The Power of the Written Word by Naomi Liles Crouse A thesis submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Master of Science Auburn, Alabama August 9, 2010 Copyright 2010 by Naomi Liles Crouse Approved by Anita All, Chair, Professor of Nursing Ramona Lazenby, Professor of Nursing Francine Parker, Associate Professor of Nursing ii Abstract A diagnosis of cancer can catapult a person into a time period dominated by uncertainty and a perceived decrease in quality of life. Literature supports that writing honestly about emotions associated with traumas may address and reverse negative emotions and increase people?s positive functioning. The purpose of this study was to investigate if expressive writing made a difference in uncertainty and quality of life for newly-diagnosed cancer patients. Fourteen patients just beginning their treatment phases were recruited and randomized into writing and non-writing groups. The writing group was given Pennebaker?s writing prompt. Mishel?s Uncertainty in Illness scale and Ferrans and Powers Quality of Life Index were administered at entry into the study and after three months. Findings were not statistically significant, however the small sample size and having multiple cancer diagnoses may have contributed to this. Findings reinforced uniqueness of illness experience and quality of life for each individual. iii Acknowledgments This thesis represents my personal ?K2,? a mountain I declared I would never ?climb.? But now, standing on the summit, I can honestly say I?m glad I did. However, I could not have done it without several invaluable ?sherpas? guiding me along this journey. Therefore I would like to extend my deepest gratitude and appreciation to? Dr. Anita All ? my committee chair, who has become a mentor and friend. Thank you for your understanding, your belief in me, and for challenging me to meet your high standards. Your guidance has directed me toward multiple opportunities for professional growth and your confidence in me has given me the courage to pursue them. Dr. Francine Parker and Dr. Ramona Lazenby ? my committee, who were sources of encouragement and advice, including not to take this assignment too seriously. Finally, thank you to my three favorite guys - Gil, Lewis and Mark ? for being my personal cheerleaders; for your understanding and patience with the unending school and homework; and your love, encouragement, support (and editing help). You have decreased my uncertainty and increased my quality of life. You give my life meaning. I love you more. iv Table of Contents Abstract ............................................................................................................................... ii Acknowledgments.............................................................................................................. iii Table of Contents ............................................................................................................... iv List of Tables ..................................................................................................................... vi List of Figures ................................................................................................................... vii List of Abbreviations ....................................................................................................... viii Chapter I Introduction .........................................................................................................1 Chapter II Literature Review ............................................................................................13 Chapter III Methodology ..................................................................................................59 Chapter IV Analysis of Data .............................................................................................74 Chapter V Discussion, Implications, Summary ................................................................84 References ......................................................................................................................103 Appendix A East Alabama Medical Center Institutional Review Board Approval ......123 Appendix B Montgomery Cancer Center Institutional Review Board Approval ..........125 Appendix C Auburn University Institutional Review Board Approval.........................129 Appendix D Study Flier .................................................................................................141 Appendix E Recruitment Script .....................................................................................143 Appendix F East Alabama Medical Center Informed Consent .....................................145 Appendix G Montgomery Cancer Center Informed Consent ........................................148 v Appendix H Personal Address Sheet .............................................................................151 Appendix I Demographic Survey ..................................................................................153 Appendix J Mishel Uncertainty in Illness Scale ? Community form ............................155 Appendix K Ferrans and Powers Quality of Life Index ? Cancer Version III ..............160 Appendix L Writing Information Sheet .........................................................................165 Appendix M Auburn Psychological Referral List .........................................................167 Appendix N Montgomery Psychological Referral List .................................................169 Appendix O Expressive Writing Information Sheet ......................................................171 vi List of Tables Table 1 Comparison of Intervention and Control Group Demographics ..........................76 Table 2 Group Total, Mean and Standard Deviation for Uncertainty in Illness Scale ......80 Table 3 Group Total, Mean and Standard Deviation for Quality of Life Index ................81 Table 4 Comparison of Item Means of Highest Uncertainty and Lowest Uncertainty in Intervention Group at Baseline and Three Months ............................................................86 Table 5 Comparison of Item Means of Highest Uncertainty and Lowest Uncertainty in Control Group at Baseline and Three Months ...................................................................87 Table 6 Comparison of Subscale Means of Intervention Group and Control Group at Baseline and Three Months ...............................................................................................92 Table 7 Comparison of Weighted Means Showing Items with Highest Importance and Satisfaction in Intervention and Control Groups at Baseline and Three Months ..............94 Table 8 Comparison of Weighted Means Showing Items with Highest Importance but Lowest Satisfaction in Intervention and Control Groups at Baseline and Three Months .95 vii List of Figures Figure 1 Comparison of Mean Uncertainty Tool Question Scores from Baseline to Three Months for Intervention Group ..........................................................................................89 Figure 2 Comparison of Mean Uncertainty Tool Question Scores from Baseline to Three Months for Control Group .................................................................................................90 Figure 3 Comparison of Weighted Mean Quality of Life Question Scores from Baseline and Three Months for Intervention Group .........................................................................96 Figure 4 Comparison of Weighted Mean Quality of Life Question Scores from Baseline and Three Months for Control Group ................................................................................97 Figure 5 Changes in Combined Uncertainty and Quality of Life Scores for Each Participant from Baseline and Three Months ....................................................................99 viii List of Abbreviations M or m = mean S.D. or sd = standard deviation N = group sample n = subgroup sample ANOVA = analysis of variance MCC = Montgomery Cancer Center EAMC = East Alabama Medical Center IRB = Institutional Review Board MUIS = Mishel Uncertainty in Illness Scale MUIS-C = Mishel Uncertainty in Illness Scale ? Community form QLI = Quality of Life Index QLI-C = Quality of Life Index ? Cancer Version III QOL ? CS = City of Hope Quality of Life ? Cancer Patient version USS = Uncertainty Stress Scale IOM = Institute of Medicine QL-Index = Spitzer Quality of Life Index 1 CHAPTER I INTRODUCTION Personal injury or illness is consistently listed as one of life?s top stressors, along with death of a spouse and divorce. According to the American Cancer Society (2009), over 1.4 million people will be diagnosed with cancer this year. For these recently- diagnosed cancer patients, life turns upside down. Their physical body is under attack, their future uncertain, and emotionally, psychologically and cognitively they are reeling (Ferrell et al., 2005; Morgan, Graves, Poggi, & Cheson, 2008). A diagnosis of cancer generates overwhelming uncertainty. ?Will I die? Is this really happening? What if they mixed up the test results and these are not even mine? How will cancer affect my life? I have plans! When will life get back to normal? Will I live to see my children grown up and married? Will my husband or wife leave me? Does this doctor know what he or she is doing? Will I have much pain?? These questions and the uncertainty inherent in them can lead to a variety of emotional reactions. Those newly diagnosed with cancer may feel fear, sadness, hopelessness, helplessness (Ferrell & Coyle, 2008), shock (Morgan et al., 2008), anxiety, anger, grief, (Hochwald, 2008), betrayal by their bodies, a loss of control over their lives (both for the present and the fore-seeable future) (Boehmke & Dickerson, 2006), denial, and many other human emotions (Pennebaker & Chung, 2007; Penrod & Morse, 1997). The experience of cancer can affect all aspects of life: multiple roles and relationships, the ability to work, the luxury of relaxing and having some control over a 2 daily schedule, one?s spirituality, and the personal vision of a future. All of these factors influence one?s perception of uncertainty and quality of life (Ferrell et al., 2005). Quality of life is characterized as a multidimensional concept (Andersen, Kiecolt- Glaser, & Glaser, 1994; Ferrell, Wisdom, & Wenzl, 1989; Osaba, 1994). Literature described it as having several components: physical or health, general functioning, psychological, spiritual, social, economic, and family or significant others (Ferrans & Powers, 1992; Ferrell, Grant, Padilla, Vemuri, & Rhiner, 1991). Introducing cancer into a life has a negative impact on quality of life, at least temporarily (Stanton et al., 2002). An individual?s response to this diagnosis depends on their personality and normal coping behaviors. Some people cope through expressing themselves by crying, talking, writing (journaling) or more recently, by blogging (Lyubomirsky, Sousa, & Dickerhoff, 2006). Others internalize their emotions and maintain a staid exterior, and there are those who stay in denial, avoiding and ignoring their emotions altogether (Pennebaker & Beall, 1986). While each is a normal reaction, literature has revealed that not expressing emotions can negatively affect one?s health, physically as well as psychologically (Lyubomirsky et al., 2006; Pennebaker & Beall, 1986). Expressive writing is a patient-focused intervention which has gained increasing attention in clinical settings in the last two decades. Research has suggested that expressing thoughts on paper has health benefits to those who participate (Frattoroli, 2006; Pennebaker & Beall, 1986). There are differing opinions as to why writing makes a difference in the writers? health, but writing out one?s feelings regarding a traumatic event, whether in the past or present, can result in both physiological and psychological benefits. Some of the benefits identified have included less intentional isolation and more 3 honest social interactions (Niederhoffer & Pennebaker, 2002), decreased anger, distress, depression and anxiety (Frattoroli, 2006), improved positive overall functioning and outlook (Frattoroli, 2006; Smyth, Stone, Hurewitz, & Kaell, 1999), improvements in liver and immune functioning (Frattoroli, 2006), and fewer doctor visits (Pennebaker & Beall, 1986). In order to give holistic care, nurses need to address not only the physical symptoms of cancer, but also assess for and identify other areas of the individual?s life that might also have been affected by the cancer, including psychological, social/family/significant others, spirituality (Ferrans & Powers, 1985; Ferrell et al., 2005) and socioeconomic (Ferrans & Powers, 1985). Chemotherapy, radiation and medications treat the physical body, but therapies are needed for addressing other areas ?wounded? by the uncertainty and perceived threats to the individual?s quality of life (Ferrell et al., 2005). Current Research Review of the literature revealed research that has recently been conducted on expressive writing with cancer patients ( Frattaroli, 2006; K?llay & B?ban, 2008; Laccetti, 2007; Morgan et al., 2008), and uncertainty and quality of life in patients (Eastwood, Doering, Roper, & Hays, 2008; Sammarco, 2001; Wallace, 2003: Wonghongkul, Dechaprom, Phumivichuvate, & Losawatkul, 2006). However, there have been very few studies (Laccetti, 2007; Rosenberg et al., 2002; Stanton et al., 2002) which looked at the effect of expressive writing on perceived quality of life in cancer patients, and none which additionally measured specifically for uncertainty. This lack of literature suggested a need to conduct research in this area. 4 Purpose The purpose of this research was to investigate whether expressive writing makes a positive difference in the lives of those newly diagnosed with cancer with regard to their feelings of uncertainty and perceived quality of life. Information obtained from this research will add to the growing amount of literature which reports the link between expressive writing and positive outcomes. This research can contribute to the literature on the effects of expressive writing in a general cancer population. Prior studies were composed mainly of breast cancer or prostate cancer patients, with only a few studies involving the general cancer population. This research will add to the combined measurement of quality of life and uncertainty in the general cancer population. Study questions Research measuring uncertainty and quality of life in patients was found in the literature (Eastwood et al., 2008; Sammarco, 2001; Wallace, 2003), as well as research on expressive writing in various population groups (Frattaroli, 2006; K?llay & B?ban, 2008; Laccetti, 2007; Morgan et al., 2008; Smyth et al., 1999) however, no studies were located which looked specifically at writing?s effects on uncertainty and quality of life in the lives of newly-diagnosed cancer patients. A non-experimental, causal-comparative study design was chosen to guide this research; therefore study questions were appropriate instead of a hypothesis (Gall et al., 2007). The following study questions guided the research: 1. Does expressive writing cause a difference in cancer patients? perceived level of uncertainty after writing for three months? 5 2. Does expressive writing cause a difference in cancer patients? perception of their quality of life after writing for three months? 3. Does expressive writing affect the relationship between quality of life and uncertainty in the intervention group as compared to the control group? Theoretical and Operational Definitions Before a study can commence, important terms must be defined. Theoretical definitions describe the abstract meaning of a concept, while operational definitions delineate how the concept is used and measured in a study (Polit & Beck, 2008). Uncertainty Life is full of events which cause uncertainty. From the mundane and immediate, such as wondering which cereal will taste better, to the life-altering and lingering questions where life may literally hang in the balance, people can face varying degrees of uncertainty every day. Theoretical definition. Uncertainty has been defined as ?the quality or state of being uncertain, or in doubt; something that is uncertain? (Merriam-Webster Online Dictionary, 2008). It is a ?dynamic, discomforting, non-normative state?, one which is unique and based on each individual?s perception of their situation (Penrod, 2007, p. 663). Uncertainty causes a person to become stuck in the present; unsure of how to proceed and move forward to his or her future goals. In essence, ?hope was paralyzed? (Penrod, 2001, p. 243). Operational definition. Uncertainty has been measured through Mishel?s Uncertainty in Illness scale (1989). Since this study looked beyond uncertainty to the more specific application of 6 uncertainty in illness, this scale was chosen instead of other scales which looked at uncertainty in a more general way. Uncertainty in Illness Uncertainty is a very real part of any illness, but especially in the cancer experience. It is highest at time of diagnosis, and gradually decreases as the causes of the uncertainty are taken care of, or over time as the individual learns how to cope and adapt (Hj?rleifsd?ttir, Hallberg, Gunnarsd?ttir, & Bolmsj?, 2008; Stanton et al., 2002). Uncertainty never goes away completely however, but remains to some degree with the individual for the rest of his or her life (Ferrell, Grant, Funk, Otis-Green, &Garcia, 1998; Hj?rleifsd?ttir et al., 2008). Theoretical definition. Mishel (1988) described uncertainty in illness as ?the inability to determine the meaning of illness-related events? (p. 225). These uncertainty-causing events have the characteristics of being ?vague, ambiguous, unpredictable, unfamiliar, inconsistent, or lacking information? (Mishel, 1984, p.163). Based on these characteristics, three attributes of uncertainty were identified: ?probability, temporality and perception? (McCormick, 2002, p.129). Operational definition. Mishel?s Uncertainty in Illness scale ? Community form was used to define the concept in this study (Mishel, 1989). Four key characteristics are addressed in this scale. These characteristics include ?inconsistency, unpredictability, ambiguity, and complexity? (Mishel, 1997b, p. 6). The Uncertainty in Illness scale ? Community form is composed of 23 questions covering these four uncertainty characteristics. A 5-point 7 Likert-type scale is used to answer each question. The scale consists of numbers 1-5, where 1 = strongly disagree and 5 = strongly agree. The higher the score, the greater the feelings of uncertainty. The domains of this scale include: Inconsistency. This domain includes questions regarding the inconsistency of the information being given the patient by various health care providers; Unpredictability. This domain includes a question regarding the inconsistencies between treatments given and anticipated and actual outcomes; Ambiguity. This domain includes questions regarding the ambiguity of the state or stage of the disease; Complexity. This domain includes questions regarding the treatments and care given to the patient (Mishel, 1997b, p. 8). Quality of Life Each individual has a unique perspective on his own quality of life. It cannot be assessed for him nor assumed, for no one fully knows another person?s deepest satisfactions or highest priorities except for that individual. Theoretical definition. Quality. (noun) a peculiar and essential character, an inherent feature, a distinguishing attribute; (adjective) being of high quality (Merriam-Webster Online Dictionary, 2008). Life. (noun) a way or manner of living (Merriam-Webster Online Dictionary, 2008). 8 Quality of life. (noun) your personal satisfaction (or dissatisfaction) with the cultural or intellectual conditions under which you live (as distinct from material comfort) (WordNet 1..7.1, 2001). Quality of life is a multidimensional concept (Andersen et al., 1994; Ferrell et al., 1989; Osaba, 1994). Ferrans and Powers (1992) defined quality of life as ?a person?s sense of well-being that stems from satisfaction or dissatisfaction with the areas of life that are important to him/her? (p. 29). This definition highlights the uniqueness of each individual?s perception of his experiences and how these perceptions impact his quality of life. According to Ferrans and Powers (1992) quality of life comprises four domains: health and functioning, psychological/spiritual, social and economic, and family. Operational definition. Ferrans and Powers Quality of Life Index - Cancer Version III was used in this study to measure and define quality of life (Ferrans & Powers, 1998a). This index, a self- report survey, is composed of 66 questions covering the four domains of ?health and functioning, psychological/spiritual, socioeconomic and family?: 33 questions on how satisfied the individual is with an item in their life, and 33 questions on how important that item is to them. A 6-point Likert-type scale is used in this survey, with 1 = very dissatisfied or very unimportant, and 6 = very satisfied or very important. The domains of this scale include: Health and functioning. This domain includes questions about the usefulness to others, physical independence, ability to meet family responsibilities, own health, pain, energy (fatigue), stress or worries, 9 control over own life, leisure-time activities, potential for a long life/retirement, ability to travel on vacations, sex life, health care; Psychological and spiritual. This domain includes questions about satisfaction with life and self, happiness in general, achievement of personal goals, peace of mind, personal appearance and personal faith in God; Social and economic. This domain includes questions about standard of living, financial independence, home (house or apartment), neighborhood, job/unemployment, friends, emotional support from others, education, and influence in government; Family. This domain includes questions on family happiness and health, children, relationship with spouse or significant other, and emotional support from family (Ferrans & Powers, 1992, p. 295). Expressive Writing Expressive writing, reflective journaling, experimental or emotional disclosure: all are names describing the act of writing down one?s private thoughts. It is the baring of one?s soul on paper, sometimes to be kept private, and other times to be read or shared with others. Research has suggested that expressive writing is associated with positive physical and psychological effects for clinical populations (Frisina, Borod, & Lepore, 2004). 10 Theoretical definition. Expressive. (adjective) 1. of or relating to expression, 2. serving to express, utter or represent, 3. effectively conveying meaning or feeling (Merriam Webster Online Dictionary, 2008). Writing. (noun) 1. the act or process of one who writes as: a) the act or art of forming visible letters or characters; b) the act or practice of literary or musical composition 2. something written as: a) letters or characters that serve as visible signs of ideas, words or symbols; b) a letter, note or notice used to communicate or record; c) a written composition (Merriam-Webster Online Dictionary, 2008). Expressive writing. Writing with the intent to divulge honest thoughts and emotions. Operational definition. Countless studies found in the literature used Pennebaker?s original writing prompt or a variation of it for their research study (e.g. Kovac & Range, 2002; Morgan et al., 2008; Rosenberg et al, 2002; Smyth et al., 1999). Following their example, the current study participants were asked to write about ?their very deepest thoughts and feelings? about past or present traumas, conflicts or stressors. However, they were also given the option to write about anything else, negative or positive, provided they included honest thoughts, feelings or emotions. Participants were asked to write at a minimum of twice a week, with no specific amount of time set per episode. 11 Study Assumptions Assumptions are defined as ?principles which are accepted as being true based on logic or reason, without proof? (Polit & Beck, 2008, p. 748). Assumptions for this study were: 1. Participants would actually do some type of writing on a regular basis - regular being at least twice a week - throughout the three months of being in the study. 2. Participants would answer survey items honestly and correctly. 3. Participants would return surveys in the time and manner requested. 4. Data collected would contribute to the knowledge base about beneficial interventions for cancer patients. Study Limitations Limitations are those things over which the researcher has no control, but which can affect the accuracy or purity of the study findings and which must be taken into account during the interpretations of the findings. Limitations for this study included, but were not limited to: 1. Participants? responses may have been influenced by their diagnosis and prognosis. 2. Participants? responses may have been influenced by the progression of their disease, or by metastasis of disease. 3. Participants? responses may have been influenced by their treatment regime and the intensity or lack of side effects. 4. Participants? responses may have been influenced by their degree of spirituality, which was not assessed in this study. 12 5. Participants? responses may have been influenced by the presence or lack of a support system, which was not assessed in this study. 6. Participants? responses may have been influenced by their age. 7. Participants? responses may have been influenced by previous and ongoing uncertainties and quality of life. 8. Participants? responses may have had limitations imposed by self-identification and self-report of uncertainties and perceived quality of life. 9. Participants? responses may have been influenced by the Hawthorne effect. Gall, Gall and Borg (2007) defined the Hawthorne effect as any situation which causes participants to respond in a different way (than they would have) due to the fact that they know they are in a study, they know the goals of the study, or they are receiving special attention for being in the study. 10. This researcher mailed surveys to participants, therefore the time, place and setting in which the surveys were completed could not be controlled beyond requesting that the outlined procedures be followed. Summary A diagnosis of cancer will always bring with it accompanying emotions such as uncertainty, anxiety and distress. Keeping these emotions to oneself has been found to negatively impact one?s health, both physiologically as well as psychologically. In contrast, allowing these emotions to be recognized and processed through writing has shown positive health benefits. For cancer patients, writing could be an inexpensive intervention; a way for nurses or others to give more holistic care. 13 CHAPTER II LITERATURE REVIEW Improvements in cancer outcomes continue as new technology and treatment options are discovered and become available. Whereas a few decades ago a cancer diagnosis would have been a death sentence, thankfully now the prognosis for some cancers can be less grim. That said, a diagnosis of cancer, and the uncertainty which accompanies it, still evokes a myriad of intense emotions (Boehmke & Dickerson, 2006; Hj?rleifsd?ttir et al., 2008). These emotions in turn often have a negative effect on an individual?s perceived quality of life (Wallace, 2003). Quality of life is important as it is what gives life meaning. Treatments can wear a person down physically, and the uncertainty and reduced perceived quality of life can be psychologically draining. Overwhelmed by it all, patients can experience helplessness and a feeling of powerlessness (Rancour & Brauer, 2003). Research has reported that writing expressively about one?s emotions can help reduce negative emotions such as anxiety, distress and depression (Frattaroli, 2006). As these emotions decrease, there is a corresponding increase in the perceived quality of life (Stanton et al, 2002). To empower, by definition, is to ?enable? and/or ?promote the self-actualization of? someone or something (Merriam-Webster?s Online Dictionary, 2009). Therefore, the act of writing empowers individuals to be a part of their own treatment plan. While medical treatments fight cancer on a physiological level, writing can fight the effects of cancer on a psychological 14 level. This review of literature will describe the evolution of the various components and their current state. Review of the Literature An extensive literature review was conducted prior to initiating this research project. A careful review of literature must be done in order to guarantee the proposed research project will contribute to an expanding body of knowledge (Gall et al., 2007). Literature review focused on four main topics: individuals with cancer, the concept of uncertainty, the concept of quality of life, and expressive writing/reflective journaling. This search was conducted not only in nursing literature, but also that of psychology, sociology and medicine. This review of literature found very few studies which examined the effect of expressive writing in a general cancer population. Most studies were conducted with specific cancer populations. There were also few studies which explored individuals? feelings of uncertainty and quality of life. Most studies looked at one or the other, rarely both. Individuals with Cancer It is estimated that in 2009 there will be nearly 1.5 million adults in America diagnosed with some type of cancer (excluding squamous and basal cell skin cancer and all in-situ carcinomas except urinary bladder) (American Cancer Society, 2009). The state of Alabama is projected to have a little over 24,000 newly diagnosed cases (American Cancer Society, 2009). The main cancer sites are breast cancer (for women) and prostate cancer (for men), followed by cancer of the lung or bronchus, and colon or rectal cancer. According to the American Cancer Society (2009), over the course of a 15 lifetime and with the chances increasing with age, women have about a 33% chance of getting cancer, whereas men are closer to a 50% chance that they will get cancer. No matter when it comes, the diagnosis of cancer brings with it an incredible variety of intense emotions. Studies found that even before the actual diagnosis was made, individuals experienced feelings of uncertainty (Morgan et al., 2008), anxiety, and fear of the unknown (Hochwald, 2008). Once the diagnosis was official, emotions often intensified with the shock of this new reality (Morgan et al., 2008). Whereas the uncertainty of ?what is wrong with me? had been settled, suddenly a whole new set of fears and uncertainties plagued an individual: ?Will treatments work? Will I be horribly disfigured? Will I die? Will treatments and their side effects be too painful for me to stand? What will happen to my family? Can I continue working? Who will stay and support me, and who will abandon me when they hear of the diagnosis?? Individuals felt overwhelmed with their new reality and all of the decisions which must be made (Beatty, Oxlad, Koczwara, & Wade, 2008; Boehmke & Dickerson, 2006; Penrod & Morse, 1997). Some individuals experienced denial (Boehmke & Dickerson, 2006; Doell, 2008; Pennebaker & Chung, 2007; Penrod & Morse,1997; Carver et al., 1993), others behavioral disengagement (Pennebaker & Chung, 2007; Carver et al., 1993), grief, and/or anger (Hochwald, 2008) fear, anguish, self blame and/or shock (Beatty et al., 2008) and feelings of being betrayed by their bodies (Boehmke & Dickerson, 2006). Individuals often cope through the expression of these emotions, either verbally or through writing. Family members, friends, and medical healthcare professionals can all listen and help an individual work through their emotions, but individuals do not always feel comfortable expressing emotions to those closest to them. Individuals may feel they 16 need to be strong (Smith, Anderson-Hanley, Langrock, & Compas, 2005) or stay positive (Beatty et al., 2008). Others may feel embarrassed by the diagnosis or afraid of being treated differently. Some individuals get tired of saying the same thing over and over ? and perhaps think that others get tired of listening to the same thing ? and yet are still dealing with whatever the issue or emotion is and thus need to continue to process what is happening. Support groups were an outlet mentioned in the literature which allowed and encouraged verbalization of emotions. Here individuals found others who were going through the same or very similar situation. Often individuals experienced new strength from knowing that they were not the only one going through this crisis or feeling a certain side effect or emotion. They felt freer to verbalize what they were thinking or feeling, knowing they would be understood. Support groups can offer suggestions for ways to cope with various therapies or the uncomfortable side effects, as well as general encouragement. These support groups can meet in a physical location, or be held online in chat rooms. The American Cancer Society local chapter in Lee county, Alabama, offers a variety of support groups including the ?I Can Cope? group which meets in Opelika, two ?Look Good, Feel Better? support groups - one which meets in Montgomery and the other in Opelika, a ?Reach to Recovery? support group which meets in Montgomery, and two general support groups, one which meets in Opelika and one in Alexander City. There are two support groups specifically for those with breast cancer: one in Opelika sponsored by the American Cancer Society, and the other in Montgomery called ?Women of Hope? sponsored by Frazer Memorial United Methodist Church. 17 For those individuals who chose not to bare their soul verbally to loved ones around them, some found release through writing. Writing can be very private through the use of a journal, or more open through an online journal or blog. Through writing privately in a journal, individuals are free to express any and all emotions as intensely as they are being experienced. No words need to be checked, no tongue bit. Raw emotions can pour forth from the pen, coming from the depths of the soul (Rancour & Brauer, 2003). If individuals desire to share their emotions and experiences with others, then an online journal or web log (blog) can be created. A patient can set up his or her own blog, or utilize websites which have been created specifically for patients going through a health crisis where the patient or family members can keep friends and family updated on the patient?s condition. The Caring Bridge website is an example of a website of blogs created specifically for those going through illnesses. Blogs allow patients to diary about daily happenings, how they are feeling both physically and psychologically, and anything else they want to share, and they can also post pictures. Most blogs are set up to allow readers to post responses and encouragement, thus creating a virtual circle of support around the patient, unrestricted by distance (Ko & Kuo, 2009). There are individuals however who choose not to express any emotion. Their coping mechanism is to either stay on the island of denial, or to keep everything bottled up inside, maintaining a stoic or positive front, and never acknowledging the depth and breadth of their emotions. This style of coping may be adequate during normal life happenings, but may not give effective support during major life changes such as a diagnosis of cancer (Iwamitsu et al., 2003). Additionally, research has suggested that 18 talking or writing about traumatic events could improve ones? physical and mental health (Pennebaker, 1999), a finding which should be relevant to anyone facing a cancer diagnosis. Uncertainty The definition of uncertainty varies depending on the subject matter. WordNet (2006), a dictionary website sponsored by Princeton University, defines uncertainty as ?being unsettled or in doubt or dependent on chance; the state of being unsure of something?. The APA Dictionary of Psychology (2007) defines uncertainty as: ?1. the state or condition in which something (e.g. the probability of a particular outcome) is not accurately or precisely known. 2. lack of confidence or clarity in one?s ideas, decisions or intentions? (p. 966). A definition of uncertainty used in cancer research is ?the lack of sureness or confidence in results of measurements or predictions of quantities owing to stochastic variation or to a lack of knowledge founded on an incomplete characterization, understanding or measurement of a system? (Genetic Consequences of Cancer Treatment, 2008, p. 1). Dictionary.com Unabridged (n.d.) added another dimension to the definition with its? third definition: ?unpredictability; indeterminacy; indefiniteness?. In this literature search, the articles fell into one of three groups: uncertainty as a concept, intolerance of uncertainty, and uncertainty in illness. Interestingly, each of these groups has at least one established scale or tool with which to measure or study its particular focus of uncertainty. Uncertainty concept. While exploring the relationship between the concepts of enduring, suffering and hope, Morse and Penrod (1999) discovered uncertainty was a related concept. 19 Uncertainty was described as a middle ground where the ultimate goal was known, but the best route to that goal was not known. The individual was stuck in the present with an uncertain future, somewhere between the states of suffering and hope. A lack of knowledge and information about the various options which could lead to reaching the goal contributed to the uncertainty. In this description, hope was ?paralyzed? (Morse & Penrod, 1999, p.148). Penrod studied the general concept of uncertainty further, and defined uncertainty?s attributes as a ?dynamic? and ?pervasive but non-normative state,? having a ?discomforting, uneasy sensation,? and as one which was impacted by an individual?s perceptions of their degree of control and confidence (Penrod, 2001, 2007, p. 663). While not using the term ?dynamic,? Hilton (1994) described the flexible nature of uncertainty as one which could wax and wane and believed that uncertainty was always accompanied by both positive and negative emotions (p. 18). McCormick, however, identified three different defining attributes: ?temporality, probability and perception? (McCormick, 2002, p. 129). Temporality was described as not knowing how long the source of uncertainty would last. The end was not in sight; the future was unknown. Probability spoke to the likelihood of something happening, and in uncertainty that probability could not be pinned down. Perception was how one saw the uncertainty at that point in time; it was unique and individualized. If an individual tried to classify the perception of uncertainty, they could not. Was the situation causing the uncertainty seen as a threat or an opportunity (Mishel, 1988)? Would it produce a positive outcome or a negative one? Penrod (2007) identified these same three attributes but categorized them as strategies to help manage uncertainty. 20 Uncertainty could have a positive outcome as an individual perceived greater control and confidence, and experienced growth towards a ?new normal? (Penrod, 2007, p. 663). Individuals saw life in a new way (Mishel, 1988; Penrod, 2007) and experienced hope (Hilton, 1994; Mishel, 1988). Uncertainty could also be seen as a negative outcome where the perception was of a decline of control and confidence which resulted in the individual feeling immobilized (Penrod, 2007), both with hope (Morse & Penrod, 1999) and in their coping mechanisms (Hilton, 1994). Anxiety, depression (Hilton, 1994; Leite & Kuiper, 2008), frustration, fear, helplessness (Hilton 1994) and distress (Berenbaurm, Bredemeier, & Thompson, 2008) were all identified as negative outcomes. Intolerance of uncertainty. ?Intolerance of uncertainty? has been explored in the halls of psychology (Berenbaum et al., 2008; Bredemeier & Berenbaum, 2008; Leite & Kuiper, 2008). Intolerance of uncertainty can be linked to one?s perception of self and the ability to tolerate change and uncertainty, both in life and inside one?s self. Those more intolerant of uncertainty experience greater discomfort and anxiety as a result of the uncertainty (Leite & Kuiper, 2008), as well as worry, depression, distress and a feeling of being ?paralyzed? (Berenbaum et al., 2008). Bredemeier and Berenbaum (2008) found that those with a higher intolerance for uncertainty tended to concentrate more on the potential negative outcomes of options instead of the positive ones, ignoring the probabilities attached to the options. Uncertainty in illness. Mishel (1981) studied uncertainty in illness and identified contributing factors such as ambiguity of symptom patterns, unpredictability of the disease prognosis and 21 progression, insufficient information regarding the diagnosis, and the lack of clarity regarding the treatment plans (p. 259). Further research revealed additional contributing factors to this uncertainty came from events which were ?vague, unfamiliar and inconsistent? (Mishel, 1984, p. 163). While Mishel (1997a) thought that uncertainty was a ?neutral cognitive state? (p. 66), Hilton (1994) disagreed and believed that both positive and negative emotions accompanied uncertainty. Uncertainty in illness is a unique and individualized experience and the issues or questions causing it may differ depending on if one is asking a health care provider or an individual (Kasper, Geiger, Freiberger, & Schmidt, 2008; Penrod, 2007). Seven categories of patient uncertainty identified by Kasper et al. and Donovan-Kicken and Bute (2008) included uncertainty regarding ?social integration, diagnosis and prognosis, deciphering information, mastering of requirement, causal attribution, own preferred degree of involvement, physician?s trustability, and treatment? (Kasper et al., 2008, p. 46). An individual?s reaction to uncertainty often depends on the potential diagnosis, for instance, if the individual thinks it is a minor illness or cancer. Cancer survivorship is a new interest in research and one of its identified attributes is uncertainty (Doyle, 2008). Interestingly, while some patients focus on hope and move towards the goal of ending the uncertainty (Morse & Penrod, 1999), others with chronic illnesses switched from a goal of reducing uncertainty to just trying to manage it (Donovan-Kicken & Bute, 2008). Individual patients face many sources of uncertainty. Some of it comes from hours of waiting, not only in waiting rooms, but also in the time lapses between discovery, confirmation and treatments (Fogarty & Cronin, 2008). During these periods of waiting, the perceived threat to their health and life is unknown, their symptoms might 22 be unpredictable and distressing, and the timetable of when a diagnosis will be known or symptoms gone is not known (Fogarty & Cronin, 2008). Individuals experience feelings of uncertainty and powerlessness (Fogarty & Cronin, 2008). Measurement of uncertainty. In the review of literature, there were no expressive writing studies found which looked specifically at uncertainty. The majority of the expressive writing studies were conducted with cancer survivors and measured the survivors? general mood; only two looked at patients during their active treatment phase. The most common tool utilized was the Profile of Mood States (McNair, Lorr, & Droppleman, 1971). This tool is composed of 65 adjectives which describe the following domains: tension/anxiety, depression/dejection, anger/hostility, vigor/activity, fatigue/inertia and confusion/bewilderment. Using the above tool, female cancer patients reported being depressed and furious (K?llay & B?ban, 2008) and prostate patients reported little mood change over the course of the study (Rosenberg et al., 2002). This research study wanted to specifically look at uncertainty, therefore a search was made for such a tool. There were four tools or scales found in the literature which had been developed and used to measure the concept of uncertainty. Two focused specifically at uncertainty and illness, and the other two looked at uncertainty and life in general. Uncertainty Stress Scale. The Uncertainty Stress Scale (USS) was designed to measure uncertainty in illness-related situations and to specifically look at the stress, threat, and positive feelings which are created by the situation (Hilton, 1994). Hilton believed that uncertainty had an 23 emotional component and thus included questions probing for positive feeling along with perceptions of threat that accompanied the state of uncertainty. As of 1994, the USS has undergone three revisions. Since then it has been translated into several languages, and used in a variety of clinical populations with adapted versions of the USS. A sampling of groups which have used this tool includes Portugese individuals with type 2 diabetes (Ap?stolo, Viveiros, Nunes, & Domingues, 2007), high-risk pregnant women (Clauson, 1996), and adult survivors of cancer (Hilton & Skrutkowski, 2002). The USS is composed of three parts (Hilton & Skrutkowski, 2002). Part A consists of 54 items. Statements about the patient?s health condition and coping are presented and participants are asked to rate their level of uncertainty regarding these items. Each item is answered on a 5-point Likert scale. The choices range from 0 = no uncertainty, to 4 = a great deal of uncertainty. Participants are also given ?not applicable? as an answer option. Part B looks at the same 54 items, but now the participant is asked to rate the items based on how much stress is being generated by the uncertainty. Each item is answered on a 3-point Likert scale. The choices are 0 = no stress, to 2 = very high stress. Part C consists of four 10-cm long visual analog scales. These measure ?global uncertainty, global stress, global threat and positive feelings? (Hilton & Skrutkowski, 2002, p. 4). The USS has questions pertaining to the following uncertainty factors: "uncertainty/clarity/reliability/dependability, symptom uncertainty, doubt regarding present and future state of the condition, and doubts about coping and understanding" (Hilton, 1994, p. 23). 24 Reliability and validity. Scoring of the USS consists of summing Part A for an overall look at uncertainty, or summing each of the individual subscales. Part B can also be scored in that same way. Part C is scored by measuring where the individual marked on the 10 cm line. Reliability of the USS is supported by a Cronbach alpha of 0.96 (Hilton, 1994). Hilton (1994) demonstrated construct validity of this scale by testing it on a variety of patient populations and situations. Through factor analysis the original scale with eight factors was reduced to four, and testing with a new group of sample patients validated the two major theoretical themes. Multidimensional scaling confirmed both the factors and reflected similar themes. Convergent validity for the uncertainty component was tested through correlating the USS with Mishel?s Uncertainty in Illness scale (r = .69, p = .00). Convergent validity for the stress component was tested through correlating the USS with the Spielberger State Trait Anxiety Inventory. Results were r = .43 (p = .01) for total stress, and r = .63 (p = .00) for the stress visual analogue items. Hypothesis testing was conducted in several different people groups and results confirmed that there was a positive correlation between uncertainty and stress (r = .50, p = .00). The data also revealed a positive correlation between uncertainty and stress (r = .54, p = .00) and uncertainty and threat (r = .55, p = .00). Multiple regression analysis revealed that cancer recurrence, shorter time interval since treatment, lower education level and poorer state of health were all associated with greater uncertainty. Finally, the use of contrasted-groups also supported the scale. Content validity was demonstrated by having the USS reviewed for ?appropriateness and clarity? by known experts in the field including physicians, nurses, 25 cancer patients, individuals with medical disorders, psychometricians, and researchers on uncertainty (Hilton, 1994). Revisions were made per the reviewers? recommendations. Mishel Uncertainty in Illness Scale. The Mishel Uncertainty in Illness Scale (MUIS) was originally developed in 1980 to assess for uncertainty in ill, hospitalized patients (Mishel, 1981). In the last twenty years the scale has been translated into over 10 languages, used in many clinical populations, and spawned four versions. One of the versions is the Mishel Uncertainty in Illness Scale ? Community form (MUIS-C). While the MUIS was created for hospitalized patients, the MUIS-C was created to assess for the pervasiveness of uncertainty caused by illness as perceived by a non-hospitalized patient (Mishel, 1997b). The MUIS-C was developed in 1986, and revised in 1989 to its current form. A small sampling of population groups which have utilized this tool includes breast cancer survivors (Sammarco, 2001; Wonghongkul et al., 2006), patients with prostate cancer (Wallace, 2003), patients who had coronary angiography (Eastwood et al., 2008), patients undergoing coronary artery bypass surgery (Staples & Jeffrey, 1997), and patients who survived life-threatening arrhythmias (Carroll, Hamilton, & McGovern, 1999). Permission was obtained from Mishel (personal communication, January 15, 2009) to use the scale as part of this current research study. The MUIS-C consists of 23 items. Each item is answered on a 5-point Likert scale. The choices are 1 = strongly disagree, to 5 = strongly agree. Scores of the MUIS-C can range from 23-115, with higher scores indicating greater perceived uncertainty. The scale has questions pertaining to the following uncertainty factors: ambiguity, complexity, inconsistency and unpredictability (Mishel, 1997b). 26 Reliability and validity. Scores of the MUIS-C can range from 23 to 115, with 69 being the midrange score in the combined 20 studies. The reliability of the MUIS-C has been supported by Cronbach?s alphas ranging from 0.74 to 0.92 across 20 studies according to the manual, but data tables indicated two other coefficient alpha scores of 0.53 and 0.65 (Mishel, 1997b). Construct validity was demonstrated by testing the tool on a variety of patient populations and situations (Mishel, 1997b). Initial efforts to support the validity included factor analysis work which confirmed the MUIS to be multi-dimensional, while the MUIS-C form was uni-dimensional. Convergent validity was tested through correlating the MUIS with another questionnaire, and while the author reports that there was ?significant correlation,? no specifics are provided. Several other relationships have been verified through testing in various populations. These include the sensitivity of the MUIS to reflect differences between groups and show the passage of time. Theoretical antecedents of uncertainty were also underscored by responses to the MUIS, as was the connection between uncertainty and emotion-focused coping, and uncertainty and anxiety. Studies also supported the idea that psycho-social distress could be predicted by the degree of uncertainty experienced. Finally the MUIS is purported to be ?significantly related to theoretically proposed outcomes? (Mishel, 1997b, p. 43); however no data was provided by the author. Intolerance of Uncertainty Scale. The Intolerance of Uncertainty Scale was originally written in French and developed to measure an individual?s intolerance of uncertainty (Freeston, Rh?aume, 27 Letarte, Dugas, & Ladoucer, 1994). This French scale is composed of 27 items and measures five factors: ?Unacceptability and avoidance of uncertainty, Negative social evaluation caused by uncertainty, Uncertainty-related frustration, Uncertainty causes stress, and Uncertainty preventing action? (Freeston et al., 1994, p. 797). Each item is answered on a 5-point Likert scale. The choices range from 1 = not at all characteristic of me, to 5 = entirely characteristic of me. Scores on this scale can range from 27-135 with higher scores indicating greater intolerance of uncertainty. An English translation of the Intolerance of Uncertainty Scale was completed in 2002 by Buhr and Dugas (Buhr & Dugas, 2002). It was translated using the back translating method. Testing revealed four factors: ?Uncertainty leading to inability to act, Uncertainty being stressful and upsetting, Unexpected events are negative and should be avoided, and Uncertainty being unfair? (Buhr & Dugas, 2002, p. 940). This English version also consists of 27 items with each one being answered on a 5-point Likert scale. The choices range from 1 = ?not at all characteristic of me?, to 5 = ?entirely characteristic of me.? Scores on this scale can range from 27-135 with higher scores indicating greater intolerance of uncertainty. Carleton, Norton and Asmundson?s scale (2007) is a shortened version of the Intolerance of Uncertainty Scale. Through confirmatory factor analyses, the 27 item scale was reduced to 12 items. The original five factors were reduced to two: Prospective anxiety and Inhibitory anxiety (Carleton, Norton, & Asmundson, 2007, p. 113). Each of the 12 items is answered on a 5-point Likert scale. The choices range from 1 = ?not at all characteristic of me? to 5 = ?entirely characteristic of me,? Scores on this scale can range from 12-60, with higher scores indicting greater intolerance of uncertainty. 28 Reliability and validity. The original French version had very high internal consistency (a = 0.91) and a good test-retest reliability (r = .74) (Freeston et al, 1994). Convergent validity was tested and confirmed through correlations to several other well-known scales (r = 0.52-0.63, p < .0001). Buhr and Dugas? English version also had a high internal consistency (a = 0.94) and a good test-retest reliability (r = .74) (Buhr & Dugas, 2002). Convergent validity was tested and confirmed through correlations with several other well-known scales (r = 0.55- 0.60, p < .001). The shortened version of the Intolerance of Uncertainty Scale has a high internal consistence (a = .91) (Carleton et al., 2007). Cross validation was confirmed through correlating the 12 item scale with original 27 item scale (r = .96). Convergent validity was tested through correlations with scales used in the 1994 and 2002 versions and results were significant (r = 0.54 ? 0.61, p < .01). Regression techniques also confirmed that the 12 item scale measured the same domains as the 27 item scale. Quality of Life Quality of life is not a one-dimensional static concept, but rather a multi- dimensional dynamic concept. Each individual has his own unique definition of quality of life based on his values, needs and experiences. What brings meaning to one individual may not bring meaning to another. Therefore the only true way to measure someone?s quality of life is to ask him (Osoba, 1994). According to Andrews and Withey (1976), the term ?quality of life? became part of the American vocabulary sometime after World War II. They proposed that the term 29 suggested a change in thinking; that ?having it all? referred to more than just having possessions but instead included the totality of one?s life. Webster?s Online Dictionary (2009) defines quality of life as one?s ?personal satisfaction (or dissatisfaction) with the cultural or intellectual conditions under which you live (as distinct from material comfort)? (p. 1). Ferrans and Powers (1992) defined quality of life as the marriage of satisfaction with various life domains and the importance placed on those same domains. Why is quality of life important? Quality of life speaks to the essence of life, to what gives someone hope and pleasure, to what gives life meaning for that individual. General quality of life. Quality of life is a dynamic topic in psychology, medicine, nursing, and sociology. In 2009, a search of the term ?quality of life? in MEDLINE produced 107,504 hits taken from the years 1950-2009. A similar search of CINAHL, Academic Search Premier, PsycInfo, PsycARTICLES, and Master File Premier produced 115,272 hits, taken from the years 1818-2009. While combining these lists might take out some duplicates, and weeding out irrelevant articles may reduce the number still further, the number left would still be impressive. A plethora of psychology articles were located which looked at various aspects and populations with regard to their quality of life. Studies were found on college students (Xiao, Tang, & Shim, 2009; Zullig, Huebner, & Pun, 2009), and the elderly (Palacios, Torres, & Mena, 2009). Other articles described parents juggling the balance of work and home responsibilities (Jang, 2009), and made suggestions on how organizations could be more family friendly (Fatimah, Jemain, Ibrahim, Nasir, & Anuar, 2009). One article described a tool to measure life?s meaning by looking at one?s accomplishments, 30 values and principles, and how one balances having a purposeful life with a need for excitement (Morgan & Farsides, 2009). Articles in medicine and nursing focused more on health-related quality of life issues. Quality of life was an outcome watched closely in randomized controlled trials for drugs (Araujo et al., 2009), surgical treatments (M?ller-Stich et al., 2009) and medical treatments (Cohen, Firth, Biddle, & Lewis, 2009). Articles were found describing quality of life issues for patients (Carroll et al., 1999; Eastwood et al., 2008), their family members (Elliot & Berry, 2009) and sometimes comparing the two (Holm, Schonberger, Poulson, & Caetano, 2009). Quality of life has become an important outcome in health care and is part of the mission statement for Healthy People 2010 (U.S. Department of Health and Human Services, 2000). Quality of life issues such as patient-assisted suicide and when to terminate life-saving treatments are complex and controversial and continue to be debated around this country. Researchers have not reached consensus on how quality of life should be defined or measured. Ferrans and Powers (1985) identified three areas of difficulty encountered in creating a tool to measure quality of life: 1) there was no agreement on which items best describe quality of life; 2) there was no agreement on how to measure quality of life, as to whether it should be measured subjectively or objectively; and 3) a critical piece of information was lacking if the importance of the quality of life items were not assessed (p. 16). Ferrans and Powers compiled a list of factors taken from tools which have been created to measure an individual?s quality of life. These included: socioeconomic status, physical health, friendships and family, marriage, perceived stresses, satisfaction with where they live as well as the nation as a whole, satisfaction with themselves and with 31 their life, and feelings of depression and means of coping (Ferrans & Powers, 1985, p.22). Obviously there is potential for much variability in tools and their scores (Osoba, 1994). It is the researcher?s job therefore to search for or create the tool which best measures the aspect or aspects which he or she is studying. Quality of life and cancer. For much of history, medical research focused on increasing the quantity of life: discovering and inventing vaccines, anesthetics and antibiotics, finding cures for diseases - or at least treatment options which could prolong life, and perfecting surgical procedures. Now that individuals are living longer after a cancer diagnosis, it has become more important to focus on their quality of life during the treatment phase and when they earn the title of ?survivor.? Research also continues to be done on quality of life and palliative care. Over the last 40 years there has been an increasing interest in exploring individuals? perceptions of their quality of life. Articles abound in cancer research on patients? quality of life. While there are articles on quality of life as specifically related to new drugs, treatment options and surgical procedures, quality of life issues are also becoming a top priority and a more common discussion point between the oncologist and the patient. An acknowledgement of quality of life issues signals a more holistic way of caring for each patient (Padilla & Grant, 1985). Quality of life issues are similar, yet unique to each individual and do not seem to depend on the diagnosis (Visser et al., 2006). Quality of life in the newly diagnosed patient. Two studies found those individuals with the highest distress were those who were newly diagnosed (Ferrell et al., 2005; Visser et al., 2006). These individuals 32 experienced ?global distress? - significant distress across the physical, psychological and social subscales. Visser (2006) also found that neither the diagnosis nor cancer site affected the distress registered in the tool. Another study on newly-diagnosed patients reported that those who lived alone had a lower quality of life than those who lived with someone (Rustoen, Moum, Wiklund, & Hanestad, 1999). In addition, the data also suggested that older patients (age 60-78) who lived alone reported greater quality of life than younger patients (age 19-39) who lived alone. An examination of qualitative research studies gave a unique and perhaps more comprehensive view of patients? experiences (Beatty et al., 2008; Boehmke & Dickerson, 2006; Ferrell, Smith, Cullinane & Melancon, 2003a; Hj?rleifsd?ttir et al., 2008; Hochwald, 2008). While the majority of these studies were on breast cancer, the quality of life issues faced are similar for most cancer diagnosis. Patients battled physical effects from the cancer and the treatments including nausea and appetite changes, fatigue, numbness and tingling, pain, menopausal symptoms, specific cancer symptoms, constipation and diarrhea, loss of hair, sleep problems, and skin changes. Psychological issues included loss of control, changes in memory or ?chemo brain,? changes in body- image and self-concept, fertility issues, attempts to cope, intense emotions of fear, anguish/distress, shock, disbelief, anger, self-blame, uncertainty and isolation. Social issues included changes in social roles and support, and managing expectations and emotions of family and friends. Family issues included changes in home roles and dealing with expectations and emotions of spouse/significant other and children. Spiritual issues raised included a need to pray and to be involved in their religion. Financial issues 33 were also raised, from fear of losing a job and insurance to fear of not being able to afford treatments (Bradley et al., 2007). Quality of life in the cancer survivor. Survivors faced their own issues. Some issues were left over from the treatment phase; others developed once they were determined ?cancer free? and a survivor. Survivors were often torn between wanting to celebrate beating the cancer and living in fear of recurrence (Ferrell et al., 2003b). They celebrated the return to ?normal? life and activities and the return of their hair. At the same time though, they also expressed distress, anxiety, vulnerability and the feeling of ?living on borrowed time? (Ferrell et al., 2003b, p. 1065). Four qualitative studies revealed intimate details about quality of life in the survivor phase (Ferrell, Grant, Funk, Otis-Green, & Garcia, 1997, 1998; Ferrell et al., 2003b; Wonghongkul et al., 2006). Physically, survivors faced weight gain and fatigue, menopause (for women), permanent side effects of the cancer or treatments, and many sources and types of pain. They also spoke of the relief of getting their strength back after treatments were done. Psychologically, survivors continued to struggle with distress over their initial diagnosis, surgery and treatments, fears of cancer recurrence or a new cancer, financial issues, anxiety and depression, uncertainties, and attempts to balance the fear of death with reality and hope. On the positive side, survivors also spoke of regaining control, positive coping and a focus on living, improvements in their moods, hope, and a return to normalcy. They learned the importance of expressing their emotions and of humor. Socially, the main concerns were fears that their relatives would get cancer, distress over the way their illness had affected their family, and distress over changes in 34 the area of their sexuality. Positive social issues included the importance of social and family support, and returning to work and their other roles. Spiritually, survivors spoke of being plagued by uncertainty. Not all survivors became spiritual or religious, yet many spoke of a deepening of their faith and a return to their religious upbringing. They spoke of hope, reorganized priorities, a new appreciation of life, and a new meaning to life. Several studies looked for connections between demographic data and quality of life. A recent study of ovarian cancer survivors found that higher quality of life was associated with being over the age of 60 and still working, living with someone, having a household income of $50,000 or more, and having no sign of recurrent cancer (Ferrell et al., 2005). Sammarco (2001), in a study of younger breast cancer survivors, found strong positive correlations between perceived social support and reported quality of life. Another study of breast cancer survivors found that uncertainty, perception of harm and years of survival influenced the quality of life (Wonghongkul et al., 2006). Quality of life in the dying patient. One of the recommendations put forth by the Institute of Medicine (IOM) in its 1997 report on end of life care was to ?strengthen methods for measuring the quality of life and other outcomes of care for dying patients and those close to them,? and to ?develop better tools and strategies for improving the quality of care? (Field & Cassel, 1997, p. 267). This report highlighted the inadequacies in end of life care and resources, and the lack of health care provider education and research in this area. Quality of life issues for dying patients are in some sense more simplified and yet just as important (McCahill, Ferrell, & Virani, 2001). Physically, they need relief from pain and other sequela of their disease processes. They need help managing their fatigue 35 and weakness, difficulties breathing and eating, sexuality problems, and skin and bowel problems. They need knowledgeable care providers who can give comprehensive and compassionate care up and through the time of death. Psychologically, they need comfort and some sense of control over their treatments. They need care which is culturally sensitive and honesty regarding treatment options and prognosis. Socially, they may have issues with roles again, need help dealing with emotional distress from their family and friends or feelings of isolation, and have financial worries. They may need assistance to do things which they enjoy. Spiritually, some patients struggle, others give up and get angry at God, and finally there are those who accept their prognosis and are at peace with it (Ferrell et al., 2003b). In response to the 1997 IOM report, City of Hope investigators in collaboration with the American Association of College of Nursing created a program entitled The End of Life Nursing Education Consortium (ELNEC) in 2000. The goal of this program was to ?enhance understanding? among nurses of end of life issues and how to give quality and compassionate care to the dying (American Association of Colleges of Nursing, 2009). A year earlier, a similar program was initiated for physicians and healthcare professionals called The End of Life Physician Education project (EPEC) (The EPEC Project, 2009). As the movement grows to create a better death, hospice organizations are becoming more common and accepted. Some are found in hospitals, others are free- standing. In Auburn, Alabama, there is one free-standing hospice associated with East Alabama Medical Center called Bethany House. There are also several local home health businesses which also serve hospice patients. 36 Quality of life for the caregivers. Spouses, significant others, and family members are not immune to effects of the cancer illness. While they may not be experiencing things themselves, they do experience them through watching and living with their loved one. Northouse et al. (2007) found that a spouse?s ?risk for distress? was the same as that of the patients. Spouses also reported ?significantly less self-efficacy and social support compared with patients in managing the effects of prostate cancer across all phases of illness? (p. 4176). Based on this data, a suggestion was made to include spouses in the patient?s program of care and assist them to be more open in communicating how they were feeling. Caregivers can experience some of the same quality of life issues as the patient (McCahill et al., 2001; Ferrell, Ervin, Smith, Marek, & Melancon, 2002). Physically, they too battle fatigue and have changes in sleep patterns and appetite. Psychologically, they feel uncertain, anxious, depressed, fearful, distracted, helpless and distressed. They too long to be in control and have their ?normal? life back. They may have difficulty coping. Socially, they may feel isolated and tired from caring for their loved one. Some have financial and work problems caused by the illness. Roles change and families often go through a period of distress. They are in need of a support system. Spiritually, they may be struggling between their feelings of uncertainty and their hope and faith in God. Resources for caregivers can be found through the local American Cancer Society and often also through many religious organizations. Quality of life and uncertainty. Literature has indicated that there is an inverse relationship between uncertainty and quality of life. Studying these two concepts in the area of cardiology, and using 37 variations of the same tools being used in the current research study on expressive writing, data suggested that the higher the uncertainty, the greater impact there was on quality of life, even up to six months after a life-threatening arrhythmia (r = -0.61, p = .001) (Carroll et al., 1999) and up to one year after a coronary angiography (adjusted r? = 0.54, p <.001) (Eastwood et al., 2008). In a study conducted on patients and their spouses before coronary artery bypass surgery, uncertainty was found to be inversely related to both hope and quality of life, with spouses reporting more uncertainty about the illness than the patient (patients r = -0.30, not statistically significant at the 95% level; spouses r = -0.49, p < .01) (Staples & Jeffery, 1997). Cancer nursing has also looked at these two concepts together. Two studies were located that used the scales which will also be used in this research project; one study was on patients with prostate cancer (Wallace, 2003), and the other was on young breast cancer survivors (Sammarco, 2001). Data indicated that for the breast cancer patients, quality of life was positively correlated with perceived social support (r = .418, p = .000) and negatively correlated with uncertainty (r = -.436, p = .000) (Sammarco, 2001). The data also indicated a negative correlation between social support and uncertainty (r = - .288, p = .002). Men with prostate cancer who were undergoing watchful waiting experienced uncertainty which was positively correlated with the health functioning domain of quality of life (r = 0.601, p = .008) (Wallace, 2003). Measuring quality of life. A search of the literature revealed more than twenty quality of life tools. They ranged from one question to over 130 questions. Some were to be completed by the physician or health care provider, others were to be answered by the individual. Some 38 focused only on those items which fell under a nurses? control, others were more broad and asked about anything which might affect quality of life. Grant, Padilla, Ferrell and Rhiner (1990) suggested several reasons for the importance of assessing quality of life in cancer patients. First, tools can record patient reactions to their cancer. Secondly, tools can record patient responses to symptom interventions. Thirdly, responses on tools can be used to compare different treatment options. Finally, tools can record the effects of different rehabilitation options (p. 262). By measuring a patients? quality of life and basing treatment plans on the results, health professionals are caring for that patient in a more holistic manner. Spitzer Quality of Life Index. The Spitzer Quality of Life Index (QL-Index) was developed in 1981 (Spitzer et al., 1981). It was created to be simple, comprehensive and applicable to numerous people groups and settings. While originally designed to be completed by a health care professional, the index is now self administered by the patient. This index has been used to measure quality of life in cancer patients (Corn et al., 2008; Perez, McGee, Campbell, Christensen, & Williams, 1997), those with lower back pain (Beaulieu, Wood- Dauphinee, Abenhaim, & Arahmowicz, 1997; Toledo, Alexandre, & Rodrigues, 2008), stem-cell recipients (Loberiza et al., 2002), and others. The QL-Index is composed of 5 items which measure an individual?s social support, psychological status and activity level. Each item is rated on a 3-point Likert scale, where 0 = a more negative response. Scores of the index can range from 0-10, with lower scores indicating less quality of life. 39 Reliability and validity. The reliability of the index was supported by Cronbach?s a = .775 (Spitzer et al., 1981). Inter-rater comparison was also performed which supported the reliability of the index (Spearman rank 0.81, p < .001; 0.84, p < .001; and 0.74, p < .005). Construct validity was demonstrated by testing the index on a variety of patient populations and situations (Spitzer et al., 1981). Initial efforts to support validity included factor analysis work which confirmed the QL-Index to be multi-dimensional. Convergent validity was tested through correlating the QL-Index with other questionnaires and in five different population groups. Data correlations were consistently significant in three groups: chronically diseased (r = 0.42-0.60, p < .001), cancer patients (r = 0.46-0.72, p < .001), and seriously ill (r = 0.34-0.53, p < .005). Discriminant construct validity was verified through a gradient of diminishing mean quality of life scores. Scores ranged from high scores for healthy people, to a mix of scores for those who had diseases or cancer, to low scores for those who were seriously ill. City of Hope Quality of Life- Cancer Patient Version. Ferrell developed the City of Hope Quality of Life- Cancer Patient Version (QOL-CS) in 1995 (Ferrell, Hassey-Dow, & Grant, 1995). It is described as multi- dimensional scale which measures psychological, physical, social, and spiritual well- being in a cancer patient. The tool, based on previous versions created by researchers at the City of Hope National Medical Center, grew out of research on pain and was revised to its current form (Ferrell et al., 1989; Padilla & Grant, 1985; Padilla et al., 1983). It has been adapted for long-term cancer survivors and translated into Spanish. A small sampling of studies which have utilized this tool includes studies on cancer survivors 40 (Ferrell et al., 2005; Ferrell et al., 1997,1998; Wonghongkul et al., 2006), and cancer patients (Juarez, Ferrell, & Borneman, 1998). The QOL-CS consists of 41 items set on an ordinal scale with 0 = worst outcome and 10 = best outcome. Scores can range from 0- 410, with higher scores indicating better quality of life. Reliability and validity. The reliability of the QOL-CS is supported by Cronbach?s alpha of .93 (Ferrell & Grant, n.d.). Cronbach?s alpha scores for the subscales were spiritual (a = .71), physical (a = .77), social (a = .81) and psychological (a = .89). Test-retest reliability had an overall score of .89, with subscales scores of spiritual (r = .90), physical (r = .88), social (r = .81), and psychological (r = .88). Construct validity of this scale was demonstrated by having it reviewed by a panel of nurses and physicians who were experts in the cancer field (Ferrell & Grant, n.d.). Step-wide multiple regressions were used for factor analysis and identified 17 factors which were statistically significant. Validity was also tested and confirmed through correlating the QOL-CS with another scale, showing an overall correlation of r = .78. Subscale correlations ranged from r = 0.44 ? 0.74. Tests on the QOL-SC have also revealed a sensitivity to discriminate between known groups of cancer survivors (Ferrell & Grant, n.d.). Quality of Life Index ? Cancer Version III. Ferrans and Powers (1985) created a Quality of Life Index (QLI) to measure an individual?s satisfaction with various domains in life, and the importance of those same domains to that individual. While originally designed for healthy adults, variations of the scale have since been created to look at specific disorders and illnesses. All variations 41 were created from the same original scale but with items specific for the disorder or illness. The Quality of Life Index ? Cancer Version (QLI-C) was developed in 1990 (Ferrans, 1990). In the past twenty years the QLI and QLI-C have been revised twice and now are in their third versions and currently identical. A recent sampling of studies which have utilized the scale with cancer populations includes studies on breast cancer survivors, both younger and older (Sammarco, 2001, 2003), women undergoing a mastectomy (Xiaokun, 2002), patients with prostate cancer (Schneider, Hsieh, Sprod, Carter, & Hayward, 2007), patients undergoing stem cell transplantation (Hacker et al., 2006) and patients with thyroid cancer (Huang, Lee, Chien, Liu, & Tai, 2004). Permission was obtained from Ferrans to use this scale for research purposes (A. All, personal communication, May 1996). The QLI-C consists of a total of 66 items on two forms (Ferrans & Powers, 1998a). The first form consisting of 33 items measures how satisfied an individual is with various domains of their life, and the second form, again 33 items, measures the importance of those same domains to the individual. Each item is answered on a 6-point Likert scale. On the first form the choices range from 1 = very dissatisfied, to 6 = very satisfied. On the second form the choices range from 1 = very unimportant, to 6 = very important. Scores can total 33 to 198 per form. The scores of satisfaction are weighted by the scores of importance giving a range of scores from 0-30. The highest scores indicate high satisfaction with those things of high importance, and thus greater quality of life. The lowest scores are those of low satisfaction with things of high importance, and point to lower quality of life. The scales can be broken down into subscales which look 42 specifically at health and functioning, social and economic, psychological/spiritual, and family or significant others. Reliability and validity. Currently the QLI and the QLI-C scales are identical; thus reliability and validity can be drawn from either tool being used with the cancer population. The reliability of the QLI tool is high, with Cronbach?s alphas ranging from 0.87 to 0.97 across nine studies (Ferrans & Powers, 1998b). Internal consistency reliability for the QLI has been demonstrated in eight studies through measurement of the four subscales (Ferrans & Powers, 1998b). Alphas ranged from 0.77 - 0.92 for health and functioning, 0.71 - 0.88 for the social and economic subscale, and 0.77-0.96 for the psychological/spiritual subscale. Alphas for family subscale were more in the mid range, 0.66 - 0.83, with one outlier of 0.18. Content validity was demonstrated through extensive literature reviews and reports from patients regarding their quality of life (Ferrans & Powers, 1985). The QLI was also tested with the Content Validity Index and received an ?acceptably high rating,? though no scores were given (Ferrans & Powers, 1998b). Construct validity was tested through correlation of QLI with another questionnaire and the r scores ranged from 0.61 to 0.93. Construct validity was also verified through testing of the subscales in a contrasting-groups approach. The QLI scores correlated with each specific group. Factor analysis was performed which supported the four subscales or factors and the overarching factor of quality of life. 43 Expressive Writing/Reflective Journaling Writing has been used for centuries to express emotions. In the Bible, David wrote songs expressing a range of emotions from praise, to fear, anger, and remorse. There are many songs (also known as psalms) which start out with David ranting at God or his enemies, and end with him being comforted and acknowledging the sovereignty of God. It was through the penning of thoughts, the acknowledgement of emotions and the wrestling through them, that comfort came. Diaries are a private means of expressing thoughts and emotions. Consider The Diary of Ann Frank (Frank, 1989) or other famous memoirs written by those who have gone through horrific experiences, whether war, death, imprisonment, cancer, or a multitude of other painful experiences. Memoirs often chronicle not only particular events, but also the emotions which accompanied and helped define the events and the individual. Some diaries are written to be read by others. Some diaries are written and locked away for the writer?s eyes only. Letter writing can also be a method of communicating with oneself. The discipline of psychology has used letter writing to complement one-on-one therapy with individuals needing medical or surgical interventions and who are dealing with emotions of anger and grief (Rancour & Brauer, 2003). Psychologists have found letter writing to be a healing strategy, allowing session work to continue after the individual leaves the office. A more recent public version of a journal or letter is called a blog. A blog is a type of online journal which is typically available for anyone in the world to read. Blogs can be found on any subject, whether it be cooking, politics, caring for plants or 44 parenting. What was almost unheard of 10 years ago has now become mainstream with blogging conferences and a proliferation of blogging websites. Blogs can be set up to only allow the creator to write, or the set up can allow comments to be added by readers. In scholarly literature, writing can be found in two main areas: academia and psychology/medicine. In both settings the purpose of writing is to explore events and emotions and a means with which to chart change. In academia the writing process is referred to ?reflective journaling,? in psychology/medicine the process is most often referred to as ?expressive writing.? Reflective journaling. There were many examples found in the literature of academia utilizing reflective journaling in classes ranging from psychology to computer science. For example, West Point used journaling in its General Psychology for Leaders class to help the students learn and retain their course concepts (Hampton & Morrow, 2003). Management undergraduate teams used journaling when conducting project management research projects to help foster honest self appraisals, identify areas that the student needed improvement in, and ultimately improved both individual and team performance (Loo, 2002). Choral students used open or group journaling to help them with the transition from student to teacher (Williams & Watkins, n.d.). Nursing students have used journaling to track the development of their critical thinking skills, as well as the progress made in their comfort and skill level over the course of a semester (Hayes, 2005; Morgan, Johnson, & Garrison, 2005; Schaefer & Curley, 2005). Reflective journaling has been found to enhance nursing students? critical thinking skills (Hubbs & Brand, 2005; Morgan et al., 2005). Journaling helped students 45 ?bridge the gap between theory and practice? (Morgan et al., 2005, p. 110). Journaling did this by providing students an opportunity to reflect on the way they practiced nursing, allowed them to reflect on events, and make connections between what they already knew and the new knowledge they were learning (Morgan et al., 2005). Hampton and Morrow (2003) and Dunlap (2006) noted that by keeping a journal, students were allowed the opportunity to reflect back over what they had learned, and thus see how much they had grown and developed. If the journal was started at the beginning of their academic time, then the student was given a wonderful overview of how far they had come both in knowledge as well as competence. Thorpe and Loo (1999) found a link between journaling and empowerment through what the student had learned, and Loo (2002) later did another study on journaling that documented recurrent themes including ?personal criticisms and improvement? and ?lessons learned and future actions to improve one?s performance? (p. 62). Journaling also allowed a student to voice questions or doubts in a safe environment, reflect on an event, and praise themselves on an accomplishment. Through feedback, the instructor could assess the level of comprehension and mastery of the class content and goals (Hubbs & Brand, 2005), identify problems and suggest recommendations to solve them, (Loo, 2002), and give the student advice garnered through living and working outside of the classroom (Cyboran, 2005). Also, through the giving of feedback, a mentoring relationship has the potential to develop between the student and instructor (Cyboran, 2005). 46 General expressive writing. In the psychology/medical literature writing was most often referred to as ?expressive writing? but also could be found under the title of ?emotional disclosure.? Pennebaker and Beall (1986) are generally credited with igniting the interest on studying the connections between writing and its outcomes. Pennebaker had conducted prior studies (Pennebaker & O?Heeron, 1984; Pennebaker & Hoover, 1986) searching for a relationship between the inhibition of emotions and health outcomes. In one study spouses who had lost their spouse to suicide or an accidental death were more likely to become sick in the year following the death if they did not tell others about their experience (Pennebaker & O?Heeron 1984). In a second study, Pennebaker and Hoover (1986) found that college students who had experienced a childhood trauma and had not told anyone about it were more likely to experience health problems as compared to those who had told others about the trauma. The results of these two studies led to the hypothesis that disclosing traumatic events - and the emotions which accompany those events - through writing has health benefits. The writing experiment was conducted on college students who were randomly put into one of four groups: control group (no writing); trauma-fact group; trauma-emotions group, and trauma-combination group. Blood pressure readings and mood assessments taken immediately after completing the writing were negatively impacted in the trauma-combination and trauma-emotions group. However, data taken over the next two years indicated that the students who wrote about traumatic events and their emotions were healthier than were those who wrote about trivial things (control group and trauma-fact) (Pennebaker & Beall, 1986). 47 The results were intriguing enough to encourage Pennebaker to continue to explore the relationship between writing and health outcomes. College employees who wrote about traumatic events showed a significant decrease (p <0.05) in their liver enzymes, as well as a positive change in their uric acid, cholesterol numbers, and two enzymes linked to immune function (Francis & Pennebaker, 1992). The study also revealed that writing benefits were greater for those with low emotional inhibition than those who had high emotional inhibition. A subsequent study noted that newly fired professionals who did emotional writing were reemployed more quickly than those who did not write or who wrote without emotion (Spera, Buhrfeind, & Pennebaker, 1994). In 1998 Smyth did a meta-analysis of expressive writing studies to look for overall significance and effect size (Smyth, 1998). Various outcome measures were also examined for effect size and specific moderators were examined for their positive or negative effect on writing outcomes. Inclusion criteria for this meta analysis were as follows: each study had to use some variation of the Pennebaker prompt, have a control and experimental writing group, have outcomes which fell under five health types, contain statistical information needed to calculate an effect size, and be a randomized experiment. After applying the inclusion criteria, 13 studies out of 19 remained in the analysis. All 13 studies had been conducted on healthy adults or college students. After combining the results, the five outcomes types reported in these studies were: ?reported health, psychological well-being, physiological functioning, general functioning and health behaviors? (Smyth, 1998, p. 177). Moderators in these studies included the characteristics of the participants, what they were asked to write, the writing session number, length and duration, the outcome type, and type of publication in which the 48 study was located. Smyth reported overall significant health benefits to those who participated in the studies. The overall effect size was d = 0.47, (r = .23) significant at the p < .000 level. All health outcomes were statistically significant for those who wrote, except for health behaviors. Expressive writing in clinical groups. While studies continue to be done with healthy volunteers, other researchers have experimented with writing in clinical diagnostic groups and looked for similar health benefits. One such study examined the effects on patients with asthma or rheumatoid arthritis (Smyth et al., 1999). This study followed patients over a four month period, and at the end of that time both experimental groups showed a statistically significant reduction in their symptoms and disease process compared to those in the control groups. Those with asthma who were in the experimental group showed improvements in lung function (measured by the mean percentage of predicted forced expiratory volume in 1 second), from a baseline of 63.9% to 76.3% after four months (p < .001). The rheumatoid arthritis experimental group measured symptoms on a scale where 0 = asymptomatic to 4 = severe and improved in overall disease activity from a baseline mean of 1.65 to a four month mean of 1.19 (p = .001). There was no improvement in either of the control groups. Two of the earliest studies of breast cancer patients found conflicting results. While Walker, Nail and Croyle (1999) did not find any significant effect of writing, Stanton et al. (2002) reported that those women who participated in the expressive writing group complained of fewer physical symptoms and had fewer medical appointments for cancer-related morbidities at the end of three months. It should be mentioned however that the studies were set up differently, including the amount and 49 timing of the writing, and having different writing prompts. A study of writing with prostate cancer patients found improvements in physical symptoms and health care utilization, but no change in psychological variables or immune factors (Rosenberg et al., 2002). As a follow up to the Smyth analysis, Frisina, Borod and Lepore (2004) analyzed nine expressive writing studies done with diagnostic clinical populations to see if the same health effects would be found. Smyth?s methodology was duplicated in the analysis, however these nine studies only assessed for physical and psychological outcomes, therefore no comparison could be made to the other three outcome types mentioned in the Smyth analysis. The results of the 2004 analysis revealed significant changes in physical health outcomes (d = .21, p < .01), but not in psychological health outcomes (d = .07, p > .10). While the overall effect size (d = 0.19, r = .101, p < .05) was less than what Smyth had found (d = 0.47, r = .23, p < .000), the data did suggest physical health improvements in both chronic and potentially fatal diseases. And while no significant effect was seen in the overall psychological domain, the subscales of depression, mood, anxiety and sleep quality did show improvement. The most recent meta-analysis of expressive writing was published in 2006 and was the largest to date with almost 11,000 participants (Frattaroli, 2006). It included 146 published and unpublished studies, combining studies of healthy adults with those of chronically-ill and potentially fatally-ill populations. Statistical computations for each study included an overall effect size in addition to an effect size for each of the six outcome types: ?psychological health, physiological functioning, reported health, health behaviors, subjective impact of the intervention, and general functioning/life outcome? 50 (Frattaroli, 2006, p. 841). The overall effect size (d = .151, r = .075, p < .000) was statistically significant. Effect sizes of the health outcome types were statistical significance for psychological health (r = .056, p < .000), physiological health (r = .060, p < .01), reported health (r = .072, p < .000), general functioning/life outcomes (r = .046, p < .01), and subjective impact of intervention (r = .159, p < .000). Congruent with the Smyth analysis, the health outcome type of health behaviors did not show changes which were statistically significant (r = .007, p > .44). A more recent study conducted by Gellaitry, Peters, Bloomfield and Horne (2009) looked at the effect of expressive writing on perceptions of emotional support in women with breast cancer. After six months, the data suggested that women who wrote expressively perceived their actual emotional support to be closer to their desired ideal support. The discrepancy between the actual and ideal emotional support decreased from 0.91 at baseline to 0.70 at the end of six months. This was compared to the control group which revealed a widening gap between their ideal and actual emotional support. The discrepancy between actual and ideal emotional support climbed from 1.27 at baseline to 1.53 at six months. Additional insights were found in participants? comments. Some participants noted that writing gave them a venue to say things which they were afraid to say to anyone else for fear of upsetting the listener. Others were even afraid to admit things to themselves, but through writing they were able to be honest with themselves about their reality and work through the experiences in their journal. Methods of expressive writing. Researchers continue to search for the perfect combination of moderating variables which will produce optimum health outcomes for writing participants. 51 Differences in moderators can make it difficult to compare studies head-to-head. Below is a sampling of just a few moderators and the variations which have been experimented with in different studies. There have been variations in the participants of the study. Initial studies were conducted on healthy volunteers such as college students (Pennebaker & Beall, 1986) university employees (Francis & Pennebaker, 1992), professionals who had just lost their jobs (Spera et al., 1994) and those employed (Barclay & Skarlicki, 2009). Studies have been conducted on those with psychiatric health problems such as inmates in a psychiatric prison (Richards, Beal, Seagal, & Pennebaker, 2000), college students with suicidal behaviors (Kovac & Range, 2002) and those with posttraumatic stress disorder (Gidron, Peri, Connolly, & Shalev, 1996). Studies have also been conducted on those with chronic health problems such as asthma and rheumatoid arthritis (Smyth et al., 1999), as well as more acute and potentially fatal health problems such as breast cancer (Gellaitry et al., 2009; K?llay & B?ban, 2008), prostate cancer (Rosenberg et al., 2002; Zakowski, Ramati, Morton, Johnson, & Flanigan, 2004), and metastatic renal cell carcinoma (de Moor et al., 2002). According to Frattaroli (2006), those studies which had participants with physical health problems had larger health effects sizes (health criteria, r =.131; no health criteria, r = .054). There have been variations in the location in which the writing takes place. Studies have been conducted in a lab or college classroom (Pennebaker & Beall, 1986; Burton & King, 2008), in the participant?s homes (Schoutrop, Lange, Hanewald, Davidovich, & Salomon, 2002), in a cancer clinic (Morgan et al., 2008) and in a location of the participant?s choosing (Laccetti, 2007). Frattaroli (2006) analysis suggested that a 52 higher psychological health effect size was produced when participants wrote at home (home, r = .122; controlled setting, r = .034) and writing privately affected the overall effect size (private room, r = .085; public room, r = .034). There have been variations on the writing instructions given to the experimental groups. Pennebaker?s original study placed participants into one of four groups (Pennebaker & Beall, 1986). The control group wrote on assigned trivial topics, one experimental group wrote on the facts of a trauma, another experimental group wrote on just the emotions evoked by a trauma, and the third group wrote about the whole experience. Burton and King (2008) randomly assigned participants to write about a control topic, a trauma, or a positive experience. Another study asked participants to write about their ?worst? or ?happiest? experience (Lyubomirsky et al., 2006). Sometimes participants have been asked to write on a specific event such as an injustice done to them by a manager (Barclay & Skarlicki, 2009), but in most studies participants were given some latitude to their topic as long as it stayed within the set guidelines. Giving participants a direct question to answer made a difference in the overall effect size (directed questions, r = .090; no directed questions, r = .052), and in the psychological health effect size (directed questions, r = .094; no directed questions, r = .011) (Frattaroli, 2006). The amount of time spent writing is another variable. Perhaps the least amount of writing was requested by Burton and King (2008), who designed their experiment to see if writing for just two minutes a day for two consecutive days could make a difference in the participant?s health. Examples of more moderate requests include one study which asked participants to write for 15 minutes a day for three consecutive days (Lyubomirsky 53 et al., 2006), and another which asked for 20 minutes once a week for four consecutive weeks (Francis & Pennebaker, 1992). On the longer side of writing is a study in which participants were asked to write for 45 minutes five different times over a two week period (Schoutrop et al., 2002) and a study which asked their participants to write three times a week for 12 weeks (Smith et al., 2005). Data from the Frattaroli analysis (2006) revealed that larger effect sizes were found when participants had to write at least 15 minutes (less than 15 minutes, r = -.007; at least 15 minutes, r = .080). Frattaroli (2006) examined a comprehensive list of moderating variables and summarized the findings on the moderators as such: ?The successful study tended to use participants with a health problem or with a history of trauma, to make sure participants were very comfortable during disclosure (e.g. by allowing them to disclose at home), to pay participants, to administer a large dose of disclosure (e.g. by requiring at least three disclosure sessions), to have participants disclose events that had yet to be fully processed (e.g. more recent events), to provide very detailed and specific disclosure instructions (e.g. directed questions), and to have relatively short followup periods (e.g. less than one month)? (p. 860). Consequently, these findings impacted and guided the design of the current research study on expressive writing. Theoretical Framework This research study was guided by the empowerment theory. Most definitions of empowerment are similar to that of Guti?rrez (1994): ?a process of increasing personal, interpersonal or political power so that individuals, families, and communities can take action to improve their situations? (p. 202). However, Rappaport (1984) asserted that empowerment cannot or should not be defined, as empowerment changes to fit each 54 unique person and context. This theory was chosen for several reasons. These reasons included the belief that by participating in expressive writing, individuals were doing something proactively which could improve their health. In addition, participants potentially could make a positive difference in their psychological state without needing to turn to medication. Finally, by exploring their emotions and all that is being impacted by their cancer, they might be able to identify that which is important to them and brings them satisfactions, and be empowered to work with intention towards living their own unique fulfilled life. This theory, first written about by Freire (1970) when describing oppressed peoples in Brazil, was later used to describe movements of social and political reform (Napier, 2006). Empowerment is often associated with progress being made in improving the status of women and minorities (Busch & Valentine, 2000; Hutchinson & Wexler, 2007; Lisovicz et al., 2006). Lee (1996) presented some basic assumptions of empowerment. One assumption is that if empowerment is needed, there must be a situation with an absence of power or control. The ripple-effects of oppression are far-reaching, extending to individuals as well as to communities. Secondly, oppressed individuals must be viewed as being capable of change, of becoming empowered. They have within themselves the strength and ability to identify problems, come up with solutions, and join with other like-minded individuals to force action. By participating in this process, individuals empower themselves. Thirdly, empowerment takes action. One is not empowered through inaction; rather empowerment comes by standing up, taking personal responsibility and working hard with others towards a common goal (p. 229). 55 Swift and Levin (1987) compared the concept of empowerment to that of prevention. Viewed in that context, the description of empowerment is compelling. The word empowerment suggests that something positive is developing; it focuses primarily on the process not the outcome; it is a dynamic state; it is connected with a holistic world view; it requires individuals to participate; it necessitates change and with change comes more control for the individual; and empowered individuals create goals and are future minded (Swift & Levin, 1987, p. 89). According to Lee (2001), there are ?three interlocking dimensions of empowerment: 1) the development of a more positive and potent sense of self, 2) the construction of knowledge and capacity for a more critical comprehension of the web of social and political realities of one?s environment, and 3) the cultivation of resources and strategies, or more functional competence, for attainment of personal and collective goals? (p. 34.). From this description, empowerment can be divided into three interrelated levels. The first level was referred to by several terms: psychological, micro or intrapersonal. This level represented the empowerment of the individual. Increasing self efficacy, learning skills, and developing a consciousness of his or her history and place in society (both present and future) was an important part of this level (Guti?rrez, 1994). Knowledge of how the individual had adapted to oppression and what coping mechanisms had been utilized in the past was also important (Lee, 2001). The second level was the interpersonal or mezzo level. Here the individual joined with their family, special interest group or community to plan and take action towards a common goal. Participants empowered themselves as they connected with their group and experienced ?collective action? (Lee, 2001). 56 The highest level was the political level. Communities and groups, united and empowered through their common goals and ability to cause action, became involved in the political arena in order to raise awareness about their specific oppression issue or situation, and work to alleviate the problem. In order to be effective politically, groups had to develop the knowledge, skills and network affiliations necessary to make their voices heard and spur on political action (Lee, 2001). Empowerment and Health Promotion More recently empowerment has been used as a foundation for health promotion (Lisovicz et al., 2006; Rissel, 1994; Wallerstein, 1992). The World Health Organization (1986) defined health promotion as ?the process of enabling people to increase control over, and to improve, their health? (p. 1). While this definition describes a potential outcome of the current research project on expressive writing, it is the definition proposed by Page and Czuba (1999) which most closely matches the elements of this research project: ?Empowerment is a multi-dimensional social process that helps people gain control over their own lives? (para. 11). Since Page and Czuba?s quote acts as a guiding light to this study on expressive writing, it is important to examine the three descriptors within the quote and how they apply to this study. ?Multi-dimensional?: this speaks to the many dimensions which impact each person, as well as the different levels in which they interact (Page & Czuba, 1999, para. 12). In the same way, the diagnosis of cancer impacts many - if not all - aspects and dimensions of a person?s life. Multi-dimensional can also refer to the care which is given a patient. Treatments, tests, and medications all focus on the physical, whereas listening and writing focus more on the emotional and psychological, thus giving 57 the patient more holistic care. ?Social?: empowerment is seen as occurring in and through relationships with others (Page & Czuba, para. 12; also Lee, 1996). While expressive writing in this study may primarily be done in private, the emotions worked through on paper can impact real people and relationships. ?Process?: empowerment is seen as a journey, as well as an outcome (Page & Czuba, para. 12; also Rissel 1994; Swift & Levin, 1987). It can also been described as an intervention (Guti?rrez, DeLois, & GlenMaye, 1995). Likewise, receiving a diagnosis of cancer starts a person on a journey - a journey not of their choosing, but one which must be traveled none the less. Writing also takes a person on a journey; one through emotions to a place of better understanding and positive functioning (Frattaoli, 2006). In order to stay true to the definition of empowerment, one acting as a ?health promoter? must not be a director, but rather more of a facilitator (Rissel, 1994). Lee (1996) agreed that the empowerment process originates from within the individual, not through anyone on the outside. In this research study, the investigator will be facilitating the writing and providing a broad writing prompt but offering no other directions. In the discipline of social work, social workers have four empowering roles, those of enabling, catalyzing, priming and linking (Lum, 1996, p. 253). Of these four, enabling is most applicable to this research study. An enabler provides information, knowledge, or contacts which the recipient uses to gain more control over their life. Likewise, in this research study the investigator will be introducing participants to writing, which according to the literature, can help them influence their own overall health functioning (Frattaroli, 2006). 58 Literature has suggested some specific ways in which writing has been empowering. It gave patients something to do which could have a positive impact on their health ? physical outcomes as well as psychological health (Frattaroli, 2006). Writing was found to decrease the need for pain medications in prostate cancer patients (Rosenberg et al., 2002), decrease physical symptoms and associated doctor visits in breast cancer patients (Stanton et al., 2002), and help patients with renal cell cancer have a better night?s sleep (de Moor et al., 2002). Hypothetically writing might also provide a patient with the opportunity to explore their goals and dreams and assess their priorities. With this new self awareness, they will be empowered to make informed decisions when given treatment options which impact their priorities and future (Swift & Levin, 1987). Knowing what is important to them and making choices based on that knowledge increases their feelings of power and control over their lives (Zimmerman, 1995). Summary In this chapter, the investigator introduced the population and characteristics which define those newly-diagnosed with cancer. The concepts of uncertainty and specifically uncertainty in illness were also explored. The concept of quality of life was examined, both in general and as it is measured in clinical populations. Finally, the history and benefits of expressive writing were identified. The results of this study will contribute to the expanding knowledge of the benefits of expressive writing in a general cancer population, as well as writings? specific influence on perceptions of uncertainty and quality of life. 59 CHAPTER III METHODOLOGY This chapter describes and explains the research methodology used during this study of expressive writing and cancer patients. Methods for this study included the: (1) research approach and design, (2) study questions, (3) a description of the setting, population, and sample, (4) details about the tools used to measure the variables, (5) ethical considerations, (6) steps and procedures followed to conduct the research, and (7) data analysis plan. The chapter concludes with the expected outcomes of this study based on the review of literature. Research Design and Approach The causal comparative research design was utilized in this study. This design?s purpose was to explain phenomena through an investigation of cause and effect relationships (Gall et al., 2007). The use of a causal comparative design enabled the researcher to examine the cause and effect relationship between expressive writing and feelings of uncertainty and perceived quality of life. This design enabled the researcher to look for changes in the outcome variables by comparing the pre-writing or baseline measurements with post-writing outcome measurements. The approach to this study was completion of tools prior to beginning to journal, followed by completion of tools after journaling for three months. Participants were randomly put into two groups: the control and the intervention group. Pretest measurements for each group included demographic data and completion of the MUIS-C 60 and QLI-C tools. These were collected when the participant entered the study. At three months, these same tools were sent to each participant for completion, in addition to an expressive writing information sheet. The data received from these tools was analyzed to determine if changes in perceptions of uncertainty and quality of life had occurred individually and as a collective group over the course of the study. Study Questions The emotions experienced by cancer patients, feelings of uncertainty in illness, perceptions of quality of life, and the act of expressive writing were all explored in the review of literature. However, there was a gap in the literature with regard to exploring the effects of expressive writing on feelings of uncertainty and perceived quality of life in a general cancer population. Therefore this study explored these relationships through the following questions: 1. Does expressive writing cause a difference in cancer patients? perceived level of uncertainty after writing for three months? 2. Does expressive writing cause a difference in cancer patients? perception of their quality of life after writing for three months? 3. Does expressive writing affect the relationship between quality of life and uncertainty in the intervention group as compared to the control group? Setting, Population, and Sample To be successful and produce valid results, careful selection of the setting, population and sample is imperitive. For this study, two local cancer treatment facilities were chosen as the setting. (A third local facility was approached but declined to participate.) Conducting research in these settings allowed for easier access to the 61 population of interest: those newly diagnosed with cancer. The research sample was chosen through a volunteer, convenience method. Setting The settings chosen for this research project were ones in which the target population was most easily accessible, and where inclusion criteria could be met. Due to time constraints and the desire to have a sample size of at least 60, two local cancer treatment facilities were approached as settings for the research study: the Cancer Center at East Alabama Medical Center and the Montgomery Cancer Center. East Alabama Medical Center (EAMC) is a public, non-profit 352-bed regional referral center located in Opelika, Alabama. It has served the local community for over fifty years. It is serviced by over 140 physicians, many of whom also see patients in the cancer center. The Cancer Center was opened to the public in 1992 and earned the designation of Community Cancer Center by the American College of Surgeons Commission on Cancer in 1995. Due to its proximity to EAMC, the Cancer Center offers the services of surgeons and regular medical physicians, as well as medical and radiation oncologists. Opened in 1990, Montgomery Cancer Center (MCC) is a state-of-the-art freestanding facility located in Montgomery, Alabama. It offers advanced diagnostic services, advanced radiation therapy and chemotherapy for cancer, as well as treatments for diseases of the blood. It also hosts the Montgomery Breast Center, a diagnosis and treatment facility for breast cancer. The center is staffed by 10 medical and radiation oncologists, as well as 7 nurse practitioners. The center also staffs three satellite facilities 62 in neighboring cities, offering patients the convenience of seeing an oncologist closer to home. Population At the time this study was conducted, it was estimated that within the year approximately 1.5 million adults living in America would be diagnosed with cancer (American Cancer Society, 2009) and with the diagnosis would experience ?global? distress (Visser, 2006). Of those, nearly 24,000 would be from Alabama (American Cancer Society, 2009). A portion of those from Alabama would be treated at the two local cancer centers involved in this study. MCC generally sees approximately 100 newly-diagnosed cancer patients a month (L. Hamilton, personal communication, March 30, 2010). The majority of patients seen in the infusion room come in with a diagnosis of breast, lung or colon cancer. This corresponds to data reported by the American Cancer Society (2009) which shows the most prevalent cancers in Alabama are female breast, colon/rectal, lung/bronchus and prostate. EAMC?s cancer patient numbers ranged from 40-70 per month, with the lower numbers in November and December due to the holidays (C. Tate, personal communication, April 1, 2010). EAMC too reported the majority of their chemotherapy patients having a cancer diagnosis involving the breast, lung, colon, or head/neck. Sample The sample consisted of participants selected through the method of convenience sampling. From October through December 2009. all newly-diagnosed patients at EAMC and MCC were given a flyer about the project by their physician or nurse practitioner on the days the investigator was at the facility. Patients then had the option of approaching 63 the investigator in the waiting room to get more information. All participants who met the inclusion criteria were invited to join the study. Inclusion criteria. Several criteria were identified as necessary in order for patients to participate and provide meaningful data for this study (Gall et al., 2007). These inclusion criteria included: 1. Participants must be over 19 years old (per Alabama law), 2. Participants could not be pregnant or prisoners, 3. Participants must be within the first three weeks of chemotherapy or radiation treatment, 4. Participants must be able to complete a written or typed journal. Exclusion criteria. Several criteria were also identified which would exclude patients from participating in this study. These exclusion criteria included: 1. Patients who did not meet the inclusion criteria, 2. Patients who cognitively could not give their consent, and 3. Patients who could not speak or read English and had no access to a translator. Data Collection Tools After the review of literature, the following tools were rejected for use in this study. Hilton?s Uncertainty Stress Scale was a valid scale to use with this research project; however few studies were found which had used it, despite the scale being available for over 15 years. Therefore, it was not chosen. The shortened version of the Intolerance of Uncertainty scale has only been tested to detect anxiety disorders in 64 college students, not in any clinical populations. Therefore, this scale was not chosen. The Spitzer Quality of Life Index was not chosen because this investigator believed it was not as comprehensive a measure as others. The City of Hope Quality of Life-Cancer Patient Version has been widely used and has established reliability and validity. However, this investigator believed that the importance of certain aspects of life should also be taken into consideration in a quality of life scale, therefore this scale was not chosen. Four tools were utilized in this study, all self-administered. These tools included a demographic information sheet, the expressive writing information sheet, the Mishel Uncertainty in Illness Scale ? Community form (MUIS-C) (Mishel, 1989), and the Quality of Life Index ? Cancer Version III (Ferrans & Powers, 1998a). The use of self- administered tools as a data collection technique offered the following benefits: patients could pick a convenient time to fill out the tools, choose to fill them out all at once or over several sittings, and determine for themselves what order to fill out the questions (Gall et al., 2007). Self-administered tools are also cheaper to administer and do not require a time commitment from the participants to meet with the interviewer (Gall et al., 2007). Demographic Information Sheet Demographic data is commonly collected in order to identify trends. The Demographic Information sheet was created by the investigator and requested the following data: gender, age, marital status, race, highest level of education (number of years of education), primary employment status (full-time, part-time, not employed), cancer type, first cancer diagnosis, and cancer diagnosis as first major illness. 65 Expressive Writing Information Sheet The Expressive Writing Information sheet was created by the investigator to ask how often the participant did expressive writing. This researcher-designed information sheet was included with the three-month follow-up packet. The tool comprised the following three questions: ?Did you do any expressive writing? How many times a week? Did it help you feel differently?? The participants were also given the opportunity to share any additional comments under the ?Comments? section at the end of the sheet. Mishel Uncertainty in Illness Scale The Mishel Uncertainty in Illness Scale ? Community form is reported to measure the areas of ambiguity, complexity, inconsistency and unpredictability in a non- hospitalized ill individual or their family member (Mishel, 1997b). This scale looks specifically at individuals? levels of uncertainty as they progress through their illness. This scale had been used with various patient populations including those with cancer, multiple sclerosis, rheumatoid arthritis and others. Due to its specific measuring of uncertainty in an ill population, this scale was chosen to be used in this research study. Quality of Life Index Ferrans and Powers (1985) believed the best way to measure an individual?s perceived quality of life was by considering an individual?s satisfaction with certain variables coupled with how important those same variables were to that individual. Ferrans and Powers (1992) developed the Quality of Life Index ? Cancer Version III with the goal of helping physicians and other health care providers identify problem areas (areas of low quality of life) and then plan interventions for those identified areas. This multi-dimensional tool looks at the subscales of family, psychological/spiritual, social 66 and economic, and health and functioning. It appeared to measure an individual?s quality of life most comprehensively; therefore it was chosen to be used in this research study. Ethical Considerations Participants in research studies should be able to participate without fear of physical or psychological harm (Gall et al., 2007). Careful consideration was taken to protect the anonymity and confidentiality of each participant. Prior to meeting with the patients, packets of information were assembled. Included in these packets were two consent forms (one for the patient to keep and one for the investigator), a personal address sheet, the demographic information sheet and the QLI-C and the MUIS-C data tools. On the packet itself and on each of the sheets within ? except the consents - was written a number. This number became the identifier of the participant in the study. As individuals agreed to participate, they were assigned the next number in sequence. For instance, the first individual who agreed to participate was given the packet with the number ?one? on it. The second individual who agreed to participate was given the packet with the number ?two? on it, and so on. Those with an odd number (one, three, five, seven?) were assigned to the intervention group. Those with an even numbe r (two, four, six, eight?) were assigned to the control group. The consent forms, personal address sheets, completed data tools, coding sheet and demographic information sheets were kept in a locked home filing cabinet by the investigator. The investigator kept a calendar recording when each participant entered the study, and when their three-month marks would be. At the three-month mark, the investigator sent out the data tools and expressive writing information sheet to the participant. The completed data tools and information sheet were returned in 67 preaddressed and stamped envelopes to the investigator at a post office (P.O.) box address. The investigator next entered the data into the SPSS software program. An outside statistician was employed to review the study?s design, data collection methods, perform statistical calculations and help with the interpretation of the results. The statistician only saw the identifier numbers, thus protecting the anonymity of the participants. There were no anticipated physical risks for the participants; however there were some psychological risks including becoming emotional (crying) or very upset by stirred memories and associated emotions (Pennebaker, 2003). These were explained in the initial meeting and included on the consent form. Participants were warned of these potential risks and instructed that if the negative emotions lasted more than four weeks or got progressively worse, it was recommended they seek professional help. To address this risk, participants were given a list of local counselors. Potential benefits included an improved perception of quality of life and decreased feelings of uncertainty. Data Collection Steps and Procedures Review of other studies utilizing both tools (Sammarco, 2001; Staples & Jeffrey, 1997; Wallace, 2003) assisted in guiding the procedures used in this study. Careful documentation was kept in order to allow others to critique, repeat or improve upon this study. Step One Step one was approval of the research project by the thesis committee. Approval was then requested of and granted by the Institutional Review Boards of EAMC 68 (Appendix A), MCC (Appendix B), and Auburn University (Appendix C). The investigator also met with the oncologist or nurse practitioner in charge of each cancer clinic to provide an overview of the project and proposed benefits for their patients. These oncologists and nurse practitioners also gave verbal approval for the study to be conducted in their facility. Step Two The original plan was for the investigator to visit each site once a week and sit in a specific spot in the waiting room along with the patients. Oncologists or their staff would give each new patient an informational flier about the study (Appendix D) during their office visit and the patient was then given the choice to seek out the investigator on their way out or just leave. If patients were interested in being a part of the study, they could come to the investigator for more information. Once recruitment actually began however, it became apparent that the plan would have to change and would be different at each facility. At the first facility, patients were approached by their physician and were given a study flier. If the patient was interested, the physician notified the investigator and the investigator called the patient to set up a meeting at the facility during a regularly scheduled infusion. Care was taken to remain within the initial three week window of the inclusion criteria. At the meeting, the investigator described the study in more detail. At the second facility, the patient educator gave each new patient a study flier in their information packet. On the days the investigator was at the facility, the investigator asked each infusion nurse if they had a new patient that day or one which would fall within the three week window criteria. If there were new patients, the investigator met 69 with them and inquired of their interest in participating in the study. If they were interested, the investigator then described the study in more detail. Each interested patient was given an explanation of the research project by the investigator as per the recruitment script (Appendix E): what would be expected of them, what data would be collected and how it would be used, and what risks and benefits would be faced by participating (Gall et al., 2007). Risks for participating included potentially experiencing some psychological distress. Potential benefits included an improved perception of quality of life and decreased feelings of uncertainty, among other benefits. After being given the opportunity to ask questions, each patient could consent or not consent. Those who wanted to participate and met the inclusion criteria signed a written consent form with the stated understanding that they could leave the study at any time without fearing any negative consequences. Those at EAMC signed the EAMC consent form (Appendix F) and those at MCC signed the MCC consent (Appendix G). Patients were given a copy of their consent form. The consent forms were inside a numbered folder which also contained a personal address sheet, the demographic sheet, and the two data tools, each numbered with the same number as was on the folder, except the consent forms which had no numbers so as to protect participants? anonymity. The number on the folder became the participant?s personal identifier number. Odd numbers were assigned to the intervention group; even numbers were assigned to the control group. Step Three Once consent had been given the participant filled out the personal address sheet (Appendix H). Then the investigator read the demographic tool (Appendix I), the MUIS- 70 C (Appendix J) and the QLI-C (Appendix K) to each participant. This enabled the investigator to explain how the data tools were to be filled out and to ensure no data points were left blank. The investigator then recorded the participant?s answers on the data tool forms. For those in the intervention group (odd numbers), a verbal explanation of expressive writing was given and questions were answered at this time. Participants in this group were given a Pennebaker prompt to help start their writing: Over the next three months I would like for you to write about your very deepest thoughts and feelings about an extremely important emotional issue that has affected you and your life. In your writing, I?d like you to really let go and explore your very deepest emotions and thoughts. You might tie your topic to your relationships with others, including parents, lovers, friends or relatives; to your past, your present, or your future; or to who you have been, who you would like to be, or who you are now. You may write about the same general issues or experiences on all days of writing or on different topics each day. All of your writing will be completely confidential. Don?t worry about spelling, sentence structure or grammar (Pennebaker, 2003, p. 362). Participants were assured they could write about anything, negative or positive, as long as they included honest thoughts and emotions. They could write in a journal or type on a computer or typewriter. Participants were asked to write a minimum of twice a week, with no specific amount of time set per episode. Each participant was given a written instruction sheet which covered the same information (Appendix L) as well as a list of local counselors for them to call if they became psychologically distressed. Those 71 at EAMC received a list of Auburn facilities (Appendix M) and those at MCC received a list of Montgomery facilities (Appendix N). They were also offered a journal in which to write. These participants were told that the journals were theirs to keep and that the journals would not be requested by the investigator at the end of the study. While the investigator would never read what had been written, the participant could choose to allow others to read the journals or could keep their writings private. Participants in the intervention group were informed that the MUIS-C and the QLI-C tools plus an expressive writing information sheet (Appendix O) would be sent to them after three months for them to self-administer and return to the investigator at a P.O. Box address in pre-addressed and stamped envelopes. Participants were given two weeks to fill out and return the tools. If the tools were not received within two weeks post- mailing, a second set of surveys was sent. Participants in the control group (even numbers) were told they were in the non- journaling group, but the investigator could not ask them to abstain from any type of journaling for the next three months while participating in the study. They too were told that at three months they would be sent the MUIS-C and QLI-C tools and the expressive writing information sheet for them to self-administer and return to the investigator at a P.O. Box address in pre-addressed and stamped envelopes. Participants were given two weeks to fill out and return the tools. If the tools were not received within two weeks post-mailing, a second set of surveys was sent. The control group participants were also offered a journal, but theirs would be sent to them once the investigator had received their three-month surveys. If they accepted the journal offer, they picked one out of the assortment provided by the investigator and it was put aside for them. 72 Data Analysis Plan All data analysis for this causal-comparative study was done using SPSS version 17.0 software. All tests had a significance level set at p < 0.05. Descriptive statistics including group means and standard deviations were performed first on all tools and variables. The following are the research study questions and the statistical tests which were used to measure the data. 1. Does expressive writing cause a difference in cancer patients? perceived level of uncertainty after writing for three months? There are certain statistical tests which are commonly used with a causal- comparative research design, one of which is the repeated measure analysis of variance (ANOVA) (Gall et al., 2007). ANOVA is used to measure and compare the amount of variance within a group and between groups by looking at their means. If the variance proves statistically significant, then there is more variance between groups than there is within each group. With repeated measures ANOVA, the test is repeated on the same group of subjects more than once, often as a pre and post test. (The original plan was to use the ANOVA but due to the small sample size the statistical test had to be changed to a paired t-test.) Scores from the MUIS-C scale (Mishel, 1997b) were utilized to answer this question. The investigator tested the intervention group and control group and compared the means of their levels of uncertainty at baseline and three months later. 2. Does expressive writing cause a difference in cancer patients? perception of their quality of life after writing for three months? Originally, the data from a repeated measures ANOVA was also supposed to answer this question, however again due to the small sample size the statistical test had to 73 be changed to a paired t-test. Scores from Ferrans and Power?s QLI-C tool (Ferrans & Powers, 1998a) were measured in the intervention and control group, and the means of their perceptions of quality of life were compared at baseline and three months later. 3. Does expressive writing affect the relationship between quality of life and uncertainty in the intervention group as compared to the control group? Relationships require statistical tests which look at correlations, therefore Pearson?s r was chosen to answer this question. It measured the strength of the relationships between each of the two tools and the intervention and control groups. Expected Results Literature reported that many who have participated in expressive writing have experienced decreased levels of uncertainty. Expressive writing has also been found to increase many writing participant?s perceptions of their quality of life. Accordingly, this study expected data to show decreasing levels of uncertainty and increasing levels of quality of life in those of the intervention group who participated in the writing exercise. If any positive changes were noted in the control group, they would not be as significant as those of the intervention group. 74 CHAPTER IV ANALYSIS OF DATA This chapter will give an overview of the process and procedures utilized in conducting this research study. The study sample will be described based on the data collected on the demographic form. Each research study question, the statistical procedures utilized to measure that question, and the results from the data collected will be presented. Data Collection As described earlier, data was collected at two different facilities and the data collection procedure differed slightly between the facilities. At both facilities though, once consent was given, the patient filled out the personal address form and the demographic sheet and the investigator administered the MUIS-C and QLI-C tools. The patient was then randomized into either the intervention or control group based on their folder number. The participants who were in the intervention group were given the writing instructions verbally as well as on a written sheet to take home, and were given a list of local counselors to call if they experienced prolonged psychological distress. They were also invited to choose a journal provided by the investigator in which to write. Both groups were told they would receive another set of tools at three months to fill out and return. The participants in the control group were promised a journal once their three- month forms were returned to the investigator. 75 At the end of three months, each participant was sent the MUIS-C and QLI-C tools as well as the expressive writing information sheet and a pre-addressed and stamped envelope. The forms were returned to the investigator at a local P.O. Box address. Seven participants returned their forms on time. The other seven did not respond within two and a half weeks and were sent a second complete set of forms. Two participants responded from that second group. Those who responded from the control group were sent their journal as promised. Once the forms were returned, the investigator entered all of the data into an SPSS version 17.0 software program. The data was then sent to a statistician to run the required statistical analysis tests with the significance level set at p < .05. Of the nine returned three-month survey forms, six were from the intervention group and three were from the control group. Due to the small total number and control group response, the repeated measures ANOVA tests originally planned were replaced by paired t-tests. Description of Sample Fourteen patients - one male and thirteen females - made up the study sample and were recruited during the time period of October through December, 2009. Nine were being treated at the Cancer Center at East Alabama Medical Center and five at the Montgomery Cancer Center. The age range of the total sample was 35-88, with 55.9 being the mean age (see Table 1). The majority of the participants were married (n=9) and Caucasian (n=12). The level of education ranged from 9 years to 24 years, with a mean of 15.4 years. More than half were currently working (n=9). For all but two this was their first major illness. Nearly half of the participants had a diagnosis of breast cancer (n=6), with the rest evenly split between colon/rectal and head/neck (under 76 ?other?). Finally, only one fifth of the sample (n=3) stated they normally did some type of journaling. The intervention group consisted of six females and the only male, all Caucasian. The majority were married (n=5) and employed full time (n=6), and all had at least some college education. Six of the seven returned their three-month survey forms (see Table 1). The control group consisted of seven females, five Caucasians and two African Americans. The majority were married (n=4) and unemployed (n=4), and most had at least some college education (n=5). Three of the seven returned their three-month survey forms (see Table 1). TABLE 1 Comparison of Intervention and Control Group Demographics Demographic Intervention Group Control Group Age 31-45 years 2 2 46-60 years 2 1 61-75 years 2 4 76-90 years 1 0 Gender Male 1 0 Female 6 7 Marital Status Married 5 4 Divorced 1 1 77 Widowed 1 1 Separated 0 1 Race Caucasian 7 5 African American 0 2 Level of Education 7-12 years 0 2 13-16 years 5 3 17+ years 2 2 Employment Status Full time 6 2 Part time 0 1 Not employed 1 4 First Major Illness Yes 6 6 No 1 1 Cancer Type Breast 2 4 Colon / rectal 4 0 Other 1 3 Normally Journal Yes 1 2 No 6 5 78 Reliability of Instruments Cronbach?s alphas are calculated to estimate the internal consistency of a tool (Polit & Beck, 2008). The higher the reliability, the higher the Cronbach?s alpha. Accordingly, Cronbach?s alphas for both tools were examined at baseline and three months. In previous studies, Cronbach?s alphas for Mishel?s Uncertainty in Illness scale ? Community form ranged from 0.74 to 0.92 (Mishel, 1997b). For this study, at baseline, the MUIS-C tool?s Cronbach?s alpha was comparable and acceptable at 0.86, but it decreased to 0.64 at three months. This decrease could be attributed to the smaller three- month sample size. Ferrans and Powers reported Cronbach?s alphas from 0.87 to 0.97 for the Quality of Life Index ? Cancer Version III (Ferrrans & Powers, 1998b). For this study, the QLI-C Cronbach?s alphas remained high throughout, with 0.93 at baseline and reducing very slightly to 0.91 at three months. Both of these alphas were comparable to those previously reported. Results The results are organized by responses to each study question. Statistical results are presented for each question as well as tables illustrating the findings. Response to Research Study Questions Research question one. Research question one: Does expressive writing cause a difference in cancer patients? perceived level of uncertainty after writing for three months? Every participant in the study filled out Mishel?s Uncertainty in Illness scale ? Community form. This scale assessed the respondent?s feelings of uncertainty as it pertained to their illness, which in this case was their cancer diagnosis, by exploring four characteristics of uncertainty: 79 ?inconsistency, unpredictability, ambiguity, and complexity? (Mishel, 1997b, p. 6). The scale consists of 23 questions answered with a 5-point Likert-like scale, where 1 = strongly disagree and 5 = strongly agree. The higher the score, the more uncertainty the person is feeling. Scores can range from 23-115, with mean scores from previous studies ranging from 42.4 - 85.5 (Mishel, 1997b). To assess for any changes in their level of uncertainty and to answer this study question, a comparison was made between the baseline and three-month data of the intervention group. Setting the significance value at p < .05, a paired t-test was performed. The results were t(5) = -0.583, p = .56. While there was a three point increase in the mean of the intervention group from baseline to three months (see Table 2), due to the small sample size no significant statistical difference was found in the perceived level of uncertainty after expressively writing for three months. For comparison purposes, data from the total group and control group were also included in the table. 80 Table 2 Group Total, Mean and Standard Deviation for Uncertainty in Illness Scale Groups n M S.D. Intervention Group Baseline 6 46.8 11.8 Three Months 6 49.8 5.5 Control Group Baseline 3 54.3 5.1 Three Months 3 55.3 6.5 Total Group Baseline 9 49.3 10.3 Three Months 9 51.7 6.1 Research question two. Research question two: Does expressive writing cause a difference in cancer patients? perception of their quality of life after writing for three months? Every participant in the study filled out Ferrans and Power?s Quality of Life Index ? Cancer Version III. This index is composed of 66 questions looking at the domains of ?health and functioning, psychological/spiritual, socioeconomic, and family? (Ferrans & Powers, 1998a). The first 33 questions ask how satisfied the individual is with an item in their life, and the second 33 questions ask how important that same item is to them. A 6-point Likert-type scale is used in this survey, with 1 = very dissatisfied or very unimportant, and 6 = very satisfied or very important. The answers are weighted so that the scale 81 ranges from 0 - 30, with higher numbers indicating higher quality of life. Those items which have high satisfaction and importance give a higher score indicative of greater quality of life, and those items which have high importance but low satisfaction get a lower score indicating a lower quality of life. To answer this study question, a comparison was made between the intervention group?s data collected at baseline and three months (see Table 3). The table also includes the data from the total group and control group for comparison. Setting the significance value at p < .05, a paired t-test was performed on the data. The results were t(5) = 1.01, p = .36. Therefore, after writing expressively for three months, no statistically significant difference was found between the two sets of quality of life data for either the total index or the subscales. Table 3 Group Total, Mean and Standard Deviation for Quality of Life Index Groups n M S.D. Intervention Group Baseline 6 24.6 2.9 Three Months 6 23.6 2.8 Control Group Baseline 3 21.4 1.3 Three Months 3 20.4 3.9 Total Group Baseline 9 23.6 2.9 Three Months 9 22.5 3.4 82 Research question three. Research question three: Does expressive writing affect the relationship between quality of life and uncertainty in the intervention group as compared to the control group? A Pearson?s r correlation was used to test for relationships between the variables and they were not statistically significant. Therefore, the data did not support the question. However, when asked if expressive writing had helped them ?feel differently,? five out of the six intervention group members who finished the study answered ?yes.? One participant clarified the ?yes? by writing, ?I feel that it always helps to express your inner feelings by putting it down on paper. It has helped me with my stress of being sick and the hope of a better future!? Another commented, ?The more chemo I had and the worse the side effects, the less I wrote. But I did not give up my spirit.? Summary Fourteen participants consented to be in the study and were randomly placed into either the intervention group or the control group. However, only nine returned the three- month data tools and completed the study. Of those who completed the study, six were in the intervention group and three were in the control group. Due to the small sample and even smaller response size, a change was made in the statistical tests done to analyze the data. In order to run a repeated measures ANOVA, a group needs at least 30 people. Therefore a paired t-test was substituted for the ANOVA and was used to answer the first and second research questions. The uncertainty t-test was not statistically significant, though examining the means of the intervention and control groups at baseline and three months revealed an increase in both groups at three months over baseline. This suggests an increase in all respondent?s levels of uncertainty during the course of the study. The 83 quality of life t-test was also not statistically significant. Examination of the control and intervention group means at baseline and three months showed a slight decrease in both group means at three months compared to baseline. This indicates a slight decrease in perceived quality of life for each group. Pearson?s r was also not statistically significant, showing no correlation between expressive writing, uncertainty and perceived quality of life. While the quantitative data was insignificant, the qualitative data was interesting and will be discussed further in the next chapter. 84 CHAPTER V DISCUSSION, IMPLICATIONS, SUMMARY The time period following a cancer diagnosis and start of treatment is fraught with uncertainty and a decrease in perceived quality of life. Expressive writing has the potential to improve cancer patients? mood, anxiety and depression (Frisina, 2004). This study looked at a general cancer population through the method of convenience sampling to see if expressive writing would be beneficial to them ? changing their perception of their quality of life and amount of uncertainty. Despite non-significant statistical findings, the majority of the intervention group stated that they felt differently after writing. This last chapter explored the qualitative data obtained from the study, as well as the encountered limitations, implications for further research and practice, and a summary of the study. Discussion A non-significant t-test or Pearson?s r indicates that the research questions are not supported by the data. In this study there were several factors which could have affected or caused this outcome including the small sample size and resulting low power of the study, high attrition rate which led to an even smaller response size, the length of the study time being too long (contributing to attrition rate), or the study time being too short to capture a true change in uncertainty or perceived quality of life caused by expressive writing. Or perhaps no matter what, expressive writing would not have made a statistically significant difference in this samples? feelings of uncertainty and perceived 85 quality of life. Despite the lack of statistical significance however, some interesting qualitative information about this sample was gleaned after a closer inspection of the data. Mishel?s Uncertainty in Life Scale ? Community form All fourteen participants completed a MUIS-C tool at the beginning of their time in the study. The reported mean at baseline was 48.6 (S.D. = 11.6) where the possible scores could range from 23-115. This original sample appears to be moderately uncertain. The study mean falls between two means reported by Mishel (1997b) also from studies with mixed type cancers: M = 44.6 (S.D. 10.9) and M = 85.5 (S.D. = 6.9). Their sample numbers were larger however with N=55 and N=70 respectively. Based on the current study?s baseline data, the items which received the highest scores for uncertainty were: ?The course of my illness keeps changing; I have good and bad days? (M = 3.7, S.D. = 1.3); ?I am certain they will not find anything else wrong with me? (M = 3.4, S.D. = 1.9); and ?It is unclear how bad my pain will be? (M = 3.0, S.D. = 1.2). These statements echo statements made by newly diagnosed breast cancer patients (Beatty et al., 2008; Boehmke & Dickerson, 2006; Hochwald, 2008). Paring the group total down to only those who returned the three-month forms ( n= 9), uncertainty due to the changing illness course remained the number one cause of uncertainty in both groups at baseline and three months (see Tables 4 and 5). Another issue producing high uncertainty in both groups was how treatments affected what they could or could not do. 86 Table 4 Comparison of Item Means of Highest Uncertainty and Lowest Uncertainty in Intervention Group at Baseline and Three Months Baseline Three Months Greatest Uncertainty Greatest Uncertainty Course of illness keeps changing (3.5) Course of illness keeps changing (4.0) Certain to find nothing else wrong (3.2) Hard to know if treatments are helping (2.8) Unclear how bad pain will get (3.2) Unsure if illness is better or worse (2.7) Treatment effectiveness undetermined (2.7) Treatment affects what I can do/not do (2.7) Least Uncertainty Least Uncertainty Don?t know what?s wrong with me (1.2) Don?t know what?s wrong with me (1.2) Explanations are hazy (1.5) Purpose of treatment is unclear (1.7) Purpose of treatment is unclear (1.5) Drs? explanations have many meanings (1.7) Drs? explanations have many meanings (1.5) Differing opinions on what is wrong (1.7) Treatments too complex to understand (1.5) No specific diagnosis (1.7) Differing opinions on what is wrong (1.5) Drs don?t use everyday language (1.7) 87 Test results inconsistent (1.5) Treatments don?t have known probability of success (1.5) No specific diagnosis (1.5) Table 5 Comparison of Item Means of Highest Uncertainty and Lowest Uncertainty in Control Group at Baseline and Three Months Baseline Three Months Greatest Uncertainty Greatest Uncertainty Course of illness keeps changing (4.0) Course of illness keeps changing (4.0) Unpredictability of illness affects ability to plan for future (3.7) Treatments affect what I can do/not do (4.0) Treatment affect what I can do/not do (3.3) Symptoms change unpredictably (3.3) Least Uncertainty Least Uncertainty Treatments don?t have known probability of success (1.3) Don?t know what?s wrong with me (1.3) No specific diagnosis (1.3) Unsure if illness is better or worse (1.3) Seriousness of illness is undetermined (1.3) Unpredictability of illness affects ability to plan for future (1.7) Drs don?t use everyday language (1.3) Treatments don?t have known probability of success (1.7) Seriousness of illness is undetermined (1.7) 88 In looking at the above statements, some reflect the normal unpredictable side of the disease and treatment process. For instance, unfortunately, it is normal for a course of illness to change; it is normal to wonder how bad the pain will get ? whether caused by the disease or side effects; it is normal for treatments to affect - sometimes dictate - ones daily life, and to wonder if treatments are working and one is getting better. Even though these are normal concerns, they can and should be addressed by health care providers. Patients may not necessarily expect or receive an answer, but they need someone to ask how they are feeling, acknowledge the uncertainties and address them as valid concerns (Boehmke & Dickerson, 2006; Vitek, Rosenzweig, & Stollings, 2007). Examining the amount of change in responses on the uncertainty tool from baseline to three months also revealed some interesting details. The intervention group (n = 6) had two questions whose scores showed a significant downward change from baseline to three months. The change indicated a decrease in uncertainty about anticipated intensity of pain (U4) and fear of additional health problems (U19). The decrease in uncertainty in pain may be explained by the passage of time: patients are usually past needed surgeries and they have had enough treatments that they know what to expect for side effects. The uncertainty regarding the chance of something else being found wrong is also reported among breast cancer survivors (Beatty et al., 2008). Three questions? scores that increased the most at three months indicated a greater uncertainty about illness progression (U3), effectiveness of treatments (U11), and potential success of treatments (U20) (see Figure 1). These uncertainties were supported by statements from breast cancer survivors who told of fears of reoccurrence, yet were also challenged by the 89 same survivors who also spoke of the importance of staying positive and seeing cancer as a temporary problem (Beatty et al., 2008), This positive viewpoint was echoed by other cancer patients still in the middle of their treatments who spoke of their hopes and belief that the treatments would work (Hj?rleifsd?ttir & Gunnarsd?ttir, 2008). Figure 1. Comparison of Mean Uncertainty Tool Question Scores from Baseline to Three Months for Intervention Group There were three questions for the control group (n = 3) which were significantly higher at baseline than three months. This change indicated a decrease in uncertainty about illness prognosis (U3), anticipated intensity of pain (U4), and inability to plan for the future (U12). By three months into treatments, patients have gotten into a routine with their treatments and are learning to live their life around them (Hj?rleifsd?ttir & Gunnarsd?ttir, 2008). Three other question?s scores were much higher at three months. These indicated an increase in uncertainty related to symptom unpredictability (U7), and 90 not understanding explanations, apparently due to doctors and nurses lack of using everyday language in their explanations (U8 and U23) (see Figure 2). One explanation for the symptom unpredictability could be the compounding intensity of side effects from chemotherapy which often worsens over time (Hochwald, 2008). The inability to understand explanations is troubling however, since open communication between a patient and their health care provider can minimize uncertainties and provide hope (Hj?rleifsd?ttir & Gunnarsd?ttir, 2008). Figure 2. Comparison of Mean Uncertainty Tool Question Scores from Baseline to Three Months for Control Group Ferrans and Powers Quality of Life Index ? Cancer Version III All fourteen participants filled out the QLI-C tool when they entered the study. The group mean at baseline was 22.5 (S.D. = 3.8) which is considered moderately high (range was 0-30). This was comparable to another study with newly diagnosed cancer 91 patients in Norway which reported a mean score of 21.5 (S.D. = 3.8) (Rustoen et al., 1999), and a sample of American breast cancer survivors (M = 21.96, S.D. = 4.5) (Sammarco, 2001). A closer look at the individual questions revealed that the top three items the fourteen participants were most satisfied with were their health care, emotional support from their family, and their faith in God (all tied with M = 5.9). The top three items they felt were most important were their family?s health, children, and their family?s happiness (all tied with M = 6.0). Faith in God was a close fourth (M = 5.9). Closer examination of the intervention group?s data revealed that the family subscale had the highest mean at both baseline (M = 27.6, S.D. = 4.8) and three months (M = 26.7, S.D. = 2.8). Within the family subscale, the items related to highest quality of life were family?s happiness, children, emotional support from family, and family?s health. The health and functioning subscale had the lowest mean at both baseline (M = 21.8, S.D. = 3.2) and three months (M = 19.7, S.D. = 4.2). Health, energy level, and worries were the items which received the lowest scores. These ratings were not surprising as most participants were in the middle of their chemotherapy treatments at their three month survey time. Other studies reported similar mean ratings with newly diagnosed cancer patients (Rustoen et al., 1999) and young breast cancer survivors (Sammarco, 2001). Across the board, the intervention group had higher means than did the control group, though there are many factors besides expressive writing which could have influenced the difference in the scores. While the intervention means were higher, it is interesting to note that all but the psychological/spiritual subscale mean decreased at three months (see Table 6). In contrast, the control group?s psychological/spiritual and 92 family subscales were both increased at three months. A decrease in the quality of life scores was similar to two other studies which did not use the Ferrans and Powers Quality of Life Index but did measure quality of life with a different tool over a period of time. Both studies looked at the effects of expressive writing with breast cancer patients and reported an initial decrease in quality of life followed later by an increase which surpassed the baseline mean (Gellaitry et al., 2009; Stanton et al., 2002). Table 6 Comparison of Subscale Means of Intervention Group and Control Group at Baseline and Three Months Subscales Baseline Three Months Intervention Group QLI - total 24.6 (S.D. = 2.9) 23.6 (S.D. = 2.8) Health/functioning 21.9 (S.D. = 3.2) 19.7 (S.D. = 4.2) Social/economic 25.9 (S.D. = 2.9) 25.5 (S.D. = 3.7) Psychological/spiritual 26.5 (S.D. = 2.9) 26.7 (S.D. = 3.3) Family 27.6 (S.D. = 4.8) 26.7 (S.D. = 2.8) Control Group QLI - total 21.4 (S.D. = 1.3) 20.4 (S.D. = 3.9) Health/functioning 19.5 (S.D.= 3.3) 16.7 (S.D. = 5.3) Social/economic 23.4 (S.D. = 0.9) 21.9 (S.D. = 5.2) Psychological/spiritual 21.4 (S.D. = 2.3) 23.3 (S.D. = 1.0) Family 24.1 (S.D. = 2.7) 26.0 (S.D. = 3.2) 93 Examining individual question weighted ratings revealed some items in common between the two groups as well as others unique to each group. All participants rated their faith in God as the number one item at both measurement times. This might have to do with the location of the study ? deep south in the Bible belt ? since a Norwegian study of a similar population reported faith in God as the least important item (Rustoen et al., 1999). Other items which received high ratings included their health care and emotional support from family (see Tables 7 and 8), which were both mentioned as important in other studies (Beatty et al., 2008; Hj?rleifsd?ttir & Gunnarsd?ttir, 2008). Participants rated their lack of energy as their lowest quality of life item which indicated they thought it was very important but were very dissatisfied with it. This is to be expected as they were in the midst of chemotherapy treatments and the resulting side effects. Other items which received a low quality of life rating in both groups included their health and their current amount of worries. Interesting to note that the intervention group ? in which five out of the six were employed ? listed ?not having a job? as one of their three lowest quality of life items. Perhaps they were concerned that the cancer would cause them to lose their job. 94 Table 7 Comparison of Weighted Means Showing Items with Highest Importance and Satisfaction in Intervention and Control Groups at Baseline and Three Months Baseline Three Months Intervention Group Intervention Group Faith in God (30.0) Faith in God (30.0) Health care (29.6) Health care (29.6) Children (29.0) Job (28.1) Emotional support from family (29.0) Control Group Control Group Faith in God (30.0) Faith in God (30.0) Health care (30.0) Emotional support from family (29.2) Emotional support from family (29.2) Peace of mind (28.9) 95 Table 8 Comparison of Weighted Means Showing Items with Highest Importance but Lowest Satisfaction in Intervention and Control Groups at Baseline and Three Months Baseline Three Months Intervention Group Intervention Group Health (11.6) Energy (9.9) Not having a job (15.0) Not having a job (11.6 ) Amount of worries (16.8) Amount of worries (14.1) Control Group Control Group Energy level (7.8) Energy level (6.3) Personal appearance (9.8) Health (12.0) Health (12.0) Ability to fulfill family responsibilities (12.0) Usefulness to others (12.0) Examination of the change in quality of life ratings over time revealed much less variance within the intervention group than the control group. This could have been attributed to many factors including difference in group sizes, differences in chemotherapy and radiation treatment regimes, or individual disease progression. The biggest change in scores recorded for the intervention group was the drop in satisfaction of their energy level (Q4) (see Figure 3). The next three largest drops were a decrease in satisfaction with their ability to manage family responsibilities (Q16), the amount of pain they were experiencing (Q3), and their ability to be independent with self care (Q5). All 96 of these were to be expected considering where these patients were in their treatment regimes. The largest jump in ratings was due to an increase of both importance and satisfaction with their health (Q1) and ability to care for their needs financially (Q24). While being more satisfied with their health seems odd, it may reflect their recovery from pre-chemotherapy surgeries, or perhaps the participants were expressing more of an appreciation for health now that their health status had changed. The third highest jump was in satisfaction with their home, apartment or wherever they lived (Q20). Figure 3. Comparison of Weighted Mean Quality of Life Question Scores from Baseline and Three Months for Intervention Group The control group showed much more variance in their ratings. As suggested above, this variance could be attributed to the smaller group size, difference in disease treatments or progression, or many other factors. The most dramatic drop was caused by 97 a decrease in both satisfaction with and importance of their neighborhood (Q19) (see Figure 4). A large drop in satisfaction was also noted in the amount of control they had over their lives (Q6), the amount of pain they were in (Q3), and their home, apartment or place where they lived (Q20). Interestingly, at three months, being useful to others went up in importance yet they were less satisfied with it (Q17). The most significant jump was seen in their satisfaction with their children (Q9), followed by their peace of mind (Q27) and themselves in general (Q33). Friends (Q13) and their personal appearance (Q32) increased in both importance and satisfaction. Figure 4. Comparison of Weighted Mean Quality of Life Question Scores from Baseline and Three Months for Control Group 98 Correlations between Uncertainty and Quality of Life Pearson?s r was used to test the relationship between expressive writing, uncertainty and quality of life in the intervention group and in the control group. There are several factors which could have caused the results to be statistically non-significant, including having multiple diagnoses and thus different chemotherapy drugs and treatment protocols, the small sample size and resulting low power of the study, the length of the study not allowing adequate time to process effects of writing, each participant?s life situations, and many other unknown factors. Figure 5 graphically portrays the changes in combined uncertainty and quality of life scores for each participant from baseline to three months. This illustrates the unique response of each participant to their cancer experience. Four participants experienced a drop in their uncertainty over the time period. This could be attributed to being over their initial shock and settling down into a treatment routine, or being completely over with their treatments. Five showed an increase in uncertainty. This could be due to the appearance of new side effects, treatments not going well, or having a difficult time performing at their job or fulfilling family responsibilities. Four reported an increase or no change in their perceived quality of life. This could be credited to a reprioritizing of their lives, a change in focus of what they thought was important. This increase could also speak to their coping skills. Five reported a drop in their quality of life. Treatments could have been one contributing factor, but also the global impact of cancer on their life (Ferrans, 1990). 99 Figure 5. Changes in Combined Uncertainty and Quality of Life Scores of Each Participant from Baseline and Three Months Limitations Limitations are those things over which the researcher has no control, but which can affect the accuracy or purity of the study findings and which must be taken into account during the interpretations of the findings. A list of potential limitations was included in chapter one. The following three limitations are discussed in more detail here due to their obvious impact on the study. The first limitation encountered almost immediately was a difficulty in obtaining buy-in from all providers at the facilities. Some actively encouraged patients to join the study, whereas other providers merely included the study flier in the large stack of information given to each newly-diagnosed patient. Obviously patients who just received a flier in a stack of papers were unlikely to respond compared with those whose providers encouraged their participation. This lack of buy-in 100 may have contributed to the small total sample size, which in turn affected the results obtained through the statistical analysis. Another limitation encountered was the attrition rate. Five participants did not return the three-month data tools, thus effectively dropping out of the study. This attrition rate may be explained by progression of their disease, metastasis or death; treatment regime and intensity of side effects; presence or lack of a support system; or current life situations including previous and ongoing uncertainties or issues pertaining to their quality of life. The attrition rate might also have been impacted by the fact that the investigator met with the participants at baseline and had them fill out the data forms at that time, whereas the three-month forms were mailed out and participants did not have the investigator there in person encouraging them to complete the forms. Perhaps more participants would have responded if the investigator had offered to meet with them face- to-face at three months to fill out the forms or had contacted them through a personal telephone call. This attrition rate significantly affected the control group data and made it difficult to compare the control group to the intervention group. A third limitation in this study worth noting was the lack of control over what the control group participants did while in this study. Since the participants were randomized into their groups, the investigator was unable to ensure that those who usually journal were put into the expressive writing group. In fact, two of the three participants who said they already did journaling were randomized into the non-writing control group. Of the three control group participants who responded at three months, one said that they did some journaling during the study time period. While the control group did not receive the writing prompt, just the act of journaling may have impacted their response data. 101 Implications for Further Research Frattaroli (2006) demonstrated that expressive writing can have a positive effect on a person?s psychological health, physiological health, reported health, general functioning, and life outcomes. Specific benefits of expressive writing for cancer patients with regard to their uncertainty and perceived quality of life, both as a general cancer patient population as well as by specific diagnosis, have yet to be determined. Repeating this study with a larger sample size may provide sufficient power to produce valid data. Additionally, limiting the sample to one type of cancer population would allow results to be compared more equally. A lack of uniformity of this sample meant that some participants were in the middle of their treatments at three months but others might have been finished ? two very different spots psychologically. Reading the journals and doing word and content analysis might also provide researchers with data enabling them to draw correlations between the content in the journals and the data results. Furthermore, investigators might want to experiment with different writing instructions, such as varying the number of days participants are asked to write or requiring a specific amount of time spent writing. Implications for Practice Alternative and complementary therapies are beginning to be researched and accepted by health care providers and insurance companies. In today?s uncertain economy, having an intervention which is inexpensive, non-invasive and proven beneficial should be welcomed by both providers and patients alike. It is important to understand the potential benefits of expressive writing and the impact writing might have on a patient?s uncertainty and perceived quality of life. This knowledge and use of a 102 writing intervention can assist health care providers in providing more holistic care by addressing the psychological effects associated with a cancer diagnosis. Summary The first goal of Healthy People 2010 is to ?increase quality and years of healthy life? (U.S. Department of Health and Human Services, 2000, para.1). Oncologists have developed cancer treatment regimes to fight for a patient?s chance at a longer life, but do not always include treatments or interventions to address their patient?s perceived uncertainties and quality of life. By recommending expressive writing to patients, health care professionals acknowledge the psychological ramifications of the disease and treat each patient in a more holistic manner (Laccetti, 2007). Expressive writing empowers patients to improve their own positive functioning by freeing themselves of bondage from past and current traumas. Participants in this research study were given the opportunity to practice expressive writing and through the completion of the two tools give voice to their uncertainties and quality of life. This sample was most uncertain about disease progression ? the disconcertion caused by fluctuations in how they felt and what they were able to do, and the possibility of a ?cure?. The uncertainty data served health care professionals a reminder to use vocabulary easily understood by their patients when giving explanations. This sample revealed the high importance they placed in their faith in God and their family. Not surprisingly, as a whole they were also frustrated over their lack of energy and good health. The responses revealed the uniqueness of each person manifested by their response to illness and their choices of what mattered most when they considered their quality of life. 103 REFERENCES American Association of Colleges of Nursing. (2009). End-of-life nursing education consortium. Retrieved April 1, 2009 from http://www.aacn.nche.edu/ELNEC/factsheet.htm American Cancer Society. (2009). Cancer facts and figures 2009. Retrieved June 22, 2009 from http://www.cancer.org/downloads/STT/500809web.pdf Anderson, B. L., Kiecolt-Glaser, J. K., & Glaser, R. (1994). A biobehavioral model of cancer stress and disease course. American Psychologist, 49(5), 389-404. Andrews, F., & Withey, S. (1976). Social indicators of well being. New York, NY: Plenum Press. Ap?stolo, J. L. A., Viveiros, C. S. C., Nunes, H. I. R., & Domingues, H. R. F. (2007). Illness, uncertainty and treatment motivation in type 2 diabetes patients. Rev Latino- am Enfermagem, 15(4), 575-582. Araujo, A. M. F., Mendez, J. C., Coelho, A. L., Sousa, B., Barata, F., Figueiredo, A., et al. (2009). Phase II study of celecoxib with cisplatin plus etoposide in extensive- stage small cell lung cancer. Cancer Investigation, 27(4), 391-396. Barclay, L. J., & Skarlicki, D. P. (2009). Healing the wounds of organizational injustice: Examining the benefits of expressive writing. Journal of Applied Psychology, 94(2), 511-523. 104 Beatty, L., Oxlad, M., Koczwara, B., & Wade, T. D. (2008). The psychosocial concerns and needs of women recently diagnosed with breast cancer: A qualitative study of patient, nurse, and volunteer perspectives. Health Expectations, 11(4), 331-342. Beaulieu, S., Wood-Dauphinee, S., Abenhaim, L., & Abrahamowicz, M. (1997). Development and preliminary testing of a low back pain module to accompany a generic quality of life index. Canadian Journal of Rehabilitation, 10(3), 167-83. Berenbaurm, H., Bredemeier, K., & Thompson, R. J. (2008). Intolerance of uncertainty: Exploring its dimensionality and associations with need for cognitive closure, psychopathology, and personality. Journal of Anxiety Disorders, 22(1), 117-125. Boehmke, M. M., & Dickerson, S. S. (2006). The diagnosis of breast cancer: Transition from health to illness. Oncology Nursing Forum, 33(6), 1121-1127. Bradley, S., Sherwood, P. R., Donovan, H. S., Hamilton, R., Rosenzweig, M., Hricik, A., et al. (2007). I could lose everything: Understanding the cost of a brain tumor. Journal of Neuro-Oncology, 85(3), 329-338. Bredemeier, K., & Berenbaum, H. (2008). Intolerance of uncertainty and perceived threat. Behaviour Research and Therapy, 46(1), 28-38. Buhr, K., & Dugas, M. J. (2002). The intolerance of uncertainty scale: Psychometric properties of the English version. Behaviour Research and Therapy, 40(8), 931-45. Burton, C. M., & King, L. A. (2008). Effects of (very) brief writing on health: The two- minute miracle. British Journal of Health Psychology, 13(1), 9-14. Busch, N. B., & Valentine, D. (2000). Empowerment practice: A focus on battered women. AFFILIA, 15(1), 82-95. 105 Carleton, R. N., Norton, M. A. P. J., & Asmundson, G. J. G. (2007). Fearing the unknown: A short version of the intolerance of uncertainty scale. Journal of Anxiety Disorders, 21(1), 105-117. Carroll, D.L., Hamilton, G.A., & McGovern, B.A. (1999). Changes in health status and quality of life and the impact of uncertainty in patients who survive life-threatening arrhythmias. Heart & Lung, 28(4), 251-260. Carver, C. S., Pozo, C., Harris, S. D., Noriega, V., Scheier, M. F., Robinson, D. S., et al. (1993). How coping mediates the effect of optimism on distress: A study of women with early stage breast cancer. Journal of Personality and Social Psychology, 65(2), 375-390. Clauson, M. I. (1996). Uncertainty and stress in women hospitalized with high-risk pregnancy. Clinical Nursing Research, 5(3), 309-325. Cohen, G. D., Firth, K. M., Biddle, S., & Lewis, M. J. L. (2009). The first therapeutic game specifically designed and evaluated for Alzheimer?s disease. American Journal of Alzheimer?s Disease and Other Dementias, 23(6), 540-551. Corn, B.W., Moughan, J., Knisely, J.P.S., Fox, S.W., Chakravarti, A., Yung, W.K.A., et al. (2008). Prospective evaluation of quality of life and neurocognitive effects in patients with multiple brain metastases receiving whole-brain radiotherapy with or without thalidomide on radiation therapy oncology group (RTOG) trial 0118. International Journal of Radiation Oncology, Biology, Physics, 71(1), 71-78. Cyboran, V. L. (2005). Fostering workplace learning through online journaling. Performance Improvement, 44(7), 34-39. 106 de Moor, C., Sterner, J., Hall, M., Warneke, C., Gilani, Z., Amato, R., et al. (2002). A pilot study of the effects of expressive writing on psychological and behavioral adjustment in patients enrolled in a phase II trial of vaccine therapy for metastatic renal cell carcinoma. Health Psychology, 21(6), 615-619. Doell, H. J. (2008). Three little words no one wants to hear. Clinical Journal of Oncology Nursing, 12(4), 551-554. Donovan-Kicken, E., & Bute, J. J. (2008). Uncertainty of social network members in the case of communication-debilitating illness or injury. Qualitative Health Research, 18(1), 5-18. Doyle, N. (2008). Cancer survivorship: Evolutionary concept analysis. Journal of Advanced Nursing, 62(4), 499-509. Dunlap, J. C. (2006). Using guided reflective journaling activities to capture students? changing perceptions. TechTrends, 50(6), 20-26. Eastwood, J., Doering, L., Roper, J., & Hays, R.D. (2008). Uncertainty and health-related quality of life 1 year after coronary angiography. American Journal of Critical Care, 17(3), 232-243. Elliot, T. R., & Berry, J. W. (2009). Brief problem-solving training for family caregivers of persons with recent-onset spinal cord injuries: A randomized controlled trial. Journal of Clinical Psychology, 65(4), 406-422. empower. (2009). In Merriam-Webster Online Dictionary. Retrieved April 20, 2009, from http://www.merriam-webster.com/dictionary/empower expressive. (2008). In Merriam-Webster Online Dictionary. Retrieved December 6, 2008, from http://www.merriam-webster.com/dictionary/expressive 107 Fatimah, P. R., Jemain, A., Ibrahim, K., Nasir, M. S., & Anuar, K. M. (2009). Quality family deployment: A new perspective in determining priority importance for improving work performance in organization. Social Indicators Research, 92(1), 131- 149. Ferrans, C. E. (1990). Development of a quality of life index for patients with cancer. Oncology Nursing Forum, 17(3)Supplement, 15-19. Ferrans, C. E. (1996). Development of a conceptual model of quality of life. Scholarly Inquiry for Nursing Practice: An International Journal, 10(3), 293-304. Ferrans, C. E., & Powers, M. J. (1985). Quality of life index: Development and psychometric properties. Advances in Nursing Science, 8(1), 15-24. Ferrans, C. E., & Powers, M. J. (1992). Psychometric assessment of the quality of life index. Research in Nursing and Health, 15(1), 29-38. Ferrans, C. E., & Powers, M. (1998a). Quality of life index. Retrieved January 17, 2009 from http://www.uic.edu/orgs/qli/index.htm Ferrans, C.E., & Powers, M. (1998b). Reliability and validity. Retrieved January, 17, 2009 from http://www.uic.edu/orgs/qli/reliability/reliabilityhome.htm Ferrell, B.R., & Coyle, N. (2008). The nature of suffering and the goals of nursing. Oncology Nursing Forum, 35(2), 241-247. Ferrell, B., Cullinane, C. A., Ervin, K., Melancon, C., Uman, G. C., & Juarez, G. (2005). Perspectives on the impact of ovarian cancer: Women?s views of quality of life. Oncology Nursing Forum, 32(6), 1143-1149. Ferrell, B., Ervin, K., Smith, S., Marek, T., & Melancon, C. (2002). Family perspectives of ovarian cancer. Cancer Practice, 10(6), 269-276. 108 Ferrell, B. R., & Grant, M. (n.d.). Quality of life patient/Cancer survivor version manual. Retrieved October 19, 2008, from http://prc.coh.org/QOL-CS.pdf Ferrell, B. R., Grant, M., Funk, B., Otis-Green, S., & Garcia, N. (1997). Quality of life in breast cancer: Part 1: Physical and social well-being. Cancer Nursing, 20(6), 398- 408. Ferrell, B. R., Grant, M., Funk, B., Otis-Green, S., & Garcia, N. (1998). Quality of life in breast cancer: Part II: Psychological and spiritual well-being. Cancer Nursing, 21(1), 1-9. Ferrell, B., Grant, M., Padilla, G., Vemuri, S., & Rhiner, M. (1991).The experience of pain and perceptions of quality of life: Validation of a conceptual model. The Hospice Journal, 7(3), 9-24. Ferrell, B. R., Hassey-Dow, K., & Grant, M. (1995). Measurement of the quality of life in cancer survivors. Quality of Life Research, 4(6), 523-531. Ferrell, B., Smith, S., Cullinane, C., & Melancon, C. (2003a). Symptom concerns of women with ovarian cancer. Journal of Pain and Symptom Management, 25(6), 528- 538. Ferrell, B., Smith, S., Cullinane, C., & Melancon, C. (2003b). Psychological well being and quality of life in ovarian cancer survivors. Cancer, 98(5), 1061-1071. Ferrell, B.R., Wisdom, C., & Wenzl, C. (1989). Quality of life as an outcome variable in the management of cancer pain. Cancer, 63(11), 2321-2327. Field, M. J., & Cassel, C. K. (Eds.). (1997). Institute of Medicine: Approaching death: Improving care at the end of life. Washington D.C.: National Academy Press. 109 Retrieved June 6, 2009 from http://www.nap.edu/openbook.php?record_id=5801&page=R1 Fogarty, C., & Cronin, P. (2008). Waiting for healthcare: A concept analysis. Journal of Advanced Nursing, 61(4), 463-471. Francis, M. E., & Pennebaker, J. W. (1992). Putting stress into words: The impact of writing on physiological, absentee, and self-reported emotional well-being measures. American Journal of Health Promotion, 6(4), 280-287. Frank, A. (1989). The diary of Anne Frank: The critical edition. New York, NY: Doubleday. Frattaroli, J. (2006). Experimental disclosure and its moderators: A meta-analysis. Psychological Bulletin, 132(6), 823-865. Freeston, M. H., Rh?aume, J., Letarte, H., Dugas, M. J., & Ladouceur, R. (1994). Why do people worry? Personality and Individual Differences, 17(6), 791-802. Freire, P. (1970) Pedagogy of the oppressed. New York, NY: Herder and Herder. Frisina, P.G., Borod, J. C., & Lepore, S. J. (2004). A meta-analysis of the effects of written emotional disclosure on the health outcomes of clinical populations. The Journal of Nervous and Mental Disease, 192(9), 629-634. Gall, M.D., Gall, J.P, & Borg, W.R. (2007). Educational research, an introduction (8th ed.). Boston, MA: Pearson Education. Gellaitry, G., Peters, K., Bloomfield, D., & Horne, R. (2009). Narrowing the gap: The effects of an expressive writing intervention on perceptions of actual and ideal emotional support in women who have completed treatment for early stage breast 110 cancer. Psycho-Oncology, 19(1), 77-84. Retrieved March, 6, 2009 from http://www3.interscience.wiley.com/cgi-bin/fulltext/121681094/PDFSTART Gidron, Y., Peri, T., Connolly, J. F., & Shalev, A. Y. (1996). Written disclosure in posttraumatic stress disorder: Is it beneficial for the patient? The Journal of Nervous and Mental Disease, 184(8), 505-507. Grant, M., Padilla, G. V., Ferrell, B. R., & Rhiner, M. (1990). Assessment of quality of life with a single instrument. Seminars in Oncology Nursing, 6(4), 260-270. Guti?rrez, L. M. (1994). Beyond coping: An empowerment perspective on stressful life events. Journal of Sociology & Social Welfare, 21(3), 201-220. Guti?rrez, L. M., DeLois, K. A., & GlenMaye, L. (1995). Understanding empowerment practice: Building on practitioner-based knowledge. Families in Society, 76(9), 534- 542. Hacker, E., Ferrans, C., Verlen, E., Ravandi, F., van Besien, K., Gelms, J., et al. (2006). Fatigue and physical activity in patients receiving hematopoietic stem cell transplantation. Oncology Nursing Forum, 33(3), 614-624. Hampton, S. E., & Morrow, C. (2003). Reflective journaling and assessment. Journal of Professional Issues in Engineering Education and Practice, 129(4), 186-189. Hayes, A. (2005). A mental health nursing clinical experience with hospice patients. Nurse Educator, 30(2), 85-88. Hilton, B. A. (1994). The uncertainty stress scale: Its development and psychometric properties. Canadian Journal of Nursing Research, 26(3), 15-30. Hilton, A., & Skrutkowski, M. (2002). Translating instruments into other languages. Cancer Nursing, 25(1), 1-7. 111 Hj?rleifsd?ttir, E., Hallberg, I. R., Gunnarsd?ttir, E. D., & Bolmsj?, I. A. (2008). Living with cancer and perception of care: Icelandic oncology outpatients, a qualitative study. Support Care Cancer, 16(5), 515-524. Hochwald, L. (2008). A head-to-toe look at breast cancer?s real effects. Retrieved October 15, 2008 from www.cnn.com/2008/HEALTH/conditions/10/15/healthmag.breast.cancer.experience/i ndex.html Holm, S., Schonberger, M., Poulson, I., & Caetano, C. (2009). Patients? and relatives? experience of difficulties following severe traumatic brain injury: The sub-acute stage. Neuropsychological Rehabilitation, 19(3), 444-460. Huang, S. M., Lee, C. H., Chien, L. Y., Liu, H. E., & Tai, C. J. (2004). Postoperative quality of life among patients with thyroid cancer. Journal of Advanced Nursing, 47(5), 492-499. Hubbs, D. L., & Brand, C. F. (2005). The paper mirror: Understanding reflective journaling. Journal of Experiential Education, 28(1), 60-71. Hutchinson, S. L., & Wexler, B. (2007). Is ?raging? good for health?: Older women?s participation in the raging grannies. Health Care for Women International, 28(1), 88- 118. Iwamitsu, Y., Shimoda, K., Abe, H., Tani, T., Kodama, M., & Okawa, M. (2003). Differences in emotional distress between breast tumor patients with emotional inhibition and those with emotional expression. Psychiatry and Clinical Neurosciences, 57(3), 289-294. 112 Jang, S. J. (2009). The relationships of flexible work schedules, workplace support, supervisory support, work-life balance, and the well-being of working parents. Journal of Social Service Research, 35(2), 93-104. Juarez, G., Ferrell, B., & Borneman, T. (1998). Cultural considerations in education for cancer pain management. Journal of Cancer Education, 14(3), 168-173. K?llay, E., & B?ban, A. (2008). Emotional benefits of expressive writing in a sample of Romanian female cancer patients. Cognition, Brain & Behavior, 12(1), 115-129. Kasper, J., Geiger, F., Freiberger, S., & Schmidt, A. (2008). Decision-related uncertainties perceived by people with cancer--modelling the subject of shared decision making. Psycho-Oncology, 17(1), 42-48. Ko, H. C., & Kuo, F. Y. (2009). Can blogging enhance subjective well-being through self-disclosure? CyberPsychology & Behavior, 12(1), 75-79. Kovac, S. H., & Range, L. M. (2002). Does writing about suicidal thoughts and feelings reduce them? Suicide and Life-Threatening Behavior, 32(4), 428-440. Laccetti, M. (2007). Expressive writing in women with advanced breast cancer. Oncology Nursing Forum, 34(5), 1019-1024. Lee, A.B.J. (1996). The Empowerment approach to social work practice. In F.J. Turner (Ed.), Social work treatment: Interlocking theoretical approaches (4th ed., pp. 218- 249). New York, NY: The Free Press. Lee, A.B.J. (2001). The empowerment approach to social work practice (2nd ed.). New York: Columbia University Press. 113 Leite, C., & Kuiper, N. A. (2008). Client uncertainty and the process of change in psychotherapy: The impact of individual differences in self-concept, clarity and intolerance of uncertainty. Journal of Contemporary Psychotherapy, 38(2), 55-64. life. (2008). In Merriam-Webster Online Dictionary. Retrieved September 20, 2008, from http://www.merriam-webster.com/dictionary/life Lisovicz, N., Johnson, R. E., Higginbotham, J., Downey, J. A., Hardy, C. M., Fouad, M. N., et al. (2006). The deep south network for cancer control. Cancer, 107(S8), 1971- 1979. Loberiza, F.R., Rizzo, JD., Bredeson, C.N., Antin, J.H., Horowitz, M.M., Weeks, J.C., et al. (2002). Association of depressive syndrome and early deaths among patients after stem-cell transplantation for malignant diseases. Journal of Clinical Oncology, 20(8), 2118-2126. Loo, R. (2002). Journaling: A learning tool for project management training and team- building. Project Management Journal, 33(4), 61-66. Lum, D. (1996). Social work practice and people of color: A process-stage approach (3rd ed.). Pacific Grove, CA: Brooks/Cole Publishing Company. Lyubomirsky, S., Sousa, L., & Dickerhoof, R. (2006). The costs and benefits of writing, talking and thinking about life?s triumphs and defeats. Journal of Personality and Social Psychology, 90(4), 692-708. McCahill, L., Ferrell, B. R., & Virani, R. (2001). Improving cancer care at the end of life. The Lancet Oncology, 2(2), 103-108. McCormick, K.M. (2002). A concept analysis of uncertainty in illness. IMAGE: Journal of Nursing Scholarship, 34(2), 127-131. 114 McNair, D.M., Lorr, M., & Droppelman, L.F. (1971). Profile of mood states. San Diego, CA: Educational and Industrial Testing Service. Mishel, M. H. (1981). The measurement of uncertainty in illness. Nursing Research, 30(5), 258-263. Mishel, M. H. (1984). Perceived uncertainty and stress in illness. Research in Nursing and Health, 7, 163-171. Mishel, M. H. (1988). Uncertainty in illness. IMAGE: Journal of Nursing Scholarship, 20(4), 225-232. Mishel, M.H. (1989) Uncertainty in illness scale ? community form. Received from M. Mishel on January 15, 2009. Mishel, M.H. (1997a). Uncertainty in acute illness. In J. J. Fitzpatrick and J. Norbeck (Eds.), Annual Review of Nursing Research, 15, 57-80. Mishel, M.H. (1997b) Uncertainty in illness scales manual. Received from M. Mishel on January 15, 2009. Morgan, J., & Farsides, T. (2009). Measuring meaning in life. Journal of Happiness Studies, 10(2), 197-214. Morgan, N.P., Graves, K.D., Poggi, E.A., & Cheson, B.D. (2008). Implementing an expressive writing study in a cancer clinic. The Oncologist, 13(2), 196-204. Morgan. D., Johnson, J., & Garrison, D. (2005). Reflective journaling: Bridging the theory-practice gap. In H. Fieldman & M. Greenberg (Eds.), Educating nurses for leadership (pp. 110-118). New York: Springer Publishing. Morse, J. M., & Penrod, J. (1999). Linking concepts of enduring, uncertainty, suffering and hope. Image: Journal of Nursing Scholarship, 31(2), 145-150. 115 M?ller-Stich, B., Reiter, M. A., Mehrabi, A., Wente, M. N., Fischer, L., K?ninger, J., et al. (2009). No relevant difference in quality of life and functional outcome at 12 months? follow-up ? a randomized controlled trial comparing robot-assisted versus conventional laparoscopic Nissen fundoplication. Langerbeck?s Archives of Surgery, 394(3), 441-446. Napier, A. (2006). Empowerment theory. Retrieved April 21, 2009 from http://www3.malone.edu/media/1/7/71/Empowerment_Presentation_A_Napier.s06.pd f Niederhoffer, K.G., & Pennebaker J.W. (2002). Sharing one?s story: On the benefits of writing or talking about emotional experience. In C.R. Snyder & S.J. Lopez (Eds.), Handbook of positive psychology (pp. 573-583). New York, NY: Oxford University Press. Northouse, L. L., Mood, D. W., Montie, J. E., Sandler, H. M., Forman, J. D., Hussain, M., et al. (2007). Living with prostate cancer: Patients? and spouses? psychosocial status and quality of life. Journal of Clinical Oncology, 25(27), 4171-4177. Osaba, D. (1994). Lessons learned from measuring health-related quality of life in oncology. Journal of Clinical Oncology, 12(3), 608-616. Padilla, G. V., & Grant, M. M. (1985). Quality of life as a cancer nursing outcome variable. Advances in Nursing Science, 8(1), 45-60. Padilla, G. V., Presant, C., Grant, M. M., Metter, G., Lipsett, J., & Heide, F. (1983). Quality of life index for patients with cancer. Research in Nursing and Health, 6(3), 117-126. 116 Page, N., & Czuba, C. E. (1999). Empowerment: What is it? Journal of Extension, 37(5). Retrieved April 21, 2009 from http://www.joe.org/joe/1999october/comm1.php Palacios, C. S., Torres, M. V. T., & Mena, M. J. B. (2009). Negative aging stereotypes and their relation with psychosocial variables in the elderly population. Archives of Gerontology & Geriatrics, 48(3), 385-390. Pennebaker, J. W. (1999). The effects of traumatic disclosure on physical and mental health: The values of writing and talking about upsetting events. International Journal of Emergency Mental Health, 1(1), 9-18. Pennebaker, J. W. (2003). Writing about emotional experiences as a therapeutic process. In P. Salovey & A. J. Rothman (Eds.), Social psychology of health (pp. 362-368). New York, NY: Psychology Press. Pennebaker, J. W., & Beall, S. K. (1986). Confronting a traumatic event: Toward an understanding of inhibition and disease. Journal of Abnormal Psychology, 95(3), 274- 281. Pennebaker, J.W., & Chung, C. K. (2007). Expressive writing, emotional upheavals, and health. In H. S. Friedman & R. C. Silver (Eds.), Foundations of health psychology (pp. 263-284). New York, NY: Oxford University Press. Pennebaker, J. W., & Hoover, C. W. (1986). Inhibition and cognition: Toward an understanding of trauma and disease. In R. J. Davidson, G. E. Schwartz, & D. Shapiro (Eds.), Consciousness and self-regulation (Vol. 4, pp. 107-136). New York, NY: Plenum Press. 117 Pennebaker, J. W., & O?Heeron, R. C. (1984). Confiding in others and illness rates among spouses of suicide and accidental-death victims. Journal of Abnormal Psychology, 93(4), 473-476. Penrod, J. (2001). Refinement of the concept of uncertainty. Journal of Advanced Nursing, 34(2), 238-245. Penrod, J. (2007). Living with uncertainty: Concept advancement. Journal of Advanced Nursing, 57(6), 658-667. Penrod, J., & Morse, J. M. (1997). Strategies for assessing and fostering hope: The hope assessment guide. Oncology Nursing Forum, 24(6), 1055-1063. Perez, D.J., McGee, R., Campbell, A.V., Christensen, E. A., & Williams, S. (1997). A comparison of time trade-off and quality of life measures in patients with advanced cancer. Quality of Life Research, 6(2), 133-138. Polit, D. F., & Beck, C.T. (2008). Nursing research: Generating and assessing evidence for nursing practice (8th ed.). Philadelphia, PA: Lippincott Williams & Wilkins. quality. (2008). In Merriam-Webster Online Dictionary. Retrieved September 20, 2008, from http://www.merriam-webster.com/dictionary/quality quality of life. (2001). In WordNet 1.7.1. Retrieved September 20, 2008, from http://www.websters-online-dictionary.org/definition/quality+of+life Rancour, P., & Brauer, K. (2003). Use of letter writing as a means of integrating an altered body image: A case study. Oncology Nursing Forum, 30(5), 841-846. Rappaport, J. (1984). Studies in empowerment: Introduction to the issue. Prevention in Human Services, 3(2), 1-7. 118 Richards, J. M., Beal, W. E., Seagal, J. D., & Pennebaker, J. W. (2000). Effects of disclosure of traumatic events on illness behavior among psychiatric prison inmates. Journal of Abnormal Psychology, 109(1), 156-160. Rissell, C. (1994). Empowerment: The holy grail of health promotion? Health Promotion International, 9(1), 39-47. Rosenberg, H. J., Rosenberg, S. D., Ernstoff, M. S., Wolford, G. L., Amdur, R. J., Elshamy, M. R., et al. (2002). Expressive disclosure and health outcomes in a prostate cancer population. International Journal of Psychiatry in Medicine, 32(1), 37-53. Rustoen, T., Moum, T., Wiklund, I., & Hanestad, B. R. (1999). Quality of life in newly diagnosed cancer patients. Journal of Advanced Nursing, 29(2), 490-498. Sammarco, A. (2001). Perceived social support, uncertainty, and quality of life of younger breast cancer survivors. Cancer Nursing, 24(3), 212-219. Sammarco, A. (2003). Quality of life among older survivors of breast cancer. Cancer Nursing, 26(6), 431-438. Schaefer, K. M., & Curley, C. (2005). Use of focus groups and reflective journaling in a hospice experience. Nurse Educator, 30(30), 93-94. Schneider, C.M., Hsieh, C.C., Sprod, L.K., Cater, S.D., & Hayward, R. (2007). Cancer treatment-induced alterations in muscular fitness and quality of life: The role of exercise training. Annals of Oncology, 18(12), 1957-1962. Schoutrop, M. J. A., Lange, A., Hanewald, G., Davidovich, U., & Salomon, H. (2002). Structured writing and processing major stressful events: A controlled trial. Psychotherapy and Psychosomatics, 71(3), 151-157. 119 Smith, S., Anderson-Hanley, C., Langrock, A., & Compas, B. (2005). The effects of journaling for women with newly diagnosed breast cancer. Psycho-Oncology, 14(12), 1075-1082. Smyth, J. M. (1998). Written emotional expression: Effect sizes, outcome types, and moderating variables. Journal of Consulting and Clinical Psychology, 66(1), 174- 184. Smyth, J. M., Stone, A. A., Hurewitz, A., & Kaell, A. (1999). Effects of writing about stressful experiences on symptom reduction in patients with asthma or rheumatoid arthritis: A randomized trial. The Journal of the American Medical Association, 281(14), 1304-1309. Spera, S. P., Buhrfeind, E. D., & Pennebaker, J. W. (1994). Expressive writing and coping with job loss. Academy of Management Journal, 37(3), 722-732. Spitzer, W. O., Dobson, A. J., Hall, J., Chesterman, E., Levi, J., Shepherd, R., et al. (1981). Measuring the quality of life of cancer patients. Journal of Chronic Disease, 34(12), 585-597. Stanton, A. L., Danoff-Burg, S., Sworowski, L. A., Collins, C. A., Branstetter, A. D., Rodriguez-Hanley, A., et al. (2002). Randomized, controlled trial of written emotional expression and benefit finding in breast cancer patients. Journal of Clinical Oncology, 20(20), 4160-4168. Staples, P., & Jeffery, J. (1997). Quality of life, hope, and uncertainty of cardiac patients and their spouses before coronary artery bypass surgery. Canadian Journal of Cardiovascular Nursing, 8(1), 7-16. 120 Swift, C., & Levin, G. (1987). Empowerment: An emerging mental health technology. Journal of Primary Prevention, 8(1&2), 71-94. The EPEC Project. (2009). The EPEC project. Retrieved April 1, 2009 from http://www.epec.net/EPEC/Webpages/about.cfm Thorpe, R., & Loo, K. (1999). A psychometric investigation of scores on the Watson- Glaser critical thinking appraisal new Form S. Educational and Psychological Measurement, 59(6), 995-1003. Toldeo, R. C. M., Alexandre, N. M. C., & Rodrigues, R. C. M. (2008). Psychometric evaluation of a Brazilian Portuguese version of the Spitzer quality of life index in patients with low back pain. Revista Latino-Americana de Enfermaem, 16(6), 943- 950. U.S. Department of Health and Human Services. (2000). Healthy People 2010: What are its goals? Retrieved February 9, 2009 from http://www.healthypeople.gov/About/goals.htm uncertainty. (2007). APA Dictionary of Psychology. Washington, D.C.: American Psychological Association. uncertainty. (2008). Genetic Consequences of Cancer Treatment. Retrieved June 8, 2008 from http://www.gcct.org/Glossary.html uncertainty. (n.d.). Dictionary.com Unabridged. Retrieved June 6, 2008, from Dictionary.com website: http://dictionary.reference.com/browse/uncertainty uncertainty. (2008). In Merriam-Webster Online Dictionary. Retrieved September 20, 2008, from http://www.merriam-webster.com/dictionary/uncertainty 121 uncertainty. (2006). WordNet? 3.0. Retrieved June 6, 2008, from Princeton University WordNet website: http://wordnetweb.princeton.edu/perl/webwn?s=uncertainty&sub=Search+WordNet& o2=&o0=1&o7=&o5=&o1=1&o6=&o4=&o3=&h= Visser, M. R. M., van Lanschot, J. J. B., van der Velden, J., Kloek, J. J., Gouma, D. J., & Sprangers, M. A. G. (2006). Quality of life in newly diagnosed cancer patients waiting for surgery is seriously impaired. Journal of Surgical Oncology, 93(7), 571- 577. Vitek, L., Rosenzweig, M. Q., & Stollings, S. (2007). Distress in patients with cancer: Definition, assessment, and suggested interventions. Clinical Journal of Oncology Nursing, 11(3), 413-418. Walker, B. L., Nail, L. M., & Croyle, R. T. (1999). Does emotional expression make a difference in reactions to breast cancer? Oncology Nursing Forum, 26(6), 1025-1032. Wallace, M. (2003). Uncertainty and quality of life of older men who undergo watchful waiting for prostate cancer. Oncology Nursing Forum, 30(2), 303-309. Wallerstein, N. (1992). Powerlessness, empowerment, and health: Implications for health promotion programs. American Journal of Health Promotion, 6(3), 197-205. Williams, K., & Watkins, R. (n.d.). A descriptive study of music student teachers? issues derived from postings on an electronic bulletin board. Retrieved June 27, 2007, from The University of Texas at San Antonio, Institute for Music Research website: http://music.utsa.edu/tdml/conf-VII/VII-Williams&Watkins.html 122 Wonghongkul, T., Dechaprom, N., Phumivichuvate, L., & Losawatkul, S. (2006). Uncertainty appraisal coping and quality of life in breast cancer survivors. Cancer Nursing, 29(3), 250-257. World Health Organization. (1986). The Ottawa charter for health promotion. Retrieved March 7, 2009 from http://www.who.int/healthpromotion/conferences/previous/ottawa/en/ writing. (2008). In Merriam-Webster Online Dictionary. Retrieved December 6, 2008, from http://www.merriam-webster.com/dictionary/writing Xiao, J., Tang, C., & Shim, S. (2009). Acting for happiness: Financial behavior and life satisfaction of college students. Social Indicators Research, 92(1), 53-68. Xiaokun, L. (2002). Quality of life among mastectomy patients receiving radiotherapy. Contemporary Nurse: A Journal for the Australian Nursing Profession, 13(2/3), 198- 208. Zakowski, S. G., Ramati, A., Morton, C., Johnson, P., & Flanigan, R. (2004). Written emotional disclosure buffers the effects of social constraints on distress among cancer patients. Health Psychology, 23(6), 555-563. Zimmerman, M.A. (1995). Psychological empowerment: Issues and illustrations. American Journal of Community Psychology, 23(5), 581-599. Zullig, K., Huebner, E., & Pun, S. (2009). Demographic correlates of domain-based life satisfaction reports of college students. Journal of Happiness Studies, 10(2), 229-238. 123 APPENDIX A EAST ALABAMA MEDICAL CENTER INSTITUTIONAL REVIEW BOARD APPROVAL 124 125 APPENDIX B MONTGOMERY CANCER CENTER INSTITUTIONAL REVIEW BOARD APPROVAL 126 127 128 129 APPENDIX C AUBURN UNIVERSITY INSTITUTIONAL REVIEW BOARD APPROVAL 130 131 132 133 134 135 136 137 138 139 140 141 APPENDIX D STUDY FLIER 142 143 APPENDIX E RECRUITMENT SCRIPT 144 RECRUITMENT SCRIPT My name is Naomi Crouse, and I am a graduate student in the Department of Nursing at Auburn University. I would like to invite you to participate in my research study. I want to see if writing makes a difference in your feelings of uncertainty and perceived quality of life. You may participate if you are over 19, newly diagnosed with cancer, just starting your treatment phase, not pregnant or a prisoner, and able to complete a written or typed journal. Please do not participate if you do not meet the inclusion criteria, cannot give legal consent, and cannot speak or read English. As a participant, you may be asked to write in a journal twice a week for three months. You will be asked to fill out three survey tools when you enter the study and again at the end of three months. I will not read any journals nor will I ask for them at the end of the study. The information gathered from the survey tools will be analyzed and written up in my master?s thesis, and perhaps later in a publication or presentation. Nothing will be said or written that can be linked to you. Risks of being in this study include experiencing emotional distress, sadness or anger as you write about emotional issues of your choice. Benefits may include feeling less uncertain and distressed, and perceiving a better quality of life. Compensation includes being given a journal, and there will be no costs encountered by you if you agree to participate. If you would like to participate in this research study, please request a consent form and read through it. Do you have any questions now? If you have questions later, please contact me at crousnl@auburn.edu or (334) 821-8618 or you may contact my advisor, Dr. Anita All, at (334) 844-6135 or aca0001@auburn.edu 145 APPENDIX F EAST ALABAMA MEDICAL CENTER INFORMED CONSENT 146 147 148 APPENDIX G MONTGOMERY CANCER CENTER INFORMED CONSENT 149 150 151 APPENDIX H PERSONAL ADDRESS SHEET 152 153 APPENDIX I DEMOGRAPHIC SURVEY 154 155 APPENDIX J MISHEL UNCERTAINTY IN ILLNESS SCALE ? COMMUNITY FORM 156 157 158 159 160 APPENDIX K FERRANS AND POWERS QUALITY OF LIFE INDEX ? CANCER VERSION III 161 162 163 164 165 APPENDIX L WRITING INFORMATION SHEET 166 167 APPENDIX M AUBURN PSYCHOLOGICAL REFERRAL LIST 168 169 APPENDIX N MONTGOMERY PSYCHOLOGICAL REFERRAL LIST 170 171 APPENDIX O EXPRESSIVE WRITING INFORMATION SHEET 172