CHAPTER THREE PHASE 1 RESEARCH DESIGN AND METHODOLOGY
This chapter describes the research design and methodology used in phase 1 of the study. The aim of the study was to develop a strategic plan that could facilitate the implementation of an integrated family planning and HIV services at public health centre level in Addis Ababa, Ethiopia.
In order to achieve the aim, the objectives of the study were to:
Describe the challenges and opportunities of clients using integrated family planning and HIV services in public health centres in Addis Ababa, Ethiopia.
Describe the challenges and opportunities of service providers regarding integrated family planning and HIV services in public health centres in Addis Ababa, Ethiopia.
Identify strategies that could facilitate the implementation of an integrated family planning and HIV services at the public health centre level.
Develop a strategic plan that could facilitate the implementation of an integrated family planning and HIV services at the public health centre level in Addis Ababa.
The researcher selected an exploratory sequential mixed-method research design in this study. The researcher used quantitative and qualitative research methods (Hesse-Biber 2010:3). Quantitative data were obtained from clients and service providers (phase 1) and provided the scientific evidence to share with stakeholders, namely programme officers. This data and the literature review formed the basis for identifying and voting on strategies qualitatively (see chapter 5), by means of a nominal group discussion (phase 2), and the development of the strategic plan (see chapter 7, section 7.2.3).
An exploratory sequential mixed-method design with quantitative data collection was used in the first phase followed by qualitative data collection in the second phase. The quantitative data informed the qualitative data collection in phase 2 (see chapters 5 & 6). The researcher used quantitative and qualitative research methods (Hesse-Biber 2010:3). This chapter describes the methodology of phase 1.
OBJECTIVES OF PHASE 1
The objectives of phase 1 were to describe The challenges of clients using integrated family planning and HIV services in public health centres in Addis Ababa, Ethiopia.
The opportunities of clients using integrated family planning and HIV services in public health centres in Addis Ababa, Ethiopia.
The challenges of service providers’ regarding integrated family planning and HIV services in public health centres in Addis Ababa, Ethiopia.
The opportunities of service providers’ regarding integrated family planning and HIV services in public health centres in Addis Ababa, Ethiopia.
A research population refers to the entire set of elements, individuals or objects having some common characteristics in which a researcher is interested (Polit & Beck 2014:274). The population of phase 1 was clients who had access to family planning and HIV services as well as service providers who rendered the care at the public health centre. Addis Ababa has a total of 80 public health centres that serve a population of approximately 3 million (Addis Ababa City Administration Health Bureau 2014:10).
Public health centres: The 80 public health centres in Addis Ababa serve approximately 690, 000 clients in the reproductive age group (18-49 years).
Clients: Approximately 301,316 clients utilised the family planning and HIV services in the public health centres in 2014. Of these, 225,987 used HIV services and 75,329 clients used family planning services (Addis Ababa City Administration Health Bureau 2014:12).
Service providers: A total of 1,200 service providers (680 nurses, 296 midwives, 200 health officers and 24 physicians) were responsible for family planning and HIV services in the 80 public health centres in Addis Ababa (Addis Ababa City Administration Health Bureau 2014:12).
Sampling is a process of selecting cases to represent the entire population so that inferences about the population can be made (Creswell 2014:142-143). The researcher used probability sampling because it allows an equal chance of selection. In probability sampling, the study subjects selected are representative of a certain population that allows equal chance of selection (Creswell 2014:142-143). This sampling method allowed the researcher to randomly select the samples of public health centres using a lottery method, clients by simple random sampling (SRS), and service providers by stratified simple random sampling.
Public health centres: Thirty-one (31) of the 80 public health centres were randomly selected, using a lottery method. Only 31 public health centres were selected due to limited resources available for the study. However, these public health centres were representative of public health centres in Addis Ababa as they were all public health centres that provide similar services to all clients and fall under the same level of function.
Clients: Four hundred and three (403) clients were selected from the 31 public health centres using simple random sampling (SRS). Simple random sampling (SRS) is the process of random selection of samples from a defined population (Loiselle, Profetto-McGrath, Polit & Beck 2011:212). The required sample size was determined by using single population proportion formula (LoBiondo-Wood & Haber 2010:230). The application of 50% proportion is assumed to yield a maximum sample size calculation (Naing, Winn & Rusli 2006:11). Accordingly, the researcher applied P=50%, 95% confidence interval, margin of error = 5%, and non-respondent rate = 5 %.
The formula used to calculate the sample size was
The target populations of clients were 116,760, that is greater than 10,000 populations and was calculated as 384 without adjustment. Considering the non-response rate of 5% (19.2) clients, the actual sample size was 384+19 = 403 clients. The researcher utilized a proportional allocation to each public health centre to draw the 403 service providers depending on the client flow of each centre (see table 3.1). The researcher listed all clients who had appointments for family planning and HIV services between 15 and 26 June, 2015 and 2,687 clients were appointed to get family planning and HIV services from the registers. Thus, every seventh (7th) client was selected and asked to volunteer to participate (see table 3.1).
Service providers: Stratified simple random sampling was used to determine the sample size of each group of service providers. Fifty percent (50%) of each group of service providers (nurses, midwives, public health officers and physicians) were randomly selected (see table 3.1). A total of three hundred and five (305) service providers (173 nurses, 83 midwifery nurses, 44 public health officers and 5 physicians) were selected from the 31 public health centres, using stratified simple random sampling (SRS).
The required sample size was determined by using a single population proportion formula (LoBiondo-Wood & Haber 2010:230). To yield a maximum sample size calculation, the researcher applied P=50%, 95% confidence interval, margin of error = 5%, and non-respondents rate = 5% (Naing et al., 2006:11).
The population of service providers who provide family planning and HIV services in the 80 public health centres was <10,000 population; that is, 1,200. To get the required sample, it was adjusted with a formula of n/1+ (n/N) = 290. Considering the non-response rate of 5% (14.5), the actual sample size was 290+15 = 305 service providers.
The researcher used a 50% proportional allocation to each group of service providers at the public health centres to draw the 305 from among the 610 service providers. Every second service provider from each group was asked to participate in the study.
The distribution of the sampled service providers was
Public health officers=88/610*305=44
Data was collected by means of questionnaires. The researcher developed the questionnaires for data collection based on the literature review. A self-administered questionnaire was developed to collect data from service providers between 1 and 12 June 2015 (see Annexure 2 for questionnaire). A questionnaire was also developed for structured interviews to collect data from clients between 15 and 26 June 2015 (see Annexure 1). The questionnaires were completed by trained data collectors in the interviews. This method was decided on due to the literacy level of some respondents who would not have been able to complete the questionnaire themselves.
The researcher consulted experts in the field to review the content of the draft questionnaires. One was an expert in questionnaire development and the research topic and the other was a statistician. Both experts accepted the content and structure of the questionnaire.
The researcher conducted a pre-test or pilot study to test the questionnaire for clarity of the questions to improve the reliability of the study (Delport & Roestenburg 2011:195). The researcher conducted pre-testing at two public health centres in non-study areas, close to Addis Ababa. Ten (10) randomly selected service providers who were working in family planning and HIV services and 10 clients who came for family planning and/or HIV services were asked to voluntarily participate in the pre-test. They all received the information sheet and consent form (see Annexures 3 and 4) to ensure voluntary participation and informed consent. The researcher revised the questionnaires according to the feedback from the experts, service providers and clients so that the wording, content and context of some questions were clear. The following changes were made to the questionnaires:
The following six questions were not clear to all the clients:
Section A: Q3 What is your occupation? “Civil servant’’ was not on the list of choices/options and was therefore added.
Section A: Q7 How much is the total monthly income of your family? Clients understood this question on monthly income as referring to only the client’s income and not the combined income of all in the household. The local language words were elaborated on to provide clarity in the local context.
Section B: Q27 Do you think that any of the following aspects are challenges to the use of family planning or contraceptive methods? The statement “I did not observe any challenges’’ was added as an option.
Section C: Q3 What was the result of your HIV test? The statement ‘’I do not want to disclose’’ was added.
Section C: Q6 What kind of HIV service(s) have you received from this health facility today? The statement “I did not receive any HIV services today’’ was added.
Section D: Q1 Have you received an integrated family planning and HIV service today? Some clients understood this to mean only integrated family planning and HIV services through a “one-stop shop’’ approach. This was corrected and elaborated to include integrated family planning and HIV service provision using other approaches.
Some clients complained about the length of the questionnaire. Data collectors explained the purpose of the study and that it would only take 20 minutes to complete the questionnaire. Clients then volunteered to participate in the study and signed written consent.
Some service providers did not answer all the questions because they were engaged in service provision, although they were asked to complete it in their free time. The supervisors again encouraged the service providers to complete the questionnaires in their coffee break or lunch break. The questionnaires were also collected in their break time to ensure that the work schedule was not unethically interrupted.
Some service providers did not follow the instructions to skip questions that were not relevant to them. This was discussed with the data collectors during the training and they explained to the service providers that they should read all the instructions and encouraged them to ask questions if they had any uncertainty.
Some service providers expected payment for their participation. However, the researcher emphasized that participation was 100% voluntary and no incentives would be paid. They were informed that they had the choice not to participate, without any penalty and all the service providers agreed to voluntarily participate in the study.
The questionnaires were initially prepared in English and translated into the local language (Amharic). The local language was again translated back to English to make sure that each question had the same meaning in both languages. The final questionnaires for clients and service providers had a consent form attached to each questionnaire (see Annexures 1 and 2).
Each respondent signed the consent form after they read through the information letter and understood the purpose of study as well as that their participation was entirely voluntary. Data collectors read the information sheet and consent form to the illiterate clients to ensure they understood and voluntarily participate. They were informed that they can withdraw any time if they want to do so and the information they provided could be kept anonymous and confidential at all times.
The two final questionnaires (for clients and service providers) had four sections (see Annexures 1 and 2).
Section 1: Biographical information
Section 2: Fertility and family planning
Section 3: HIV/AIDS services
Section 4: Integration of family planning and HIV services
The self-administered questionnaire (for service providers) consisted of 86 questions and the questionnaire used for the structured interview (for clients) consisted of 61 questions. Both questionnaires included statements and dichotomous/multi-chotomous questions. The most response categories for the dichotomous questions were ‘Yes/No’ and ‘Agree/Disagree’. There were also questions where a justification was required, following the dichotomous or multi-chotomous questions and statements to know the reasons why the respondents chose specific answers/options. Both questionnaires were used to provide a quantitative or numeric description of the sample and contained qualitative questions for qualitative enhancement.
VALIDITY AND RELIABILITY
The quality of research is determined by its validity and reliability.
Validity is the degree to which an instrument measures what it is supposed to measure (Polit & Beck 2014:194). Validity refers to the extent to which an empirical measure adequately reflects the real meaning of the phenomenon under consideration (De Vos et al., 2011:172). There are various types of validity such as face, content, criterion-related and construct validity. In this study the questionnaires were only tested for content and face validity (Polit & Beck 2014:198).
Content validity represents the universe of content, or the domain of a given construct. The questionnaire was judged for content validity (LoBiondo-Wood & Haber 2002:314). The researcher concentrated on the content of the questionnaires, which was derived from the literature review, consultation with experts, and the study objectives. In addition, the questionnaires were tested in the pre-test and corrected where necessary (Polit & Beck 2014:198).
Face validity basically verifies that the instrument gives the appearance of measuring the concept. It is an intuitive type of validity in which experts in the field and the supervisors were asked to read the instrument and evaluate the content in terms of whether it appeared to reflect the concept the researcher intended to study (LoBiondo-Wood & Haber 2002:315). Face validity was ensured during the development of the questionnaires and was evaluated by the scientific review committee of the Department of Health Studies, University of South Africa (UNISA). Trained supervisors did the data cleaning and a statistician and a data analyst entered the data for statistical analysis to ensure the validity of the study. Double entry method (EPI info and SPSS software) was also applied to cross-check the data quality. Open-ended questions were open coded by the researcher.
LoBiondo-Wood and Haber (2002:319) refer to reliability as the extent to which the instrument yields the same results on repeated measures. Reliability refers to “the degree of consistency or dependability with which the instrument measures the attribute it is designed to measure. If the instrument is reliable, the results will be the same each time the test is repeated” (Polit & Beck 2014:194).
De Vos et al. (2011:178) describe reliability as the consistency or stability of the measurements. In this study, the researcher ensured that measures were taken so that the data-collection methods were consistent and did not distort the findings in order to meet the requirements for reliability. The researcher ensured that all the questions had the same meaning for all the respondents. Pre-testing was conducted to ensure that the questionnaires were well understood by all respondents to improve the reliability of the study. The problems observed in the pre-test were addressed to increase the reliability of the study. Two public health officials supervised data collection to ensure reliability (see section 3.11 and 3.12 for details). The researcher was convinced that the data collection process was reliable and enhanced the reliability of the results.
DATA COLLECTOR AND SUPERVISOR RECRUITMENT AND TRAINING
Six (6) data collectors who had completed a diploma nursing programme were recruited and trained to collect the data for the first phase of the study. Only data collectors who had previously been involved in structured interviewing or collecting data through self-administered questionnaires for other quantitative studies in the past 3 to 5 years were recruited. The supervisors were recruited based on the recommendation of previous PhD fellows who witnessed their good experience in supervision of data collection.
Two days’ training was provided to the data collectors and supervisors on 27 and 28 May 2015 at a public health centre, in Addis Ababa. The researcher facilitated the two days’ training. The training included a power point presentation, group discussion, and role play. The content of presentation included the objectives of the study, methodology and the questionnaires. The questionnaires were discussed individually with role play on each section.
Before data collection commenced, the researcher obtained ethical approval from the Research Ethics Committee, Department of Health Studies at the University of South Africa (see Annexure 6) as well as from the Addis Ababa City Administration Health Bureau (see Annexure 7). The data collectors and supervisors were recruited and trained prior to data collection.
Service providers: Data was collected from the service providers between 1 and 12 June 2015. The heads of the selected public health centres received an institutional support letter from the Addis Ababa City Administration Health Bureau to inform them about the study prior to data collection. After approval, the researcher and supervisors together with data collectors identified who would be eligible for recruitment to possibly participate in the study. The first service provider was selected randomly and after that every second service provider was selected using a lottery method. The data collectors gave the information letter as well as the voluntary consent form to the selected service providers at the public health centres. The service providers signed the voluntary consent form after they had read and understood the information sheet on the consent form and completed the self-administered questionnaires written in the local language, Amharic. All the service providers read and completed the self-administered questionnaires and returned the completed questionnaires to the data collectors either after the lunch or coffee breaks on the same day. A total of 305 questionnaires were distributed and 305 completed questionnaires were returned, which reflected a 100% response rate.
Clients: Data was collected from the clients between 15 and 26 June, 2015. The researcher identified clients who participated in the study by selecting every 7th client out of 2,687 from the list of appointments (see table 3.1) and informed the data collectors as well as the supervisors to ensure the correct sampling technique applied. The first client was selected randomly using a lottery method. The data collectors read and explained the purpose of the study using the information leaflet as guidance (see Annexures 3 and 4). The participants were selected from the 31 public health centres. After the information sheet and consent form were explained to them, the respondents signed if they agreed to participate in the study. The data collectors interviewed the respondents at the public health centres when they came for family planning and/or HIV services, but after they had received the service. The respondents were interviewed in a private room at the public health centre to maintain confidentiality and privacy. The completed questionnaires were received on a daily basis in the evening of each data collection day by the supervisors and the researcher. The supervisors were two public health professionals who had more than five years’ experience in monitoring and supervising data collection in different studies. The researcher was also responsible to oversee the overall data-collection process. The role of the supervisors was to assist the researcher and visit the data collection process at the public health centre where the researcher would not be able to visit all public health centres on the same day. On average, the supervisors and the researcher visited 3 to 4 public health centres per day. A total of 403 questionnaires were distributed and all were returned.
The researcher obtained permission to conduct the study and -upheld the ethical principles of beneficence, privacy, confidentiality, justice and consent (De Vos et al., 2011:119).
Ethical approval and permission to conduct the study was obtained from the Research Ethics Committee, Department of Health Studies at the University of South Africa (Annexure 6). Institutional consent was obtained from the Addis Ababa City Administration Health Bureau and a copy of the institutional support letter was given to each public health centre and access to the facility was provided (Annexure 7). The information sheets and consent forms (Annexures 3 and 4) provided all the respondents with information on the ethical aspects that applied in the study.
TABLE OF CONTENTS
ORIENTATION TO THE STUDY
1.3 PROBLEM STATEMENT
1.4 THEORETICAL GROUNDING
1.5 AIM AND OBJECTIVES
1.6 RESEARCH DESIGN AND METHODOLOGY
1.7 ETHICAL CONSIDERATIONS
1.8 DEFINITIONS OF KEY TERMS
1.9 OPERATIONAL DEFINITIONS
1.10 CHAPTER LAYOUT
CHAPTER TWO LITERATURE REVIEW
2.2 CHANGE LOGIC MODEL
2.3 FAMILY PLANNING
2.5 INTEGRATION OF HEALTH CARE SERVICES
CHAPTER THREE PHASE 1 RESEARCH DESIGN AND METHODOLOGY
3.2 RESEARCH DESIGN
3.3 OBJECTIVES OF PHASE 1
3.4 RESEARCH PROCESS
3.7 DATA-COLLECTION INSTRUMENT
3.9 VALIDITY AND RELIABILITY
3.10 DATA COLLECTOR AND SUPERVISOR RECRUITMENT AND TRAINING
3.11 DATA COLLECTION
3.12 ETHICAL CONSIDERATIONS
3.13 DATA ANALYSIS
CHAPTER FOUR PHASE 1: DATA ANALYSIS AND INTERPRETATION AND RESULTS
4.2 CLIENT RESPONDENTS (N=403)
4.3 RESPONDENTS’ CHALLENGES (OUTPUTS)
4.4 RESPONDENTS’ OPPORTUNITIES (OUTPUTS)
4.5 SERVICE PROVIDER RESPONDENTS (N=305)
4.6 RESPONDENTS’ CHALLENGES IDENTIFIED (OUTPUTS)
4.7 ADVANTAGES AND DISADVANTAGES OF FAMILY PLANNING AND HIV SERVICE INTEGRATION (N=305)
CHAPTER FIVE PHASE 2 RESEARCH DESIGN AND METHODOLOGY
5.2 UNIT OF ANALYSIS
5.3 EXPLORATORY INTERVIEW
5.4 NOMINAL GROUP TECHNIQUE (NGT)
5.5 MULTIPLE GROUP DATA ANALYSIS
5.7 ETHICAL CONSIDERATIONS
CHAPTER SIX PHASE 2: DATA COLLECTION AND ANALYSIS, AND INTERPRETATION
6.4 MULTIPLE NOMINAL GROUP ANALYSIS
6.5 DATA INTERPRETATION AND LITERATURE CONTROL
CHAPTER SEVEN STRATEGIC PLAN DEVELOPMENT AND VALIDATION (OUTCOME)
7.2 PART I: STRATEGIC PLAN DEVELOPMENT
7.3 PART II: VALIDATION OF DRAFT STRATEGIC PLAN
7.4 FINAL STRATEGIC PLAN
CHAPTER EIGHT CONCLUSION, LIMITATIONS AND RECOMMENDATIONS
LIST OF REFERENCES
GET THE COMPLETE PROJECT
STRATEGIES TO FACILITATE THE INTEGRATION OF FAMILY PLANNING AND HIV SERVICES AT THE PUBLIC HEALTH CENTRE LEVEL IN ADDIS ABABA, ETHIOPIA