Abstract
Prospective longitudinal mixed methods cohort study of parents’ recollection of stress-invoking events during their child's pediatric intensive care unit (PICU) admission. Respectively, 234, 190, and 71 parents of 188 children were assessed at PICU discharge, 3 to 9 months, and 18 to 30 months postdischarge, with substantial attrition observed at the final follow-up timepoint. Parental responses were analyzed using content analysis methodology and consensus coding. Scores of parental posttraumatic stress disorder/symptoms (PTSD/PTSS) and acute stress disorder measures were analyzed and compared to frequencies of each descriptive theme. Parents who responded within the theme “Sensing immediate need for additional care for my child” were more likely to meet criteria for PTSD qualification at 3 to 9 months (P = .02). At 3 to 9 months and 18 to 30 months, parents who reported no stressors related to the PICU stay were less likely to meet PTSS score qualification (P < .05). Parents experience significant stress during and following their child's PICU stay, with many stressors remaining consistent for up to 30 months; findings at T3 should be interpreted cautiously because of marked participant attrition.
Keywords
Background
Increased survival rates in pediatric critical care have shifted the focus from survivorship to morbidity outcomes, collectively known as postintensive care syndrome-pediatrics (PICS-p). 1 Parental psychological stress experienced during a child's admission to the pediatric intensive care unit (PICU) is one component of PICS-p, with significant levels of acute stress disorder (ASD) and posttraumatic stress disorder (PSTD) being reported.2-10 Post-PICU parental traumatic stress can have significant adverse effects on the family, including increased child psychological stress and familial dysfunction with financial and marital strain.3,6,8,11-13
Quantitative risk factors for PICS-p have been previously described. 3 Qualitative studies have been instrumental in highlighting parent insight into stressors experienced during or after a PICU admission, adding to our understanding of the syndrome.10,14-17 Most of these studies have been limited to a single assessment and thus are unable to address duration or changes in parental perceptions of PICU-associated stressors.
This report is from a 2-site, mixed methods longitudinal cohort study documenting the multiple dimensions of the parent PICU experience and associations with parental ASD and PTSD.2,6 We qualitatively examined the parents’ reports of their most stressful PICU experiences, their post-PICU stress symptoms, and the persistence of those symptoms up to 18 to 30 months following discharge from the PICU. Further, we descriptively examined the potential match of the qualitative findings with the criteria for ASD, posttraumatic stress symptoms (PTSS) and PTSD.
Methods
Full description of the larger study and study procedures have been reported previously.2,6 In brief, we conducted a prospective mixed methods cohort study involving parents of patients admitted to the PICUs at 2 tertiary children's hospitals from June 2016 through January 2020. This study, “Prevalence of long-term PTSD symptoms in parents of children admitted to the PICU,” was approved by the IRB at each hospital (L-11,839) on 07 December 2015 (00003891) on 23 May 2016. All procedures followed the standards of both institutional review boards. Respondents gave written consent for review and signature prior to starting interviews.
Inclusion and Exclusion Criteria
Within 3 to 14 days of PICU admission, we approached a convenience sample of parents whose children (0-17 years) had an unexpected admission to the PICU for > 48 h. All available parents of children within the convenience sample were approached to participate. We excluded parents with (1) a prior history of a psychiatric disorder that resulted in hospitalization in the past 2 years; (2) a current diagnosis of PTSD; (3) a history of suspected or proven child abuse; (4) an inability to speak English; and (5) a child who died during the study period. Written consent was obtained from all participants.
Study Design and Measures
We used the Acute Stress Disorder Scale for DSM-5 (ASDS-5) to evaluate for probable ASD at T1 (peri-discharge) and the validated scale PTSD Symptom Scale-Interview for DSM-5 (PSS-I-5) to evaluate for PTSD at T2 (3-9 months postdischarge) and T3 (18-30 months postdischarge) (Supplemental Table S1). These timepoints were chosen to capture the different phases of parental adaptation to their child's critical illness. Peri-discharge representing the acute response, 3 to 9 months postdischarge the intermediate adaptation response and 18 to 30 months postdischarge the long-term adaptation response. 18
At T1, parents responded in writing or verbally to the following question adapted from the ASDS-5 19 : “Briefly describe, if any, the most traumatic experience during your PICU stay?.” The adaptations to this interview question were assessed for face validity in a focus group of PICU parents, PICU physicians and a clinical therapist before their use in this study. Parents at T2 and T3 responded verbally during a telephone interview with a trained research team member to the following question adapted from the PSS-I-5 20 : “Please briefly describe, if any, the experience during your PICU stay which was the most distressing and/or the most haunting for you currently?.” Parent response was immediately verified through a read back technique. Parents also completed the ASDS-5 (T1) and PSS-I-5 (T2 and T3) to evaluate for acute and posttraumatic stress. Medical and social histories of the parent and child were obtained including socioeconomic status, history of parent/caregiver mental and physical illness, parent prior traumatic life events, and traumatic events between assessments.
Trauma Conceptual and Operational Definitions
Definitions are based on the criteria established by the developers of the ASDS-5 and PSS-I-5. Conceptually, the stress response is a continuum with impairing stress occurring within 4 weeks following a trauma diagnosed as ASD and stress occurring beyond 4 weeks diagnosed as PTSD.
21
1. ASD qualification: a score of ≥ 3 on at least 9 of the 14 symptoms on the ASDS-5.
19
2. PTSD qualification: total PSS-I-5 score ≥ 23 and/or reporting a symptom profile of at least 1 intrusion, 1 avoidance, 3 cognition/mood and 3 arousal/reactivity symptoms along with impairment in functioning.
20
A PTSS which does not meet criteria for PTSD may still impact a parent's functioning, we also assessed parents who met criteria for more than minimal PTSS (including parents who met PTSD criteria).
3. More than minimal PTSS: Score ≥ 9 on the PSS-I-5.
20
Qualitative Analysis
A semantic descriptive content analysis methodology 20 and consensus coding were used to analyze the interview responses. Subject data were coded individually regardless of relation to each other (ie, spouse/co-parent). All interview responses were initially reviewed by 2 coders experienced with pediatric critical care, PICS-p, PTSS, PTSD, and qualitative data analysis. After all responses were reviewed, major impressions were discussed and then coding commenced. Both coders were blinded to the parental scores of ASD or PTSS/PTSD. As per the analytic methods, the unit of response was decided a priori to be the full response to the interview question. The unit of analysis was the phrase. First level codes were developed along with key attributes; codes that co-occurred and had related meaning were combined to form themes. Conceptual definitions for each theme were then developed.
Quantitative Analyses
All statistical analyses were performed using SAS 9.4 (SAS Institute, Inc., Cary, North Carolina). Scores from the ASDS-5 and the PSS-I-5 were computed for each parent participant at each time point, all subjects/parents were treated as an individual entry. Based on scores, parent cohorts were identified indicating qualification or not for probable ASD and for probable PTSD. Associations of the diagnosis of probable ASD, PTSS and PTSD with the qualitative themes were assessed using chi-square test or Fisher exact test. A P-value <.05 was considered statistically significant.
Results
A total of 234/234 (100%) of eligible parents assessed for ASD at discharge completed the interview question. One hundred ninety out of 195 (97.4%) parents at 3 to 9-months and 71/175 (40.6%) at 18 to 30-months completed both the qualitative question and the PTSD assessment. Demographic data are reported in Table 1. The large drop in respondents at T3 was due to an error in data collection that omitted the interview question at the Westchester site at the 18 to 30-month data point. Of note, Penn state enrollment at the ASD discharge assessment was 92/234 (39%), making T3 attrition at Penn State 21/92 (22.8%). On Pearson's chi-test the T3 cohort had significantly higher rates of anxiety (P = .004) and depression (P = .044) on admission, and ASDS (P = .049) at discharge, otherwise the cohorts were similar.
Parent and Child Baseline Demographics.
Numbers and percentages less than total are due to missing data or rounding.
Median value with IQR.
Illness or diagnosis that requires ongoing medical attention or medication outside of normal preventative care.
Most Frequently Reported Thematic Findings
at Discharge (T1)
The total number of coded interview phrases was 520 at T1; zero were uncoded. From the coded phrases, 11 themes were identified. The most frequently reported theme was parents “sensing immediate need for additional care for my child,” reported 141 times by 126 parents (Supplemental Table S2). This theme represented parents’ reported awareness of their child's urgent need for medical attention or of the medical staff's concern. One mother reported stress due to the staff response, “…they had to cut his shirt off, it really hit me….” Additional examples of the most reported themes with brief descriptions are below; the full list with description and examples are in Supplemental Table S2. ‘Physical symptoms or circumstances of the child's injury conveying seriousness”: Parent- observed changes in their child (eg, bleeding) or witnessing of the injury (eg, car accident) that impressed upon them their child's vulnerability. One mother described the time pre-hospital “My husband pulled our unresponsive daughter out of the pool.’ While another described observations in the PICU ‘…she had become flaccid and unable to move.’ ‘Thinking back on the impact of being told the medical diagnosis’: Many parents stated that the diagnosis was the most stressful. One mother reported ‘… found out she had lobar holoprosencephaly.’
Themes Associated With Parental ASD at Discharge
None of the themes at discharge were significantly associated with the ASD qualifying scores. However, parents who reported the length of the PICU stay as the most stressful experience trended toward a significant association with the ASD scores (P = .069) (Supplemental Table S3).
Most Frequently Reported Thematic Findings at 3 to 9-Months Postdischarge (T2)
There was a total code frequency of 215 and 6 uncoded responses. Seventeen unique themes were identified; their brief descriptions and examples are reported in Supplemental Table S4. Of these, the most frequently reported was “child's physical symptoms/circumstances of injury conveying seriousness of situation” with 41 instances from 36 parents (Table 2). Prevalence of probable PTSD and PTSS qualification was 29/190 (15.2%) and 80/190 (50%) parents, respectively.
Most Common Codes and Themes of Stressors Identified at 3 to 9 Months Postdischarge.
Examples of the most reported themes with brief descriptions are below; full list in Table 2. ‘Sensing immediate need for additional care for my child’ and ‘Physical symptoms or circumstances of child's injury conveying seriousness’, both themes were defined as at T1. ‘Initial time surrounding admission was the worst’: Parents noted the pre arrival period, waiting at the hospital and/or initial uncertainty about their child's condition were the most stressful. One parent stated, ‘transport to the hospital is the most stressful part.’ Others specified the initial period of waiting in the hospital as quite stressful, ‘Being in the ED, not knowing what will happen.’ ‘Thought my child was dying’: this theme, as described previously, also emerged at T2. ‘I am feeling ok’: This new theme emerged at T2 and represented those parents who denied stress related to the index PICU admission. As one parent reported ‘Not currently bothered or stressed.’
Themes Associated With Parental PTSD at 3 to 9-Months Postdischarge
Parents who reported stress due to “sensing immediate need for additional care” during their child's admission were significantly more likely to meet qualification for probable parental PTSD (P = .02) (Table 3). Parents who reported “I’m feeling ok” were significantly less likely to meet qualification for probable parental PTSD (P = .03) (Table 3).
Theme Association With PTSD and PTSS Qualification at 3 to 9 Months.
PICU, pediatric intensive care unit; PTSD, posttraumatic stress disorder; PTSS, posttraumatic stress symptoms.
Themes Associated With Parental PTSS at 3 to 9-Months Postdischarge
Parents who reported stress due to “initial time surrounding admission” were significantly more likely to meet qualification for probable parental PTSS (P = .001) (Table 3). Parents who reported “I’m feeling ok” were significantly less likely to meet qualification for probable parental PTSS (P = .003) (Table 3).
Most Frequently Reported Thematic Findings at 18 to 30-Months (Supplemental Table S5)
Reports from 71 Penn State parents who completed the assessment at 18 to 30-months post-PICU discharge were represented by 16 unique themes, reported a total theme frequency of 109, 1 uncoded response. The most frequently reported was “Physical symptoms or circumstances of child's injury conveying seriousness” with 26 instances from 17 parents (Table 4). Prevalence of probable PTSD and PTSS qualification was 13/71 (18.3%) and 26/71 (36.6%) respectively.
Codes and Themes of Stressors Identified at 18 to 30 Months Postdischarge.
PICU, pediatric intensive care unit.
The full list of themes and examples are noted in Supplemental Table S5, no parent reported the theme “Blaming the Child” or “Age of patient.” Otherwise, theme frequency followed similar trends as the 3 to 9 months postdischarge cohort.
Themes Associated With Parental PTSD at 18 to 30 Months Postdischarge
No theme was significantly associated with probable PTSD qualification at 18 to 30-months postdischarge. However, parents who reported “I’m feeling ok” were significantly less likely to meet qualification for probable parental PTSS (P = .009) (Table 5).
Theme Association With PTSD and PTSS Qualification at 18 to 30 Months.
PICU, pediatric intensive care unit; PTSD, posttraumatic stress disorder; PTSS, posttraumatic stress symptoms.
Consistency of Stressor at PTSS/PTSD Assessment
Seventy-one parents completed the qualitative question at both postdischarge timepoints. Consistency of reported stressor code at 3 to 9-months and 18 to 30-months postdischarge was high ranging from 71% to 100% per code.
Parents who reported “Sensing immediate need for additional care” as the stressor associated with their PICU stay at both T2 and T3 were significantly more likely to meet criteria for probable parental PTSD at 18 to 30 months (P = .03). Parents who consistently reported being worried about damage to child/uncertain future trended toward being more likely to meet probable PTSD qualification at 18 to 30 months postdischarge (P = .08). Regarding PTSS, parents who consistently reported “Sensing immediate need for additional care for my child” at both assessment time points trended toward being more likely to meet probable PTSS qualification at 18 to 30 months postdischarge (P = .06).
Discussion
This is the first longitudinal, 2-site study of the PICU-associated parental PTSD to use both qualitative and quantitative methods to examine parents’ reports of stressors associated with their child's PICU admission. While many parents continue to report current stressors related to the PICU admission, only a few of the identified themes were consistently associated with probable ASD, PTSD/PTSS qualification. However, parents who reported no current stressors during assessments were significantly less likely to meet qualification for probable PTSD/PTSS, indicating that parents may accurately report their level of current stress using quantitative and qualitative methods. Consistent with the updated Integrative Trajectory Model of Pediatric Medical Traumatic Stress (PMTS), this finding suggests that parents follow heterogeneous recovery trajectories, and that contemporaneous stress appraisal may reflect parents’ position along resilient, recovery, or chronic distress pathways. 22 Our prior work noted these heterogeneous responses are multifactorial including parent previous life stressors, acute stress response and long term PICU/illness sequelae for both family and patient. However, all data suggest that clinicians may be able to correctly screen for parents who are most in need of intervention using parent reports after discharge.
When examining themes that were positively associated with probable parental PTSS and PTSD, we noted that sensing urgency for medical treatment and the initial period of illness/diagnosis both convey parents’ uncertainty and worry during a period of high threat and decline in their child's health. Within the PMTS framework, 22 this peritraumatic phase represents a critical window during which heightened threat appraisal may influence longer-term stress trajectories. These thematic findings suggest that the initial care period including diagnosis may be the starting point for an intervention designed to address parental perceptions that could aggravate an already difficult situation. Interventions known in PTSD literature such as cognitive behavioral therapy, stress management and coping techniques 23 may be helpful in this population during this timeframe. More studies would be needed to explore methods and timing of interventions.
Parents’ perceptions and experiences of medical events and/or their child's illness during the hospital stay were the most reported stressors. This equates to what Jakobsen et al described as “The challenging PICU stay.” 24 Within this theme parents felt stress at how quickly events/illness happened and their child's physical experience, especially around the admission timeframe. They recalled with vivid detail even at greater than 18 months postdischarge those specific events that evoked stress. Within the PMTS trajectory-based model, such persistent and emotionally salient recollections may reflect sustained distress rather than full recovery.
As reported by Dahav et al, the child's current and future health vulnerability is a significant parental stressor. 17 Parents expressed worry about survivability, present and future functioning during the PICU stay, and these concerns often persisted over time. In the PMTS model, ongoing uncertainty during the long-term adaptation phase may perpetuate traumatic stress even after medical stabilization. This highlights ongoing assessment or screening may be beneficial to allow for intervention and mitigation of PTSS/PTSD.
Similar to the conceptualization of medical traumatic stress and our quantitative report, parents in this cohort showed varied stress responses. 3 Qualitative analysis identified parents who reported no measurable stress response and those who had sustained stress response. However, much like quantitative data, no one parent-reported stressor was universally associated with probable ASD, PTSS, or PTSD. This variability is consistent with distinct PMTS recovery trajectories and supports the use of qualitative assessment alongside quantitative screening to tailor intervention strategies.
We hope to use these data in combination with our quantitative data to create a model identifying parents most at risk for PICU associated parental PTSD.
Limitations
There are limitations to this study. (1) There was a steep decline in the number of participants who completed the qualitative question at 18 to 30 months, in large part by one site erroneously omitting this data point. The attrition rate at the site that completed the 18 to 30 month data point was somewhat modest at 22.8%. However, the overall effect of attrition cannot be fully assessed. (2) Respondents were interviewed months to almost 3 years after discharge, risking recall bias. Despite participants being asked to describe their current stressors, reframing of memories with information processed since a previous data collection point could have altered current reporting. (3) There may have been life experiences that could increase risk of probable PTSD qualification for which we did not account including long term patient health outcomes. (4) The considerable time (18-30 months) for T3 allows for potential shifts in probable PTSS/PTSD diagnosis. (5) The PSS-I-5, despite demonstrated concurrent validity with the Clinician-Administered PTSD Scale (rs > .72), is not the gold standard for a PTSD diagnosis. Therefore, probable PTSD/PTSS qualification and not diagnosis is represented. Similarly, the ASDS-5 has not been fully validated, though communication with the scale developer suggests that the minor changes from the ASDS-4 are unlikely to significantly impact validity. (6) Our adaptation to each scales’ original question, while vetted by PICU parents, physicians, and a clinical counselor for face validity, are not fully validated questions. (7) Validated screens were only available in English; therefore, non-English speaking parents were excluded. (8) Lastly, this study was designed to establish potential risk factors and not causality.
Conclusion
Parents’ perception of stress evoking events in the PICU is an important contributor to understanding PICU-associated parental ASD and PTSD. We theorize that specific themes derived from parents’ reports represent risk factors or periods most associated with the development of parental PTSS/PTSD and thus threaten family well-being. We anticipate that these qualitative findings may inform the future development and timing of clinical interventions to prevent or minimize this experience for parents.
Supplemental Material
sj-docx-1-jpx-10.1177_23743735261449569 - Supplemental material for Pediatric Intensive Care-Associated Parental Traumatic Stressors by Parent Report: Beyond the First Year
Supplemental material, sj-docx-1-jpx-10.1177_23743735261449569 for Pediatric Intensive Care-Associated Parental Traumatic Stressors by Parent Report: Beyond the First Year by Mekela Whyte-Nesfield, Jiaxiang Gai, Daniel Kaplan, Simon Li and Pamela S. Hinds in Journal of Patient Experience
Footnotes
Acknowledgments
Pamela P. Siller, MD
Institutions at Which Study Was Completed
Data were collected at Penn State Health Children's Hospital and Maria Fareri Children's Hospital, New York Medical College. Data analysis was completed at Children's National Hospital.
Address for Reprints
111 Michigan Avenue NW Washington DC 20010. No reprints will be ordered.
Ethical Adherence
The authors confirm that there are no known conflicts of interest associated with this publication and there is no significant financial support for this work that would alter its results.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study received financial support for statistical analyses from the DC Lawyer's Care for Children Endowment Fund to Children's National Hospital.
Statement of Authorship
All the authors of this article are responsible for the reported research. We confirm that the article has been read and approved by all named authors. All the authors participated in the concept and design, analysis and interpretation of data, drafting and revising the article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
