Interventions to Reduce Psychologic Morbidity After PICU Discharge: Challenges to Establishing Efficacy*

    loading  Checking for direct PDF access through Ovid


Although mortality from critical illness during childhood is low, survivors of critical illness are at risk for adverse psychologic and behavioral outcomes. For example, posttraumatic stress disorder (PTSD) has been estimated to occur in 5–28% of children following a PICU admission, and posttraumatic stress symptoms (PTSS) occur in 35–62% (1). Among parents of PICU patients, PTSD has been reported in 10–21% and PTSS in as many as 84%. Psychologic morbidity following a PICU admission can lead to serious disturbances in child and family functioning; these disturbances, in turn, can have a major impact on child health.
In this issue of Pediatric Critical Care Medicine, Baker and Gledhill (2) report the results of a systematic review of interventions to reduce psychologic morbidity in parents and children after discharge from the PICU. Only six studies were identified that could be included in the review; three of these were reported as full randomized controlled trials (RCT) (3–5) and three as feasibility studies (6–8). Two of the feasibility studies also had RCT designs (6, 7), and one was an intervention without a control arm (8). Participants included parents and children. In some studies, all families discharged from the PICU were eligible; in others, only families at high risk for psychologic morbidity were included. The interventions evaluated primarily included psychoeducation and parental support in written and verbal formats (in-person or telephone) delivered during the hospital stay or at a follow-up clinic visit. Outcomes included self-report scales of mental health focusing on PTSD, depression, and anxiety. Overall, family participation in the interventions was low and attrition high, especially for those studies in which the intervention was administered after hospital discharge. None of the studies reported harm, but most showed only a small treatment effect, if any.
In light of the increasing awareness of psychologic morbidities after a child’s PICU stay, why have so few interventions been developed and tested for efficacy? The answer is perhaps partly related to the many challenges investigators face in attempting to conduct RCTs of psychosocial interventions in the PICU setting (9). In designing such trials, an interdisciplinary team including, for example, physicians, nurses, and psychologists, can be beneficial; however, managing diverse disciplinary perspectives is not always easy. The resultant experimental methods need to be accepted as scientifically legitimate by all team members. Input from patients and families may also be needed to design interventions that can be implemented without undue family burden and to identify meaningful outcomes.
As demonstrated in Dr. Baker’s review (2), low participation rates are typical of interventional trials aimed at reducing psychologic morbidity in PICUs, and participation is greater among mothers than fathers and among whites than minorities. To adequately recruit subjects with a range of backgrounds, recruitment methods must be developed with gender, culture, and language sensitivity, which increases the cost and complexity of the trial. Patients and families at high risk for psychologic morbidity may show the largest treatment effect in an interventional trial. However, restricting recruitment to those at high risk requires screening of patients and families (typically with self-report measures) during their PICU stay, at a time when many patients and families will be unwilling or unable to participate in screening. Inclusion of all PICU patients and families may waste study resources and dilute findings as many families will adjust without psychologic morbidity.
Randomization of patients and families to psychosocial interventions may itself be problematic. Even if the concept of random assignment is understood and accepted, patients and families may have a preference for one study arm or the other.

Related Topics

    loading  Loading Related Articles