Table 3 Sleep outcome variables, potential explanatory variables and their pre-specified associations in the Tromsø Study

Initial screening of all eligible participants in the Tromsø Study:

   • Prevalence of sleep disturbances reported in the initial questionnaire (Questionnaire 1) containing a four-item response option, where sleep disturbance is defined as the average occurrence of sleep problems "more than once a week".

10 to 50% in the adult population.

[2, 76]

   • Sleep disturbances and their possible relationships, including demographic variables, health-related variables, and lifestyle and socioeconomic variables, and biological markers.

Increased risk for impairments in health, decreased quality of life and increased healthcare utilization.

[19, 65, 77-80]

Further mapping with validated sleep assessment questionnaires among subjects with sleep complaints selected in the initial screening and their control group:

   • Sleep outcome variables related to the use (dose and frequency) of specific prescription drugs, derived from the Norwegian Prescription Database (NorPD).

Frequent use of hypnotics and over-the-counter (OTC) medications.

[36, 37]

   • Sleep outcome variables related to socioeconomic variables.

Insomnia related to unemployment and socioeconomic deprivation, more common in women, elderly and individuals living alone.

[81, 82]

   • Sleep outcome variables and excessive daytime sleepiness related to objective medical diagnosis, such as diabetes and coronary disease.

A possible link between short sleep, diabetes mellitus, coronary heart disease and all over mortality.

[21, 25-28]

   • Sleep outcome variables related to the diagnosis of restless legs.

Relation between symptoms of restless legs and insomnia

[15]

   • Sleep outcome variables related to self-reported complaints of musculoskeletal symptoms.

Relation between symptoms of musculoskeletal symptoms and insomnia

[83]

   • Sleep outcome variables related to self-reported psychiatric symptoms

Elevated risk of depression and anxiety disorders in people with persistent insomnia

[58, 77, 84-86]

   • Sleep outcome variables related to self-reported frequency of nightmares.

Possible association between nightmares and being a woman, feeling depressed after a poor night's sleep, and complaining of a long sleep latency.

[87]

   • Sleep outcome variables related to excessive daytime sleepiness and reported actual sleep length.

Daytime sleepiness inversely related to hours of sleep and positively related to the ease of falling asleep at night, especially among young adults.

[88, 89]

   • Sleep outcome variables and excessive daytime sleepiness related to subjective perceptions of pain

Chronic pain populations are more likely to experience chronic insomnia, sleep maintenance problems, and/or nonrestorative sleep.

[17, 35, 90]

   • Sleep outcome variables and excessive daytime sleepiness related to doctor-seeking behaviour.

Although insomnia is related to more consultations with GPs, many individuals hesitate to consult their doctor about insomnia

[72, 82]