our search used a combination of terms for the concept of “critical care/illness” combined with additional terms for the prespecified domains of social, cognitive, emotional, physical, health-related quality of life (HRQL), and family functioning. Results We identified 71 articles reporting on the postPICU experience of more than 2400 parents and 3600 families of PICU survivors in 8 countries. These articles used 101 different metrics to assess the various aspects of family outcomes; 34 articles also included open-ended interviews. Overall, most families experienced significant disruption in at least five out of six of our family outcomes subdomains, with themes of decline in mental health, physical health, family cohesion, and family finances identified. Almost all articles represented relatively small, single-center, or disease-specific observational studies. There was a disproportionate representation of families of higher socioeconomic status (SES) and Caucasian race, and there was much more data about mothers compared to fathers. There was also very limited information regarding outcomes for siblings and extended family members after a child's PICU stay. Conclusions Significant opportunities remain for research exploring family functioning after PICU discharge. We recommend that future work include more diverse populations with respect to the critically ill child as well as family characteristics, include more intervention studies, and enrich existing knowledge about outcomes for siblings and extended family.


Introduction
With mortality rates in pediatric intensive care units (PICUs) falling below 10% in many countries, [1][2][3] intensivists are increasingly attuned to the quality of life and multiple other discharge outcomes experienced by PICU survivors and their families. Pediatric postintensive care syndrome (PICS-p) 4 collectively describes the psychological, cognitive, and health-related quality of life (HRQL) challenges experienced by survivors of childhood critical illness which includes parents, siblings, and extended family members. Much of the related literature has focused on patientcentered outcomes, however, in a systematic review of 33 studies from 2005 to 2019, Abela et al. found that many parents of critically ill children reported psychological, physical, and social impacts after their child's hospitalization, sometimes for years after discharge.6 Importantly, parental health and well-being

Family Functioning
Within the family outcomes domain, nearly a third of articles out of 71 (n = 23, 32%) measured family functioning,  with dominant themes that included changes in family function after the child's illness, interpersonal family relationships, parenting, and family burden (online supplemental Table 1). Most of the articles in this subdomain (n = 19 of 23) assessed family function as one of several outcomes, particularly the intersection of family function with parental mental health and wellbeing. The 23 articles used a total of 41 unique outcome measures, including 13 unique semi-structured interviews, to assess family functioning in 1066 families and 477 parents  Parents transitioned from primary concern for child's medical condition, stress, and fear to financial concerns, focus on family functioning, and long-term recovery of the child.
Lewis 2014 FF, PMH 52 parents of children who had been supported on ECMO a minimum of 6 months prior from one center in the UK.

IES-R, FAD, HADS
19 to 34 months after child's hospital discharge. 21% of parents scored above the cutoff criteria for PTSD. Only 9.6% of the sample reported no posttraumatic stress symptoms.
Lutz 2012 FF, PMH 15 parents of NICU and PICU graduates under the age of 4 years with feeding concern, and 10 professionals who care for this population in the United States.
Not stated.
Feeding problems greatly impact families. System barriers and financial concerns escalated parents' distress. Evaluations were performed prior to surgery, at hospital discharge, 12 to 18 months later, and at 5 years of age.

Majnemer
Developmental delays were common in children with heart defects, and these deficits persisted to school entry. Many factors were significantly associated with developmental outcomes, suggesting a multifactorial etiology.
(continued)  after general PICU admission (n = 21 articles, including 6 focused on survivors of severe TBI), in mixed NICU/PICU populations (n = 1 article, 15 parents), and in a CICU/congenital heart disease population (n = 1 article, 29 parents). All articles except one included families whose children had survived their ICU stay. Nearly two-thirds of the articles (n = 13 of 23) reported on family outcomes beyond 6 months from hospital discharge; the longest interval reported was 11 years postdischarge. Two of the articles did not specify the timeframe of their family evaluations in relation to the child's illness. The impact of a child's PICU admission on general family functioning was mixed, and no consistent risk factors for lower family cohesion were clearly identified. While primary PICU diagnosis and illness severity were associated with greater disruption of family function in some studies, 22,26 factors most associated with improved family cohesion included lower baseline psychological distress, greater social support, and being in a twoparent family. 24 Interestingly, family adaptability was higher in families with a greater No. of children, greater baseline financial concerns, and greater stress about the hospital staff behavior. 25,27 Studies also described challenges in the relationship between parents 8,15,22 and between siblings 22 after a PICU admission for a child in the family, and several highlighted the difference in emotional responses of mothers compared to fathers. 8 Attitudes toward parenting were also shown to shift after PICU discharge, 6,9,18,21 and the effects of PICU admission on parentchild bonding were reported as being both improved 12 and impaired. 8 Fear of a child incurring a new injury or another illness appears to cause some parents to worry about or limit their child's return to normal activities. 12,19 Assessments of family burden quantified how anxietyprovoking a child's postdischarge health and behaviors can be to their parents, the time constraints of these characteristics on parental time, and how the child's health and behavior impacted family interaction, activities, and finances. 22 Increased family burden scores after PICU discharge were associated with worsening child functional status in several studies, and many parents noted that they had to provide more physical and emotional care to a previously independent child. 6 These effects of the child on the family, and the family on the child, are interrelated, and it is notable that the family functioning subdomain had significant overlap with all five of the other subdomains.

Parent Mental Health
Parental mental health was the focus of 62 manuscripts (86%), representing the largest component of family-centered outcomes within this data set. 6,7,[12][13][14][15][16]18,20,21,23,24, These 62 articles used 43 unique validated outcome measures, physical exam, laboratory testing, and 18 qualitative interviews to report on parental mental health and well-being after general PICU admission (n = 45), after specific illnesses/interventions requiring PICU admission (n = 11), or CICU admission (n = 7). More than three-quarters of the articles (79%) evaluated multiple parent outcomes using more than one measure, and the majority of measures were obtained via interview [in person (45%), Mail/email (22%), telephone (0.07%), or a mix (19%)]. All but two of the articles specified the timing of measured outcomes, which ranged from 24 h postevent to 16 years posthospital discharge. Nearly half of the studies (n = 27, 45%) assessed outcome within 6 months of hospital discharge, and 75% of all the studies specifying a time interval for outcome assessment were completed within a year of hospital discharge.
In the parental mental health subdomain, there were 45 quantitative or mixed methods articles that reported having a child admitted to the PICU as a stressful event for parents and evaluated the risk of mental health effects such as posttraumatic stress disorder (PTSD) (n = 23), acute stress disorder (n = 3), depression (n = 20), anxiety (n = 22), general psychological dysfunction (n = 23) and traumatic growth/coping (n = 2). Most of the articles (62%) primarily reported postdischarge outcomes for parents and examined the risk of developing parental mental health issues following PICU admission in the context of preadmission parental factors and/or in-hospital parental and child/ patient factors. Thirteen qualitative articles explored several subthemes, including feeling disempowered during the child's admission, significant distress during and after PICU admission, unmet information and emotional support needs, and psychological issues such as PTSD and anxiety. Only 6 of the 58 articles (10%), including 1 qualitative study addressing parental mental health, were interventional studies. Of these, three protocols implemented in-hospital coping education and expectation setting, one implemented both in-hospital education and a follow-up telephone call, and one implemented the offer of a follow-up visit. The qualitative study identified processes used by parents to make sense of their child's death during bereavement meetings. Only two of the five quantitative articles were adequately powered to evaluate the effect of the intervention; both of these studies showed improvement of parental mental health and a correlation between parental mental health and the child's psychological function.
Grief. Eight articles in the parental mental health subdomain were related to parental grief and the need for support surrounding a child's death in the PICU, [56][57][58][59][60][61][62][74][75][76] and one additional article focused on the parents' views on their other child/children's response to their sibling's death. 28 Most (5 of 8) reported mental health outcomes within a year of the child's death, but one study contacted families at a mean interval of 3.4 years from their child's death (range: 2.3 to 5.7 years). 58 In addition to open interviews where parents and siblings were asked to share their thoughts and feelings, the eight articles used four standardized measures of grief, and several additional measures of mental health, 75 caregiver burden, 76 and strength of relationships. 76 Studies showed that parents experience high levels of complicated grief symptoms; for some parents, symptoms may decrease within 6 to 18 months of their child's death, for others complicated grief may persist. Parents reported negative physical and mental health outcomes such as depression, PTSD, and anxiety-related hospitalizations. In a few studies, parent grief was associated with the child's cause of death, the chronicity of their child's illness, and parent demographics.

Parental Physical Health
Out of 71 articles, we identified 4 articles that reported on parental physical health, including 3 articles that also assessed the surviving child 24,37,53 and 1 that focused on grieving parents of children that did not survive intensive care. 75 The articles included a heterogeneous population of 484 families from 3 different countries, assessed at a variety of time points ranging from several months to more than a decade postPICU discharge. The articles that assessed parents of PICU survivors included parents of children who had undergone corrective surgery for congenital heart disease (n = 113 mothers and 88 fathers), 53 parents of meningococcemia patients (n = 134 parents), 77 and parents of cardiac arrest survivors (57 families). 24 Ethnicity was specified in only one article. 75 Overall, there was a common theme of impaired physical health among parents at the time of PICU discharge, with improved self-reported health by parents over time. Several explanations for these outcomes were explored, including: "posttraumatic growth" or "response shift" due to changed perspectives or internal standards, finding meaning in their child's illness, and use of available resources and coping strengths.

Family Economics
Family economics or parental work was evaluated in 7 of the 71 articles within the family domain, primarily among families of general PICU patients, with multiple articles assessing the families of children who had sustained a traumatic brain injury. 10,11,22,24,36,75,78 These outcomes were assessed from 6 weeks to 16 years following PICU discharge, with the majority of follow-up within the first 5 years following admission (four articles assessed multiple time points within 1 year, 11,22,36,75 two were between 2 and 5 years, 10,24 and one was between 4 and 16 years). 78 Overall themes were that lower socioeconomic status (SES) was associated with less developmental follow-up of former PICU patients, and that higher financial spending and healthcare utilization and work and/or school days missed correlated with severity and length of PICU admission. However, when considering employment and finances, overall parent-reported QOL was more related to the presence of chronic underlying conditions among the children than the acute PICU admission itself. Additionally, parental PTSD was associated with stressful work conditions and unemployment. 36 One additional article commented on a greater predominance of fathers compared to mothers, to return to work quickly after a child's NICU/PICU death with a range of return to employment between 0 and 427 days. 75

Siblings and Extended Family
There is a severe lack of information regarding siblings' outcomes after a child's PICU stay, and even less information about extended family members. Only 3 out of the 71 studies included in this systematic review evaluated sibling outcomes, 18,22,28 and only one of these studies was exclusively focused on siblings. 28 In that 2013 study, 27 parents were interviewed approximately seven months after their child's death, with identification of several themes, including confusion among preschoolers, behavioral changes among school-age children and adolescents, and the tendency of siblings to recognize the parent's grief, and attempt to provide comfort to the adult rather than seeking comfort for themselves. 28 In a 2002 study designed to assess outcomes of children with traumatic brain injury after PICU stay, 82 families were assessed at multiple time points the following year. Parents perceived that the relationship between the sibling and patient deteriorated over time, and 21% of siblings were reported to have worsening behavior, and deterioration in relationship with parents. Approximately 40% of families reported that siblings could not stay in their own home and/or missed school due to their sibling's ongoing medical needs. 22 Finally, a small qualitative study in Denmark assessed the effect of PICU diaries on family recovery at 4 to 6 months after discharge among three families. 18 The siblings remarked that the diaries provided answers to questions they had about events that had taken place while their sibling was in the PICU. No studies focused on extended family members, including grandparents, stepparents, aunts, or uncles.

Outcomes Instruments Used by the Articles in the Family Domain
In addition to 34 unique unstructured and semi-structured interviews, the articles identified in this scoping review used 101 unique instruments to evaluate mental health (n = 50), family functioning (n = 5), coping and/or grief (n = 8), family economics (n = 4), cognition (n = 9), and HRQL or general health (n = 25). The overwhelming majority of these instruments were commercially or publicly available, and were previously validated instruments, with only 6 (4%) representing novel instruments created by study investigators. Specific measures for stress (n = 10) and posttraumatic stress symptoms (PTSS) or PTSD (n = 6) were the most common themes reported in assessments of child and parent mental health, but there were also a variety of instruments used to assess anxiety (n = 5), depression (n = 4), coping (n = 3), and other mood states (n = 3). Only three of the instruments were specific to health care/PICU exposure (Glasgow Meningococcal Septicaemia Prognostic Score, Parent Stressor Scale:PICU, and the Posthospitalization Behavior Questionnaire), while the remaining either adapted scales to measure trauma by defining the PICU admission as the traumatic event under investigation, or used nonspecific measures of mood, economics, or function without reference to the PICU admission specifically. Most of the instruments relied on parent reports, including 7 of the 12 instruments used to assess child behavior and cognition postPICU discharge.

Discussion
In a multinational collaboration extending over a several year period, our group conducted a scoping review of the medical literature since 1970 to identify all articles pertaining to family functioning or family member well-being after a child's PICU admission. This review identified 71 articles reporting on the late effects of critical illness in childhood on more than 2400 parents and 3600 families of PICU survivors in 8 countries, and an additional 600 parents and families recovering from the death of their child in the PICU. Articles identified in this review used 101 different metrics to assess the various aspects of family outcomes, and 34 articles additionally included open-ended interviews that provided a rich narrative of the PICU experience and postPICU environment from parents' and caregivers' perspectives. Overall, the body of literature identified in this scoping review indicated that most families experienced significant disruption in at least 5 out of 6 of our proposed family outcomes subdomains, with strong themes of decline in mental health, physical health, family cohesion, and family finances identified. However, there was also a clear demonstration of families' resilience, recovery, and growth after their child's critical illness. Without question, a family's exposure to the PICU environment is often life changing, and we are only beginning to understand the multifaceted and long-term effects of PICU survivorship for children and their families.
Of the 71 articles identified in the family domain, it is notable that 29 (40%) of these were published in the last 10 years. Although it is encouraging that there has been a significant increase in the No. of studies that address the effects of a child's critical illness on the entire family over the past 50 years, the fact that only 71 articles were identified in total indicates that there is still much in this area that remains unknown. It is also an encouraging sign that research appears to be moving toward greater inclusion of metrics that assess family function and other aspects of family health and well-being, such as family finances, social connections, and the effect of a child's illness on his or her siblings. However, the significant gaps in all three of these content areas, as well as the complete absence of literature that includes more extended family, such as grandparents, are notable. It is also notable that almost all of the identified articles reported the results of relatively small, single-center, or disease-specific observational studies, there was a disproportionate representation of families of higher SES and Caucasian race, and there was much more data about mothers compared to fathers. In addition, the metrics used to measure many functional and financial outcomes were not healthcare specific, and failed to incorporate the different sources of cost to families, including, for example, loss of work productivity and subsequent potential loss of insurance. Focused investigation of the impact of interventions on the overall dynamics of postPICU family function and economics in a variety of settings is desperately needed.
The results of this scoping review also indicate that the predominance of existing literature on family outcomes has reported on parental mental health; this was overwhelmingly the largest of our six subdomains, with very few articles identified that addressed the physical and mental health outcomes of other family members or the larger societal issues of family finances and social connectivity. However, many of the assessment instruments used by the articles in this scoping review capture data about other family subdomains which could be used to facilitate broader research in this area without significant increases in cost or duration of follow-up. For example, while the SRS-PTSD is a common instrument for assessing psychological distress, it is possible to analyze the "health problems score" separately, which may give more granular insight into a variety of aspects of family outcomes. The SF-36 and Caregiver Burden Assessment are other examples of "broad spectrum" instruments with subset composite scores that could be used in this way. This approach may be a practical method for defining the scope of family outcomes, as postdischarge PTSD and other stressors are commonly reported by parents, and are associated with a high rate of comorbid medical disorders and social dysfunction.
We acknowledge several limitations inherent to a scoping review, with perhaps the most significant limitation being the time it takes to actually perform such a broad review. By definition, that means there will always be delays between the original search and the ultimate publication of the results. This review included articles published through 2017; given the rapid expansion of research in this area, there are almost certainly newer publications that are not included. A possible approach to this perennial problem with scoping reviews may be to use a similar strategy to Cochrane Reviews, in which updates are published at structured intervals from each other. In addition, no search strategy is perfect. We may have missed articles because of the search terms used, or inappropriately excluded articles during the title and article review. We attempted to mitigate both of these concerns using wellvalidated methods for scoping reviews, 5 which include very broad literature searches and dual review by investigators that requires agreement for an article to be eliminated from consideration. Given the exhaustive nature of this review, we are confident that despite these limitations, this report represents the most comprehensive review of this topic to date.

Conclusions
The bulk of postPICU outcomes research has rightfully focused on the surviving child, however, many factors that influence the quality of that survival fall within a larger family paradigm that is often altered indefinitely by the circumstances of the child's PICU admission. Family outcomes research has increasingly provided data on parental mental health and family functioning in recent years, but the results of this scoping review demonstrate that much less is known about parental physical health, parental work and family finances, and the impact of a child's illness on siblings and extended family members. The majority of available work is observational and not interventional, reports on homogenous and limited disease-specific populations, and offers limited causal connections between the child's PICU exposure and ultimate family outcomes. Significant opportunities remain for research exploring family functioning after PICU discharge and possible effective interventions that can be implemented to mitigate the decline in family function following PICU discharge. Dr Kelly Michelson has funding from the National Palliative Care Research Center, the National Institutes of Health, and the Greenwall Foundation for unrelated research.

Declaration of Conflicting Interests
All other authors declare that there are no additional conflicts of interest.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval
Not applicable, because this article does not contain any studies with human or animal subjects.