SEPREVEN: a Stepped-wedge Randomised Controlled Trial

Assessment of the effectiveness of the program on reducing the rates of adverse events (number of adverse events/1000 patient-days) (AEs) measured using a retrospective chart review with a NICU trigger tool (Sharek PJ et al. Pediatrics 2006;118(4):1332-1340). Out of the pool of patients included in the study, 3600 charts (60 charts per 4 months and per unit) will be randomly assigned to be reviewed using the NICU trigger tool. An adverse event is defined here as an injury, large or small, caused by the use (including non-use) of a drug, test, or medical treatment.(Trigger tool kit, Raju Ped Research). Preventable adverse events and non preventable adverse events secondary to expected complications or side effects are both included.

Assessment of the effectiveness of the program on reducing the rate of adverse events (percentage of patient admissions with an adverse event) measured using a retrospective chart review with a NICU trigger tool (Sharek PJ et al. Pediatrics 2006;118(4):1332-1340), over time during the inclusion period

Assessment of the effectiveness of the program on reducing the rates of adverse events (number of adverse events per 100 admissions) (AEs) measured using a retrospective chart review with a NICU trigger tool (Sharek PJ et al. Pediatrics 2006;118(4):1332-1340) over time

Assessment of the effectiveness of the program on reducing the rates of preventable adverse events (number of adverse events/1000 patient-days) (AEs) measured using a retrospective chart review with a NICU trigger tool (Sharek PJ et al. Pediatrics 2006;118(4):1332-1340). Out of the pool of patients included in the study, 3600 charts (60 charts per 4 months and per unit) will be randomly assigned to be reviewed using the NICU trigger tool. An adverse event is defined here as an injury, large or small, caused by the use (including non-use) of a drug, test, or medical treatment. (see Trigger tool kit definitions and Raju Ped Research) Preventability is assessed according to the existence of an "error". Expected complications or side effects from therapeutic or diagnostic interventions are generally non-preventable, and hence are considered complications. (see Raju, Pediatric Research) Healthcare associated infections are considered preventable adverse events.

Assessment of the effectiveness of the program on reducing the rates of Central line associated blood stream infections defined using the Primary Blood Stream Infections and LCBSI criteria (CDC/NHSN Device-Associated Module BSI January 2015), collected prospectively in the 12 participating NICUs. Analysis will include subgroups according to patient's characteristics and contextual factors (nurse staffing, unit occupancy, type of unit).

Assessment of the effectiveness of the program on reducing the rates of adverse events caused/lead by unplanned extubations measured using the retrospective trigger tool chart review (Sharek PJ et al. Pediatrics 2006;118(4):1332-1340). Out of the pool of patients included in the study, 3600 charts (60 charts per 4 months and per unit) will be randomly assigned to be reviewed using the NICU trigger tool.

Unprogrammed extubation: description of prospectively reported factors associated with UE (context and cause), and patients' characteristics, as well as HCP's assessment of preventability of this UE (multiple choices questionnaire) in the participating units. Prospective collection of outcome 30 minutes and 24 hours after UE (reintubation/ non invasive ventilation or spontaneous breathing and outcome severity grading) according to reported causal factors and patients' characteristics

Assessment of the effectiveness of the program on reducing the rates of severe adverse events collected using a multisource prospective voluntary anonymous collection. Results will include control charts, and include patients' characteristics and contextual factors over time.

Assessment of the effectiveness of the program on reducing the rates of adverse events related to medication errors, measured using the retrospective trigger tool chart review (Sharek PJ et al. Pediatrics 2006;118(4):1332-1340). Out of the pool of patients included in the study, 3600 charts (60 charts per 4 months and per unit) will be randomly assigned to be reviewed using the NICU trigger tool; A medication error is defined here as any preventable event that causes or leads to inappropriate medication use with patient harm while the medication is in the control of the health care professional in the NICU. (see Raju Ped Research).

Type of medication, of error and consequences and patients' and unit's characteristics are collected prospectively in the participating NICUs

Type of adverse event, consequence and patients' and unit's characteristics are collected prospectively in the participating NICUs

Assessment of the effectiveness of the program on reducing the rates of catheter-related adverse events collected prospectively and using the retrospective trigger tool chart review (Sharek PJ et al. Pediatrics 2006;118(4):1332-1340) in the 12 participating NICUs

Number and types of medical errors are defined as catheter related (central line), medication, ventilation, skin lesions, nasal injury, extravasation, enteral feeding, temperature control, unprogrammed extubation, surgical complications, diagnostic error or delay, parenteral fluid/nutrition adverse event, and, in existing, others. These "errors" are collected prospectively in each participating unit during the 20 months period, using a multi-source collection file (manual, anonymous and voluntary and/or institutional). A Medical Error is defined here as an error in health care, due to a failure of the planned action to be completed as intended, or using a wrong plan of action to achieve the goal, and excluding near misses. (see Raju Pediatric Research) Errors are included independently of the harm caused to the patient. Descriptions will include patients' characteristics and contextual factors.

Identify Severity of adverse events, rated prospectively using the National Coordinating Council for Medication Error Reporting and Prevention Index for Categorizing Errors (NCC MERP, 2001)(including infectious and non infectious adverse events). Severity is described according to contextual factors and patients' characteristics.

Late-onset infections are defined according to the January 2015 CDC/NHSN Surveillance Definitions for all types of Specific Types of infections, for Primary Blood Stream Infections and for neonatal Ventilator associated pneumonia (Hospital-Acquired infections for neonates). Clinical sepsis without laboratory proven infection is included. Description includes type on infection, clinical and bacteriological characteristics and type of consequence - data are collected prospectively by physicians. Descriptions will be done including patients' characteristics and contextual factors.

Dermatitis scoring is based on the existing Dermatitis score used by Garland JS (Garland JS et al. J Perinatol 2009), traduced in french (Score sheet available on demand). Score ranges from 0 (no dermatitis) to 4, with severe dermatitis >= 2. Score is completed propectively by the nurse in care, before and untill 5 after the procedure. Scores wil be compared according to antiseptic procedure and patient's characteristics.

The nasal trauma score is based on the score used by Collins (Collins CL et al, Eur J Ped 2014). Score is completed prospectively by the nurse in care of the patient, in the first week of nasal ventilation (after extubation or not). Rates and type of nasal lesions according to interface and patient's characteristics will be described.

Extravasation injuries from peripherally inserted short catheter score is rated from I to IV based on the classification described b are collected prospectively using the score described by Casanova (Casanova D et al, Br J Plastic Surgery 2001). Score is rated prospectively by nurse in care; for severe extravasation, context of extravasation (infused substances, site,...), patients' characteristics and consequences are collected prospectively.

For each collected adverse event, caregivers self-administered written questionnaire assessing of adverse event has been told to parent(s), heath care professional's reasons to disclosure it or not, and in case of disclosure, parental reactions according to health care professional (multiple choices)

Mortality (yes/no)

NICU length of stay (days)

Neurological complications (Intraventricular hemorrhage, leucomalacia) for prematures

Hospitalization data: morbidity for prematures bronchopulmonary dysplasia at 36 weeks (yes/no); severe retinopathia (yes/no); severe enterocolitis (yes/no)

Prospective record on volume of blood culture and results (positive/negative)

Safety culture is measured by a questionnaire filled in by the healthcare providers of the 12 units, in the weeks before the beginning and after the end of the patients' inclusion period using the french traduction of the Hospital Survey On Patient Safety Culture (HSOPSC)(http://www.ahrq.gov/qual/patientsafetyculture), named Questionnaire de mesure de la culture de sécurité des soins en milieu hospitalier, validated by the CCECQA. A comparison of the results before and after the study will be described globally and per unit.