Project Topics
Get Complete Project Material File(s) Now! »
CHAPTER 3 – RESEARCH DESIGN AND METHOD
INTRODUCTION
The main findings of the literature review indicate that in Mozambique the MWH strategy is one of the priorities of the Ministry of Health. However, considerable limitations exist regarding the availability of records regarding the activities of this service. Within this context, can be highlighted the absence of both quantitative and qualitative assessments of the impact of MWHs in increasing institutional deliveries. In addition, the literature review has noted the increasing potential of GIS technology in Public Health, as a powerful tool to assess health interventions and programs.
Considering these aspects, the research design and method proposed in this study is coherent with the research purpose: To assess the effectiveness of MHWs in increasing coverage of institutional deliveries using Geographical Information Systems (GIS) in six districts of Cabo Delgado Province (Mozambique).
Within this context, the main research paradigm is described, considering the aim of the study, but also considering the availability of MWHs data in Mozambique. In this chapter, the research method is introduced, with special attention to the sampling process, the determination of the inclusion and exclusion criteria, the data collection approach and method, the data analysis and the internal and external validity of the study.
RESEARCH DESIGN
The current research is based on a quantitative paradigm, as it is an objective to use a formal and systematic process that uses numerical data in order to obtain information about the effectiveness of the MWHs in a determined geographical region under study (Burns and Grove, 2005:23). Quantitative research can be used to describe variables, examine relationships among variables and also, to examine cause-and-effect interactions between variables. In this case, numerical data was provided from three main sources of Cabo Delgado Province: Provincial Health Information System, Geo-coded administrative and Geographical data and Population Census.
In this study, the researcher did not intend to change the course of events, but rather to gather data in relation to a phenomenon of interest. Consequently, it should be defined as an observational study in which the researcher measures, records, counts, observes or classifies events (Stommel and Wills, 2004:118). In order to ensure good reporting, the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement has been followed (Ven Elm et al, 2007:806-808). This statement defines a checklist of items that should be included in observational studies, namely: (1) Introduction: background and objectives; (2) Methods: study design, setting, participants, variables, data sources, bias, study size, quantitative variables and statistical methods; (3) Results: participants, descriptive data, outcome data and main results; (4) Discussion: key results, limitations, interpretation and generalisability. These aspects have been considered in the report of the study.
An ecological or correlational research design has been considered the most suitable research design, as the units of analysis were groups of people rather than individuals. The main advantages of ecological studies are based on the low time and cost required, because they use existing data. On the other hand, confounding is high and control for all the potential confounders should be used in order to exclude the possibility that other characteristics account for any relationships found (Bonita, Beaglehole and Kjellström, 2006:40).
In Mozambique, up until the time of the study, registers used in MWHs are weak, because official MWHs guidelines have only just been launched. Therefore, it is difficult to study individually (using subjects, elements) if health facilities with MWH have a broader institutional deliveries coverage than those centres without MWH. In addition, the aim of the present study is to assess the impact, on health facilities coverage, of the set up of MWH. For this reason, clusters of population have been defined according to the distance from a health facility. This group of population was used to estimate annual deliveries and compare with registered institutional deliveries, determining institutional delivery coverage. This ecological study attempts to study if there is an association between the presence of a MWH (exposure) and better institutional deliveries coverage in a group of health facilities.
The research did not just try to study the impact of MWH on health facilities, but also tries to compare if health facilities with MWH have improved their area of influence over a determined population. Consequently, the research also compared two groups, or clusters of health facilities with similar characteristics: those with MWH and those without the service. As detailed below, inferential statistics have been used in order to determine if there is a significant difference in institutional deliveries coverage between health facilities with and without MWH.
The main aim of correlational studies is to describe the nature of relationships, not to determine cause and effect. This design involves the systematic research of relationships between or among two or more variables (Burns and Grove, 2005:26). Three variables have been studied: (a) Institutional deliveries coverage; (b) distance of the population to the health facility and (c) presence of MWH.
Within this context, a main hypothesis has been formulated:
Hypothesis1: health facilities type II with MWH have better coverage of institutional deliveries than those without MWH.
A null hypothesis has also been formulated:
Null Hypothesis1: there is no significant difference between the coverage of institutional deliveries in Health Facilities type II with MWH and without.
In order to exclude confounders, a second hypothesis has been tested:
Across time analyses shows that MWH services increase institutional deliveries in health facilities.
It was considered that time series may reduce some of the socioeconomic confounding that are a potential problem in this type of design (Bonita et all, 2006:42). This analysis should allow us to undercover institutional deliveries patterns of selected health facilities in order to assess if the set up of MWH services have induced changes in institutional deliveries.
RESEARCH METHOD
Sampling
As Burns and Grove indicate (2005:341) sampling involves selecting a group of people, group, events, behaviours or other elements that will be under study. A sampling plan describes the crucial aspects of making the sample selections, mainly: the elements and population, the sampling criteria, the ethical issues related to sampling and finally the sample. These key aspects will be described later in this document.
Population and sampling
Elements are the individual units of a population. The population (in some cases named target population) is the entire set of elements who meet the sampling criteria, which are a list of characteristics indispensable for membership or in the population. Target population, or just population is selected from the population universum (Burns and Grove, 2005:342).
This study focusses on six districts of Cabo Delgado province, namely Montepuez, Balama, Namuno, Meluco, Macomia and Ancuabe. The selection of these districts was based on two main factors: first, these six districts have the support of several NGOs, including medicusmundi, which have the mandate of strengthening the Health Information System. As a result, we expected a high quality of the institutional deliveries data registers. The second reason is that it was predictable that these districts have a major number of MWHs, due to the support that have received during several years from several NGOs.
In 2009, the health facilities network of the Southern and Central area of Cabo Delgado province (districts of Montepuez, Balama, Namuno, Meluco, Macomia and Ancuabe) was composed by 39 health units or elements, as individual units of the population universum (Burns and Grove, 2005:341), classified as:
The selection of the health facilities under study was one of the key aspects of the sampling process. In 2001 the government of Mozambique launched a classification and characterization of health facilities pertaining to the National Health System (MISAU, 2001). This document establishes, among other aspects, the specific functions, criteria and mechanism for the cataloguing and description of the health facilities pertaining to the three health care levels, namely: primary, secondary and tertiary.
Based on this characterization, the crucial aspect was to select a type of health facility that, initially, did not receive obstetrics reference from any other lower level. This means that, in essence, this health facility serves its direct population. In this sense, health facility of type I attend populations with complications referred from health facilities of type II. The research methodology was based on the determination of institutional delivery coverage based on clusters of populations. In this context it is critical to work with a population that, in the first instance, goes to the most nearby health facility as its first option. This allows us (as will be detailed in chapter 4) to compare the registered institutional deliveries with the expected deliveries, considering a catchment of population that, initially, accesses the proximal health facility.
In this sense, with the aim of assessing the effectiveness of MWHs in increasing institutional deliveries coverage, the following sampling criterion has been defined:
Inclusion criteria:
Functional health facilities managed by the Ministry of Health (providing health care during the 6 months prior to data collection).
Health facilities that offer maternal health care, including Basic Obstetric Care (this excludes health posts).
Health facilities pertaining to primary health network (this excludes tertiary and secondary health facilities as provincial and rural hospitals).
Health facilities that will not receive obstetrics reference from other health facilities.
Exclusion criteria:
Health facilities that started their activities during the last six months prior to data collection.
Health facilities pertaining to secondary or tertiary health network.
Health facilities without maternal health care services.
Health facilities that receive obstetrics reference from lower levels.
Ethical issues related to sampling
The elements of the sample are health facilities pertaining to the national health system. In this sense, before starting the collection data, a request for this information was sent to the Provincial Health Board, in order to obtain informed consent on behalf this body. In addition, authorization was requested on behalf of the Agriculture Provincial Board to use the geo-coded data, including population location and census. Both requests for data collection and usage in order to implement the research were accepted (see annex 1 and 2).
The data used for the research were mainly, geo-referenced data (including population location and census) and institutional deliveries in the health facilities under study. In both cases, the population’s clusters of pregnant women that use a health facility to deliver, were anonymous to the researcher, respecting the Right to Autonomy and Confidentiality (Burns and Grove, 2005:190).
TABLE OF CONTENTS
ACKNOWLEDGEMENTS
ABSTRACT
LIST OF ACRONYMS
LIST OF FIGURES
LIST OF TABLES
CHAPTER 1 – ORIENTATION TO THE STUDY
1.1INTRODUCTION
1.2BACKGROUND INFORMATION ABOUT THE RESEARCH PROBLEM
1.3 RESEARCH PROBLEM
1.4 AIM OF THE STUDY
1.5 SIGNIFICANCE OF THE STUDY
1.6 DEFINITIONS OF TERMS
1.7 FOUNDATIONS OF THE STUDY: THE THREE DELAYS MODEL
1.8 RESERACH DESIGN
1.9 RESEARCH METHODS
1.10SCOPE OF THE STUDY
1.11 STRUCTURE OF THE DISSERTATION
1.12 CONCLUSION
CHAPTER 2 – A LITERATURE REVIEW OF MATERNITY WAITINGS HOMES AND APPLICATIONS OF GEOGRAPHIC INFORMATION SYSTEMS IN PUBLIC HEALTH
2.1 INTRODUCTION
2.2MATERNAL MORTALITY
2.3 THE THREE DELAYS MODEL
2.4 SAFE MOTHERHOOD INTERVENTIONS IN MOZAMBIQUE
2.5 MATERNITY WAITING HOMES
2.6 GEOGRAPHICAL INFORMATION SYSTEMS IN PUBLIC HEALTH
2.7 CONCLUSIONS OF THE LITERATURE REVIEW
CHAPTER 3 – RESEARCH DESIGN AND METHOD
3.1 INTRODUCTION
3.2 RESEARCH DESIGN
3.3 RESEARCH METHOD
3.4 INTERNAL AND EXTERNAL VALIDITY OF THE STUDY
3.5 CONCLUSION
CHAPTER 4 – 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 – CONCLUSIONS AND RECOMMENDATIONS
5.1 INTRODUCTION
5.2 RESEARCH DESIGN AND METHOD
5.3 SUMMARY AND INTERPRETATION OF THE RESEARCH FINDINGS
5.4 CONCLUSIONS
5.5 RESEARCH RECOMMENDATIONS
5.6 RECOMMENDATIONS FOR FUTHER RESEARCH
5.7 CONTRIBUTIONS OF THE STUDY
5.8 LIMITATIONS OF THE STUDY
5.9 CONCLUDING REMARKS
GET THE COMPLETE PROJECT
EFFECTIVENESS ASSESSMENT OF MATERNITY WAITING HOMES IN INCREASING COVERAGE OF INSTITUTIONAL DELIVERIES USING GEOGRAPHICAL INFORMATION SYSTEMS IN SIX DISTRICTS OF CABO DELGADO PROVINCE (MOZAMBIQUE)
