August 19, 2007

...and I would like a side order of Holy Grail while I am at it too

One of my quests is for a diabetes management software application that is worth using. I would like the cure and the Holy Grail while I am at it too.


They seem just about as likely.


One look at the marketing materials and you kind of just know it isn’t going to work out. In Of Independence and Angry Mobs I wrote about the doctor centered mentality of diabetes equipment manufacturers. I think that mind set is one of the reasons we can’t find this Grail.

So what would this holy grail of software do?

It would help families adapt to varying diabetes by:
Tracking for reporting all the data users create in their daily diabetes care without creating a lot more sets of family tasks. We have diabetes – that has enough tasks as it is.

All the BG data, including CGM info. All the basal data. All the carbs entered into the pump to calculate all the bolus data. Any user defined flags on those carbs like pizza. User defined variables that could be things like heavy exercise, set pulled off, ketones, freaky weather, stress, menstrual cycle, weekday, weekend and what ever else people dream up. The key here is being user definable.

Then we need reporting ability and the ability to include or exclude data based on those use defined fields. Build reports that average the midnight to 6 am BD data from meters or CGM for nights following a day with hard exercise. There you have info for tailoring basals for a post activity dip in BG.

How about sorting the data for a school girl’s weekdays, excluding days with PE class to build a normal school day profile and one only for PE days?

Maybe we can make some sense out of Pizza.

The key is to get all that data into one application, WITH some user defined life style information, and then slice and dice to provide intelligent reports for decision making. The doctor doesn’t have the time to do it and we won’t either without some decent tools.

The reports need more than breakfast lunch dinner night time periods. I vote for 24 hourly time slots on the log report with numbers and a graph. But hey why not let the users pick what works for them? 4 may work for some folks 12 for other I want 24. It is software; variables are part of the deal.
Reports should be able to show more than one day’s data, show average days in a range or average user defined days. Toss in some standard deviations too.

Those of us who use a bolus calculation in a pump are putting in a ton of information. Things are moving in the direction of saving and using that data but from what I have seen that movement is in baby steps. Compared to the price of a pump memory chips are dirt cheap and small, load them into the pump. Make the pump talk with the CGM and all of it interface with a simple to use but user controllable desktop reporting engine.

If it was real cool it would let you save or track one data set and use it as a benchmark for another set of data to let the user test to see before and after adjustment. Maybe we can have a few ‘Wizards’ to help calculate ISF, I:C and a partridge in a pear tree.



Since we are talking about a a pipe dream here, all of this should be based on industry standard data structures so that patients (and their doctors) can choose any combination of meter, pump that suites individual needs. As opposed hampering patient freedom and health for proprietary gains.