Health issues & The effect of lifestyle on some non-communicable diseases

SPEC 4.2.2.5–4.2.2.6 • DATA SKILLS

Lifestyle Risk Factors & Data Interpretation

The content of risk factors is covered. This lesson is about the SKILLS — reading tables, interpreting graphs, distinguishing correlation from causation, and writing answers that actually score marks in compare-style questions.

The Comparison Failure

Risk-factor data questions are a reliable section of Paper 1. They're also where students lose the most marks unnecessarily.

Examiner reports show that on data-interpretation questions about smoking and cardiovascular disease, roughly half of students score zero. The reason isn't lack of content knowledge — they've memorised which lifestyle factors cause which diseases. The reason is they misread the data or fail to use comparative language. — pattern recurring across AQA risk-factor data questions

What You Get From This Lesson

How to read column headers properly. How to spot correlation vs causation. How to interpret scatter diagrams. How to write a comparison answer that actually uses comparative language. And the difference between "risk of developing X" and "risk of X" — that one phrase decides marks.

What You'll Learn

By the end of this lesson, you'll be able to:

Lesson Objectives

Distinguish correlation from causation, and explain when a causal mechanism has been proven.

Interpret risk factor data from tables, bar charts, and scatter diagrams.

Apply the principles of sampling to scientific data about risk factors.

Write comparison-style answers that use the specific language mark schemes credit.

Spec ref: 4.2.2.5 (Health issues) and 4.2.2.6 (Lifestyle effect on NCDs) — skills focus. Paper 1, Higher tier.

Quick Recap — The Named Factors

A reminder of the spec-named risk factors and what they're linked to (full detail in Lesson 2.07):

Risk factorLinked to

Diet, smoking, exerciseCardiovascular disease

ObesityType 2 diabetes

AlcoholLiver and brain function

SmokingLung disease and lung cancer

Smoking + alcohol (during pregnancy)Unborn babies

Carcinogens (incl. ionising radiation)Cancer

This Lesson's Focus

You already know these. This lesson is about how to USE this knowledge when AQA hands you a data table, graph, or scatter diagram and asks you to interpret it.

Correlation vs Causation

Two terms that look similar but mean very different things. Mark schemes are strict about which one applies to a given dataset.

Correlation

Two factors change together — but one does NOT necessarily cause the other. A correlation shows a link, not a cause.

Example: ice cream sales and drownings both rise in summer. They're correlated. Ice cream doesn't cause drowning — both are caused by warm weather.

Causation

One factor directly causes the other. Proven only by understanding the biological mechanism linking them.

Example: smoking causes lung cancer — proven via mechanism: tar contains carcinogens → DNA damage → mutations → uncontrolled cell division → tumour.

Mark Scheme Trap

When data shows a correlation, you can describe the correlation. But unless the question gives you a mechanism, don't claim causation. A scatter diagram showing two variables moving together is correlation evidence — not proof of cause.

"A Causal Mechanism Has Been Proven..."

The spec uses this exact phrase. It means: for SOME risk factors, scientists have identified the biological process that explains HOW the factor causes the disease. For others, only correlation is known.

Risk factor → DiseaseMechanism (proven?)

Smoking → lung cancer✓ Causal mechanism proven : tar contains carcinogens → DNA damage in lung cells → mutations → uncontrolled cell division → tumour.

Saturated fat in diet → CHD✓ Causal mechanism proven : raises blood cholesterol → fatty material builds up in coronary arteries → narrowing → reduced blood flow to heart muscle.

Obesity → Type 2 diabetes✓ Causal mechanism proven : cells become less responsive to insulin → blood glucose stays elevated.

Carcinogens (incl. ionising radiation) → cancer✓ Causal mechanism proven : carcinogens or radiation damage DNA in cells → mutations → uncontrolled cell division → tumour formation.

Other lifestyle/disease linksSome are correlations only — no proven mechanism. The data shows a link; the biology is still being investigated.

Why It Matters

An "explain" question wants the mechanism. A "describe" or "interpret" question wants the data pattern. Match the answer type to the question.

Reading Tables — The Column Header Trap

Before you write a single word, read what's actually being measured. The most common mistake in data questions is misreading the column header.

"Risk of developing X" ≠ "Risk of X"

If the column header says "% increase in risk of developing CHD", your answer must talk about risk of developing CHD. Writing "smoking is the most dangerous" is a different claim and scores zero — that's about general danger, not increased risk of a specific outcome.

Before Answering, Identify

What's the variable on each axis / column?

What units? (% increase? Per 1000 people? Per year?)

What's the reference point? (compared to non-smokers? to the general population?)

Comparison Questions — The Language That Scores

When a question says "compare", the mark scheme demands comparative language. Stating two values isn't comparing — it's describing.

Doesn't Score

"Disease A increases by 40%. Disease B increases by 15%."

Why: Two facts stated. No actual comparison made.

Scores

"Smoking increases the risk of developing Disease A the most (40%), and Disease B the least (15%) — a difference of 25 percentage points."

Why: Uses superlatives + values + reference point.

Comparative Words That Mark Schemes Credit

Superlatives: highest, lowest, most, least, greatest, smallest. Comparators: more than, less than, X times more, twice as much, half as much. Always pair them with the reference point ("compared to non-smokers", "out of the four diseases").

Scatter Diagrams — Correlation Type

The spec requires you to use a scatter diagram to identify a correlation between two variables (MS 2g). Three correlation types to recognise:

Reading a Scatter Diagram

Positive correlation: as one variable increases, the other increases. Points trend from bottom-left to top-right.

Negative correlation: as one variable increases, the other decreases. Points trend from top-left to bottom-right.

No correlation: points scattered randomly — no consistent trend.

Strong vs Weak

A strong correlation has points tightly clustered around a line of best fit. A weak correlation has points more scattered, though a trend is still visible.

Sampling Principles

The spec requires you to understand the principles of sampling as applied to scientific data, including epidemiological data (MS 2d). Epidemiology is the study of how diseases spread and which factors affect them across populations — risk factor studies are based on samples, and sample design affects how much you can trust the results.

What Makes a Sample Reliable

Large sample size — more people = more reliable patterns; reduces effect of individual outliers.

Random selection — avoids bias (e.g. only studying healthy people would skew results).

Representative — sample should match the population (age, sex, ethnicity, lifestyle) you want to draw conclusions about.

Control group — for comparison (e.g. people without the risk factor) so you can isolate its effect.

Sample Size Trap

If a study uses only 50 people and finds smoking increases CHD risk, that's suggestive but not conclusive — small samples can show patterns by chance. A study with 500,000 people finding the same pattern is much more reliable. Exam questions sometimes give two studies with different sample sizes — comment on which is more reliable.

Incidence Varies By Location

The spec also asks you to explain how lifestyle factors affect the incidence (frequency) of non-communicable diseases at local, national, and global levels. For example: smoking rates differ between countries, so lung cancer incidence varies internationally. A high-fat diet is more common in some regions, so CHD incidence is higher there. Epidemiological studies compare these incidence rates to identify risk factors.

Human & Financial Costs of NCDs

The spec asks you to discuss the human and financial cost of non-communicable diseases to an individual, a local community, a nation, or globally (WS 1.4).

LevelHuman costFinancial cost

IndividualPain, reduced quality of life, premature death, dependence on othersMedical bills (if not covered), loss of earnings, treatment costs

Local communityFamily stress, carer burden, loss of community membersLocal NHS / care services costs, loss of workers

NationalReduced average life expectancy; inequalities between regionsNHS treatment costs, lost economic productivity, benefits paid

GlobalDisease burden in developing nations; preventable deaths worldwideWHO/international healthcare costs, lost global productivity

"Discuss" Questions

If asked to "discuss" the cost of NCDs, mention BOTH human AND financial costs, and at AT LEAST TWO different levels (e.g. individual + national). Single-level or single-type answers cap your marks.

Worked Example 1 — Comparing Risks

Disease% increase in risk of developing the disease

Disease W45%

Disease X180%

Disease Y22%

Disease Z95%

Compare the increased risk of developing the four diseases caused by smoking.

2 marks

POINT

"Compare" + table data = use superlatives. Identify the highest and lowest from the table.

EVIDENCE

Highest: Disease X (180%). Lowest: Disease Y (22%). The phrasing must use "risk of developing", not just "risk".

EXPLAIN

Model answer: "Smoking caused the highest increase in the risk of developing Disease X (180%) and the lowest increase in the risk of developing Disease Y (22%). The increase for Disease X is roughly 8 times larger than for Disease Y." ✓✓

Two paired comparisons (highest + lowest), each using superlative + specific value + correct phrase "risk of developing". Bonus credit-worthy detail: the "8 times" comparator quantifies the difference. The wrong version: "Disease X is the most dangerous" — that's a different claim about general danger, not about increased risk of developing the disease.

Worked Example 2 — Scatter Diagram

Scientists plotted the number of cigarettes smoked per day against lung cancer death rate (per 100,000 men) for a large sample.

The scatter diagram shows a strong positive correlation between cigarettes smoked per day and lung cancer death rate.

(a) Describe what the scatter diagram shows. 2 marks

(b) Can you conclude from this scatter diagram alone that smoking causes lung cancer? Explain your answer. 2 marks

PART A

Model answer: "There is a strong positive correlation — as the number of cigarettes smoked per day increases, the lung cancer death rate increases. The points lie close to a line of best fit." ✓✓

PART B

Model answer: "No — a scatter diagram only shows correlation, not causation. To prove smoking causes lung cancer you would also need to identify the causal mechanism — e.g. carcinogens in tar damage DNA in lung cells, leading to mutations and uncontrolled cell division." ✓✓

Part (a): names the correlation type ("strong positive"), describes the relationship in words, refers to data trend. Part (b): correctly distinguishes correlation from causation, then names what additional evidence would be needed (the mechanism). Saying "yes, the graph proves smoking causes cancer" scores zero — even though it's medically correct, a single scatter diagram does NOT prove causation.

Three Examiner Alerts

Alert 1 — "Risk of developing X" vs "Risk of X"

When the column header says "risk of developing [disease]", your answer must use that exact phrase. Writing "more dangerous" or "more likely to die from" describes something different. AQA mark schemes are strict on this — wrong phrase = no mark.

Alert 2 — Comparison needs comparative language

Stating two values doesn't compare them. Always include superlatives ("highest", "lowest") OR comparators ("twice as much", "X times more") AND a reference point ("out of the four diseases", "compared to non-smokers").

Alert 3 — Correlation ≠ Causation

A scatter diagram or correlated data shows a link, not a cause. To claim causation you need a biological mechanism. Saying "the graph proves X causes Y" without naming the mechanism scores zero on causation questions.

Common Errors — Wrong vs Right

Wrong

"Smoking is the most dangerous risk factor."

Right

"Smoking causes the highest increase in the risk of developing [named disease] — XX% compared to non-smokers."

Wrong

"The graph proves that smoking causes lung cancer."

Right

"The graph shows a strong positive correlation. To prove causation, the biological mechanism would also need to be identified."

Wrong

"Disease A is 40%, Disease B is 15%."

Right

"Disease A has the highest increased risk (40%); Disease B has the lowest (15%). A is more than twice as high as B."

Wrong

"The study only tested 50 people but the result is still valid."

Right

"A small sample size of 50 may not be representative. A larger sample would give more reliable results — patterns from small samples may occur by chance."

Wrong

"Non-communicable diseases cost money."

Right

"NCDs cause human costs (reduced quality of life, premature death) AND financial costs (NHS treatment, lost productivity) at individual, local, national and global levels."

Three Things To Remember

1. Correlation vs Causation

Correlation = two variables change together. Causation = one factor directly causes the other, proven by biological mechanism. Scatter diagrams show correlation; they don't prove causation. The spec phrase: "A causal mechanism has been proven for SOME risk factors, but not in others."

2. Read The Column Header

"Risk of developing X" ≠ "Risk of X" ≠ "Danger from X". Mark schemes credit only the exact phrasing in the data. Before answering any data question — identify the variable, units, and reference point.

3. Compare With Superlatives

"Compare" questions need highest / lowest / most / least OR X times more / twice as much — paired with a reference point. Two values stated side-by-side isn't a comparison; it's a description.

Next Lesson

Lesson 2.10 — Module 2 Masterclass. Bringing together everything from Module 2 (digestion, circulation, CHD/health, cancer, plants) into the 6-mark synoptic question style that AQA loves on Paper 1.

Self-Check Before Moving On

Try Without Looking

What's the difference between correlation and causation? (Correlation = link; causation = direct cause, proven by mechanism.)

What's the difference between "risk of developing CHD" and "risk of CHD"? ("Developing" specifies the outcome being measured — different from general "danger" or "death rate".)

What's needed in a compare-from-data answer? (Superlatives + values + reference point.)

Why might a small sample size be unreliable? (Patterns may occur by chance; not representative of population.)

Name two levels at which NCDs have costs. (Any two of: individual, local, national, global — and you need to mention BOTH human and financial.)

Module 2 — Almost Done

What You've Covered Across Module 2

2.01–2.04: Cells, tissues, organs, organ systems — and the digestive system in detail.

2.05: Enzymes and digestion.

2.06: Heart, blood vessels, blood — circulation anatomy.

2.07: CHD, health issues, lifestyle risk factors, cancer.

2.08: Plant tissues, transpiration, translocation.

2.09: Data interpretation skills for risk factors (this lesson).

Next: Lesson 2.10 — Module 2 Masterclass

The 6-mark synoptic-style questions that bring multiple topics together: heart disease + lifestyle factors + risk data; or plant transport + photosynthesis + osmosis. Practice on the question style that decides whether you finish Paper 1 at Grade 7 or Grade 9.