EPI7-2:STUDY DESIGN - Observational And Experimental Studies

OBJECTIVES

At the end of this session you should:
  • understand the differences between experimental studies and observational studies. Occupational epidemiology is most concerned with the latter.

OBSERVATIONAL AND EXPERIMENTAL STUDIES:

Another axis of classification of studies by their design relates to whether they are experimental in the allocation of exposures or observational.

Most but not all studies in occupational epidemiology are observational studies where people are observed over time to determine if exposures of interest are associated with outcomes of interest.

For example, it is not ethical to expose people to substances that are known to be hazardous to determine if there is an effect with time. If it were, one could expose workers in the same way that laboratory animals are to known doses of say asbestos or silica and then measure their respiratory function over time.

There are instances however where such experimental studies may be ethical. An example would be to study the effect of fatigue on driving skills by working with volunteer drivers who agree to be deprived of sleep and to have their performance measured on a simulator where their mistakes are recorded. In this way it has been possible to express sleep deprivation in hours in blood alcohol equivalents and show in an ethical experimental manner that it is hazardous.

The most familiar form of experimental study is the randomised controlled trial of a drug or medical intervention. These are rarely performed in occupational health although there is need for them. One example would be a drug trial for winter influenza which could be ethically performed by randomly allocating the vaccine to some workers and a placebo to another group.

NOTE: The rest of this course is concerned with observational studies.

It can easily be seen that there are links between cohort or follow up studies comparing exposed and uenxposed workers and experimental designs, where the exposure is randomly allocated. Such random allocation helps to eliminate confounding by unmeasured factors. Confounding will be dealt with later in the course.

More generally most types of study design are linked. The next page deals with the relationships between study design in a simpler way. Those who are interested can page on and find a more complex approach to study design in the ensuing pages and also in the reference material which is Pearce Textbook*, Chapter 2.

*Permission to use obtained




Creative Commons 

License
General Introduction to Occupational Health: Occupational Hygiene, Epidemiology & Biostatistics by Prof Jonny Myers is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 2.5 South Africa License
.