Link between Childhood Sleep & Adolescent Substance Use?

Medical student Akshay Krishnan and Dr. David Reichenberger explore the associations between childhood weeknight sleep and later alcohol and marijuana use. Their paper, “Childhood sleep is prospectively associated with adolescent alcohol and marijuana use,” can be found in the October 2024 (Vol. 98) issue of Annals of Epidemiology. 

Read the full article here:
https://www.sciencedirect.com/science/article/abs/pii/S1047279724001674

UCSF-Bay Area CFAR’s Mentoring the Mentors Workshop:
https://cfar.ucsf.edu/mentoring/events/mentoring-mentors-workshop-hiv-researchers

Episode Credits:

• Executive Producer: Sabrina Debas
• Technical Producer: Paula Burrows
• Annals of Epidemiology is published by Elsevier.
  
  
  

Chapters

• 9:33 Behind the Paper

Transcript

Patrick Sullivan: 0:12

Hello, you're listening to EpTalk Behind the Paper, a monthly podcast from the Annals of Epidemiology. I'm Patrick Sullivan, editor-in-chief of the journal, and in this series we take you behind the scenes of some of the latest epidemiologic research featured in our journal. Today we're talking with Mr Akshay Krishnan and Dr David Reichenberger about their article Childhood Sleep is Prospectively Associated with Adolescent Alcohol and Marijuana Use, respectively associated with adolescent alcohol and marijuana use. You can read the full article online in the October 2024 issue of the journal at wwwanalystofepidemiologyorg. First I'd like to introduce our guests.

Patrick Sullivan: 0:56

Akshay Krishnan is a second-year medical student at Sidney Kimmel Medical College at Thomas Jefferson University. He's part of the seven-year accelerated Pennsylvania State Sydney Medical College BS/MD program and his research centers on health-related social needs, telehealth efficacy, and the intersection of sleep and health. Dr. David Reichenberger, is a postdoctoral fellow at the Knight Cardiovascular Institute and the Oregon Institute of Occupational Health Sciences at the Oregon Health and Science University. His research examines the micro- and macro- longitudinal associations of daily health-related behaviors, like substance use or screen time, with nightly sleep health and, in turn, how changes in sleep influence cardiovascular health. Dr Reichenberger is currently looking for tenure-track faculty positions.

Patrick Sullivan: 1:46

Thank you both for joining us today. So this is a little bit of a different structure than we usually do, and I'm so happy we're doing it this way, because we have both the author of the paper and the mentor. So we're first going to talk a little bit about the study itself and then we'll spend a little bit of time talking about your mentoring relationship and how that worked, preparing this particular work. Akshay, I first just want to ask about the purpose of the study. What questions were you really interested in answering?

Akshay Krishnan: 2:15

Yeah, the purpose of our study was to investigate how childhood sleep predicts adolescent alcohol and marijuana use. So when we were doing our literature search we found many studies that looked at longitudinal associations. But those that were looking at longitudinal associations only looked at two time points one time in childhood and one time point in adolescence. But we were curious to see how this longitudinal relationship would span all of childhood.

Patrick Sullivan: 2:37

Great. So you've sort of alluded to this a little bit, but what is the study design and what methods did you have to, you know, select in order to answer the question that you were seeking to answer?

David Reichenberger : 2:49

I'll actually answer that one. Our study used data from the Future of Families and Child Well-Being Study. This was a longitudinal birth cohort that had started in 1998 and has collected data from children and their parents across multiple waves. This includes collecting data at birth, when the child was at ages 3, 5, 9, and age 15, and most recently when the child was age 22. So collecting data when they're now young adults. So this study is just rich with data. Some of the variables that we used were parent reports of the child's sleep, which has been collected ever since the age three wave and then during the age 15 wave, when the children were teens. They were finally old enough to self-report their own sleep and whether they had tried alcohol or marijuana by that point in their life. We applied a logistic regression framework to evaluate whether later bedtimes or shorter sleep duration at each wave across childhood increased the odds of drinking alcohol or trying marijuana by age 15.

Patrick Sullivan: 3:54

Great. And so what was the main finding? And then did it sort of line up with what you were hypothesizing when you started?

David Reichenberger : 4:01

Yeah, in general, we found that later bedtimes and shorter sleep durations at ages 5, 9, and 15 were associated with increased odds of substance use by age 15. And this ultimately aligned with our hypothesis.

Patrick Sullivan: 4:16

So in terms of how you came to that hypothesis, I mean, was there sort of previous literature or why would you postulate that such a relationship might exist?

Akshay Krishnan: 4:26

So when we looked at previous literature, we did find some associations between childhood sleep and adolescent alcohol and marijuana use, and what we found with our studies is that our findings mostly agreed with the previous literature. We found that insufficient sleep was cross-sectionally associated with increased adolescent alcohol use, but we also found no association between adolescent sleep duration and marijuana use, which prior studies have found. We think that the difference between our studies and prior studies could be due to how sleep was assessed, with prior studies focusing more on sleep problems and our studies focusing more on sleep duration and bedtime.

Patrick Sullivan: 5:00

Great. Can you say a little bit about what you see as the strengths and limitations of your study?

Akshay Krishnan: 5:05

I think one of the biggest strengths of our study is the sample population. The Future of Families cohort is a large, diverse national sample of urban children and the study has been following them since birth. And I think one of the limitations to our study is how the data was collected, because it was parent report or self-report data. There could be some recall bias there and also the way sleep was assessed in each wave of the study was slightly different, and daytime sleep, such as naps, was not asked about.

Patrick Sullivan: 5:32

And can I ask, was the alcohol drinking also reported by the parents or by the children?

David Reichenberger : 5:38

So that was actually answered by the adolescents.

Patrick Sullivan: 5:42

By the adolescents, so also the potential for some recall bias there or some misclassification, and whether they agreed to volunteer that information. Sure. So what prompted you to investigate childhood sleep as a risk factor for early adolescent substance use?

David Reichenberger : 5:58

So we really wanted to look at childhood sleep as a risk factor, because childhood is just such a crucial period for growth and development. Sleep is often supporting many of the changes that children undergo as they age, with one of the notable changes being brain maturation, and this is especially important as the child transitions from childhood into adolescence. So there have been many studies that have linked poor sleep with worse decision making, and worse decision making may be linked to the decision to engage in substance use when they're an adolescent. So really we're most interested in examining the possibility of sleep as the predetermining factor of this more distal outcome.

Patrick Sullivan: 6:44

Thanks. So you've talked a little bit about the nature of the exposure here, which is the amount of sleep, but I wonder whether it's possible that there may be other characteristics of households that might be confounded with the sleep exposure. So what are your thoughts about whether you think this is really leaning towards a causal relationship, or do you think there might be some confounding here with other household exposures in this period of adolescence?

David Reichenberger : 7:11

It's certainly possible that other factors are confounding this relationship, but we did our best really to try to control for many of those potential confounds by controlling for many sociodemographic characteristics as well as socioeconomic status. In the past we've looked at whether perhaps bedtime routines could be related to sleep in particular, especially because without a bedtime routine it may mean that sleep is less of a consistent habit and it could reflect more potentially household chaos or some other aspect that the child is experiencing in their life.

Patrick Sullivan: 7:53

Yeah, thanks for thinking that through. So I want to last ask about a couple findings that were just sort of striking to me. And again, I know this is an observational study so we're not going to attribute causality too strictly, but just your thoughts about or whether you were surprised by these.

Patrick Sullivan: 8:10

So one is that the participants who had later bedtimes at age three had a lower odds of drinking alcohol by age 15. And, in contrast, the participants that had later bedtimes at age nine had greater odds of drinking alcohol by that age. So do you have any thoughts about why bedtimes at early versus later ages might appear to have different associations with the drinking alcohol outcome?

David Reichenberger : 8:37

Yeah, this was a really counterintuitive finding that we didn't expect what we think and really we're speculating here because we were not able to assess daytime sleep. The question just simply wasn't asked at age three. But what we suspect is that it's perhaps that these children are having daytime sleep. They're taking naps throughout the day and that is pushing back their bedtime. So they're still getting sufficient sleep, but the later bedtime is potentially reflecting better sleep health at that age.

Patrick Sullivan: 9:14

Interesting. I know that was a little risky to speculate when you have these kinds of associations, but I think that's part of the process of seeing what you learn, figuring out what the hypotheses are for the next round and then either designing a new study to try and answer that, collecting additional data elements. So thanks for that. So now we're going to move to a segment we call Behind the Paper, which really gets to you as people as well as researchers, and I think it's really interesting to understand how ideas for studies come about and how people work with mentors and work in teams. So we want to talk a little bit about that. So first I wonder for either of you how you came to be interested in studying sleep health.

Akshay Krishnan: 9:57

I think my interest in sleep health started because I myself have a lot of trouble sleeping, so I was really interested in learning more about how I could both improve my sleep and also why sleep was so important to having a healthy lifestyle.

Patrick Sullivan: 10:09

Yeah, I think sometimes, when there are topics that we can relate to one, there's interest in studying them. But you may also have insight into asking the right questions or how you design something, and I think we bring our own experiences for sure to our research in that way. What was the biggest challenge you faced in conducting the analysis and how did you overcome it?

Akshay Krishnan: 10:32

One of the biggest challenges I faced while doing this research study came at the end, when it came time to write the paper. I'd never written a paper from start to finish before, so even though I knew all the parts that had to go into a paper, I didn't really know how to get it there on my own, and to overcome this challenge I worked a lot with Dr. Reichenberger and Dr. Anne- Marie Chang, who was David's mentor at the time as well. They spent a lot of time focusing on how to structure an abstract, how to structure the introduction, how to do the statistical analysis, how to structure your figures, and they both really spent countless hours teaching me how to go through this whole process, and it really shaped me into the researcher I am today.

Patrick Sullivan: 11:05

Great, it's great to hear about that process and can I just ask like for those sections, was part of that process looking at examples of introductions, or was it more by example, or was it more by sort of brainstorming together what the components of each segment might be?

Akshay Krishnan: 11:20

I think it was more a combination of both, because before I started writing the paper we did a literature search. So as I was doing literature search I was reading a lot of papers on similar topics. So from there I kind of got an idea of what needed to go into each section. And then, as we were writing the paper, we would go through section by section and pick out what was good from each section and what could use more work, and then we'd brainstorm ideas on how to make it better.

Patrick Sullivan: 11:40

That sounds like a great process and actually, even when you're later in your career, I think writing with colleagues is more fun than writing alone, so that's a great habit to start. This again might be for either of you, but were you surprised by any of the findings? We talked a little bit about a couple of the findings, but anything else surprise you, or did you have any associations that you predicted that didn't pan out?

Akshay Krishnan: 12:02

Yeah, when we first started this project I didn't believe that there would be associations between childhood sleep and adolescent alcohol and marijuana use, especially if we were measuring the sleep 10 to 12 years before the substance use. But while we were doing the project we found that it was associated. So it was really interesting to kind of like have your opinions tested and then be able to do some digging into why that may have happened.

Patrick Sullivan: 12:23

And does that suggest any sort of next step questions? You have this association now in observational data. Does it prompt you to think about any next step questions or any grant ideas, dissertation ideas or any of those ways we might pursue next steps?

David Reichenberger : 12:41

Uh so, not yet for grants, certainly, but it definitely inspires me to really consider the bidirectional relationship between sleep and substance use, that sleep can indeed influence our choices, potentially, of course, and that substance use can affect subsequent sleep, and then looking at how that reciprocity may change over time.

Patrick Sullivan: 13:10

Interesting. So you've sort of already alluded to this, but we're lucky to have both of you on the call today. I love interviewing trainees- Hopefully we're all learners at any age but someone in that training phase and a mentor. But it sounds like there's another sort of colleague that's in the mix here of your work. So I wonder if you could talk about that mentorship situation either how you work together or it sounds like there may be another colleague who's also contributing to this and how does the structure of that mentoring work and how do you think that's impacted your career development?

Akshay Krishnan: 13:41

Yeah.

Akshay Krishnan: 13:41

So when I was at Penn State, I really had the privilege of being mentored by both Dr. Reichenberger, who was a graduate student when he was mentoring me, and then also Dr. Anne- Marie Chang, who was Dr. Reichenberger's graduate advisor.

Akshay Krishnan: 13:55

They both had a lot of research experience which they really imparted onto me. So during this mentorship process I worked with them during different times of day and they would help me with different things, and we'd also have some time where I'll be working together. So when I talked to Dr. Reichenberger, we talked a lot about the research process itself how to come up with a good question, how to fill out an IRB, how to design study methods and how to do statistical analysis once you finish the study. And then, when I talked to Dr. Chang, we talked a lot about how to structure a paper, what type of language I should be using and how to design figures, and so, because I was able to get advice and mentorship from both of these amazing researchers, I was really able to develop my own passion for research, which I've continued now, even in medical school, when classes are a lot more busy.

Patrick Sullivan: 14:38

It's great that you're able to keep that research interest while you're in medical school. It's juggling a lot. And, Dr. Reichenberger, can I just ask you, in that relationship you're sort of probably most frequently working with Akshay, but you also have this other resource, so did you get sort of mentoring on mentoring? Was there any sort of coaching in that relationship with Dr. Chang?

David Reichenberger : 14:59

Yeah, absolutely. There were very few times where Akshay gave me trouble, but I still had questions about, like, how to interact with him or how to broach a certain topic, and so Dr. Chang was always there for me to teach me how to mentor Akshay. And Dr Chang was just an amazing advisor as well, and so just by working with her, I kind of already knew the kind of mentor I wanted to be and how I wanted to interact with Akshay as my mentee.

Patrick Sullivan: 15:32

And I'm going to shout out Dr. Ted McDonald, who is my major professor, who I'll send this podcast to but it really is like that investment that Dr. Chang made in you enables this relationship that you can have with Akshay, and so I think it's really worth thinking about why these relationships work well when they do, and this sort of tiered mentorship maybe because it's not intuitive always how to be the best mentor even if you've had great ones being able to go back and and being mentored on mentor.

Patrick Sullivan: 16:04

There's also a great curriculum in UCSF called Mentoring the Mentors. That is like a two or three day program that you could find online that really addresses a lot of these things for earlier career mentors. So thank you both for sharing that and I guess I'll just sort of wrap up by saying, maybe for each of you, what advice do you have for aspiring students or for earlier career researchers who want to pursue research? You know what steps were helpful to you. How did you seek and find mentorship? What advice would you have?

Akshay Krishnan: 16:35

I think my advice for other students trying to pursue research is to find a topic they're truly passionate about, because if you're passionate about the field you're researching, the hours it takes to do data collection and writing won't feel like a burden and you'll be really able to explore something you want to learn more about. I would also advise students to find a good mentor, because there's a lot that goes into having a successful study, not just you going and talking to the participants. It's a lot more about the whole process behind it and it's really hard to learn that by yourself. But if you have a strong mentor, a strong set of mentors, it makes the process go a lot easier and really helps you take a lot more out of the process and can really help you ignite your passion for research and keep doing it even later on in your career.

David Reichenberger : 17:14

Yeah, I totally echo that sentiment. Really having the right mentor for you as an individual can make or break your research experience or your experience in graduate school.

Patrick Sullivan: 17:27

All right, any last thoughts that either of you would like to share with our listeners?

David Reichenberger : 17:30

No, I just really want to take the time to thank you for giving us this opportunity to chat.

Patrick Sullivan: 17:37

Thank you. It's our pleasure and we appreciate you bringing your work to Annals and appreciate the chance to you know, spread the word a little bit more about your work, but also about this process. So that brings us to the end of this episode. I want to thank again Akshay Krishnan and Dr. David Reichenberger for joining us today. It's such a pleasure to have you on the podcast.

David Reichenberger : 17:59

Thank you so much.

Patrick Sullivan: 18:00

Thank you. I'm your host, Patrick Sullivan. Thanks for tuning in to this episode and see you next time on EPITalk, brought to you by Annals of Epidemiology, the official journal of the American College of Epidemiology. For a transcript of this podcast or to read the article featured on this episode and more from the journal, you can visit us online at www. annalsofepidemiology. org.