AIMS: To assess the feasibility of research into the occurrence, causation and prevention of adverse events (AEs) in New Zealand public hospitals. METHODS: A two-stage retrospective review was carried out on 1,575 medical records selected by systematic list sample from admissions for 1995 in three public hospitals in the Auckland region. Following initial screening, medical records were subject to structured implicit review using a standardised protocol. Feasibility measures, using international benchmarks where possible, were: adequacy of sample selection; completeness of medical records; reliability and validity of screener and reviewer judgements; internal consistency and face validity of AE determination and preventability assessment. RESULTS: The sample selection procedure was effective, although nearly 10% of records could not be secured. Information in medical records was sufficient for the identification and analysis of AEs. Adequate levels of agreement were achieved for screener and reviewer judgements, with kappa scores ranging between 0.302 and 0.622 and positive predictive values between 50.0% and 89.7%. The criteria for AE determination showed internal consistency and face validity, as did those for preventability. CONCLUSIONS: Research into the occurrence, causation and prevention of AEs in New Zealand health care settings is methodologically feasible and meets international benchmark standards.
AIMS: To assess the feasibility of research into the occurrence, causation and prevention of adverse events (AEs) in New Zealand public hospitals. METHODS: A two-stage retrospective review was carried out on 1,575 medical records selected by systematic list sample from admissions for 1995 in three public hospitals in the Auckland region. Following initial screening, medical records were subject to structured implicit review using a standardised protocol. Feasibility measures, using international benchmarks where possible, were: adequacy of sample selection; completeness of medical records; reliability and validity of screener and reviewer judgements; internal consistency and face validity of AE determination and preventability assessment. RESULTS: The sample selection procedure was effective, although nearly 10% of records could not be secured. Information in medical records was sufficient for the identification and analysis of AEs. Adequate levels of agreement were achieved for screener and reviewer judgements, with kappa scores ranging between 0.302 and 0.622 and positive predictive values between 50.0% and 89.7%. The criteria for AE determination showed internal consistency and face validity, as did those for preventability. CONCLUSIONS: Research into the occurrence, causation and prevention of AEs in New Zealand health care settings is methodologically feasible and meets international benchmark standards.
Authors: Helen Hogan; Frances Healey; Graham Neale; Richard Thomson; Charles Vincent; Nick Black Journal: BMJ Qual Saf Date: 2012-09 Impact factor: 7.035