Sleep in the elderly cancer patient

Sleep quality disturbances are reported in 15% to 30% of the U.S. adult population, including half of those aged ? 65. Sleep disruptions in individuals with cancer have been reported clinically but are yet to be verified in the literature. Sixty percent of all cancer occurs in persons aged ? 65. Because elderly persons are more likely to report sleep disruptions and more likely to be diagnosed with cancer, this population may be more vulnerable to sleep disruptions than the general oncology population. The purpose of this study was to determine if there are differences in sleep quantity and sleep quality between individuals aged < 65 and those aged ? 65 and to determine if demographic and clinical factors affect sleep quality in the elderly cancer population. Patterns of sleep disruption in the elderly population were also reported. Secondary analysis was conducted on data from a descriptive correlational study on sleep, pain, and fatigue in cancer patients. The Pittsburgh Sleep Quality Index (PSQI) was used to determine perceived sleep disruption, use of medication, and daytime dysfunction. Fifty-seven of the 214 participants in the study were ? 65. No differences were found in sleep quantity or overall sleep quality between the older and younger groups. Sleep duration, a categorical measurement of sleep quantity, was significantly different between groups(X2 = 14.339, p = <c.002), with those aged < 65 showing shorter sleep duration. No associations were found between sleep quality and demographic factors in those aged ? 65. Stage of disease in this age group was the only clinical factor have significance to sleep quality (x=X2 = 8.491, p < .014), with regional and advanced disease more problematic for sleep than localized disease. Comparisons between those aged 65 to 74 and those ? 75 showed no differences in sleep quality. Nearly 60% of older persons reported awakening at least three times during the past week. Use of the bathroom was the most common reason cited for nighttime awakening in the elderly population, followed by coughing or snoring and then pain. It is recommended that further research be conducted to verify sleep disturbances in oncology patients. Those with advance stage disease should be identified as most likely to report sleep disturbances in the oncology population and should be targeted for education about sleep hygiene and for interventional studies.

SLEEP IN THE ELDERLY CANCER PATIENT by Marsha Tadje A thesis submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Master of Science College of Nursing The University of Utah August 1999 Copyright C Marsha Tadje 1999 All Rights Reserved THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a thesis submitted by Marsha T. Tadje This thesis has been read by each member of the following supervisory committee and by majority vote has been found to be satisfactory. Chair: Susan L. Beck Rosemary B. Field 1- /z- Cfq SY-edi£L& U . {2/cit CUtOLa(]'~ f Stephanie J. Richardson THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPRO V AL To the Graduate Council of the University of Utah: I have read the thesis of Marsha T. Tadje in its final form and have found that (1) its format, citations, and bibliographic style are consistent and acceptable; (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final manuscript is satisfactory to the supervisory connnittee and is ready for submission to The Graduate School. Susan L. Beck Chair, Supervisory Committee Approved for the Major Department Chair/Dean Approved for the Graduate Council ~~ <; ,c..Q..~--- David S. Chapman Dean of The Graduate School ABSTRACT Sleep quality disturbances are reported in 15% to 30% of the U.S. adult population, including half of those aged > 65. Sleep disruptions in individuals with cancer have been reported clinically but are yet to be verified in the literature. Sixty percent of all cancers occur in persons aged :2: 65. Because elderly persons are more likely to report sleep disruptions and are more likely to be diagnosed with cancer, this population may be more vulnerable to sleep disruptions than the general oncology' population. The purpose of this study was to determine if there are differences in sleep quantity and sleep quality between individuals aged < 65 and those aged :2: 65 and to determine if demographic and clinical factors affect sleep quality in the elderly cancer population. Patterns of sleep disruption in the elderly population were also reported. Secondary analysis was conducted on data from a descriptive correlational study on sleep, pain, and fatigue in cancer patients. The Pittsburgh Sleep Quality Index (PSQI) was used to determine perceived sleep quality, sleep latency, sleep efficiency, sleep duration, sleep disruption, use of medication, and daytime dysfunction. Fifty-seven of the 214 participants in the study were > 65. No differences were found in sleep quantity or overall sleep quality between the older and younger groups. Sleep duration, a categorical measurement of sleep quantity, was significantly different between groups (X2 = 14.339, n < .002), with those aged < 65 showing shorter sleep duration. No associations were found between sleep quality and demographic factors in those aged ;;:: 65. Stage of disease in this age group was the only clinical factor having significance to sleep quality (X2 = 8.491, 12 < .014), with regional and advanced disease more problematic for sleep than localized disease. Comparisons between those aged 65 to 74 and those aged ;;:: 75 showed no differences in sleep quantity or qUality. Nearly 60 % of older persons reported awakening at least three times during the past week. Use of the bathroom was the most common reason cited for nighttime awakening in the elderly population, followed by coughing or snoring and then pain. It is recommended that further research be conducted to verify sleep disturbances in oncology patients. Those with advanced stage disease should be identified as most likely to report sleep disturbances in the oncology population and should be targeted for education about sleep hygiene and for interventional studies. v TO ROB, MY "SOUL MATE," AND ERIN, LISA, AND LINN THANKS FOR YOUR SUPPORT! TABLE OF CONTENTS Page ABSTRACT ......................................... iv LIST OF TABLES ..................................... x Chapter I. INTRODUCTION................................. 1 Problem ........................................ 1 Sleep Disturbances in Elderly Populations ................... 1 Cancer in Elderly Populations .. ~ . . . . . . . . . . . . . . . . . . . . . . . 2 Sleep Disturbance in Cancer Patients ...................... 3 Significance for Nursing .............................. 4 Cellular and Symptom Effects .......................... 5 Quality of Life ................................... 5 Status of Research ................................. 7 Purpose ........................................ 8 Research Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 DeflIlitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 II. LITERATURE REVIEW ............................ 13 Sleep Theory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 14 Normal Sleep ................................... 16 Normal Wakefulness . . . . . . . . . . . . . . . . . . . . . . . . . .. 17 Normal Nonrapid Eye Movement Sleep .......... . . . .. 17 Normal Rapid Eye Movement Sleep ................. 18 Mechanisms of Sleep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 18 The Physiology of Sleep ........................ 19 The Biochemical Basis of Sleep . . . . . . . . . . . . . . . . . . .. 20 Circadian Rhythms . . . . . . . . . . . . . . . . . . . . . . . . . . .. 23 Manifestations of Sleep ............................. 24 Manifestations of Wakefulness . . . . . . . . . . . . . . . . . . . .. 24 Manifestations of Stage I Sleep .................... 25 Manifestations of Stage II Sleep . . . . . . . . . . . . . . . . . . .. 25 Manifestations of Delta Sleep ..................... 25 Chapter Page Manifestations of Rapid Eye Movement Sleep ........... 26 Normal Sleep Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 26 The Purpose of Sleep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 27 Sleep Disturbances ................................ 30 Insomnia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 30 Disorders of Excessive Daytime Somnolence . . . . . . . . . . .. 31 Disorders of the Sleep-Wake Schedule . . . . . . . . . . . . . . .. 32 Sleep in Elderly Populations .......................... 33 Sleep in Cancer Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 37 Sleep in Elderly Persons With Cancer . . . . . . . . . . . . . . . . . . . .. 41 Conceptual Framework ............................. 42 Age and Sleep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 44 Gender and Sleep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 45 Clinical Variables and Sleep .......................... 47 Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 47 Stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 48 Hospitalization .............................. 49 III. METHODOLOGy ................................ 51 Design ....................................... 51 Sample ....................................... 51 Setting ....................................... 52 Instrument ..................................... 52 Procedure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 53 Data Analysis ................................... 54 Sleep Quantity and Overall Sleep Quality . . . . . . . . . . . . . . . . . .. 54 Demographic and Clinical Factors . . . . . . . . . . . . . . . . . . . . . .. 55 Sleep Components ................................ 55 Subgroup Analysis ................................ 55 IV. RESULTS ..................................... 56 Sample Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 56 Comparisons of Sleep Quantity . . . . . . . . . . . . . . . . . . . . . . . .. 58 Comparisons of Sleep Quality ......................... 59 Demographic Comparisons in Those Aged More Than Sixty-Five. . .. 60 Gender Differences ........................... 60 Cultural Differences . . . . . . . . . . . . . . . . . . . . . . . . . .. 60 Marital Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 61 Socioeconomic Status .......................... 62 Educational Status ............................ 62 viii Chapter Page Comparison of Clinical Factors. . . . . . . . . . . . . . . . . . . . . . . .. 64 Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 64 Stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 64 Treatment Modality ........................... 66 Hospitalization Status .......................... 68 Responses To the Components of Sleep Quality . . . . . . . . . . . . . .. 68 Perceived Sleep Quality . . . . . . . . . . . . . . . . . . . . . . . .. 68 Sleep Latency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 68 Sleep Efficiency ............................. 68 Sleep Disturbances . . . . . . . . . . . . . . . . . . . . . . . . . . .. 68 Use of Medication . . . . . . . . . . . . . . . . . . . . . . . . . . .. 70 Daytime Dysfunction .......................... 70 Differences in Aging Subgroups .... . . . . . . . . . . . . . . . . . . .. 73 V. DISCUSSION................................... 75 Nursing Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 81 Recommendations for Further Research . . . . . . . . . . . . . . . . . . .. 82 Recommendations· for the Questiorinaire . . . . . . . . . . . . . .. 84 Recommendations for Recruitment .................. 85 Implications for Theory ............................. 86 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 87 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 89 ix LIST OF TABLES Table Page 1. Demographic Variables for < 65 and ~ 65 Age Groups . . . . . . . .. 57 2. Comparison of Sleep Duration in < 65 and > Age Groups ....... 59 3. Gender and Sleep Status in > 65 Age Group . . . . . . . . . . . . . . .. 61 4. Marital and Sleep Status in ~ 65 Age Group . . . . . . . . . . . . . . .. 63 5. Income and Sleep Status in ~ 65 Age Group . . . . . . . . . . . . . . .. 63 6. Education and Sleep Status in ~ 65 Age Group .............. 65 7. Site of Disease and Sleep Status in > 65 Age Group ........... 65 8. Stage of Disease and Sleep Status in ~ 65 Age Group .......... 66 9. Treatment Modality and Sleep Status in > 65 Age Group ........ 67 10. Sleep Quality in ~ 65 Age Group . . . . . . . . . . . . . . . . . . . . . .. 69 11. Sleep Latency in ~ 65 Age Group ...................... 69 12. Habitual Sleep Efficiency in ~ 65 Age Group ............... 69 13. Frequencies of Sleep Disturbances in ~ 65 Age Group . . . . . . . . .. 71 14. Use of Sleeping Medication in > 65 Age Group . . . . . . . . . . . . .. 72 15. Enthusiasm in ~ 65 Age Group . . . . . . . . . . . . . . . . . . . . . . .. 72 16. Difficulty Staying Awake During Day in > 65 Age Group. . . . . . .. 72 17. Daytime Dysfunction in ~ 65 Age Group . . . . . . . . . . . . . . . . .. 73 18. Comparison of Sleep Quantity and Quality in Elderly Subgroups .... 74 CHAPTER I INTRODUCTION Problem Epidemiological surveys indicate that from 15 % to 35 % of the adult population complain of frequent sleep quality disturbance such as difficulty falling asleep or difficulty maintaining sleep (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989; Kales, Kales, & Soldatos, 1982; Mant & Eyland, 1988; Walsleben, 1982). These disturbances are often associated with situational stress, illness, aging, and drug treatment. Half of the 19 million Americans aged ;a: 65 experience poor sleep and often report disturbances such as insomnia, excessive daytime sleepiness, circadian rhythm disruptions, and sleep-related disorders (Dowling, 1995). Sleep Disturbances in Elderly Populations Complaints of sleep disturbances such as difficulty initiating and maintaining sleep and daytime drowsiness are more common in elderly populations than in any other age group and are recognized as being among the most common health problems of the elderly. Asplund (1996) found that there is an age-related increase in daytime sleepiness and napping despite a good night's sleep. This research found that individuals suffering from poor health, with diseases such as 2 diabetes, heart disease, or arthritis, were more often troubled by sleepiness in the daytime. The number and duration of nocturnal awakenings in these individuals were associated with the amount of daytime sleepiness. The use of hypnotics was more prevalent among elderly populations, but their use did not prevent these individuals from the effects of daytime sleepiness. Klink, Quan, Kaltenborn, and Lebowitz (1992) showed that insomnia was reliably predicted by increasing age and health problems, suggesting that elderly adults with less than excellent health may be at an increased risk for sleep disturbances. These results were also confrrmed by other studies identifying deteriorating health as more likely to affect sleep quality than age (Reyner & Home, 1994; Schmitt, Phillips, Cook, Berry, & Wekstein, 1996), placing the elderly oncology patients at risk by virtue of their diagnosis. Cancer in Elderly Populations The elderly population in the United States has been steadily increasing during the last century in absolute numbers and in percentage of the total national population. Current projections estimate that one in five Americans will be 65 or older by the year 2033 (Trimble et al., 1994); that is, the aging population will double in less than 40 years. Unfortunately, malignant disease, especially solid tumors, occurs more frequently in this subset of the population. Current calculations suggest that approximately 60 % of all cancers occur in individuals aged > 65 (Trimble et al., 1994). This percentage is expected to increase significantly as the U.S. population ages. In addition, 11 % to 12% of persons 3 aged ~ 70 have a history of prior malignancy, whereas the general population rate with prior malignancy is only 2% (Byrne & Carney, 1993). Sleep Disturbance in Cancer Patients Sleep disturbance in the oncology population is likely to be a multifactorial problem. Beszterczeyand Lipowski (1977) found that approximately 45% of the cancer patients seen for radiotherapy slept less than 50 hours per week, and 23 % of these same patients slept less than 40 hours per week. Physical illness, pain, hospitalization, and the psychological impact of malignant disease may disrupt the sleeping patterns of individuals with cancer. In tum, poor sleep adversely affects daytime mood and performance. In the general population, insomnia has been associated with an increased risk of developing anxiety or depression, which may further contribute to insomnia (Kales et aI., 1982). Payne and Massie (1997) reported that individuals with a new diagnosis of cancer or a relapse of the disease might show mixed symptoms of anxiety and depression, irritability, and disruption of sleep. Contrary to expectations, Beszterczey and Lipowski (1977) found that insomnia was positively correlated with anxiety and depression but not with pain. These studies indicate the need for more comprehensive studies of the prevalence and severity of sleep disturbances in various categories of oncology populations. Relative cancer survival rates for virtually every major site are improving (Cunningham, 1997). As cancer changes from an acute to a chronic disease, sleep disturbances related to the effects of the disease are likely to become more prevalent in the surviving oncology population. Because cancer is mainly a disease of aging, elderly populations may be those most influenced by the sleep disturbances related to the disease process. Sienificance for Nursina: 4 Sleep disturbance over prolonged periods, along with other health disruptions, can have an impact on cellular and immune functions, symptom severity, and memory and cognitive functions. They can make affected individuals vulnerable to accidents or injury and can interfere with productivity and socialization. Disruptions to sleep may reflect disruptions to circadian rhythms. These disruptions not only have a detrimental effect on daytime functioning and quality of life but also have the potential to interfere with the delivery of treatment protocols that are influenced by circadian oscillations. Sleep disturbances may suggest high levels of stress or anxiety or poor adaptation to diagnosis and treatment. Oncology nurses can play an important role in the identification of susceptible individuals at vulnerable times and provide education to high-risk individuals about the management of sleep disturbances. Oncology nurses can contribute to interventional studies in order to identify the appropriate management of sleep problems in oncology populations. It is essential that all members of the health care team be familiar with the evaluation of sleep-related complaints and creative management strategies. That is, there is likely an important relationship among sleep, quality of life, and possibly even immune function (Yellen & Dyonzak, 1996). Cellular and Symptom Effects Sustained sleep disturbances may have an effect on cellular function. Interleukin-1 activity is dramatically increased during sleep and is associated with the onset of slow-wave sleep (Moldofsky, Lue, Eisen, Keystone, & Gorczynski, 1986). According to Palmblad, Petrini, Wasserman, and Akerstedt (1979), deprivation of sleep for 48-hour time periods caused a reduction in deoxyribonucleic acid (DNA) synthesis, suggesting alterations in cell-mediated immune reactions. The reduction of natural killer cell and lymphokine-activated killer cell numbers and activity is also associated with sleep deprivation (Irwin et al., 1994; Irwin et al., 1996). This effect could be significant in many cancer patients since immunity may already be compromised from the effects of the disease, treatment modalities, or both. A weakened immune system may make these individuals even more vulnerable to infections or, possibly, to relapse of the disease as the immune system does surveillance on "nonself" cells. In addition, sleep disturbances may also aggravate symptom severity, making cancer patients with sustained sleep disturbances less tolerant to pain, fatigue, and nausea. Quality of Life Quality of life refers to those aspects of life and human function considered essential for living fully. With the realization that mere survival is often an inappropriate goal in a society dominated by chronic disease, quality-of-life considerations have taken center stage in decisions about treatment (Mor, Allen, & Malin, 1994). 5 Myers and Badia (1995) reported a positive correlation between nighttime sleep quality in aging and cognitive performance. Other research indicates a decrease in sleep quality with aging. In general, elderly persons spend more time in bed but less time asleep. They experience an increased amount of wakefulness after sleep onset, more Stage I sleep, and less Stages III and IV sleep (Dement, Miles, & Carskadon, 1982; Myers & Badia, 1995; Robinson, 1993). The decreases in Stages III and IV may be especially important since it is believed that these stages are necessary for cerebral restitution (Adam, 1980; Myers & Badia, 1995; Oswald, 1980; Schoessler, Ludwig-Beymer, & Huether, 1990). Asplund (1996) reported that elderly individuals who were awakened frequently during the night due to symptoms of chronic disease such as pain and nocturia were more likely to suffer from daytime sleepiness than healthy asymptomatic elderly individuals. Changes in daytime alertness and cognitive performance due to sleep disturbance may place older populations at risk for accidents and injury. 6 Fatigue due to prolonged sleep disturbance may influence the quality of life in cancer patients. Reports in the literature suggest that fatigue decreases activity across all aspects of daily life, including physical and social activity, interpersonal interaction, recreational activity, and home and family care. Information reception and processing may also be influenced. In fact, affected individuals may believe that more energy is required for these types of tasks after diagnosis and treatment (Nail & Jones, 1995). Fatigue is often an outcome of disrupted sleep that is especially noted by those who have trouble initiating or maintaining sleep (Dement et aI., 1982; Robinson, 1993). Dement and colleagues (1982) reported that daytime sleepiness is a reflection of the quality of sleep at night. Status of Research 7 Individuals working in oncology recognize that the symptoms associated with cancer and its treatment often interfere with the ability to sleep throughout the night. However, documentation to validate the phenomenon of sleep disturbances in cancer patients is paltry at best. In addition, there are currently no published reports describing sleep in elderly persons with cancer. There is evidence that the sleep patterns of elderly adults are different than those of younger adults, especially in the chronically ill. However, it is not known if sleep in elderly cancer patients differs from that in younger cancer patients. It is also not known if gender, culture, marital, or economic status influence sleep in these individuals. Treatment modalities, medications, disease site, disease stage, and hospitalization may influence sleep quality in elderly persons with cancer as well, but little has been done to document it. Baseline research may help to identify individuals at risk in the hope that interventions and education about sleep disturbances will be initiated in a timely manner. Research will continue to contribute to one's awareness of sleep disturbances as side effects of cancer and its treatment, and it may also provide clues about the etiology and treatment. Understanding sleep in cancer patients and designing effective interventions for sleep disturbances are important in developing practices that will improve the patients' quality of life. 8 Purpose From a clinical standpoint, one would expect to fmd sleep disturbances a common complaint among persons with cancer. They experience many symptoms related to disease and treatment that might interfere with sleep. Certain medications used in treatment may affect circadian cycles. In addition, cancer fatigue may have implications for sleep. Sleep disturbances in this population may take the form of insomnia at nighttime or daytime sleepiness, which are also known as disorders of excessive somnolence. Many factors may contribute to sleep difficulties in cancer patients, including disease site and treatment modality, gender, culture, socioeconomic status, and available support systems. Until these conditions are understood, little can be done to alleviate or attenuate the problem. Because healthy elderly persons frequently report disturbances in sleep, it is likely that elderly cancer patients may be affected significantly by sleep disturbances. The purpose of this research was to describe sleep length and its perceived quality in elderly persons with cancer and to examine factors related to sleep disturbances in these individuals. It was hoped that the components of sleep that might be related to sleep disturbances specific to the elderly person with cancer would also be identified. Research Questions 1. Does nocturnal sleep quantity in elderly persons (aged ~ 65) with cancer differ from sleep quantity in younger adults (aged < 65) with cancer? 2. Is the overall sleep quality (Pittsburgh Sleep Quality Index [PSQI] global score) in cancer patients aged < 65 different than for those aged > 65? 3. Is there an association between demographic factors (gender, culture, socioeconomic, and marital status) and sleep status (good versus poor sleepers) in those aged > 65? 9 4. Is there an association between clinical factors (site and stage of disease, treatment modality, and hospitalization) and sleep status (good versus poor sleepers) in those aged ;;::: 65? 5. What are the elderly cancer patients' patterns of response to the components of sleep quality identified in the PSQI (sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbance, use of medication, and daytime dysfunction)? 6. Do differences in sleep quantity, quality, or both occur for different age groups of elderly cancer patients (young-old persons, 65 to 74; middle-old persons, 75 to 84; and old-old persons, 85 +)? Dermitions For the purpose of this study, the following defInitions were used: Sleep is a conlplex physiopsychic state involving the dimming of consciousness, relaxation of skeletal muscles, and temporary changes in sensory­motor functions (Lamb, 1982) "during which the powers of the body are restored" (Richardson, 1986, p. 66). Carskadon and Dement (1994) defined sleep as a reversible behavior state of perceptual disengagement from and unresponsiveness to the environment usually (but not necessarily) accompanied by postural recumbency, quiescence, closed eyes, and all the other indicators one commonly associates with sleeping. (p. 16) Nocturnal sleep quantity is the estimated amount of time spent sleeping during the night (Pacini & Fitzpatrick, 1982). 10 Sleep latency is the length of time required to fall asleep after going to bed or turning the lights off (subjective) or the alert stage to Stage I sleep (objective)-usually lasting 5 to 15 minutes (Dement et al., 1982; Knab & Engel- Sittenfeld, 1983; Pacini & Fitzpatrick, 1982; Richardson, 1997; Robinson, 1993). Sleep efficiency is the percentage of time actually asleep while in bed: time asleep/time in bed (Buysse et aI., 1989; Dement et al., 1982; Morewitz, 1988; Silberfarb, Hauri, Oxman, & Schnurr, 1993), Sleep disturbance is an interruption in the sleep cycle such as when awakening occurs. Disruptions may be a result of anxiety, noises, dreams, nocturia, pain, snoring, pruritus, sleep apnea or dyspnea, or temperature fluctuations (Buysse et al., 1989; Yellen & Dyonzak, 1996). Sleep pattern disturbance is the state in which the individual experiences or is at risk of experiencing a change in the quantity or quality of rest pattern as related to biological and emotional needs (Carpenito, 1989). Sleep duration is the amount of actual sleep received during the night in hours, not including time awake while in bed (Buysse et al., 1989). Sleep maintenance is the ability to sustain sleep throughout the night without awakening. The use of medication specifically assists one in falling asleep (Buysse et al., 1989) or in maintaining sleep throughout the night or whose side effects are known to promote sleep_ Daytime dysfunction is the lack of subjective energy available to perform usual daily tasks such as eating, working, driving, or socializing and the reported difficulty in staying awake to perform such tasks (Buysse et al., 1989). 11 Sleep quality score is a subjective measurement that encompasses quantitative· aspects of sleep such as sleep duration, sleep latency, and number of arousals, as well as more subjective aspects such as the depth or perceived restfulness of sleep. The PSQI asks participants to rate their sleep during the past week as very good, fairly good, fairly bad, or very bad (Buysse et aI., 1989). The PSQI global score is a score computed from the sum of the components of sleep studied in the questionnaire, including sleep latency, sleep duration, sleep efficiency, sleep disturbance, use of sleeping aids, and daytime dysfunction (Buysse et aI., 1989). A review of the literature indicates that physiological age is more important than chronological age in determining sleep disturbances and daytime sleepiness (Asplund, 1996; Bliwise, 1993; Myers & Badia, 1995; Schmitt et al., 1996). Most demographic studies of sleep in elderly populations consist of subjects aged ;;:: 60 years. For the purpose of this study, the elderly population consisted of adults aged > 65 years. This cohort was further categorized into the young-old (65 to 74), the middle-old (75 to 84), and the old-old (85+). 12 CHAPTERn LITERATURE REVIEW Physiology textbooks were referenced in order to defme and validate the sleep process. A search of the terms "sleep and aging," "sleep and neoplasms," and "sleep disorders" was conducted on medical and nursing literature databases. Searches were limited to English-speaking journals and to research conducted on human subjects. Sou~ces older than 1990 were not searched, but older references from select articles were researched when applicable. Sleep disorders were also researched on the National Cancer Institute on-line database. Information about normal sleep was drawn from Bahr (1984), Carskadon and Dement (1994), Chuman (1983), Leo and Huether (1998), Ludwig-Beymer, Huether, and Schoessler (1994), Richardson (1986, 1997), Robinson (1993), Schoessler and colleagues (1990), Shaver and Landis (1994), and Shaver and Landis (1995). Information about sleep in elderly populations was taken from Bahr (1984), Bliwise (1993), Carskadon and Dement (1994), Dement and colleagues (1982), Dowling (1995), Morewitz (1988), Myers and Badia (1995) and Robinson (1993). The theory of sleep was searched using "sleep theory," "sleep and repair," and "sleep and restoration." Information for sleep theory was largely drawn from Adam (1980), Adam and Oswald (1983), Oswald (1980), Rechtschaffen (1998), Richardson (1996), and Landis and Whitney (1997). 14 A review of the literature provided numerous choices in terminology to describe problems associated with sleep. Included in those choices were "sleep quality," "sleep quantity," "insomnia," "sleep disorders," "sleep maintenance, " "sleep disruption," and "sleep disturbance." In order to maintain consistency throughout this thesis, the term "sleep disturbance" was selected to describe the phenomenon in cancer patients, since their sleep is likely to be disturbed by symptoms associated with disease and treatment. Likewise, the terms "excessive daytime somnolence," "daytime napping," "hypersomnolence," and "fatigue" describe a similar phenomenon in different populations. The term "fatigue" was selected to describe the sequelae of sleep disturbances, since it is most consistently associated with the oncology literature. However, in order to maintain validity of the research, terminology of cited text is consistent with the authors' reports. Sleep Theory The functions of sleep have been proposed as restorative, protective, instinctive, or adaptive. Sleep may serve as a means to conserve energy or to adjust biological systenls periodically. Sleep has even been prescribed as a folk remedy for illness. It is also seen as a cyclical suspension of consciousness that generally occurs at night. Throughout the ages, sleep has been an important behavioral state in preserving personal safety and an adaptation to the environment as night ensued. Rechtschaffen (1998) recently argued that sleep is important enough that it "ultimately enhances surviVal." The purpose of sleep is seen mainly as a restorative or a reparative process, according to current nursing literature. 15 Because certain physiological correlates of sleep are believed to aid healing (Adam & Oswald, 1983), sleep may be necessary for tissue repair. Increased peripheral cell mitosis has been associated with nighttime sleep, and increased protein synthesis has been observed with deep sleep and REM sleep (Adam, 1980; Adam & Oswald, 1983; Oswald, 1980). In the daily cycle, the greatest amount of growth hormone is released during sleep, and disrupting, delaying, or prohibiting sleep reduces circulating levels (Sassin et al., 1969). Growth hormone exerts its anabolic effects indirectly by stimulating the liver to produce growth factors or large peptides that mediate growth (Lee & Stotts, 1990). Systemic or the local administration of growth hormone can significantly increase both wound strength (Hollander, Devereux, Marafmo, & Hoppe, 1984) and the amount of granulation tissue in the wound (Steenfos & Jansson, 1992). Reduction in growth hormone secretion accompanying sleep loss may reduce the anabolic effects of promoting cellular proliferation and protein synthesis necessary to repair the effects of normal living on cellular and immune functions. Sleep loss is associated with altered thermoregulation and hypothermia, and it can lead to increased food intake, increased energy expenditure, and altered metabolism (Landis & Whitney, 1997). These researchers also indicated that sleep loss is associated with increased heart rate and peripheral norepinephrine levels indicative of sympathetic nervous system activation. Within the restorative framework, wake-sleep cycles are defmed according to the use of energy in catabolic-anabolic and recharge processes. High amounts of energy are expended 16 over the course of a day in carrying out activities necessary for survival and maintaining one's place in the world. When one sleeps, energy is directed inwardly and used to restore body functions (Oswald, 1980; Richardson, 1996) or to recharge neural-cerebral units (Adam, 1980; Richardson, 1996). The inertia and unresponsiveness of sleep actually mask active and predominantly anabolic processes. Health is, in large part, contingent on how effectively the balance is struck between the functions of wakefulness-activity and the functions of sleep­restoration. According to Richardson, these functions are bound in reciprocal arrangements. Restorative theorists of sleep have proposed that sleep is to offset the metabolic costs of maintaining a constant body temperature in mammals (Walker & Berger, 1980) or to force inactivity and "balance the energy budget" (Zepelin, 1994). Taylor, Rogers, and Driver (1997) stated that sleep is a compensatory mechanism following the catabolic process of daytime activity. Theorists of the restorative process of sleep view sleep as an integral component to sustaining the functioning of the body and mind, and they believe the use of the body and mind induces the demand for sleep. Normal Sleep Human beings alternate between a wakeful and a sleeping state. Sleep itself can be divided into two main parts: (a) rapid eye movement (REM) and (b) non­REM (NREM) sleep. Thus, there is alternation between three states: (a) wakefulness, (b) NREM sleep, and (c) REM sleep. Each phase has distinctive characteristics. 17 Normal Wakefulness Wakefulness is characterized by spontaneous, low-voltage, random, and fast electrical activity on an electroencephalogram (EEG). The pattern is desynchronized. The level of muscle activity is usually high. During periods of alertness, only REMs are recorded on an EEG, and slower REMs are associated with a decrease in the level of alertness. An alert individual has the capacity to respond to a multitude of internal and external stimuli, and varied behavioral and physiological responses are possible. Normal Nonrapid Eye Movement Sleep NREM sleep is divided into four stages, each showing decreased electrical activity. Stages III and IV are characterized by slow-wave sleep. During non­REM sleep, an EEG response to stimuli is reduced, explaining decreased cerebral responsiveness. Muscle tone and activity are diminished on an EEG, and oxygen consumption in the muscles decreases. During NREM sleep, there is a decrease in body temperature, blood pressure, cardiac output and heart rate, and heart rate variability. The lowest blood pressure levels occur during deep non-REM sleep. Ventilatory changes occur as well, with decreases in minute ventilation and slight increases in tidal volume. Cerebral blood flow decreases during NREM sleep, affecting primarily the brainstem and cerebellum in Stages I and II and the cortex during Stages III and IV (Leo & Huether, 1998). There appears to be decreased cerebral vasomotor carbon dioxide responsiveness during NREM sleep that causes a slight elevation in the PaC02 (Robinson, 1993). 18 Normal Rapid Eye Movement Sleep Normal REM sleep is characterized by REMs, muscle atonia, low voltage, fast, and desynchronized cortical activity. This is the period of sleep when most dreaming occurs. REM sleep is sometimes referred to as paradoxical sleep, since there is rapid EEG activity, which is usually associated with wakefulness. There is also diminished muscle tone, which is usually associated with non-REM sleep, although the atonia of REM sleep is different from the relaxation of non-REM sleep. There is inhibition of the spinal neurons and diminished or absent deep tendon reflexes. Eye movements are rapid, and there may be episodes of muscle twitching. The pupils dilate and constrict in phases. Vital signs are variable, and there is increased consumption of oxygen. During REM sleep, regional cerebral blood flow increases, but responses to hypercapnia and hypoxia vary (Robinson, 1993). Irregular breathing is associated with REM sleep, although minute ventilation, tidal volume, and PaC02 approach waking levels. Brain synthesis, which is believed to organize and store information, occurs during REM sleep and may indicate the importance of sleep on memory (Schoessler et al., 1990). Mechanisms of Sleep Although much research has been conducted on sleep architecture, the functional and structural mechanisms of sleep are not fully understood. Many theories have been developed in an effort to explain sleep phenomena, but no one theory answers all questions or eliminates all controversies. Certain aspects of sleep have been well-studied, however, and are discussed in the following sections. 19 The Physiolo&y of Sleep The state of wakefulness requires intact cerebral hemispheres to interact with the thalamus, hypothalamus, and brainstem. Consciousness is mediated by the reticular activating system (RAS), located in the central core of the brainstem. The RAS is activated by many areas of the cortex, including the limbic system, which may explain why emotional status can affect the sleep process. Ascending sensory fibers serve to stimulate the RAS. Stimuli such as light or noise, which ascend these sensory fibers, are translated into neural transmissions that synapse onto the RAS and maintain or prolong consciousness. Stimulation of the ascending RAS results in prompt EEG stimulation and arousal. Destruction of the ascending RAS results in slow, synchronized EEG waves and coma, which is not reversed by sensory stimulation. An activated RAS is required for wakefulness. Since activation of the ascending RAS results in wakefulness, one would believe that sleep must be the process of deactivation of the ascending RAS. However, decreased RAS activity does not automatically cause sleep to occur (Chuman, 1983). Sleep is identified as an active process that is stimulated by its own neurotransmitters (Leo & Huether, 1998; Schoessler et al., 1990). While wakefulness is maintained by nerve cell activity of the reticular structures of the upper brainstem and the posterior hypothalamus, hypnogenic or sleep-producing areas appear to be located in the thalamus, hypothalamus, midbrain raphe, cortex, and locus ceruleus. The two systems for wakefulness and sleep are interrelated. They are antagonistic systems and are influenced by 20 neuronal input from the central and peripheral nervous system. The reticular formation is primarily responsible for generating REM sleep, and projections from the reticular formation and other areas of the mesencephalon and brainstem produce non-REM sleep. The Biochemical Basis of Sleep Sleep is a biochemical process, as well as a physiologic one. Biochemical substances assist with the regulation of sleep, and there is no single chemical that accounts for all sleep phenomena nor is it likely that all chemicals involved in the process have been identified. Differing theories attempt to explain how chemical transmitters alter the excitability of postsynaptic nerve cells and how neurons are stimulated or inhibited by these transmitters. Other chemical substances have a more general effect by altering or modulating the metabolic activity of neurons. The chemical substances known to have a function in the regulation of sleep and wakefulness include catecholamines (epinephrine, norepinephrine, and dopamine); acetylcholine; serotonin; histamine; L-tryptophan; prostaglandins; and adenosine (Leo & Huether, 1998; Robinson, 1993). Melatonin plays a role in circadian rhythms (Bliwise, 1993; Brzezinski, 1997; Myers & Badia, 1995). Growth hormone is associated with the initiation of sleep, and cortisol levels rise in the morning as awakening occurs. Acetylcholine and somatostatin playa role in the transition of one stage of sleep to another (Leo & Huether, 1998). Serotonin. Serotonin appears to be involved with sleep induction and the maintenance of non-REM sleep. The median raphe nuclei in the medulla, pons, 21 and midbrain secrete serotonin in a cyclical manner that is part of one's circadian rhythm. It is believed that the raphe system initiates sleep onset, maintains non­REM, and primes the brain for transition to REM sleep. The onset of REM sleep is linked with the secretion of norepinephrine in the locus ceruleus. Serotonin inhibits the RAS. The RAS then becomes inactive and ceases to signal the cerebrum, and sleep ensues (Chuman, 1983; Richardson, 1986; Robinson, 1993). An afferent neural pathway originating in the raphe nuclei of the brainstem overlaps visual pathways that provide circadian rhythms based on light input from photoreceptors in the retina. This circuit ends in the hypothalamus and thalamus. Serotonin is the primary neurotransmitter involved in this pathway. Neurotoxic destruction of serotonin at this pathway alters circadian rhythms (Myers & Badia, 1995). L-Tryptophan. L-tryptophan is the amino acid precursor of serotonin. Studies have shown that oral administration of L-tryptophan induces sedation, decreases levels of anxiety before sleep, decreases total awake time, reduces the frequency of nighttime arousals, and increases the amount of non-REM sleep. Nonspecific stress may produce sleep disturbances by shunting L-tryptophan down a different metabolic pathway than the serotonin-forming pathway. The inhibition of the synthesis of serotonin leads to sleep disturbances, which are quickly reversed by small doses of precursors of serotonin (Robinson, 1993). Melatonin. Researchers have recently used melatonin levels as markers of the circadian system. While its exact mechanism is not understood, melatonin is 22 synthesized and released in the pineal gland primarily during the nighttime hours but can occur even when individuals are awake. Melatonin has been shown to induce sleep (Brzezinski, 1997; Myers & Badia, 1995; Robinson, 1993; Wurtman, Axelrod, & Fischer, 1964). The release of melatonin usually begins at 2200, peaks at 0200, and terminates at 0800. Melatonin is stable across many nights, and it is unaffected by activity. In elderly populations, nighttime serum melatonin levels are reduced by at least 50%, whereas daytime levels are unchanged. Larger melatonin amplitudes are associated with longer durations of sleep that may explain, in part, why older adults have difficulty sleeping through an entire night uninterrupted. An appropriately timed administration of exogenous melatonin has been used to treat sleep disturbances associated with circadian phasing, to increase circadian amplitude, to hasten resynchronization after transmeridian travel, and to enhance nighttime sleep quality (Brzezinski, 1997; Myers & Badia, 1995). These findings have implications for populations in which sleep disturbances are problematic. Growth hormone. Growth hormone is secreted from the anterior pituitary during NREM sleep (Adam, 1980; Adam & Oswald, 1983; Chuman, 1983; Leo & Huether, 1998; Ludwig-Beymer et al., 1994; Oswald, 1980) and usually peaks during Stages III and IV early in the night. This secretion also occurs after meals, physical exertion, and psychological stress. The highest levels, however, occur during non-REM sleep. The amplitude and peak patterns of growth hormone secretion change with age. Growth hormone secretion is associated with sleep 23 primarily in adolescence. The honnone's levels increase during puberty, decrease in amplitude in adults, and are nearly absent in the elderly (Robinson, 1993). Growth honnone enhances anlino acid transport into cells and promotes protein synthesis. Thus, non-REM sleep is believed to enhance tissue restoration (Adam, 1980; Chuman, 1983; Leo & Huether, 1998; Ludwig-Beymer et aI., 1994; Oswald, 1980). Catecholamine and corticosteroid levels are depressed during non­REM sleep. These substances are highly catabolic, and their cyclic suppression during sleep periods further enhances protein synthesis (Oswald, 1980). Circadian Rhythms Circadian rhythms are psychophysiological measures that periodically adjust every 24 hours (Felton, 1987; Myers & Badia, 1995; Robinson, 1993), These rhythms serve to organize a living system's physiologic function temporally and adaptation to fluctuating conditions in the environment and are, therefore, important to the health of an organism (Ellmore & Burr, 1993). Wakefulness, temperature, serum cortisol, honnones, sensitivity of the respiratory center to carbon dioxide, and urinary potassium excretion are just a few examples of the numerous biological processes that exhibit circadian rhythmicity (Dement et al., 1982; Felton, 1987; Myers & Badia, 1995; Robinson, 1993). The timing of sleep onset and sleep duration also affects circadian rhythms. EEG changes during REM and non-REM sleep have been associated with 'the melatonin and the circadian rhythms of sleep (Dijk, Shanahan, Duffy, Ronda, & Czeisler, 1997). 24 Manifestations of Sleep Behavioral indicators of sleep include closed eyes, body movement, respiratory pattern, and a response to arousal stimuli. These indicators have been used in sleep research to assess the presence of human sleep (Carskadon & Dement, 1994). A greater understanding of the sleep process has been obtained through the polysomnographic recording of sleep in the laboratory. This process includes the use of an EEG, electromyogram, and electrooculogram to record brain waves during sleep, along with muscle and eye activity. Data from these instruments are then analyzed to provide researchers with the information needed to evaluate the sleep process" Manifestations of Wakefulness During wakefulness, the EEG is characterized by spontaneous, low-voltage, random, and fast electrical activity. The level of muscle activity is normally high, and only REMs are recorded on an electrooculogram. Beta waves occur from 13 to 35 hertz per second (hz/sec). Slow, rolling movements of the eyeballs are associated with a decrease in the level of alertness. An alpha wave is the pattern characteristic on an EEG during relaxed wakeful periods. These waves occur at a rhythm of 8 to 13 hz/sec and originate mainly in the occipital area of the brain. Large amplitudes are common, but as one moves into Stage I sleep, the alpha amplitude attenuates. 25 Manifestations of Stage I Sleep Sleep usually begins when the lights dim and is followed by less than 20 minutes of awake time. Stage I sleep is characterized by attenuated alpha waves and interspersed low-frequency theta waves. Eye movements are slow and rolling. Parasympathetic activity begins with a slowing of respiration, heart rate, and a drop in blood pressure. Less than 5% of the sleep cycle is included in this stage. Individuals in this stage of sleep are still aware of aural stimuli. Manifestations of Staee IT Sleep Stage II sleep follows with almost absent eye movements. There is mixed­frequency, low voltage corticoelectrical activity with bursts of sleep spindles and K complexes, then a large-amplitude negative wave followed by a positive wave. Two of these events herald the onset of Stage II sleep. Parasympathetic activity increases during Stage II. Myoclonic jerks occur, and individuals are easily awakened from sleep during this stage. Manifestations of Delta Sleep Stages III and IV sleep, commonly called delta sleep or slow-wave sleep, are defmed by high voltage (75 microvolts or greater), low-frequency waveforms of less than 4 hz/sec. Stage III sleep is identified when at least 20% to 50% of the recorded patterns show these characteristics. Occasional sleep spindles may occur. At this point, the subject becomes difficult to arouse. Parasympathetic activity increases. In Stage IV sleep, more than 50% of the wavelengths per recorded page 26 are delta waves. Parasympathetic activity reaches its peale Heart and respiratory rates are slow and even, and muscle tone is low. Very few eye movements are recorded. Sleepwalking, sleeptalking, and nocturnal enuresis occur during this stage. Some dreaming may occur. Manifestations of Rapid Eye Movement Sleep REM sleep looks much like the alert state in humans. Sympathetic activity is high, with blood pressure variability, and heart and respiratory rates increase. electromyogram and electrooculogram fmdings show that large muscles become hypotonic and REMs occur. The EEG shows predominantly beta waves. This is the stage of sleep in which dreaming is most likely to occur. Normal Sleep Patterns An active RAS maintains the alert state until external stimuli, internal stimuli, or both inhibit the system. Then the regulatory structures responsible for sleep are activated. The period of time it takes to go from the alert state to Stage I sleep is known as "sleep latency." Laypersons call the period "falling asleep." The length of this period varies in individuals. Sleep then progresses from Stage I to Stage II, then to the deep sleep of Stages III and IV, and finally to REM sleep, which is controlled by mechanisms in the pontine reticular formation. A new cycle begins with Stages II, III, IV, III, II, and REM sleep. A person cycles through these various stages approximately every 60 to 90 minutes or four to eight times per night. The initial latency to REM onset is the longest of the night, with 27 greater periods of non-REM sleep early in the night and much denser REM later in the sleep period (most commonly the early morning hours), By the end of the night, the pontine reticular formation is inhibited, and the RAS is activated-causing arousal and awakening. The Purpose of Sleep The importance of sleep to health is often tested through determining the effects of sleep deprivation. It is believed that sleep deprivation must be greater than 30 hours of total wakefulness to create decreased performance or physical efficiency in healthy individuals (Shaver & Landis, 1995). Sleep disturbances that prevent sleep cycle completion or a change in usual patterns have been found to decrease physical performance, provoke negative mood changes, and cause energy disruption. Performance can usually be maintained if the motivation to perform is sustained. However, when engaged in monotonous activities, concentration is easily lost. At this point, it is easy to lapse into "micro-sleep," placing these individuals at risk for injury (Shaver & Landis, 1995). Studies in rats have shown that prolonged total sleep deprivation leads to debilitated appearance, weight loss, increased energy expenditure, skin lesions, increased food intake, increased plasma norepinephrine, decreased plasma thyroxine, and (in late stages) decreased body temperature and death (Rechtschaffen, Bergmann, Everson, Kushida, & Gilliland, 1989). Human total sleep deprivation studies have occurred for up to 11 days, but most studies have been conducted over a period of 5 days or less. This period is too short to demonstrate severe physiological changes (Rechtschaffen et aI., 1989). Home (1985) voiced opinion that there are currently no deflnitive data in humans to suggest that lack of sleep is detrimental to any organ other than the brain. 28 Adequate sleep is important in maintaining an alert state and maximum cognitive function. Investigators have found that deprivation of sleep causes individuals to have episodes of minisleep for a few seconds throughout the day (Chuman, 1983; Morewitz, 1988; Shaver & Landis, 1995). During these "minisleep" episodes, there are short lapses of attention. The EEG shows Stage I waves or other non-REM or REM waves. The frequency and duration of these events increase·as sleep deprivation increases. These individuals may have difflculty with memory, concentration, or motor skills. Tasks requiring speed or perseverance are done with difficulty. Secretion of interleukin-1 is dramatically increased during sleep and is associated with the onset of slow-wave sleep (Moldofsky et aI., 1986). As a result, ongoing sleep disturbances may interfere with normal immune functions. Palmblad and colleagues (1979) documented reductions of DNA synthesis in 12 subjects after 48 hours of sleep deprivation, suggesting that cell-mediated immune reactions may be decreased. Irwin and colleagues (1994, 1996) reported a reduction of natural killer cell and lymphokine-activated killer cell number and activity. Shaver and Landis (1995), however, cited studies that indicate sleep deprivation may actually stimulate acute phase responses of the immune system. 29 It is believed that non-REM sleep is important for body tissue restitution following the "wear and tear" on the body during waking activities, and it may also be important for the regulation of normal body temperature and the metabolism of energy. REM sleep, on the other hand, is believed to be important for psychic restitution and stimulation of the brain (Shaver & Landis, 1995). Deprivation of only REM sleep appears to affect mood and basic biological drives, and it may cause hyperactivity, emotional liability, agitation, mood disruption, and decreased impulse control (Chuman, 1983). Selective non-REM deprivation tends to cause hyporesponsiveness, excessive sleepiness, and decreased motivation (Chuman, 1983). Deprivation of sleep in all stages appears to cause mood changes. There is also increased irritability and aggressiveness. A person may even experience hallucinations, disorientation, personality disorders, or a loss of emotional control. It takes longer to perform tasks, and individuals may appear somber and listless. Sleep-deprived individuals may be observed to have mild nystagmus, hand tremor, ptosis of the eyelids, or expressionless faces. Speech quality and word pronunciation deteriorates; pain tolerance is decreased. Corticosteroid and catecholamine output levels increase. It is important to note that sleep stages do not occur in isolation. Alterations in one phase are likely to affect all phases of sleep. Most people are not fully deprived of sleep for more than a few days. Partial sleep deprivation over long periods of time can affect mood and performance, however. 30 Sleep Disturbances Sleep disturbances may occur in the form of an inability to fall asleep, difficulty staying asleep, or awakening before readiness to rise. Sleep disturbances may also take the form of excessive daytime somnolence, daytime sleepiness, hypersomnolence, or fatigue. Elderly populations have reported changes in patterns affecting total sleep time and the perceived quality of that sleep. In addition, clinical observation has identified individuals with cancer as those at risk for sleep disturbances due to the symptoms associated with the disease and its treatment. Insomnia Insomnia is the most frequent disturbance of sleep in the general population (Kales et aI., 1982). They reported that complaints of insomnia in people aged > 50 years approach 40%. Mant and Eyland (1988) related that 20% of persons aged ~ 55 complain of difficulty sleeping, whereas an additional 17 % require hypnotics to avoid having sleep disturbances. Insomnia is usually viewed as an inability to sleep, but the term is inadequate for the different disorders associated with it. Insomnia includes sleep onset insomnia, early morning awakening, and an inability to maintain sleep throughout the night. These insomnias are classified under disorders in initiating and maintaining sleep. Onset insomnia. Onset insomnia is a sleep disturbance in which an individual has difficulty falling asleep. The period from wakefulness to Stage I 31 sleep tends to be longer. This period of time is known as sleep latency. Critically ill individuals are more likely to experience this type of insomnia due to various internal and external stimuli on the RAS. Anxiety and pain are examples of possible stimuli responsible for onset insomnia. In onset insomnia, sleep tends to be normal, but progression through the different stages of sleep still occurs. However, there tends to be a decrease in the number of sleep cycles due to the lengthened latency period. Conclusion insomnia. In contrast to onset insomnia, conclusion insomnia is the interruption of sleep before the usual awakening time. This type of insomnia is accompanied by a decrease in REM sleep. The number of sleep cycles may be affected, as well, if awakening occurs too early. Early morning awakening can be an important diagnostic indicator of depression. Interruption insomnia. Interruption insomnia is characterized by frequent aWakenings throughout the night. These interruptions can be caused by internal or external stimuli. Nocturia, nausea, and pain are examples of internal stimuli, whereas procedural events might awaken individuals in the acute care setting. Whenever sleep is interrupted, it has to begin again with Stage I sleep; then REM sleep is compromised. If interruptions come repeatedly, deep, restorative sleep will also be affected. Disorders of Excessive Daytime Somnolence Disorders of excessive daytime somnolence include excessive sleepiness caused by sleep apneas and multiple awakenings. This type of sleep disorder is 32 seen predominantly in overweight men, results from sleep-associated relaxation of the upper airway, and causes obstructive sleep apnea. This disturbance can also be seen in individuals who have frequent nighttime awakenings, thus preventing sufficient nocturnal sleep. This disturbance is consistent with the terminology of "daytime sleepiness" or "hypersomnolence" and, in fact, may also be responsible for daytime fatigue. Disorders of the Sleep-Wake Schedule Sleep-wake disturbances may result from external influences or from neurological disorders. The internal circadian pacemaker is probably the suprachiasmatic nucleus of the hypothalamus. The suprachiasmatic nucleus establishes and maintains the exogenous rhythm, but it can be altered by exogenous factors such as jet travel, shift work, or medications. Rhythms in plasma cortisol levels, melatonin, and temperature are the best indicators researchers have for monitoring circadian cycles in normal sleep. This type of sleep disturbance is common in the acute care setting in which the circadian rhythms may be disturbed by light, procedures, and medications. Many of the medications administered by nurses disrupt biological rhythms. Cortisol induces an enzyme that shunts tryptophan away from the serotonin pathway and causes sleep-onset insomnia (Robinson, 1993). Stress can also lead to an increase in serum cortisol by the same mechanism as exogenous steroids. Many pain medications deprive users of REM sleep. When these drugs are withdrawn and normal sleep is initiated, REM rebound may occur. As a result, a higher percentage of the total sleep time is spent in REM sleep. Since catecholamines are released during REM sleep, these individuals will have lengthened catecholamine exposure until REM sleep returns to normal, making them vulnerable to the effects of these neurotransmitters for a lengthened period (Robinson, 1993). 33 Onset insomnia was evaluated in this research by questioning individuals about the length of time it generally took them to fall asleep during the past week. Interruption insomnia was evaluated by asking about specific reasons for nighttime awakening. Disorders of excessive daytime somnolence were evaluated through questions about daytime sleepiness and daytime enthusiasm. Sleep in Elderly Populations Researchers have found that there are many changes in sleep physiology and alterations in daytime functioning as aging occurs. The nighttime sleep of elderly individuals is characterized by a number of changes from the sleep of younger adults. There is a reduction in Stages III and IV sleep. There is also increased sleep latency at bedtime; that is, it takes elderly individuals longer to fall asleep. In addition, the number of arousals during sleep is increased, and the length of those arousals is increased. Increased fatigue is also seen that may be explained, in part, by the aforementioned changes. Total sleep time and the proportion of REM to NREM sleep remain fairly constant from ages 20 to 60. As adults reach middle age (40 to 59 years), circadian patterns change in the direction of a more disturbed sleep. Dement and colleagues (1982) reported that elderly individuals have an increased total duration 34 of Stage I sleep and an increase in the number of shifts into Stage I sleep. They reported very little change in Stage II. Restorative Stage IV sleep is reduced to 50% by age 50. By the 6th decade of life, little or no Stage IV sleep can be found in 25% of the population (Bahr, 1984). By the time adults reach their 70s, one third have lost Stage III sleep, and 100% have lost Stage IV sleep_ Recent studies indicate that the loss of deep sleep in older populations may be related to a measurement artifact. The amplitude of their delta waves may be too low to be detected by laboratory equipment (Bliwise, 1993; Robinson, 1993; Webb, 1982). Increasing amplifier sensitivity eliminated age differences in the length of deep sleep in cross-sectional and in longitudinal groups (Bliwise, 1993). The most commonly reported disturbances found in elderly populations include poorer quality of sleep, longer sleep latencies, more nighttime awakenings, shorter sleep stage periods, and more frequent use of drugs (Middelkoop, Smilde-van den Doel, Neven, Kamphuisen, & Springer, 1996; Webb, 1982). Circadian rhythms can be determined by monitoring fluctuations in body temperature, endogenous melatonin, and polysomnographic recordings in the sleep laboratory. As aging occurs, the cycles associated with sleep tend to occur at an early hour in the day _ These rhythms also have decreased amplitudes and shortened phases in aging adults. There is an increase in plasma catecholamines and a decrease in glomerular filtration and renal plasma flow. Morphological and chemical changes are also evident in the suprachiasmatic nuclei and in the pineal gland. Polysomnographic studies by Myers and Badia (1995) showed that age- 35 related alterations in sleep include reduced amplitude, an advanced phase, and a shortened period and are likely due to changes in the circadian rhythms of temperature, melatonin, and sleep-wakefulness. A sleep study by Haimov and Lavie (1997) showed that aging did not affect the overall amount of daytime sleep or the general shape of the polysomnographic sleep propensity curve. The study did show that aging attenuated circadian amplitude and, in fact, even advanced it. These circadian alterations are associated with the decreased sleep quality found in elderly popUlations. There appears to be no difference in the total sleep time between younger and older adults (Bliwise, 1993), but younger people have a consolidated night's sleep with few or no daytime naps. Elderly adults tend to exhibit a sleep distribution across the 24-hour day (Dowling, 1995; Robinson, 1993). Given that information, it is important that daytime napping be analyzed in this population. Changes in behavioral rhythms such as physical activity can cause sleep disturbances. External cues such as changes in light intensity, mealtimes, or particular television shows are strong regulators for one's internal circadian clocks. Disturbances or alterations in these rhythms are more common in elderly popUlations, causing increased disruption of previously established patterns. These disruptions can be attributed to retirement, isolation, boredom, grief, anxiety, loneliness, or depression. The manifestation of these disruptions on sleep is evident in the gradual phase advancement of the sleep period compared to the traditional sleep time. As a result of phase advancement, elderly individuals report 36 awakening in the middle of the night or early morning drowsiness. Many elderly individuals also show a more polyphasic sleep pattern, which is similar to that found in very young children. According to Prinz, Vitiello, Roskind, and Michael (1990), healthy elderly persons require more time to fall asleep, awaken more frequently during the night, and awaken earlier in the morning than young adults. Reyner and Home (1994) reported that older adults had an earlier sleep onset time, move less during sleep, and lie awake in bed longer upon awakening than younger subjects. As aging continues, the frequency and duration of awakenings increase, and sleep efficiency is lost (Haimov & Lavie, 1997).· Morewitz (1988) reported that elderly persons have multiple miniature arousals during a normal night's sleep. Scarcely an hour will go by without awakening, with deep, restorative sleep lost. It also takes older individuals longer to return to sleep after a nighttime awakening. Thus, it takes the elderly more total time in bed to achieve the same amount of restorative sleep that a young adult achieves in a short amount of time. Qualitative and quantitative modifications in this basic structure and continuity of the sleep-wake cycle most likely contribute to excessive daytime sleepiness (disorders of excessive somnolence). As a result, increased daytime napping occurs (Morewitz, 1988). There is also an increase in daytime fatigue and a decrease in daytime sleep latency in elderly populations. Daytime napping was more common among elderly persons with chronic illness than in healthy elderly adults, however, especially among older men (Asplund, 1996). Morewitz (1988) reported that there is 37 evidence of brief sleeping episodes (microsleeps) during the day. These periods are characterized by momentary losses of attention and EEG evidence of Stage II sleep bursts lasting up to 10 seconds. It is not known if these episodes result from partial sleep deprivation due to poorly sustained nocturnal sleep or if they are a manifestation of aging changes in sleep architecture (Morewitz, 1988). Not all researchers report the same phenomena regarding sleep changes in elderly populations, but it seems apparent that older adults are more likely to report sleep disturbances than young adults. They are more likely to report difficulty falling asleep and staying asleep throughout the night. Their attenuated sleep propensity curve is less likely to provide them with deep sleep. Sleep disturbances are likely to be most problematic in individuals with health problems or chronic disease. Problems with daytime drowsiness may be related to sleep disturbances, but they may also be caused by changes in lifestyle brought on by retirement or poor health. Regardless of the cause, Kales and Kales (1970) reported that elderly adults are less able to tolerate prolonged sleep disturbances than younger individuals. Sleep in Cancer Patients Sleep disturbances related to cancer diagnosis and treatment are often reported clinically but have rarely been documented. In most cases, research about sleep disturbance has been limited to a single item on a symptom questionnaire. According to Yellen and Dyonzak (1996), sleep disturbances can result from a number of sources, including anxiety, depression, pain, metabolic disturbances, 38 and changes in surroundings while undergoing treatment. Sleep disturbances may take the form of insomnia or somnolence. Sleep disturbances may last for a few days or may even persist for weeks or months, affecting the overall quality of life and suggesting poor adaptation to the diagnosis and treatment of cancer. Depression and anxiety are frequently correlated with sleep disturbances (Kales et aI., 1982; Leo & Huether, 1998; Ludwig-Beymer et aI., 1994; Robinson, 1993). Depression and anxiety are associated with increased mobility during sleep, more disruptive sleep, and an absence of a rested feeling upon awakening. For many individuals, the diagnosis of cancer is synonymous with suffering, disfigurement, pain, long hospitalization, separation from loved ones, and death. Research by Lamb (1982) revealed that hospitalized oncology patients experience significantly higher levels of anxiety and depression than hospitalized patients with benign conditions. However, no significant increase in sleep disturbances was found for malignant conditions. Lamb suggested that the small sample size and the reliability of the tool in oncology patients might have contributed to these results. Beszterczey and Lipowski (1977) found that insomnia was positively correlated with reported symptoms of anxiety and depression but, contrary to expectations, not with pain. These studies indicate that clinical impressions need to be verified through research, since assumed relationships have not necessarily been validated. Because oncology patients often report symptoms of pain, nausea and vomiting, and fatigue, one would expect to fmd sleep disturbances to be a common complaint. This symptomatology associated with cancer suggests a clinical 39 impression that sleep is more often and more severely disturbed in cancer patients than in the general population or in individuals with nonmalignant conditions. The actual incidence of insomnia in cancer patients ranges from 29 %, which was reported by Andrykowski and Henslee (1988) in those undergoing autologous bone marrow transplant, to 95 % in a sample of 300 cancer patients, which was reported by Yellen and Dyonzak (1996). These wide ranges in fmdings indicate that there may be great variability based on site. Silberfarb and colleagues (1993) studied breast and lung cancer patients compared to healthy control subjects and healthy insomniacs. These subjects kept a sleep log for 2 weeks and spent 3 nights evaluating polysomnographic recordings. Researchers found that the sleep of breast cancer patients was similar to that of the healthy control group, whereas the lung cancer patients' sleep was more like that of healthy insomniacs. In addition, the lung cancer patients consistently underestimated their sleep difficulty despite objective verification in the laboratory. The breast cancer patients tended to overestimate their problems with sleep. These researchers pointed out that individuals with lung cancer may have a form of denial because their disease is mainly self-induced, whereas complaints of insomnia tend to be inaccurate and exaggerated. These studies indicate great variability in sleep disturbances between these two sites, both by report and by laboratory fmdings. Hu and Silberfarb (1991) found that the literature reporting relationships between sleep disturbances and cancer are much more prevalent than those reporting no differences. On the basis of their research, they reported the 40 following: (a) Cancer patients are approximately three times more likely to have sleep maintenance problems compared to a control group; (b) cancer patients have greater difficulty with sleep onset and maintenance than the general population; (c) the sleep of the cancer patient is similar to that of the suicidally depressed patient but dissimilar in depressive symptomatology; and (d) the second most commonly prescribed psychotropic medications after antiemetics are those intended to assist in sleep. Pain, hypoxia, urinary frequency, fever, pruritus, and alterations in endocrine function have all been associated with insomnia in cancer patients (Massie & Lesko, 1989). The diagnosis of malignancy is often associated with stress, anxiety, and depression, and there is an inverse relationship between those feelings and the amount of sleep obtained (Thomas, 1987). The diagnosis of insomnia has also been found to have an adverse effect on the quality of life (Cannici, Malcolm, & Peek, 1983), not only during the time of its occurrence but also by producing fatigue and irritability. This fmding is particularly problematic for those experiencing pain or other side effects associated with cancer and its treatment. While insomnia is the most frequently reported disturbance related to sleep in cancer patients and the general population, sleep disturbances can also take the form of hyper somnolence or fatigue. In cancer patients, fatigue is often associated with cranial radiation. "Somnolence syndrome" was first described as acute encephalopathy in 1929 (Yellen & Dyonzak, 1996); it consisted of symptoms 41 ranging from mild drowsiness to marked lethargy. Affected individuals sleep for prolonged periods of time 3 to 12 weeks after treatment. They may feel exhausted with minimal activity and have no energy for daily activities. The syndrome is also seen in persons following bone marrow transplantation and certain types of chemotherapy. Individuals reported reduced strength, memory difficulties, and difficulties with nighttime sleep. Greenberg and colleagues (1992) reported that patients receiving radiation for breast and prostate cancer suffered from excessive fatigue. Literature reviews by Nail and Jones (1995) and Richardson (1995) reported a high prevalence of fatigue in individuals receiving radiotherapy or chemotherapy, which became more significant as treatment progressed. Sleep in Elderly Persons With Cancer Sleep disturbances secondary to cancer or its treatment are likely to occur in elderly adults. Berlin (1984) reported that anxiety, depression, loss of social support, and a diagnosis of cancer are contributory factors in sleep disturbances in the elderly. Preexisting sleep disturbances or tendencies can be aggravated by illness. Age-related sleep changes, normally well-tolerated, can become problematic in the context of illness. Cancers, especially solid tumors, are more prevalent as aging occurs. As the American population ages, the incidence of cancers will continue to increase. In addition, improving treatment modalities will increase survival rates; that is, individuals will live longer with the effects of the disease. From a clinical standpoint, sleep disturbances are often associated with malignancy and aging. However, there are currently no published studies ECCLES HEALTH SCJrrJ"~S L!~"·~ RV .... ,,\;.. lu,j,n. describing the sleep phenomena of older cancer patients. There is no published literature to indicate that their sleep is more problematic than younger adults, although it would be expected because sleep is more problematic in elderly individuals with chronic disease. It is important to verify and document sleep disturbances in this group and to identify those at risk. This study provides a baseline for further research, including interventional studies designed to manage sleep disturbances in elderly cancer patients. Conceptual Framework 42 The conceptual framework for this research is founded on the theory that there are many factors contributing to sleep pattern disturbance. The variables within the framework were included because they are known to influence or are believed to influence sleep or the circadian rhythms associated with it. Demographic variables such as age, gender, culture, and economic status may influence sleep patterns. In addition, clinical variables such as site and stage of disease, treatment modality, and in-patient or out-patient status can have different effects on sleep patterns. These sleep disturbances can be broken down into the components of sleep identified by the PSQI, that is, sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbance, use of hypnotic medication, and daytime dysfunction (see the Figure). Demographic Variables Age Gender Marital status Economic status Educational status Culture Clinical Variables Disease site Disease stage Treatment modality Hospitalization status Conceptual Model Components of Sleep Sleep latency Sleep efficiency Sleep duration Sleep disturbance Use of medication Daytime dysfunction Subjective sleep quality 43 44 AKe and Sleep "Aging" is subject to semantic confusion. Chronological age does not necessarily approximate physiological age. The decline in slow-wave sleep may occur at a chronological age far earlier than most age-related declines in other biological functions. Some gerontology researchers have noted that distance from death may be a far better approximation of the aging process, but too few longitudinal sleep studies exist to yield these fmdings (Bliwise, 1993). Studying elderly persons in subgroups is important because sleep changes as aging occurs. The amount of Stages III and IV sleep begins to decline at age 50. Stage III is absent in one third of those age· 70 to 79, and Stages III and IV may be completely absent by age 90. Sleep efficiency decreases to approximately 70% to 80 %, and the percentage of Stage I sleep increases to approximately 8 % to 15 % as aging occurs. The amplitude and duration of sleep episodes decline with aging, along with the overall amount of nighttime sleep (Bliwise, 1993; Webb, 1982). In addition, elderly individuals are more likely to suffer from poor health. Denlent and colleagues (1982) reported that in respondents aged ~ 65 there was a significant increase in the proportion who claimed to sleep fewer than 5 hours each night. Individuals aged ~ 65 reported frequent nocturnal awakenings and early morning arousal significantly more often than younger subjects. Complaints of nonspecific sleep disturbances and awakenings during the night, along with the use of sedative-hypnotic medications, appear to increase with age. These complaints are substantiated by polysomnographic measures, including lower sleep efficiency, decreased slow-wave sleep, and less time spent asleep compared to healthy younger sUbjects. 45 A longitudinal study of sleep in aging cohorts (> 61) found that most measures of EEG sleep and sleep quality tended to be relatively stable across all age groups, with the exception of increased prevalence of mild sleep-disordered breathing in the oldest cohort. After 3 years, individuals in the oldest cohort showed greater decline in sleep quality, continuity, and depth compared with the two younger groups who showed a relatively unchanging profile from study entry to fmal follow-up assessment (Hoch et al., 1997). They believed it was plausible that increasing health difficulties associated with aging could account for the greater sleep-related decline among the "old-old" cohort. Buysse and colleagues (1991) found that measures of habitual sleep quality in elderly subjects did not correlate strongly with most polysomnographic sleep measures, number of medications used, or circadian measures. The results showed that subjective sleep quality deteriorates in healthy elderly adults but not to the level seen in individuals with preexisting sleep disorders. Extremely healthy elderly subjects appeared to adapt their perceptions of objectively disturbed sleep. Since sleep architecture and perceptions of sleep change with aging, especially in relation to health problems, it is important to verify differences in sleep disturbances in each age cohort. Gender and Sleep Aging women have earlier bedtimes and wake times than men but report more daytime sleepiness (Dowling, 1995). Women also report more sleep 46 disturbances, but their polysomnographic sleep patterns appear better. Men in their 40s tend to increase their awake time and to decrease their REM time; they also have longer sleep latency. Women lose REM sleep time in their 50s, but they maintain more of the deep sleep associated with Stages III and IV than men (Robinson, 1993). Dement and associates (1982) found that Stage III sleep tends to be normal or even increased in elderly women. Reyner and Home (1994) reported that elderly women especially were more likely to have earlier sleep onset times, move less during sleep, and lie awake in bed longer upon awakening than younger subjects. Women have a higher number of stage shifts probably because they have retained a greater number of absolute stages during sleep time. Both men and women in old age experience multiple miniature arousals (Stage I or awake) during a normal night's sleep. Rediehs, Reis, and Creason (1990) reported greater variability in total sleep time for men. They also found that men spend more time awake after initial sleep, particularly in the later part of the night. Women awoke more frequently, but their awake periods were generally brief. Based on this information, it would be expected to see differences in sleep patterns between men and women. These differences may be dependent on whether or not the evaluation comes in the form of a questionnaire or laboratory data. Ford and Kamerow (1989) studied the prevalence and incidence of sleep disorders in the general adult community. They found that women, separated or widowed, the unemployed, and those in the lowest socioeconomic quartile had significantly higher rates of insomnia, whereas the never married or never 47 employed were more likely to report hypersomnia. No significant difference was found when comparing White populations to non-White populations. Kales and colleagues (1982) reported that insomnia is associated with increasing age, female gender, psychological disturbances, and individuals of lower educational and socioeconomic status. Klink and colleagues (1992) reported that female gender, advancing age, concomitant health problems, and prior history of insomnia were significant risk factors for complaints of insomnia in the general population. These investigators did not compare cultural or economic factors. The sociodemographic variables contributing to sleep pattern disturbances in elderly cancer patients have not been studied. Clinical Variables and Sleep It would be inappropriate to assume that all individuals with cancer suffered with sleep pattern disturbances. Likewise, it would be inappropriate to assume that those who suffer with sleep pattern disturbances all suffer with the same type of disturbance. Differences are likely to be different based on the site and stage of disease, the delivered treatment modality, and hospitalization status. Silberfarb and associates (1993) reported that lung cancer patients slept as poorly as insomniacs but that they underreported their sleep difficulties. Breast cancer patients slept similarly to normal-sleeping volunteers. They found that lung cancer patients appear to underestimate their objectively verified sleep 48 disturbances. These fmdings indicate that the site of the disease may be responsible for problematic sleep. Sleep disorders in cancer patients have ranged from a reported low of 29% (Andrykowski & Henslee, 1988) to a high of 95% in transient or persistent insomnia from a sample of 300 cancer patients (Yellen & Dyonzak, 1996), indicating that there may be a wide variation in sleep disturbances related to site. Sleep disturbances may be exacerbated by paraneoplastic syndromes associated with steroid production (Berlin, 1984). Endocrine changes, which might interfere with sleep, can occur as a result of surgical procedures, hormone therapy, or reproductive changes induced by chemotherapy or radiotherapy. Sleep disturbances may result from fever, pruritus, endocrine disorders, medications, and anxiety, which are likely to vary according to site of disease. Fever may be a manifestation of the disease itself or an infection from immunosuppressive therapy. Pruritus occurs in 10% to 30% of individuals with Hodgkin's disease (Seiz & Yarbro, 1996) but may occur in other types of lymphoma, leukemia, or hepatic tumors. The site and stage of disease are responsible for determining which treatment protocols are used that, in tum, will determine which side effects are likely to be problematic. Sleep disturbances are, therefore, likely to be influenced by the location, pathology, and characteristics of the primary disease. Stage of the disease and the management of symptoms associated with advanced disease may cause sleep pattern disturbances. Sleep disturbances might 49 be exacerbated by symptoms associated with tumor invasion such as draining lesions, gastrointestinal or genitourinary alterations, pain, obstruction, fever, cough dyspnea, pruritus, and fatigue (Berlin, 1984). Pain is experienced in 60% to 90% of patients with advanced cancer, and poorly controlled pain results in high levels of anxiety and depression (Massie & Holland, 1989) that, in turn, can cause sleep pattern disturbances (Kales et al., 1982; Klink et al., 1992). Carrol, Fine, Ruff, and Stepnick (1994) reported that individuals with head and neck cancers placed on a four-drug pain regimen increased their average sleep time of 4.4 hours to 7.5 hours. Symptom severity such as pain, constipation, and restlessness can increase with disease progression (Storey, 1998). Dohno, Paskewitz, Lynch, Gimbel, and Thomas (1979) indicated that illness severity is linked to sleep deprivation. Hospitalization Hospitalization status can cause sleep pattern disturbances in oncology patients. Hospitalization is likely to subject individuals to interruptions due to treatment schedules, hospital routines, or roommates. Other factors influencing sleep in the hospital setting include noise, ambient temperature, comfort, pain, and anxiety (Webster & Thompson, 1986). Sheely (1996) reported that the number and duration of nocturnal disturbances, as well as the level of patient participation in nocturnal care, were negatively correlated with sleep quality in hospitalized oncology patients. Ulander, Grahn, Sundahl, and Jeppsson (1991) said that patients with cancer reported difficulty getting to sleep and staying asleep when hospitalized. However, Lamb (1982) found no significant difference between the 50 sleep of hospitalized patients with cancer and matched patients with nonmalignant diagnosis. Consideration should be given to the fact that clinical variables may be impossible to analyze in isolation. Treatment modalities for cancer are determined by the site and stage of disease. Hospitalization status may be determined, in part, by the stage of the disease and the treatment modality that are, again, determined by the site of the disease. When reviewing clinical data about each of these variables, it is important to note that one variable may be influencing the other and should, therefore, be considered before conclusions are drawn. CHAPTERll METHODOLOGY Desien This study was part of a larger descriptive correlational study conducted on sleep, pain, mood, and fatigue in cancer patients. The larger study (Beck & Schwartz, 1998) was designed to establish the prevalence rate and nature of sleep disturbances in 250 cancer patients from three settings within a large tertiary care center, including an assessment of the relationship of sleep quality to other symptoms. Secondary analysis for this research was conducted on the data. This cross-sectional research consisted of recruiting volunteers to complete a questionnaire that asks about sleep patterns, mood, fatigue, and pain. Demographic data were also collected as was clinical information regarding disease site and stage. The study met Institutional Review Board approval. Data were collected at specific dates during several months in 1998. Sample Two hundred fourteen cancer patients were recruited into the larger study. The prospective, consecutive sample consisted of individuals aged;;:: 18 who had been diagnosed with cancer and were currently receiving care during the prescribed time and were willing to participate. Individuals who were too ill (mentally or physically) were excluded. No pregnant women, prisoners, children, or other vulnerable groups were included. Settine 52 The study took place at the University of Utah Health Sciences Center in Salt Lake City, Utah. This large tertiary care center is a teaching facility for medical, nursing, and pharmacy students. Participants were recruited from three areas within the institution. These settings consisted of the inpatient oncology unit, the outpatient oncology clinic, and the radiation therapy department, including a main site and a satellite community center. Participation in the study was voluntary. Instrument The PSQI is a self-rated questionnaire that assesses the sleep quality of a 1- month time interval. The questionnaire was adapted to evaluate a I-week interval of sleep (Buysse et al., 1989), The questionnaire contains 19 self-rated questions and 5 questions rated by the bed partner or roommate (if available). Only self­rated questions are included in the scoring. The 19 self-rated items are combined to form seven component scores, each of which has a range of 0 to 3 points. In all cases, a score of 0 indicates no difficulty, whereas a score of 3 indicates severe difficulty. The seven component scores are then added to yield one global score, with a range of 0 to 21 points. 53 The 19 individually rated items generate seven "component" scores: (a) subjective sleep quality, (b) sleep latency, (c) sleep duration, (d) sleep efficiency, (e) sleep disturbances, (f) use of sleeping medication, and (g) daytime dysfunction. Buysse and colleagues (1989) compared healthy subjects, depressed patients, and sleep-disordered patients for 18 months using the PSQI. Acceptable measures of internal homogeneity, consistency (test-retest reliability), and validity were obtained. A global PSQI score of ;::: 5 yielded a diagnostic sensitivity of 86.5% (kappa = 0.75, J2 < 0.001) in distinguishing good and poor sleepers. The clinometric and clinical properties of the PSQI suggest its utility both in psychiatric clinical practice and research activities. The tool has been recommended for a simple screening measure to identify the presence of significant sleep disturbance, and it has also been recommended in studying the relationship between sleep quality and other variables such as age, gender, health status, medical or psychiatric conditions, and performance on other psychological variables. Its validity in cancer patients has not been documented. The PSQI does not measure frequency or duration of napping, which may be an important consideration in elderly populations since they tend to exhibit a sleep distribution across the 24-hour day (Bliwise, 1993). Procedure Individuals who were being treated in the inpatient, outpatient, and radiation therapy departments at the University of Utah Health Sciences Center were questioned about their willingness to participate in this study. Recruitment of 54 participants was conducted by nursing personnel and researchers, including myself, after being instructed on the correct procedure for data collection. The purpose and procedure for completing the questionnaire were explained. Completion of the survey implied consent. Demographic data were retrieved from the medical record or asked of the specific participant. Information was also collected about disease site and stage, presence or absence of pain, and current medications. Records were kept regarding enrollment, refusal, and completion of the questionnaire. Data Analysis Each of the questions from the sleep survey was encoded, and the responses from participants were categorized for analysis. Data from the completed sleep surveys were entered into the Statistical Package of the Social Sciences (SPSS)® version 8.0 and analyzed using this statistical program to answer the specific research questions addressed earlier. The seven component scores and the global score were computed for each subject. Significance was computed to an alpha level of .05. Sleep Quantity and Qverall Sleep Quality Mean scores of sleep quantity in elderly adults (aged ~ 65) were compared with the mean scores of sleep quantity in younger adults (aged < 65). Mean scores of overall sleep quality (PSQI) in elderly adults were compared to the mean scores of overall sleep quality in younger adults. Independent i-test analysis was conducted to answer each of these questions by comparing those aged ~ 65 or < 65. Demoeraphic and Clinical Factors According to the PSQI, a global score of ;:: 5 yields an 89.6% sensitivity and an 86.5% specificity in identifying poor sleepers. Demographic factors in elderly adults such as gender, culture, marital status, and socioeconomic status were compared to good and poor sleepers based on the PSQI global score. Clinical factors such as stage, site, treatment modality, and hospitalization were compared to good versus poor sleepers in a similar manner. Sleep Components 55 The seven components of the PSQI (subjective sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbances, use of medication, and daytime dysfunction) were tabulated for all older subjects. Frequency distributions of the sleep disturbances found in the elderly participants were computed. Suheroup Analysis It was intended that the sample of elderly participants would be broken down into three subgroups (aged 65 to 74, 75 to 84, and ;:: 85) and analyzed using one-way analysis of variance (ANOVA). Only 1 participant was;:: 85; consequently, the participants were reassigned into two groups (aged 65 to 74 and ;:: 75). The mean PSQI global score was computed for the two subgroups. These scores were compared using chi-square analysis to determine if differences related to the aging process were found. CHAPrERIV RESULTS Sample Description Inpatient data were collected over a 6-week period. Five days were spent in the outpatient clinic and 5 days were spent in the radiation therapy department. Of the 410 eligible patients, 279 agreed to participate in the study. The final sample included 214 subjects, but only 210 reported their age. Fifty-seven • >_ 1 ~'~ _ ~ .. (27.1 %) were aged ~ 65 (Table 1). The participants were fairly equally distributed between males and females in each age group. The sample was primarily White. Of those aged > 65, only 1 subject represented a population other than White. Marital status was also similar between the two age groups. Educational background for this group was relatively high, with almost 50 % having at least some college or technical training and less than 10% having less than a high school degree. Educational levels were similar in each age group. Income distributions indicated that subjects aged < 65 tended to have a higher net family income. In both groups, approximately half of the participants fell into the middle income bracket. The elderly group had a slightly higher percentage of participants with less than a high school degree and a slightly lower percentage with more than an associate degree. The elderly sample had a higher percentage who were of single 57 Table 1 Demographic Variables for < 65 and > 65 Age Groups < 65 ~ 65 Total Variable % !! % !! % N Gender Male 51.0 77 52.6 30 51.0 107 Female 49.0 74 47.4 27 49.0 103 Ethnicity Anglo 90.0 135 98.2 54 92.3 150 Other 10.0 15 1.8 1 7.6 55 Educational status < High school degree 7.9 12 12.3 7 9.2 19 Some college or associate 61.5 93 63.2 36 61.6 129 degree College graduate or more 30.5 46 24.6 14 28.9 60 Marital status Partner 70.9 107 61.4 35 67.7 142 No partner 29.1 44 38.7 22 31.6 66 Annual income < $15,000 17.5 24 36.4 16 22.9 42 $15,000 to $49,999 41.6 57 45.4 20 42.0 77 ~ $50,000 40.8 56 18.2 8 35.0 64 Note. Sample size varies due to missing data. marital status, which would be consistent with the loss of a spouse secondary to death. The elderly group also had a lower net family income, which would be consistent with retirement. Comparisons of Sleep Quantity Sleep quantity was evaluated in terms of the number of hours slept and sleep efficiency. The mean for hours of sleep was 6.8 (SO = 1.8) in those aged 58 < 65 and 6.9 (SO = 1.8) in those aged > 65. Independent 1 tests showed no significant differences between the groups in the number of hours slept (! = -.402, I! = .688). The mean sleep efficiency was .81 (SO = .17) in those aged < 65 and .79 (SO = .18) in those aged > 65. Independent 1 tests showed no significant difference in sleep efficiency between the two groups (! = .9, I! = .369). A comparison of sleep duration was made between the two groups. Chi­square analysis showed a significant difference in sleep duration between the two groups based on hours of sleep (Table 2). Interestingly, sleep duration is essentially the same measurement as sleep quantity. In order to simplify calculations for the PSQI global score, sleep quantity is changed from an interval scale to a 0 to 3 ordinal scale (known as sleep duration). No differences were found in sleep quantity between the two groups, but differences were found in sleep duration between the two groups. The younger group showed shorter sleep duration, with 41.6 % sleeping 6 hours or less compared with 27.2 % of the elderly participants. The distribution of scores may have evened out the means, explaining why no difference was found in sleep 59 Table 2 Comparison of Sleep Duration in < 65 and > Age Groups < 65 > 65 Sleep duration % n % n > 7 hours 43.6 65 36.4 20 6 to 7 hours 14.8 22 36.4 20 5 to 6 hours 29.5 44 12.7 7 < 5 hours 12.1 18 14.5 8 Note. Pearson X2 = 14.339, df = 3, 12 = .002. quantity. A larger sample may have seen different results. Comparisons of Sleep Quality The PSQI contains 19 self-rated questions and 5 questions rated by the bed partner or roommate (if available). Only self-rated questions are included in the scoring. The 19 self-rated items are combined to form seven "component" scores, each of which has a range of 0 to 3 points. In all cases, a score of 0 indicates no difficulty, whereas a score of 3 indicates severe difficulty. The seven component scores are then added to yield one global score, with a range of 0 to 21 points, 0 indicating no difficulty and 21 indicating severe difficulties in all areas. A score of ;::: 5 yields a diagnostic sensitivity of 89.6 % and a specificity of 86.5 % in predicting poor sleepers. The PSQI global mean score was 8.06 (SD = 4.7) for individuals aged < 65 and 8.52 (SD = 4.8) for those aged > 65. Independent 1- 60 test analysis showed no differences in the global score (t = -.556, 12 = .579). Subjects evaluated their perceived sleep quality during the past week by categorizing it as very bad, fairly bad, fairly good, or very good. A score of 0 to 3 was possible, with 3 indicating poorer perceived sleep quality. A comparison of perceived sleep quality showed a mean score of 1.23 (SD = .81) in those aged < 65 and 1.05 (SD = .83) in those aged ~ 65. Independent 1-test analysis showed no differences in perceived sleep quality between the two groups (1 = 1.4, 12 = .164). Demographic Comparisons in Those Aged More Than Sixty-Five Gender Differences No differences were found between good and poor sleeper status based on gender (Pearson X2 = 1.591, 12 = .207) (Table 3). Cultural Differences Subjects were asked to categorize their ethnic background into African American, Native American, Asian/Pacific Islander American, Hispanic American, or Anglo American (White). A percentage of 92.3 of all respondents identified themselves as Anglo American (White), and a percentage of 0.5 of all respondents identified themselves as African American, 2.4% as Native American, 1.4% as Asian/Pacific Islander American, and 3.3 % as Hispanic American. Of those aged > 65, 98.2 % were Anglo American, and only 1 subject (1.8%) was Asian/Pacific Islander American. Analysis of cultural differences in this sample would be 61 Table 3 Gender and Sleep Status in > 65 Age Group Sleeper status Gender Good sleeper Poor sleeper Male !!=4 !l = 15 Within gender 21.1% 78.9% Within sleep status 30.8% 51.7% Female !!=9 !l = 14 Within gender 39.1 % 60.9% Within sleep status 69.2% 48.3% Note. Sample size varies due to missing data. inappropriate due to the homogeneity of the sample. Marital Status The research tool asked subjects to categorize their marital status as single, married, living with a partner, separated, divorced, and widowed. Statistical breakdown showed that more individuals aged < 65 were single (16.6 % ) compared to those aged ~ 65 (5.3%). Those aged ~ 65 were more likely to be widowed than those aged < 65 (24.6% versus 2.0%). In the aged ~ 65 group, 5.3% were single, 61.4% married, none separated or living with a partner, 8.8% divorced, and 24.6% widowed. In order to increase the strength of analysis, marital status was compressed into two categories: (a) married/partner and (b) no partner (Table 4). These categories were compared to good and poor sleeper status 62 based on the PSQI global score (~ 5 = poor sleeper). Nonparametric analysis showed no differences in the two categories based on marital status (Pearson X2 = .736, J2 = .305). Socioeconomic Status Net family income on the research tool was divided into eight categories from < $5,000 per year to ~ $70,000 per year. These values were compressed into three categories in order to increase statistical strength. Of those aged < 65, 17.5% made less than $15,000 per year, 41.6% earned $15,000 to $50,000 per year, and 40.8% earned more than $50,000 per year. A percentage of 36.4 of those aged ~ 65 earned less than $15,000 per year, 45.4 % earned $15,000 to $50,000, and 18.2 % earned more than $50,000. Income categories were compared to good and poor sleeper status based on the PSQI global score in the aged ~ 65 category (Table 5). No statistical differences were found between good and poor sleepers based on income status in those aged ~ 65 (Pearson X2 = 1.041, J2 = .594). Educational Status The research tool divided educational level into seven categories: (a) less than an eighth-grade education, (b) 1 to 3 years of high school, (c) a high school or equivalent degree, (d) some college or technical training, (e) an associate degree, (t) a bachelor's degree, and (g) postgraduate education. These categories were compressed into three categories in order to increase the strength of analysis 63 Table 4 Marital and Sleep Status in > 65 Age Group Sleeper status Marital status Good sleeper Poor sleeper Married/Qartnered g=9 g = 16 Within marital status 36.0% 64.0% Within sleep status 69.2% 55.2% Single/widowed/divorced n=4 n = 13 Within marital status 23.5% 76.5% Within sleep status 30.8% 44.8% Note. Sample size varies due to missing data. Table 5 Income and SlegJ Status in ;;:: 65 Age Group Sleeper status Family income Good sleeper Poor sleeper < $15,000 n=4 n=7 Within income 36.4% 63.6% Within sleep status 36.4% 29.2% ~152ooo to ~502000 n=4 n = 13 Within income 23.5% 76.5% Within sleep status 36.4% 54.2% $50,000 n=3 n=4 Within income 42.9% 57.1% Within sleep status 27.3% 16.7% Note. Sample size varies due to missing data. (Table 6). No statistical differences were found between good and poor sleepers based on educational status in the older group (Pearson X2 = 1.506, n = .471). Comparison of Clinical Factors 64 Fifteen cancer sites were represented in the group aged > 65. The most frequently reported cancer site was the breast, with 14 cases reported. Prostate cancer followed, with 8 cases reported. In order to increase statistical significance, leukemias, lymphomas, and myelomas were compressed into a single category (11 individuals). All solid tumors were placed in a category named "others." This category included genitourinary tumors, gastrointestinal tumors, head and neck tumors, and lung cancer. Twenty-four subjects fell into this category. Poor sleepers were most likely to fall in the solid tumor category, followed by hematologic malignancies, and then breast and prostate cancers. No significant differences were found for site of disease and sleeper status in the group aged > 65 (Pearson X2 = 4.674, n = .197) (Table 7). Stage of disease was separated into three categories: (a) local, (b) local/regional, and (c) advanced/metastatic. Stage of disease was found to be statistically significant in determining good versus poor sleepers in the group aged ;;:: 65 (Table 8). Table 6 Education and Sleep Status in > 65 Age Group Sleeper status Educational status Good sleeper Poor sleeper < High school degree High school/associate degree/some college More than associate degree Note. Sample size varies due to missing data. Table 7 Site of Disease and Sleep Status in > 65 Age Group Site Good sleeper Breast !!=6 Within site 54.5% Within sleep status 46.2% Prostate !! = 1 Within site 25.0% Within sleep status 7.7% Hematologic !! = 1 Within site 11.1% Within sleep status 7.7% Other !!=5 Within site 27.8% Within sleep status 38.5% Note. Sample size varies due to missing data. !!=o !!=8 !!=5 Sleeper status !! = 3 !! = 17 !! = 9 Poor sleeper !!= 5 45.5% 17.2% !!= 3 75.0% 10.3% !!= 8 88.9% 27.6% !! = 13 72.2% 44.8% 65 66 Table 8 Stage of Disease and Sleg> Status in > 65 Age Group Sleeper status Stage Good sleeper Poor sleeper Local n=6 n= 2 Within stage 75.0% 25.0% Within sleep status 50.0% 8.0% Regional n=4 n = 14 Within stage 22.2% 77.8% Within sleep status 33.3% 56.0% Advanced n=2 n=9 Within stage 18.2% 81.8% Within sleep status 16.7% 36.0% Note. Pearson X2 = 8.491, df = 2, 12 = .014. Treatment Modality Subjects aged ~ 65 were categorized according to treatment modality (Table 9). Only 7.4% of the poor sleepers came from the category receiving both treatments, but it also had the smallest representation of all groups. No significant differences were found between sleep status and current therapy in the group aged < 65 (Pearson X2 = 4.748,12 = .191). 67 Table 9 Treatment Modality and Slem Status in > 65 Age Group Sleeper status Treatment modality Good sleeper Poor sleeper No current therap~ n=5 n = 10 Within treatment 33.3% 66.7% Within sleep stanis 38.5% 37.0% Chemothera~ onll! n=O n= 7 Within treatment 100.0% Within sleep status 25.9% Radiotherapl! therapl! onll! n=7 n= 8 Within treatment 46.7% 53.3% Within sleep status 53.8% 29.6% Chemotherap~ Rlus n = 1 n= 2 radiotherap~ Within treatment 33.3% 66.7% Within sleep status 7.7% 7.4% Note. Sample size varies due to missing data. Hospitalization Status Only 3 of 47 inpatients were aged > 65. Therefore, it was deemed inappropriate to evaluate this group due to the small sample size. Responses To the Components of Sleep Quality Perceived Sleep Quality Sleep quality was evaluated by each participant based on a 4-point scale (Table 10). Sleep Latency Each participant estimated the length of time required to fall asleep after going to bed (Table 11). Sleep Efficiency 68 Sleep efficiency is the ratio of time asleep over time in bed. This ratio was calculated by using the reported hours slept for the amount of time asleep. The amount of time spent in bed was calculated by subtracting getting-up time from bedtime (Table 12). Sleep Disturbances Nearly 60% of the elderly population reported awakening at least three times during the past week. Getting up to use the bathroom was the most common reason for waking during the night. Coughing or snoring was also responsible for a number of sleep disturbances. Individuals aged;;::: 65 were more likely to report 69 Table 10 Sleep Quality in > 65 Ag;e Group Perceived sleep quality Very good Fairly good Fairly bad Very bad % !l % !l % !l % !l 23.6 13 54.5 30 14.5 8 7.3 4 Table 11 Sleep Latency in > 65 Ag;e Group Sleep latency in minutes < 15 minutes 16 to 30 minutes 31 to 60 minutes ~ 60 minutes % !l % !l % !l % !l 30.2 16 39.2 21 18.9 10 11.3 6 Table 12 Habitual Sleep Efficiency in > 65 Ag;e Group Habitual sleep efficiency ~ 85% 75% to 84% 65% to 74% < 65% % !l % !l % !l % !l 46.0 23 14.0 7 16.0 8 24.0 12 70 feeling too cold as a reason for awakening than feeling too hot. Pain was responsible for disrupting sleep in about half of those reporting. More than one third reported pain three or more times during the past week. Approximately one third reported sleep disruption related to nausea during the past week (Table 13). Use of Medication The medications evaluated in this category are those used specifically to assist in ability to fall asleep or to maintain sleep throughout the night (Table 14). Daytime Dysfunction The score for daytime dysfunction is based on the evaluation of enthusiasm for daily activities and difficulty staying awake (Table 15). The scores for enthusiasm range from 0 to 3, with 0 indicating no problem and 3 indicating a very big problem. Difficulty staying awake to eat, drive, or socialize is scored from 0 to 3, with 3 indicating more difficulty (Table 16). Not during the past week and more than once during the past week are essentially the same measurement. This question was overlooked when the tool was converted from a I-month evaluation period to a I-week evaluation period. This oversight may have confounded the data to some degree. However, since only 5.3% responded to more than once during the past week, the impact may not be significant. Future studies should correct the fields in order to prevent overlapping data. Scores for difficulty staying awake and enthusiasm for getting things done were combined to form a composite score ranging from 0 to 6, with 6 indicating ~ Table 13 Freguencies of SleeJ1 Disturbances in > 65 Age GrouJ1 Not during past week More than once during Disruption past week % !! % !! Sleep latency ~ 30 minutes 43.6 24 20.0 11 Wake up during the night 14.0 8 5.3 3 Get up to use bathroom 7.0 4 7.0 4 Cannot breathe comfortably 80.0 44 9.1 5 Cough or snore loudly 55.6 30 13.0 7 Feel too hot 60.7 34 10.7 6 Feel too cold 54.5 30 5.5 3 Have bad dreams 73.6 39 9.4 5 Have pain 50.9 28 3.6 2 Have nausea 66.1 37 14.3 8 Once or twice during past week % !! 12.7 7 21.1 12 19.3 11 3.6 2 5.6 3 14.3 8 20.0 11 11.3 6 10.9 6 10.7 6 Three or more times during past week % !! 23.6 13 59.6 34 66.7 38 7.3 4 25.9 14 14.3 8 20.0 11 5.7 3 34.5 19 8.9 5 ..........J. 72 Table 14 Use of Sleeping Medication in > 65 Age Group Use of sleeping medication Not during past week Once during past Two or three times More than three times week during past week during past week % !! % !! % !! % !! 50.9 29 1.8 1 14.0 8 33.3 19 Table 15 Enthusiasm in > 65 Age Group Enthusiasm for getting things done No problem at all Only slight problem Somewhat a problem A big problem % !! % !! % !! % !! 17.9 10 32.1 18 14.3 8 35.7 20 Table 16 Difficulty Staying Awake During Day in > 65 Age Group Difficulty staying awake driving, eating, or socializing during past week Not during past week < Once during past Once or twice during More than three times week past week during past week % !! % !! % !! % !! 84.2 48 5.3 3 7.0 4 3.5 2 73 more significant problem (Table 17). Differences in Aeine Suberoups One-way ANOVA was originally planned on age groups from 65 to 74, 75 to 84, and 85 + . Due to insufficient representation in the oldest group, it was determined that analysis would be best served by subdividing those aged > 65 into two subgroups. Those 65 to 74 were placed in one group, and those aged > 75 were placed in the remaining group (Table 18). Independent 1 tests showed no significant differences in sleep quantity or sleep quality between the two elderly subgroups. Table 17 Daytime Dysfunction in > 65 Age Group Daytime dysfunction Zero One or two Three or four Five or six % !! % !! % !! % !! 16.1 9 44.6 25 33.9 19 5.4 3 74 Table 18 Comparison of Sleep Quantity and Ouality in Elderly Subgroups Variable !! X SD ! test Significance (two-tailed) Hours of sl~ 65 to' 74 34 6.6 1.80 ~ 75 25 7.4 1.70 -1.625 .110 Sl~ efficiency 65 to 74 31 .77 .20 ~ 75 24 .81 .15 -.950 .347 PSQI global score 65 to 74 26 9.1 5.20 ~ 75 18 7.4 4.10 1.124 .267 Sl~ guali!y 65 to 74 34 2.2 .87 ~ 75 26 1.9 .86 1.3 .201 CHAPTER V DISCUSSION The generalization of this study would be inappropriate due to the size and relative homogeneity of the sample in terms of ethnicity. Only 30% of this sample were aged ~ 65, which conflicts with data reported earlier suggesting that 60 % of all cancers occur in elderly individuals. These numbers may be consistent with the population served at this institution and, therefore, may not be consistent with the general population. This rmding may indicate a higher refusal rate in this population for participation in the study. Sensory changes accompanied with aging may have affected participation in the study, since elderly individuals are more likely to have hearing and visual impairments. Because elderly adults are more likely to suffer with comorbid conditions, current health status may have affected participation in this study. This convenience sample may also have missed many individuals who are most affected by sleep disturbances. "I'm too tired" was a common response to solicitation into the study, indicating that these particular individuals may have been suffering with sleep disturbances and the fatigue associated with them. Although this response would have underestimated the problem in all groups, elderly individuals were more likely to decline due to fatigue. 76 Differences in sleep between the elderly sample and its younger counterpart were, in general, not seen. In the general population, females, single or widowed individuals, or those of low socioeconomic status are more likely to report sleep disturbances. No differences were seen in sleep efficiency, sleep quality, or the PSQI global score in the elderly subgroup in relation to these variables. One explanation may be that since none of the subjects was considered healthy (due to the diagnosis of cancer), sleep was problematic for the entire group regardless of age. Mean PSQI scores of 8.52 in the elderly sample and 8.06 in the younger sample reinforces that finding. Interestingly, sleep duration showed significant differences between the < 65 age group and the > 65 age group (12 < .002), with younger participants demonstrating shorter sleep duration. A percentage of 41.6 of those aged < 65 slept 6 hours or less compared with 27.2 % of those aged ~ 65. The research tool did not question participants about lifestyle factors that may affect sleep duration. Elderly individuals are less likely to be employed or to have young children driving their activity. They may have the lUxury of going to bed and awakening at their convenience rather than being driven by family activities and alarm clocks. The differences seen in sleep duration did not translate into differences in sleep efficiency, however, even though duration is a component of efficiency. This finding may be due to the fact that sleep duration is based on an interval scale, whereas sleep efficiency was categorical. The distribution of scores in sleep quantity may have evened out the means, thus demonstrating no difference between 77 the groups. Small sample size may have also contributed to this effect. Economic and educational variables may have shown differences if compressed into two categories rather than three, which allowed for more participants in each cell. Correlation studies would be more likely to identify relationships between the demographic variables and sleep. Relationships between clinical variables and sleep in the elderly were also not seen, except in the disease stage. Those with local/regional or advanced disease were more likely to be poor sleepers than those having localized disease only <n < .014). Symptoms associated with tumor invasion seen in regional or advanced disease such as draining lesions, gastrointestinal or genitourinary alterations, pain, obstruction, fever, cough, dyspnea, pruritus, and fatigue may all influence sleep. Differences would have been expected related to site of the disease based on the comparison of breast and lung cancers by Silberfarb and colleagues (1993). A small sample size necessitated the compression of many sites into one group. Only 5 individuals were identified as having lung or bronchial tumors. These small groupings may also have contributed to the results of the analysis. A larger sample may have identified differences related to site in which each site could be studied on its own merit. Differences would also have been expected as a result of hospitalization (Sheely, 1996) but could not be examined due to the small number of elderly participants in the inpatient sample (n = 3). Treatment modality also did not appear to be responsible for differences in sleep quality. In all clinical analyses, the strength of analysis was weakened by the fact 78 that several cells contained fewer than 5 subjects. An analysis of the sample in the larger study may have identified relationships between clinical variables and sleep. Of those aged ~ 65, 21. 8 % reported sleep that was fairly bad to very bad. The median score fell in the fairly good to fairly bad category. Nearly 50 % of the elderly subjects reported the use of sleeping aids at least twice per week, with 33.3% taking medications for sleep three or more times per week. This fmding is in contrast to the elderly population at large that usually sees the use of medication as a sign of weakness and is reluctant to use pain or sleeping aids. However, the frequency of using sleeping aids in this sample may explain, in part, why it tends to report good sleep_ " Daytime dysfunction is a combination of difficulty staying awake and lack of enthusiasm. Approximately 85 % of the group reported no difficulty staying awake during the past week, yet nearly 40 % had a score of ~ 3 on daytime dysfunction, suggesting that daytime dysfunction is at least somewhat of a problem but is not related to sleepiness. Fifty percent of the elderly participants also reported lack of enthusiasm to be at least somewhat of a problem. Daytime dysfunction in this population may have more to do with disease or treatment related fatigue than sleep quantity or quality. Daytime dysfunction might also be influenced by environmental factors that were not evaluated in the tool. People with many demands on their time may not have the lUXUry of discovering if they are suffering with fatigue, whereas someone who is bored may fmd it problematic. 79 Nearly 60% of the older subjects reported nocturnal awakening at least three times during the week. Most of the nocturnal awakenings were related to the use of the bathroom, pain, coughing, or snoring, respectively. Despite the frequency of reports of nocturnal awakenings, nearly 80 % of the participants still reported very good to fairly good sleep. These individuals may have learned to adapt to the nighttime interruptions and not to see them as problematic. Buysse and colleagues (1991) reported that healthy elderly adults tend to report "good" sleep despite verification otherwise in the laboratory and on the PSQI measurement tool. It is believed that they tend to change their perceptions of sleep and learn to adapt despite· evidence of disturbed sleep. There may be a similar phenomenon in these subjects, which is in contrast to a characteristic found in individuals with chronic insomnia who tend to have a poorer perceived sleep than laboratory data would indicate (Dement et aI., 1982; Silberfarb et aI., 1993). Healthy adults usually fall asleep approximately 10 to 15 minutes after retiring. By that criterion, 80 % of the sample have problematic sleep latency. Thirty percent of the elderly sample reported their sleep latency to be greater than 30 minutes, and 24 % reported that this had been a problem at least three times during the past week. Clearly, since only 20 % of the sample fell within the normal range, these results indicate that sleep latency is an issue that needs to be addressed in the elderly oncology population. It is difficult to say at which point sleep efficiency becomes problematic. In healthy adults, sleep efficiency is usually approximately 90% (Haimov & Lavie, 80 1997). Only 46% of the elderly sample fell within this category. If looked at from another perspective, 40% of the sample spent at least one fourth of their time in bed in wakefulness, and 24 % spent at least one third of their time in bed in wakefulness. Therefore, it would appear that sleep efficiency is a problem in the elderly group. A discrepancy was found; that is, the mean PSQI global score in the older sample was 8.52, which is considered poor sleep according to the authors of the tool. Despite that score, nearly 80% of the subjects reported very good to fairly good sleep. A wider range in scoring options for perceived quality may have found a closer relationship between the global score and the perceived score. The nlean sleep efficiency for this group was 78 %, which is also below average for healthy adults (90 %). Only about half (46 %) of the older sample were found to have a sleep efficiency of greater than 85 % . While scores for sleep latency, sleep efficiency, and overall sleep quality seem to indicate poor sleep, perceived sleep quality seems to indicate otherwise. This finding may indicate that perceptions about sleep have changed. Further research is recommended to determine the validity of this tool in oncology populations. Correlation studies would be more likely to identify relationships between the given variables and sleep. No questions on the tool were related to daytime napping. According to the literature review, elderly persons are more inclined to distribute sleep across a 24- hour day. The PSQI does not question the frequency or duration of napping. Less-than-healthy individuals of all age groups may be more inclined to nap in 81 order to compensate for a poor night's sleep. Inclusion of sleep time in the form of naps may have had a significant impact on the overall sleep quantity, sleep quality, or both. The PSQI also does not question participants about a prior history of sleep disturbances or depression, which may influence outcomes. These individuals tend to score higher on the PSQI global score (Buysse et al., 1989). No differences were found in different age subgroups in the older subjects. Only 1 participant was aged;;;;:: 80. According to Hoch and colleagues (1997), old-old cohorts are those most likely to show reduced sleep, especially as related to deteriorating health. This sample showed a greater mean in those aged > 75 (7.3) than in those aged 65 to 74- (6.6) .. Sleep efficiency was also greater in the oldest subgroup (.82 and .76, respectively), The older subgroup had better subjective sleep quality scores and better global scores than those aged 65 to 74. More representation in this group, especially among those aged > 85, may have led to different findings. A percentage of 69.6 of the elderly participants also completed the survey themselves compared with 85.9 % of those aged < 65, which may have misrepresented the fmdings in the older sample. Elderly hospitalized persons were more likely than younger hospitalized persons to be identified as being "too sick" to participate in the study. Understanding sleep disturbances in cancer patients is incomplete without their contribution. Nursine Implications Findings related to sleep and stage of disease have implications for nurses working in oncology. The early identification of those at risk will help nurses 82 target information to people most likely to be affected by disturbed sleep, particularly those with advanced stage disease. Information about the sleep process and sleep hygiene may help elderly adults with regional or advanced disease to develop healthy sleep patterns, thus avoiding the sleep disturbances associated with advancing disease in this population. This study indicates that oncology patients are likely to have disruptions in sleep several times during a given week. Identifying the causes of sleep disturbances in elderly oncology populations can help nurses to develop interventions to help patients minimize nighttime interruptions. Before interventions can occur, however, nurses need to incorporate a thorough sleep assessment when evaluating their clients. Depending on the type of sleep disturbance, information and intervention can assist patients in minimizing the number and duration of those interruptions. Information regarding fluid intake and medications causing urinary retention may help elderly patients to decrease the number of nocturnal awakenings associated with the need to void. Older participants continue to report nocturnal awakenings related to pain, indicating that nurses must do a better job at intervening in pain management and at helping their patients know how to manage pain. When sleep disturbances are identified through thorough assessment, specific interventions can be designed for each individual. Recommendations for Further Research Sleep studies need to be replicated in other samples of cancer patients in order to validate these findings and to increase application to the general public. 83 This study is limited; that is, it does not reflect the percentage of elderly with cancer in the general population, and it also does not reflect diversity in culture. The number of hospitalized elderly was also insufficient to analyze. Different tools associated with sleep should be studied to determine if the fmdings in this study can be replicated. Of particular value would be the development of sleep evaluation tools specific to oncology patients in order to capture the sleep experience as it relates to them. The development of tools to help clinicians assess sleep in their clients is also recommended. A thorough, ongoing evaluation of sleep and sleep disruption sho