Qualidade na Atenção Primária Acesso livre

Abstrato

A Quality Initiative to Implement a Managed Sepsis Protocol in a Public Hospital Based on the IHI Quality Improvement Model: Experience Report

Leidy Katerine Calvo Nates, Adriano J Pereira, Antônio C Neto, Eliezer Silva

Background: Although 13 years have passed since the first publication of the international treatment guidelines, sepsis continues to be the leading cause of death in Brazilian Intensive Care Units (ICU), with lethality rates of 55%. The main causes of such high rates are poor adherence to the Surviving Sepsis Campaign (SSC) guidelines and insufficient implementation of the measures indicated in the care package. The present study reports our experience in implementing a Managed Sepsis Protocol (MSP) in a public hospital.

Methods: In order to improve the care for septic patients, we based our actions on the Quality Improvement Model (QIM), the methodology widespread by the Institute for Healthcare Improvement (IHI-USA), using the Plan-Do-Study- Act (PDSA) cycle to deploy and standardize processes.

Results: Main obstacles during in the implementation of a Managed Sepsis Protocol (MSP) were: low levels of staff engagement with setting of poor priorities; inadequate notification of new sepsis cases and data collection; lack of automatic alarms system for sepsis detection and of a reference multidisciplinary team, unclear patient flow and definition of team member roles. After a series of interventions focused on the standardization of sepsis management processes, we observed 70% compliance with the SSC care-bundle and that the reporting of suspected cases increased by 60%. In addition, the time interval between the opening of the MSP for a given patient and the arrival of the initial standard sepsis-evaluating lab tests was shortened to 30 min.

Conclusion: We conclude that the IHI quality-improvement model seems to be a suitable tool to implement sepsis management protocols.

Isenção de responsabilidade: Este resumo foi traduzido usando ferramentas de inteligência artificial e ainda não foi revisado ou verificado