Home
Resource
Utilizing robust statistical methods for maximum likelihood estimation in clinical informatics for obstetrics research in the community hospital setting
PDF
PDF format is widely accepted and good for printing.
Plug-in required
PDF-1(918.29 kb)
Citation & Export
View Usage Statistics
Staff View

Citation & Export
Hide

Simple citation

Martingano, Daniel JS. Utilizing robust statistical methods for maximum likelihood estimation in clinical informatics for obstetrics research in the community hospital setting. Retrieved from https://doi.org/doi:10.7282/t3-gfpa-vw17

Export

Click here for information about Citation Management Tools at Rutgers.

Statistics
Hide

Description

TitleUtilizing robust statistical methods for maximum likelihood estimation in clinical informatics for obstetrics research in the community hospital setting
NameMartingano, Daniel JS (author); Mitrofanova, Antonina (chair); Vyas, Riddhi (member); Coffman, Frederick (member); Rutgers University; School of Health Professions
Date Created2022
Other Date2022-05 (degree)
SubjectMedicine, Mathematics, Biostatistics, Antibiotics, Azithromycin, Chorioamnionitis, Community hospital, High risk pregnancy, Obstetrics
Extent74 pages
DescriptionBACKGROUND:
Research in the field of obstetrics can be very challenging because the nature of studying treatments and interventions in pregnancy patients poses several ethical and practical limitations. Pregnancy research in the community hospital setting provides unique challenges but so too provides much needed answers to niche clinical problems, necessitating understanding, development, and implementation of statistical methods that are best suited to this scenario.
OBJECTIVE:
The objective of this dissertation is to provide a solution to these limitations by defining and implementing a statistical framework utilizing robust methods for conducting high-quality, practical research in the field of obstetrics and prove its effectiveness though original research with novel statistical methods.
METHODS:
Two prospective observational cohort studies were conducted: one evaluating antibiotic regimens in the management of preterm prelabor rupture of membranes and the other studying alternative antibiotic regimens for surgical prophylaxis for non-elective cesarean deliveries during the COVID-19 pandemic. The results of both studies were analyzed using robust statistical analysis to yield original results.
RESULTS:
The first study demonstrated a decreased risk was noted for the development of clinical chorioamnionitis (p=0.003), neonatal sepsis (p<0.001), and postpartum endometritis (p=0.010) when comparing azithromycin to erythromycin regimens. Pregnancy latency by regimen was not significantly different (p=0.90). The second study demonstrated that patients receiving clarithromycin had significantly lower rates and a decreased risk of postpartum endometritis as compared to those who did not receive adjunct prophylaxis (p=0.034). When evaluating robust statistical methods, the recommended statistical analysis framework for generalized linear regression and survival analysis under these unique circumstances includes Welch two-sample t-tests for continuous variables, G-Test and Fischer’s exact test for categorical variables, Quasi-likelihood Poisson regression with robust error variance, robust Cox proportional hazards model, Aalen-Johansen estimator with IJ variance for survival curve, and direct approach to adjusted survival curves.
CONCLUSION:
Utilizing this novel approach to statistical analysis as demonstrated by original research in this dissertation proposal for clinical research in high risk obstetrics in the community hospital setting may provide more accurate and appropriate results.
NotePh.D.
NoteIncludes bibliographical references
Genretheses
Persistent URLhttps://doi.org/doi:10.7282/t3-gfpa-vw17
LanguageEnglish
CollectionSchool of Health Professions ETD Collection
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Version 8.5.5
Rutgers University Libraries - Copyright ©2024