The problem certainly lies in the way the research is done. You are not finding differences in results displayed depending on the brand of the browser. You are encountering differences depending on who is using the system.
It's even possible to have different data being presented in different browsers, but then it shows that your pages are more complex than they should be and are poorly tested. I'm not saying that you have this problem, I just say it is possible, although highly unlikely. If you think it might be this, that there is something on the page that might be hiding data in IE, open a specific question and show your code that you think is doing this. There is no way to analyze without seeing code. But reinforcement that is unlikely to have any problem.
But answering your question objectively incompatibility between the database and any browser not because they even talk to each other.
Either way you can test this easily. Use IE and see if anything changes. And / or have your clients test with another browser (I know this is a bit trickier to do depending on the client) and the result will be the same. That is, I would discard any problem with the browser, but try to be sure, there is no better way to find out.
Lastly you may still be curious about the problem. If there is really a problem and there is no error of interpretation of the data by the clients or by you, the error is probably in the server-side application. Probably under certain circumstances it recovers the data one way or another.
Can the browser influence this? Yes, it may, but again it is highly unlikely. Your application would need to treat the data differently depending on the information of which browser the user is using being captured through JavaScript and sent to the server. But all this is speculation and if it were this, you would know.