Independent Variable: The dosage of the lipopolysaccharide (LPS).
This is the variable that is being changed by the experimenters; it will be given to the subjects in order to test for a reaction.
Dependent Variable: The levels of Inflammation.
This is the variable that is being measured; it would change as a result of the LPS dosage provided to the subjects.
Experimental Design
Groups
There will be 5 groups, in which 0 mg (control, given as a placebo), 2.5 mg, 5.0 mg, 7.5 mg, and 10 mg of the LPS will be given to each group respectively.
5 groups will be used in order to minimize the cost of the experiment as much as possible. The dosages of LPS will each increase by 2.5 mg in order to maintain a constant drug dose increase and to produce reliable and usable data (as opposed to increasing the dosage by a random amount). By keeping the dosages low, it is possible to maximize the results while minimizing cost.
Age
The age range for this experiment will be 65 years old and over.
Number of subjects per group
There will be 15 subjects in each group.
The subject size was chosen in order to keep the costs of administering the LPS low, while also having a large enough sample size to produce sufficient data.
Subject Selection
Since this experiment is testing the levels of Inflammotin in the elderly, the age range for this experiment will be 65 years old and over. The subjects must consist of a mix of genders, races, and social classes in order to eliminate any subject bias. In order to select our subjects, a randomized selection will be taken from retirement facilities, assisted living, and hospitals. Once the subjects are chosen, they will be medically profiled to see if any of them have a high risk for inflammation. Those that do will be evenly dispersed among the 5 groups to reduce the chance of experimental error.
Sources of Error and Bias
1. Prior medical history may impact the protein levels regardless of the LPS.
All subjects will be medically profiled prior to the experiment in order to identify if there is any medical history to be aware of. Rather than exclude these subjects from the experiment, they will be divided equally between the 5 groups based on health (i.e. prior/current inflammatory disease). Including these subjects will provide a better sample of the elderly population, in which a number of the population has an unhealthy medical history. Dividing them evenly between the groups ensures that all groups have a similar ratio of healthy subjects to unhealthy subjects.
2. Taking subjects from hospitals, retirement facilities, etc. in specific cities/areas with high poverty may result in more subjects with poor health care which could produce skewed results.
The selection of subjects will be randomized in an attempt to control these sources of error. This can be done by broadening the area of selection, such as taking the subjects from multiple cities with different economic rankings.
3. The group size is not large enough to produce an accurate portrayal of the results.
While this could potentially be easily fixed by adding more subjects to the groups, because this experiment is financially constrained, this option is not possible. Instead, subjects will be chosen as randomly as possible in order to get the most accurate and unbiased portrayal of the population as is possible within the confines of the 15 person test groups.
4. The experiment may be biased because the experimenter may lean towards using the least amount of LPS to make it more cost-effective.
While this error may be hard to control, it can be attempted by setting up a team of multiple experimenters as a way to create a system of checks and balances so a single experimenter will not be able to sway the data.
BME 100 Fall 2015
