[Author's personal comments are in italics:]
I recently attended SLEEP 2018, the Annual meeting of the Associated Professional Sleep Societies. This took place in Baltimore from June 3 - June 6, and my purpose there was to represent the Circadian Sleep Disorders Network.
I tried to network as much as possible. However, the conference schedule was jam packed, so it turned out to be quite hard to network as people rushed from one presentation to another. If I were to attend such a conference again, I would either go to more of the networking events or perhaps staff a booth there.
There were many wonderful talks, so I have detailed only a selection of some of the most intriguing ones.
In this study, 14 participants stayed in a lab for seven days in dim light conditions. After a baseline day, they were randomly divided into a day shift group and a night shift group. Each group was on the shift for three days, followed by a 24 hour constant routine protocol, and then a recovery day. The researchers tried to keep things as consistent as possible by feeding the participants isocaloric snacks and meals. As is typical, the participants in this study were not sleep disordered and were healthy.
Their results found that while cortisol and serotonin rhythm did not change significantly, other metabolites did. Those whose circadian rhythm was most misaligned with their work schedule had the most elevated risk of metabolic disturbances. This was a fairly short and small study, but it was illuminating nonetheless.
This paper looked at sex differences between appetite in male and female night shift workers, and the presentation began with a literature review. From the body of research so far, we know that women are more likely to be larks. They also tend to have earlier peaks in body temperature and melatonin. However, researchers noted that there are contradictory results from various studies on whether shift work increases obesity risk in women more than men.
In this study, 14 participants took part in simulated night shifts over an eight day period and had their leptin and ghrelin levels assessed. Leptin is a hormone that controls satiety, and thus decreases desire for food. Ghrelin is a hormone that controls hunger and thus increases desire for food. The results of this study showed that leptin levels decreased in women and increased in men. Ghrelin levels increased in women, but did not change in men. Participants reported no change in perceived hunger levels. Therefore, the researchers concluded that changes in hormones favoured energy increase in women but not in men.
At the Q&A session, the researchers stated that while all the women in the study were in the follicular phase of their cycles, some of these women were on birth control pills. This difference in hormone levels was not included as a confounding variable however, so I would be very hesitant about interpreting this study as evidence that women's appetite increases when sleep deprived or during circadian mis-alignment. After all, other research has shown that circadian misalignment or sleep deprivation has been shown to decrease appetite in certain women. Since this study has a significant confounding variable that has been ignored, I would be sceptical about the results of this study.
At the beginning of this talk, the researchers spoke about how it's well known that short sleeping impairs glucose tolerance. This effect has been replicated over and over so much that it is almost scientific fact at this point. However, while this phenomenon has been studied repeatedly in normal weight individuals it has not been as well studied in the obese. So McHill and colleagues recruited 17 participants (seven of whom were obese) for a circadian misalignment protocol through use of dim lighting in a lab. All were good sleepers and slept eight hours each night before the study began. The protocol consisted of ten "days" made up of five hours and 20 minutes. Participants were woken up at intervals throughout these "days".
Results showed that the glucose peak occurred circa three hours earlier (7am) in the obese group than in the non-obese group, and that post meal insulin levels were highest in the obese group. Researchers also found that the more hours melatonin remained elevated during wake times, the worse the metabolic outcomes. The latter finding is very relevant to those of us with circadian sleep disorders, as if we are forced to live outside our natural sleep cycle our melatonin levels will be high for a significant portion of the day.
Overall, the researchers concluded that there is a strong link between the circadian system and glucose metabolism.
Dr Eve Van Cauter gave a lively and entertaining speech. She emphasised how important the recent Nobel Prize win was for sleep and circadian medicine research, and stated that it should energise all researchers, doctors, and patients in the field. She also noted that one of the most cited studies in circadian research (Speigel et al., 2004) was originally repeatedly rejected, yet is now considered to be one of the most important papers ever published in sleep medicine. Her message was to keep pushing to get your research out there, no matter what the obstacles. These are wise words to keep in mind when we look to publishing our registry results.
Dr Van Cauter also emphasised that sleep itself is an important synchroniser of the circadian system. This could explain how good sleepers can develop sleep problems when doing shift work, and could also explain how phase-delay chronotherapy can move a person from DSPD to Non-24. In her words, "sleep health may promote circadian health."
She then moved on to talk about the links between sleep deprivation, hunger, and food intake. Research has shown that food reward (the pleasure we get from food) increases when we are sleep deprived. This is because sleep restriction amplifies and extends the peak of circulating cannabinoid receptors in the brain. Thus sleep deprived people tend to eat more and to eat less nutritious food. Perhaps this increased food reward would explain why short sleepers have a massively increased risk of obesity - in the odds of 45% generally.
I spoke to her about this afterwards and she said that while appetite almost always increased in sleep deprived men, a subset of women report nausea and decreased appetite when sleep deprived.
We also know from hundreds of studies that sleep deprived people almost always have an increase in glucose and a decrease in insulin sensitivity. In laymen's terms, this means those with circadian disorders who are sleep deprived have an increased risk of diabetes.
Overall, this was a fascinating talk that emphasised over and over that sleep deprivation has a significant effect on hormones. I also note that normal sleep time in the myriad of studied covered in the talk seemed to be midnight to 8am. Yet the vast majority of people must rise at 6-7am during the week. Overall, it would appear that a significant percentage of the workforce is sleep deprived. The health costs of this are troubling.
This presentation began with an overview of the research on chronotypes and lifestyle factors. Late chronotypes unsurprisingly consume more caffeine, alcohol, and cigarettes. They also tend to have shorter sleep durations. I would hypothesise that these factors all feed into one another. Most late sleepers are forced to rise early due to work or school commitments. Of course they then become sleep deprived, and sleep deprived people need more stimulants to keep functioning.
In Baron's research, she studied those with DSPD who tended to sleep from 3am to 10am. She compared them to a group of average sleepers who slept from midnight to 8am. It was unclear if her late sleepers group were living according to their natural schedule or were forced to rise early.
When compared to the average sleepers, those in the late sleepers group:
Late sleepers were also heavier and more depressed.
Baron conducted a separate study where she recruited as many night owls as possible. In this study she found that timing of dim light melatonin onset was related to fast food intake, particularly in men. She also found that night owls had lower amounts of morning activities. Hardly surprising since most delayed sleepers tend to be fatigued in the morning!
Overall, Baron concluded that circadian misalignment leads to increased amount of insulin resistance and that it limits sleep duration. She also found that it leads to poorer diet and exercise behaviours. Of course, this does not have to be the case. If we lived in a more inclusive society late sleeping would not automatically limit sleep duration.
This session with the esteemed circadian researcher was quite illuminating. Dr Wright was very personable, and he engaged with everyone. Many people posed very technical questions, which were not so relevant to circadian sleep disorders. However, I learned a lot.
Interestingly, Wright stated that it takes about three weeks to recover fully from chronic sleep deprivation. This helps to explain why weekend lie-ins are not enough to feel refreshed during the week. It also helps to illuminate why chronic sleep deprivation is associated with the development of health problems.
At the session there was much debate about whether lying in at weekends was beneficial or not. Some cited the recent Åkerstedt study which suggested people who lie-in at the weekends live longer, while others felt that weekend lie-ins interfere with the circadian rhythm and that one study didn't prove anything. My response to this is that if one needs to sleep in at the weekends they are living outside their circadian rhythm anyway, which is ultimately harmful.
Wright stated that in sleep studies the sleep period simply refers to time spent in bed. So, even if a person takes an hour to fall asleep this is still counted as sleep time. Hence, there may be some truth to the old wife's tale that lying down and resting can benefit the brain even if one is not actually able to sleep.
He also mentioned that melatonin actually increases glucose levels. This could partly be why living outside your circadian rhythm (when melatonin levels are high and you should be sleeping) increases glucose. In fact, glucose has such a strong relationship with sleep that Wright said he could take a person's baseline glucose, take it again five days later, and then predict whether a person has been sleep deprived for that period.
Wright spent some time talking about weight gain and shift workers. He has found that people doing shift work will actually gain weight on the same caloric intake that they maintained on doing daytime hours. For those of us with circadian disorders, this would be akin to working a day shift.
Finally, Wright mentioned a still ongoing study by Céline Vetter at the University of Colorado. Vetter has divided employees into chronotypes and then put them on their body's ideal shift schedule. Her results are bound to be intriguing!
There was no information on the conference booklet about the content of this talk, so I assumed it would be a critique of the lack of sleep disordered patients in research. We simply are not studied enough. However, it actually turned out to be a talk on gender, race, and ethnic diversity in the sleep professional field.
Still, as with all the talks I learned something. The researcher echoed Dr Van Cauter's thoughts on academic rejection and stated that it was normal to have your research repeatedly rejected. The message again was to keep going and to be persistent. Finally, one of the speakers talked about how Twitter had helped her to make connections and to get ahead in her career. She suggested that everyone should consider using it. I believe this information is useful to CSD-N too.
This study looked at the impact of shift duration on medical errors and adverse events in a group of medical residents. Before the policy change, residents had been working an average of 71 hours per week. This was then reduced to 61 hours. After the reduction, the residents made fewer fatal errors (60% reduction) and were involved in fewer motor vehicle accidents. This paper also found that even one long shift increased the risk of adverse events.
This study intrigued me because it was yet another clear example of the dangers of sleep deprivation. Obviously, doctors are tasked with taking care of vulnerable patients, but even non-medical professionals can be dangerous, as demonstrated by the increase in motor vehicle accidents. The researchers did however note that this study was limited by its observational and self-reported nature.
This paper looked at the health costs associated with sleep durations. In this paper, short sleepers were classified as <6 hours; average as 7-8 hours; and long as 9+ hours. Analysis indicated that long sleepers have the highest total health expenditures. I would posit that this is because long sleeping is a symptom rather than an outcome. In other words, many people need to sleep a lot because they are not healthy. It's not that sleeping a lot makes them unhealthy. This research also noted that both short and long sleepers have the highest prescription expenditure and overall health costs. I would assert that this is because many/most short sleepers are not choosing to be short sleepers. Economic circumstance, illness, and sleep disorders are instead shortening their sleeping periods.
This was an absolutely fascinating talk and others in the room appeared to agree. The researchers noted that there is a higher prevalence of poor sleep amongst racial ethnic minorities and those of low socioeconomic status. So they decided to try a low cost yoga intervention to see if it would improve sleep.
Researchers simply held a series of yoga classes, including one class on home-based yoga. They also held one to three sessions on sleep hygiene. These classes took place within the low income housing unit itself over a six week period. Results indicated that an astonishing 80% of participants had improvements in sleep.
While yoga won't cure our circadian disorders it seems to me that it could be useful in helping sleep quality. Might be worth a try for some!
This was another interesting and lively presentation. In this study, researchers analysed sleep outcomes in adolescents before and after a 27 minute delay in school start times. Dr Sharkey mentioned that before the delay, the students were essentially on a rotating shift schedule as their weekday and weekend sleep times were so different. The original school run time was 7.33am to 1.59pm, and the experimental time was 8am to 2.25pm.
292 students completed the baseline questionnaire and 371 students completed it after the new start time. 127 students in total completed both.
Results showed that there was no significant change in the adolescents' bedtimes during the week, but that their wake times were delayed by 28 minutes, which was a nearly identical amount to the actual delay. Given that it's estimated 70% of teenagers suffer with a temporary form of DSPD, it makes complete sense that their bedtime would not change and that instead their wake time would be later.
Rather tellingly, those students in favour of the change had longer sleep latencies and got more sleep than those against it. It was unclear to me why non-delayed students were suddenly getting less sleep with a later school start time, but perhaps this was due to the timing of extracurricular activities being pursued later.
Dr Sharkey noted that this study was conducted in the town she lives in and that she personally had to deal with the fallout. Apparently, certain school staffers were not happy with the change. She reflected that dislike of change is a common human problem in general!
This study looked at the association between insulin resistance/glucose metabolism, circadian delay, and forced early morning awakening in a group of obese adolescents. These female adolescents were aged 14-19 and attended the same high school. Salivary melatonin levels were assessed throughout the day and serum glucose and insulin levels were also collected.
There were several illumining results:
This study started with an overview of the current literature on the links between shift work, food intake and appetite. The researchers noted that rotating shifts particularly seem to have an impact on food intake. Overall, shift work leads to extended periods of wakefulness, increased meal opportunities, and increased energy intake compared to day-shift work.
Therefore, the hypothesis for this study was that extended wakefulness would increase energy intake in a group of shift workers. Inclusion criteria specified that the workers must not have any existing sleep disorders.
This study looked at rotating shift workers on the following schedule:
Participants filled in a food recall questionnaire and results were adjusted for age, BMI, and physical activity. Sleep duration was recorded by an actigraph. Analysis showed that the shift schedule affected sleep duration, energy intake, and eating duration. The free day between the afternoon and the night shift showed the greatest increase in energy intake, as well as the shortest sleeping period, and the longest eating duration. The researchers also noted that carbohydrate and fat intake were highest during their 24 hour free period. They hypothesised that would be from fatigue from the night shift and perhaps familial activities. Interestingly, sleep duration was not associated with energy intake.
This study once again showed that shift work greatly impacts food intake.
This intriguing eight week study examined whether cautious use of Zolpidem would reduce suicidal ideation in persons with insomnia and depression. 103 people with depression, insomnia, and suicidal ideation were recruited for this double blind placebo controlled trial. Half the group received Zolpidem, while the other half received a placebo. All participants in the trial were also on Fluoxetine. The researchers wanted to investigate whether phonotype or seasonality played a role in suicidal ideation in this population.
This was a novel study in that the researchers recruited people who were moderately suicidal, yet were also out-patients.
All participants filled in questionnaires and wore actigraphs throughout the study periods. Results showed that delayed sleep timing was associated with greater suicidality. Eveningness was also associated with increased suicidality prior to treatment, as was total sleep time. The researchers did indeed point out that the actigraphs may reflect sedentary behaviours common in depression rather than sleep time itself. Even after adjusting for depression and insomnia, eveningness was still related to suicidal ideation, however.
I approached the researcher afterwards to ask how delayed the group of delayed sleepers were. She informed me that they did not meet criteria for DSPD. However, personally I feel that even mild delays or "night owls" suffer significantly in this 9-5 society. They are still living outside their chronotype. Perhaps living within their chronotype could decrease suicidal ideation? This was however beyond the bounds of the study.
This study re-examined the results of a previous trial that looked at the effect of antidepressants plus cognitive behavioural therapy for insomnia (CBT-I) in people with major depressive disorder and insomnia. For this study, the researchers looked at whether chronotype affected the improvement in depression and insomnia symptoms during the study.
139 people were randomised to receive either antidepressants and CBT-I or antidepressants and control therapy over a period of 16 weeks. Symptoms of depression were assessed via the Hamilton Depression Rating Scale (HDRS).
Chronotype appeared to significantly impact the outcomes in this study. Evening types appeared to benefit more from CBT-I treatment than other chronotypes. However, greater eveningness was still associated with worse sleep quality, shorter sleep duration, non-refreshing sleep on working days, more depressive symptoms, and less depression remission.
An important point to note is that the researchers concluded that evening types had poorer responses to antidepressants, but better responses to CBT-I for insomnia. It was unclear to me if this was because CBT-I improved their sleep duration/sleep quality, or if it was something else altogether. Since insomnia symptom improvement did not differ from chronotype to chronotype, I would wonder why the CBT-I helped so much.
This study examined the link between suicide and sleep. We know that suicide is the leading cause of death in people aged 15-24, and this study examined whether sleep would impact suicidal ideation. Students completed questionnaires which assessed how many days per week they received sufficient sleep. Only 11% of the entire sample reported insufficient sleep on 0-1 days. In other words, 89% of the sample were spending at least two days a week sleep deprived. Considering that most of the people in this group are still growing, this is disturbing.
The results indicated that sleep difficulty and insomnia were all significantly associated with suicidal ideation, even when people were not depressed. They were no differences in findings between athletes and non-athletes.
This study looked at depression risk in good sleepers who developed either acute or chronic insomnia. It is well established that there is a strong link between insomnia and depression. However, these researchers wanted to look at depression incidence and severity in those who developed either transient or chronic insomnia. 1069 people participated in the study, which is quite a large group for any study. They were followed for a year. Over the year, 75% remained good sleepers, 25% developed acute insomnia, and 6% developed chronic insomnia.
Results indicated that poor sleepers had significantly more depression. Depression severity was also marginally higher in the groups affected by insomnia, especially in those with insomnia for the first time.
I tried to attend as many talks that were related to chronotypes, sleep deprivation, and circadian disorders as possible. These are issues we regularly deal with, so some themes related to them came up repeatedly over and over.
1) The epidemic of sleep deprivation due to work and school start times
If researchers, doctors, and governments really want to work on sleep health, they need to start with the work/school environment. Throughout the conference, researchers regularly mentioned that normal/average sleepers tend to sleep between midnight and 8am, and that only about 10% of people would choose to rise before 7am. They also talked about the epidemic of sleep deprivation. Yet start times are changing only in schools. It seems nobody is connecting the dots and realising that work start times also need to change. There need to be large policy changes everywhere.
Perhaps it's easier to work with school districts on changing school start times (although Dr Sharkey mentioned how difficult this actually was). However, aside from one on-going study by Dr Vetter at the University of Colorado nobody mentioned the idea of changing work start times.
Many people with young children are sleep deprived due to their children's sleep schedules, but the other main cause of sleep deprivation is work start times. Few people naturally rise at 6am or 7am. Yet occupational demands mean they must do this five days a week. Thus the epidemic of sleep deprivation. As a result, sleep deprivation is normalised and even glorified in western culture.
This effects those with circadian sleep disorders, as people conflate their having to rise an hour earlier or go to bed earlier with circadian sleep disorders. Clearly they are not the same, but this acceptance of circadian disturbance and chronic sleep deprivation impacts the way those with circadian disorders are perceived.
If most people in society were allowed to sleep to their natural rhythm and were rarely sleep deprived, there might be more outrage at the idea that there is a subset of people who are forced to be massively sleep deprived every day for decades.
There has been decades of research on the link between glucose metabolism and sleep deprivation/disturbance. In fact, we now know that almost everyone's blood glucose levels rise when sleep deprived. Glucose intolerance has also been linked with weight gain, as has sleep deprivation. Combined with other risk factors such as poor diet, family history, and chronic stress it's easy to see how both diabetes and weight gain have become epidemics in the western world. Interestingly, although sleep deprivation increases appetite in almost everyone, there are outliers in whom sleep deprivation decreases appetite. These outliers are almost all female.
Overall, the conference was highly educational and inspiring. There are just so many dedicated and caring sleep doctors and researchers out there. However, I came away feeling slightly frustrated. It seems that we know so much about the importance of sleep, yet little is being done to ensure people actually get enough rest. Unless there are policy changes (particularly in the workplace), this knowledge is fairly useless. It's why non-profits like ours are so important. As a group, we can advocate for change. It may take some time, but I believe we can work together and make a real difference.