correlating risk factors

• Non-communicable disease


  Diseases not spread from person to person, instead caused by risk factors
• In the 1930s, rate of lung cancer began to increase sharply and scientists couldn't explain this
• Scientists could not carry out experiments on humans to try to work out what causes lung cancer as that would be unethical
• Epidemiology


  Studying the patterns of disease to determine risk factors
• Lung cancer is much more common among cigarette smokers than among non-smokers


  Scientists look at how many cigarettes people smoked each day and then how many of these people developed lung cancer
• Correlation


  A link between two variables
• A correlation does not prove cause, it simply suggests that two variables might be linked
• As the number of cigarettes smoked per day increases

  The risk of developing lung cancer also increases
• Positive correlation


  When two variables increase or decrease together
• As the number of years a person smoked increases

  The risk of developing lung cancer also increases
• Causal mechanism


  The scientific explanation for how one variable causes another
• Cigarette smoke contains chemicals which damage DNA and increase the risk of cancer, called carcinogens
• Sampling


  Investigating a group of people to draw conclusions about the whole population
• Ideally, we'd look at every single person in a population, but in practice it's not possible, so we sample a group of people instead
• To avoid bias, we need to take as large a sample as possible and it must be as random as possible
• We cannot draw conclusions from a small or non-random sample