Elsevier

European Journal of Cancer

Volume 45, Issue 18, December 2009, Pages 3262-3270
European Journal of Cancer

Prevalence and predictors of insomnia in women with invasive ovarian cancer: Anxiety a major factor

https://doi.org/10.1016/j.ejca.2009.05.030Get rights and content

Abstract

The estimated prevalence of insomnia in cancer patients varies between 20% and 50%, which is substantially higher than the general population. To date, little is known about the risk factors for insomnia in patients with cancer. This study examines the prevalence and predictors of insomnia in a population-based sample of women with ovarian cancer.

Participants were 772 women participating in the Australian Ovarian Cancer Study – Quality of Life Study. Insomnia was assessed using the Insomnia Severity Index (ISI). Demographic, disease and treatment variables, and psychosocial variables, including anxiety and depression, support care needs and social support and coping, were investigated as potential predictors of insomnia. Twenty-seven percent of women reported sub-clinical symptoms of insomnia (ISI score 8–14) and 17% reported clinically significant insomnia (ISI score 15–28). Three variables were significant predictors of clinically significant insomnia: young age (<50 years: Odds Ratio (OR) = 2.36; Confidence Interval (CI) 1.06–5.26; 50–59 years: OR = 2.73; CI 1.33–5.64) relative to 70+ years; higher unmet needs in the physical/daily living domain (OR = 1.02; CI 1.01–1.03) and elevated anxiety (sub-clinical anxiety (Hospital Anxiety and Depression Scale (HADS) score 8–10): OR = 1.83; CI 1.04–3.24; clinical anxiety (HADS score 11–21): OR = 2.03; CI 1.08–3.85). In contrast to predictors of primary insomnia, women with cancer aged <60 years were more likely to report clinical levels of insomnia than women of 70+ years. Consistent with primary insomnia, elevated anxiety predicted insomnia in women with ovarian cancer. Given that both anxiety and insomnia are relatively common, and the relationship may potentially be bi-directional, the efficacy of interventions targeting insomnia and anxiety, rather than insomnia alone, is worthy of consideration.

Introduction

Sleep disturbance is both a common occurrence and a neglected problem in cancer patients.1, 2 Insomnia is the most common type of sleep disturbance and refers to a range of sleep-related symptoms. Symptoms include: difficulty in falling and staying asleep; early morning wakening with inability to resume sleep and non-restorative sleep. These symptoms can lead to increased levels of distress and impairment in daytime functioning.1, 3 Psychological and behavioural consequences include daytime fatigue, mood disturbance and psychiatric disorders, and difficulties in memory and concentration.1, 4 Physiological and physical consequences include, greater perception of pain, immunosuppression and may possibly adversely impact on survival.1, 5

In their 2001 review, Savard and Morin1 estimate that between 30% and 50% of newly diagnosed cancer patients report insomnia. Several years after treatment, 23–44% of cancer survivors were still reporting symptoms of insomnia, indicating that insomnia may become a chronic problem if left untreated.2, 6 Many of these studies assessed insomnia using a single item measure, with heterogeneous samples of cancer patients and/or small sample sizes. Using a single item to assess insomnia has been criticised as insufficient to capture the full symptomatology of insomnia and therefore to adequately distinguish between sub-clinical and clinically significant insomnia.7

More recent studies have used clinical interviews or validated questionnaires to assess and classify symptoms of insomnia. Savard and colleagues2 reported that 51% of non-metastatic breast cancer survivors who received radiotherapy had symptoms of insomnia, with 19% meeting criteria for clinically significant insomnia. Davidson and colleagues8 reported that 30% of cancer survivors attending out-patient clinics had clinical levels of insomnia. Insomnia was most common in breast (38%), lung (37%), gastrointestinal (32%) and gynaecological (29%) cancer patients. Similarly, Bardwell and colleagues9 reported that 39% of women with stages I–IIIA breast cancer had clinically significant insomnia up to 4 years post-diagnosis. While definitions and measures vary, sleep disturbance and clinically significant insomnia are a substantial problem in cancer patients, and the prevalence of insomnia is greater than the 7.0–9.5% reported in general population studies.10

Three studies have reported data on sleeping difficulties in women with ovarian cancer. All three used a single item question among a list of symptoms. The estimates varied between 46% and 60% of ovarian cancer patients and survivors,11, 12 while Donovan and colleagues13 found that current ovarian cancer patients reported sleep disturbance as the third most severe symptom from a list of 22 symptoms.

From the general insomnia literature, various risk factors have been identified, including older age, female gender, lower socio-economic status/education, divorced/single status, poorer physical health and physical symptoms such as pain, poorer mental health and life event stress.10, 14, 15 Despite the high prevalence of insomnia in cancer patients, and the likely negative impact of this on quality of life,5 few data are available regarding risk factors for insomnia in cancer patients, and none specifically for women with ovarian cancer.

In 300 non-metastatic breast cancer survivors, Savard and colleagues2 identified sick leave, unemployment, widowhood, lumpectomy, chemotherapy and surprisingly, a lower stage of cancer that were associated with sub-clinical and clinically significant insomnia. Davidson and colleagues8 reported that fatigue, younger age, leg restlessness, use of sleeping pills, low mood, frightening dreams, concerns and recent cancer surgery, were associated with insomnia in a heterogeneous group of cancer survivors. In patients registering for palliative care, Akechi and colleagues16 reported younger age, diarrhoea and living alone as associated with current insomnia. Increased psychological distress between baseline and follow-up significantly predicted insomnia at follow-up, while increasing physical symptoms, declining physical functioning and use of new drugs potentially impacting on sleep, either positively or negatively, were not predictive.

Bardwell and colleagues9 have undertaken the only comprehensive study of risk factors for insomnia in cancer patients. In 2645 early breast cancer survivors, cancer-related variables, personal characteristics, health behaviours, physical symptoms and psychosocial variables were examined as potential predictors. Several factors were associated with insomnia in univariate analyses, including lower education, less physical activity, more pain, vasomotor, genitourinary, gastrointestinal symptoms, less social support, more social strain, life events and depressive symptoms. However, in multivariate analysis, only depressive symptoms and vasomotor symptoms (night sweats in particular) were significant predictors of insomnia.

In summary, research to date indicates that insomnia is a significant problem for cancer patients and survivors, and suggests that the prevalence of insomnia is significantly greater than in the general population. We identified only four studies exploring predictors of insomnia, two in heterogeneous samples of cancer patients and two in women with early breast cancer. Younger age and low mood were the only significant predictors of insomnia in more than one study.

The aims of the current study were to:

  • 1.

    Describe the prevalence of insomnia in a population-based sample of women with both early and late stage ovarian cancer.

  • 2.

    Identify demographic, medical and psychological predictors of insomnia in women with ovarian cancer.

Section snippets

Australian Ovarian Cancer Study Quality of Life Study (AOCS QoL Study)

The Australian Ovarian Cancer Study (AOCS) is a population-based study, recruiting women aged 18–79 years with epithelial ovarian cancer (including fallopian tube and primary peritoneal cancers) between 2002 and 2006. Details of the study have been described elsewhere.17 In short, women were recruited through major treatment centres (prior to surgery when possible) and state-based cancer registries. AOCS collects detailed behavioural, environmental and dietary data relating to the period ending

Participant demographic and disease characteristics

Between May 2005 and December 2006, 1204 eligible AOCS participants were invited into the QoL study. Of these women, 293 women declined and a further 115 did not complete the questionnaire. A total of 796 women completed the baseline questionnaire (66%). Clinical and histopathological characteristics of the cohort are displayed in Table 1. Responders and non-responders (including refusals, deaths and loss of contact) were similar in terms of age and marital status. Responders were more likely

Discussion

In this large and representative sample of women with ovarian cancer, insomnia was reported by 44% of the sample, with 17% (130/772) experiencing clinically significant insomnia. This result is similar to the earlier studies of cancer patients, in which 20–50% have sub-clinical symptoms of insomnia and about 20% with clinically significant insomnia.1, 2, 8, 9, 16, 24 Previous studies of women with ovarian cancer have not focused specifically on insomnia but have reported rates of sleeping

Ethics statement

The work has been approved by the appropriate ethical committees related to the institutions in which it was performed, and participants have given informed consent.

Role of funding source

No involvement in study.

Funding

This study was funded by The Cancer Council New South Wales and The Queensland Cancer Fund (RG 36/05). Financial support for the parent study was provided by the US Army Medical Research and Materiel Command under DAMD17-01-1-0729, the National Health and Medical Research Council of Australia (400413 and 400281) and the Cancer Councils of New South Wales, Queensland, South Australia, Tasmania and Victoria and the Cancer Foundation of Western Australia. Additional recruitment was conducted under

Conflict of interest statement

None declared.

Acknowledgements

Full membership of the AOCS Group is listed at http://www.aocstudy.org/. We gratefully acknowledge the contribution of all our clinical and scientific collaborators and the cooperation of the following institutions: New South Wales: John Hunter Hospital, North Shore Private Hospital, Royal Hospital for Women, Royal North Shore Hospital, Royal Prince Alfred Hospital, Westmead Hospital and New South Wales Cancer Registry; Queensland: Mater Misericordiae Hospital, Royal Brisbane and Women’s

References (32)

  • A.M. Berger et al.

    Current methodological approaches to the study of sleep disturbances and quality of life in adults with cancer: a systematic review

    Psychooncology

    (2007)
  • J. Savard et al.

    Randomized study on the efficacy of cognitive-behavioral therapy for insomnia secondary to breast cancer, part II: Immunologic effects

    J Clin Oncol

    (2005)
  • K.A. Lee et al.

    Critical components of a sleep assessment for clinical practice settings

    Issues Mental Health Nurs

    (2005)
  • J.R. Davidson et al.

    Sleep disturbance in cancer patients

    Soc Sci Med

    (2002)
  • W.A. Bardwell et al.

    The relative importance of specific risk factors for insomnia in women treated for early-stage breast cancer

    Psychooncology

    (2008)
  • M.I. Fitch et al.

    Canadian women’s perspective on ovarian cancer

    Cancer Prev Control

    (1999)
  • Cited by (41)

    • Insomnia and its association with quality of life in women with ovarian cancer

      2020, Gynecologic Oncology
      Citation Excerpt :

      Thus, overall more than one-in-three women likely experienced clinical insomnia and 58% of these were treated with sleep medication. The prevalence of insomnia we observed at each time point was similar to that reported by other cross-sectional studies (14–24%) in women with ovarian cancer [9,10] and the proportion of women with a history of insomnia was consistent with that of the general population (6–10%) [25]. Among the subgroup of women without pre-existing insomnia at diagnosis over one-in-five women experienced new-onset clinical insomnia symptoms after their cancer diagnosis, 17% after the removal of the effect of chemotherapy.

    View all citing articles on Scopus
    View full text