Implementation of a Sepsis Chart and Its Effects on Mortality Rate in Indonesian Primary Referral Hospitals

Background

Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection. If not recognized and treated promptly, sepsis may progress to septic shock, resulting in multiple organ failure and death. Sepsis is a serious complication of infectious diseases and remains one of the leading causes of death worldwide, particularly in low- and middle-income countries.

Globally, an estimated 48.9 million cases of sepsis occurred in 2017, with approximately 11 million sepsis-related deaths, accounting for 19.7% of all global deaths. Data from 16 Asian countries reported an intensive care unit mortality rate of 44.5% among patients with sepsis. At Dr. Cipto Mangunkusumo National Referral Hospital (RSCM), a 2017 study reported that 64% of adult patients with infection had sepsis, defined as a Sequential Organ Failure Assessment (SOFA) score >2. Another study conducted between 2016 and 2022 reported a mortality rate of 69% among patients with sepsis at RSCM, with a higher risk of mortality among patients with metabolic comorbidities. Data from the Community Services Coordinator of the Department of Internal Medicine at RSCM also showed that 10.3% of internal medicine inpatients were diagnosed with sepsis.

The One-hour Sepsis Bundle is a set of interventions recommended to be initiated within the first hour of suspected sepsis. These interventions include measurement of serum lactate, collection of blood cultures, administration of broad-spectrum antibiotics, fluid resuscitation, and hemodynamic management, with the aim of improving tissue perfusion and achieving early infection control.

Although several studies have demonstrated the effectiveness of the One-hour Sepsis Bundle in reducing mortality, its implementation at RSCM has not been systematically evaluated. Currently, there is no specific sepsis scoring system integrated into the Hospital Information System (HIS) or clinical pathway for patients with sepsis. In addition, a review by the Medical Committee indicated that the One-hour Sepsis Bundle had not been implemented in the majority of sepsis cases presenting to the Emergency Department (ED).

Previous studies have demonstrated that the use of sepsis forms and implementation of sepsis codes may reduce mortality and length of hospital stay. One approach to improving implementation is Koran Sepsis, a paper-based order set designed to guide healthcare professionals in delivering the One-hour Sepsis Bundle. This study aims to evaluate the effectiveness of Koran Sepsis implementation on short-term mortality, defined as 7-day mortality, among patients with sepsis at during the period of September 2026 - 2027

Objectives

The primary objective of this study is to evaluate the effectiveness of Koran Sepsis implementation in improving adherence to the One-hour Sepsis Bundle and reducing 7-day mortality among patients with sepsis.

Specifically, this study aims to determine the proportion of patients receiving initial sepsis management according to the One-hour Sepsis Bundle prior to Koran Sepsis implementation and to determine the proportion of healthcare professionals adhering to the One-hour Sepsis Bundle following its implementation. The study will also compare adherence to the One-hour Sepsis Bundle before and after Koran Sepsis implementation and assess differences in 7-day mortality between the two periods.

In addition, this study aims to evaluate the association between individual components of the One-hour Sepsis Bundle, including serum lactate measurement, blood culture collection, administration of broad-spectrum antibiotics, fluid administration, and vasopressor administration, and 7-day mortality among patients with sepsis.

Study Design

This study is an observational study using a pragmatic stepped-wedge cluster design. The participating hospitals will transition sequentially from the standard-of-care period to the sepsis chart implementation period. The sequence of hospital implementation will be determined according to the order in which approval are obtained.

Sample Size Calculation

Based on the estimated number of participating hospitals, the number of steps in the stepped-wedge design, and the anticipated difference in proportions, the sample size was calculated using the formula proposed by Hemming et al., resulting in a total sample size of 304 patients from 2 hospitals.

Study Procedures and Data Collection

Patients meeting the operational definition of sepsis will be recruited consecutively during the study period. Their clinical management during the initial hour of sepsis care will be documented, including the implementation of each component of the One-hour Sepsis Bundle. Patients will subsequently be followed for seven days to determine their mortality outcome.

Data will be collected using a case report form completed by clinicians or trained research personnel based on information obtained from the patients' medical records. Information regarding sepsis management will be obtained from the electronic health record and Koran Sepsis documentation. All collected data will be processed anonymously to maintain patient confidentiality.

Study Variables and Parameters

The primary outcome of this study is 7-day mortality, defined as death occurring during hospitalization within seven days following admission.

Secondary outcomes include adherence to the One-hour Sepsis Bundle and the time to administration of broad-spectrum antibiotics. We will assess adherence to the One-hour Sepsis Bundle based on documentation of the relevant interventions in the electronic health record and Koran Sepsis. The components assessed will include serum lactate measurement, blood culture collection before antibiotic administration, administration of broad-spectrum antibiotics, administration of 30 mL/kg crystalloid fluid, and vasopressor administration when fluid resuscitation is inadequate or unsuccessful. The time to administration of broad-spectrum antibiotics will be defined as the time elapsed between sepsis diagnosis and antibiotic administration.

Patient-level characteristics that will be collected include age, sex, Charlson Comorbidity Index (CCI), lactate-modified qSOFA score, SOFA score, presence of septic shock, source of infection, and identified pathogen. Age will be categorized as 18-59 years and ≥60 years. The CCI will be calculated based on the patient's documented comorbidities and subsequently categorized as high or low according to the optimal cut-off derived from receiver operating characteristic (ROC) curve analysis. The lactate-modified qSOFA score will be determined based on clinical and laboratory parameters and categorized as high or low according to the cut-off derived from ROC curve analysis.

Hospital-level characteristics will also be collected to account for potential differences between participating centers. These variables will include the total number of inpatient admissions per month, total number of ED visits per month, number of inpatient beds, and the types of pathogens commonly identified at each hospital.

The primary exposure of interest is the implementation of Koran Sepsis. The primary clinical outcome is 7-day mortality. Adherence to the One-hour Sepsis Bundle will be evaluated as an implementation-related outcome and will also be assessed in relation to mortality. Potential confounding variables will include patient-level characteristics and hospital-level factors.

Data and Statistical Analysis

Data will undergo data entry, editing, coding, processing, and cleaning. Data quality will be assessed by identifying missing values, inconsistencies, and implausible variations before statistical analysis.

Statistical analyses will be performed using STATA version 15.1. Numerical variables will be presented as mean ± standard deviation (SD) for normally distributed data or as appropriate measures of central tendency and dispersion when the data are not normally distributed. Categorical variables will be presented as frequencies and percentages.

The primary and secondary outcomes will be analyzed using multilevel regression models to account for the clustering of observations within participating hospitals. The analysis will incorporate hospital-level random effects to account for between-hospital variability and intracluster correlation. The stepped-wedge design will also be taken into consideration in the analysis by accounting for the study period and intervention status.

This approach is intended to control for differences between participating hospitals and minimize potential bias arising from correlation among patients treated within the same hospital. Multivariable analysis will additionally be performed to adjust for relevant patient-level and hospital-level covariates. The association between individual components of the One-hour Sepsis Bundle and 7-day mortality will also be assessed. Relative risks will be calculated to compare mortality between patients who received management consistent with the One-hour Sepsis Bundle and those who did not.

Benefits

This study is expected to contribute to improving the quality of sepsis care at major hospitals in Indonesia by providing evidence regarding the implementation of the One-hour Sepsis Bundle in routine clinical practice.

The study may also identify specific aspects of early sepsis management that can be optimized to reduce mortality among patients with sepsis. Furthermore, the findings are expected to provide scientific evidence regarding the effectiveness of the One-hour Sepsis Bundle and Koran Sepsis, which may support the development of an effective and efficient system for the implementation of standardized sepsis management at participating hospitals.