org355: Sleep & Heart Disease

1. What is the epidemiologic study design?

2. What would be the key eligibility criteria to be part of the analysis?

More specifically, those who fell asleep at midnight or later had a 25% higher risk of going on to develop cardiovascular disease, while those who fell asleep before 10pm had a 24% increased risk.

3. What were the comparison groups?

4. What sort of RR figures correspond to 25% and 24%?

The team say the findings appear to be stronger in women than men...

5. What does this mean in terms of the RRs?

“Because we also adjusted for all of the other more common cardiovascular risk factors, it’s clear that this association is significant in some way,” said Plans.

6. What does "adjusted for" mean, and why is this important?

Plans said further research, with larger numbers of participants, is needed to examine the findings, adding there was not enough evidence at present to prescribe a particular bedtime to the public.

7. What is the advantage of doing another, larger study?

8. What do we need to know before we can say that the timing of bedtime has a causal relationship to heart health?

Re: Sleep & Heart Disease

by | SOUJANYA KAUP - Tuesday, 9 November 2021, 4:34 PM

Hello Sujit,

Thanks for yet another great learning opportunity.

My answers are as follows:

1. What is the epidemiologic study design?

Cohort study

2. What would be the key eligibility criteria to be part of the analysis?

Adults enumerated in UK Biobank who do not have cardiovascular disease or sleep disorders.

3. What were the comparison groups?

Those who fell asleep:

a. after midnight or later

b. 11 to 11.59 PM

c. 10 to 10.59 PM

d. before 10 PM

4. What sort of RR figures correspond to 25% and 24%?

Stratum specific RR

5. What does this mean in terms of the RRs?

Higher the RR, stronger the association.

6. What does "adjusted for" mean, and why is this important?

The authors have identified factors which are known independent risk factors for the outcome, cardiovascular disease. By 'adjusting' for these factors, the effect of sleep can be independently assessed by comparing the effect of sleep among those with all other factors which are at same level.

7. What is the advantage of doing another, larger study?

As cardiovascular diseases are multifactorial, it would be necessary to adjust for these variables to ascertain the effect of sleep. The power of the study would decrease with each additional variable adjusted for.

Thus the study might not be powered enough to detect a true difference if it exists. A larger study might be able to ascertain a true effect of sleep.

8. What do we need to know before we can say that the timing of bedtime has a causal relationship to heart health?

For causality to be established, the Bradford Hill criteria needs to be fulfilled.

Soujanya Kaup

Re: Sleep & Heart Disease

by | MADHUTANDRA SARKAR - Wednesday, 10 November 2021, 6:58 PM

1. This is a cohort study.
2. The key eligibility criteria to be part of the analysis would be adults aged 43 to 79 of the UK Biobank with none of them having the heart disease or a sleep disorder at the start of the study.
3. The comparison groups were 1) participants going to sleep at 10pm or shortly after and 2) participants falling asleep earlier or later at night.
4. These are stratum-specific measures.
5. RR is a measure of the strength of the association between the exposure and the outcome. So, “the findings appear to be stronger in women than men” means the risk of heart disease was higher among women than men.
6. “Adjusted for” means the researchers controlled for or removed the effects of all other more common cardiovascular risk factors which might confound this observed association.
This is important because this adjustment made the association between timing of sleep and heart disease stronger than it really was.
7. There was not enough evidence at present to prescribe a particular bedtime to the public. Further research with larger numbers of participants is needed to examine the findings.
8. We need to know the Bradford Hill criteria for causation before we can say that the sleep timing has a causal relationship to heart health.

Re: Sleep & Heart Disease

by | JUDITH MARGARET BURCHARDT - Saturday, 13 November 2021, 10:24 AM

Thank you Soujanya, Madhutandra and Sujit,

I agree with the previous comments.

1. What is the epidemiologic study design?

Cohort study
2. What would be the key eligibility criteria to be part of the analysis?

UK biobank members aged 43 - 79 with Data on sleep time and CVS outcome

More specifically, those who fell asleep at midnight or later had a 25% higher risk of going on to develop cardiovascular disease, while those who fell asleep before 10pm had a 24% increased risk.

3. What were the comparison groups?

Those who went to sleep from 10pm - 10.59pm
4. What sort of RR figures correspond to 25% and 24%?
1.25 and 1.24
The team say the findings appear to be stronger in women than men...

5. What does this mean in terms of the RRs?
The RRs will be a little higher in women and a little lower in men
“Because we also adjusted for all of the other more common cardiovascular risk factors, it’s clear that this association is significant in some way,” said Plans.

6. What does "adjusted for" mean, and why is this important?
Confounders have been considered and allowance has been made for them. This means that the association between sleep time and outcomes can be considered without confounding. This assumes there is no residual confounding.

Plans said further research, with larger numbers of participants, is needed to examine the findings, adding there was not enough evidence at present to prescribe a particular bedtime to the public.

7. What is the advantage of doing another, larger study?

The confidence intervals around the estimates will become narrower.

8. What do we need to know before we can say that the timing of bedtime has a causal relationship to heart health?

Bradford Hill criteria. This would include - Is this biologically plausible? (the article says it is about resetting biological clocks with morning light). Ideally a randomised controlled trial asking some people to go to sleep earlier and others later and then measuring outcomes. (It is hard to see how this would work in practice.)

Judith

Re: Sleep & Heart Disease

by | SOUJANYA KAUP - Sunday, 14 November 2021, 11:51 AM

Thanks, Judith!
Regarding Q4: I just realized these are not stratum-specific RRs. I did not understand the question clearly. Your answer seems to be right to me now. A RR of 1.25 would mean there would be a 25% higher risk among exposed com[ared to unexposed.

Soujanya