I think that Mark Amundson's business will provide this service for those who wish to purchase it.
The whole question of drug interactions is an interesting one. There are databases one can purchase, but when it comes to implementing them, this is an issue which virtually all medical records systems in large hospitals, particularly teaching hospitals, wrestle with. The severity of the interaction is rated by the company distributing the database, but I gather the majority of the time there is not even close to unanimity of opinion about the advisability of using their rating system to generate an alert since the severity of the interaction does not always match the clinical importance of it. Since most systems that pop up an alert will expect an entry to explain the over-ride, they don't want to harass their physicians with marginally helpful alerts or alerts they do not feel rise to the level of necessitating this sort of added interaction. Generally there is a process established where folks can recommend alerts and then there are one or more levels of review committees set up to determine what will go into the record system and also accept feedback and evaluate how well they work. This is true for things other than drug interactions as well, such as reminders. As far as I know, the VA does its own interaction testing and own entries which involve an entry and a double check by a different person to be sure the orderable item file is correct. I don't know it for sure, but I suspect there are committees like the ones I mentioned above at the VA. This whole question is a big one for the reasons Kevin has already cited and especially because of VOE. VOE will go out the door with a drugs and interactions, but how will this be maintained? It was populated with an existing database, but will that continue? I am not aware of any plans to see that it will or of attempts being made to seek the cooperation of the VA in doing it. That does not mean it hasn't happened. I just have not heard about it. So I would guess that Mark's business prospects should be quite good. Let's hope that he prices his services reasonably. On Saturday 25 March 2006 08:36, Kevin Toppenberg wrote: I have a question for the pharmacists on the message board: As I am processing the 280,000 entries in the FDA database, I am matching first by NDC. If that fails, I then have a drug that potentially needs to be added to the VA PRODUCT file. In order for the drug to work with drug-interaction testing later, the drug INGREDIENTS of the new addtion need to match with existing VistA ingredients. But here lies my problem. A great many drugs fail to match ingredients. Here is one example: FDA: ceftazidime pentahydride VistA: ceftazidime In this example, I feel fairly comfortable matching these. But if I make a rule that says, "if simple comparison doesn't match, then allow matching based on FIRST word only", then I worry that I will create a match between two ingredients that shouldn't be matched. So my question is: Is this worry valid? Are there ingredients where the second or third words DO MAKE A DIFFERENCE? Next, there are going to be new substances/ingredients that have been released by the FDA that have not previously been put into the VA PRODUCT file, or the INGREDIENTS file. For example, Spiriva (tiotropium), I don't know if this is in my VA PRODUCT file or not, but let's assume it is new enough that it is not already there. To add this as a new drug, I will need to add a new ingredient, TIOTROPIUM. But then what about the DRUG INTERACTIONS functionality? I haven't studied this part yet, but I assume there is a file somewhere that lists drug interactions and ingredients that cause them. If I don't have drug interaction data for new drugs, but do have it for old drugs, then providers will have a system that only works some of the time --> dangerous. In the VA system, I assume that new drugs are added one at a time by a pharmacist, on an as-needed basis. Question: Where do they get their information about drug interactions? There must be some database that they access, which should be continually updated. If so, is this a commercial product, or is it public and I could download it somewhere? I guess another option would be to simply add the new ingredients and then shut off the drug interaction system so that no warnings are generated (rather than just some of the time), and we would have the same situation as we do now, where doctors write prescriptions on paper and the pharmacist at the pharmacy fills the order (and *might* do some interaction checking). This is not a bad solution... Perhaps my work to automatically add missing drugs from the FDA database has been doomed from the start because there is not enough data there. (Other missing data is the maximum and minum doses, though I think our site can get by without this.) But frankly, the pharmacy system can't work if the doctors can't order ANY drug. And I feel I have established that we are not going to be able to tap into the up-to-date VA PRODUCT file maintained by the VA somewhere. And even this will not likely contain ALL the drugs (i.e. pediatric drugs, birth control tablets, labor & delivery drugs etc etc.) Your thoughts would be appreciated Kevin ------------------------------------------------------- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid0944&bid$1720&dat1642 _______________________________________________ Hardhats-members mailing list Hardhats-members@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/hardhats-members -- Nancy Anthracite ------------------------------------------------------- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642 _______________________________________________ Hardhats-members mailing list Hardhats-members@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/hardhats-members
