Six Hours of Sleep: Is It Enough for Some Patients?

Aaron B. Holley, MD


April 02, 2015

Sleep as a Vital Sign

Obtaining sufficient sleep is a crucial component of proper health maintenance. Some authors have gone so far as to recommend that sleep be considered an additional "vital sign" to be assessed at each healthcare provider visit.

There are multiple facets to a comprehensive sleep evaluation for an individual patient; however, one of the most important measures is average sleep duration each night. Although all might agree on the importance of total sleep time,[1] it isn't clear what the nightly target should be. In other words, how much sleep is enough for your patient? Before providing an answer to this surprisingly complex question, it's worth taking the time to review the science of health outcomes as they relate to sleep.

Sleep affects many aspects of health and function,[2,3] which I like to separate into three broad categories: cognitive, mental/behavioral, and physiologic. Cognitive function includes such processes as memory, learning, and abstract reasoning; whereas mental/behavioral health can include general quality-of-life indicators (see the functional outcomes in sleep questionnaire[4] for more detail) and risks for such specific diagnoses as anxiety or depression. The category of general physiology includes appetite, weight gain, and metabolism,[5] along with processing of carbohydrates,[6] and effects on cardiovascular health.[7] I use these three general categories when I discuss the effects of sleep deprivation with my patients because I feel that they are easily comprehended by most people. Insufficient sleep has far-reaching effects on human function, so these three categories should not be considered all inclusive.

Setting Sleep Targets

A group of experts from multiple organizations has been meeting to set targets for sleep duration in the general population. They were tasked with assessing eight different domains of sleep-related health outcomes when assessing the literature. Most of the domains are included in the three categories described above. A consensus recommendation was required for each domain, and recommendations across all domains were summarized to provide targets for specific age groups.[1]

The National Sleep Foundation just published the results, which are essentially age-based recommendations for nightly sleep duration.[8] For sleep physicians who care for adults, a quick glance at the abstract confirms what most in the field already know—adults should be targeting 7-9 hours per night. The text, although very brief and short on detail, is more nuanced on these adult targets. A new category has been created—6 hours per night "may be appropriate" for adults.

Before discussing the implications of this new category, I would like to review what we know about how it was derived. Experts were asked to rate the "appropriateness" of a given sleep duration by using a scale of 1 to 9 (1 = "extremely inappropriate number of hours of sleep," and 9 = "extremely appropriate number of hours of sleep"). Using the RAND/UCLA Appropriateness Method,[9] they synthesized the responses to achieve the final targets. Of interest, for "young adults" and "adults," 6 hours per night received a median score of 6 on their scale,[8,10] hence the creation of the "may be appropriate" category.

Although the consensus authors adjusted the sleep targets by age, everyone in the sleep community knows that sleep requirements vary on the basis of many factors. Some factors (such as genetics and comorbid disease) are specific to the host, whereas others (such as stress and physical activity levels) are dependent on an individual's environment. Furthermore, the required sleep duration may vary according to the desired outcome. A given patient may optimize cognitive function at one duration but physical function at another.[11] So in one sense, it's perfectly logical to say that 6 hours may be appropriate in certain situations. The questions are: In what situations are 6 hours appropriate, for how long, and for whom?

Who Can Get By on 6 Hours of Sleep?

The stakes are fairly high. From 10% to 30% of the US adult population sleeps fewer than 6 hours per night,[12,13] and members of the US military average slightly more than 6 hours of nightly sleep. In short, a large portion of the adult population in the United States averages 6 hours of nightly sleep. Given what we know about the subjective response to chronic sleep debt, it's fair to assume that most if not all of these people do not appreciate the deficits associated with the cumulative sleep debt that they have accumulated.[2] Furthermore, considerable data show that getting only 6 hours of sleep has real consequences. Although deficits tend to manifest more quickly and prominently with restriction to fewer than 4 hours, cumulative deficits do occur with 6 hours of sleep per night.[2,14,15]

Who are these people for whom 6 hours of sleep "may be appropriate?" The new recommendations, as currently published, don't provide much detail on this point, though they promise that a comprehensive account of their systematic review will be published "in the near future." A reasonable place to start might be to determine what percentage of the population can function on 6 hours of sleep. If we knew this number, we would have a pretest probability to start with to calculate the likelihood that a given patient could retain reasonable health and function at this average sleep duration.

There are data to suggest that susceptibility to sleep debt exhibits "trait-like" characteristics,[11,15,16] meaning that individuals have an inherent ability (or lack thereof) to perform well under varying amounts of sleep restriction. Translation—"one person's 6 is another's 8 hours per night for optimal performance on a given task." Unfortunately, we are a long way away from being able to perform genetic testing to predict resistance to sleep debt, and it's unlikely that a single gene is responsible.[17] It's difficult to identify a proportion of the population for whom 6 hours "may be appropriate" because most studies of sleep restriction have enrolled small patient numbers. Considerable sampling error might result from the attempt to extrapolate the proportion of resistant patients in these small studies to the entire adult population. Furthermore, most studies on the cognitive and neurobehavioral responses to sleep debt have only enrolled healthy, younger adults. Older adults with comorbid diseases known to affect sleep might show very different responses. Therefore, hypothesizing a population-based pretest probability on the basis of prevalence data from existing studies is problematic.

What about clinical correlates? For example, could differences in sleep architecture be identified through polysomnography (PSG) or specific clinical characteristics (sleep need or total sleep time) that might reflect these traits? Because so much variability exists across traits and sleep need (or pattern) does not correlate with resistance to deprivation, it seems unlikely that clinical history or PSG data could provide specific guidance. Early data indicate that neither would be helpful for identifying those with adequate resistance to sleep debt.[18,19]

The Outcomes of Sleep Deprivation

Recently, one group studied whether trait-like vulnerability to acute, total sleep deprivation correlates with vulnerability to chronic sleep deprivation.[16] They studied healthy volunteers after 2 nights of acute sleep deprivation and after 7 days of sleep restriction to 3 hours per night. The two experimental periods were 2-4 weeks apart and administered in randomized order. A variety of cognitive and neurobehavioral outcomes were assessed. Trait-like susceptibility to acute and chronic sleep deprivation was correlated in the subjects. This finding suggests that in situations in which resistance to sleep debt will have critical implications for job performance, a 2-day exposure to sleep deprivation might predict the individual's capacity to accomplish the necessary tasks. This is very promising research, albeit with many limitations.[15]

I have yet to mention the physiologic consequences of sleep loss (the last of the three categories I talk to my patients about). Do trait-like characteristics in the response to chronic sleep deprivation extend to physiologic effects, such as weight gain or appetite change? One would assume that they do, but this area hasn't been well studied. We simply know that these physiologic changes are more common in patients who sleep fewer than 7 hours per night.[14]

In summary, recommending a specific duration of sleep for a specific individual will be problematic for the primary care practitioner. Even if reliable tests were available to identify the patient who can maintain task performance on 6 hours of sleep, it's not likely that such tests would extend to predicting performance on other tasks or to predicting physiologic responses. We have good data on historical and occupational risk factors for behavioral sleep restriction, which will help clinicians identify the patients who are likely to have chronic sleep debt.[20,21] The degree to which sleep debt affects performance and overall health will be difficult to tease out. For employers in certain industries, subjecting applicants or employees to a brief period of total sleep deprivation to predict their future performance might make sense, but implementing such a policy would probably be associated with significant legal and economic challenges.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.