(8) Evaluating data, statistics, graphs and correlation for diseases e.g. risk factors for non-communicable diseases

A correlation is a relationship between two random sets of data e.g. set A and set B for two variables.

A correlation does not automatically mean that A may affect B, B may affect A or both A and B are affected by a third variable.

 

Medical research scientists have to decide:

(i) whether links between possible causes and effects are valid, and

(ii) whether the link between treatments and cures is actually valid,

and this all about real correlation and just pure coincidence or a third unknown factor.

 

There is also the question of data collection and its evaluation and analysis.

(1) How reliable is the data and compared with an appropriate control group e.g. as in drug testing.

(2) How valid is the data - was it correctly measured and collected without bias? Where the results repeatable?

(3) How big was the sample size? Was it big enough for clear statistical significance?

(4) If repeated, how could a survey or set of experiment trials be improved?

(5) Where there any anomalies in the data? Can they be explained? Are they significant i.e. is the some unknown factor influencing the results in some way? Was sufficient time allowed for other possible effects to be clearly seen?

(6) What is the confidence of the conclusion? Have you proved the relationship between two variables?

(7) Could there be a bias in the interpretation of the results - vested personal or economic interests? I'm afraid this can happen in the pharmaceutical industry.

Graph 2 A negative correlation graph

Again, despite the scatter of points (X), quite clearly, on average, an increase in variable B follows an decrease in variable A. The graph has a negative gradient - don't take the word negative to mean no correlation!

Graph 3 A no correlation graph

No particular pattern or correlation - just a scattered set of points (X).

There is no clear trend of A influencing B or B influencing A.

 

Graph 4 example

If variable A is time, years of the 20th century, the variable B could be quantity of cigarettes smoked (blue line) and the number of lung cancer cases (purple line). Lung cancer takes many years to develop and there is a 20 year lag between the two sets of data. This is good evidence for the connection between the non-communicable disease of lung cancer and smoking.

Graph 5 example

Linear relationships, but one (blue line) increasing proportionately greater than the other (purple line) - steeper gradient observed.

Graph 6 example

Correlation of smoking and lung cancer. A = time, B blue = amount of smoking and B purple = lung cancer deaths.

This has some similarities with graph 4, but in this case, although there is a lag, some other factor has intervened to bring both trends down.

Hopefully, this is what will happen in the future. At the moment the two curves have just dropped a bit from their peak values.

There has been a world-wide health campaign against smoking and some countries ban the advertising of cigarettes and it does seem to be having some effect.

Apparently during the Covid-19 corona flue pandemic in 2020. large numbers of people stopped smoking for fear it would make the more susceptible to this virulent and potentially deadly virus.

They hope, and quite correctly, it will help their lings to healthier, they are also more reluctant to go out to shops to buy cigarettes, others have lost their jobs and need to cut down on expenses.

Doctors in China have evidence that survival rates of non-smokers exceed that of smokers.

All in all, 2020 is a good time to stop smoking! and hope the two curves continue to fall.

 

Graph 7 example

Variable A could represent time in years. The purple B line could represent the number of people vaccinated in a population - increasing coverage. The blue B line could represent the fall in cases of the disease being vaccinated against - increasing immunisation of a population.

Graph 8 example

Graph 8 is a typical graph for when a new vaccine is introduced to combat some disease.

A represents time in years. B represents the incidence of the disease (infection) in a given population.

Once the vaccine is developed and the immunisation programme instigated, the effect can be quite dramatic.

This happened in1940s with the development of a vaccine against diphtheria, producing a rapid decline in the number of cases.

Diphtheria is an acute and highly contagious bacterial disease causing inflammation of the mucous membranes, formation of a false membrane in the throat which hinders breathing and swallowing, and potentially fatal heart and nerve damage by a bacterial toxin in the blood. It is now rare in developed countries owing to immunization.

 

There other similar patterns for communicable diseases e.g. measles is an infectious viral disease causing fever and a red rash, typically occurring in childhood. This has been greatly minimised by the MMR vaccination of young children which is very effective at protecting people against measles, mumps, and rubella, and preventing the complications caused by these diseases.

Graph 9 examples

Graph 9 represents a trend that rises, reaches a peak and falls again to a similar start value.

If variable A represents a year in time, the blue graph line could represent the peak of some infection cases e.g. salmonella food poisoning tends reach a peak in the summer months. Warmer temperatures encourage the growth of salmonella bacteria in food causing the disease. Careless cooking and food storage add to the effect.

---

Key points Source of information is based on textbooks & syllabus-specifications for students taking the AQA GCSE, Edexcel GCSE and OCR GCSE level biology examinations (~US grades 9-10).

Notes for biology students on evaluating data, statistics, graphs, and correlation in the context of non-communicable diseases (NCDs) like cancer, smoking-related illnesses, alcohol-related liver disease, and more.

---

Evaluating Data, Statistics, Graphs & Correlation in Human Biology

Why Stats Matter

Understanding how to interpret and evaluate data is essential in biology. It helps scientists:

• Identify risk factors for diseases.

• Understand correlations and causal relationships.

• Inform public health policies.

• Evaluate the effectiveness of treatments.

---

Types of Data Presentation

1. Tables

• Organize raw data clearly.

• Useful for comparing values (e.g. number of cancer cases in smokers versus non-smokers).

2. Bar Charts

• Show categorical data (e.g. deaths by disease type).

• Bars do not touch.

• Height of bars = frequency or number.

3. Histograms

• Show continuous data (e.g. age ranges of liver cirrhosis patients).

• Bars touch to show data ranges.

• Area of bar = frequency.

4. Line Graphs

• Show trends over time (e.g. smoking rates versus lung cancer cases over decades).

5. Scatter Graphs

• Show correlation between two variables.

• Each point = one data pair (e.g. alcohol consumption versus liver disease incidence).

---

Correlation versus Causation

Important: Correlation does not always mean causation. Scientists look for causal mechanisms to prove cause.

---

Evaluating Risk Factors for NCDs

1. Smoking

• Contains carcinogens (e.g. tar, benzene).

• Strong correlation with lung cancer, COPD, and heart disease.

• Causal mechanism: chemicals damage DNA and lung tissue.

2. Alcohol

• Excessive drinking damages liver cells.

• Leads to fatty liver, alcoholic hepatitis, and cirrhosis (scarring of liver tissue).

• Also increases risk of mouth, liver, and breast cancer.

3. Obesity

• Risk factor for type 2 diabetes, heart disease, and some cancers.

• Often evaluated using BMI (Body Mass Index).

4. Poor Diet

• High sugar/fat intake linked to diabetes and cardiovascular disease.

---

How to Evaluate Data in Exams

1. Look for patterns – Is there a trend? (e.g. more alcohol = more liver disease?)

2. Identify anomalies – Any outliers or unexpected results?

3. Use correct terminology – e.g. “positive correlation”, “causal link”, “risk factor”.

4. Interpret graphs – What does the slope or shape suggest?

5. Draw conclusions – Based on evidence, what can you infer?

---

Why This Is Important in Biology

• Helps understand how lifestyle affects health.

• Supports evidence-based medicine.

• Encourages critical thinking and scientific literacy.

• Prepares students to evaluate real-world health data.

---

Keywords, phrases and learning objectives for this part on medical statistics

For non-communicable diseases be able to analyse medical statistics data e.g. graphs correlating causes of medical conditions, negative and positive correlation shown by graphs, data reliability? and  valid conclusions.

Be able to analyse data for the for medical conditions such as obesity and diet, lung cancer and smoking, cancer factors, CVD factors

Given data, be able to assess the effectiveness of vaccination by looking at the incidence of infectious disease and the role of MMR vaccination for measles, mumps and rubella.

---

WHAT NEXT?

TOP OF PAGE

INDEX of notes on communicable diseases

INDEX of all my BIOLOGY NOTES

BIG website, try using the [SEARCH BOX], maybe quicker than the many indexes!

Basic Science Quizzes for UK KS3 science students aged ~12-14, ~US grades 6-8

Biology * Chemistry * Physics for UK GCSE level students aged ~14-16, ~US grades 9-10

Advanced Level Chemistry for pre-university age ~16-18 ~US grades 11-12, K12 Honors

Find your GCSE/IGCSE science course for more help links to all science revision notes

email doc brown - comments - query?

---

graphs correlating possible causes & effects of disease for syllabus-specifications for students taking the IGCSE/GCSE level biology examinations, summary revision notes key points on graphs correlating possible causes & effects of disease for students studying AQA igcse/gcse biology notes on graphs correlating possible causes & effects of disease, Edexcel gcse biology notes on graphs correlating possible causes & effects of disease,  OCR 21st century GCSE biology notes on graphs correlating possible causes & effects of disease, OCR gateway GCSE biology notes on graphs correlating possible causes & effects of disease, WJEC gcse biology notes on graphs correlating possible causes & effects of disease, CCEA gcse biology notes on graphs correlating possible causes & effects of disease, CIE Cambridge igcse biology, notes on graphs correlating possible causes & effects of disease useful for US grade 9-10 biology student courses, Explaining importance of examples of different types graphs of medical statistics explained & interpreted in GCSE level biology, What you need to know about examples of different types graphs of medical statistics explained & interpreted for GCSE level biology, Explaining use of examples of different types graphs of medical statistics explained & interpreted knowledge in GCSE level biology, Examples of examples of different types graphs of medical statistics explained & interpreted explained when studying GCSE level biology, What is the significance of examples of different types graphs of medical statistics explained & interpreted in GCSE level biology, describing explaining theory of examples of different types graphs of medical statistics explained & interpreted when studying GCSE level biology, exam revision notes for examples of different types graphs of medical statistics explained & interpreted, online help for understanding examples of different types graphs of medical statistics explained & interpreted in GCSE biology, what do I need to learn about examples of different types graphs of medical statistics explained & interpreted? what do I need to know about examples of different types graphs of medical statistics explained & interpreted for GCSE biology exams, how to prepare for questions on examples of different types graphs of medical statistics explained & interpreted in GCSE biology examination?

---