Measuring and interpreting health risk (1.8, 1.9, 1.10)

A correlation between a factor (e.g., a high-fat diet) and CVD suggests a link, but controlled experiments are needed to show causation.

Confounding variables (e.g., lifestyle) may distort results. These are variables that affect both the variables being studied.

• Moderate alcohol consumption correlates with lower CVD risk. There is no evidence of causation: the correlation may be due to other confounding lifestyle factors (e.g., diet, exercise).
• Smoking correlates with a higher risk of CVD. This is supported by evidence:
  • Smoking damages endothelium…
  • …which promotes atherosclerosis…
  • …and this increases the chance of clotting or rupture.

For smoking and higher CVD risk, there is correlation and causation.

Scientific studies can show conflicting trends because:

• Different study designs can lead to different conclusions.
• Population differences (age, genetics, lifestyle).
• Measurement bias or data collection errors.
• Short-term vs long-term studies may give different results.

Conflicting evidence doesn’t mean results should be discarded, but their universality should be reconsidered. To increase confidence in findings, further research, larger sample sizes, or better experimental control are required.

Health studies are used to identify risk factors for diseases such as CVD, cancer, and diabetes.

To draw valid conclusions, studies must be well-designed and controlled.

Longitudinal studies track outcomes over time.

Cohort studies are a specific type of longitudinal study:

1. A large group of healthy individuals are observed over time.
2. Natural exposure to risk factors (e.g., diet, smoking, activity) is recorded.
3. Outcome is recorded: (e.g., does each individual develop the disease?)

Advantages:

• Shows time-based (temporal) relationships.
• Can help suggest causal links through multivariate analysis.

Limitations:

• Long-term and expensive.
• Risk of participant drop-out.

There are also ethical considerations in cohort studies in which participants are not deterred from making decisions that are detrimental to their health (e.g., smoking).

Case-control studies compare people with a disease (cases) to similar people without it (controls).

These studies look back at past exposures or lifestyles.

Advantages:

• Quicker and cheaper than cohort studies.
• Useful for rare diseases.

Limitations:

• Relies on memory (recall bias).
• Can only show correlation, not causation.

Risk is the probability that a harmful event (e.g., heart disease, stroke) will occur.

• Actual risk is based on scientific data and epidemiological studies.
• Perceived risk is how dangerous people think something is, and it can be influenced by emotions, the media, and experience.

Overestimation of risk is when people think something is more dangerous than it really is. Overestimation of risk can occur when:

• Something is unfamiliar or new (e.g., new cholesterol drugs).
• Something is highly publicised (e.g., media stories about heart attacks in young athletes).
• The danger seems beyond personal control (e.g., pollution or genetics).
• There are emotional associations (fear of sudden death).
• The person knows someone personally affected.

Underestimation of risk occurs when people think something is less dangerous than it really is. Underestimation of risk can occur when:

• The risk is familiar or voluntary (e.g., eating unhealthy food, smoking).
• Effects take years to develop; there is no immediate consequence.
• The person believes they can control the outcome (e.g., “I exercise, so I’m safe”).
• The person doesn’t understand or trust scientific statistics.
• The benefits (pleasure from food, convenience) seem to outweigh the risks.

Where risks are not accurately perceived, decision-making becomes compromised.

• Overestimation leads to unnecessary anxiety or fad behaviour
  (e.g., avoiding all fats, leading to nutrient imbalance).
• Underestimation leads to poor lifestyle choices
  (e.g., continuing to smoke or eat high-fat foods).

Public communications can help people perceive risk more accurately by:

• Presenting data as absolute risk (e.g., 1 in 1000) rather than relative risk (e.g., “doubles the risk”).
• Using clear, evidence-based messages.
• Educating about long-term cumulative risks (e.g., diet, inactivity, smoking).
• Address misinformation spread by media or social platforms.