OPPORTUNITIES FOR NEONATAL HEALTH SERVICES IN ETHIOPIA

Get Complete Project Material File(s) Now! »

CHAPTER 4 QUANTITATIVE ANALYSIS, PRESENTATION AND DESCRIPTION OF THE RESEARCH FINDINGS

INTRODUCTION

In this chapter, the findings of the quantitative data are presented, interpreted, and discussed as per the analysis of the researcher-administered survey and register abstracted data. In this study, descriptive and inferential analysis methods were utilised. In the conclusion, the chapter is synthesised and key findings from the quantitative data are provided.

DATA MANAGEMENT AND ANALYSIS

Administration of questionnaire

The researcher-administered closed-ended structured survey questionnaire was applied for health workers in the primary hospitals and health centres, and health extensions workers in the health posts. In addition, structured document analysis checklist was used to collect sick newborn service statistics data from sick young infants’ registers in the primary healthcare units (Chapter 3, Section 3.4.3.1). For the data entry purpose, everyone of studied health facilities was coded with a unique identification number as per the systematically given code to their respective woreda/district, and followed by primary health care units. Therefore, the questionnaires for each facility were numbered consecutively. This helped to conduct a quality control in each of the data by checking the completeness and consistency of the questionnaire in the process of data entry and cleaning. However, study participants were not assigned any identification code.
Based on the calculated sample size (Chapter 3, Section 3.4.1), data collection was done from a total 142 health facilities (3 primary hospitals, 76 health centres and 63 health posts), and all the 15 woredas in the West Gojjam zone were represented.
Consequently, interviewer-administered questionnaires were administered to 221 health workers and health extension workers in 142 health facilities including hospitals, health centres (HCs) and health posts (HPs); 2 health workers per primary hospital and health centre, and one health extension worker per health post. In addition, 767 sick young infants’ case management service statistics were abstracted from the sick young infants’ registers in health facilities.

Data analysis

Statistical analysis programme

In this research data was analysed and presented by manipulating the quality assured data and valid percentage was taken consistently throughout the analysis. In general, data was presented in a sequenced manner according to the three main sections of the questionnaires which are: maternity services, neonatal health services and management of sick neonates, and document review for neonatal health services in the primary health care units.
After the data entry was entered by a data entry clerk by using the computer software EpiData 3.1, data was then exported to Statistical Package for Social Science (SPSS windows version 23), and STATA 15. The analysis was done by the researcher with close technical support from an experienced researcher specialized in data management and analysis. SPSS software greatly helped to run descriptive analysis including frequency; and STATA also used to recode the variables and create new variables, construct graphs and carry-out the regression multivariate analysis in response to the research objectives and associated dependant and independent variables. To consider whether the association between or among variables has statistical significance or not, a p-value less than 0.05, was considered as major test value. In most of the variables, a mean was used to report the findings with 95% confidence interval (CI).

Data cleaning, checking for completeness and consistency

In this study, data was cleaned at three different stages namely at the stage of data collection and entry by chekcign manually and employed the software before running the analysis of the data with a through manually. As adapted procedure, the data entry clerk checked all the questionnaires before commencing the data entry and those questionnaires not fulling the completeness and consistency criteria were discarded and not used for any further actions. In addition, the consistency each data was ensured after the data entry by SPSS data, by running a simple frequency, split and select analysis. For example, with the simple frequency running, if ultrasound scan for pregnant women was available at the health post level, the hard copy of the specified code of the questionnaire and entered data were verified.

RESEARCH RESULTS

Results from descriptive statistics

Descriptive statistics was used to provide a comprehensive explanation of the data, and inferential statistics was used to interpret the generalizability of the data into the general population (Polit & Beck 2010:392). The establishment of associations between variables is an important part of descriptive analysis; and explanatory analysis can be conducted on both bivariate and multivariate relationships (Blaikie 2003:120). Variable oriented analysis involves identifying the relationships among variables, often probabilistic in nature, which leads to external statistical generalization (Onwugbuzie et al 2009: 17). The analysis was presented in tables with number, percentage, frequencies, means, p-value, and confidence interval.

Characteristics of surveyed health facilities

Type of health facilities

Out of the total 142 surveyed health facilities in the West Gojjam zone, 63 (44.4%) were rural health centres, 63 (44.4%) health posts, 13 (9.2%) urban health centres, and 3 (2 %) primary hospitals (Table 4.1).
As shown in Table 4.2, all fifteen woredas/districts from West Gojjam zone are represented by the surveyed health facilities. In the simple random sampling procedure, 2 (1.4%) of health facilities were surveyed from Burie, and Finote selam town administrations, whereas 18 (12.7%), 17(12%), and 16 (11.3%) of health facilities were surveyed from Jabi Tehnan, Mecha, and Yilmana Densa woredas respectively. The two town administrations are represented by the smaller number of health facilities in the West Gojjam Zone; whereas the biggest populated woredas like Jabi Tehnan, Mecha, and Yilmana Densa have shared the bigger representation of health facilities in the zone. The simple random sampling procedure reflected the population demographic of the targeted woredas in that most of health facilities selected were from larger woredas such as Jabi Tehna, and Yilmana Densa; while, the remaining selected health facilities were from woredas with fewer number of health facilities serving a smaller population catchment.

Profile of health providers who were targeted for the researcher administered questionnaire

To address the two major sections of the researcher – administered questionnaire, 2 providers (1 for each unit) working in maternity units and under-five clinics for the management of maternal and neonatal health services were targeted for researcher-administered questionnaire in primary hospitals (PHs) and health centres (HCs); however, in the health posts (HPs), a single health extension worker (HEWs) was targeted for both sections of the questionnaire. From the total HCs and PHs assessed, 43 (31.2%) of interviewed health providers in the maternity units were had midwife diplomas, while 13 (9.4%) and 10 (7.2%) had Bachelor of Science in a Midwife degree and Nursing diploma respectively. Bachelor of Science in Nursing degree represented 1.4% and health officer 5.1%. In addition, HEWs accounted 63 (45.7%) from the total health providers interviewed for maternal health section of the questionnaire.
Regarding the roles of the interviewed health providers in the maternity units, 39 (28.3%) were maternity unit staff, 27 (19.6%) were heads of the maternity units, 61
(44.2%) were health post staff, 4.3% were heads or deputy heads of the health facility, and 3.6% were heads of the maternal and child health units of the health facility (Table 4.4).
After completing the maternal health section of the questionnaire, the second major section to address the management of neonatal health services including the management of sick newborns was handled with health providers who were working in under-five clinics in PHs and HCs; and any available HEWs in the targeted health posts.
From the total HCs and PHs assessed, 51 (35.9%) of the surveyed health providers had some kinds of Nursing qualification. More specifically, 22 (15.5%) were Health officers, 5 (3.5%) were Midwives all types and 1 (0.7%) was a Medical doctor interviewed regarding the management of sick newborns. Apparently, from the total health providers interviewed for newborn health services, 63 (44.4%) were HEWs (Table 4.5).
Additionally, 37 (26.2%) were heads of the under-five clinics, 34 (24.1%) were staff in the under-five clinics, 4 (2.8%) were staff in the maternal and child health units, 3 (2.1%) were heads of the maternal and child health units, and 1 (0.7%) was head or deputy head of the health facility. In fact, 63 (44.4%) of the interviewed health providers were health post staff who were providing full packages of the health extension programme including preventive, promotive, and selective curative care (Table 4.6).

Key maternal health services contributed to improving the outcome of newborn health

Antenatal care

The availability of antennal care (ANC) service components is summarized in Table 4.7. ANC service was provided in 129 (95.6%) of surveyed health facilities in West Gojjam zone. Most of ANC components were available more 90 percent. The provision of bed nets for malaria prevention and presumptive deworming for pregnant women during ANC visits were 57(44.2%) and (62.8%) respectively.
Other ANC service components that are supposed to be provided at health centres and hospitals, such as haemoglobin, urine protein, urine sugar and syphilis screening tests, were available in 58 (73.4%), 60 (75.9%), 38 (48.1%) and 73 (92.4%) health facilities respectively. The percentage mean score from the 22 components of ANC services shows that, 87.9 [95%CI: 84.883-90.874], 87.4 [95%CI: 84.507-90.317] and 82.8 [95%CI: 80.750-84.906] were available respectively in primary hospitals (PHs), urban health centres (UHCs), and rural health centres (RHCs). Considering, the availability of 15 expected components of ANC services in the health posts (HPs), the percentage mean score was 50.9 [95%CI: 44.051-57.747]. It shows that, the percentage mean score of ANC service components in HPs level was much lower than PHs and health centres (HCs). Moreover, the mean availability of the ANC service components, more available in PHs and HCs as compared with HPs.

Basic Emergency Obstetric and Newborn Care (BEmONC)

In general, all (79) PHs, urban and rural health centres provided delivery services. Even though HPs are not expected to provide delivery services, 11 (17.5%) of HPs offered clean and safe delivery services. In addition, monitoring and management of labour by using partograph was available in 74.7% of HCs and hospitals. The availability of the basic emergency obstetric and newborn care (BEmONC) service in the PHs and HCs was assessed by asking the basic seven signal functions including parenteral administration of antibiotics, oxytocics, and anticonvulsants; and applying manual removal of placenta, retained products, assisted vaginal delivery, and neonatal resuscitation. The administration of parenteral oxytocics was provided in 65 (84.4%) of health facilities at least once during the six months before the survey, followed by 63 (81.8%) and 54 (79.4%) of health facilities applied manual removal of placenta and carried-out assisted vaginal delivery respectively. However, among the seven basic signal functions, 40 (54.1%) of health facilities administered parenteral anticonvulsants (Table 4.8). The signal functions were less practiced in RHCs when compared to UHCs and PHs. In addition to the seven basic BEmONC signal functions, 60 (76.9%) of health facilities applied intravenous fluid administration and 53 (73.6%) administered injectable quinine or artemether for complicated malaria at least once during the six months before the survey.
As shown in (Figure 4.1) the provision of BEmONC signal functions at least three months before the survey was also assessed. Among the HCs and PHs facilities that provided delivery services, 62 (95.4%) performed parenteral uterotonics and 40 (95.2%) performed removal of retained products. Consistently, within the three and six months before the survey, among the seven signal functions, the least performed basic signal function was parenteral anticonvulsants (54.1% and 72.2%) and the highest performed was parenteral uterotonics (84.4% and 95.4%).
As illustrated in Figure 4.2, the mean was computed for the readiness of health facilities to provide BEmONC signal functions, and the provision of BEmONC signal functions before three and six months of the survey. In terms of the mean readiness of health facilities to provide BEmONC signal functions, it was found that PHs were ready to provide all seven signal functions; however, the readiness mean score for the UHCs and RHCs shows 6.8 [95%CI: 6.64-7.0] and 6.5 [95%CI: 6.224-6.799] respectively. Regarding the provision of the seven BEmONC signal functions in last three months before the survey, PHs had the highest score, 6.7 [95%CI: 6.007-7.325] and RHCs had lowest mean score, 3.5 [95%CI: 2.980-4.035]. This shows that RHCs provided only half of the signal functions in the last three months. The mean score of the UHCs was 6.2 [95%CI: 5.208-7.099], which is slightly lower than the mean score of PHs. The mean score in the provision of BEmONC signal functions in last six months was also computed; PHs had a mean score of 6.7 [95%CI: 6.007-7.325], 6.461 [95%CI: 5.698-7.224] and 4.4 [95%CI: 3.933-4.896] signal functions mean scores were respectively for UHCs and RHCs. The comparison of the mean scores in the three and six months shows that RHCs had a slight better experience of provision of the seven signal functions in the last six months than the three months before the survey. In three parameters including health facility readiness to provide BEmONC signal functions, and provisions of BEmONC signal functions in last three and six months, RHCs were only providing about 3-4 BEmONC signal functions, and not completely ready to provide all the seven signal functions. In contrast, PHs and UHCs were in a better position to provide BEmONC signal functions.
As a part of the immediate postpartum care in the management of third stage labour, controlled cord traction, oxytocin injection on the thigh within 1 minute after delivery and uterine massage after the delivery were always practiced in 73 (90%), 73 (90%) and 69 (87.3%) of assessed PHs, UHCs and RHCs The mean score from the three expected actions for the active management of third stage labour show that PHs carried out all three actions; followed by the RHCs with the mean score of 2.7 [95%CI: 2.549-2.879]; and UHCs with the mean score of 2.7 [95%CI: 2.344-3.04]. Overall the respondents’ response shows that the experience in the active management of third stage labour was found to be a common practice in all types of health facilities.

Availability of trained skilled birth attendances in the health facilities

To understand the availability of trained skilled birth attendants in health facilities, total numbers of health professionals who had direct exposure in attending delivery including all types of doctors, midwifes and nurses, and health officers and their training status on BEmONC or CEMONC were collected from each health facility. Consequently, the percentage of trained skilled birth attendants’ availability shows that urban and rural health centres had 25.2 [95%CI: 17.365-33.101] and 20.9 [95%CI: 17.365-33.101] percent of trained skilled birth attendants in their respective health facilities respectively. Despite 4.1 percent of skilled birth attendants were available in the hospitals, the CI intervals crossing the point of 0 [95%CI: -2.016-10.312].

Neonatal health care services

Emergency Newborn care (EmNeC)

Emergency newborn care (EmNeC) was assessed by asking the four signal functions including the neonatal resuscitation, kangaroo mother care for premature or very low birth weight, injectable antibiotics for neonatal sepsis, and corticosteroids in preterm labour. The highest score was for newborn resuscitation with bag and mask with 56 (71.8%) and followed by injectable antibiotics for newborn sepsis with 85 (61.6%) of the health facilities. However, corticosteroids for preterm labour was only practiced in 5 (6.5%) HCs and PHs. It is also the lowest score in all signal functions (Table 4.9). Likewise, only 6 (7.7%) of health facilities were administered intravenous fluids for the newborn and 41 (30%) of health facilities including HPs were teaching mothers to express breast milk and feed with small cup/spoon if the newborn is unable to feed. As per the report of 11 (18.3%) HPs, injectable antibiotics for the management of newborn sepsis was not practiced because of the lack of cases.
As it shown in the Figure 4.3, in addition to the above five actions shown in the Table 4.9, intravenous fluids for newborns and newborn resuscitation with bag and mask with oxygen were also included and used to compute the mean provision of EmNEC signal functions in last six months. From the seven signal functions, PHs had a higher mean score, 6.3 [95%CI: 6.007-7.325] and RHCs had a lowest mean score, 2.3 [95%CI: 2.043-2.623] in the provision of EmNeC signal functions in the last six months before the survey. On the other hand, the UHCs mean score (4.2 [95%CI: 3.517-4.944] was below the PHs and higher than the UHCs. It implies that, PHs were providing more than 6 EmNEC signal functions from the seven expected; in the contrary, UHCs and RHCs were providing slightly more than 4 and 2 signal functions respectively. The likely of getting EmNeC signal functions to save the lives of the sick young infants in the RHCs were critically low and was better provided in PHs.

Essential immediate newborn care

Health workers were asked about the experience of essential newborn care practices including breastfeeding, applying antibiotics in the eye, bath, cord care and weighing the baby. The application of chlorohexidine (CHX) in the umbilical cord and putting the baby in the breast within one hour of delivery practices were reported by 28 (35.4%) and 34 (43%) of health facilities respectively. However, the rest of essential and immediate newborn care services availability including putting the baby on the abdomen of the mother once the baby is delivered, delaying bathing baby for thermal protection, applying of antibiotics in the eyes and taking the weight of the baby were reported to be more than 90% (Table 4.10). In addition, 4 (6.3%) of rural health centres were also applying gentian violet in the umbilical cord.
In addition, the mean score was computed from the seven immediate actions for essential newborn care. From Table 4.10, nothing applied in the cord/dry cord care and or apply chlorhexidine (CHX jel) the umbilical cord of the baby was merged into one variable and in the same scenario, first put the baby on the breast immediately after delivery and first put the baby on the breast within one hour of delivery was considered both are correct and recoded into one variable.
As per the expectations, the PHs were providing all the seven-immediate essential newborn actions (mean score 7), and similarly both urban and rural health centres were providing the essential mean score of 6.8 [95%CI: 6.527-7.011] and 6.8 [95%CI: 6.607-6.916] respectively from the seven expected actions. It implies that, the respondents of the health providers in the maternity ward confirmed that almost all essential newborn care actions were practiced for newborns immediately after birth as a routine care (Figure 4.4).

TABLE OF CONTENTS
CHAPTER 1  ORIENTATION TO THE STUDY
1.1 INTRODUCTION
1.2 BACKGROUND INFORMATION ABOUT THE RESEARCH PROBLEM
1.3 RESEARCH PROBLEM
1.4 AIM OF THE STUDY
1.5 SIGNIFICANCE OF THE STUDY
1.6 DEFINITION OF CONCEPTS
1.7 THEORETICAL FOUNDATIONS OF THE STUDY
1.8 RESEARCH DESIGN AND METHOD
1.9 ETHICAL CONSIDERATIONS
1.10 SCOPE OF THE STUDY
1.11 STRUCTURE OF THE THESIS
1.12 CONCLUSION
CHAPTER 2 LITERATURE REVIEW
2.1 INTRODUCTION
2.2 NEWBORN HEALTH IN ETHIOPIA
2.4 Determinants of neonatal mortality
2.3 OPPORTUNITIES FOR NEONATAL HEALTH SERVICES IN ETHIOPIA
2.4 CONCLUSION
CHAPTER 3 RESEARCH DESIGN AND METHOD
3.1 INTRODUCTION
3.2 RATIONALITY OF MIXED METHOD APPROACH
3.3 RESEARCH DESIGN
3.4 RESEARCH METHOD
3.5 INTERNAL AND EXTERNAL VALIDITY OF THE STUDY
3.6 ETHICAL ISSUES RELATED TO SAMPLING AND DATA COLLECTION
3.7 CONCLUSION
CHAPTER 4  QUANTITATIVE ANALYSIS, PRESENTATION AND DESCRIPTION OF THE RESEARCH FINDINGS
4.1 INTRODUCTION
4.2 DATA MANAGEMENT AND ANALYSIS
4.3 RESEARCH RESULTS
4.4 OVERVIEW OF RESEARCH FINDINGS
4.5 CONCLUSION
CHAPTER 5  QUALITATIVE ANALYSIS, PRESENTATION AND DESCRIPTION OF THE RESEARCH FINDINGS
5.1 INTRODUCTION
5.2 DATA MANAGEMENT AND ANALYSIS
5.3 RESEARCH RESULTS
5.4 OVERVIEW OF RESEARCH FINDINGS
5.5 CONCLUSION
CHAPTER 6  THE INTEGRATION OF QUANTITATIVE AND QUALITATIVE RESULT
6.1 INTRODUCTION
6.2 MIXED-METHOD BACKGROUND
6.3 RESEARCH QUESTIONS
6.4 DATA COLLECTION PROCESS
6.5 MIXED-METHOD ANALYSES
6.6 FINDINGS OF THE INTEGRATION OF THE MIXED METHOD RESEARCH RESULT
6.7 OVERVIEW OF THE INTEGRATION OF THE RESULTS
6.8 CONCLUDING REMARKS
6.9 CONCLUSION
CHAPTER 7 PROPOSED GUIDELINES TO IMPROVE THE NEONATAL HEALTH CARE SERVICES IN THE PRIMARY HEALTH CARE UNITS IN THE NORTH-WEST OF ETHIOPIA
7.1 INTRODUCTION
7.2 RESEARCH METHODS
7.3 THE NEED FOR DEVELOPING THE GUIDELINES
7.4 GUIDING PRINCIPLES
7.5 OBJECTIVES
7.6 SCOPE
7.7 METHODOLOGY
7.8 STRATEGIC OBJECTIVES AND ACTIVITIES
7.9 MONITORING AND EVALUATION
7.10 CONCLUSION
CHAPTER 8 CONCLUSION, RECOMMENDATIONS AND LIMITATIONS OF THE STUDY
8.1 INTRODUCTION
8.2 RESEARCH DESIGN AND METHOD
8.3 CONCLUSIONS
8.4 RECOMMENDATIONS
8.5 CONTRIBUTION OF THE STUDY
8.6 LIMITATION OF THE STUDY
8.7 STRENGTH OF THE STUDY AND IMPLICATION OF THE FUTURE RESEARCH
8.8 CONCLUDING REMARKS
LIST OF REFERENCES
GET THE COMPLETE PROJECT