ABSTRACT
IMPORTANCE: Surveys based on hypothetical situations suggest that health-care providers agree that disclosure of errors and adverse events to patients and families is a professional obligation but do not always disclose them. Disclosure rates and reasons for the choice have not previously been studied. OBJECTIVE: To measure the proportion of errors disclosed by neonatal intensive care unit (NICU) professionals to parents and identify motives for and barriers to disclosure. DESIGN: Prospective, observational study nested in a randomised controlled trial (Study on Preventing Adverse Events in Neonates (SEPREVEN); ClinicalTrials.gov). Event disclosure was not intended to be related to the intervention tested. SETTING: 10 NICUs in France with a 20-month follow-up, starting November 2015. PARTICIPANTS: n=1019 patients with NICU stay ≥2 days with ≥1 error. EXPOSURE: Characteristics of errors (type, severity, timing of discovery), patients and professionals, self-reported motives for disclosure and non-disclosure. MAIN OUTCOME AND MEASURES: Rate of error disclosure reported anonymously and voluntarily by physicians and nurses; perceived parental reaction to disclosure. RESULTS: Among 1822 errors concerning 1019 patients (mean gestational age: 30.8±4.5 weeks), 752 (41.3%) were disclosed. Independent risk factors for non-disclosure were nighttime discovery of error (OR 2.40; 95% CI 1.75 to 3.30), milder consequence (for moderate consequence: OR 1.85; 95% CI 0.89 to 3.86; no consequence: OR 6.49; 95% CI 2.99 to 14.11), a shorter interval between admission and error, error type and fewer beds. The most frequent reported reasons for non-disclosure were parental absence at its discovery and a perceived lack of serious consequence. CONCLUSION AND RELEVANCE: In the particular context of the SEPREVEN randomised controlled trial of NICUs, staff did not disclose the majority of errors to parents, especially in the absence of moderate consequence for the infant. TRIAL REGISTRATION NUMBER: NCT02598609.
