Phase 1: Prepare
Step 1 Warm-up recall
From M5.L1: what is the formula for odds, and how does an odds ratio differ from a probability ratio?
Private: nothing you type here is sent or stored.
Step 2 Learning goal
By the end of this lesson you can:
- state what Cohen, Bavishi and Rozanski (2016) pooled and found (Foundation);
- read a relative risk and convert it into a percentage change (Core);
- tell relative risk from absolute risk (Core).
Phase 2: Teach
Step 3 A familiar example
A shop offers “20% off” on a coffee costing 2 pounds. You save 40 pence. A different shop offers 20% off a car costing 20,000 pounds, and you save 4,000. The same percentage means very different amounts. Risk works the same way.
Step 4 Clear explanation
Cohen, Bavishi and Rozanski (2016) pooled 10 prospective studies with 136,265 participants. After adjustment, a high sense of purpose in life was associated with a pooled relative risk of 0.83 for all-cause mortality (95% confidence interval 0.75 to 0.91) and of 0.83 for cardiovascular events (95% confidence interval 0.75 to 0.92).
Relative risk (RR) is the risk in one group divided by the risk in another. An RR of 0.83 means the risk in the high-purpose group was 83% of the risk in the comparison group: a 17% lower risk, since (1 − 0.83) × 100 = 17. Unlike an odds ratio, a relative risk compares probabilities directly.
Absolute risk is the actual chance. If the comparison group had a 10% risk over the follow-up period, an RR of 0.83 means the high-purpose group had 8.3%: a difference of 1.7 percentage points. The relative figure sounds bigger than the absolute one, and good reports give both.
This is an association across observational studies. Purpose could protect health, but illness could also reduce purpose, and a third factor such as income or social support could affect both, as in M5.L3.
Step 5 Worked example (4 marks)
A study reports RR = 0.75 for a cardiovascular event among people with high purpose. The comparison group’s risk is 8%.
- Relative reduction = (1 − 0.75) × 100 = 25%. (1 mark)
- Absolute risk in the high-purpose group = 0.75 × 8% = 6%. (1 mark)
- Absolute difference = 8 − 6 = 2 percentage points. (1 mark)
- Fair statement: high purpose was associated with a 25% lower relative risk, about 2 points lower in absolute terms, with cause not established. (1 mark)
Phase 3: Practise
Step 6 Learner practice (instant-marked)
2. How many prospective studies and participants did Cohen, Bavishi and Rozanski (2016) pool?
Ten prospective studies with 136,265 participants.
3. What pooled relative risk for all-cause mortality was reported for high purpose?
RR 0.83, 95% CI 0.75 to 0.91.
4. A relative risk of 0.83 applied to a baseline risk of 10% gives an absolute risk of:
0.83 × 10% = 8.3%.
Phase 4: Close
Step 7 Recap
- Cohen and colleagues (2016): 10 prospective studies, 136,265 participants; RR 0.83 for mortality and for cardiovascular events.
- Percentage change in risk = (1 − RR) × 100.
- Relative and absolute risk differ; report both.
Step 8 Exit ticket (ungraded)
Write one sentence reporting this finding with both a relative and an absolute figure, using an invented baseline.
Private: nothing you type here is sent or stored.
Step 9 Go further (Extension)
Read the abstract of Cohen, Bavishi and Rozanski (2016) in Psychosomatic Medicine. Find how many studies reported on cardiovascular events and what the authors say about limitations.
Common mistakes
| Mistake | Why it fails |
|---|---|
| Treating a relative risk as an absolute risk | A 17% relative drop is not a 17-point drop. |
| Using odds-ratio language for a relative risk | They are different measures. |
| Assuming purpose causes health | Observational data leave direction of cause open. |
Limits. Figures come from the paper’s abstract. The baseline risk used here is invented for teaching.