Background
Fungal sepsis is a serious healthcare-associated infection that poses a significant threat to newborns, particularly those born prematurely or with low birth weight. Despite advances in neonatal intensive care, fungal infections remain associated with prolonged hospitalization, major complications, and high mortality.
Early recognition and timely treatment are critical to improving outcomes. However, identifying which neonates are most at risk can be challenging because multiple clinical and treatment-related factors often interact throughout hospitalization.
Although fungal sepsis has been widely studied internationally, local evidence from the Philippines, especially from Mindanao, remains limited. This study sought to address that gap by examining the clinical characteristics, treatment practices, and in-hospital outcomes of neonates with fungal sepsis while identifying factors associated with adverse clinical outcomes.
The Research Challenge
The research team sought to describe the clinical profile and treatment practices of neonates with culture-confirmed or clinically managed fungal sepsis and determine which demographic, clinical, and treatment-related factors were associated with adverse outcomes.
The study explored questions such as:
- Which newborns are at greater risk of mortality and major complications from fungal sepsis?
- What clinical and treatment-related factors influence the length of hospital stay?
- How can retrospective hospital records be used to identify independent predictors of poor outcomes?
- Which statistical approaches are most appropriate for evaluating multiple interacting risk factors?
Beyond describing patient characteristics, the project addressed a broader methodological challenge frequently encountered in retrospective medical research: how can researchers draw reliable conclusions from complex hospital data while accounting for missing information, multiple predictors, and different types of clinical outcomes?
Data2Stats’ Role
Data2Stats provided comprehensive statistical support throughout the project, helping transform retrospective hospital records into meaningful evidence that can support neonatal care and future clinical research.
The team supported the study through:
- Research consultation and statistical planning;
- Development of the statistical analysis plan;
- Data management planning, validation, and quality assessment;
- Missing data assessment and handling;
- Statistical analysis and interpretation of findings;
- Preparation of publication-ready statistical tables;
- Technical review of statistical methods and manuscript; and
- Statistical reporting support.
By ensuring that the data were carefully managed and analyzed using appropriate statistical techniques, the project was designed to generate reliable evidence on neonatal fungal sepsis and its associated outcomes.
Statistical Methods Used
To address the research objectives, several statistical methods were incorporated into the study.
Descriptive statistics were used to summarize demographic characteristics, clinical features, treatment practices, and in-hospital outcomes among neonates with fungal sepsis.
Normality was assessed using the Shapiro-Wilk test to guide the selection of appropriate statistical analyses. Depending on the characteristics of the data, Chi-square tests, Fisher’s Exact tests, Independent t-tests, and Mann-Whitney U tests were performed to examine differences between patient groups.
To identify independent predictors of mortality and major complications, multivariable logistic regression was employed, with Variance Inflation Factors (VIF) used to assess multicollinearity and the Hosmer-Lemeshow test used to evaluate model fit.
Length of hospital stay was analyzed using either linear regression or generalized linear modeling, depending on the distribution of the outcome variable, ensuring that each analysis appropriately reflected the underlying data.
What the Data Revealed
The study protocol has been developed to generate comprehensive local evidence on neonatal fungal sepsis among Filipino newborns admitted to intensive care.
Upon completion, the analyses are expected to describe the clinical characteristics and treatment practices of affected neonates while identifying factors associated with mortality, major complications, and prolonged hospitalization.
Rather than examining individual risk factors in isolation, the study applies multivariable statistical methods to determine which variables independently influence clinical outcomes after accounting for other contributing factors.
The findings are expected to provide clinicians with evidence that supports earlier recognition of high-risk patients, more informed treatment decisions, and improved management of neonatal fungal sepsis within local healthcare settings.
A Valuable Lesson in Clinical Research
Retrospective hospital records contain a wealth of clinical information, but transforming that information into reliable evidence requires more than simply comparing groups.
This study highlights an important lesson in medical research:
Understanding clinical outcomes requires looking beyond individual risk factors.
Neonatal outcomes are rarely influenced by a single characteristic. Prematurity, birth weight, treatment interventions, underlying conditions, and other clinical variables often interact in complex ways.
While descriptive analyses and simple statistical tests help identify potential patterns, multivariable regression allows researchers to determine which factors remain independently associated with adverse outcomes after accounting for other variables.
Equally important is planning for missing data, selecting statistical methods appropriate for each type of outcome, and validating regression models. Together, these steps strengthen the reliability of research findings and support more evidence-based clinical decision-making.
Outcome and Impact
The study protocol has been developed to examine the epidemiology, clinical characteristics, treatment practices, and in-hospital outcomes of neonatal fungal sepsis within a local hospital setting.
More importantly, the project demonstrates the value of rigorous statistical support in helping researchers:
- Analyze complex retrospective clinical data;
- Identify independent predictors of adverse outcomes;
- Apply appropriate statistical methods for different types of clinical endpoints; and
- Generate evidence that can guide future neonatal care and research.
Upon completion, the findings are expected to contribute valuable local evidence that supports earlier recognition of high-risk neonates, improves clinical decision-making, and strengthens the management of neonatal fungal sepsis in the Philippines.
Key Takeaways
- Retrospective hospital records can provide valuable insights into neonatal outcomes when analyzed using appropriate statistical methods.
- Neonatal fungal sepsis is influenced by multiple interacting demographic, clinical, and treatment-related factors.
- Multivariable regression helps distinguish independent predictors from simple associations.
- Careful statistical planning, data validation, and regression diagnostics improve the reliability of clinical research.
- Local evidence is essential for informing future neonatal care, infection control practices, and evidence-based treatment strategies.
At Data2Stats, we believe that statistical analysis is about more than identifying risk factors. It is about transforming complex clinical data into meaningful evidence that helps healthcare professionals make informed decisions, improve patient care, and advance neonatal health research.