[Previous Months][Date Index][Thread Index][Join - Register][Login]
[Message Prev][Message Next][Thread Prev][Thread Next]

[IP] Fwd: NYTimes.com Article: A Guessing Game to Rally the Diabetic Child

Slow afternoon at work. Here's a message that I
sent to the professor whose sponsoring the 
development of the game. Heh heh heh.


Professor Pentland --

I read an article in the New York Times today 
about the game for diabetic children that youre 
designing with Vikram Sheel Kumar. I am an adult 
with type 1 diabetes, a medical writer, and a 
former designer of childrens educational software, 
and I have some concerns about the structure of 
the game. All of my comments are based on the 
description of the game in the Times article, 
which I realize may not be accurate or complete. 
If Im misunderstanding the structure and scope 
of the game, I apologize.

First, it sounds as though there is a basic flaw 
in game design: 

Since the child is only rewarded for accurate 
predictions that fall within the target range, 
she or he will be motivated to predict only numbers 
that fall within that range, whether or not that 
is likely. (To maximize points in the game, the 
obvious thing to do is always to predict a blood 
sugar in the middle of the target range.) This 
undermines the stated point of the game: to accu-
rately predict blood glucose levels. Among other 
things there is great utility in being able to 
predict that ones blood sugar will be high, since 
short-term high blood sugar does not usually produce 
symptoms (unlike low blood sugar).

Even if the child were honestly trying to predict 
his or her blood sugar, it would be hard to do so 
very accurately. Most peoples target ranges are 
fairly narrow, so to score higher a child would 
have to be able to predict the blood sugar within 
just a few points. In my experience, this is impossible. 
Yesterday, for example, my pre-lunch blood sugar was 
113. Today it was 95. Both were within my fasting 
target range, but in neither case could I have pre-
dicted what my blood sugar was, beyond saying that 
I was sure that it wasnt below 70, and I was fairly 
sure it was below 120. I certainly cant predict my 
blood sugar accurately to within 20 points, unless 
its in the hypoglycemic range. 

Second, the game seems to skate into some pretty 
dangerous emotional territory. 

I understand the desire not to "reward" out-of-range 
results, but by rewarding only target range predictions, 
the game would in essence be punishing children for 
having highs and lows. I think that most diabetes 
educators (and diabetics) would tell you that this 
is a bad idea. Diabetes educators these days spend a 
lot of time trying to convince kids that the only 
"bad" number is a number that you dont know about 
(which is why its important to check regularly). High 
and low blood sugars happen, and that its as important 
to learn to deal with them as it is to learn to prevent 
them. I realize that its not your intention to punish 
kids for not getting perfect numbers, but the game 
seems structured to do just that. Even in team compe-
titions it could be a problem: what would happen if 
one kid on the team was having poor control? Would 
the other kids on the team oust him or her? It sounds 
as though youre hoping to incorporate mutual support 
into the team portion of the game; I wonder, though, 
how that would mesh with the competitive aspect of 
the game.

The emphasis on predictability also concerns me. Control 
of diabetes is more art than science. Many things can 
affect diabetic control: illness, stress, and hormonal 
fluctuations (a notorious problem for exactly the target 
age group for this game). In addition, its been well 
documented that the action of long-acting insulins can 
vary greatly from day to day in the same person, causing 
unpredictable highs and lows. (As you may know, most type
1 diabetics take a combination of long- and short-acting
insulin, unless, like me, they use an insulin pump, which 
uses only short-acting insulin .) Im a well-educated, 
disciplined adult. I keep good records, I am careful about 
diet and exercise, Im a savvy pump user, and I check my 
blood sugar an average of 7 times a day. But even so, 
there are times when my blood sugar is high or low un-
expectedly. Thats just the way it is.

I think that a better game would focus on problem-solving 
rather than predictions. The game as described does very 
little to address the issue of how to achieve target-range 
blood sugars. It does very little to address the issue 
of how to deal with, or prevent, highs and lows. (The 
only mention of these issues is, by implication, the team 
games in which points are given for offering advice on 
strategy.) My experience in going to diabetes support 
groups and talking to other diabetics online is that 
90% of what we do is problem-solving: discussing ways 
to prevent or cope with highs and lows, and to deal with 
travel, celebrations, new relationships, social pressures, 
pregnancy, parenthood, long term complications, and so forth.

All of the above is in the spirit of constructive criticism. 
I hope that you find it useful in developing the game.


Janet Lafler
for HELP or to subscribe/unsubscribe, contact: HELP@insulin-pumpers.org
send a DONATION http://www.Insulin-Pumpers.org/donate.shtml