Academic journal article Psychological Test and Assessment Modeling

Measurement Properties of PROMIS Sleep Disturbance Short Forms in a Large, Ethnically Diverse Cancer Cohort

Academic journal article Psychological Test and Assessment Modeling

Measurement Properties of PROMIS Sleep Disturbance Short Forms in a Large, Ethnically Diverse Cancer Cohort

Article excerpt


Sleep problems are common for cancer patients, both during and after treatment (Garland et al., 2014). The prevalence of insomnia ranges from 30 % to 60 % (Savard, Simard, Blanchet, Ivers, & Morin, 2001); moreover, during chemotherapy, patients are three times more likely to report insomnia than the general population (Palesh et al., 2010). After treatment, insomnia symptoms can persist for up to 2 to 5 years (Savard & Morin, 2001). For cancer survivors, symptoms of insomnia frequently result in a higher risk of future physical and mental health problems, and subsequently, in a poorer quality of life (Garland et al., 2014). The availability of a valid and reliable self-report measure of sleep disturbance can help screen and identify cancer patients with clinically-relevant problems, monitor and evaluate supportive care services, and identify effective interventions; with the goal of improving the overall quality of life and functional ability of those assessed.

The Patient-Reported Outcomes Measurement Information System® (PROMIS®), a U.S. National Institutes of Health Common Fund initiative, includes a number of extensive item response theory (IRT)-calibrated item banks. Researchers can choose to administer domain items by computerized adaptive testing (CAT) or can select a subset of items for use as fixed length short forms (Buysse et al., 2010). The PROMIS Sleep Disturbance IRTcalibrated item bank includes items that measure perception of sleep quality, depth of sleep, satisfaction with sleep, and perception of difficulty getting and staying asleep (Cella et al., 2010). Qualitative methods were used to ensure content validity of the sleep disturbance domain, including issues commonly identified by cancer patients (Flynn et al., 2010). Currently, four PROMIS Sleep Disturbance fixed-item short forms are widely available. These short forms vary in number of items (e.g., 4-, 6-, and 8- items) and are scored on the same t-score metric (50 = U.S. population mean score), with the longer short forms reporting higher reliability (Yu et al., 2011). Initial validation work has provided evidence that supports the reliability and validity of the PROMIS Sleep Disturbance measures in small clinical populations (Cella et al., 2010), and internet-based general samples (Buysse et al., 2010). However, to date, the PROMIS Sleep Disturbance measures have not been validated in an ethnically diverse, community-based sample, nor among oncology patients. Demonstrating the validity and reliability of the PROMIS Sleep Disturbance measure in this community-based sample of cancer patients will ensure that it is appropriate for use with study participants reporting mild to severe sleep disturbance.


Sample. Partìcipants were from the Measuring Your Health (MY-Health) study cohort. Overall study design, recruitment strategy, and demographic characteristics of the MYHealth sample used in these analyses are described in the companion issue (see Jensen et al., 2016). We excluded participants without cancer stage, age, or race/ethnicity information, or those who did not answer all 10 sleep disturbance items in order to ensure all reliability and validity testing was completed using a single uniform cohort (n = 4,956).

Demographic and clinical variables. We collected registry-based information on age at diagnosis, sex, date of cancer diagnosis, cancer type, and cancer stage. Participant selfreport information was collected on receipt of chemotherapy and hormonal therapy, race/ethnicity, comorbid conditions (number and type), education level, current employment status, annual income, marital status, insurance coverage, and information concerning whether or not the participant was born in the U.S.

Sleep disturbance. We evaluated the psychometric properties of three previously established sleep disturbance forms (4a, 6a and 8b forms). The 10 item form administered in this study also includes items that are frequently selected in the online PROMIS CAT assessment (Table 1). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.