Assessing the Impact of Health Facility-Based Maternal Delivery Care on Adverse Pregnancy Outcomes in Kenya: A Multi-Factor Analysis Approach Restricted; Files Only
Bruno, Beverly (Spring 2024)
Abstract
Background: Globally, approximately 810 women die daily from preventable delivery complications, with Sub-Saharan Africa accounting for 66% of these deaths in 2017. Existing literature highlights differences in adverse maternal health outcomes based on delivery setting, method, and birth attendant type.
Methods: This study used the Theoretical Domains Framework (TDF) and Capability, Opportunity, and Motivation-Behavior (COM-B) model and utilized the 1998 Kenya Demographic Health Survey (DHS) data to examine obstetric outcomes of women aged 15-49 who gave birth in the 3-5 years preceding the survey. We quantified the impact of delivery conditions and country-level variation on obstetric hemorrhage (OH) burden, providing a foundational baseline for evaluating subsequent changes despite the data’s focused scope. Birth attendants in the study were categorized according to DHS classifications into distinct groups. Those reporting excessive bleeding during delivery were categorized as having a high likelihood of experiencing OH. Additional regression and multifactorial analyses (MFA) highlight protective factors that can substantially mitigate the likelihood of adverse pregnancy outcomes.
Results: Among a representative sample of birthing Kenyan women (N=7881), significant associations were found between physician attendance during delivery and delivery method, with rates varying between 7.6% (N=111) in urban and 4.1% (N=261) in rural areas, showing increased likelihood of cesarean delivery when a doctor was present (r = 0.175, 95% CI: [0.140, 0.209], p < 2.2e-16). Regression analyses revealed increased log-odds of a doctor or nurse/midwife present when excessive bleeding was reported (β = 1.45, p = 0.016; β = 1.32, p = 0.013) with prolonged labor as a covariate. Government Maternal and Child Health Centers were associated with heightened odds of excessive bleeding for last births (β = 0.617, p = 0.024). Prolonged labor remained a significant risk factor (β = 1.273, p < 0.001). For second to last births, delivering at Private Hospitals/Clinics was linked to reduced likelihood of excessive bleeding (β = -2.4429, p = 0.045). MFA showed doctor and nurse/midwife absence contributed 13.933% and 10.676% respectively to variability, highlighting their impact on adverse pregnancy outcomes. Excessive bleeding rates were 7.5% (N=110) in urban and 8.7% (N=559) in rural areas.
Conclusions: These findings underscore the need for examination of the association between medical personnel presence and adverse pregnancy outcomes like excessive bleeding. A physician's presence may be a necessity during delivery, and crucial for some crisis management, but may also coincide with greater use of cesareans or other invasive techniques which require more resources and communication.
Table of Contents
Introduction. 1
Influence of Past Reproductive Health Policy on Obstetric Care Quality and Facility-Based Delivery Service Utilization. 11
Present-Day Obstetric Care Quality. 13
Theoretical Framework/Evidence-Based Practice Model. 20
Methods. 27
Data Source. 27
Data Preparation And Variable Selection. 30
Analysis. 33
Results. 35
Correlations between delivery attendant, delivery location, and delivery type. 35
Discussion. 41
Influence of Provider Birth Preferences on Delivery Method Choice. 41
Strengths and Limitations. 43
Future Research. 45
Conclusion. 46
Public Health Implications. 46
Appendix: Tables and Figures. 50
Table 2a. 50
Demographic Characteristics. 50
Table 2b. 51
Delivery Characteristics, Continued. 51
Table 2b. 52
Delivery Characteristics, Continued. 52
Table 2b. 53
Delivery Characteristics, Continued. 53
Table 3a. 53
Correlations for Person Who Gave Delivery Care and Delivery Method for Last and Second to Last Birth. 53
Table 3b. 54
Correlations for Place of Delivery and Delivery Method for Last and Second to Last Birth. 54
Table 3c. 54
Correlations for Person Who Gave Delivery Care and Place of Delivery for Last and Second to Last Birth. 54
Table 4a. 55
Binomial Logistic Regressions for Person Who Gave Delivery Care and Excessive Bleeding Reports (Last Birth) 55
Table 4b. 56
Binomial Logistic Regressions for Person Who Gave Delivery Care and Excessive Bleeding Reports (Second to Last Birth) 56
Table 5a. 57
Binomial Logistic Regressions for Cesarean Delivery Method and Excessive Bleeding Reports. 57
Table 5b. 57
Binomial Logistic Regressions for Vaginal Delivery Method and Excessive Bleeding Reports. 57
Table 6a. 58
Binomial Logistic Regressions for Place of Delivery and Excessive Bleeding Reports (Last Birth) 58
Table 6b. 59
Binomial Logistic Regressions for Place of Delivery and Excessive Bleeding Reports (Second to Last Birth) 59
Figure 2a. 60
Excessive Bleeding Reports by Urban-Rural Status (Last Birth) 60
Figure 2b. 60
Excessive Bleeding Reports by Urban-Rural Status (Second to Last Birth) 60
Figure 3. 61
Individual Factor Map. 61
References. 62
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Palabra Clave | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
File download under embargo until 20 May 2025 | 2024-04-29 15:27:13 -0400 | File download under embargo until 20 May 2025 |
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|