ADVERTISEMENT

Issue Navigator

Volume 11 No. 10
Earn CME
Accepted Papers





Scientific Investigations

Feasibility and Efficacy of a Self-Management Intervention for Insomnia in Stable Heart Failure

Nancy S. Redeker, PhD, RN1; Sangchoon Jeon, PhD1; Laura Andrews, PhD, RN1; John Cline, PhD2; Daniel Jacoby, MD2; Vahid Mohsenin, MD2
1Yale University School of Nursing, New Haven, CT; 2Yale University School of Medicine, New Haven, CT

ABSTRACT

Background:

Chronic insomnia is common among patients with heart failure (HF) and may contribute to fatigue and poor function. However, to date there have been no randomized controlled trials focused on treatment of insomnia or daytime symptoms in this population.

Objective:

The purpose of this study was to examine the preliminary efficacy, feasibility, and acceptability of a self-management intervention (cognitive behavioral therapy [CBT-I]) for insomnia among patients with stable HF.

Methods:

We conducted a pilot randomized controlled trial (RCT) in which patients with stable Class I-III HF (n = 25/52.1% women; mean age = 59 ± 14.8 years) were randomized in groups to CBT-I (n = 29) or an attention control condition (HF self-management with sleep hygiene; n = 19). Participants completed 2 weeks of wrist actigraphy, the insomnia severity index, and measures of fatigue, depression, sleepiness, and functional performance at baseline and follow-up. We computed the size of the effects on the dependent variables and used MANOVA to evaluate the effects of CBT-I on insomnia and fatigue.

Results:

CBT-I was feasible and acceptable and had a statistically significant effect on insomnia and fatigue, while controlling for the effects of comorbidity and age.

Conclusions:

CBT-I has short-term efficacy as a treatment for chronic insomnia among patients with stable HF. Future studies are needed to address its sustained effects.

Citation:

Redeker NS, Jeon S, Andrews L, Cline J, Jacoby D, Mohsenin V. Feasibility and efficacy of a self-management intervention for insomnia in stable heart failure. J Clin Sleep Med 2015;11(10):1109–1119.


Heart failure (HF), a disabling chronic condition that afflicts over five million Americans,1 is associated with excess morbidity and mortality, comorbidity, poor daytime function, and high symptom burden. Almost 75% of HF patients report poor sleep.2,3 Insomnia, characterized by difficulty initiating and maintaining sleep, early morning awakenings, non-restorative sleep, and daytime dysfunction, occurs in 25% to 56% of HF patients,25 who report higher rates than “healthy” controls2 and the general population.

Chronic insomnia contributes to incident HF and death.6 It is distressing and associated with symptoms, such as fatigue, nocturnal dyspnea, depression, anxiety, pain, and excessive daytime sleepiness, poor quality of life, and decrements in functional performance.3,5,713 Health care providers frequently attribute sleep disturbance, including insomnia symptoms, to the pathophysiology (e.g., fluid congestion) and symptoms of HF itself (e.g., nocturnal dyspnea, nocturia) or to sleep disordered breathing that occurs in about half the population.14 However, sleep disordered breathing does not consistently explain insomnia, self-reported sleep quality, or fatigue; and insomnia was closely associated with daytime symptoms (fatigue, sleepiness, and depression) and both self-reported and objectively measured functional performance among patients with stable HF who received evidence based disease management.5 Although researchers and clinicians suggest the use of self-management strategies, including adherence to medications and fluid and sodium management, as ways to improve sleep disturbance and other symptoms,15 there is little evidence that these strategies improve insomnia or fatigue in HF patients. The strategies also do not correct perpetuating factors for chronic insomnia such as increased time in bed awake,2 perceptions of insufficient sleep, frequent napping,2,4 and use of the TV as white noise12 among HF patients. Negative associations between sleep hygiene behavior and sleep quality16 among HF patients underscore the importance of addressing these factors. Yet, HF patients report that health professionals rarely ask about or provide interventions for sleep or insomnia symptoms.12

BRIEF SUMMARY

Current Knowledge/Study Rationale: Although chronic insomnia is common among patients with chronic heart failure (HF), little is known about the efficacy of insomnia treatment in this population. The purpose of this study was to examine the preliminary efficacy, feasibility, and acceptability of cognitive behavioral therapy for insomnia (CBT-I) among patients with stable HF.

Study Impact: CBT-I is feasible, acceptable and improves insomnia, sleep efficiency and fatigue among HF patients, a group that often suffers from intractable fatigue and insomnia for which few efficacious self-management interventions have been available. These findings suggest the importance of identifying and treating insomnia in these patients and support the need for future randomized controlled trials of CBT-I in this population.

The most efficacious treatments for insomnia are hypnotic medications and cognitive behavioral therapy for insomnia (CBT-I), a multi-modal behavioral self-management intervention that addresses perpetuating factors for insomnia (dysfunctional thoughts, beliefs, and behaviors). CBT-I has more durable effects on insomnia than hypnotic medications17,18 and lacks the adverse effects of these drugs. CBT-I has well-documented benefits, including improvements in insomnia, sleep continuity, and daytime symptoms among patients with a number of comorbid medical conditions.1933 CBT-I also may improve sleep for HF patients who dislike the adverse effects of hypnotic medications12 and are at especially high risk for insomnia and disabling daytime symptoms34 that are often attributed to HF itself. However, the effects of CBTI in HF patient have not been examined, and it is not clear that the results of previous studies of CBT-I in people with other chronic medical conditions are generalizable to the HF population.

The aims of this study were to (1) evaluate the feasibility and acceptability of CBT-I provided in a group format to patients with HF; and (2) evaluate the size of the effects of CBT-I, compared to HF self-management education (attention control) with sleep hygiene education, on (a) insomnia and self- and actigraph- recorded sleep characteristics (quality, duration, sleep continuity); (b) daytime symptoms (fatigue, excessive daytime sleepiness, anxiety, depression), and (c) functional performance among HF patients at two weeks post-CBT-I.

METHODS

Design

We conducted a pilot randomized controlled trial (RCT), in which HF patients were randomized in groups of 4–5 to group-based CBT-I or an attention control condition, consisting of HF self-management education and sleep hygiene education (Table 1). Patients completed baseline and then follow-up assessment at the conclusion of the 8-week treatment. The institutional review board approved the study, and all participants provided written informed consent.

Components of the cognitive behavioral therapy for insomnia (CBT-I) and attention control (healthy hearts) conditions.

jcsm.11.10.1109.t01.jpg

table icon
Table 1

Components of the cognitive behavioral therapy for insomnia (CBT-I) and attention control (healthy hearts) conditions.

(more ...)

Setting and Sample

Health care providers in a HF disease management program in a tertiary care center referred HF patients to the study. We also advertised the study in local newspapers and with postings on the university clinical trials website and searched the discharge database of a tertiary care hospital and the patient database of a community primary care center after obtaining a HIPAA waiver.

Included participants were over the age of 18, cognitively intact by clinical impression, lived at home, had New York Heart Association Classification I-III HF, and an Insomnia Severity Index > 7,35 indicating at least mild chronic insomnia. Participants who had significant sleep disordered breathing and were adherent to positive airway pressure treatment for ≥ 6 h/6 nights/week were included. Individuals were excluded if they had untreated sleep disordered breathing (apnea-hypopnea index ≥ 10), restless legs syndrome, narcolepsy, performed night or rotating shift work, were cognitively impaired by clinical impression, or had renal failure. Potential participants were excluded if they had seizure disorders or were excessively sleepy (Epworth Sleepiness Scale scores > 18) because these are contraindications to sleep restriction, a component of CBT-I. We excluded individuals who had neurological or musculoskeletal disorders that restricted movement of the non-dominant arm because of the possible confounding effects on wrist actigraph recordings.

Although evaluation of statistically significant effects was not the primary goal of this study, we calculated the power to detect a group by time interaction for sleep efficiency (SE) using effect sizes obtained from similar studies that found moderate-large effects on these variables.19,36 We estimated that 48 subjects (29 in treatment vs. 19 in control) would provide 95% power to detect a time effect size of 1.10 and 76% power to detect a group by time effect size of 0.80 for SE.

Procedures

We screened potential participants by interview and medical record review to evaluate study eligibility. Potential participants who had not previously been evaluated for sleep disordered breathing were screened for 2 nights in their homes with the Apnea Risk Evaluation System (ARES) recorder (Watermark Medical, Inc.), a wireless recorder worn on the forehead. ARES measures blood oxygen saturation (SpO2) and pulse rate (reflectance pulse oximetry), airflow (nasal cannula/pressure transducer), respiratory effort (pressure transducer-forehead venous pressure), venous volume (photo-plethysmography), snoring (acoustic microphone), head movement, and position (accelerometers). Pulse rate changes indicate arousals. Snoring changes indicate respiratory-related arousals. Forehead venous pressure is a valid measure of respiratory effort,37 while head position indicates the positionality of obstructive events. ARES distinguishes REM from NREM with EEG, EOG, and EMG electrodes at FP1 and FP2.

The apnea-hypopnea index (AHI) is the sum of apneas and hypopneas/h with 4% oxygen desaturation. ARES has high sensitivity and specificity3739 and better test-test reliability40 than in-lab PSG. We recorded 2 nights and calculated respiratory events/total sleep time because of potential night-night variability and scored the data through the manufacturer's website with validated automated methods and review by a board certified sleep technician.

We paid participants $25 for completing the ARES screening. All screened participants received a report of the screening with recommendations for follow-up, if indicated.

Participants who met inclusion criteria were randomized, using an electronically generated randomization sequence, to either CBT-I or the attention-control condition in groups of 4 or 5, but were not informed of the group assignment until the group meeting. At the time of consent, participants randomized to the self-management group (attention-control) were told that improvement in HF by better self-management and sleep hygiene could improve their sleep. The clustered approach was used to assure filling of the groups in a timely fashion. Participants completed 2 weeks of baseline evaluation, including 24-h wrist actigraphy and sleep diaries and study questionnaires, and then participated in CBT-I or the attention control conditions. Measurements were repeated 2 weeks after group participation. Participants received a total of $250 in increments across the study milestones (baseline data collection through final telephone follow-up), in addition to the payment for sleep apnea screening, for their participation.

A clinical psychologist who was board-certified in sleep medicine provided the CBT-I in 4 biweekly 1-h sessions over an 8-week period with telephone calls on intervening weeks. Participants were offered the opportunity to taper hypnotic medications, and this was discussed in session 2, but it was not required for participation. The elements of the CBT-I were provided according to standard CBT-I methods and are listed in Table 1.

The Attention Control Condition included HF self-management education based on the American Association of Heart Failure Nurses' Fight against Heart Failure Handbook41 and sleep hygiene education (Table 1). This content was selected because it is recommended for HF patients,15 and HF self-management may improve insomnia and other symptoms by improving control of HF itself. An advanced practice nurse (APRN) led the attention control group in exactly the same format as the CBT-I group to control for time and attention. It is common practice for a registered nurse or an APRN to perform HF self-management education.

We provided both groups with investigator-designed workbooks that outlined the session content. All participant materials were prepared at the sixth grade reading level, formatted appropriately for older adults, and included images that reflected gender, racial, and ethnic diversity. The CBT-I group received a CD or MP3 player and a recording of the progressive muscle relaxation exercises.

We incorporated principles of treatment fidelity for studies of behavioral change in the study design and delivery of the intervention. We developed and followed protocol manuals for the sessions and provided consistent patient materials and follow-up to assure fidelity of treatment delivery and receipt. Patient materials were of identical format and length between the CBT-I and attention-control groups. The RA telephoned participants on weeks alternating with the group sessions to elicit problems with the sleep strategies, answer questions, and remind participants to complete the sleep diaries, attend the next meeting, and complete data collection activities in order to facilitate treatment enactment. The calls lasted approximately 15 minutes. The session content was presented by telephone or individually in person to participants who missed sessions. Biweekly telephone calls were also used to elicit use of the recommended strategies (treatment enactment).42

We used separate interventionists for the treatment and attention-control conditions to prevent contamination between conditions. Because this was a feasibility study, we did not set a minimum threshold for group attendance. Missed sessions were made up by telephone. Participants were invited to cross over into the alternate group after completing either the treatment or control conditions, but data from the second round of sessions were not included in the analyses.

Variables and Measures

Clinical and Demographic Variables

We used interview and chart review to track demographic and clinical data, prescribed and over-the-counter medications, health and sleep history, and comorbidity. The Charlson Comorbidity Index43 and the New York Heart Association Functional Classification were calculated with standard methods.

Measures of feasibility included rates of enrollment and retention in the study. We digitally audio-recorded all of the sessions and randomly selected and transcribed 20% to review the extent to which the treatment elements (Table 1) were addressed in each session (treatment fidelity). We tracked participant completion of study sessions and telephone calls (treatment enactment and the extent to which participants used the CBT-I elements through the biweekly telephone calls). We elicited treatment acceptability for both conditions with items eliciting acceptability of specific components, the extent to which participants viewed the treatment as helpful, and factors related to treatment delivery (scheduling, location, satisfaction with the provider). Each item was scored on a 0–10 (10 = highest) numeric rating scale.

Sleep Characteristics

Because of our interest in the multidimensional subjective and objective attributes of sleep (sleep quality, duration, continuity, latency) and patient perceptions (sleep quality, insomnia severity), we included actigraph and self-report measures of sleep. The Pittsburgh Sleep Quality Index44 includes dimensions reflecting sleep efficiency, sleep duration, sleep latency, sleep disturbances, sleep quality, and hypnotic medication use. It has well-documented reliability and validity44,45 Global sleep quality (total) ranges from 0–21. A score > 5 is considered poor sleep quality. The PSQI has diagnostic sensitivity of 89.6% and specificity of 86.5% for distinguishing “good” vs. “poor” sleepers.44 Cronbach α was 0.83 in HF patients.5

The Insomnia Severity Index (ISI) was used to evaluate insomnia severity. The ISI is internally consistent (0.74–0.88) and sensitive to treatment.35 The score range is from 0–28. Score levels indicate no (0–7), mild clinical insomnia (8–14), moderate (15–21), and severe clinical (22–28) insomnia. We used the continuous, as well as categorical scores, depending on the analyses. Cronbach α was 0.70 in the current study.

Objective sleep characteristics were measured with the Respironics Minimitter Actiwatch AW2, a wrist-worn accelerometer over 2-week periods at baseline and at the conclusion of treatment. Correlations between actigraphy and polysomnographic (PSG) measures range from 0.82–0.98 (sleep efficiency) and 0.90–0.97 (sleep duration) in normal sleepers.4649 Actigraphy corresponded with PSG on sleep efficiency, awakenings, wake after sleep onset, and total sleep time among people with insomnia50 and is sensitive to changes over time and treatment.51 Although actigraph-recorded sleep latency is not always a stable variable, we included sleep latency because it was sensitive to differences in NY Heart class among HF patients in our previous work (unpublished data).

A daily sleep/symptom diary was used for 14 nights at baseline and follow-up to assist in interpreting the wrist actigraph data (e.g., lights out time, lights on time) and to record nightly use of hypnotic medications. Diaries are reliable and valid and frequently used for this purpose.

Symptoms

We measured common daytime symptoms experienced by HF patients and associated with chronic insomnia, including fatigue, excessive daytime sleepiness, anxiety, and depression. The Multidimensional Assessment of Fatigue Index (MAF) that elicits severity, distress, degree of interference and timing of fatigue.52 The MAF has concurrent (r = 0.84) and divergent (r = −0.62) validity, and internal consistency (α = 0.84) in HF patients.53 The Epworth Sleepiness Scale (ESS) was used to measure self-reported daytime sleepiness, and has well-documented reliability and validity.5456 Coefficient α was 0.87 in HF patients.53 The Center for Epidemiological Studies Depression Scale (CESD)57,58 was used to measure depressive symptoms. Cronbach α was 0.83 in HF patients.53 The 40-item Spielberger State Inventory Form Y was used to measure state anxiety. Test-test reliability exceeds 0.7 for both the state and trait inventory and it distinguishes patient groups.59

Functional Performance

The Medical Outcomes Study SF36 V2 physical function component (PFC) was used to measure functional performance the day-to-day corporeal activities people do in the normal course of their lives to meet basic needs, fulfill usual roles, and maintain health and wellbeing.60,61

Data Analysis

Actigraph data were downloaded into Actiware v. 5 software (Respironics Minimitter, Inc.), and nocturnal sleep duration, sleep efficiency, and sleep latency were computed for each day of data collection. The mean values of these variables were computed over each of the 2-week intervals of data collection. Descriptive statistics were used to summarize each of the primary sleep variables (actigraph) (time in bed, sleep efficiency, sleep latency, and duration) over each of the 2-week intervals.

Data were committed to an MS Access database with the data obtained through the interviews and questionnaires. We used logical data checks after importing the data into SAS to identify outliers and data errors. Univariate statistics were computed on all variables to examine frequency distributions of categorical variables and normal distributions for continuous variables.

Baseline equivalency of demographics and potential confounding variables including comorbidity, NYHA, and sleep apnea was assessed between 2 intervention groups to determine the success of randomization. Variables found to differ by group assignment were included in the models as covariates. We computed descriptive statistics on feasibility and acceptability criteria. We also computed intra-class correlation coefficients to assess the degree to which randomization into treatment groups influenced the outcomes.

The means and standard deviations of the changed outcomes (delta) from pre- to post-intervention were calculated for each intervention group, and the effect sizes were calculated by dividing the differences between the 2 change scores with the pooled standard deviations. Because insomnia severity and fatigue are correlated, we used multivariate analysis of variance (MANOVA) to compare the treatment effects simultaneously on these variables. We performed the analyses with and without adjustment for age and baseline comorbidity. The multivariate models were performed with all subjects, and after removing outliers diagnosed using a criterion (DFFITS ≥ 1) that indicates a large influence on estimations, the results were confirmed. The Kolmogorov-Smirnov test was performed to check the normality of the residuals.

RESULTS

We contacted a total of 433 potential participants (Figure 1). Among those who were ineligible for the study, approximately half did not have insomnia. The second most common reason for exclusion from the study was untreated sleep disordered breathing. Of the 52 randomized patients 48 (92%) completed the study (CBT-I, n = 29: attention-control, n = 19). In the CBT-I group, one patient was hospitalized for a heart transplant immediately after randomization, and a second was lost to follow-up before completing any sessions. Among the patients in the attention-control condition, one patient dropped out because he was unhappy with randomization to the control condition; a second started shift-work after randomization and was then ineligible for the study.

Consort flow diagram for participant recruitment.

jcsm.11.10.1109a.jpg

jcsm.11.10.1109a.jpg
Figure 1

Consort flow diagram for participant recruitment.

(more ...)

Clinical and demographic characteristics of the overall sample and each of the treatment conditions are presented in Table 2. Participants in the CBT-I group were somewhat older (p = ns) and had significantly more comorbidity than those in the attention control condition (p < 0.05), but there were no statistically significant differences between groups on gender, race, or New York Heart Association Functional Classification.

Clinical and demographic characteristics of the total sample, and the CBT-I and Healthy Hearts Groups.

jcsm.11.10.1109.t02.jpg

table icon
Table 2

Clinical and demographic characteristics of the total sample, and the CBT-I and Healthy Hearts Groups.

(more ...)

Based on review of audio-recordings of the sessions by 2 independent assessors, the interventionists addressed the session-specific components, as listed in Table 1, in all relevant biweekly meetings and telephone calls. The rates of attendance and completion of the study meetings and telephone calls are in Table 3. Among the entire sample, 32 (68%) participants completed all 4 meetings; the remainder made up missed sessions by telephone, while 45 (95%) completed all of the telephone calls. Reasons for missing sessions included inclement weather and personal matters (e.g., death of a family member; household repairs). At the completion of the study, participants in the CBT-I group reporting using the following on a regular basis: avoid daytime naps (n = 21/73%), consistent bedtime/get up time (n = 27/93%), avoid television at bed time (n = 20/69%), strategies to manage bedroom environment (n = 26/89%), relaxation CD or other methods of relaxation at bedtime (n = 20/69%).

Attendance and call completion.

jcsm.11.10.1109.t03.jpg

table icon
Table 3

Attendance and call completion.

(more ...)

At baseline, 7 (23.3%) of the CBT-I group used prescribed or over-the-counter hypnotic medications (Table 2). Seven (35%) of the attention control group used prescribed or over-the-counter medications at baseline, with all but one using them half or more of the nights. At follow-up, one CBT-I participant discontinued hypnotics and one decreased from more than half to less than half of the 14 diary nights; while one participant in the attention control group began using hypnotics and one increased the frequency from less than half to more than half of the diary nights.

Satisfaction/acceptability with treatment was high among both groups (Table 4). Median scores of of helpfulness of the treatment ranged from 6–10 out of 10. Overall, perceptions about the short-term and long-term helpfulness of the treatment and its effects on daytime function were somewhat higher for the CBT-I group. Both groups had high levels of satisfaction with the logistical details, such as scheduling, location, adequacy of payment, study materials, and effectiveness of the group leader. Among the components of the CBT-I intervention, the relaxation CD had the lowest rating (median = 8).

Satisfaction and acceptability with treatment components and logistics.

jcsm.11.10.1109.t04.jpg

table icon
Table 4

Satisfaction and acceptability with treatment components and logistics.

(more ...)

The descriptive statistics for the primary study variables, change scores, and effect sizes are presented in Table 5. There was a moderate-large effect of CBT-I on insomnia severity (p = 0.03). The mean decrease of more than 7 points in insomnia severity in the CBT-I group, compared with a smaller decrease in the attention-control group exceeded the 6-point criterion indicative of clinical improvement and improved daytime function.62 The mean ISI score at follow-up indicated that the typical patient experienced remission of insomnia. In the CBT-I group, 76% improved across one or more levels of insomnia severity and no one deteriorated, while 41% of the attention-control group improved, and 10% deteriorated. At post intervention, 15 participants (51.7%) in the CBT-I group had no insomnia (ISI ≤ 7), compared with 31.6% in the attention-control group (p = 0.17). After controlling for age and comorbidity, the odds ratio for the effects of treatment on insomnia was 0.41 (95% CI = 0.11, 1.48).

Descriptive statistics, change, standard deviation of change and p value of change in self-reported and actigraphic measures of sleep and symptom and function performance between baseline and 2 weeks post-intervention.

jcsm.11.10.1109.t05.jpg

table icon
Table 5

Descriptive statistics, change, standard deviation of change and p value of change in self-reported and actigraphic measures of sleep and symptom and function performance between baseline and 2 weeks post-intervention.

(more ...)

The effects on self-reported sleep quality, sleep latency, and sleep efficiency were small-moderate. There was a moderate-large effect on actigraph-recorded sleep efficiency (p = 0.03), but the mean sleep efficiency was low at all time points in both groups. There were also moderate CBT-I effects on actigraph-recorded sleep latency (p = 0.09) and duration (p = 0.15). There was a large improvement in fatigue (p = 0.04) and a small improvement in physical function, but no meaningful differences in sleepiness, depression, or anxiety at follow-up.

To examine the possible effect of the clustering of individuals by group membership within each treatment condition, we calculated the intra-class correlation coefficients (ICC) for the change in insomnia severity and fatigue, the 2 variables with the largest effect sizes across the treatment groups. The ICCs for these variables approached zero. Therefore, there did not appear to be a significant effect of clustering.

At baseline and follow-up there was a positive association between insomnia severity and fatigue (r = 0.494, p = 0.0004) and (r = 0.33, p = 0.0021), respectively. However, there was no statistically significant association between the changes in insomnia severity and changes in fatigue from baseline to follow-up (r = 0.12, p = 0.42).

The results of the MANOVA appear in Table 6. Age and gender were included as covariates in these analyses because of the trends suggesting group differences in these variables. We did not include the New York Heart Association Function Classification because there were no group differences in this variable and it was not related to fatigue or changes in fatigue (data not shown). There was a statistically significant effect of CBT-I, compared with the attention-control condition, on insomnia severity and fatigue (p = 0.02). This effect remained similar when controlling for age and comorbidity.

Multivariate analysis of variance (MANOVA) for the effects of CBT-I vs. attention-control on changed insomnia severity and fatigue at post intervention after controlling for age and comorbidity at baseline.

jcsm.11.10.1109.t06.jpg

table icon
Table 6

Multivariate analysis of variance (MANOVA) for the effects of CBT-I vs. attention-control on changed insomnia severity and fatigue at post intervention after controlling for age and comorbidity at baseline.

(more ...)

DISCUSSION

CBT-I was feasible and highly acceptable to patients with stable HF and resulted in large improvements in insomnia and fatigue. These effects were statistically significant after controlling for the effects of comorbidity and gender. The importance of these findings is demonstrated by improvements in the mean level of insomnia severity in the CBT-I group to near normal levels at follow-up and improvements in 76% of the CBT-I group, compared to no change or deterioration in almost half of participants in the attention-control condition, in contrast to no change or deterioration in a larger percentage of the attention-control group.

The effect of CBT-I on sleep efficiency was large, despite the fact that sleep efficiency continued to be low. Overall the small percentage of increase may not be clinically significant. However, assuming a normal distribution for this variable, the standard deviation suggests that 15% of the population would have an increase of 9.4%, a clinically meaningful increase. Future studies are needed to further validate this finding and to evaluate clinical, demographic, or treatment-related characteristics that might predict improvement of this magnitude.

The large effect of CBT-I on fatigue is especially notable, given the often intractable nature of fatigue among patients with HF even when using evidence-based disease management, and suggests the causal contributions of insomnia to fatigue in these patients. This finding also exends previous cross-sectional relationships between insomnia and fatigue.5 The strength of this effect, given comparison with the results of the attention control condition that included standard HF self-management strategies (adherence to medications and fluid and sodium restriction, diet, relaxation, and exercise) that may improve HF itself and resulting symptoms, further supports the importance of insomnia to these patients.

Although the effect of CBT-I on functional performance was small, the magnitude of the effect of CBT-I on insomnia was consistent with the size of the effect associated with improvement in function in a non-HF population.62 These findings merit future study given our previous cross-sectional work in which insomnia was associated with both self-reported and objective functional performance, including a 100-foot decrement in six-minute walk distance.5

Consistent with previous work,5 the HF patients were not overall excessively sleepy, a finding possibly related to high levels of sympathetic arousal associated with both insomnia and HF. Future studies should be conducted in larger samples to identify the characteristics of patients who are most likely to be sleepy, a symptom that was not associated with sleep disordered breathing in previous work.14 The lack of change in depressive symptoms evident in both groups is consistent with a previous study of CBT for depression and CBT with exercise in HF patients63 and warrants further study in a larger group, given the usual high prevalence of depression among patients with HF and the beneficial effects of CBT-I on depression found in other studies.31

Sleep disordered breathing, including both central and obstructive sleep apnea, are common among patients with HF, but did not explain insomnia or sleep quality in cross-sectional work.5,14 Although the sample was too small to conduct sensitivity analysis, future studies should compare the effects of CBT-I among those with insomnia and SDB who were adherent to positive airway pressure and those without significant SDB. From an anecdotal perspective, frequent requests to participate in this trial by HF patients who had significant SDB and insomnia, but were not adherent to CPAP, as well as their frequent use of prescribed hypnotic medications, suggest the critical importance of patient education about the distinctions between these two conditions and careful assessment and referral for sleep disorders in this population who are at high risk for both sleep disorders.

Our findings support the potential efficacy of CBT-I among patients with HF. While a larger fully powered RCT is needed to further confirm these short-term effects and to established sustained efficacy, our results suggest the feasibility of the design and approach used in this preliminary study, as well as high levels of patient acceptability and use of the CBT-I components. Given recent evidence that insomnia predicts incident HF and death over a mean 11 year follow-up,6 future studies should also focus on the effects of insomnia treatment, including CBT-I, on the pathophysiology of HF and event-free survival.

The use of an attention-control condition that included education on standard HF self-management strategies that may influence insomnia and symptoms provided a comparison that reflected practice recommendations for HF.15 Sleep hygiene was included, but unlike some studies that included sleep hygiene or usual care alone, this design controlled for time and attention that may have important influences on behavioral outcomes and may contribute to high levels of acceptability

We used the group approach to CBT-I to maximize social support and group learning, but scheduling was a challenge due to participants' busy schedules, weather and other unanticipated events. Although this design may be subject to the effects of group clustering we found none in this study. Although recent CBT-I trials have incorporated web-based and telehealth interventions, and these may be useful for some HF patients, limitations of these delivery mechanisms may include lack of familiarity or access to suitable technology by older adults or those of low socioeconomic status.

Hypnotic use was somewhat less common at baseline in the CBT-I, compared to the attention-control condition, and at follow-up two patients decreased used in the former, compared to increased use by two in the latter group. Although the sample was too small to determine significant change, the extent to which hypnotic use changes as a result of CBT-I should be examined in future studies, given the efficacy of hypnotic taper as a component of CBT-I and previous evidence that HF patients dislike the effects of hypnotics.12

The results of this study are generalizable to patients with stable HF who had comorbid insomnia and for the most part, did not have advanced disease. However, one participant who was awaiting heart transplant successfullycompleted the sessions at her hospital bedside. Although she was not able to complete follow-up assessment, this observation suggests that CBT-I may be be appropriate for these patients, especially if accommodations can be made to address difficulty in attending sessions.

Limitations of the study included the inequality of the group sizes due to the need to randomize by group to ensure timely filling of the groups, given available resources and the possible effect of group. Although participants made up missed sessions by telephone, the sample is too small to determine differences in outcomes for those who attended all of the face-to-face sessions compared with those who required make-up sessions. Although treatment fidelity and the levels of satisfaction with the interventionist for each control condition were high, it is possible that the specific therapist or the disciplinary perspective of the therapist influenced the treatment outcomes. On the other hand, the disciplinary perspectives of the therapist reflect real-world clinical practice. Future studies should incorporate evaluation of these factors.

CBT-I is feasible, acceptable, and has short term efficacy among patients with stable HF and chronic comorbid insomnia. Future study is needed to establish its sustained effects on symptoms, event-free survival, and other outcomes.

DISCLOSURE STATEMENT

This was not an industry supported study. This study was funded by NIH grant R21NR011387 (N. Redeker, PI) and P20 NR014126 (N. Redeker, PI). The authors have indicated no financial conflicts of interest. The study did not employ any off-label use of medications.

ABBREVIATIONS

AHI

apnea-hypopnea index

APRN

advanced practice registered nurse

ARES

Apnea Risk Evaluation system

CESD

Centers for the Epidemiological Studies of Depression Scale

CBT-I

cognitive behavioral therapy for insomnia

DFFITS

change in the predicted value for a point when that point is left out of the regression

EEG

electro-encephalography

EMG

electromyography

EOG

electro-oculography

FP1

frontal electrode

FP2

frontal electrode

HF

heart failure

ISI

insomnia severity index

MAF

multi-dimensional assessment of fatigue scale

MANOVA

multivariate analysis of variance

NYHA

New York Heart Association Functional Classification

NREM

non-rapid eye movement

PFC

physical function component of the Medical Outcomes Study SF36 Questionnaire

PSG

polysomnography

PSQI

Pittsburgh Sleep Quality Index

RCT

randomized controlled trial

REM

rapid eye movement

SE

sleep efficiency

SpO2

pulse oxygen saturation

ACKNOWLEDGMENTS

The authors gratefully acknowledge the assistance of Joanne Pacelli, Research Assistant, who provided excellent support in recruiting and retaining the study participants.

REFERENCES

1 

Go AS, Mozaffarian D, Roger VL, et al., authors. Heart disease and stroke statistics--2014 update: a report from the American Heart Association. Circulation. 2014;129:e28–292. [PubMed]

2 

Redeker NS, Stein S, authors. Characteristics of sleep in patients with stable heart failure versus a comparison group. Heart Lung. 2006;35:252–61. [PubMed]

3 

Gau FY, Chen XP, Wu HY, Lin ML, Chao YF, authors. Sleep-related predictors of quality of life in the elderly versus younger heart failure patients: a questionnaire survey. Int J Nurs Stud. 2011;48:419–28. [PubMed]

4 

Brostrom A, Stromberg A, Dalstrom U, Fridlund B, authors. Sleep difficulties, daytime sleepiness, and health-related quality of life in patients with chronic heart failure. J Cardiovasc Nurs. 2004;19:234–42. [PubMed]

5 

Redeker NS, Jeon S, Muench U, Campbell D, Walsleben J, Rapoport DM, authors. Insomnia symptoms and daytime function in stable heart failure. Sleep. 2010;33:1210–6. [PubMed Central][PubMed]

6 

Laugsand LE, Strand LB, Platou C, Vatten LJ, Janszky I, authors. Insomnia and the risk of incident heart failure: a population study. Eur Heart J. 2014;35:1382–93. [PubMed]

7 

Lee KS, Song EK, Lennie TA, et al., authors. Symptom clusters in men and women with heart failure and their impact on cardiac event-free survival. J Cardiovasc Nurs. 2010;25:263–72. [PubMed]

8 

Johansson P, Riegel B, Svensson E, et al., authors. The contribution of heart failure to sleep disturbances and depressive symptoms in older adults. J Geriatr Psychiatry Neurol. 2012;25:179–87. [PubMed]

9 

Fink AM, Gonzalez RC, Lisowski T, et al., authors. Fatigue, inflammation, and projected mortality in heart failure. J Card Fail. 2012;18:711–6. [PubMed Central][PubMed]

10 

Song EK, Moser DK, Rayens MK, Lennie TA, authors. Symptom clusters predict event-free survival in patients with heart failure. J Cardiovasc Nurs. 2010;25:284–91. [PubMed Central][PubMed]

11 

Freedland KE, Carney RM, Rich MW, authors. Effect of depression on prognosis in heart failure. Heart Fail Clin. 2011;7:11–21. [PubMed Central][PubMed]

12 

Andrews LK, Coviello J, Hurley E, Rose L, Redeker NS, authors. “I'd eat a bucket of nails if you told me it would help me sleep:” perceptions of insomnia and its treatment in patients with stable heart failure. Heart Lung. 2013;42:339–45. [PubMed Central][PubMed]

13 

Johansson P, Arestedt K, Alehagen U, Svanborg E, Dahlstrom U, Brostrom A, authors. Sleep disordered breathing, insomnia, and health related quality of life -- a comparison between age and gender matched elderly with heart failure or without cardiovascular disease. Eur J Cardiovasc Nurs. 2010;9:108–17. [PubMed]

14 

Redeker NS, Muench U, Zucker MJ, et al., authors. Sleep disordered breathing, daytime symptoms, and functional performance in stable heart failure. Sleep. 2010;33:551–60. [PubMed Central][PubMed]

15 

Yancy CW, Jessup M, Bozkurt B, et al., authors. 2013 ACCF/AHA guideline for the management of heart failure: executive summary: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation. 2013;128:1810–52. [PubMed]

16 

Riegel B, Glaser D, Richards K, et al., authors. Modifiable factors associated with sleep dysfunction in adults with heart failure. Eur J Cardiovasc Nurs. 2012;11:402–9. [PubMed Central][PubMed]

17 

Morin CM, Bootzin RR, Buysse DJ, Edinger JD, Espie CA, Lichstein KL, authors. Psychological and behavioral treatment of insomnia:Update of the recent evidence (1998-2004). Sleep. 2006;29:1398–414. [PubMed]

18 

Morin CM, author. Cognitive-behavioral approaches to the treatment of insomnia. J Clin Psychiatry. 2004;65 Suppl 16:33–40. [PubMed]

19 

Rybarczyk B, Stepanski E, Fogg L, Lopez M, Barry P, Davis A, authors. A placebo-controlled test of cognitive-behavioral therapy for comorbid insomnia in older adults. J Consult Clin Psychol. 2005;73:1164–74. [PubMed]

20 

Edinger JD, Olsen MK, Stechuchak KM, et al., authors. Cognitive behavioral therapy for patients with primary insomnia or insomnia associated predominantly with mixed psychiatric disorders: a randomized clinical trial. Sleep. 2009;32:499–510. [PubMed Central][PubMed]

21 

Currie SR, Clark S, Hodgins DC, El-Guebaly N, authors. Randomized controlled trial of brief cognitive-behavioural interventions for insomnia in recovering alcoholics. Addiction. 2004;99:1121–32. [PubMed]

22 

Espie CA, MacMahon KM, Kelly HL, et al., authors. Randomized clinical effectiveness trial of nurse-administered small-group cognitive behavior therapy for persistent insomnia in general practice. Sleep. 2007;30:574–84. [PubMed]

23 

Germain A, Moul DE, Franzen PL, et al., authors. Effects of a brief behavioral treatment for late-life insomnia: preliminary findings. J Clin Sleep Med. 2006;2:403–6. [PubMed]

24 

Buysse DJ, Germain A, Moul DE, et al., authors. Efficacy of brief behavioral treatment for chronic insomnia in older adults. Arch Int Med. 2011;171:887–95.

25 

Espie CA, Fleming L, Cassidy J, et al., authors. Randomized controlled clinical effectiveness trial of cognitive behavior therapy compared with treatment as usual for persistent insomnia in patients with cancer. J Clin Oncol. 2008;26:4651–8. [PubMed]

26 

Epstein DR, Dirksen SR, authors. Randomized trial of a cognitive-behavioral intervention for insomnia in breast cancer survivors. Oncol Nurs Forum. 2007;34:E51–9. [PubMed]

27 

Chen HY, Chiang CK, Wang HH, et al., authors. Cognitive-behavioral therapy for sleep disturbance in patients undergoing peritoneal dialysis: a pilot randomized controlled trial. Am J Kidney Dis. 2008;52:314–23. [PubMed]

28 

Manber R, Edinger JD, Gress JL, San Pedro-Salcedo MG, Kuo TF, Kalista T, authors. Cognitive behavioral therapy for insomnia enhances depression outcome in patients with comorbid major depressive disorder and insomnia. Sleep. 2008;31:489–95. [PubMed Central][PubMed]

29 

Ulmer CS, Edinger JD, Calhoun PS, authors. A multi-component cognitive-behavioral intervention for sleep disturbance in veterans with PTSD: a pilot study. J Clin Sleep Med. 2011;7:57–68. [PubMed Central][PubMed]

30 

Kapella MC, Herdegen JJ, Perlis ML, et al., authors. Cognitive behavioral therapy for insomnia comorbid with COPD is feasible with preliminary evidence of positive sleep and fatigue effects. Int J Chron Obstruct Pulm Dis. 2011;6:625–35.

31 

Smith MT, Huang MI, Manber R, authors. Cognitive behavior therapy for chronic insomnia occurring within the context of medical and psychiatric disorders. Clin Psychol Rev. 2005;25:559–92. [PubMed]

32 

McCrae CS, Bramoweth AD, Williams J, Roth A, Mosti C, authors. Impact of brief cognitive behavioral treatment for insomnia on health care utilization and costs. J Clin Sleep Med. 2014;10:127–35. [PubMed Central][PubMed]

33 

Lichstein KL, Nau SD, Wilson NM, et al., authors. Psychological treatment of hypnotic-dependent insomnia in a primarily older adult sample. Behav Res Ther. 2013;51:787–96. [PubMed]

34 

Hayes D Jr.; Anstead MI, Ho J, Phillips BA, authors. Insomnia and chronic heart failure. Heart Fail Rev. 2009;14:171–82. [PubMed]

35 

Bastien CH, Vallieres A, Morin CM, authors. Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med. 2001;2:297–307. [PubMed]

36 

Edinger JD, Sampson WS, authors. A primary care “friendly” cognitive behavioral insomnia therapy. Sleep. 2003;26:177–82. [PubMed]

37 

Popovic D, King C, Guerrero M, Levendowski DJ, Henninger D, Westbrook PR, authors. Validation of forehead venous pressure as a measure of respiratory effort for the diagnosis of sleep apnea. J Clin Monit Comput. 2009;23:1–10. [PubMed]

38 

Westbrook PR, Levendowski DJ, Cvetinovic M, et al., authors. Description and validation of the apnea risk evaluation system: a novel method to diagnose sleep apneahypopnea in the home. Chest. 2005;128:2166–75. [PubMed]

39 

Ayappa I, Norman RG, Seelall V, Rapoport DM, authors. Validation of a self-applied unattended monitor for sleep disordered breathing. J Clin Sleep Med. 2008;4:26–37. [PubMed Central][PubMed]

40 

Levendowski D, Steward D, Woodson BT, Olmstead R, Popovic D, Westbrook P, authors. The impact of obstructive sleep apnea variability measured in-lab versus in-home on sample size calculations. Inter Arch Med. 2009;2:2.

41 

American Association of Heart Failure Nurses, PCNA, Society of Chest Pain Centers. Fight Against Heart Failure Handbook, 2008.

42 

Bellg AJ, Borrelli B, Resnick B, et al., authors. Enhancing treatment fidelity in health behavior change studies: best practices and recommendations from the NIH Behavior Change Consortium. Health Psychol. 2004;23:443–51. [PubMed]

43 

Charlson ME, Pompei P, Ales KL, MacKenzie CR, authors. A new method of classifying prognostic comorbidity in longitudinal studies. Chron Dis. 1987;40:373–83.

44 

Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ, authors. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28:193–213. [PubMed]

45 

Buysse DJ, Reynolds CF, Monk TH, Hoch CC, Yeager AL, Kupfer DJ, authors. Quantification of subjective sleep quality in healthy elderly men and women using the Pittsburgh Sleep Quality Index (PSQI). Sleep. 1991;14:331–8. [PubMed]

46 

Kushida CA, Chang A, Gadkary C, Guilleminault C, Carrillo O, Dement WC, authors. Comparison of actigraphic, polysomnographic, and subjective assessment of sleep parameters in sleep-disordered patients. Sleep Med. 2001;2:389–96. [PubMed]

47 

Blood MR, Sack RL, Percy DC, Pen J, authors. A comparison of sleep detection by wrist actigraph, behavioral response, and polysomnography. Sleep. 1997;20:388–95. [PubMed]

48 

Edinger JD, Means MK, Stechuchak KM, Olsen MK, authors. A pilot study of inexpensive sleep-assessment devices. Behav Sleep Med. 2004;2:41–9. [PubMed]

49 

Benson K, Friedman L, Noda A, Wicks D, Wakabayashi E, Yesavage J, authors. The measurement of sleep by actigraphy: direct comparison of 2 commercially available actigraphs in a nonclinical population. Sleep. 2004;27:986–9. [PubMed]

50 

Lichstein KL, Stone KC, Donaldson J, et al., authors. Actigraphy validation with insomnia. Sleep. 2006;29:232–9. [PubMed]

51 

Littner M, Kushida CA, Anderson WM, et al., authors. Practice parameters for the role of actigraphy in the study of sleep and circadian rhythms: an update for 2002. Sleep. 2003;26:337–41. [PubMed]

52 

Belza BL, author. Multidimensional Assessment of Fatigue (MAF) Scale Users Guide. Seattle: University of Washington, 1990.

53 

Redeker NS, author. Somatic symptoms explain differences in psychological distress in heart failure patients vs a comparison group. Prog Cardiovasc Nurs. 2006;21:182–9. [PubMed]

54 

Johns MW, author. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep. 1991;14:540–5. [PubMed]

55 

Johns MW, author. Daytime sleepiness, snoring, and obstructive sleep apnea, the Epworth Sleepiness Scale. Chest. 1993;103:30–6. [PubMed]

56 

Johns MW, author. Reliability and factor analysis of the Epworth Sleepiness Scale. Sleep. 1992;15:376–81. [PubMed]

57 

Radloff LS, Teri L, authors. Use of the Center for Epidemiological Studies-Depression Scale witholder adults. Clin Gerontol. 1986;5:119–36.

58 

Devins G, Orme C, authors. Center for Epidemiologic Studies Depression Scale. In: Keyser D, Sweetland R, editors. Test Critiques. Kansas City, MO: Test Corporation of America, 1985.

59 

Spielberger CD, author. Manual for the State-Trait Anxiety Inventory STAI-Form Y. Palo Alto, CA: Consulting Psychologists Press, 1983.

60 

McHorney CA, Ware JE, Raczek AE, authors. The MOS 36-Item short-form health survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31:247–63. [PubMed]

61 

Leidy NK, author. Functional status and the forward progress of merry-go-rounds: toward a coherent analytical framework. Nurs Res. 1994;43:196–202. [PubMed]

62 

Yang M, Morin CM, Schaefer K, Wallenstein GV, authors. Interpreting score differences in the Insomnia Severity Index: using health-related outcomes to define the minimally important difference. Curr Med Res Opin. 2009;25:2487–94. [PubMed]

63 

Gary RA, Dunbar SB, Higgins MK, Musselman DL, Smith AL, authors. Combined exercise and cognitive behavioral therapy improves outcomes in patients with heart failure. J Psychosom Res. 2010;69:119–31. [PubMed Central][PubMed]