Are We Really Getting Less Sleep than We Did in 1975?
Is sleep harder to get today than 30 years ago?
Posted Jan 18, 2010
We frequently hear that life has become more complex, stressful and time pressured over the past few decades. It is typically assumed that this has led to people getting less sleep than in the past. This is a concern because short sleep durations have been linked to major health issues such as obesity and hypertension. This raises concern about what happens when people with a biological need for 7 or 8 hours of sleep a night are regularly getting less than this amount due to the demands of modern living.
A recent article by Kristen Knutson and her colleagues published in the journal "Sleep" tried to address this issue. These researchers looked at diaries collected as a part of eight previous national time use surveys conducted between 1975 and 2006. Subjects in these studies were adults in the United States 18 years of age or older. All of the surveys were designed to be representative of the adult population of the US and were kept for a 24 hour period. In none of these surveys was there any particular focus for the diaries, such as sleep. This reduced the potential for bias of respondents in overestimating the amount of time spent on a specified activity because their attention has been drawn to it.
Previous reports do indeed indicate an increase in short sleep periods (less than 6 hours of sleep per 24 hour day) over this time frame. For example, the American Cancer Society survey of over one million people in 1959 to 1960 found that 2% of the sample reported getting less than 6 hours of sleep per night. A number of later surveys in the 1960s and 1980s indicated significant increases in the number of adults reporting short sleep durations. By 2004 the National Health Interview Survey indicated that for adults between 30 and 64 years of age about 30% were getting 6 hours or less of sleep per night. Given what we know about the impact of sleep loss on health and the potential for accidents, such increases are indeed alarming.
Knutson and colleagues were interested in identifying factors such as marital status and socioeconomic level that might influence the chance of people reporting short sleep. They reviewed the diary data from the eight previous time use surveys and calculated the amount of time respondents spent sleeping, napping or resting. They defined short sleepers as those reporting less than 6 hours of sleep per day in their journals. This was a conservative estimate as they combined sleeping, napping and resting into the sleep category as this would mean that the individuals could not have got more than 6 hours of actual sleep in the 24 hour period. They also looked at variable such as employment status, age, sex, education, income, marital status and daily activities.
They noted some changes in the diary-keeping samples from the 1970s to the 2000s such as an increase in the proportion of non-whites and those with some college education. These changes reflected well-known national tends. The percentage of short sleepers increased from 7.6% in 1975 to 9.3% in 2006. They also calculated the odds ratio for short sleep in each of the studies as compared to the 1975 study after adjusting for such variables as age, sex and employment. The odds ratio calculates the ratio of the odds of an event occurring in one group to the odds of it occurring in another group, in this case later samples as compared to the baseline 1975 sample.
This analysis showed that there was indeed a significant change in the odds of short sleep from 1975 to 2006. Much of the increase in short sleep appears to be due to the increase in extremely short sleep periods of less than 5.5 hours. Demographically, people 45 to 54 years of age were most likely to be short sleepers, women were less likely to be short sleepers than men, single and divorced individuals were more likely to be short sleepers than married people, those with some college education were more likely to be short sleepers than those with less education, and African Americans were more likely to be short sleepers than whites, Asians and Hispanics.
Perhaps most interesting of all the findings was that the only factor associated with increased risk of short sleep over the time from 1975 to 2006 was full-time work status. In other words, factors such as sex and age showed similar odds of short sleep over the 31 year period except for working full time. Full time workers, based on the 24 hour time use diaries collected over these different time periods, have greater odds of being short sleepers in 2006 then they did in 1975.
These findings do call into question the results of the other surveys noted above that have shown an alarming increase in short sleep in recent decades. The change found in this review was only from 7.6% to 9.3% and was accounted for by those working full time. The primary limitations of these time use diary studies are that there are no data related to physiological sleep as opposed to self report of sleep and there may be biases in the samples such as more time-pressured individuals being less likely to complete this kind of survey.
There are data showing that working overtime is associated with problems such as high blood pressure, depression and obesity. Knutson et al point out that short sleep could be a factor in this finding. Given the large population of full time workers, even after the effects of the Great Recession, the impact of even mildly decreased sleep duration could be significant for our country and the health of workers.