Skip to main content
Free access
ORIGINAL REPORTS
August 08, 2011

Natural Course of Insomnia Comorbid With Cancer: An 18-Month Longitudinal Study

Publication: Journal of Clinical Oncology

Abstract

Purpose

This study aimed to assess the prevalence and natural course (incidence, persistence, remission, and relapse) of insomnia comorbid with cancer during an 18-month period.

Patients and Methods

All patients scheduled to receive a curative surgery for a first diagnosis of nonmetastatic cancer were approached on the day of their preoperative visit to participate in the study. A total of 962 patients with cancer (mixed sites) completed an insomnia diagnostic interview at the perioperative phase (T1), as well as at 2 (T2), 6 (T3) 10 (T4), 14 (T5), and 18 (T6) months after surgery.

Results

Findings revealed high rates of insomnia at baseline (59%), including 28% with an insomnia syndrome. The prevalence of insomnia generally declined over time but remained pervasive even at the end of the 18-month period (36%). Rates were greater in patients with breast (42% to 69%) and gynecologic (33% to 68%) cancer and lower in men with prostate cancer (25% to 39%) throughout the study. Nearly 15% of patients had a first incidence of insomnia during the study, and 19.5% experienced relapse. The evolution of symptoms varied according to sleep status. Remissions (patients becoming good sleepers) were much less likely for patients with an insomnia syndrome (10.8% to 14.9%) than for those with insomnia symptoms (42.0% to 51.3%). Most frequently (37.6%), patients with an insomnia syndrome at baseline kept that status throughout the 18-month period.

Conclusion

Insomnia is a frequent and enduring problem in patients with cancer, particularly at the syndrome level. Early intervention strategies, such as cognitive-behavioral therapy, could prevent the problem from becoming more severe and chronic.

Introduction

Early descriptive studies have revealed that sleep disturbances affect between 30% and 50% of patients with cancer 1,2 and that 18% and 19% of survivors of prostate cancer and breast cancer, respectively, meet the diagnostic criteria for an insomnia syndrome.3,4 Important limitations of these studies are the use of small, convenience samples, often years after treatment, and their cross-sectional nature. In recent years, there have been an increasing number of longitudinal studies on the psychosocial aspects of cancer treatments integrating measures of sleep disturbances. Some of them found a general decrease of sleep impairments over time,57 whereas others revealed relatively stable scores.810 However, the majority of these studies used small samples of patients with specific cancer sites, which makes comparisons across cancer sites impossible, and used sleep quality items from general quality-of-life or physical symptoms questionnaires that do not distinguish the presence of clinically significant sleep impairments.
An exception is the study by Palesh et al11 conducted in 823 patients receiving chemotherapy for various types of cancer (all stages). According to answers to sleep items from a depression questionnaire, 79.6% of the patients displayed insomnia symptoms at cycle 1 including 43.0% who had an insomnia syndrome; these rates decreased to 68.3% and 35.2%, respectively, by cycle 2. Among good sleepers at cycle 1, 34.6% developed insomnia symptoms by cycle 2 (10% developed an insomnia syndrome). The prevalence of insomnia symptoms was the highest in patients with breast and gynecologic cancer, whereas rates of insomnia syndrome were greater in patients with lung cancer. Nonetheless, more studies are needed to characterize the evolution of insomnia throughout the cancer care trajectory.
The goal of this large, population-based, longitudinal study was to assess the prevalence of insomnia across cancer sites and document its natural history (ie, incidence, persistence, remission, and relapse) in an 18-month period in patients undergoing treatment for nonmetastatic cancer. No a priori hypothesis was established given the observational nature of the study. Findings from the first two time points (perisurgical period and 2 months later) have recently been published.12 The current report presents results from all six time points.

Patients and Methods

Participants

Recruitment.

Inclusion criteria included a confirmed first diagnosis of nonmetastatic cancer, that patients be scheduled to receive curative surgery, patients between 18 to 80 years of age, and that patients be able to read and understand French. Exclusion criteria included administration of neoadjuvant cancer treatment; that upcoming surgery was part of brachytherapy for prostate cancer; severe cognitive impairments (eg, Alzheimer's disease) or severe psychiatric disorder (eg, psychosis) as noted in the medical chart, observed at recruitment, or self-reported; patients to have been diagnosed by a physician or to be treated for a sleep disorder other than insomnia (eg, sleep apnea); and severe visual, hearing, or language defects.
Patients were recruited at L'Hôtel-Dieu de Québec and Hôpital du St-Sacrement. All patients meeting the initial inclusion criteria (ie, age, upcoming surgery) were approached by a research assistant on the day of their preoperative visit to explain the study goals and assess the patients' eligibility for the study. Then, eligible patients were invited to provide their written consent. The study was approved by the ethic's review boards of both hospitals.
Of the 3,196 patients approached at the clinics, 1,519 were excluded and 715 refused to participate in the study, thus giving a participation rate of 57.4% (n = 962; Fig 1; see Appendix for differences on refusal and dropout rates). The mean age was 57.0 years (range, 23-79 years); 64.4% of the participants were females; 67.4% were married/cohabitating; 51.3% had completed a college degree or more; 39.6% were employed. Patients had various cancer sites, including breast (49%), prostate (27%), gynecologic (12%), head and neck (2%), urinary or GI (UGI; 7%), and other (3%). Cancer stages were stage 0 (5%), stage I (36%), stage II (38%), stage III (19%), and stage IV (3%). Overall, 51.2% received radiation therapy, 30.6% chemotherapy, and 38.5% hormone therapy, whereas 46.5% received no adjuvant treatment. On the basis of the clinical interview, 11.9% of the participants reported antecedents of an insomnia syndrome and 12.6% of insomnia symptoms.
Fig 1. Participants' flowchart and cancer treatments received between each time point (T; baseline, 2, 6, 10, 14, and 18 months). CTX, chemotherapy; HTX, hormone therapy; RTX, radiation therapy; SRG, surgery.

Study Design

This study used a prospective longitudinal design comprising six time assessments at baseline (perioperative phase; T1), 2 (T2), 6 (T3), 10 (T4), 14 (T5), and 18 months (T6). Although patients were recruited before surgery, 81.2% of them completed baseline measures afterward (20 days afterward on average). Comparisons of patients who completed measures before and after surgery are presented in the Appendix.

Measure

An adaptation of the insomnia interview schedule (IIS),13 a semistructured interview, was used to categorize patients into three groups, at each time point: (a) good sleepers: no subjective complaint of sleep difficulties or use a hypnotic medication less than one night per week; (b) patients with insomnia symptoms: complaint of sleep difficulties but do not meet the criteria for an insomnia syndrome or use a hypnotic medication one or two nights per week; patients with insomnia syndrome: subjective complaint of sleep difficulties, sleep onset latency or wake after sleep onset greater than 30 minutes, ≥ 3 nights per week, duration ≥ 1 month, associated with impaired daytime functioning or marked distress or using a hypnotic medication ≥ 3 nights per week for ≥ 1 month.

Procedure

At each time point, a phone interview was conducted to administer the IIS. Interjudge agreements were conducted on 8.1% of the interviews (agreement rate, 83.1%; weighted kappa, .81; 95% CI, .72 to .89). To avoid overlapping with a subsequent time point, a maximum interval of 8 weeks was allowed to return the self-report scales completed (not reported here) and to conduct the interview, but in general, these procedures were all done within 4 weeks. More details on the procedure are available elsewhere.12

Statistical Analyses

Only patients who had completed study measures on at least two consecutive time points were included in the analyses (n = 856). The prevalence of insomnia symptoms and syndrome was computed for the total sample and across cancer sites. Generalized mixed models using a split plot factorial design (cancer sites × time) were performed to test the significance of temporal changes, adjusting prevalence estimates for missing data. To estimate incidence, persistence, remission, and relapse rates, each pair of consecutive time points was analyzed, using the definitions we describe in this section. Incident patients were those who had no symptoms at one time point and had insomnia symptoms or syndrome at the subsequent time point. Remission was defined as a change from a status of either insomnia symptoms or syndrome at one time point to a no-symptom status (good sleepers) at the subsequent time point. Persistent insomnia was defined as the presence of insomnia symptoms or syndrome at two consecutive time points. Relapse was defined by a resurgence of insomnia after a patient went into remission. Unless otherwise specified, the term insomnia will be used hereafter to include both the insomnia syndrome and insomnia symptoms groups. To describe visually the most common sleep trajectories, 555 patients with all 6 time points completed and 187 additional patients who did not miss two consecutive time points were included. For that specific purpose only, missing data were imputed according to the most likely pattern of results using the most similar complete case patient (Hot Deck Imputation).14 Finally, the probability of changing/keeping the same sleep status at the subsequent time point for each possible previous sleep status (good sleepers, insomnia symptoms, and insomnia syndrome) was computed for the total sample and for each cancer site (n = 856).

Results

Point Prevalence of Insomnia at Each Assessment

At T1, 28% of the patients met the criteria for an insomnia syndrome and 31% had insomnia symptoms for a total of 59% (Fig 2). Then the prevalence of insomnia decreased steadily and significantly over the 18-month period, ending at T6 with rates of insomnia syndrome of 21% [F(5,3725) = 7.58; P < .001] and of insomnia symptoms of 15% [F(5,3725) = 19.78; P < .001 (total of 36%)].
Fig 2. Overall prevalence of insomnia over time.

Point Prevalence of Insomnia Across Cancer Sites at Each Assessment

As shown in Figure 3, prevalence rates of insomnia were the highest in patients with breast (42% to 69%) and gynecologic (33% to 68%) cancer and the lowest in patients with prostate cancer (25% to 39%). Likewise, prevalence rates of insomnia syndrome were the highest in patients with breast (25% to 36%) and gynecologic (18% to 29%) cancer and the lowest in patients with prostate cancer (12% to 15%). Rates of insomnia were also high in patients of the other three groups (head and neck, UGI, and other), but given that many cells were composed of fewer than 20 patients, variations observed in these groups need to be interpreted cautiously. Significant time effects on rates of insomnia were found in patients with breast [F(5,1873) = 29.58; P < .01], prostate [F(5,1027) = 5.38; P < .01], gynecologic [F(5,397) = 9.17; P < .01], and UGI cancer [F(5,220) = 5.23; P < .01], and the general pattern was in the direction of a progressive reduction over the 18-month period.
Fig 3. Prevalence of insomnia across cancer sites. (*) These cells have less than 20 patients. Gyn, gynecologic; UGI, urinary and GI.

Incidence, Persistence, Remission and Relapse Rates

Overall, 14.4% of patients had a first incidence of insomnia during the study (9.4% to 15.3%; Table 1). The general persistence rate was 50.7% (28.9% to 63.9%), and 35.0% of patients had insomnia persisting for at least three consecutive time points (2 time intervals and more). The global remission rate was 45.8% (35.7% to 51.0% across cancer sites), and most often (35.9%), the remission lasted only one time interval. Overall, 19.5% of the participants experienced relapse of insomnia during the course of the study. Relapse rates varied according to cancer site between 12.8% (prostate) and 24.0% (gynecologic), and patients with gynecologic cancer were particularly likely to experience enduring relapses (a relapse lasting 2 time intervals: 4.2% v 0.9 [prostate] and 1.4% [breast]). When combining data from patients with a first incidence of insomnia and those who experienced relapse (and eliminating the overlapping patients), the total incidence was 31.8% (25.5% to 35.9%).
Table 1. Incidence, Persistence, Remission, and Relapse Rates for the Total Sample and Each Cancer Site
 Total Sample(N = 856)Breast (n = 426)Prostate (n = 235)Gynecologic(n = 96)
No.%No.%No.%No.%
Incidence
    First episode12314.46214.53615.399.4
    Total*27231.815335.96025.53031.3
Persistence, time intervals43450.727263.96828.94951.0
    113415.78820.7187.71616.7
    2859.95512.9125.11111.5
    3758.84911.5146.066.3
    4263.0102.473.044.2
    511413.37016.4177.21212.5
Remission, time intervals39245.821750.98435.74951.0
    130735.916939.77029.83132.3
    2829.64610.8135.51818.8
    330.420.510.400.0
Relapse, time intervals16719.59722.83012.82324.0
    115518.19121.42811.91919.8
    2121.461.420.944.2
*
Total incidence combines first incidences of insomnia and relapses and eliminates overlapping participants.
Each time interval includes two consecutive time points.

Most Common Sleep Trajectories

Figure 4 shows the most frequent sleep trajectories obtained according to sleep status at baseline. Among patients who were good sleepers at T1, 62% remained good sleepers throughout the study. The other trajectories in that group were much less frequent and indicate the development of transient insomnia symptoms. Among patients who had insomnia symptoms at T1, patients most frequently (23.5%) went into a full remission. The most frequent trajectory in patients with an insomnia syndrome at T1 indicated the persistence of the insomnia syndrome throughout the 18-month period (37.6% of patients).
Fig 4. Most frequent trajectories for each group: (A) Patients with insomnia syndromes (SYN), (B) patients with insomnia symptoms (SX), (C) good sleepers (GS). Curves represent the trajectories with the greatest proportion of patients for each sleep status at baseline. For instance, 37.6% of all patients with SYN at baseline kept that status throughout the duration of the study. Given the three sleep groups and the 5 time points after baseline, we found 49 distinct trajectories for GS, 72 distinct trajectories for patients with SX and 65 distinct trajectories for patients with SYN. Because it was impossible to show all of them, only trajectories comprising at least six patients were illustrated. Trajectories shown represent 236 (78.1%) of the 302 GS, 103 (44.8%) of the 230 patients with SX, and 103 (49.0%) of 210 patients with SYN at baseline.
In this section, we also report the likelihood of changing status at some point during the course of the study (data not shown) as compared with T1. These rates are greater and sometimes exceed 100%, because a patient could switch several times from one status to the other during the 18-month period. Overall, 35.4% of the good sleepers at T1 developed insomnia symptoms at some point during the study, and 10.9% of them developed an insomnia syndrome. In addition, 33.9% of patients with insomnia symptoms at T1 developed an insomnia syndrome at some point during the study. Among patients with an insomnia syndrome at T1, 45.7% became good sleepers and 44.8% progressed to a less severe form of insomnia (insomnia symptoms) at some point during the study.

Probability of Changing Sleep Status

Good sleepers.

Table 2 lists probability rates of changing or keeping the same sleep status at the subsequent time point for each group (good sleepers, insomnia symptoms, and insomnia syndrome) at each time point and for each cancer site. Good sleepers were likely to remain good sleepers (80.0% to 87.1%) at the subsequent time point and less likely to develop an insomnia syndrome (3.1% to 4.7%) than patients with insomnia symptoms (9.3% to 16.0%). Between 13% and 20% of good sleepers at a specific time point developed insomnia at the subsequent assessment. Patients with prostate cancer were the most likely to remain good sleepers between two consecutive time points throughout the study (87.0% to 93.2%) as compared with patients with breast (72.4% to 84.7%) or gynecologic (76.3% to 86.7%) cancer.
Table 2. Probability of Having a Particular Sleep Status According to the Preceding Sleep Status
Preceding StatusCurrent Status (%)
Total Sample (N = 856)Breast (n = 426)Prostate (n = 235)Gynecologic (n =96)
GSSXSYGSSXSYGSSXSYGSSXSY
T2: 2 months
    GS81.814.73.574.221.74.289.49.11.576.915.47.7
    SX47.138.314.640.145.814.164.717.717.748.741.010.3
    SY14.915.769.411.619.269.214.714.770.625.04.270.8
T3: 6 months
    GS80.015.34.772.423.74.088.88.42.876.315.87.9
    SX42.641.416.039.542.218.453.942.33.955.035.010.0
    SY11.615.872.612.513.474.115.221.263.64.833.361.9
T4: 10 months
    GS84.611.63.881.612.95.590.87.71.580.015.05.0
    SX47.743.19.350.040.29.846.753.30.026.347.426.3
    SY12.110.477.512.310.477.48.08.084.018.86.375.0
T5: 14 months
    GS84.812.13.181.813.84.487.010.72.382.515.02.5
    SX51.336.312.453.237.19.743.534.821.758.829.411.8
    SY11.014.874.211.616.871.64.48.787.023.517.758.8
T6: 18 months
    GS87.19.33.784.711.14.293.24.62.386.711.12.2
    SX42.042.915.236.542.920.644.044.012.053.346.70.0
    SY10.89.280.013.39.377.315.411.573.10.00.0100.0
NOTE. For each row in each large column (total sample, breast, prostate, and gynecologic), percentages add up to 100%. These percentages were computed using the available number of participants, which varied according to time point and cancer diagnosis. Numbers shown represent sample sizes available at baseline only.
Abbreviations: GS, good sleepers; SX, insomnia symptoms; SY, insomnia syndrome.

Insomnia patients.

Patients with insomnia symptoms were much more likely to experience remission (ie, become a good sleeper) at the subsequent time point (42.0% to 51.3%), than those with an insomnia syndrome (10.8% to 14.9%). Patients with insomnia symptoms were particularly at risk to develop an insomnia syndrome at T2 (14.6%) and T3 (16.0%) in the total sample, but important differences were found across cancer sites. More precisely, conversions from an insomnia symptoms status to an insomnia syndrome status were the most likely at T3 (18.4%) and T6 (20.6%) for patients with breast cancer, T4 for patients with gynecologic cancer (26.3%), and T2 (17.7%) and T5 (21.7%) for patients with prostate cancer.

Discussion

This study aimed to assess the prevalence and the natural course of insomnia comorbid with cancer over an 18-month period. Overall rates of insomnia (36% to 59%) were even higher than most that have been found in the post-treatment phase.1 The much greater prevalence of the insomnia syndrome throughout the duration of the study (21% to 28%) as compared with the general population (6% to 10%)15,16 is particularly striking. Recent findings11 revealed rates of insomnia syndrome varying from 35% to 43% during chemotherapy, but it is important to note that a more liberal definition was used (assessed with a depression questionnaire and not taking into account its impact on daytime functioning).
As in our previous report on the first two time points,12 final results of this study emphasize the pervasiveness of insomnia at the time of cancer surgery. Prevalence rates significantly decreased throughout the cancer care trajectory but remained elevated even at T6. Moreover, differences across cancer sites that were found at T1 and T212 were still apparent at the subsequent time points. More specifically, patients with breast and gynecologic cancer continued to report more insomnia than patients with prostate cancer. On the basis of a lack of differences between men and women of the UGI group, we previously concluded that these between-group differences were not simply the result of a gender effect,12 but additional studies are needed on this issue. Age (patients with prostate cancer were older) and the presence of nocturnal hot flashes (only 10.4% of patients with prostate cancer in this study received hormone therapy, and 0.8% received chemotherapy) are other possible explaining factors.
The rate of first incidence of insomnia during the study (14.4%) is lower than in the general population (30.7% in a year in LeBlanc et al17), but this can be explained by the high prevalence rates found at baseline. When combining first incidences with relapses, the rate increased to 31.8%. Although incidence rates tended to decrease during the 18-month period, a significant proportion of patients still developed insomnia months after surgery. This suggests that factors occurring later during the cancer care trajectory, such as adverse effects of cancer treatments (eg, hot flashes, fatigue), act as new triggers for the development of insomnia.
Insomnia remissions were frequent during the course of the study (45.8% overall), but persistence rates were even higher (50.7% overall). Patients with insomnia symptoms consistently had a greater probability of remitting at the subsequent time point (ie, becoming good sleepers; 42.0% to 51.3% overall) than those with an insomnia syndrome (10.8% to 14.9% overall). Likewise, persistence rates were much greater in patients with an insomnia syndrome (69.4% to 80.0% overall) than in patients with insomnia symptoms (36.3% to 43.1% overall). In addition, patients with an insomnia syndrome at T1 most frequently kept their insomnia syndrome status throughout the 18-month period (37.6% of all trajectories). These data suggest that insomnia, especially when criteria for an insomnia syndrome are met, is an enduring problem in patients with cancer. Besides being a persistent problem, insomnia is also highly recurrent. Nearly 20% of the patients who previously experienced an insomnia remission had a relapse later on during the study.
Altogether, these data highlight the importance of offering effective interventions specifically targeting insomnia, both at the symptoms and syndrome level, to patients with cancer. Cognitive-behavioral therapy has consistently been found effective in treating insomnia in this population.18,19 On the basis of the data of the present study, it seems that there are several possible windows of opportunity for intervention depending on the sleep status and the cancer site. Clearly, the presence of an insomnia syndrome at the time of diagnosis and surgery should represent a top priority as this disorder is enduring and may lead to several negative consequences. Insomnia symptoms at that time could also represent a target for intervention, although natural remissions are much more likely for these patients. Perhaps a less intense form of intervention (eg, self-help) would be particularly relevant in that case. On the basis of conversion rates from the insomnia symptoms to the insomnia syndrome status, it seems that good windows of opportunity for intervention would be T4 (10 months) for patients with gynecologic cancer and T2 and T5 (2 and 14 months) for patients with prostate cancer. T3 and T6 (6 and 18 months) also seemto be appropriate timing for intervention in patients with breast cancer, although differences across time were less marked in this group.
Strengths of this study include a large sample of patients with diverse cancer sites who were recruited using a population-based procedure, the longitudinal nature of the study, and the use of a semistructured interview and of operational criteria for diagnosing insomnia. This study is also characterized by the limitations that include significant differences on many characteristics between study participants and those who declined, underrepresentation of some cancer sites (too few patients in GI and head and neck categories to make reliable comparisons, no patients with lung cancer), first measure was taken either before or, in most cases, after surgery, and because of the large diversity in cancer treatment regimens, measurements could not be taken in relation to their timing. The lack of a noncancer group to assess to what extent the evolution of insomnia differs when cancer is not present is another limitation. Finally, the fact that the most common sleep trajectories described represented only between 44.8% and 78.1% of all patients needs to be taken into account when interpreting these findings. Future analyses of this data set will explore sociodemographic, medical, and psychological factors to explain variations in incidence, persistence, remission, relapse, and persistence of insomnia over time and across cancer sites.

Acknowledgment

We thank Valérie Tremblay, Lucie Casault, Caroline Desautels, Geneviève Dumont, Dave Flanagan, Nathalie Gagnon, Catherine Gonthier, Geneviève Laurent, Marie-Ève Le May, Julie Maheux, Marie-Esther Paradis, Sylvie Perron, Julie Roy, Sophie Ruel, Élaine Thériault, Claudia Trudel-Fitzgerald, and Maude Villeneuve, who were involved in the recruitment and assessment of the participants or the data entry, as well as the participants who volunteered their time for this study.

Authors' Disclosures of Potential Conflicts of Interest

The author(s) indicated no potential conflicts of interest.

References

1.
J Savard, CM Morin: Insomnia in the context of cancer: A review of a neglected problem J Clin Oncol 19: 895– 908,2001
2.
L Fiorentino, S Ancoli-Israel: Sleep dysfunction in patients with cancer Current Treatment Options in Neurology 9: 337– 346,2007
3.
J Savard, S Simard, J Blanchet, etal : Prevalence, clinical characteristics, and risk factors for insomnia in the context of breast cancer Sleep 24: 583– 589,2001
4.
J Savard, S Simard, S Hervouet, etal : Insomnia in men treated with radical prostatectomy for prostate cancer Psycho-Oncology 14: 147– 156,2005
5.
ME Cooley, TH Short, HJ Moriarty: Symptom prevalence, distress, and change over time in adults receiving treatment for lung cancer Psycho-oncology 12: 694– 708,2003
6.
AL Kenefick: Patterns of symptom distress in older women after surgical treatment for breast cancer Oncol Nurs Forum 33: 327– 335,2006
7.
KS Thomas, J Bower, MA Hoyt, etal : Disrupted sleep in breast and prostate cancer patients undergoing radiation therapy: The role of coping processes Psychooncology 19: 767– 776,2010
8.
K Ahlberg, T Ekman, F Gaston-Johansson: The experience of fatigue, other symptoms and global quality of life during radiotherapy for uterine cancer Int J Nurs Stud 42: 377– 386,2005
9.
JT Hickok, GR Morrow, JA Roscoe, etal : Occurrence, severity, and longitudinal course of twelve common symptoms in 1129 consecutive patients during radiotherapy for cancer J Pain Symptom Manage 30: 433– 442,2005
10.
K Härtl, J Engel, P Herschbach, etal : Personality traits and psychosocial stress: Quality of life over 2 years following breast cancer diagnosis and psychological impact factors Psychooncology 19: 160– 169,2010
11.
OG Palesh, JA Roscoe, KM Mustian, etal : Prevalence, demographics, and psychological associations of sleep disruption in patients with cancer: University of Rochester Cancer Center-Community Clinical Oncology Program J Clin Oncol 28: 292– 298,2010
12.
J Savard, J Villa, H Ivers, etal : Prevalence, natural course, and risk factors of insomnia comorbid with cancer over a 2-month period J Clin Oncol 27: 5233– 5239,2009
13.
CM Morin: Insomnia: Psychological Assessment and Management 1993 New York, NY Guilford Press
14.
PL Roth: Missing data: A conceptual review for applied psychologists Pers Psychol 47: 537– 560,1994
15.
CM Morin, M LeBlanc, M Daley, etal : Epidemiology of insomnia: Prevalence, self-help treatments, consultations, and determinants of help-seeking behaviors Sleep Med 7: 123– 130,2006
16.
MM Ohayon: Epidemiology of insomnia: What we know and what we still need to learn Sleep Med Rev 6: 97– 111,2002
17.
M LeBlanc, C Mérette, J Savard, etal : Incidence and risk factors of insomnia in a population-based sample Sleep 32: 1027– 1037,2009
18.
J Savard, S Simard, H Ivers, etal : Randomized study on the efficacy of cognitive-behavioral therapy for insomnia secondary to breast cancer, part I: Sleep and psychological effects J Clin Oncol 23: 6083– 6096,2005
19.
CA Espie, L Fleming, J Cassidy, etal : Randomized controlled clinical effectiveness trial of cognitive behavior therapy compared with treatment as usual for persistent insomnia in patients with cancer J Clin Oncol 26: 4651– 4658,2008

Appendix

Differences in Refusal and Dropout Rates

Patients who refused to participate were older (age 60.6 v 57.0 years; t[1,668] = −7.1; P < .001) and more likely to have head and neck (χ2[1; n = 1,677] = 5.6; P = .02) or UGI cancer (χ2[1; n = 1,677] = 13.4; P < .001) but less likely to have prostate cancer (χ2[1; n = 1,677] = 15.7; P < .001). Overall, 7.3% of patients experienced cancer recurrence or progression during the course of the study. Among them, 59.4% did not complete all of the study's six assessments compared with 41.0% of patients with no evidence of recurrence (χ2[1] = 8.87; P = .003). No significant difference was found between completing patients and noncompleting patients on their sleep status at baseline (χ2[2] = 1.11; P = .57).

Comparison of Presurgery and Postsurgery Patients

Complementary χ2 analyses were conducted to delineate the influence of cancer surgery on sleep status. As previously reported,12 results at 2 months (T1) revealed that patients who had received surgery before T1 (ie, postsurgery) had a significantly higher prevalence of insomnia syndrome (30.5%) as compared with the presurgery subgroup (20.7%; χ2[1] = 6.98; P = .008). Likewise, presurgery patients were significantly more likely to be good sleepers (53.3%) than patients from the postsurgery subgroup (37.4%; χ2[1] = 5.54; P < .001). No significant between-group differences on sleep status were found at 6 months. Except for relapse rates, which were significantly greater in the postsurgery subgroup (21.5% v 11.3%; χ2[1] = 8.81; P = .003), incidence, persistence, and remission rates did not significantly differ between these two subgroups of patients.

Information & Authors

Information

Published In

Journal of Clinical Oncology
Pages: 3580 - 3586
PubMed: 21825267

History

Published online: August 08, 2011
Published in print: September 10, 2011

Permissions

Request permissions for this article.

Authors

Affiliations

Josée Savard [email protected]
Josée Savard, Hans Ivers, Aude Caplette-Gingras, Charles Morin, Université Laval; Josée Savard, Hans Ivers, Julie Villa, Aude Caplette-Gingras, Laval University Cancer Research Center; Québec, Québec, Canada.
Hans Ivers
Josée Savard, Hans Ivers, Aude Caplette-Gingras, Charles Morin, Université Laval; Josée Savard, Hans Ivers, Julie Villa, Aude Caplette-Gingras, Laval University Cancer Research Center; Québec, Québec, Canada.
Julie Villa
Josée Savard, Hans Ivers, Aude Caplette-Gingras, Charles Morin, Université Laval; Josée Savard, Hans Ivers, Julie Villa, Aude Caplette-Gingras, Laval University Cancer Research Center; Québec, Québec, Canada.
Aude Caplette-Gingras
Josée Savard, Hans Ivers, Aude Caplette-Gingras, Charles Morin, Université Laval; Josée Savard, Hans Ivers, Julie Villa, Aude Caplette-Gingras, Laval University Cancer Research Center; Québec, Québec, Canada.
Charles M. Morin
Josée Savard, Hans Ivers, Aude Caplette-Gingras, Charles Morin, Université Laval; Josée Savard, Hans Ivers, Julie Villa, Aude Caplette-Gingras, Laval University Cancer Research Center; Québec, Québec, Canada.

Notes

Corresponding author: Josée Savard, PhD, Laval University Cancer Research Center, 11 Côte du Palais, Québec, Québec, Canada G1R 2J6; e-mail: [email protected].

Author Contributions

Conception and design: Josée Savard, Charles M. Morin
Collection and assembly of data: Julie Villa, Aude Caplette-Gingras
Data analysis and interpretation: Josée Savard, Hans Ivers,Charles M. Morin
Manuscript writing: All authors
Final approval of manuscript: All authors

Disclosures

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

Funding Information

Supported in part by Grant No. MOP-69073 from the Canadian Institutes of Health Research and a research scientist award from the Fonds de la recherche en santé du Québec (J.S.).

Metrics & Citations

Metrics

Altmetric

Citations

Article Citation

Download Citation

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

For more information or tips please see 'Downloading to a citation manager' in the Help menu.

Format





Download article citation data for:
Josée Savard, Hans Ivers, Julie Villa, Aude Caplette-Gingras, Charles M. Morin
Journal of Clinical Oncology 2011 29:26, 3580-3586

View Options

View options

PDF

View PDF

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Personal login Institutional Login

Purchase Options

Purchase this article to get full access to it.

Purchase this Article

Subscribe

Subscribe to this Journal
Renew Your Subscription
Become a Member

Media

Figures

Other

Tables

Share

Share

Share article link

Share