Ethnic differences in adverse events in Dutch hospital care? A record review study.

Objectives: US studies showed an increased risk of adverse events (AEs) in hospital care among ethnic minorities, but in Europe ethnic inequities in patient safety have never been analysed. In the Netherlands, approx.12% of the population is of non-Western ethnic origin. Because ethnic inequities in patient safety have been found in the US, and potential inequities in quality of care have been found in Europe, there is an urgent need to know whether ethnic inequities in quality of care exist in the Netherlands. The objective of this study is to analyse incidence, type, nature, impact and preventability of AEs by ethnic origin. Methods: We conducted a prospective mixed methods cohort study in four Dutch urban hospitals. Inclusion criteria: Dutch or non-western ethnic origin, age range 45-75 and admission to hospital for at least 1 night. Eligible patients completed a questionnaire at hospital admission and signed a consent form for record review. The questionnaire contained items on ethnic origin (based on country of birth criteria) and on potential determinants of ethnicity-related risks of AEs (e.g. language and health literacy). After discharge, a two-stage medical record review study was conducted.The method of determining AEs was comparable with those of other international studies. First, a nurse screened the record for the presence of 16 triggers (e.g. an adverse drug reaction or an unplanned readmission after discharge from the index admission). In case one or more triggers were present, a medical specialist reviewed the record to determine whether an AE was present, and if yes, to assess causes, preventability, location, timing, classification, most responsible discipline, patient and admission characteristics and the quality of the record. We compare incidence, type (e.g. diagnostic AEs), impact, and nature of AEs between ethnic Dutch patients and patients of non-western ethnic origin, while adjusting for case mix (age, sex and primary diagnosis). Subsequently, we analyse the contribution of patient related determinants to the risk differences. Results: 1380 patients who met the inclusion criteria were willing to participate (790 Dutch, 590 of non-western origin). Record review is currently being finalised; over 1300 records have been reviewed up to now. Preliminary analysis suggested that AE rates do not differ significantly between groups, but that AEs in non-western patients may be often preventable. Also, slight differences in types of AEs may be present: more surgical AEs in Dutch patients versus more diagnostic AEs in non-western patients. No firm conclusions can be drawn yet because of incomplete data. Results of the complete dataset will be available by June 2013. Conclusion: This study quantifies differences in AEs between Dutch and non-western patients in hospital care and provides insight in potential ethnicity-related differences in patient characteristics which determine in the incidence, nature and preventability of AEs.