CHAPTER 4 ANALYSIS AND DESCRIPTION OF RESEARCH FINDINGS (PHASE I)
The previous chapter discussed the research methodology of this study. This chapter deals with data analysis and presents the findings of phase 1 of the study. The chapter presents data that were collected by conducting structured interviews with 400 pregnant women at 16 health facilities in the Buikwe district (as discussed in section 3.4.2. of this thesis).
The data were analysed with the assistance of a statistician using the SPSS (version program and Excel 2010 to produce charts and graphs (see Annexure F – the statistician’s letter).
The purpose of the study was to identify factors influencing pregnant women’s utilisation of anti-malaria services in the Buikwe district of Uganda. The specific objectives of phase 1 of the study were to identify:
- social-demographic characteristics like age, parity, education level, employment status and income levels that might influence pregnant women’s utilisation of anti-malaria services
- environmental factors like geographic access to health facilities, availability of quality health services, traditional healers’ and peers’ potential influence on pregnant women’s utilisation of anti-malaria services
- personal cognitive characteristics of pregnant women like knowledge about malaria during pregnancy and sources of malaria-related information that might influence their utilisation of anti-malaria services
- behaviour/health practices of pregnant women like IPT utilisation, utilisation of LLINs, use of anti-malaria drugs and other malaria control measures related to the utilisation of anti-malaria services
In this chapter, findings based on the analysis of the data obtained during the structured interviews conducted with 400 pregnant women, will be discussed according to the sections of the interview schedule (see Annexure B1). Where applicable, bivariate correlations were calculated for variables using Pearson’s correlation coefficients and the corresponding probability values (p-values). The closer the correlation coefficient value is to +1 or -1, the stronger the positive or negative association respectively between variables. According to Burns and Grove (2005:137), 0.1-0.3 means a weak relationship, greater than 0.3-0.5 means a moderate relationship and greater than 0.5 implies a strong relationship. Where applicable, Pearson’s correlation coefficient was calculated for all variables with IPT use by pregnant women. Furthermore, Pearson’s Chi-square test (X2) or Fisher’s exact test was performed to statistically study the association of categorical variables with IPT use by pregnant women. The level of significance was set at .05.
Findings related to objective 1 of phase 1 of the study: social-demographic variables that might influence pregnant women’s utilisation of anti-malaria services
The first objective of the study was to identify social-demographic characteristics like age, parity, education level, employment status and income levels that might influence pregnant women’s utilisation of anti-malaria services. This study was based on the concept of reciprocal determination of Bandura’s social cognitive theory (please see section 1.9.2 of this thesis). In relation to that theoretical framework, objective 1’s findings are related to personal factors that influence health seeking behaviours of pregnant women, as reported in response to questions in section A of the data collection tool.
The respondents’ ages ranged from 21 to 45 years and were categorised into five age intervals. Out of 400 respondents, 56.0% (f=224) were in the age group of 21 to 25 years, 35.8% (f=143) were 26 to 30 years old and 8.3% (f=33) were older than 30 years as shown in figure 4.1. The mean age of the interviewed pregnant women was 25.5 years, the median age was 24.0 years and the mode was 22.0 years (see table 4.1). The standard deviation was 3.93 and the range was 22.0 years. In a similar study conducted in Uganda involving 1 069 pregnant women, their median age was 24 years (De Beaudrap et al 2013:[9-10]), corresponding to the median age group of the current study’s respondents.
Pearson’s correlation coefficient was -0.019 and the p-value was 0.480. This meant that there was no correlation between the age of pregnant women and their utilisation of IPT. Pearson’s Chi-square test (X2) was 0.955 meaning that there was no association between the age of pregnant women and their IPT utilisation.
Out of 400 respondents, 72.3% (f=289) were married and 11.3% (f =45) were cohabiting with their partners implying that 83.5% (f=334) of all respondents were in some kind of relationship. However, 0.8% (f=3) were widowed, 3.3% (f=13) were separated, 1.8% (f=7) were divorced and 10.8% (f =43) had never been married (see figure 4.2). This implies that, at the time of data collection, 16.5% (f=66) respondents were not in long term relationships with their spouses. A similar study, conducted in Nigeria, reported that the percentage of married pregnant women was 82.0% (n=150) (Diala et al 2013:). The finding is comparable to the current study’s finding that most pregnant women were married. Pearson’s correlation coefficient was – 0.003 and the p-value was p>0.05. Therefore, there was no correlation between marital status of respondents and their IPT utilisation.
Respondents’ levels of education
Figure 4.3 indicates that the respondents’ education levels ranged from no education to university level education. Out of 400 respondents, 48.8% (f=195) had primary school education and 5.5% (f=22) had never attended school. This means that 54.3% (f=217) of all respondents had no or primary level education. Furthermore, 37.8% (f=151) had secondary while 8.0% (f=32) had post-secondary level education.
Another Ugandan study reported that 63.4% (n=453) of the respondents had no or only primary level education (De Beaudrap et al 2013:). This implies that pregnant women in Uganda with low levels of education (either primary or no schooling) might find it difficult to understand the rationale for using IPT during their pregnancies. Diala et al (2013:) reported that women might find it difficult to use medicines (such as IPT drugs) when they are not ill. Women with limited education might find it even more difficult to comprehend the benefits of IPT than educated women. In the current study, there was a positive correlation between education level and IPT use (Fisher’s exact test =0.32). This implies that, educated pregnant women with secondary school or post-secondary school education were more likely to utilise IPT drugs than the less educated ones (with no schooling or only primary level schooling).
Respondents’ employment status
Out of 400 respondents, 53.8% (f=215) were housewives with no gainful source of income. However, 27.5% (f=110) of the respondents were employed in the government sector, by non-governmental organisations (NGOs) or self-employed. Furthermore, 3.5% (f=14) of the respondents were students and 15.3% (f=61) were unemployed. There was no association between employment status and IPT use (Fisher’s exact test >0.05). This implies that employment status of pregnant women was not a significant factor influencing their IPT utilisation.
The MOH (2010a:2-4) also confirmed that there was a high unemployment rate (of 9.8% during 2009) in Uganda, similar to the current study’s findings. High unemployment rates might imply that women depended on their husbands/partners for their own and for their children’s survival, and also for money to pay for transport to reach the health care facilities.
Respondents’ average monthly incomes
Out of the 400 respondents, 80.8% (f=323) had average monthly incomes of less than Shs 200 000 (R1 = Shs 300; USD1 = Shs 3 450 during July 2014). Only 13.8% (f=55) of the respondents had monthly incomes of Shs 400 000 or more including 9.8% (f=39) of the respondents who had an average monthly income exceeding Shs 600 000. This is in agreement with the fact that most Ugandans are poor with a daily income of less than 1USD (R11.50) per day (MOH 2010a:2-3). In this study, there was no association between respondents’ monthly incomes and IPT utilisation (Pearson 2-sided Chi-square=0.153). This implies that pregnant women’s IPT utilisation was not related to their monthly incomes.
Pearson’s Chi-square test (17.978; p = 0.264) indicated that no significant correlation existed between the income groups and the number of fever episodes from which the pregnant women had reportedly suffered during their current pregnancies.
Respondents’ pregnancy trimesters
During the data collection period (April 2014), out of the 400 pregnant women, 8.3% (f= 33) were in the first, 48.8% (f=195) in the second and 43.0% (f=172) in the third trimesters of their pregnancies (see table 4.3). This means that fewer pregnant women in the Buikwe district attended ANC clinics during the first trimester than during the second and third trimesters during the data collection phase of the current study. Pearson’s correlation coefficient between the respondents’ pregnancy trimesters and their taking of IPT drugs was -0.01 (p>0.05). Thus there was no correlation between the pregnant women’s trimesters of pregnancy and their utilisation of IPT drugs.
CHAPTER 1 ORIENTATION TO THE STUDY
1.4MALARIA AS A PUBLIC HEALTH PROBLEM
1.6PURPOSE AND OBJECTIVES OF THE STUDY
1.7SIGNIFICANCE OF THE STUDY
1.8DEFINITIONS OF KEY CONCEPTS
1.9FOUNDATIONS OF THE STUDY
1.10RESEARCH DESIGN AND METHODOLOGY
1.11SCOPE AND LIMITATIONS OF THE STUDY
1.12STRUCTURE OF THE THESIS
CHAPTER TWO LITERATURE REVIEW
2.2THE EPIDEMIOLOGY OF MALARIA
2.3THE BIOLOGY OF MALARIA PARASITES
2.4CONTEXTUALISATION OF INFORMATION OBTAINED FROM THE LITERAUTRE REVIEW WITHIN THE SOCIAL LEARNING
CHAPTER 3 RESEARCH METHODOLOGY
3.2AIM OF THE STUDY
3.5VALIDITY OF THE STUDY
CHAPTER 4 ANALYSIS AND DESCRIPTION OF RESEARCH FINDINGS (PHASE I)
4.4SUMMARY OF KEY FINDINGS
4.5Objective 4 of phase 1 of the study: the influence of pregnant women’s behaviour/health practices on the utilisation of anti-malaria services
CHAPTER 5 ANALYSIS AND DESCRIPTION OF RESEARCH FINDINGS (PHASE II)
5.2RESEARCH OBJECTIVES OF PHASE 2 OF THE STUDY
5.4SUMMARY OF KEY FINDINGS OF PHASE 2 OF THE STUDY
CHAPTER 6 CONCLUSIONS, LIMITATIONS AND RECOMMENDATIONS
6.2RESEARCH DESIGN AND METHOD
6.3A COMPARISON OF RESEARCH FINDINGS OF PHASES 1 AND 2 OF THE STUDY
6.4 CONTRIBUTIONS OF THE STUDY
6.5 IMPLICATIONS OF STUDY FINDINGS FOR THE ROLL BACK MALARIA (RBM) PROGRAMME AND MALARIA CONTROL PROGRAMME (MCP) OF THE MINISTRY OF HEALTH (MOH)
6.6LIMITATIONS OF THE STUDY
6.7SUGGESTIONS FOR FUTURE RESEARCH
GET THE COMPLETE PROJECT
Factors that influence pregnant women’s utilisation of anti-malaria services in the Buikwe district of Uganda