Article Text

A multicomponent cognitive–behavioural intervention for breast cancer survivors with insomnia improved perceived sleepCommentary
  1. Gene Elizabeth Harkless
  1. University of New Hampshire Department of Nursing, Durham, New Hampshire, USA

Statistics from

D R Epstein

Dr D R Epstein, Carl T Hayden Veterans Affairs Medical Center, Phoenix, AZ, USA; dana.epstein{at}


Is a multicomponent cognitive–behavioural intervention effective for chronic insomnia in breast cancer survivors?



randomised controlled trial.





Follow-up period:

2 weeks after intervention.


a university and a medical centre {in Phoenix and Tempe, Arizona, USA}.*


81 women 29–86 years of age who completed surgery, chemotherapy, or radiation therapy for stage 1–3 breast cancer ⩾3 months before study entry and had sleep onset latency or time awake after sleep onset ⩾30 minutes for ⩾3 nights/wk for 2 weeks. Exclusion criteria included cognitive impairment and suspected sleep apnoea, restless legs syndrome, or periodic limb movement disorder; or Brief Symptom Inventory global severity index T score >70.


multicomponent intervention (stimulus control instructions, sleep restriction, and sleep education and hygiene [SEH]) (n = 40) or SEH only (control) (n = 41). Stimulus control instructions associated the bed and bedroom with falling asleep or back to sleep, strengthened the bed and bedroom as cues for sleep, and weakened them as cues for activities interfering with sleep. Sleep restriction reduced time in bed (minimum 5 h), using individualised sleep–wake schedules. Both stimulus control and sleep restriction targeted factors perpetuating insomnia. SEH included general information about sleep processes and functions, circadian rhythms, individual sleep needs, sleep deprivation, and recommendations to improve sleep. Both groups received 4 weekly in-person group sessions followed by 2 weekly individual telephone sessions, kept sleep diaries, and wore wrist actigraphs.


self-perceived improvement in sleep over the course of treatment, including 4 sleep evaluation items (improvement in sleep-onset latency, waking after sleep onset, total sleep time, and quality of sleep), and sleep-onset latency, waking after sleep onset, total sleep time, time in bed, sleep efficiency, and sleep quality.

Patient follow-up:

89% (mean age 58 y).


The multicomponent group had greater self-perceived improvement in sleep over the course of treatment than the control group (p<0.02). Sleep diaries showed that both groups improved over time in sleep-onset latency, waking after sleep onset, total sleep time, time in bed, sleep efficiency, and sleep quality (p = 0.000). However, groups did not differ for these variables, except that the multicomponent group reported less time in bed than the control group (table).

Multicomponent intervention v sleep education and hygiene only (control) for chronic insomnia in breast cancer survivors*


A multicomponent cognitive–behavioural intervention resulted in greater perceived improvement in sleep and reduced time in bed compared with a single-component intervention in breast cancer survivors with chronic insomnia.

*Information provided by author.


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

Clinical impact ratings: Oncology 6/7


Sleep is difficult for approximately one-third of adults in the US, with one-half of those eligible for a diagnosis of some type of chronic insomnia;1 this prevalence is estimated to double in women with a history of breast cancer.2 Defining insomnia as difficulty initiating or maintaining sleep, Epstein and Dirksen discuss the usefulness of non-pharmacologic interventions for reducing insomnia. They also made a good case for the need to study non-pharmacologic interventions for co-morbid insomnia in specific clinical populations such as cancer survivors.

An important clinical lesson from the findings is the potential therapeutic value of SEH. SEH improved sleep outcomes in both groups, albeit with lower change scores and lower evaluations of sleep improvement in the control group. Improvements in control group scores may be a result of both groups receiving identical SEH. Furthermore, Epstein and Dirksen postulated that in group sessions, the use of bar graphs to depict time in bed for sleep may have exposed participants in the SEH only group to some of the multicomponent sleep restriction content.

Both groups reduced time spent in bed, with the multicomponent group decreasing by as much as 71 minutes more than the control group. Notwithstanding the large between-group difference in time spent in bed, the modest effect sizes call into question the resources required to implement multiple group meetings and individual telephone support and the practicality of the interventions. Using a booklet for the SEH intervention in future research will help to address these issues. Limited generalisability is also a concern because patients were mostly white, well educated, and on average 5–7 years out from their diagnosis, with no evidence of active disease. Future research should target more heterogeneous populations and address whether sleep experiences differ in women of similar ages who have had more recent diagnoses or have not had breast cancer.


View Abstract


  • Source of funding: National Cancer Institute.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.