Dev-Picayune

It appears that Quicken no longer is supporting importing it’s really sorry QIF format (a format that I have written simple parsers for a couple of times). It seems that the real emerging and increasingly more accepted standard is OFX. Since it is XML based (now) and more institutions support it, things are getting more standardized. But what I immediately thought of in the midst of researching a little about OFX (since I may try to support it for my San Antonio project), is how badly we need a standard for Diabetes monitoring.

As it stands, every pump and meter seems to have it’s own standard for an interface back to the computer (and that is to be expected). Every doctor’s office has a standard for how they want the information from a patient to be presented to them. What we really need is an XML-based standard for diabetes statistics. The nice thing is that since you can specify elements as optional you can collect standardized information that contains as little or as much as you need or want. For instance, a diabetes meter should by default have the ability (either natively or via its custom software) to create an XML file that conforms to this standard that gives all the readings from the meter. A simple insulin pump might produce data pertaining to just doses given (and what ratios and other data was related to the doses). And a food diary program might supply food intake data. Eventually, you’ll end up seeing a merging of some of this data. Already there are pumps that either contain meters or interface directly with meters and also allow you to input the carbohydrates taken in for a meal. In this instance the pump (or pump software) should be producing a lot of the data we’re talking about here.

The real beauty of the standard, though would be that doctors, diabetes researchers, diabetes-related software and hardware creators, patients, hospitals, etc… would all have a common standard for this information. For instance, today in Grayson’s endocrinologist office they have a chart that you fill out and either you call in the doses and meter readings or you fax in the chart. This is completely prone to errors, misunderstandings, and is entirely paper based. Imagine if you could just hook your meter or pump up to your computer or network cable or wireless network or to a telephone or something and the data could be sent to the office in a standard XML format which of course could be formatted and displayed for the doctor’s in any format they wanted. Since it would be a true standard, it wouldn’t matter which brand of pump or meter the person used (they should all end up supporting it). Tons of time could be saved, long-term records could be electronically stored. Doctors could call in from home and look at the numbers electronically. Office staff and after-hours support could have a subset of the numbers sent to the doctor’s phone or PDA.

In fact, you could expand the standard to be 2-way. You could have programming changes for a pump (ratios and insulin absorption algorithms) sent back down to the patients pump. The possibilities are limitless. If you pitched the standard correctly, the manufacturers would race to make their products support the new standard (most would be able to quickly enable downloading in the new format from their product’s software). Now, if I could figure out how to get everyone together to sit down and come up with the standard and the framework for revising it.

As the scope of products grows, this is only going to get more important. The next big thing should be continuous blood sugar monitoring. This seems to me to be an extremely important step towards proper care. This standard would accomodate this too by easily allowing the thousands of records (or a subset) to be sent within the file. A lot of analytical study could be done to adjust doses, ratios, pumps, etc… based on this data to more accurately similate normal insulin production and aborption.