This paper reviews the types of research methods used by researchers to understand and analyze bipolar disorder. Bipolar disorder is a behavioral disorder which has gained popularity in recent years due to the increasing number of cases manifesting not just in adults but also in adolescents and young adults. The goal of the present research is to identify methods of research previously used to study the effects, factors affecting, and relationships of such factors manifesting in bipolar disorder. Quantitative and qualitative research designs will be assessed through a review of previous research conducted. Such a review will reflect not only the value of the methods but also the range of viable methods available to a researcher studying bipolar disorder.
Research Methodologies on Bipolar Disorder
Bipolar disorder is characterized by a manifestation of alternating manic, hypomanic, depressed and mixed episodes (American Psychiatric Association, 2000). The DSM-IV actually describes bipolar disorder into four categories based on the specific manifestations observable in a patient. It is important to understand the causes, manifestations, and methods of treatment in cases of bipolar disorder because of the prevalence of this disorder. Bipolar disorder has been shown to affect roughly 5.7 million American adults or 2.6 percent of the population over the age of 18 years (U.S. Census Bureau, 2005).
In order to better understand this disorder, the effect that the same has on the lives of patients needs to be understood. Moreover, factors which may improve or worsen the situation of diagnosed patients also need to be identified and assessed. By having a clear understanding of these aspects of the lives of those diagnosed with bipolar disorder strategies of coping as well as treatment strategies may be created and implemented.
The first step in such a process is by understanding the research methods which are applicable and which have been shown to be applicable in the situation of those diagnosed with bipolar disorder. Thus, in the succeeding sections several previous studies will be discussed in detail. Particular focus will be placed on the research methodologies applied by the researchers in deconstructing bipolar disorder.
The discussion will first flow from an analysis of quantitative research methods which have been applied to research on bipolar disorder then it will move to applicable qualitative methods. For each study there will be a discussion as to the goals of the study and the methods applied by the researchers in attaining such goals. The efficacy and appropriateness of the method applied will be discussed as well as the method’s potential for application in future studies. The discussion will give a firm understanding of the range of methodologies which may be applied to research involving bipolar disorder.
Quantitative research designs involve systematic investigations of quantitative properties and behaviors and their relationships (Creswell, 2003). Quantitative designs involve the gathering and analysis of numerical data – as opposed to the non-numerical nature of the data gathered for qualitative designs (Creswell, 2003). Since data is numerical in nature, quantitative research needs to be more purposive in its identification of variables to be studied. The research goals are therefore more specific and the purpose more limited than in qualitative research. However, because of the greater structure in quantitative research, the findings are more externally valid than those in qualitative designs.
Research on bipolar disorder has traditionally been quantitative due to the technical nature of most studies conducted on the topic. Most research conducted has focused on the biological aspect of the disorder and has therefore required strong reliance on quantitative analysis and numerical data. The shift of research focus away from the biological roots of the disorder and towards a social and behavioral understanding has caused the application of more quantitative research methods and, more importantly, the introduction of qualitative methods as well.
One of the most common quantitative methods is the use of a scale. In the study conducted by Goossens, Knoppert-Van Der Klein, Kroon, and Achterberg (2007) a scale was used in the form of two questionnaires. One questionnaire was structured by the researchers themselves in order to gain information about the demographics and clinical data of the participants. The second questionnaire was a pre-established scale constructed to measure the need for care of respondents. The first questionnaire provided information which would be tallied by the researchers in order to come up with a trend regarding the composition of their participant pool. The data obtained from the same were thus coded into numerical output. The second questionnaire came with a procedure for calculation of the scores of respondents based on the different parts of the questionnaire measuring for particular aspects of need for care.
A scaled instrument is one of the clearest ways to conduct a quantitative study. When the scale is one previously recognized and established then the reliability of the results are easily verifiable because of the prior use of the instrument. In the study of Goossens et al. (2007) the use of the two questionnaires was helpful in the sense that the goal of the study was to determine the needs of patients with bipolar disorder. The study was able to identify areas of need as a result of the application of the need for care questionnaire.
However, the use of only the two questionnaires limited the study because no other conclusions could be drawn other than those which the scale had been crafted to measure. The effect of not meeting such needs on the condition of the patients was not determined, neither was the perceived lack of care verified with the hospital care-givers. Overall, given the purpose of the study – to identify care needs – the use of the two questionnaires was appropriate and sufficient.
Another study aimed at identifying mood fluctuations as a core characteristic of bipolar disorder through the use of scaled questionnaires (Hofmann & Meyer, 2006). The researchers used the Hypomanic Personality Scale to identify participants who were at risk for bipolar disorder. The participants were then asked to complete a 28-day diary to measure their mood fluctuations. While diary entries are usually used in qualitative studies, this is because diary entries are usually subjected to content analysis. In the present case however the diary entries accomplished by the participants were more a response to scaled questions. The researchers could therefore easily score the diary entries and assess them based on quantitative means.
The research showed that indeed mood fluctuations were greater in participants who were more at risk for bipolar disorder. The researchers concluded that mood fluctuations might be one of the key characteristics of bipolar disorder – given the prevalence of the same in the participants and given the consistency of such finding between the participants (Hoffman & Meyer, 2006). The researchers correctly resorted to the use of a scale in measuring the mood levels of the participants. Given the very specific focus of the study – finding out the level of mood fluctuations and comparing this with the degree of risk for developing bipolar disorder – the use of qualitative methods or of non-questionnaire formatted scales would have yielded not only less reliable but also more resource-consuming results.
A study conducted by Lewinsohn, Seeley, and Klein (2003) employed a semi-structured interview as a means of obtaining data. The structure of the interview was similar to the scale used in the study of Goossens et al. and Hoffman and Meyer in the sense that the structured interview was based on a pre-established diagnostic structure constructed for the characteristics of bipolar disorder according to the DSM-IV definition (Lewinsohn et al., 2003). The goal of the study was to examine the incidence, prevalence, onset, and course of bipolar disorder in adolescents. In order to attain this goal, diagnostic interviews were conducted with a large group of high school students. The researchers were able to determine the long-term incidence of bipolar disorder by re-interviewing the participants after several years. The relatives of the participants were also diagnosed in order to determine the prevalence of the disorder within the families of participants.
The diagnostic nature of the interview made it easy to scale and reduce to numerical data because of the already identified characteristics of bipolar disorder. The study concluded by finding that the first lifetime onset of bipolarity most often occurred during adolescence. It was also found that relatives of adolescents with bipolar disorder or major depressive disorder were often indicative of the manifestation of bipolar disorder in adolescents.
The researchers were correct in applying a semi-structured interview in their research because by doing so, more participants were accommodated within the time necessary to complete the study. This made the findings more externally valid. However, one weakness of the study is its failure to adequately assess the relationship between presence of bipolar disorder in relatives and adolescents. The diagnostic interview could only account for the presence of the disorder in either or both the relatives and adolescents but could not determine the nature of the co-occurrence.
One study examining the co-occurrence of specific personality disorders by assessing the presence of specific clinical features also used semi-structured interviews coupled with regression analyses to obtain data (Skodol, Stout, McGlashan, Grilo, Gunderson, Shea, Morey, Zanarini, Dyck, & Oldham, 1999). The interviews were clinical in nature and allowed the researchers to scale the clinical characteristics of the participants according to the DSM-IV Axis. The data obtained from the interview was put through a series of regression analyses in order to determine the relationship of each clinical characteristic to the co-occurrence of other mood disorders.
By applying such a method, the researchers were able to identify particular mood disorders, specifically depressive personality disorder, as a co-occurring disorder with bipolarity (Skodol et al., 1999). The research method applied was apt considering that the particular clinical characteristics were already identified and needed only to be paired with the rates of recurrence of bipolarity as well as occurrence of other disorders.
A study conducted by Albuquerque, Deshauer, Fergusson, Doucette, MacWilliam, and Kaufmann (2009) examined the recurrence rates of bipolar disorder in a group of physicians who had been monitored for major depression and bipolar disorder. The researchers used a retrospective cohort design to obtain and assess data. A cohort design is one which studies a group of people who share a common characteristic during a particular period of time (Mann, 2003). In a retrospective cohort design all the data has been gathered prior to the onset of the study itself. The researchers thus rely on existing records to confirm or reject the study hypothesis.
In the present study existing medical records and charts from the monitoring sessions conducted with the participants were obtained and assessed for rates of recurrence. The findings showed that the rates of recurrence were indeed high and in fact in cases of bipolar disorder the physicians were eventually incapacitated to continue with their medical practice (Albuquerque et al., 2009).
The use of the retrospective cohort design was efficient in the sense that the researchers had easily accessible data which could be analyzed for the purposes of the study. However, the restricted population comprising the study pool makes the findings less externally valid despite the quantitative nature of the research. Also, the lack of control over the process of soliciting the data from the participants, i.e. the danger of extraneous variables coming into play during the collection of data and the inability of the researchers to control for this makes the findings less reliable.
Qualitative research designs are geared towards gaining in-depth understanding of human behavior and the factors which govern such behavior (Creswell, 2003). This type of research is linked with the collection and analysis of non-numerical data. Unlike quantitative research designs, qualitative designs have broader study goals. These designs are less rigid than quantitative designs because researchers obtain data without isolating research variables from the context where they may be found. However, the lack of control for extraneous variables causes qualitative research to generally be less reliable than quantitative designs. It is also less externally valid because of the context-rich nature of the data gathered. It should be noted however that qualitative designs provide a rich source of information regarding the relationships of different variables with each other.
In the previous discussion it was shown that the interview can be used for quantitative research designs. This method can also be applied for qualitative research designs. A study examining the meaning of life for individuals diagnosed with bipolar disorder used interviews to obtain information about the participants’ views regarding their disorder and their future (Jonsson, Wijk, Skarsater, & Danielson, 2008). The interview was non-structured and the main questions were simply the two study goals. Follow-up questions were propounded when necessary to prod the participants to expound on their answers or to be more accurate in their description of situations.
The data obtained from the interviews was analyzed through qualitative data analysis. Text relevant to the study goals was marked by the researchers and coded based on similarity in themes. The discussion of the findings was thus guided by the identified themes in the study. The findings revealed that persons diagnosed with bipolar disorder experienced insecurity, as well as underwent challenges in accepting, understanding and managing their condition (Jonsson et al., 2008). Given the reliance on the personal experience of participants in order to attain research goals, the use of a qualitative design was not only appropriate but necessary. If the researchers had used readily available scales to measure the general contentment of the participants with life or other such variables, the particular experience of individuals suffering from bipolar disorder may have been missed out on. The absence of a scale particularly geared to the experience of individuals suffering from bipolar disorder justified the resort to a qualitative interview method.
Another study examining the experience of individuals in coping with bipolar disorder looked into the experience of individuals newly diagnosed with bipolar disorder (Proudfoot, Parker, Benoit, Manicavasagar, Smith, & Gayed, 2009). The data for the study was obtained through online communications between the participants and the researchers. Electronic mails (e-mails) expressing the concerns and difficulties faced by patients newly diagnosed with bipolar disorder were sent to experts who gave informed support to these patients. These e-mails were obtained by the researchers and the communications underwent data analysis in order to give the researchers an idea of the prevalent concerns among newly diagnosed patients.
The study showed that the main concerns of participants included unwanted side-effects of medicines, coping mechanisms for unpleasant symptoms, loss of sense of self, uncertainty about their future, stigma, and others (Proudfoot et al., 2009). It should be observed however that e-mail as a medium of communication is rather limiting for the grievant. While some may find this medium as a comfortable one to express the whole range of emotions and concerns they are undergoing, a majority of people find this medium as restricting and would voice out only major concerns while repressing what they may deem as more mundane concerns. Also, the design was built around a help system and so more of the participants may have focused on formal complaints rather than by voicing out fears, difficulties, insecurities and the like.
The study conducted by Proudfoot et al. (2009) could be contrasted with the study conducted by Inder, Crowe, Moor, Luty, Carter, and Joyce (2008) who obtained their data through several therapy sessions with the participants. The sessions were recorded, transcribed and subjected to data analysis – just as in the previous study conducted by Jonsson et al. (2008). However, the difference between the present study and that of Jonsson et al. as well as Proudfoot et al. is the building of a relational foundation between the researchers and the participants. In the present study the researchers were able to establish a deeper understanding of the participants because of the regular therapy sessions which the participants underwent. The researchers were thus able to probe deeper into the issues voiced out by the participants thereby giving them a greater understanding of the basis for and extent of the concerns.
In the study the goal was to examine the difficulties of developing a sense of self and identity in the case of individuals suffering from bipolar disorder. The researchers found that bipolarity created experiences of confusion, contradiction, and self doubt causing the participants difficulty in establishing continuity in their sense of self (Inder et al., 2008). Unlike the previous studies, the present study was able to explore countering mechanisms which might alleviate the situation of participants. The regular therapy sessions allowed for feedback as to what methods might improve the condition of participants. The researchers found that by eliminating external definitions of self based on the illness and by promoting self-acceptance and incorporating different aspects of the self, a more integrated sense of identity could be established (Inder et al., 2008).
Qualitative designs are not all dependent on data analysis. For example, the case study is a longitudinal study which examines the experience of particular individuals possessing the characteristics being observed. A case study was performed in connection with assessment of a treatment for bipolar disorder. The goal of the study was to examine whether a psychotherapy previously used in adults suffering from bipolar disorder could likewise work for adolescents (Crowe, Inder, Joyce, Moor, Carter, & Luty, 2008).
The proposed treatment was the interpersonal social rhythm therapy and this was applied over a period of 30 sessions to the youngest identified willing participant in the first screening of the study (Crowe et al., 2008). Data was obtained by assessing the participant’s significant relationships and assessing the impact of these relationships on the participant’s mood instability. The research also held the goal of identifying any patterns in the socialization of the participant. By constantly monitoring the developments in the participant’s social interactions, these goals were met. The researchers found that the interpersonal social rhythm therapy could be customized to adapt to the needs and issues faced by an adolescent.
The study by Crowe et al. (2008) provides great insight into the challenges faced by an adolescent suffering from bipolar disorder. However, the study is low in external validity given the fact that only one individual was studied and tested. Therefore, the peculiarities of such individual’s life and personality played a significant role in determining the outcome of the study. However, the study still has clinical implications because it shows that a therapy process which had previously been applied only to adults has the capacity to cater to the developmental needs of adolescents as well.
Bipolar disorder is indeed a growing concern not only among adults but more so amongst young adults. Research has shown that the onset of bipolar disorder may be observed even in adolescence and continuing therefrom up to old age. This, compounded with the sparse research conducted in this field, shows that importance should be attached to the needs of young adults suffering from this disorder.
The discussions have shown that both quantitative and qualitative methods are applicable in studies of bipolar disorder. Quantitative methods which have been applied are scaled interviews, scaled questionnaires, regression analyses, cohort designs, and the like. When conducting research that has specific goals and areas of inquiry, quantitative methods are useful and effective. Moreover, where established scales have already been constructed and only their application to a new population is inquired into or their application alongside other scales or measures is the purpose, then quantitative designs are highly recommended.
On the other hand, where the research inquiry is broad and the researchers are seeking out answers to questions regarding experiences, relationships of variables, socially rich inquiries and context-heavy variables then qualitative research is more appropriate. Some qualitative research methods that were illustrated herein were the interview, correspondence analysis, therapy sessions, and case study. Much of qualitative research obtains data through data analysis and coding. The two designs are not mutually exclusive and research may be conducted with a mixed design – one incorporating both quantitative and qualitative methods in the design. The determining factor as to the appropriate design to be employed in any given study is the goal of the study. Thus, a good understanding of the research goals and points of inquiry is necessary to establish an appropriate research design.
Albuquerque, J, Deshauer, D, Fergusson, D, Doucette, S, MacWilliam C, & Kaufmann, M (2009). Recurrence Rates in Ontario Physicians Monitored for Major Depression and Bipolar Disorder. Canadian Journal of Psychiatry, 54(11), 777-782.
American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association.
Creswell, JW (2003). Research design: Qualitative, quantitative, and mixed method approaches. CA: Sage Publications.
Crowe, M, Inder, M, Joyce, P, Moor, S, Carter, J, & Luty, S (2008). A developmental approach to the treatment of bipolar disorder: IPSRT with an adolescent.
Goossens, PJJ, Knoppert-Van Der Klein, EAM, Kroon, H, & Van Achterberg, T (2007). Self-reported care needs of outpatients with a bipolar disorder in the Netherlands. Journal of Psychiatric and Mental Health Nursing, 14, 549-557.
Hoffman, BU, & Meyer, TD (2006). Mood fluctuations in people putatively at risk for bipolar disorders. British Journal of Clinical Psychology, 45, 105-110.
Inder, ML, Crowe, MT, Moor, S, Luty SE, Carter, JD, & Joyce, PR (2008). “I Actually Don’t Know Who I Am”: The Impact of Bipolar Disorder on the Development of Self. Psychiatry, 71(2), 123-133.
Jonsson, PD, Wijk, H, Skarsater, I, & Danielson, E (2008). Persons Living with Bipolar Disorder – Their View of the Illness and the Future. Issues in Mental Health Nursing, 29, 1217-1236.
Lewinsohn, PM, Seeley, JR, & Klein, DN (2003). Bipolar disorder during adolescence. Acta Psychiatrica Scandinavia, 108(418), 47-50.
Mann, CJ (2003). Observational research methods. Emergency Medical Journal, 20, 54-60.
Proudfoot, JG, Parker, GB, Benoit, M, Manicavasagar, V, Smith, M, & Gayed, A (2009). What happens after diagnosis? Understanding the experiences of patients with newly-diagnosed bipolar disorder. Health Expectations, 12, 120-129.
Skodol, AE, Stout, RL, McGlashan, TH, Grilo, CM, Gunderson, JG, Shea, MT, Morey, LC, Zanarini, MC, Dyck, IR, & Oldham, JM (1999). Co-occurrence of mood and personality disorders: a report from the collaborative longitudinal personality disorders study. Depression and Anxiety, 10, 175-182.
U.S. Census Bureau (2005). U.S.Census Bureau Population Estimates by Demographic Characteristics. Retrieved on 8 August 2010 from http://www.census.gov/popest/national/asrhl/