Early in the morning coffee is taken mostly to awaken. During the day coffee is taken more for conviviality (17%) and relaxation (34%) rather than for stimulation (14%); only 7% take coffee to cope with stress (Harris Research Centre, 1996). Support for this comes from a study done by Hofer et al. (1993) in which 120 students were put on a strict abstinence regimen, after which they received caffeine during 12 complete days. Although caffeine abstinence caused moderate and transient withdrawal effects, there was no so-called titration of caffeine, that is, coffee consumers did not consume more when the coffee contained less caffeine. Apparently, caffeine itself is a minor reason for coffee consumption, although the studies by Hughes' team repeatedly show that abstained coffee drinkers prefer caffeinated coffee above decaffeinated coffee (Hughes et al., 1995).
These motives to drink coffee, essentially all of a positive nature, imply that the disturbing effects of coffee on sleep are confounded by other aspects. Illustrative of this view is research by De Groen et al. (1993), who studied snoring and anxiety dreams in 98 veterans from World War II. Fifty-five of them suffered from current posttraumatic stress disorder. The outcome showed that the association between snoring and anxiety dreams was independent of many factors that were expected to be related, one of which was coffee consumption. A comparable study was done in 14,800 male twins, born between 1939 and 1995, who served the army in Vietnam between 1964 and 1975 (Fabsitz et al., 1997). Responses were collected from 8870 men on the frequency of their sleep problems as reported on the Jenkins sleep questionnaire, which inventories the prevalence of at least one sleep problem per month. Sixty-seven percent of the respondents awoke often, 61.5% awoke tired or worn out, 48.1% experienced trouble falling asleep, and 48.6% awoke early. It appeared that of the 11 conditions inventoried, coffee consumption of at least eight cups per day vs. up to seven cups per day was related only to awaking tired (odds ration [OR] 1.32), while heavy alcohol use and type A behavior were associated with a higher risk for all sleep problems. The conclusion was that a number of the risk factors associated with these sleep problems came from lifestyle characteristics or stress.
The same conclusion can be drawn from a study of locomotive engineers and their spouses (Dekker et al., 1993). Twenty-seven engineers who were working irregular work schedules and their spouses completed daily logs for 30 d. These logs were divided into workdays and nonwork-days. Workday sleep length was significantly shorter than nonworkday sleep length for both subject groups. The number of cups of coffee consumed on workdays was higher (2.75 cups per day) than on nonworking days (2.17 cups per day), but only for the locomotive engineers. The authors concluded that increased coffee consumption was correlated with longer sleep latency, increased negative mood, and decreased positive mood on both work and nonwork days. Driving a locomotive is a taxing task that demands continuous vigilance; the stress of this combined with the frequent intake of coffee to compensate for this stress may have caused this decrease in sleep quality and feelings of well-being.
The same conclusion may apply to a study by Ohayon et al. (1997), who researched the prevalence of snoring and breathing pauses during sleep in 2894 women and 2078 men aged 15 to 100 years, a representative sample of the U.K. population. Forty-five percent of this sample reported snoring regularly, which was associated with the male sex, aged 25 years or more, and consuming at least 6 cups/d (OR 1.4, p < .002). Since snoring was also associated with obesity, daytime sleepiness or naps, nighttime awakenings, and smoking, it could be that, as found in the former studies, an inadequate lifestyle was the causal factor of the sleep-related problems, and not caffeine itself.
The same line of reasoning goes for the restless legs syndrome and periodic limb movement disorder (PLMD), two other sleep-impairing disorders. Cross-sectional studies in the U.K., Spain, Italy, Portugal, and Germany among 18,980 subjects, 15 to 100 years old, revealed that caffeine intake was not associated with restless legs syndrome, although it was with PLMD (Ohayon and Roth, 2002). The specific factors associated with PLMD included being a shift or night worker, snoring, daily caffeine intake, use of hypnotics, and stress.
Depression may lead to bad sleep, but stress is not always the causative factor. Chang et al. (1997) followed 1053 men in a prospective study to assess the relationship between self-reported sleep disturbance and subsequent clinical depression and psychiatric distress over a median follow-up period of 34 years. The relative risk for depression was greater for those who reported a bad sleep at the start of the follow-up period. Coffee, however, had no influence. In this case, sleep disturbances reflected a vulnerability for depression, since even after resolution of the depressive period, sleep EEG abnormalities remained. It is unlikely that coffee as a mood enhancer and cognitive stimulant has anything to do with a genetic predisposition to vulnerability for bad sleep and depression.
Although these results may shed light on studies reporting impaired sleep quality due to caffeine intake, they may only count for those who use sedative hypnotics, which may hinder a refreshing sleep.
In general, it can be said that coffee drinking is often associated with a cluster of factors that are representative of a stressful and risky lifestyle. It is these factors that might be responsible for certain sleep-wake problems, and not coffee.
Was this article helpful?