Environmental Toxicology Diary

Prevalence and incidence rates as already discussed in previous post measure the frequency of occurence of observable effect due to exposure of suspected causative agent (etiological factor). However, as these rates are only used to provide information regarding the chance, possibility of occurence of a response, effect because of an etiological factor. But this risk, relative risk and attributable risk, can measure the probability and new cases expected by comparing exposed and non-exposed group to an etiological factor, thus will be able to determine the risk ratio and predicted case.

Relative Risk
Relative risk is a measure of probability that an individual will develop a symptom or effect when exposed to a specific etiological factor compare to a non-exposed individual. In this case, suppose you have a case, you would like to measure the chance of getting lung cancer of smoker compare to non-smoking person, then you can do a small survey. First, make a two groups, one is smoking individuals, another is non-smoking individual. During a certain period, measure individuals who develop lung cancer in a smoker group, and count the number of people as well who develop the same effect in a non-smoker group. Since you already know about prevalence and incidence rate, you can now measure the incidence rate.

I will make an example, this is not based on real example at all instead just make things esier to understand. Suppose in a 1,000 smokers, there are 10 people develop lung cancer and in 1,000 non-smoking people, there are 2 individuals develop lung cancer, that would make incidence rate is 0.01 and 0.002. Relative risk is a ratio between individuals develop lung cancer in smoking individuals and individuals who develop lung cancer in a non-smoking group. I think you have already got the answer. The ratio would be 10/2, equals 5. This means, smoking will increase the chance of getting lung cancer as many as 5 times more dangerous. You can do this simple survey for both retrospective and prospective study.

Attributable Risk
Attributable risk is a risk which a new case is expected if an individual is exposed to a specific etiological factor. This kind of risk measures how many individuals develop a particular effect due to exposure towards an etiological factor. Let us continue above example to make clear understanding about the difference of these risks.

We know from above that there are 10 individuals smoker who develop lung cancer, while on the other hand, there are 2 individuals non-smoking groupd who have lung cancer symptom. Both groups are each having 1,000 individuals being sampled. Attributable risk can be measured by substracting number individuals develop lung cancer in the smoker group with individuals with the disease in a non-smoking groupd. That would be 10-2, equal 8 individuals. What this means is, it is expected that 8 new cases will arise of individuals who develop lung cancer if they smoke.

Attributable risk is useful to be used as part of decision making, for example in jewellery factory when it usually employs many loud device that might hurt workers ears, the managers may have to decide whether it is necessary to equip an expensive earring protective from sound or not. Thus a simple experiment can be conducted to measure how many workers have less sensitivity of their ear compare to workers who have same symptoms in different place which is not having loud noise. Suppose in a loud noise place, from 100 people there are 3 workers got trouble of their hearing and from another quiet place from 100 people there are 2 workers who have the same effect, thus 3-2 is 1 worker is expected to develop trouble for his or her ears. This may not feasible and is not justified the cost needed to add procedure and ear protecting device for workers.

Bookmark to: