Last year, I started a show dedicated to parents and entrepreneurs. There were already hundreds of shows out there for parents, and for.
New shows hit iTunes every day. At this point, everyone is starting a podcast, or talking about starting one. The good news is that listenership is still increasing, and people are on the lookout for new shows. Overall, it's still a good time to create a show, and for more reasons than you might think.
Last year, I started a show dedicated to parents and entrepreneurs. There were already hundreds of shows out there for parents, and for startups, but not as many shows specifically for entrepreneurial parents. Today, nearly one year later, I've recorded more than one hundred episodes of the show. While it's not a simple project--podcasts take time, energy, and investment--it can be a worthwhile. Here's how starting a podcast might boost your career:
Speaking and distilling your ideas into clear, concise, and well-told nuggets is not easy. Just like stand-up comedians practice their sets over and over again--Chris Rock and Ali Wong come to mind--public speaking is a skill that's honed over time.
Weekly podcasting has helped me to get focused, clear about my ideas, and articulate in my questions. Talking through ideas is great prep work for anyone who wants to do panels, public speaking, or press interviews.
Many projects have a long turnaround time, which can take a while to get feedback. A book, for example, can be a multiple-year project. A podcast, in contrast, can be relatively quick to ship and get feedback on. Each week after releasing an episode, I get emails in response and learn, in real-time, from fellow parents and business owners about what's resonating.
I was not prepared for how much I would dive into the audio format of storytelling and geek out about music, sound editing, and story construction. It turns out we listen very differently than we read, and repetition, pauses, intonation, and break points are all important rhythmic elements to a show. Listening to other great storytelling shows (like This American Life, or Serial) teaches me just how beautiful a story arc can be.
Getting a great story from someone isn't always easy. You have to establish trust, navigate the conversation, allow space for people to go deeper, and ask the right questions.
Tim Ferriss has a list of favorite questions in his book, Tribe of Mentors, and Michael Bungay Stanier has seven great questions in The Coaching Habit. Krista Tippett, the host of the On Being podcast, is also a phenomenal question-asker. Learning to ask better questions is a life skill that helps you connect with others, hire great people, and dig into more meaningful conversations.
"What's the greatest trick you've used to become a better interviewer?", I asked my friend Srinivas Rao.
"Close all your tabs and applications and turn your computer dark," he said immediately. "You have to just listen."
I wondered if he took notes while interviewing.
"Nope," he said. "Just get better at listening."
Over hundreds of hours I've learned how to slow down, be more in the moment, and really listen to what the person on the other end of the microphone is saying. It takes a lot more energy and focus than, say, scrolling social media on my phone, and it's worth it.
When you start a series of conversations around a specific topic, it's likely you'll build a community of people interested in joining the conversation. Beyond your audience, however, are your peers--podcasting has opened up a fellow network of audio geeks learning about the craft alongside me.
One of the more surprising outcomes of the podcast is the network effect of connecting with so many wise individuals. If I had reached out and asked many of these folks for an hour of their time to talk, many likely would have declined.
But as the host of a show that reaches a wider audience, people are more willing to collaborate and share their ideas with me in service to the listeners. Having long-form, in depth conversations with brilliant people has been one of the highlights of my last year. By having a show, I've been able to connect to more people than I otherwise would have.
If yes, consider starting a podcast. It's a great sandbox for learning new ideas and practicing your skills. Don't think of a podcast just as a marketing tool or a growth engine (of course, it likely will be both)--it's also a tool you can use to improve yourself.
The opinions expressed here by Inc.com columnists are their own, not those of Inc.com.
BODY & SPIRIT comes to dance, just as you can learn to be a better swimmer Books, 1998), the Dalai Lama speaks of a regret from his own life: "It's still there. There are so many supplement products to choose from — where do you start?.
Despite the well-established natural shifts to later wake and sleep times that occur in adolescence, most schools retain early start times. Currently school starting times are not adjusted for the shift to later wake and sleep times that occur naturally in adolescence. This mismatch between adolescent biology and the conventional practice of starting school early leads to a systematic reduction in the amount of time available for sleep to teenagers and consequently chronic sleep deficiency. The Centers for Disease Control and Prevention and the American Academy of Pediatrics (2014) have stated that early school starts are associated with increased health risks of obesity, depression, and drug use as well as poorer academic performance (Owens et al., 2014; Wheaton et al., 2015). Their recommendation that middle and high schools should open no earlier than 8:30 a.m. is now supported by the American Medical Association (2016). Almost all studies to date, while scheduling school starting times later than before, have retained a starting time at or earlier than 9:00 a.m. (Kirkby et al., 2011). One study in New Zealand did report improved sleep in 17 to 18-year-old students when the starting time was moved from 9:00 to 10:30 a.m., as compared to younger controls who remained at 9:00 a.m. (Borlase et al., 2013). A recent study of optimal times for cognitive performance for students aged 18–19 concluded that much later times were optimal, specifically after 11 a.m. or 12 noon (Evans et al., 2017).
Evidence for increased health risks associated with early school starting times is both substantial and demonstrated through a variety of research methodologies (Hansen et al., 2005; Millman, 2005; Basch et al., 2014; de Souza and Hidalgo, 2014). The underlying biological drivers are also well established. Adolescents need anywhere from 8 to 10 h of sleep per night for full health and academic performance depending on age and inter-individual differences, yet most get far less (Iglowstein et al., 2003; Foster et al., 2013; Hirshkowitz et al., 2015). Biological changes in the timing of the 24-h circadian clock during adolescence delay the onset of wake and sleep times, and this shift does not reverse until early adulthood (Roenneberg et al., 2004). Additionally, homeostatic regulation of pressure to sleep builds more slowly, taking a longer time to reach the critical threshold required to initiate sleep (Carskadon, 2011). Adolescent sleep restriction is clearly linked to early school starts as on non-school days adolescents have wake times two or more hours later (Roenneberg et al., 2007), a finding seemingly not substantially affected by cultural factors (Gradisar et al., 2011; Foster et al., 2013).
Later school starting times provide benefits to adolescent sleep, health and learning (Curcio et al., 2006; Carskadon, 2011; Lufi et al., 2011). There is substantial evidence that later starting times benefit academic outcomes even in the early stages of puberty and this positive impact continues into late adolescence (Carrell et al., 2011; Edwards, 2012; Meltzer et al., 2014). Later starting times are associated with improved sleep that continues into the years following implementation (Borlase et al., 2013; Wahlstrom et al., 2014). Later starting times also reduce the rate of student driving accidents (Danner and Phillips, 2008), and lower reports of depression (Kirkby et al., 2011).
The principle that school start times for adolescents should be later than currently the norm in American schools (about 8:00 a.m.) is now widely accepted. Research is now needed into synchronizing school starting times more closely with adolescent biology, taking into account the increasing impact of circadian rhythm changes in adolescence (Shekleton et al., 2013; Kelley et al., 2015). While the evidence for starting middle and high schools at 8:30 a.m. or later is positive, this recommendation is based on evidence limited to studies where school times are shifted to no later than 9:00 a.m., or often earlier, leaving unanswered the question of how late school starting times should be. To explore the impact of a much later school start time, we examined the impact of a 10 a.m. starting time (representing a 1:10 h delay from an 8:50 a.m. starting time), on rates of illness of 13 to 16-year-old students and academic performance of 14 to 16-year-old students.
In light of the then-existing evidence on start times, the lead investigator of the present study (PK) designed and directed a field experimental to implement and assess the efficacy of a much later school start time. The design is summarized in Table 1, columns 1, 2, and 4 (see section Results). The school was a mixed sex, state-funded school's for 13 to 18-year-old students. At the start of the study, the school student performance was considerably below the national mean. The school was in an urban area of 0.7 million, in a region of England where achievement was lower than national average.
Table 1. Effect of experimental treatment (10 a.m. school start time) on absences due to illness.
The school had no additional funding, selective entry or other educational interventions during the duration of the study. The school operated within The Innovation Trust, a charitable trust to improve schools in partnership with leading scientists (see section Acknowledgments). Working with this group of experts, a later start time was considered with a focus on determining a starting time for adolescents aged 13–18. Initial testing of student academic performance at 10:00 and 14:00, voluntary chronotype surveys and body fat levels of students led to the conclusion there should be a core time period for all students between 11:00 and 15:00, with additional learning time before (for morning/average chronotypes) and after (for evening chronotypes) (Sussman, 2009). This appeared too radical a change and, combined with the then current National Health Service Trust judgement that schools could not legally hold any mental health data on students 16 or under including sleep and body fat data, led to limiting the age range of 13–16 and using only existing national measures of health and academic performance. Therefore analysis of the relevant sleep and circadian neuroscience research indicating 10:00–10:30 a.m. starting times would be appropriate at age 16 (Kelley et al., 2015). This process led to the final choice of 10:00 a.m. as the school start time, a 1 h 10 min delay compared to the traditional 8:50 a.m. start time then in use.
The 2-year process for changing start times was ethically approved by both The Innovation Trust, a charitable educational research trust linked to the school, and the school's Governing Body on the basis that students would not be subject to additional surveys, tests or measures other than a later starting time. A legally-required consultation with parents, students, teachers and others was conducted. This raised a number of concerns including transport to school. Unlike American schools, urban schools in England generally do not provide transport for all students, and the existing transport infrastructure made this a relatively minor issue. The school already supported parental work hours and transport by opening 1 h early and 1 h after school (this included sports and club opportunities). Therefore, with minor changes, all the times in the school were simply moved an hour and 10 min later. Although parents supported the changes, there was resistance from the external local educational administrators, a phenomenon noted in other studies (Wahlstrom et al., 2014). The concerns raised during the consultation, while often valid, were addressed as possible, but considered secondary to the raison d'etre for schools—to help healthy children realize their full academic potential.
After approval of the change, a 6-week trial run with a 10:00 a.m. start was conducted in 2010 after all annual examinations were completed and students taking examinations had left, but before the start of the new academic year. The trial moved all related scheduled events to match the 10:00 a.m. start including transport to and from school, family arrangements, and school clubs and activities. There were also some additional and unexpected benefits reported: The travel times avoided rush hour and were considered safer; and some staff could take their children to their primary school in the morning.
After 2 years with the 10 a.m. start time, a natural experiment was created by a change in local education administrators that reinstated the early start as used in all local schools. Consequently, the school starting time for 13 to 16-year olds was changed back to 8:50 a.m. due to these policy changes, not part of the study design; however we took advantage of this natural experiment.
This observational study used a before-after-before (A-B-A) design. School start times changed as follows over 4 years: Year 0 had an 8:50 a.m. start; Years 1 and 2 had a 10:00 a.m. start; and Year 3 had an 8:50 a.m. start. Over these four academic years (September 2010–August 2014) national data and school data on illness and performance of 2,049 students aged 13–16 years were collected.
English education data are collected nationally for each school and these data for all schools are published by the UK Office of National Statistics (2016). Two variables were used to assess the impact of the time change: Student absence due to illness, and overall student academic performance. English schools are required to distinguish absence (not in school for any reason) from absence due to illness (which accounts for ~60% of all absences, varying by year). Absence due to illness is recorded for each student for every morning and afternoon session, and so the measure of illness utilized here was the number of school sessions lost through illness per student per year. This measure gives a more precise measure of illness than raw absences rates (Department for Education, 2016).
At that time, the government's key measure of student academic performance was the General Certificate of Secondary Education (GCSE) examinations taken at the end of compulsory schooling at age 16. Good academic progress at 16 was defined as achieving five or more GCSE grades of C or above in English, Mathematics and at least three other subjects. The measure of school performance was the percentage of students achieving good academic progress.
In addition, a national system of value-added analysis of individual students and school potential achievement in GCSE grades and good academic progress was produced by the Fisher Family Trust for all students and schools in England. As the prior ability of student cohorts can vary year-to-year, so can the effectiveness of schools, and thus a second analysis of performance based on these value-added predictions was undertaken (Visscher and Coe, 2003; Koedel et al., 2015). The predicted results were based on the students' previous achievement in national tests. Schools exceeding predicted percentage good academic progress were deemed to have added value, given as a percentage good academic progress above the predicted outcomes (with negative results for schools below predictions). For example, if the students achieved an average of 4% higher good academic progress, it would have a value-added score of +4%. These two national measures of actual and predicted school performance were used in our analyses.
National data were used to analyse illness and performance. National data were used where available to estimate standard deviations. In addition, a data set of 2,880 similar schools was constructed where “similar” was defined as state-funded school with 13 to 16-year-old students in cohorts >100, and reporting examination data for the period 2010–2014. The data sets used are school-level data published by the UK's Office of National Statistics and other sources as indicated (UK Office of National Statistics, 2016). The achievement data have limitations due to policy and publication practices. Policy limitations include a decision to remove schools becoming academies (similar to Charter schools) from national data, changes in examinations, and in the data reported during the 4 years.
Analysis of illness and performance data during this period led to reporting both raw scores and scores adjusted to account for national variation. For national trends in illness, performance, T-tests were used to assess significance, and Cohen's d and h for effect size, taking into account Hattie's critique of educational measures to determine impact of interventions (Hattie, 2008). Hattie's critique of educational research based on synthesizing over 800 meta-analyses on raising achievement raised questions about educational data analyses and significance. Specifically, he argued that more sophisticated statistical techniques, large numbers of subjects usually possible within education research, and other factors meant effect sizes should be >0.5 in individual studies. Hattie lists 138 effect sizes found in meta-analyses of the significant educational interventions of the last 20 years, showing the highest 20 interventions had effect sizes ranging from 0.61 to 1.44. Hattie's concerns about educational research and effect sizes are shared by others (Snow, 2015; Churches et al., 2017). For illustrative purposes, the cost/ benefits of educational reforms in England and New York to increase educational performance was also analyzed.
Rates of absence due to illness of students aged 13–16 were lower with the 10 a.m. start time (see Table 1 and Figure 1).
Figure 1. Absences due to illness, differences from national average.
In Year 0, for the baseline cohort of students aged 13–16 which had experienced 8:50 start times for both years of their course in the target school, the mean number of absences due to illness was 15.4, compared to 11.5 nationally (Year 0 “Field Control Group” in Table 1, compare column 6 for the target school to column 5 for the national benchmark). That is a difference of 3.9 absences which is statistically significant at p < 0.0005 (column 7). Moreover, it is large: The rate in the target school was 34% higher than in the national benchmark (column 8).
The situation changed sharply for Year 1 (“Field Treatment group #1” in Table 1), the first year with the experimental 10 a.m. start treatment. The mean number of absences in the target school dropped from 15.4 to 11.3, a drop of 4.1 absences (column 6, compare Year 0 and Year 1). The gap between the target school absence rate and the national benchmark shrank by 3.3 absences to a mere 0.6 absences (Year 1, column 7). This shrinkage amount to a 31% decline in the gap between the target school and the national benchmark (Year 1, column 10), a decline that is statistically significant at p < 0.0005. Cohen's d for the change is 0.58, making the effect size in the medium-to-large range. This reduced absence rate left the target school a mere 0.6 absences above the national benchmark, a difference so small that it is not statistically significant (p > 0.05, column 7).
The crux of the field experiment is Year 2, when the 10 a.m. start was operating for its second full year (the target school's GCSE courses last 2 years with examinations throughout). Absences dropped further in the target school, from a mean of 11.3 to a mean of 7.9 (column 6), a drop of 3.4 absences. This brought the target school's absence rate significantly below the national benchmark (column 7). This shift from slightly above the national benchmark in Year 1 to significantly below the national benchmark in Year 2 has a Cohen's d of 0.48, a medium effect size.
The most crucial comparison is between Year 2 when the 10 a.m. start was operating for its second full year and Year 0 in which students had been on the 8:50 start for at least 2 years. The mean number of absences fell by fully 7.5 (Column 6, Year 2 vs. Year 0). The gap between the target school rate and the national benchmark shrank by 55%, amounting to a shrinkage of 4.8 absences. This change is statistically significance at p < 0.0005. The Cohen's d for the change is 1.07, a large effect size.
Thus, the field experiment demonstrates in multiple ways that the experimental treatment is associated with a decline in absences that is large and statistically significant not only relative to the control group but also relative to the national benchmark. The pattern is evident in Figure 1.
Next comes the natural experiment when the start time in the target school returned to 8:50 in Year 3 due to a change in policy which shifted control of the target school to local education authorities who imposed a uniform early start time on all comparable schools. The mean number of absences rebounded to 11.2, well above the level in the previous year (column 6). This was 20% above the national benchmark, a clear contrast with the previous year's 10% below the benchmark (column 8). Adjusting for yearly changes in the national rate of absences, the gap between the target school and the nation inflated to 2.8 absences a year, a 30% increase (columns 11 and 12). This change is statistically significant at p < 0.0005 and has a Cohen's d of 0.47, a medium effect size.
All in all, for absences the field experiment demonstrates that the change to a 10 a.m. start reduced absences due to illness by over 50 percent compared to national rates, a large and statistically significant decline. The natural experiment demonstrates that reverting to the 8:50 start induced, already in its first year, a medium sized, statistically significant increase in absences. These data are consistent with a dose-dependent response to a 10:00 start.
In the target school's Year 0 (“Field Control Group” in Table 2) the baseline cohort of students which had experienced 8:50 start times for both years of their course, the percentage who successfully completed the GCSE examinations was 34, compared to 56.2 nationally (columns 5 and 6). That is a vast difference of 22 percentage points, statistically significant at p < 0.0005 (column 7). This difference amounts to 40% of the national baseline (column 8), marking the target school as having a very low success rate. In value added terms, the actual percentage of successful students was 5 percentage points lower than the FFT estimate for the school based on the cohort's past performance (column 9), showing a negative value-added value. All in all, in Year 0 in the target school, the performance picture was grim.
Table 2. Effect of experimental treatment (10 a.m. school start time) on academic success.
The situation changed sharply for Year 1, the first year with the experimental 10 a.m. start treatment. The actual percent of academically successful students shot up by 19 percentage points to 53 (column 6, compare Year 0 and Year 1). The gap between the target school and in national benchmark in Year 1 is so small that it is not statistically significant (Year 1, column 7). Between Year 0 and Year 1, the gap between the target school success rate and the national benchmark shrank from 22 to 5 (Year 1, column 7), a shrinkage of 17 percentage points that is statistically significant at p < 0.0005. The value added by the school has also risen: The actual value added by the school is 4 percentage points more than the FFT prediction, up 9 percentage points from Year 0. This amounts to 15% of the national baseline and is statistically significant at p < 0.0005.
The heart of the field experiment is the comparison of Year 2 vs. Year 0. Here again there is extensive evidence for the positive impact of the 10 a.m. start on student success. Student success in Year 2 is 52% compared to 34% for Year 0 (column 6), a gain which is significant at p = 0.001 and of medium size, with a Cohen's H of 0.37. Between Year 0 and Year 2, the gap between the target school success rate and the national benchmark shrank from 22 to 7 (Year 2, column 7). The actual student success in Year 2 exceeds the FFT prediction by 7 percentage points (column 10). Indeed, there is a 12-percentage point gain in the value added by the school, a gain that is statistically significant at p < 0.0005 (column 13) and amounts to 20% of the national benchmark.
Thus, the field experiment demonstrates in multiple ways that the experimental treatment is associated with a statistically significant and substantial gain in performance not only relative to the control group but also relative to the national benchmark.
Then, in the natural experiment in Year 3, the school starting time reverted from 10 a.m. to 8:50 a.m. The student success rate fell slightly to 51 (Year 3, column 6) in comparison to a national benchmark of 60, giving the target school a deficit of 9 percentage points which is statistically significant at p = 0.014 (column 7). The value added by the school has shrunk by 4 percentage points which is 7% of the national baseline (column 14).
Thus, even though these students had the first year of their 2-year course with the 10 a.m. start, reverting to the 8:50 a.m. start for their second year is associated with a reduction in success relative to the national baseline and with a reduction in the value added by the school.
The whole pattern for the combined results of the field experiment and the natural experiment is clear in Figure 2.
Figure 2. Academic success: value added by target school.
In Year 0, student success in the target school was actually 5 percentage points below what would have been expected in light of the students' prior performance (FFT). Then the field experiment began. The introduction of the 10 a.m. start halfway through the Year 1 cohort's course lifted the school's value added to 9 percentage points above the expected success percent. There was a further small rise to a value added of 12 percentage points in Year 2 for the cohort which had experienced the 10 a.m. start for both years of their course. The field experiment then ended. The natural experiment then reverted the start time to 8:50, so students in the Year 3 cohort had the first year of their course with the 10 a.m. start and the second year of their course with the 8:50 start. Value added was still positive for this cohort, but had fallen by 4 percentage points compared to the previous cohort which had 2 years of 10 a.m. starts.
Based on this study, moving school start times later so that they are better aligned with adolescent sleep and chronotype patterns is practical and beneficial. Following a change to a 10:00 a.m. school start time, rates of absence due to illness in students aged 13–16 reduced, and academic performance of students aged 14–16 significantly improved. When the school start time was returned to 8:50 a.m., these benefits were abolished. These findings suggest that the general policy recommendation to start high schools no earlier than 08:30 a.m., while helpful, should not be taken as justification to exclude consideration of much later starting times. As the 8:50 a.m. would already be considered appropriate using the 8:30 a.m. recommendation, this study shows further improvements can be made when starting times are much later, in this case 10:00 a.m.
The absence due to illness rate data are consistent with a direct benefit of later school starting times on student health. The differentiation in recording absence and absence due to illness in England offers a more precise measure of illness and a large national data set. Following initiation of the 10:00 a.m. start, illnesses decreased in the second year by over 50%. With the return of an 8:50 a.m. these benefits were reversed, with illness increasing by of 30%. Academic performance of students aged 14–16 also improved with a 10:00 a.m. start. Having 2 years of later start times was particularly beneficial; showing a 12-percentage point gain in the value added that is statistically significant at p < 0.0005 and amounts to 20% of the national benchmark. There is no reason to believe that these outcomes reflected motivation changes. The pupils were studying for their final national exams, which determine their eligibility for continued study and ultimately college or university, or their competitiveness in the jobs market. These were not study-related tests but the real, once-in-lifetime exams that have a major influence on the children's futures and thus would be equally motivated. While there are several possible explanations for the poor performance of students in similar low socioeconomic status areas, the possible impact of sleep restriction linked to early school starts has rarely been considered.
One of the objections raised to changing school start times is that the change is impractical and cannot overcome other barriers, such as bus timetables or sports program scheduling. While the English legal framework makes changing to much later start times a formal process any school can undertake (and therefore more practical than in some other countries), moving all of this school's schedules later produced no practical difficulties. By choosing a 10 a.m. start time, the school aimed to provide benefits for the largest proportion of children possible, given the inter-individual range in phase delays that children experience means some substantially delayed children would still be waking too early in their circadian cycle (Iglowstein et al., 2003; Kelley et al., 2015). Even later start times might address this issue, but 10 a.m. was considered a reasonable compromise between maximizing biological benefits for most children while remaining practical. The national US recommendation that middle and high schools should start after 8:30 a.m. is a clearly justified positive step, although the evidence in this study suggests a much later start of 10 a.m.—even when replacing a school start later than the 8:30 a.m. recommendation—brings additional benefits. A recent study of university students aged 18–19 found that later starting times (after 11 a.m. or 12 noon) were optimal and much later than an 8:30 a.m. start recommended for High School students of 17–18. The study also found that using a fixed time for all students would disadvantage one or more chronotypes, and evening chronotypes in particular if starts were early (Kelley et al., 2015; Evans et al., 2017). Thus, even with the benefits accrued using a 10 a.m. wake time for all students, this approach does not address the wider variation of wake times in different chronotypes.
The limitations in this study include small sample sizes in some measures, a focus mainly on illness, and an inability to measure students' sleep duration. Although other studies have shown sleep improvements following less substantial interventions, some have tracked sleep improvements over 4 years and found that improvements persist (Borlase et al., 2013; Wahlstrom et al., 2014). The English location and measures of academic performance are difficult to contextualize with previous research using U.S. state-wide testing and graduation data. The nationally available data on illness on a school-by-school basis, focus on students in mid-adolescence and the starting time of 10:00 a.m. are, we believe, unique in this research field. A larger, more detailed study is needed of school starts after 9:00 a.m., including those that are later than 10:00 a.m. There is preliminary evidence that such changes can have benefits, particularly for older students (Carrell et al., 2011).
The school-level improvements in performance in this study, if more widely replicated, should be examined from economic and educational perspectives. For example, expenditure to reduce the English attainment gap between rich and poor students reached more than one billion pounds between 2010 and 2015, and yet had relatively little impact compared to the gains made following the 10 a.m. school start time change in this study. The change to smaller schools in key U.S. cities such as New York City have been thoroughly researched after additional expenditure of over a billion dollars to improve performance (Schneider et al., 2007; Stiefel et al., 2015). Other educational policy changes including creating new kinds of school such as Charter, Academy, and Science, Technology, Engineering and Math (STEM) schools, increasing the duration of school times, curriculum and test changes, or No Child Left Behind also have very high expenditure but minimal impact and little scientific rigor in evaluating their effects. In contrast, changing to later start times is a very cost-effective intervention to raise educational standards with substantial scientific backing (Jacob and Rockoff, 2011; Snow, 2015; Hafner et al., 2017).
The key finding in this study is reduced absences due to illness by over 50% compared to national rates (Table 1, columns 12 and 13). This huge change is both practically important and highly significant statistically (p < 0.0005 and Cohen's d for the change is 1.07, a large effect size). The broader impact of later starts on specific aspects of adolescent health, such as sleep duration and quality, mental health, and social development were not assessed, although other studies have shown potential impacts (de Souza and Hidalgo, 2014; Meltzer et al., 2014; Minges and Redeker, 2016). Additional research into much later starts should measure both actual sleep patterns and optimal performance times for individual students. The most important area for further research may be the impact of later starts on areas of social behavior development and mental health. For example, the daily sleep loss of two or more hours per day imposed by early school starts (which cannot be recovered with 10 or more extra hours of sleep at weekends), may put those with a genetic predisposition to a mental illness at greater risk given that direct links between sleep of <6 h and gene expression have been established (Möller-Levet et al., 2013). Sleep deprivation is also associated with adolescents being less perceptive readers of human emotions (van der Helm et al., 2010; Guadagni et al., 2014), during a period of greater sensitivity to sociocultural signals (Blakemore and Mills, 2014) and related brain developments in adolescence. These interrelated factors of significant sleep deprivation, genetic predisposition, the high prevalence of the onset of mental illness during adolescence for a range of disorders (Schmitt et al., 2014) and less a perceptive reading of sociocultural signals, may impact on levels of mental illness and emotional disorders in adolescence (Wulff et al., 2010, 2012).
Using a research-based approach to determine a school starting time for 13 to 16-year-old students led to the implementation of a 10 a.m. school starting time. This later starting time had a substantial benefit for rates of illness and academic performance. A research-based approach to school starting times is clearly replicable in different contexts, cultures, and countries. More importantly, a post-9:00 a.m. school start strategy is one with few costs and many potential benefits which start to accrue, quite literally, overnight. Application of sleep research in this way demonstrates the powerful impact on society and individuals of making evidence-based policy changes.
The Innovation Trust and School Governing Body approved the change in starting times in compliance with English Educational Law. Parents and students were consulted during the change as legally required.
PK, SL, and JK: study design; JK and ME: data analysis; PK, SL, JK, and ME: writing and critical review of manuscript.
This work was funded by Science + Technology in Learning (Award number 1/2017).
SL has had a number of commercial interests in the last 12 months (2016–17). None are directly related to the research or topic reported in this paper but, in the interests of full disclosure, are outlined below. SL has received consulting fees from the Atlanta Falcons, Atlanta Hawks, BHP Billiton and Slingshot Insights; has current consulting contracts with Akili Interactive; Consumer Sleep Solutions; Delos Living LLC; Environmental Light Sciences LLC; Headwaters Inc.; Hintsa Performance AG; Light Cognitive; Mental Workout; OpTerra Energy Services Inc.; Pegasus Capital Advisors LP; PlanLED; and Wyle Integrated Science and Engineering; has received unrestricted equipment gifts from Biological Illuminations LLC, Bionetics Corporation and F. Lux Software LLC; royalties from Oxford University Press; and has served as a paid expert in legal proceedings related to light, sleep and health. He holds a patent through Harvard University and Brigham and Women's Hospital for “Systems and methods for determining and/or controlling sleep quality.” He is a Program Leader for the CRC for Alertness, Safety and Productivity, Australia.
The other authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors would like to thank Professors David Reynolds, Russell Foster, and Daniel Muijs for their contribution to the thinking behind changing the start time based on sleep research, and David Reynolds contributed in its implementation as Chair of The Innovation Trust.
American Academy of Pediatrics (2014). Adolescent Sleep Working Group: school start times for adolescents. Pediatrics 134, 642–649. doi: 10.1542/peds.2014-1697
CrossRef Full Text
Basch, C. E., Basch, C. H., Ruggles, K. V., and Rajan, S. (2014). Prevalence of sleep duration on an average school night among 4 nationally representative successive samples of American high school students, 2007–2013. Prev. Chronic Dis. 11:E216. doi: 10.5888/pcd11.140383
PubMed Abstract | CrossRef Full Text | Google Scholar
Blakemore, S. J., and Mills, K. L. (2014). Is adolescence a sensitive period for sociocultural processing?. Annu. Rev. Psychol. 65, 187–207. doi: 10.1146/annurev-psych-010213-115202
PubMed Abstract | CrossRef Full Text | Google Scholar
Borlase, B. J., Gander, P. H., and Gibson, R. H. (2013). Effects of school start times and technology use on teenagers' sleep: 1999–2008. Sleep Biol. Rhythms 11, 46–54. doi: 10.1111/sbr.12003
CrossRef Full Text | Google Scholar
Carrell, S. E., Maghakian, T., and West, J. E. (2011). A's from Zzzz's? The causal effect of school start time on academic achievement of adolescents. Am. Econ. J. 3, 62–81. doi: 10.1257/pol.3.3.62
CrossRef Full Text
Churches, R., Dommett, E., and Devonshire, I. (2017). Neuroscience for Teachers: Applying Research Evidence from Brain Science. London: Crown House Publishing.
Danner, F., and Phillips, B. (2008). Adolescent sleep, school start times, and teen motor vehicle crashes. J. Clin. Sleep Med. 4, 533–535.
PubMed Abstract | Google Scholar
de Souza, C. M., and Hidalgo, M. P. (2014). Midpoint of sleep on school days is associated with depression among adolescents. Chronobiol. Int. 31, 199–205. doi: 10.3109/07420528.2013.838575
PubMed Abstract | CrossRef Full Text | Google Scholar
Department for Education (2016). Pupil Absence in Schools in England: Autumn 2015 and Spring 2016. London, UK: Department of Education, SFR 51/2016.
Edwards, F. (2012). Early to rise? The effect of daily start times on academic performance. Econ. Educ. Rev. 31, 970–983. doi: 10.1016/j.econedurev.2012.07.006
CrossRef Full Text | Google Scholar
Evans, M. D. R., Kelley, P., and Kelley, J. (2017). Identifying the best times for cognitive functioning using new methods: matching university times to undergraduate chronotypes. Front. Hum. Neurosci. 11:188. doi: 10.3389/fnhum.2017.00188
PubMed Abstract | CrossRef Full Text | Google Scholar
Foster, R. G., Peirson, S. N., Wulff, K., Winnebeck, E., Vetter, C., and Roenneberg, T. (2013). Sleep and circadian rhythm disruption in social jetlag and mental illness. Prog. Mol. Biol. Transl. Sci. 119, 325–346. doi: 10.1016/B978-0-12-396971-2.00011-7
PubMed Abstract | CrossRef Full Text | Google Scholar
Gradisar, M., Gardner, G., and Dohnt, H. (2011). Recent worldwide sleep patterns and problems during adolescence: a review and meta-analysis of age, region, and sleep. Sleep Med. 12, 110–118. doi: 10.1016/j.sleep.2010.11.008
PubMed Abstract | CrossRef Full Text | Google Scholar
Hansen, M., Janssen, I., Schiff, A., Zee, P. C., and Dubocovich, M. L. (2005). The impact of school daily schedule on adolescent sleep. Pediatrics 115, 1555–1561. doi: 10.1542/peds.2004-1649
PubMed Abstract | CrossRef Full Text | Google Scholar
Hattie, J. (2008). Visible learning: A synthesis of Over 800 Meta-Analyses Relating to Achievement. London: Routledge.
Hirshkowitz, M., Whiton, K., Albert, S. M., Alessi, C., Bruni, O., DonCarlos, L., et al. (2015). National Sleep Foundation's updated sleep duration recommendations: Final report. Sleep Health 1, 233–243. doi: 10.1016/j.sleh.2015.10.004
Things to consider
Recent research has led to some misleading headlines, some of which claim that vaping can cause cancer.
This isn’t true. There isn’t any evidence that suggests vaping causes cancer.
However, there is some evidence that suggests vaping may increase your overall risk for cancer. This is different than directly causing cancer.
We break down the tentative connection, assess the effects of different e-fluids, and more.
Have there been any diagnosed cancer cases directly tied to vaping?
There are no documented cancer diagnoses directly linked to vaping or e-cigarette use. However, this remains a difficult question to answer for a few reasons.
Not only is vaping a relatively recent phenomenon, people who vape tend to be on the younger side.
According to one , most people who use e-cigarettes are under the age of 35.
It can take decades before long-term effects appear. For example, most lung cancer diagnoses occur after the age of 65.
As a result, it could be years before we understand the link between vaping and long-term effects, such as cancer.
Another issue is that most people who vape are also current or former cigarette smokers.
The same 2018 study reported only of people who vape have never smoked cigarettes.
This presents a challenge for researchers, as it’s difficult to determine which health effects are caused by vaping, cigarette use, or a combination of the two.
How likely are you to develop cancer as a result of vaping?
It depends. If you use vaping as way to avoid or quit smoking cigarettes, vaping actually decreases your overall cancer risk.
But if you’ve never smoked cigarettes and aren’t planning on starting, vaping increases your overall cancer risk.
Although a 2018 review suggests vaping poses fewer health risks than smoking cigarettes, vaping isn’t risk-free.
And given the current lack of long-term studies, the overall health effects of vaping aren’t well-understood.
More research is needed to understand the potential implications of long-term vaping.
Does vaping increase your risk for a certain type of cancer?
Vaping has been linked to an increased risk of the following cancers:
This isn’t an exhaustive list, though. Additional research could link vaping to other types of cancer.
Most studies have focused on lung cancer. In one , researchers found that exposure to e-cigarette vapor led to DNA- and gene-level changes that could increase the risk of lung cancer.
Another animal study from 2018 concluded that smoke from e-cigarettes might contribute to lung and bladder cancer in humans.
These animal studies have significant limitations. In particular, they can’t replicate the way people actually use vaping devices. More research is needed.
Does it matter if the juice has nicotine in it?
Nicotine is what makes tobacco products addictive. Some vape juices contain nicotine while others do not.
The relationship between nicotine and cancer is a complex one. In general, research suggests nicotine exposure does present a cancer risk.
The results from a 2018 animal study suggest nicotine from e-cigarette vapor:
However, one major limitation of this study is that the animals were exposed to a dose much higher than that of typical vape use in humans.
More data is needed to understand the long-term effects of vaping with nicotine.
Does the juice flavor have an impact?
Juice flavor might have an impact on cancer risk.
One 2018 study on teens who vape found that fruit-based flavors contained higher levels of acrylonitrile, a toxic chemical.
The U.S. Environmental Protection Agency (EPA) classifies acrylonitrile as a “probable human carcinogen.”
In general, different flavors do appear to pose different health risks.
For example, one 2018 study examined the effects of common vape juice-flavoring chemicals on monocytes, a type of white blood cell.
Researchers found cinnamaldehyde (cinnamon flavor) was the most toxic to white blood cells. O-vanillin (vanilla flavor) and pentanedione (honey flavor) also had significant toxic cellular effects.
One 2016 study found that certain vape juice flavors were more toxic to lung cells. Among the flavors tested, strawberry was the most toxic. Coffee- and menthol-flavored e-juices also had toxic effects.
A also found that some common vape juice-flavoring chemicals, particularly diacetyl (butter/popcorn flavor), have been associated with severe respiratory illnesses.
Are there certain ingredients to avoid?
Vaping devices and liquids are regulated by the . Labeling requirements include a warning if the product contains nicotine.
Manufacturers aren’t required to list e-juice ingredients. However, as of 2018, they are required to submit an ingredient list to the FDA.
Juices and e-liquids contain several different types of ingredients. The main ingredients are listed below.
Different vape juices contain different nicotine concentrations.
Higher nicotine concentrations are associated with of adverse health effects.
People who are dependent on nicotine may consider gradually tapering the quantity of nicotine per milliliter.
The base is a flavorless suspension that constitutes most of the liquid in vape juice. Most manufacturers use a combination of propylene glycol (PG) or vegetable glycerin (VG), which is also referred to as glycerin or glycerol.
Both of these substances are classified as generally recognized as safe (GRAS) by the FDA. They appear in food, cosmetic, and pharmaceutical products.
However, this doesn’t mean adverse side effects aren’t possible.
One 2015 study used gas chromatography to assess the risks associated with exposure to PG and VG in a shisha pen. Researchers found the concentrations were high enough to potentially irritate the airways.
These ingredients vary according to the flavor of the juice. Some flavoring chemicals appear to be more toxic than others, while others can react with base liquids to create new and potentially toxic chemical compounds.
Research into both the short- and long-term health effects of flavoring ingredients is ongoing. More research is needed to fully understand which ingredients to avoid.
The following list includes flavoring chemicals that have been identified as potentially harmful:
It might not be possible to know the ingredients in a particular e-juice.
If you’re unable to review a product’s ingredient list, you might find it helpful to avoid the flavors that have been associated with the chemicals listed above.
These flavors include:
What about juuling?
“Juuling” is a term that comes from a popular e-cigarette brand, Juul. It’s essentially the same as vaping. The risks described in this article also apply to juuling.
Does vaping affect the lungs in the same way as cigarette smoking?
Smoking cigarettes and vaping affect the lungs differently. More research is necessary to truly understand their unique effects, though.
Cigarettes contain chemicals that irritate and damage tissue in your airways and lungs.
The tar in cigarette smoke can also build up in the lungs. This makes it more difficult to breathe.
Over time, smoking cigarettes can increase your risk for lung diseases, such as:
E-cigarettes contain fewer toxic chemicals than cigarettes. They don’t emit tar.
However, e-cigarettes still contain chemicals that could affect the lungs. More research is necessary to identify the long-term effects of extended exposure.
What about ‘popcorn lung’?
There are currently no cases linking vaping to popcorn lung.
Popcorn lung refers to a rare but serious lung condition called bronchiolitis obliterans, or fixed obstructive lung disease.
This condition inflames the lungs’ smallest airways (bronchioles), making it difficult to breathe.
The reference to popcorn comes from a chemical called diacetyl, which is used as a flavoring ingredient in microwave popcorn.
Diacetyl also appears in some vaping e-liquids.
has linked inhaling diacetyl at microwave popcorn manufacturing plants with certain lung diseases.
More research needs to be done to understand the short- and long-term effects of inhaling diacetyl in e-juice.
Are there other risks to consider?
The risks associated with vaping vary according to the device, e-juice, and habits of the user.
Some potential short-term risks include:
Some potential long-term risks include:
There’s currently no evidence that vaping increases the risk of heart or lung disease.
Research suggesting that vaping e-liquids contain high levels of heavy metals is limited.
Vaping may also present unique risks to teenagers and young adults.
There’s still a lot we don’t know about vaping. On the whole, though, it appears to present fewer risks than smoking cigarettes.
The bottom line
Based on what we know, vaping poses less of a cancer risk than smoking cigarettes. However, it might present an increased risk for people who don’t currently smoke cigarettes.
Speak to a doctor or other healthcare professional if you’re trying to quit smoking or have questions about vaping.
I sat in my high chair watching my older siblings excitedly playing with their gifts, and listening to Dad talked about me to Mom as she prepared breakfast.
Sometime this year, people in the US and Europe will start getting treated for diseases using the gene-editing tool CRISPR, but a big question remains—will it actually work?
Our primate cousins may hold the answer.
The first use of CRISPR to edit human cells in a dish was reported in 2013. It’s since been touted as an easy way to alter people’s DNA, promising to banish what are currently lethal or lifelong maladies with a single treatment that fixes them at the genetic root.
But since cutting DNA would permanently change someone’s genome, scientists need to make sure CRISPR is safe and effective before using it in people. Mice aren’t always an accurate way to predict how humans will react to a new therapy, so scientists often turn to monkeys as the gold standard in research that’s headed for the clinic.
At the moment, there isn’t much published data on CRISPR’s use in monkeys, but early results from small studies point to the prospect of a cure for some diseases—and to challenges in other cases.
Some of the ongoing CRISPR experiments in monkeys involve blood disorders like sickle-cell disease and beta thalassemia. CRISPR could be ideal for both diseases, scientists think, because each is caused by mutations in a single gene that makes hemoglobin, the protein in red blood cells that transports oxygen throughout the body. Both diseases could be repaired with a single genetic snip.
The idea is to use CRISPR to knock out a certain section of DNA and flip on a switch to produce a fetal type of hemoglobin that’s usually shut off after birth. This would give patients enough healthy hemoglobin to wipe out any symptoms of disease.
Researchers like Cynthia Dunbar at the National Institutes of Health are testing this approach in the blood-forming stem cells found in bone marrow. These cells are extracted from monkeys, altered with CRISPR in a lab, and then infused back into the monkeys’ marrow to grow and make new, healthy blood cells.
For a treatment to be effective, a certain percentage of cells need to get edited. In the monkeys she’s studying, Dunbar says, things look good initially, but after three to four months, only about 5 percent of cells have the necessary edit. To alleviate sickle-cell, “you’d probably need something approaching 20 percent,” she says.
Hans-Peter Kiem at the Fred Hutchinson Cancer Research Center is also working on the problem of sickle-cell in monkeys. His team was able to extract a certain type of stem cells from monkeys’ blood, modify the cells with CRISPR, and put them back into the animals. The editing rate was 40 percent, and the effects lasted for more than six months.
“Since monkeys are so similar to humans, I don’t think there’s going to be a huge challenge in translating this work to humans,” Kiem says. “We use the same technology as we would use in patients.”
One concern with CRISPR has been the possibility that it will make accidental cuts to other parts of the genome that aren’t being targeted. Kiem says he hasn’t yet seen any of these so-called off-target effects in monkeys, but his team is sequencing the genomes of the CRISPR-treated monkeys to make sure.
The biotech firm CRISPR Therapeutics is using a similar approach in a clinical trial for beta thalassemia that’s slated to begin this year in Europe. The company says it also plans to ask regulators to green-light a study on treating sickle-cell disease in people.
Editas Medicine, meanwhile, is using monkeys to test its CRISPR-based therapy for a kind of inherited blindness.
The company, based in Cambridge, Massachusetts, wants to correct a type of Leber congenital amaurosis, a rare genetic disorder that causes severe vision loss. Some people with it are born blind or begin to lose their vision when they’re still very young.
Charles Albright, chief scientific officer at Editas, says his team has used CRISPR to correct the disease-causing mutation in at least 10 percent of the light-sensing cells, called photoreceptors, in monkeys’ eyes. The treatment is injected directly into the eye.
That wouldn’t be enough to bring back a normal sense of sight, but Albright says it would allow patients to “see reasonably well” and perhaps live more independently.
“We think we have the potential to restore the ability of photoreceptors to detect light, which is the first step in being able to construct vision,” he says. In 2015, Editas said the therapy would be tested in people by 2017, but it’s had to delay the start of its first clinical trial because of a manufacturing problem. The company now intends to seek approval from the US Food and Drug Administration this year to begin that study.
But here’s the catch: the monkeys Dunbar and Kiem worked on didn’t have sickle-cell. Instead of looking for signs of improving symptoms, they used a blood test to measure the level of fetal hemoglobin in the animals’ bodies. Similarly, the Editas monkeys don’t have vision loss—scientists took tissue from the animals’ retinas and sequenced the DNA to confirm that they’d made the correct edit.
Researchers could get a better sense of whether a CRISPR therapy in monkeys will translate to cures for people if the animals had the same disease-causing mutations as humans.
Here, too, gene editing can help. Jon Hennebold, a researcher at the Oregon National Primate Research Center, is injecting monkey embryos with CRISPR to breed animals with more genetically precise equivalents of human diseases.
The goal is to disable certain segments of DNA to re-create mutations that are known to cause diseases in humans. Those embryos can then be used to artificially inseminate female monkeys, which will give birth to monkeys harboring genetic mutations nearly identical to those in humans.
Hennebold is trying to create these “knockout” models for human diseases like hereditary blindness and deafness, among other disorders. His lab hasn’t successfully bred such monkeys yet—the researchers are modifying embryos and testing them to make sure they’ve made the mutations they want. The next step will be transferring such embryos into female monkeys to generate live offspring that can be used for later studies.
“Right now, we’re trying to create a storehouse or library of embryos that are relevant for different diseases,” Hennebold says.
If the work appears slow and painstaking, that’s because it is. Hennebold says scientists need to move cautiously into human clinical trials of CRISPR because so little work has been done in monkeys. There’s still a question of whether CRISPR will cause immune reactions, for example, and off-target effects are a concern.
“We just don’t know enough about what it does outside of the region we want to modify,” he says.
In short, if you are thinking about starting a blog then there is no better time than If the.com version is already registered you may still be able to get the.net or.