Effects of Depression on Sleep Quality, Fatigue, and Sleepiness in Persons With Restless Legs Syndrome Norma G. Cuellar, Sarah J. Ratcliffe, and Darleen Chien

BACKGROUND: Restless legs syndrome is a sleep disorder that is associated with depression and poor sleep quality; it possibly affects sleepiness and fatigue, thereby affecting quality of life. OBJECTIVE: To determine the effect of depression on sleep quality, sleepiness, and fatigue in persons with restless legs syndrome. STUDY DESIGN: Descriptive, comparative study. Data was collected on demographics, depression, sleep quality, sleepiness, and fatigue. Participants were grouped based on depression scores. RESULTS: 40% of the sample reported depressive symptomology. Depressed participants had significantly worse sleep quality (t = 4.12, df = 40, p < .001) and fatigue (t = 3.69, df = 46, p = .001). Depression did not affect sleepiness (p = .733). CONCLUSIONS: Persons with restless legs syndrome who are depressed have poorer sleep quality and higher fatigue than nondepressed persons with restless legs syndrome. Few participants are being treated for depression. Health care providers must recognize the impact that depression has on persons with restless legs syndrome and develop innovative nonpharmacological strategies to help with depression. J Am Psychiatr Nurses Assoc, 2006; 12(5), 262-271. DOI: 10.1177/1078390306295070

Keywords:

restless legs syndrome; depression; sleep; fatigue; sleepiness

Depression is a serious mental health disorder that affects up to 19 million adults in the United States and affects people of all ages, gender (women twice the rate of men), race, ethnicity, and socioeconomic status (Kessler, Chiu, Demler, Merikangas, & Walters, 2005; Regier et al., 1993). Depression affects work performance (Haslam, Atkinson, Brown, & Haslam, 2005), with the annual cost of missed work days and lost productivity related to depression estimated at more than $83 billion, including medical and pharmaceutical bills (Han & Wang, 2005). Of all of the symptoms of depression, sleep disturbances can be the most debilitating.

Insomnia leads to daytime sleepiness and fatigue and affects coping skills, accomplishment of tasks, mood, relationships, and family and social life (Hui et al., 2002; Roehrs & Roth, 2005; Walker, Fine, & Kryger, 1995). Restless legs syndrome (RLS) is a sleep disorder with a strong association with depression; however, little is known about the relationship of depression with sleep, fatigue, and sleepiness as well as the effectiveness of treatment for depression in persons with RLS. The purpose of this secondary data analysis was to examine the effects of depression on sleep quality, sleepiness, and fatigue in persons with RLS.

Norma G. Cuellar, DSN, RN, assistant professor, University of Pennsylvania School of Nursing, Philadelphia, PA; [email protected] or [email protected].

DEPRESSION

Sarah J. Ratcliffe, PhD, is an assistant professor of biostatistics, Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania School of Medicine, Philadelphia, PA. Darleen Chien, BS, is with the University of Pennsylvania School of Nursing, Philadelphia, PA. This study was partially funded by the Hartford Center for Geriatric Nursing Excellence Jones Fund, the University of Pennsylvania School of Nursing, and the American Association of Diabetic Educators/Sigma Theta Tau International.

262

Depression is a serious medical condition, second only to hypertension, and is the leading cause of disability years (Han & Wang, 2005). According to the National Institute of Mental Health (2001), major depressive disorder is identified when a person has five or more symptoms of depression (Table 1) coinciding with impairment in everyday functioning for at least 2 weeks. Despite the high prevalence and awareness of depression, it is often underdiagnosed and results in poor outcomes of treatment (Gensichen Copyright © 2006 American Psychiatric Nurses Association

Restless Legs Syndrome TABLE 1.

Symptoms of Depression

Sad mood Loss of interest or pleasure in activities that were once enjoyed Change in appetite or weight Difficulty sleeping or oversleeping Physical slowing or agitation Energy loss Feelings of worthlessness or inappropriate guilt Difficulty thinking or concentrating Recurrent thoughts of death or suicide

ability to complete cognitive tasks like memory, learning, language processing, and sustained attention) and an increased incidence of amnesia (Durmer & Dinges, 2005). More than half of all patients with insomnia have depression, with a higher prevalence of both insomnia and depression seen in women (Thase, 2005). Of those people with breathing-related sleep disorders, 18% have major depressive disorder with odds of 5.26 of having a major depressive disorder diagnosis compared with those without sleep disorders (Ohayon, 2003). RESTLESS LEGS SYNDROME

et al., 2005; Han & Wang, 2005). Primary care physicians often overlook symptoms of depression, resulting in failure to treat depression in 50% of patients (Simon et al., 1995; Simon & VonKorff, 1995). Treatment of depression is considered successful when there is 50% reduction in depressive symptoms. However, symptoms often reoccur and are linked to relapse, disability, and suicide (Badawy, 2003; Goldney, 2005; Hall, 2006; Han & Wang, 2005; Sondergard, Kvist, Andersen, & Kessing, 2006; Tardiff, Marzuk, & Leon, 2002; Worm, Dragsholt, Simonsen, & Kringsholm, 1999). Although antidepressants are beneficial in the treatment of major depression, they are only effective in 60% to 70% of cases (Janicak et al., 2004). When depression exists with comorbid conditions, the effects of depression may be compounded and more difficult to treat (Mitchell & Subramaniam, 2005). Recent studies show that persons with repeated depressive episodes or late-onset depression are least likely to be successfully treated and more likely to experience cognitive and functional impairments (Driscoll et al., 2005). Only one third of all people with depression seek help, are diagnosed, and receive appropriate treatment (Kessler et al., 2003). This tendency to forgo treatment may be related to social stigma, which has been found to be more prevalent in minorities than Caucasians (Barney, Griffiths, Jorm, & Christensen, 2006; Das, Olfson, McCurtis, & Weissman, 2006). Treatment with antidepressant medications produces undesirable side effects including nausea and sexual dysfunction. Moreover, antidepressants have been linked to severe side effects such as suicide, psychosis, and violence (Simon, Savarino, Operskalski, & Wang, 2006). More than 80% of people suffering from depression have disordered sleeping patterns and habits, often resulting in sleep deprivation (National Institute of Mental Health, 2001). Sleep deprivation results in abnormal brain function (a decreased

Restless legs syndrome (RLS) is categorized as both a sleep disorder and a movement disorder affecting up to 15% of the population. A genetic component is seen in approximately 50% of cases (idiopathic RLS; Desautels et al., 2001), whereas secondary RLS is associated with comorbid conditions. Most persons are affected by middle age, with 40% reporting onset of symptoms before the age of 20 (Walters et al., 1996). The symptoms progressively worsen with age, increasing in severity and frequency. Symptoms of RLS cause delayed sleep latency, often preventing sleep until early morning hours, such as 2 to 3 a.m. The sleep disorder in RLS causes increased REM sleep and decreased stage I sleep as well as increased awakenings after sleep latency (Eisensehr et al., 2001; Hening, 2004). Sleep latency (or sleep onset) was measured in persons with RLS at 100.3 min (±26.3) compared with persons who do not have RLS at 59.2 min (±30.5) using polysomnography, the gold standard for sleep measurements (Mizuno, Mihara, Miyaoka, Inagaki, & Horiguchi, 2005). More than 80% of persons with RLS experience periodic limb movement disorder (PLMD) that causes awakenings throughout the sleep cycle and involves involuntary leg twitching or jerking during sleep that occurs every 10 to 60 s, often throughout the night. PLMD during sleep (PLMS) is considered pathological if greater than five per minute (Nicolas, Michaud, Lavigne, & Montplaisir, 1999; Ohayon & Roth, 2002) Depression has been reported in 67% of persons with RLS (Blattler & Muhlemann, 1982). Persons with RLS have reported significantly higher depression scores than persons without RLS (Rothdach, Trenkwalder, Haberstock, Keil, & Berger, 2000). Persons with RLS have characteristics of major depression as well as generalized anxiety disorders with a reduced quality of life (Saletu et al., 2002; Tanaka et al., 1999; Ulfberg, Nystrom, Carter, & Edling, 2001). Using electroencephalographic

Journal of the American Psychiatric Nurses Association, Vol. 12, No. 5

263

Cuellar et al.

mapping, Saletu et al. (2002) confirmed evidence of neurophysiological correlates with self-reported depressive symptoms in persons with RLS. Higher levels of depression and anxiety have been associated with a poorer quality of life among patients with RLS (Atkinson et al., 2004; Hening et al., 2004; Rothdach et al., 2000). When a person is treated with antidepressants for depression or anxiety, RLS symptoms are often exacerbated, resulting in discontinuing the medication, which leaves the person with RLS without any pharmacological treatment options. To explore the effects of depression on sleep in persons with primary and secondary RLS, we conducted a secondary analysis using self-rating scales. The effects of depression on sleep quality, sleepiness, and fatigue were examined in this descriptive, comparative study. METHODS The purpose of this study was to determine if depression had an effect on sleep quality, sleepiness, and fatigue in persons with RLS. The primary outcome was sleep quality with secondary outcomes of sleepiness and fatigue. Participants RLS participants were diagnosed according to established diagnostic criteria by the International Restless Legs Syndrome Study Group (Walters, 1995; Table 2). The sample consisted of 53 participants with RLS who were recruited from the University of Pennsylvania Rodebaugh Diabetes Center, University of Pennsylvania Sleep Center, and RLS support groups in the city of Philadelphia and surrounding suburban area. The sample came from two previous studies that were combined, which examined persons with primary and secondary RLS and diabetics with RLS who responded to self-report questionnaires between October 2004 and February 2006. Institutional review board approval was obtained. The study was described in a phone interview, and verbal consent to mail the survey to the participants was obtained. The consent was signed by the participant in the home and returned with the self-report measurements. Measurements Demographic data were collected along with family history of RLS, age of onset of RLS, comorbid conditions, health promotion behaviors, pharmacological 264

TABLE 2.

Diagnostic Criteria for RLS

1. Uncomfortable sensations in the legs that 2. Get worse with rest, 3. Get worse at night, and 4. Are relieved by movement.

history, and complementary and alternative medicine. The following self-report measures were used. The Center for Epidemiologic Studies Depression Scale (CES-D) was used to measure self-reported depressive symptoms (Radloff, 1977) including an index of cognitive, affective, and behavior depressive features and the frequency the symptoms have occurred. A score of 0 to 15 indicates no depression; 16 to 20 indicates mild depression; 21 to 30 indicates moderate depression; and 31 or higher indicates severe depression. The CES-D was developed for the nonpsychiatric population aged 18 and older and has a reported reliability of .90 (Devins & Orme, 1985; Radloff, 1977). The Pittsburgh Sleep Quality Index (PSQI) was used to measure subjective quality and sleep disturbances. Participants were asked to self-rate sleep quality on a scale of 0 to 3 with a score of 0 indicating no difficulty and a score of 3 indicating severe difficulty over the past month with a possible range of 0 to 21 points. The 19 items of the PSQI examine seven components including subjective sleep quality, sleep latency (or onset), sleep duration, habitual sleep efficiency (the ratio of total sleep time to time in bed), sleep disturbances, use of sleeping medications, and daytime dysfunction. Scoring of the PSQI is specific to each of the seven components. Higher scores indicate worse sleep quality. The PSQI has a reliability of .83 (Inoue, Nanba, Honda, Takahashi, & Arai, 2002). The Epworth Sleepiness Scale (ESS) is considered the gold standard scale for measuring daytime sleepiness (Johns, 1991, 1992). Participants are asked to rate the chances that they would doze off or fall asleep in eight situations encountered in daily life. Using a scale of 0 (no chance of dozing) to 3 (high chance of dozing), participants can determine their daytime sleepiness using a mean score, with a score of 9 and above indicating sleepiness. Test–retest reliability is .82. The Fatigue Severity Scale (FSS) is an instrument designed for participants to rate statements that distinguish fatigue from depression (Krupp & Elkins, 2000; Krupp, LaRocca, Muir-Nash, & Steinberg, 1989; Schwartz, Jandorf, & Krupp, 1993). Based on a 7-point Likert scale, the FSS consists of nine questions of

Journal of the American Psychiatric Nurses Association, Vol. 12, No. 5

Restless Legs Syndrome

self-reported fatigue. The participant is asked to read each statement and circle a number from 1 to 7 regarding feelings of fatigue in the past week, with a low fatigue rating indicating no fatigue. Scoring is done by averaging the response of the questions (adding up all the responses and dividing by 9). Persons with fatigue generally score a mean of 6.5 or higher. FSS scores were found to be unrelated to depression scores using the CES-D. Items 1, 2, 3, 4, 6, and 8 address the quality of fatigue, whereas items 5, 7, and 9 rate the social consequences of fatigue rather than fatigue itself. Cronbach alphas have been reported at .88 (Krupp et al., 1989). Statistical Methods Depression status was calculated based on a CESD score of 16 or more. The sample was split to identify two groups for comparison: the RLS group with depression (n = 21) and the RLS group without depression (n = 32). Demographic variables were compared by depression status via t tests, MannWhitney U test, chi-square test, or Fisher’s exact tests to help determine any potential confounders. These methods were also used to compare the outcomes by depression status. Regression analyses were conducted to determine if any significant differences found in the previous step were still important after adjustment for other potential confounders. RESULTS Demographic Data Fifty-three participants were included in the study. Demographic characteristics are compared in Table 3. The majority of participants were female (62.3%), were married or had a partner (59%), and were retired (60%); the average age was 68 years (range 37-93). The ethnicity of participants was Caucasian (83%), African American (9%), Asian or Pacific Islander (2%), and mixed (6%), although all non-Caucasians were grouped for analyses. Two thirds of the respondents had a high school education or higher. The majority (56%) of participants who reported an income made less than $40,000 per year. Only 21 of the 53 participants were being treated with medications for RLS. Depressive Symptoms (CES-D) The mean CES-D score from the entire sample was 13.91 with a standard deviation of ±9.52, ranging

from 0 to 36. Thirty-two (60%) were classified as not depressed (scoring 0-15) and 21 (40%) were depressed (scoring 16+); 6 were mildly depressed, 13 moderately depressed, and 2 severely depressed. Significant differences were found between depression and marital status (odds ratio [OR] = 3.26, 95% confidence interval [CI] = 1.02-10.41, p = .046), ethnicity (OR = 7.5, 95% CI = 1.38-40.84, p = .020), and education (t = 2.53, df = 20, p = .015). Single participants with RLS were approximately three times more likely to be depressed than participants who were married/ partnered. Non-Caucasians were approximately 7.5 times more likely to be depressed than participants who were Caucasian. Depressed participants had less education than nondepressed participants, on average (15.6 vs. 13.5 years). Nine (17%) of the participants (n = 53) were taking antidepressant drugs. Of this 9, 5 (56%) were depressed; thus, only 44% of antidepressant users were being successfully treated for depression. Taking antidepressants had no statistically significant effect on depression (OR = 2.2, 95% CI = 0.519.34, p = .456). Furthermore, only 5 (24%) of the 21 participants classified as depressed on the CES-D were receiving treatment for their depression. Subjective Sleep Quality (PSQI) The total PSQI score was calculated on 42 participants; because of some missing values, individual components of the PSQI had more respondents. The mean PSQI score was 10.8 (±4.4), ranging from 2 to 19. Depressed participants had significantly worse sleep quality (13.7 vs. 8.8, t = 4.19, p < .001). In addition, 37 (88.1%) of these participants were classified as poor sleepers (scoring 5+). The 5 good sleepers were not depressed (see Table 4). A regression model was generated to determine if differences in the global PSQI score by depression status were still significant after we controlled for age, ethnic group, marital status, education, body mass index (BMI), and coffee consumption. There was a significant difference in the global PSQI scores by depression status (b = 3.58, 95% CI = 0.496.67, p = .025). Depressed participants had a 3.6point higher global PSQI score than nondepressed participants, on average. Of the individual components of the PSQI, depressed participants scored significantly higher (worse) on subjective sleep quality (component 1, p = .002), habitual sleep efficiency (component 4, p = .039), sleep disturbance (component 5, p = .001), and daytime dysfunction (component 7, p = .033, see Table 4 and Figure 1).

Journal of the American Psychiatric Nurses Association, Vol. 12, No. 5

265

Cuellar et al. TABLE 3.

Demographic Characteristics of the Sample, Overall and Depression Status

Characteristica Female Age (years) Married/partnered Caucasian Years in school Employment status Full time Part time Retired Unemployed Work in home Disability Socioeconomic status ($1,000)b 0-20 20-40 40-60 60+ Cups of coffee/day 0 1 2+ Other caffeinated drinks/day 0 1+ Smokes Family history of RLS Exercise/week (hr) BMI

All Participants (n = 53)

Depressed (n = 21)

Not Depressed (n = 32)

33 (62.3) 67.6 (±11.1) 31 (58.5) 44 (83.0) 14.8 (±3.1)

15 (71.4) 64.5 (±13.1) 9 (42.9) 14 (66.7) 13.5 (±2.8)

18 (56.3) 69.7 (±9.1) 22 (71.0) 30 (93.8) 15.6 (±3.0)

4 6 32 1 3 6

(7.5) (11.3) (60.4) (1.9) (5.7) (11.3)

2 2 11 0 1 5

(9.5) (9.5) (52.4) (0.0) (0.5) (23.8)

2 4 21 1 2 1

(6.5) (12.9) (67.7) (3.2) (6.5) (3.2)

10 15 11 9

(18.9) (28.3) (20.7) (17.0)

9 7 2 0

(42.9) (33.3) (9.5) (0.0)

1 8 9 9

(3.2) (25.8) (29.0) (29.0)

p Value2 .268 .097 .046 .020 .015 .384

.001

.043 15 (28.3) 10 (18.9) 28 (52.8)

9 (42.9) 1 (4.8) 11 (52.4)

6 (18.8) 9 (28.1) 17 (53.1)

30 (56.6) 22 (41.5) 5 (9.4) 30 (56.6) 4.1 (±3.3) 29.8 (±7.4)

7 (33.3) 14 (66.7) 2 (9.5) 10 (47.6) 3.7 (±4.0) 31.3 (±6.8)

23 (71.8) 8 (25.0) 3 (9.4) 20 (64.5) 4.3 (±2.9) 28.8 (±7.8)

.005

.986 .258 .513 .235

Note. RLS = restless legs syndrome; BMI = body mass index. Frequency (%) or mean (±SD) is reported. a Responses within a characteristic may not sum to 100% because of missing values; bp values comparing depressed and not depressed participants via t test, Mann-Whitney U test, chi-square test, or Fisher's exact test, as appropriate.

TABLE 4. Comparison of Pittsburgh Sleep Quality Index (PSQI) Global and Component Scores by Depression Status

Global PSQI score Sleep category Good Poor Component scores 1. Subjective sleep quality 2. Sleep latency 3. Sleep duration 4. Habitual sleep efficiency 5. Sleep disturbance 6. Sleep medications 7. Daytime dysfunction

All Participants (n = 42)

Depressed (n = 17)

Not Depressed (n = 25)

10.8 (±4.4)

13.7 (±3.0)

8.8 (±4.1)

5 (11.9) 37 (88.1)

0 (0.0) 17 (100.0)

1.8 1.6 1.5 1.7 1.7 1.6 1.4

(±0.9) (±1.1) (±0.9) (±1.1) (±0.7) (±1.2) (±0.8)

2.2 (±0.7) 1.9 (±0.9) 1.8 (±1.0) 2.1 (±1.1) 2.1 (±0.5) 1.9 (±1.2) 1.7 (±0.7)

p Valuea