To study the frequency and types of suboptimal care and medical errors in children who died of severe bacterial infection as the first-stage procedure intended to improve quality of care.
Population-based confidential inquiry.
Two adjoining administrative districts in France.
Children older than 3 months dead from severe bacterial infection from 2000 through 2006.
The medical files were summarized on standardized forms and then evaluated independently by two experts, who determined whether the initial management before the patients’ arrival in intensive care was or was not optimal, in comparison with current guidelines.
Of 23 deaths from severe bacterial infection, 21 could be analyzed; management was considered suboptimal in 76%. The coefficient of agreement between the experts was high, with a weighted κ of 0.73. The types of errors identified included parental delay in seeking medical care (33%; 95% confidence interval, [12–54]), physicians’ delay in administering appropriate treatment (antibiotic therapy in the case of purpura; 38%; 95% confidence interval, 16–60), insufficient doses of or failure to repeat fluid resuscitation (24%; 95% confidence interval, [9 –35]), and overall underestimation of disease severity (38%; 95% confidence interval, [16–60]).
This study found a high frequency of suboptimal care in the initial management of children who died of severe bacterial infection, with four separate types of errors. Other studies are needed to assess the potential avoidability of this type of death.
From the Clinique médicale pédiatrique (EL), Urgences pédiatriques (CG-LG), and Réanimation pédiatrique et néonatale (J-CR), Hôpital de la Mère et de l’Enfant, CHU Nantes, Nantes, France; Unité d’urgences pédiatriques et de maladies infectieuses (AM), Hôpital R. Salengro, CHU de Lille, Université de Lille-Nord de France, Lille, France; Association pour le Recherche et l’Enseignement en Pédiatrie Générale (RA), Cabinet de Pédiatrie, Issy-les-Moulineaux, France; Service de pédiatrie générale (TB), Centre Hospitalier Départemental de la Roche-sur-Yon, France; Unité INSERM U953 (NM, M-HB-C, MC), Recherche épidémiologique en santé périnatale et santé des femmes et des enfants, Paris, France; CépiDc-Inserm (AA), Centre d’épidémiologie sur les causes médicales de décès, Le Vésinet, France; and Service de pédiatrie générale (MC), Hôpital Saint-Vincent-de-Paul, Paris Descartes University, AP-HP, Paris, France.
Drs. Launay and Gras-Le Guen contributed equally to this work.
The authors have not disclosed any potential conflicts of interest.
For information regarding this article, E-mail: firstname.lastname@example.org