27 Oct Depression Medication

DSS 012Y Data Visualization for the Social Sciences Theory Challenge 1 Depression Medication: Human Trials Fournier, J. C., DeRubeis, R. J., Hollon, S. D., Dimidjian, S., Amsterdam, J. D., Shelton, R. C., & Fawcett, J. (2010). Antidepressant drug effects and depression severity: A patient-level meta-analysis. JAMA: Journal of the American Medical Association, 303 (1), 47-53. Background Do all people respond to anti-depressants in the same way? Successfully treating depression can sometimes feel like a gamble. What works for some people doesn’t seem to have any effect on others. Researchers are trying to understand what variables predict the success of a treatment. If they could understand what types of people respond to medication, then they could more accurately tailor their treatments. To make these relationships harder to detect, we need to Pill (Lucas Lucas/flickr/CC-BY-2.0) keep in mind that depression can be a transitory state and that humans show a placebo effect to anti-depressant treatments. As depression is a transitory state, over a few weeks, some people will feel better just "because." As there is a possible placebo effect, some people may actually feel better just because they think that they are being treated. Procedure They recruit 120 human participants with depression. All participants answer a depression inventory (survey). Each participant’s responses to the inventory are analyzed to determine if their depression is mild, moderate or severe. Participants are selected such Bazinga! looks like this: that there are equal numbers in each depression severity group. Half of the participants in each of those depression groups are randomly assigned to be given the anti-depression medication-"Bazinga!" -and the other half of The placebo looks like this: the participants are given a placebo pill. This pill is not a treatment for depression but will allow the researchers to see any transitory recovery and placebo effects. The medication or placebo is taken daily for sixteen weeks. Then the participants are brought back for a follow-up assessment. Each participant answers the same depression inventory and their responses are analyzed. This is compared to the initial assessment to determine whether or not the depression improved.

There are three variables in this data set. One is a predictor that is under the experimenter’s control – it is the different conditions to which the participants were assigned. One is a predictor that is a characteristic of the participants when they arrived. The other is the measured outcome.  List all variable relationships – but as there is only one outcome, only investigate those with that outcome. Some questions to consider: • Is Bazinga! more effective than a placebo? • Is Bazinga! equally effective for the different severities of depression? • Do we see the same pattern of improvement due to Bazinga! by the severity of depression?