Purpose Of The Study
In sub-Saharan African, there is has been a significant rise in disease burden, loss of productivity and the emigration of medical doctors to developed counties have resulted in the in dearth of the most required health resource (Aluwihare, 2005). Most studies on medical brain drain have examined the subject from the perspective of recipient developed countries and little attention has been paid to the donor developing countries. Thus, “scarcity of data from developing country makes it difficult to fully describe the impact of migration on countries of origin” (Hagopian et al., 2004). Most studies on doctors migration from Sub-Saharan African has tended to focus on numbers, without exploring the underlying reason for migration, assessing the potential negative impact of migration on the health care systems, or considering means to alleviate the problem. In actual fact, information regarding the extent of migration is usually obtained through data from countries of destination (Stilwell et al 2003).
The migration of medical doctors from Nigeria and other countries in sub-Saharan African generates three areas of major concern. The first is a loss of the basic health services that is available to the citizen. for instance, Ghana, faced with a ratio of nine doctors to every 100 000 patients and no more than 22 paediatrician are licensed to practise in country and no more than 10 specialists of any kind practising in the remote area.
The second effect doctor migration from Nigeria is that it prevent the health sectors the ability to organize and expand. Public health institutions heavily depend on doctors to lead, develop and promote them as they work to advance health care delivery. As obtained in US, doctors are well positioned to serve their organisation by actively involved in managing resources and articulating priorities. It is therefore speculated that as the numbers of available medical doctors in developing countries reduces.
The third challenges are that doctor's migration depletes a significant element of the middle class in developing countries. As in the developed countries, medical doctors in African comprise of an important segment of the social and economic make up of the middle class. They are generally accorded a lot of respect in the society, as being above corruption, they advocate for improvement in quality of education of public schools and they play a vital role in political (Hagopian et al., 2006). In Nigeria, over 70.2% of the population lives on $1 per day (WHO, 2006)
Research Aim And Objectives
The aim of this study is to analyse the causes of medical brain drain from developing countries, benefits and burdens associated with brain drain based largely on the views of Nigerian doctors practising in UK.
The objectives are:
To identify the factors that influence the decision of medical doctors to migrate to developed countries migration
To identify the impacts of migration on healthcare in developing countries
The most suitable research design was selected to meet the aims and objectives of the study. The purpose of the study was to conduct a series of interviews to identify the factors that influence the decision the migration of medical doctors to developed countries and the impacts of migration on the health care system in developing country. The focus is analysing the causes and impacts of medical brain drain largely based on the view of Nigerian doctors practising in the United Kingdom.
Traditionally, science is uniquely quantitative (Streubert and Carpenter, 1999), however the general idea of research is about positioning the researcher in the experiential world and a more practical level it involves relating research question to available data (Punch, 2000). Basically research design is divided into major categories: quantitative and qualitative research method.
Quantitative research method entails conceptualization of problems in term of experimental comparison, which is based on some intervention or treatment and usually with a clear outcome or variables in mind. It involves the collection and analysis of numerical data and depends on numbers for its conclusion and statistical techniques are often employs for its analysis (Cormack, 1995).
Qualitative research methods are tools used in understanding and describing the world of human experience. The concept and theoretical elaboration of qualitative research emerge was derived out of words in the collection and analysis of data. Qualitative research design has been selected as the most appropriate method for this study. The reason for adopting this method is due to it creditable nature of analysis and it provides systemic for gaining insights into other persons' views of the social world, ‘putting oneself into someone else shoes'(Pogar and Thomas, 2008).
“Qualitative” as described by Pogar and Thomas, (2008) refers to the nature of data or evidence collected. Qualitative data entail detailed description based on language or recorded pictures by investigator. “Qualitative research method is a research strategy that often emphasise word rather than quantification in the collection and analysis of data. As a research strategy it is inductive, constructionist, and interpretivist” (Bryman, 2008), whereas quantitative researcher does not always subscribe to all the three features. In addition, It can be described as a method of naturalistic enquiry which is often less obstructive than quantitative investigations and does not manipulate research settings. The research method in qualitative research aims at studying people in their natural social setting and to collect a natural occurring data. Demonstrable advantages of qualitative research over quantitative method have been shown in situations in which there is a little pre-existing knowledge (Bowling, 2002).
Furthermore, Muir Gray (2001) emphasise the usefulness of qualitative research in gaining understanding of health and health services, it also complement and support a quantitative research. The research method is often used in generating hypothesis to solve a problem which be subjected to a test using either a quantitative methods or combination of qualitative and quantitative method (Muir Gray, 2001). Nonetheless, there may be some challenge delineating the nature of qualitative research usually when the link between the theory and research is ambiguous. In qualitative research, theory is usually an outcome of investigation rather than having theory that precedes it (Bryman 2008).
One of the most important decisions required for any research activities is to determine the type of participants essential for the study. As it is usually impossible to whole population, thus researcher select a sample of research subject that represent the whole research population (Polit and Hungler, 1995). Polgar and Thomas (2008), defined sample as a subset of a population while sampling involves the selection of a sample from a population. Sampling method according to Brink (1996) is a process of selecting the sample form a population in order to obtain information regarding a phenomenon in a way that represents the population of interest”. Essentially, if a sample is representative of the study population, A generalise valid sample result to the population can be obtained without going through the expense of studying the whole population.
Hesse-Biber (2005), argued that the logic of qualitative sampling is concerned with in-depth understanding, working with small sample which is usually a function of the purpose of the study; in light of its sampling frames and of practical constraints. The overall goal of qualitative sampling is to examine the “process” or “meanings” of individual attribute to their given social situation, not necessarily to make generalisations.
Although, sampling in quantitative research may be probabilistic (if representativeness is important) or purposive (if the focus of the research is to study the relationship between two variables) while qualitative research is often a non probability sampling which is generally restricted to the following method – convenience sampling, purposive sampling, snowball sampling and theoretical sampling (Punch, 2000). The aim of all qualitative method is method is to understand complex phenomena and generate hypothesis, rather than apply the findings to a wider population (Bowling, 2002)
Samples in qualitative studies are usually small, non random samples; this is due to the use of different consideration in the selection of participants. A critical step in qualitative sampling is the choice of settings with high potential for information richness (Polit and Beck 2000). For instance simple or very detailed studies, single figures may be required; for complex questions large samples and a variety of sampling techniques may be considered. However, in practise, the number of participant required becomes apparent as the study progress Marshall (2006).
Another important point is that in qualitative studies, individuals are not always considered the unit of analysis. Glaser and Strauss (1967) cited in Polit and Beck, (2000) noted that incidents or experiences are often the basis for analysis. A participant with rich information can therefore contribute dozens of incidents and likewise a small number of participants can generate large samples for analyses (Polit and Beck 2000).
A total of numbers of eight medical doctors that are of Nigeria origin practising in the UK were recruited for telephone interviews to determine their views about migration of doctors from developing countries to developed countries. Doctors of Nigerian origin were selected by the researcher because of easy accessibility and also the time constraint to undertake this research. Moreover, selection of this particular group of medical professional is special interest to the researcher due of the trend of medical brain drain in Nigeria.
The inclusion criteria for selecting interviewees were that the participant must be a medical doctor of Nigeria origin, received his/her medical training in Nigeria and practising medicine in the United Kingdom. Nigerian doctors who are engaged in other professional activities apart in UK were excluded from the study.
A purposive sampling method was used in this study because the research is limited to Nigerian doctors practising in United Kingdom. According to Bowling (2002), purposive sampling method aim at sampling a group of people or settings, with a particular characteristics. Also it consideration often apply to sampling of cases in which research will be conducted and then to people within those cases. This sampling method is essentially strategic and an entails establishing a good correspondence between the research questions and sampling. It is otherwise known as judgement sampling where the respondents are selected because their knowledge and experience is valuable to the research process (Bryman, 2008). The sampling method can be used in detecting cases within an extreme condition as for certain characteristics or cases within a wide range situation in order to maximize variation (Hesse-Biber, 2005).
As stated in Bryman (2008), “most writers on sampling in qualitative research based interviews recommend that purposive sampling is conducted”. This form of sampling is essentially strategic and entails an attempt to establish a cordial relationship between the research questions and sampling, In other words, the researcher samples on the basis of wanting to interview who are relevant to the research questions. Purposive sampling provides information-rich cases for in-depth study and analysis (Paton 1990).
However a common disadvantage of this sampling method was that it require an in-depth interviewing, time-consuming data analysis and a rich insight is required in order to understand the social phenomena unlike statistical information (Bowling, 2002).
In addition to purposive sampling, a snowball sampling method was employed to recruit other participants for the interview based on recommendation from the initials participants. Snowball sampling in not sample technique that is employ in quantitative research but within qualitative research (Bryman, 2008). The advantage of snowball sampling is that sponsorship encourages cooperation and therefore further facilitates access to other participants, it is more cost effective and practical and in addition it gives the researcher opportunities to specifies that the characteristics that is require of the new participants although the limitation of this approach is that sample may be unrepresentative because sample may be restricted to rather small group or network of contacts (Sapssford and Jupp, 1996)
In qualitative studies, data collection is more flexible than in quantitative studies and decisions about what information to collect usually evolve in the field. Qualitative researcher are often prepared for problematic situations that may arise in the process of data collection however the best prearrange plan for data collection sometimes fall through. Essentially, the primary method of collecting qualitative data is majorly through self report, by interviewing other study participants (Polit and Beck , 2009)
Interviewing remains the principal mode of data collection in qualitative research (May 1991). It is a data collection method in which one person (the interviewer) asks question from people (a responder), interviews are either conducted by telephone or face-to-face (Polit and Hungler, 1991) besides, the most widely used interview format are either by semi-structured or open interview methods. In the context of quantitative research study, they are used majorly in preparation of a standardized data collection and the development of data collection tools whereas qualitative interviews play a significant role imparting of expert knowledge about the research field in question, the recording and analysis of the informants' subjective perspective, or collection of data relating to their biography. In actual fact, as a means of acquiring information, interviewing is an element so significant to qualitative methodology (May 1991). Research areas in which qualitative interviews are frequently used include studies of gender-related questions, studies of social and political orientations of different population groups, or studies of access to professions and of professional socialization (Ficks et al, 2004).
According to Cormack, (2000), three types data are characteristically generated from interviews, these include: peoples experience and account of events; their attitude, perceptions and opinions about phenomena; and biographic and demographic details. This study aim to reflect the doctors' views of Nigerian doctors in terms of reasons, experience, attitude, perception and opinion on “Medical Brian Drain”
Initial access to Nigerian doctors practising in the UK was through an email to the secretary of Medical Association of Nigerian Specialties and General Practitioners (MANSAG) in the British Isles to seek for volunteers that would be willing to partake in the telephone interviews. Initial three participants were recruited with the assistance of MANSAG secretary while other participants who participated in the studies were recruited using a snowballing method based recommendation from the initials three participants. These resulted in the selection of eight eligible Nigerian doctors who were willing to participate in the telephone interview. Those fitting the criteria were doctors who trained in Nigeria and are currently practising the medical profession in the UK.
Individual consent was obtained through an email indicating their interest to participate in the telephone interview. Interviews were conducted at scheduled time and of the day which were convenient for the participants. Prior to the commencement of the telephone interview a reaffirmation of the consent were obtained from all the participants.
Telephone interview was the data collection felt more appropriate by the researcher for obtaining detailed information because participants works and reside in different location across the UK. Participants felt more comfortable with that approach because of their busy work schedules. The purpose of the study was made clear to the participants and was encourage through reassurance to be plain in their response. In addition, the researcher tried to be objective both during the interview and during data analysis
A semi-structured interview schedules with mainly fixed question with no response codes was designed for the interview process and was used flexibly to allow the researcher to probe and also to enable the respondents to raise other relevant issues lacking in the interview schedule. Other importance of semi-structured interview schedules is that it permit the researcher to ask the questions out of order at appropriate opportunities during the interview (Bowling, 2002). The semi-structured telephone interviews were conducted over an average of 20 minutes for each of the participants. The interview process was conversational and interactive in nature.
The interview was audio recorded and stored in a digital format for easy transcribing. It allowed the researcher to attend to the participants rather than manually recording of all responses. Furthermore, recording and transcribing interviews data has been shown to have the following advantages: it allows a more detailed examination of participants' responses, protects against accusations that analysis might have been influenced by researcher's value or biases. The recording and transcribing of interview data also opens up the data to public scrutiny by other researcher who can carry out a secondary analysis of the obtained data. However, the procedure can be time-consuming, and require good equipment for the recording and transcribing of data (Bryman 2008).
Although telephone interview is an established and acceptable approach for quantitative data collection; it is a method principally used for survey analysis (Aday, 1996) and is most widely used survey modality in industrialized nations (Bernard, 2002). According to Bowling (2002), interview conducted by telephone appear to have equal accuracy rates to face-to-face interviews in relation to the collection of data on a study on general health status and the prevalence of depressive symptoms. Other major advantages of telephone interviews is that, in theory is economic and relative to time and available resources (Bowling, 2002); there is decreased cost and travel, particularly when compared to face-to-face interviews, respondent are located in their normal environment, it has the ability to reach wide geographically dispersed respondents and greater interviewer safety (Bernard, 2002). Thus, in this study, telephone interviewing made it possible for the interviewer to stay more focused on the interviewee's responses. Furthermore, it gave the participants more anonymity and this seemed to reduce their anxiety about participating. According to Novic, (2008), “respondents in qualitative interview have been described to be relaxed on the telephone, and willing to talk freely and to disclose intimate information” and also qualitative telephone data have been judged to contain rich, clear, detailed, and of good quality information (Sturges and Hanrahan, 2004) .
Nonetheless, telephone interview is less used compared to face-to-face interviews in qualitative research (Opdenakker, 2006; Sweet, 2002), it may perhaps be an adaptable data collection tool (Carr and Worth, 2001). Notable disadvantages of telephone interviews include lack of visual cues, and the potential for possible distraction of participants by activities in their environments (Opdenakker, 2006), although such distractions may also occur during a face to face interview. Also, the duration for telephone interviews may be short compared to face-to-face interviews which may affect the in-depth of the interview. However this little evidence to support the assertion (Sturges and Hanrahan; Sweet, 2002).
The purpose of qualitative data analysis is to provide structure, organise, and elicit meaning from research data. Qualitative data analysis is labour intensive activities that require creativity, conceptual sensitivity and rigorous work (Polit and Beck, 2008). Thorme (2000), described qualitative data analysis as the most complex and mysterious of all of the phases of a qualitative project.
Although qualitative study usually relies on inductive reasoning process to interpret and structure the meaning that are derived from data. Differentiating inductive from deductive inquiry processes is an important step in identifying what is important in qualitative research. Essentially, inductive reasoning makes use of data to generate ideas (generating hypothesis), while in deductive reasoning begins with idea and makes use data to compare the idea (testing hypothesis). However, many quantitative studies uses of inductive reasoning, while good qualitative studies usually requires various types of strategies for data analysis.
Furthermore, qualitative research often takes the position that an interpretive understanding is only possible by means of revealing or deconstructing the meanings of a phenomenon. Thus, a difference between explaining how something operates (explanation) and why it operates in the manner that it does (interpretation) may be a more effective way to distinguish quantitative from qualitative analytic processes involved in any particular study (Thorme, 2000).
According to Morse and Field (1995, pg 126) cited in Polit and Beck (2008), qualitative data analysis is a “process of fitting data together, of making the invisible obvious, of liking and attributing consequence to antecedents. It is a process of conjecture and verification, of correction and modification, of suggestion and defence”.
Analysis of data in qualitative study frequently involves analysis of verbatim transcripts of dialogues and narratives (Polgar and Thomas, 2008). It normally begins with a search for broad categories or themes. Themes emerge from data which often develop within categories of data but may also cut across them. The search for themes involves not only for discoveries of common meaning across the participants but also seeking natural variation (Polit and Beck, 2008). A common point of variation for the analysis of transcribed data is developing a coding system. The coding system is used to categorise data into specific classes (Polgar and Thomas, 2008).
In analysing data, codes that identifies with common themes were used as they emerge from the transcribed data. However most researchers support different approaches to coding but it typically involves in depth studying of the transcribed data and development of close familiarity with the data. During the analytical process, all the ideas concepts and themes are noted to form major categories. Having developed codes for the transcribed data, the reporting of this process often involves detailed description of the categories and their context, with liberal use of examples from the data (Polgar and Thomas, 2008).
However, in order to identify prominent themes and the patterns among the theme, a qualitative content analysis method was adopted. It involves breaking down data into smaller units, coding and naming the units according to the content, and grouping coded material on shared concepts (Polit and Beck, 2008). According to Hsieh and Shannon (2005), “qualitative content analysis is defined as a research method for the subjective interpretation of the content of text data through the systematic classification process of coding and identifying themes or patterns”. Researchers using qualitative content analysis emphasis focus on the characteristics of language as communication with concentration to the content or contextual meaning of the text (Tesch, 1990, Flick et.al., 2005)
Semi-structured open ended questions were designed in advance for the data collection. The questions were design out of findings from theoretical and literature review interview to ensure specific research theme are covered.
Telephone interviews were conducted to determine the view of Nigerian doctors practising in the UK. The interviews were audio recorded to ensure that the interviews data are the actual verbatim responses of the participants and the audio recording was subsequently transcribed over a period of one week. The advantage of transcribing interviews is that aid interpretation by early identification of key themes and increases the familiarity of the researcher with the data (David, 2005, Bryman, 2008).
Moreover, it is strongly suggested that qualitative interviews be recorded and simultaneously transcribed, rather than depending on interview notes. Notes taking may be incomplete and biased which could be caused by a gap in the researcher's memory recalled time or personal views (Polit and Beck, 2008).
The transcribed data were read and reread to identify and index themes and also to categories the data: these centred on particular incidents, phrases, or types of activities regarding doctors' migration.
All relevant data related to each category were identified and, each data was checked and compared with the others to establish different analytical categories. This however involves a coherent and systematic approach (Pope et. al., 2000). Categorisation of the data by theme was done manually, which has the advantage of maintaining close relationship and understanding of the original data by the researcher however this could be time-consuming for large databases (Bowling, 2002).
Validity, Reliability And Rigor
Without rigor, research is worthless; it becomes fiction, and loses its effectiveness. Therefore, a great deal of consideration is given to reliability and validity in all research methods (Morse J.M. et. al., 2002). Evaluation of research studies is an essential criterion in the application of research findings. Traditionally, such evaluation has focus on assessment of reliability and validity. However, their use in qualitative work has been questioned (Long, T. and Johnson, M. 2000). The term ‘Reliability and Validity' are concept often used in testing or evaluating quality of quantitative research, nonetheless, the idea is often used in virtually all kind of research (Golafshani, 2003, Bryman, 2008).
According to Patton (2001) validity and reliability are two important factors which any qualitative researcher should put into consideration while designing a study, analysing data and judging the quality of the study. Moreover, Mason (1996:12) cited in Bryman, (2008) argued that reliability, validity and generalisability “are different kinds of measure of the quality, rigor and wider potential of research, which are achieved according to certain methodological and disciplinary conventions and principles”.
However, reliability and validity have been substituted with a parallel concept of “trustworthiness” which is made up of four different criteria: credibility, transferability, dependability, and confirmability. These contain specific methodological strategies for demonstrating rigor in qualitative studies, is often used to evaluate coding, categorizing, or confirming data results with participants, negative case analysis, peer debriefing, structural corroboration, and referential material adequacy(Lincoln & Guba, 1985). In addition to the four trustworthiness criteria, Lincoln and Guba suggest criteria of authenticity.
Thus this study can be evaluated using the suggested four criteria of trustworthiness of a qualitative inquiry as described by Lincoln and Guba in Polit and Beck (2008):
Credibility: refers to the confidence in accuracy of the data and its interpretation. The researcher ensured the research is carried out in accordance to the cannons of good practise and findings which can be subjected to further validation by external reader or investigator.
Transferability: refers essentially to the generalisation of the data, the extent to which finding can be transferred or apply in other settings or groups. The findings of this study were oriented to the appropriate uniqueness and significance of ‘Medical Brain Drain' to the aspect of the social world being studied.
Dependability: refers to the stability (reliability in quantitative research) of data over time and conditions. This entails ensuring proper records of the research process and procedures were properly kept in an accessible manner. These include details of problem formulations, selection criteria of research participants' interview transcripts and data analysis results.
Confirmability: refers to objectivity which is essential for congruence between two or more people about the data's accuracy, relevance or meaning. While recognising that complete objectivity is impossible in social research, the researcher ensures his personal values or theoretical inclination has not influenced the conduct of the research and its outcome.
Authenticity: refers to the extent of fairness and faithfulness shown of a range of different realities by the researcher. These were ensured by asking all the participants the same type of semi-structured open ended questions, all the participants belong to the same social setting and were also subjected to the same medium and method of data collections process.
The basic principle governing the ethical issues in social research is that the people should not be at risk as a result of participating in a research and an informed consent should be obtained before being allowed to participate (Richards H. M., and Schwartz L. J., 2002).
Approval for this study was obtained from Swansea University School of Health Science Ethics Committee before the commencement of the study; this was to ensure research ethics was not compromised. Initial contact to the participants was through the secretary of Medical Association of Nigerian Specialist and General Practitioner in British Isles (MANSAG). The participants were contacted and briefed about the study through an email. An information sheet alongside a written consent was sent to the participants via email 2 weeks prior to the interview. This was to allow the participants to make a voluntary decision regarding their participation. Participants gave consent by replying the email stating “I agree to take part in the study” and they also gave their preferred day and time for the telephone interview. This was to ensure there was no interfering with their work or other personal day schedule. Consent was later reaffirmed verbally at the beginning of the interview.
All the participants were assured that the data collected would be kept completely anonymous and codes will be used for each participant instead of names. The audio recording and the data transcript from the telephone interview were securely stored in a safe place. In addition participants were informed and assured prior to the interview that they retain the right to withdraw from the study at any stage of the process without any reason.