Bringing I-PASS to the Bedside: A Communication Bundle to Improve Patient Safety and Experience

Our primary outcome was the rate of medical errors, including harmful errors (preventable adverse events) and non-harmful errors. Medical errors and adverse events were measured per 1000 patient-days before and after implementation of Patient and Family Centered I-PASS using an established systematic safety surveillance methodology. Trained research clinicians reviewed patient medical charts, hospital incident reports, family safety interviews, and staff reports for potential errors and adverse events. Trained physician-reviewers blinded to pre- vs. post-intervention status then categorized all suspected incidents as either adverse events, non-harmful errors, or exclusions. Adverse events that were clearly caused by a medical error were subsequently deemed preventable and all other cases were categorized as non-preventable.

Time frame: 6 months (3 months pre, 3 months post) per site (7 sites total)

Population: Patients admitted to study units.

Family experience before and after implementation. Experience was measured using a 10-15 minute survey verbally administered prior to discharge. Parents were asked to rate various aspects of their experience with care. This included experience during and after rounds, experience with written communication, experience with physicians and nurses, and overall hospital experience. The survey was developed, cognitively tested, and piloted at a non-intervention site (Boston Children's Hospital) and translated into Arabic, Chinese, Russian, and Spanish. We compared percent top-box experience ratings pre- vs. post-intervention using a GEE chi-squared test for binary outcomes, clustered by site. Top-box score was calculated as the percentage of participants that gave the top-most response for the given survey item (e.g., 5=Extremely; 5=Excellent). Missing data was accounted for through use of multiple imputations appropriate for missing data in clustered studies.

Time frame: 6 months (3 months pre, 3 months post) per site (7 sites total)

Population: English-, Arabic-, Chinese-, Russian-, and Spanish-speaking parents/caregivers of \<18-year-old patients on the study units were eligible to participate. Of these, 6-8 per week per site were randomly selected and surveyed 24 hours prior to discharge about their experiences with various aspects of communication.

Changes in quality of communication during rounds processes were assessed before and after implementation based on: (a) real-time structured direct observations of rounds (n=653) and (b) post-hoc analyses of audio-recordings of a subset of rounds observations (n=164). Research assistants conducted 1-hour weekly in-person rounds observation sessions per site, simultaneously completing a real-time assessment tool for each patient and audio-recording rounds. Site research clinicians blinded to pre- vs. post-intervention status conducted post-hoc analyses of a subset of rounds audio recordings using a structured assessment tool to measure rounding team adherence. Percent top-box ratings pre- vs. post-intervention were compared using a GEE chi-squared test for binary outcomes, clustered by site. Top-box score was calculated as the percentage of participants that gave the top-most response (e.g., 5=Excellent). Missing data were accounted for through use of multiple imputations.

Time frame: 6 months (3 months pre, 3 months post) per site (7 sites total)

Population: Observations and audio-recordings of rounds were conducted at each site for one hour at a time on a randomly selected day each week. A median of 3 (IQR 2,5) patients were observed and recorded per session. Of these, 2 patients per week per site were randomly selected for post-hoc analysis.

An additional measure of interest was the rate of non-preventable adverse events. Medical errors and adverse events were measured per 1000 patient-days before and after implementation of Patient and Family Centered I-PASS using an established systematic safety surveillance methodology. Trained research clinicians reviewed patient medical charts, hospital incident reports, family safety interviews, and staff reports for potential errors and adverse events. Trained physician-reviewers blinded to pre- vs. post-intervention status then categorized all suspected incidents as either adverse events, non-harmful errors, or exclusions. Adverse events that were clearly caused by a medical error were subsequently deemed preventable and all other cases were categorized as non-preventable.

Time frame: 6 months (3 months pre, 3 months post) per site (7 sites total)

Population: Patients admitted to study units.

Shared understanding between parent, resident, and nurse was measured before and after implementation of the Patient and Family Centered I-PASS intervention.

Time frame: 6 months (3 months pre, 3 months post) per site (7 sites total)

Population: These data were collected but final analyses are still pending. The shared understanding measure was a novel secondary measure in this study, and the analysis was more complex than initially anticipated. Therefore, the number of participants analyzed is currently 0.