Epi in the News

Hi Sujit,

Long time reader, first time poster.

1. What is the outcome?
Outcome:
Development of depression

Comparison group:
- Those who had moderate exercise for 2.5 hours per week (recommended weekly limit)
- Those who had moderate exercise for 1.25 hours per week (half recommended)
- Those who did not do moderate exercise weekly

2. What are the Risk ratios?

25% - RR = 0.75 (Rate of depression in exercise group / rate of depression in no exercise)
18% RR = 0.82 (Rate of depression in 1.25 exercise group / rate of depression in no exercise no exercise)

3. Physical exercise reduces rate of mental health conditions. Plausibly you could say that physical exercise also decreases rates of other medical conditions such as cardiovascular disease and obesity, which are both linked to poorer mental health as well.

4. Prospective cohort study. A cohort study is preferable over a case/control trial as it is very difficult to account for all the different variables which contribute to poorer mental health outcomes.

5. People who run are conceivably more likely to have healthier lifestyles overall. Furthermore they may participate in sport (community) which may decrease loneliness and improve mental health.

6. Not sure!

Hi Sujit and hi to all the other people answering this forum, I am Ileana, from Spain, new student of the Msc and I will try to respond first and then compare my answer to the others.
Thank you in advance!
FOR NEW STUDENTS

1. What is the outcome and what are the comparison groups?
They selected cohort prospective studies, and it seems like they compared the exposure to exercice (compared to lower volume to physical activity), so the outcome would be depression. Anyways in the supplemental material the authors gave this information very well explained

2. What are the RRs which correspond to the figures 25% and 18%? They correspong to the relative risks exprresed in Table two, 4.4h/w RR 0.82 (CI 0.77-0.87) and 8.8h/wk RR 0.75 (CI:0.66-0.82)

Thanks!

Hello Sujit and others!

I'm also a new student studying MSc Public Health. I will try the first questions, please see my answers below:

q1. The outcome is depression and the comparison groups are adults who exercised (moderate physical activity) for 2.5hrs per week, adults who exercised (moderate physical activity) for 1.25hrs per week and adults who did not exercise at all.

q2. According to table 2 in the paper we can identify the RRs associated with the figures 25% and 18% as 0.75 and 0.82 respectively.

Can you clarify exactly what type of "relative risk" these are? I was thinking they are incidence risk ratios as we are looking at the development of depression in those who exercised (for a certain period of time) compared to the development of depression in those who did not exercise at all.

Thank you!
~Ashna

Hello Sujit and Everyone else,

I'd like to pitch in with my perspective on the answers:
FOR NEW STUDENTS

1. What is the outcome and what are the comparison groups?
- The outcome is depression. Because this is a meta-analysis, the data is harmonised. So qualifying as depression takes 2 forms here:
1) Major Depressive Disorder diagnosed via a criteria such as DSM5 or ISCD
2) Elevated depressive symptoms as elicited through screening tools such as the Center for Epidemiologic Studies Depression scale
> comments: interesting that diagnosis through full criteria is practically put on the same level as a screening tool.

- Comparison groups have been based on exposure to physical activity, divided into 3 brackets based on METs of exercise:
1) 4.4 mMET-h/week. Which is basically half the minimum recommended physical activity levels
2) 8.8 mMET-h/week. Which corresponds with the minimum recommended physical activity levels
3) 17.5 mMET-h/week. Which should represent most of the benefits of increased physical activity.
> Comments: a lot of effort was put in adjusting the values to remove occupational activity, non-exercise activity and resting METs. So it is attempting to only control for exercise. They even calculated it as marginal METs, which attempts to remove the resting METs that occur during exercise.

2. What are the RRs which correspond to the figures 25% and 18%?
18% corresponds to the difference in relative risk of depression in subjects who hit 4.4 mMET-h/week of exercise compared with those who did not report any physical activity
25% corresponds to the difference in relative risk of depression in subjects who hit 8.8 mMET-h/week of exercise compared with those who did not report any physical activity
> comments: without diving into the rabbit hole of checking each individual paper, I'm not sure what classified as no physical activity and how this was measured.
> comments 2: essentially, the authors are trying to elicit a dose-response relationship from amount of exercise compared with people who do not exercise for the outcome of depression.

FOR RETURNING STUDENTS

3. What is the mediator (causal pathway) hypothesised?
The authors state it's likely there is more than 1 mediator. Examples include activation of pathways to get the "runner's high", long term neuroplastic adaptations, improved body image, improved social life, exposure to green spaces, amongst many other possible mediators. There is also the possibility of reverse causality, where depressed people are less likely to go outside (anhedonia - not enjoying activities you once enjoyed -, and lethargy being some of the diagnostic criteria of MDD)

4. What is the study design, and why was this design preferable to other options?
This is a systemic review and meta-analysis of 15 studies, as far as I can tell, they are all prospective cohort studies. Meta-analysis is the gold standard for evidence in order to answer a specific research question. Given the circumstances, using prospective cohort studies is a good option, because there are ethical issues with a clinical trial forcing people to be sedentary when we know there is good evidence for harm in a sedentary lifestyle.

5. Aside their outcomes, are there differences between people who run and people who don't? Why do some of these differences matter?
- definitely. People who run are able to run because they have spare time and resources to participate in a leisure activity or invest in their health. For example, people who are struggling to get by through working multiple jobs would not have this luxury. So SES is one.
There are also other background factors like physical health. If someone is in poor health or has a disability, they may not be able to run. Chronic medical conditions and disability is also likely to be a significant risk factor for depression.

6. To prove causation, do we need to "run" a trial?
As mentioned before, I think a proper clinical trial may be difficult to pass through the ethics board. I'm going to put my opinion that we don't need to "prove" causation. It is enough to provide compelling evidence, like this paper, that there is a dose-response relationship between exercise and depression. I think it's not difficult to convince people that exercise is good for them. The more clinically relevant question is how much is enough, and what is the potential loss of not doing enough. Splashing research money to conclusively prove this may not be wise use of funding.

Hi Sujit,

Answers to (1) and (2)

The outcome was incident depression (more details about this in the paper: The outcome of interest was depression, including (1) presence of major depressive disorder indicated by self-report of physician diagnosis, registry data, or diagnostic interviews and (2) elevated depressive symptoms established using validated cutoffs for a depressive screening instrument.)

Comparison groups and RR interpretation

Compared with adults with no physical activity at all ( i.e 0 hours weekly = 0 mMET-h/wk) (baseline group),

adults with half the recommended activity (i.e 1.25 weekly hours = 4.4 mMET-h/wk) had 18 percent lower risk of depression (RR of 0,82)

and

adults with the total recommended activity (i.e 2.5 hours weekly = 8.8 mMET-h/wk) had 25 percent lower risk of depression (RR of 0.75) .

Outcome : Risk of Depression
Comparison Groups : 1) Adults who completed 2.5 hours of moderate physical activity per week
2) Adults who completed 1.25 hours of moderate physical activity per week

Hi Sujit and fellow students,

The following are my answers:

1. The outcome is depression.
The comparison groups are:
i. adults with 2.5 hours of moderate physical activity per week
ii. adults with just half of the 2.5 hours (i.e. 1.25 hours) of weekly physical activity
iii. adults who didn’t exercise

2. The RRs which correspond to the figures 25% and 18% are 0.75 and 0.82 respectively.

3. The mediator (causal pathway) hypothesised is: a group of molecules known as 'exerkines', which is released by several organ systems in response to exercise. Few studies have linked them to reductions in harmful inflammation, generation of new blood vessels and regeneration of cellular mitochondria, and therefore having beneficial effects on our physical and mental health.

4. This is a prospective cohort study. This design helps assess temporality and therefore causality of the association. Thus this design is preferable to other observational study designs.

5. The people who run are apparently healthier, aware about their health, motivated to follow a healthy lifestyle, and belong to higher socio-economic status.

These differences might confound this association.

6. We cannot run a trial due to ethical reasons in this case. A well-designed observational study (case-control or cohort) will help prove causation.

Thanks!
Madhutandra