BACKGROUND: The Hospital to Home Outcomes (H2O) trial was a 2-arm, randomized controlled trial that assessed the effects of a nurse home visit after a pediatric hospital discharge. Children randomized to the intervention had higher 30-day postdischarge reutilization rates compared with those with standard discharge. We sought to understand perspectives on why postdischarge home nurse visits resulted in higher reutilization rates and to elicit suggestions on how to improve future interventions. METHODS: We sought qualitative input using focus groups and interviews from stakeholder groups: parents, primary care physicians (PCP), hospital medicine physicians, and home care registered nurses (RNs). A multidisciplinary team coded and analyzed transcripts using an inductive, iterative approach. RESULTS: Thirty-three parents participated in interviews. Three focus groups were completed with PCPs (n = 7), 2 with hospital medicine physicians (n = 12), and 2 with RNs (n = 10). Major themes in the explanation of increased reutilization included: appropriateness of patient reutilization; impact of red flags/warning sign instructions on family’s reutilization decisions; hospital-affiliated RNs “directing traffic” back to hospital; and home visit RNs had a low threshold for escalating care. Major themes for improving design of the intervention included: need for improved postdischarge communication; individualizing home visits—one size does not fit all; and providing context and framing of red flags. CONCLUSION: Stakeholders questioned whether hospital reutilization was appropriate and whether the intervention unintentionally directed patients back to the hospital. Future interventions could individualize the visit to specific needs or diagnoses, enhance postdischarge communication, and better connect patients and home nurses to primary care.

BACKGROUND AND OBJECTIVES: Families of children with medical complexity are experts on their child's baseline behavior and temperament and may recognize changes in their hospitalized child's health before clinicians. Our objective was to develop a comprehensive understanding of how families identify and communicate their child's deteriorating health with the hospital-based health care team. METHODS:In this qualitative study, our multidisciplinary team recruited family members of hospitalized children with neurologic impairment. Interviews, conducted in the hospital, were audio recorded, deidentified, and transcribed. By using inductive thematic analysis, each transcript was independently coded by 3 or 4 team members. Members met regularly to reach consensus on coding decisions. Patterns observed were organized into themes and subthemes.RESULTS: Participants included 28 family members of 26 hospitalized children 9 months to 17 years of age. Children had a mean of 9 hospitalizations in the previous 3 years. Analysis resulted in 6 themes. First, families often reported their child "writes his own book," meaning the child's illness narrative rarely conformed to textbooks. Second, families developed informal, learned pathways to navigate the inpatient system. Third, families stressed the importance of advocacy. Fourth, families self-identified as "not your typical parents" and discussed how they learned their roles as part of the care team. Fifth, medical culture often did not support partnership. Finally, families noted they are often "running on empty" from stress, fear, and lack of sleep.CONCLUSIONS: Families of children with medical complexity employ mature, experience-based pathways to identify deteriorating health. Existing communication structures in the hospital are poorly equipped to incorporate families' expertise.

OBJECTIVES: Continuous physiologic monitors (CPMs) generate frequent alarms and are used for up to 50% of children who are hospitalized outside of the ICU. Our objective was to assess factors that influence the decision to use CPMs. METHODS: In this qualitative study, we used group-level assessment, a structured method designed to engage diverse stakeholder groups. We recruited clinicians and other staff who work on a 48-bed hospital medicine unit at a freestanding children’s hospital. We developed a list of open-ended prompts used to address CPM use on inpatient units. Demographic data were collected from each participant. We conducted 6 sessions to permit maximum participation among all groups, and themes from all sessions were merged and distilled. RESULTS: Participants (n = 78) included nurses (37%), attending physicians (17%), pediatric residents (32%), and unit staff (eg, unit coordinator; 14%). Participants identified several themes. First, there are patient factors (eg, complexity and instability) for which CPMs are useful. Second, participants perceived that alarms have negative effects on families (eg, anxiety and sleep deprivation). Third, CPMs are often used as surrogates for clinical assessments. Fourth, CPM alarms cause anxiety and fatigue for frontline staff. Fifth, the decision to use CPMs should be, but is not often, a team decision. Sixth, and finally, there are issues related to the monitor system’s setup that reduces its utility. CONCLUSIONS: Hospital medicine staff identified patient-, staff-, and system-level factors relevant to CPM use for children who were hospitalized. These data will inform the development of system-level interventions to improve CPM use and address high alarm rates.

BACKGROUND Hospitalized children generate up to 152 alarms per patient per day outside of the intensive care unit. In that setting, as few as 1% of alarms are clinically important. How nurses make decisions about responding to alarms, given an alarm's low specificity for detecting clinical deterioration, remains unclear. OBJECTIVE Our objective was to describe how bedside nurses think about and act upon monitor alarms for hospitalized children. DESIGN, SETTING, PARTICIPANTS This was a qualitative study that involved the direct observation of nurses working on a general pediatric unit at a large children's hospital. MEASUREMENTS We used a structured tool that included predetermined categories to assess nurse responses to monitor alarms. Data on alarm frequency and type were pulled from bedside monitors. RESULTS We conducted 61.3 patient‐hours of observation with nine nurses, in which we documented 207 nurse responses to patient alarms. For 67% of alarms heard outside of the room, the nurse decided not to respond without further assessment. Nurses most commonly cited reassuring clinical context (eg, medical team in room), as the rationale for alarm nonresponse. The nurse deemed clinical intervention necessary in only 14 (7%) of the observed responses. CONCLUSION Nurses rely on clinical and contextual details to determine how to respond to alarms. Few of the alarm responses in our study resulted in a clinical intervention. These findings suggest that multiple system‐level and educational interventions may be necessary to improve the efficacy and safety of continuous monitoring.

Introduction: Medical emergency teams (METs) bring critical care expertise to the bedsides of hospital ward patients who may be deteriorating. Diurnal variation in MET activation rates may identify inconsistencies in the detection of patients needing intervention. We aimed to determine whether such variation exists at our tertiary care children’s hospital. Methods: In this retrospective cohort study, we collected data including date and time of MET and disposition following MET for all inpatients at Cincinnati Children’s Hospital Medical Center with a MET call between January 2008 and May 2014. The analysis compared the MET rate between days and nights, weekdays and weekends, and before and after nursing shift change. Results: The number of METs per hour varied throughout the day. More METs were called during the day than at night (0.7 calls/shift ± 0.95 vs 0.6 ± 0.9, P < 0.001). There were also more METs per day on weekdays than weekends (1.4 ± 1 calls/d vs 1.2 ± 1, P < 0.001). Daytime METs were more likely to lead to transfer to the intensive care unit or operating room than those called at night (61.9% vs. 52.9%, P < 0 .001). MET activation rates did not differ significantly in the 2 hours before nursing shift change compared to the 2 hours after. Conclusions: At our large tertiary care children’s hospital, there are both diurnal variations and variations across weekdays versus weekends in the MET activation rate. This difference may indicate variations in our ability to detect deteriorating patients on the wards and be further studied.