The problem is that you don't know how that data is used and who is using it.

Let's say your brother is feeling depressed, you are naturally concerned, and do research on treating depression and suicide prevention and ask for advice on appropriate forums.

Unsurprisingly, it turns out that these activities correlate very highly to people with depression and low productivity, An employer then uses this information and decides to hire someone else for the job. The loan officer at the bank decides that you are not a good credit risk right now because of your recent search history and declines your loan application. At the same time, your health insurance decides to raise your rates because you are now in a "higher risk profile."

You would not be able to contest or correct these decisions because you would never know why they were made.

Furthermore, the people making the decisions based on the data would not know or care wha specific actions resulted in your profile or score.