Data Analysis for Diet & Health
- A risk factor is any factor that increases the chance of developing a particular condition or disease
- An increase in a risk factor does not guarantee the development of disease, but increases the risk
- An example of a risk factor for the incidence of CVD is increased blood cholesterol
- An increase in the presence of a risk factor would be expected to lead to increased incidence of disease
- The incidence of disease describes the number of cases of a disease that occur within a particular group of people within a given time
Cholesterol
- Cholesterol is a type of lipid produced in the body
- Cells require a source of cholesterol for cell membrane functioning, sex hormone structure, and the synthesis of bile
- Cholesterol is transported around the body as lipoproteins
- Lipoproteins are molecules composed of lipid and protein
- There are two types of lipoproteins; high density lipoproteins (HDLs) and low density lipoproteins (LDLs)
- The balance of HDLs and LDLs in the blood is thought to be an important factor in the risk of developing heart disease
High density lipoproteins
- HDLs contain unsaturated fat, cholesterol, and protein
- These molecules transport cholesterol from body tissues to the liver to be recycled or excreted; they are responsible for reducing blood cholesterol levels when it is too high
- HDLs are also thought to contribute to the removal of cholesterol from the fatty plaques that form during atherosclerosis
Low density lipoproteins
- LDLs contain saturated fat, cholesterol, and protein
- The role of LDL is to move cholesterol from the liver into the bloodstream where it remains until it is required by the cells; they increase blood cholesterol levels when it is too low
- LDLs bind to receptors on cell surface membranes, enabling them to be taken up by the cells that need them and removing cholesterol from the blood
- High levels of LDLs can lead to blockage of these membrane receptors, causing blood cholesterol to rise
Lipoproteins and heart disease
- LDLs are often thought of as 'bad cholesterol'
- When your body has too much LDL cholesterol the membrane receptors on the surface of cells are blocked, leading to an overall rise in blood cholesterol
- Cholesterol contributes to the formation of plaques in the arteries
- High density lipoproteins are often thought of as 'good cholesterol'
- This is because HDLs carry cholesterol to the liver where it is broken down and excreted, therefore lowering overall blood cholesterol levels
- Rather than one being bad and the other good, in reality it is the ratio of LDLs to HDLs that is important; the healthy ratio is thought to be roughly 3:1 of LDL:HDL
- It is thought that an LDL:HDL ratio larger than 5:1 increases the risk of heart disease
- While diet is a factor in determining the production of LDLs and HDLs in the body, the way in which the cells process lipids is also thought to be important
- It is important to be able to analyse data on the link between lipoproteins and CVD
Evaluating data linked to risk factors
- There are several things that should be taken into consideration when analysing data from studies on health risk factors; if, for example, a study is conducted on the effects of smoking on lung health, then the following should be kept in mind
- Sample size; a study with a large sample size will give results that are more likely to be representative of the target population
- Individuals in the sample; the results of a study can only be applied to parts of the population represented in the sample, e.g.
- If the sample only contains women aged 20-40 then the results cannot be used to draw conclusions about the effect of smoking on males or older women
- Control group; a study of this nature needs to contain a control group so that the effects of the variable of interest can be isolated from the effects of other factors
- In this example the control group would contain individuals who do not smoke; this gives a base line against which the results from the smoker group can be compared
- All factors other than the variable of interest should be controlled between the control group and the test group, e.g. age, diet, activity levels
- Statistical significance; the differences between groups need to be sufficiently large to be sure that they are not just the result of random chance; this is known as being statistically significant
- To determine whether a difference between results is significantly different, statistical analysis needs to be carried out
- Statistical significance can often be determined by comparing standard deviations of two data sets
- If standard deviations overlap then the difference between data sets is not statistically significant
- If the standard deviations do not overlap then the difference is statistically significant
- The influence of other variables; variables other than the variable of interest need to be controlled so that they do not impact the results; in this study such control variables might include
- Genetics
- Secondary exposure to smoking
- Exercise levels
- Diet
Worked example
The graph below shows different lipoproteins with their associated cardiovascular disease rate. Note that the bars represent the standard deviations of the data sets.
The study involved 300 men aged between 40 and 80. Their LDL and HDL levels were measured over a period of 30 weeks and mapped against developing CVD events such as atheromas and thrombosis
- Describe the data
- Draw a conclusion
- Comment on the validity of the data. Note that validity refers to a study that successfully measures the variable that it sets out to measure, without influence from any other factors.
1. Describe the data
The graph shows that the cardiovascular event rate percentage is higher with LDLs compared to HDLs
The cardiovascular event rate with LDLs is around 28 % while the event rate percentage for HDLs is around 18 %
2. Draw a conclusion
There is no significant difference in percentage of CVD events with LDL and HDL as there is an overlap of the standard deviation bars
3. Comment on the validity of the data
A sample of 300 men is very small in relation to an entire population, so needs to be larger in order to be representative
Only men between 40 and 80 were sampled, so the data is not representative of the entire population
Other factors relating to CVD are not measured or commented on, e.g. smoking, diet, exercise; such factors would need to be controlled between the two groups or measured so that their effect can be eliminated from the data
Examiner Tip
When dealing with data there are a few important things to remember
- It is always a good idea to include numbers when you are describing a data set
- Correlation is not causation; when drawing conclusions it is always best to say 'there is a correlation' or 'there is an association' between two variables rather than stating that one has caused the other
- Look for clues about the validity of the experiment. Is the sample representative? Have other variables been controlled? Has statistical analysis been carried out?