Self-report studies are known for sometimes being inaccurate, chiefly because too much responsibility is placed on the respondents. Oftentimes people forget, overestimate, underestimate or do a host of other things that result in partially or wholly inaccurate reporting. There's also something known as The Testing Effect; a situation wherein people answer a question a particular way because they feel it is the answer the test-giver is looking for. So, in sum, self-report studies aren't really the best source for data collection. I mention all of this because it was a self-report study that researchers recently used to link symptoms of depression with the development of diabetes in older adults.

Researchers from Northwestern University studied a group of over 4,600 participants who were at least 65 years of age or older and did not have diabetes at the outset of the experiment in 1989. Each year for ten years, the participants were evaluated for symptoms of depression, based on the participants' response to survey questions. Also recorded each year were clinical measures that would indicate the development of diabetes.

The researchers found that as the participants reported feelings of depression as the years went on, an increase in participants who developed diabetes occurred.

Now, aside from the problems related to the self-report study, I also wonder how the researchers can be so sure that the depression lead to the diabetes. In fact, it seems to make more sense (at least to me) that the diabetes led to the depression. A person contracting a life-altering disease such as diabetes certainly wouldn't add joy to their life, so it stands to reason that it would instead make them a bit -- if not very -- depressed. I'm no scientist, but this correlation seems to make much more sense to me.