3. Longitudinal studies and Interventions

Question 1

Cross-sectional/correlational studies measure ____ or ____ things at the same ____

Answer 1

Two or more things at the same point in time

Question 2

Longitudinal studies measure… and see how it…

Answer 2

One thing now
Predicts something at a later point in time

Question 3

Longitudinal studies don’t tell us about direct causation, but they do provide stronger evidence about causation. This is because of a temporal assumption. What does this mean?

Answer 3

That cause must come BEFORE effect

Question 4

A prospective cohort study follows a group of individuals over a period of time to see how certain ____ affect later ____

Answer 4

how certain exposures affect later outcomes

Question 5

An odds ratio (OR) in sleep research is used to q____ the strength of a____ between s____ problems and the subsequent d____ of m____ h____ d____ over t____

Answer 5

To quantify the strength of the association between sleep problems and the subsequent development of mental health diagnoses over time

Question 6

Odds Ratio (OR) is so important because…
1. It considers ____ groups
2. It tells you how ____ times more ____ it is that you would develop a ____ diagnosis if you had ____ problems compared to not having ____ problems

Answer 6

• Because it considers BOTH groups!
• The OR it tells you how MANY times more likely it is that you would develop a MH diagnosis if you had sleep problems compared to not having sleep problems

Question 7

In sleep research, what do the following odds ratios (OR) mean?
1. OR = 1
2. OR greater than 1
3. OR less than 1

Answer 7

1. Sleep problems do NOT affect the likelihood of developing a mental health diagnosis
2. Individuals with sleep problems are MORE likely to develop a mental health diagnosis compared to those without sleep problems
   (e.g. OR = 2 meaning that people with sleep problems are 2 x MORE likely to develop a mental health diagnosis over 1 year compared to those without sleep problems)
3. Individuals with sleep problems are LESS likely to develop a mental health diagnosis compared to those without sleep problems

Question 8

What is a meta-analysis?

Answer 8

A method of combining and analysing data from many studies that address the same research question

Question 9

In a forest plot, each study has what two things?

Answer 9

1. An effect size (square)
2. Confidence interval (line through square)

Question 10

What is the line of no effect in a forest plot?

Answer 10

This is placed at the point where the effect size equals no effect.

Question 11

When an Odds ratio is 1, why is the line of no effect at 1?

Answer 11

Because an OR of 1 means that there is no difference in odds between the two groups being compared.

Question 12

What did Hertenstein et al (20203) meta-analysis of longitudinal studies examine?

Answer 12

Whether insomnia predicts later mental health diagnosis

Question 13

Insomnia increases the risk of developing what four conditions at a later point in time?

Answer 13

1. Depression
2. Anxiety
3. Alcohol abuse
4. Psychotic disorders
   (strongest for anxiety and depression)

Question 14

The aim of Koren et al (2002) was to explore the role of sleep c____ post-t____ in the development of P____

Answer 14

Explore the role of sleep complaints post-trauma in the development of PTSD

Question 15

Who did Koren et al (2002) study?

Answer 15

102 survivors of motor vehicle accidents followed up over 1 year

Question 16

Of the 102 pts in Koren et al (2002), how many met the diagnostic criteria for PTSD?

Answer 16

26 (25.5%)

Question 17

What were the results of Koren et al (2002)?

Answer 17

• Most participants reported sleep difficulties 1-week after the accident
• The severity was slightly greater in those who developed PTSD
• The gap in sleep problems between PTSD and no-PTSD groups got bigger over time
• By 3-months the no-PTSD group have sleep difficulties comparable to the control group
• The PTSD group have sustained sleep difficulties over the year whereas the no-PTSD group sleep gets better and stabilises (lower than the control group at 1-year)

Question 18

Results from Koren et al (2002) suggest what?

Answer 18

• May be possible to detect who will develop PTSD from their post-trauma sleep complaints

Question 19

What is the basic question asked by interventionist-causal models?

Answer 19

Does change in X via intervention cause change in Y?

Question 20

What are the 5 components for Cognitive behavioural therapy for insomnia (CBT-I)?
1. Sleep h____
2. S____ control
3. Sleep r____
4. R____
5. C____

Answer 20

1. Sleep hygiene
2. Stimulus control
3. Sleep restriction
4. Relaxation
5. Cognitive

Question 21

What are the three most common relaxation exercises in CBT-I?
1. Progressive m____ relaxation
2. Deep b____
3. I____ training

Answer 21

1. Progressive Muscle Relaxation
2. Deep Breathing
3. Imagery Training

Question 22

What are the three common methods of cognitive therapy in CNT-I?
1. Cognitive r____
2. P____ intention
3. Thought s____

Answer 22

1. Cognitive Restructuring
2. Paradoxical Intention
3. Thought Stopping

Question 23

What was the aim of Scott et al (2021)?

Answer 23

Provide better causal evidence that sleep and mental health are causally related

Question 24

What were the three stipulations of Scott et al (2021) meta-analysis?

Answer 24

1. RCTs had to actually improve sleep
2. Studies had to look at mental health at a later point in time rather than immediately post-treatment
3. More than just CBT-I and/or anxiety and depression

Question 25

What were the results of Scott et al (2021) meta-analysis?
1. Studies ____ sleep ____ mental health
2. Evidence of ____-____ relationship

Answer 25

• studies improving sleep improved mental health
• evidence of dose-response relationship (more improve sleep, more improve mental health)

Question 26

The effect of _____ on insomnia is bigger than CBT-I and exercise

Answer 26

Acupuncture!