Sleep quality refers both the subjective assessment given by the subject of how restorative and undisturbed his/her sleep has been (via a standardized questionnaire) and to a series of objective measures which may be derived from polygraphic recordings (in the laboratory or at home) or from recordings of wrist activity movements (wrist actigraphy monitoring), and/or head movements and eyelid movements ("Nightcap" monitoring). Subjective and objective measures of sleep quality are not necessarily concordant. The most commonly used objective measure of sleep quality is an index of sleep fragmentation which may be derived from all three types of recordings. However, the amount and depth of nonREM sleep, the amount of REM sleep and the temporal organization of nonREM and REM stages are clearly major components of the complex concept of sleep quality.

Sleep exerts major modulatory effects on endocrine function (1). For some hormones (e.g. GH), 50-75% of the total daily secretion is dependent on sleep and is eliminated by total sleep deprivation. There is good evidence to indicate that sleep disorders are associated with decreased levels of IGF-1, presumable because of reduced GH secretion. For other hormonal axes (i.e. the corticotropic and thyroid axes), sleep exerts inhibitory influences, resulting in lower nocturnal concentrations when the subject is asleep than if the subject remains awake throughout the night. Finally, sleep also affects endocrine function on the following day. Following a night of partial or total sleep loss or several nights of sleep curtailment, the quiescent period of cortisol secretion is abridged and evening cortisol levels are elevated.

A well validated instrument for the measurement of subjective sleep quality is the Pittsburgh Sleep Quality Index questionnaire which provides a global score of sleep quality on a scale of 1-21 (2,3).

The most commonly used measure of objective sleep quality is "sleep efficiency", defined as the percentage of time in bed spent asleep. Thus, if a subject wishes to have an 8-hour sleep period, takes 30 min to fall asleep, wakes up 30 min before the scheduled end of the sleep period and experienced a number of awakenings interrupting sleep for a total of 60 min, the sleep efficiency is 6 hrs/8 hrs = 75%.

Since there is good evidence to indicate that reduced sleep duration and/or quality is associated with decreased activity of the somatotropic axis and increased activity of the corticotropic axis, measurements of IGF-1 concentrations and of evening cortisol levels may be considered as endocrine markers of sleep quality. Limitations on the use and interpretation of IGF-1 include the need to draw blood and the fact that the hormone is affected by multiple factors besides sleep quality. In contrast, accurate measurements of biologically active cortisol levels may be obtained on saliva samples and multiple saliva samples collected between early evening (e.g. 18:00) and bedtime provide an excellent estimation of the "quiescent period of cortisol secretion".

To the best of my knowledge, there is no published information on the possible relationship between sleep quality and SES. It is likely that individuals with high SES may have more control on their sleep duration and sleep conditions and also more opportunities to recover sleep following sleep loss. In an ongoing study, we have shown that metabolic and endocrine disturbances associated with a sleep debt accumulated over several consecutive days may be entirely reversed by extending the rest period for a few days. The opportunity to recover sleep loss is therefore likely to be an important component of health status across adulthood.

High SES is rarely found in shift workers, who have a chronic sleep debt and are at high risk for a number of pathologies, including cardiovascular, digestive, and psychiatric disease. The relative contribution of the accumulated sleep loss versus the misalignment of the imposed sleep-wake cycle and the endogenous circadian rhythms in the development of these pathologies remain to be demonstrated.'

Finally, another area where sleep quality and SES may interact is urban violence. In inner city slums, the night is a time of increased danger and violence which is not conducive to good sleep quality. Interestingly, a number of studies have shown that sleep loss increases irritability and depresses mood. The possible contribution of sleep loss to violent behavior has never been considered, let alone investigated.

The importance of good sleep for good mental and physical health may seem obvious but is still a matter of controversy in the field. Well-controlled studies have shown that total sleep deprivation kills a rat in approximately two weeks. Less drastic studies in humans have indicated that sleep deprivation impairs immune function. Sleep apnea is associated with a variety of neuroendocrine disorders. Our own studies have indicated that sleep loss is associated with an alteration in hypothalamo-pituitary-adrenal (HPA) function on the following day, consisting of an elevation of evening cortisol concentrations (4). The normal day long decline in cortisol levels which follows the early morning elevation appears to occur at a slower rate in sleep-deprived subjects or in subjects running a "sleep debt".