Monday, March 22, 2010

Importance of Accuracy

Accuracy has been a long-discussed topic in the Diabetes Community, but most recently it's been the blogging rage because of an FDA meeting on blood glucose meter accuracy. Last week, several fellow D-Bloggers posted their thoughts on this topic - KellyK, Kerri, Samantha, and ScottS to name a few.

Of course, I'd be remiss if not to mention Jim Huck, who has an incredibly funny post on the topic. Being a math teacher, he poses this scenario:

Johnny is a type 1 diabetic who tests his blood sugar at 6pm and discovers that he is 240. He had eaten 3 slices of Round Table Pizza at 5PM (when his BG was 118), which had 72 grams of carbohydrates and 30 grams of protein. He had already taken 4 units of Humalog at 4:45 PM, and his normal insulin to carb ratio is 1/13. Since the pizza has a lot of fat, Johnny expects that it will take about 3 hours for the pizza to finish digesting.

Taking into account that his meter has a 20 percent margin of error, the food label has a 20 percent margin or error, his insulin dose at 5PM could have been off by as much as 39 percent, and now that his blood glucose level is up at240 mg/dl, his correction ratio could be off by as much as 30 percent.

The question: How much beer will Johnny need to drink to forget that he has diabetes? Round your answer to the nearest six pack.
Outstanding, Jim. Simply AWEsome.

Anyhow, on to the FDA meeting that spawned this Accuracy Talk...

This two-day meeting happened March 16-17. To get at what criteria should be changed "in order to promote greater public health," the FDA sought input from physicians, nurses, healthcare providers, industry representatives, diabetes educators, professional societies, consumers and even patient advocate groups.

In covering this event, Reuters wrote it like this: "At issue is whether incorrect meter readings stem from problems with the technology or other issues such as patients or doctors and nurses not using the devices properly. Other problems can include trouble with test strips, blood samples, hand washing and other factors that can impact results."

Those Operator-Error excuses seem to be what the industry is hanging it's hat on in why inaccuracies happen, outside the mentality that any 20% deviation is allowable. A similar theme was expressed at another recent FDA meeting, where Insulin Pump Failures were explored and manufacturers and industry-insiders indicated that it's not their fault mistakes happen; it's the patients not doing something properly.

I take offense at both of these messages, and it makes me seriously wonder if the FDA is getting the full, accurate picture on all of this? Consider this statement by industry consultant Barry Ginsberg, (an MD & Ph.D who heads the New Jersey-based firm Diabetes Consultants: "How much accuracy you need depends on who you are. Those with Type 2 diabetes who treat their condition with diet changes and oral drugs don't need to monitor their blood levels as closely as those with type 1 diabetes who take insulin."

Are you serious???? Accuracy only depends on who you are? Wow.

In reflecting on this issue, I began pondering the luxury of being able to be inaccurate as much as 20% percent of the time, or being permitted to be "off" by that much at any given time....
  • Being a journalist, accuracy is my top priority. If I was allowed to always maintain a 20% error margin, this would mean that names and titles would be misspelled, and facts would simply be wrong. This doesn't instill public confidence, and I'm sure that the FDA meeting leader wouldn't appreciate having their names and job descriptions butchered.
  • As a responsible member of society, I pay taxes. Yet, I'm sure that the IRS wouldn't be pleased to receive up to 20% less of their owed money each year. Same goes for state and local income taxes, as well as property taxes.
  • Driving is a large part of my life - in the most daily of routines such as commuting to and from work. A necessary part of this means obeying traffic laws, staying in my own lane of traffic, and not running red lights or traveling at a pacer higher than the posted speed limit. Next time this doesn't happen, and the police decide to pull me over, I'll go with the 20% error margin excuse: that I was within that range above the limit, or that the specific traffic law I violated wasn't part of the 80% being obeyed that day.
  • Being on time is important and I must meet deadlines in both my personal and professional lives. When a meeting begins at 8 a.m., I must be there at that time - not 8:10, or 8:20. On Time. Or else the meeting starts without me, I get into trouble, or something else consequential occurs.\
  • As a Type 1 diabetic, I take insulin. Calculating is extremely important. If I take too much, my blood sugar levels drop. That could happen at bedtime, resulting in dangerous lows. If that happens, I might not wake up. The paramedics might have to be called, resulting in your tax dollars being spent on me and my medical emergencies rather than on your company's new driveway entrance that you've requested be constructed from the local municipal board. Nope, sorry. You don't get that business benefit because your inferior product is contributing to the tax dollars being spent on something else. Thanks to your inaccurate blood test meter readings and my subsequent Lows. Our Loss, and Yours.

  • However, this 20% error margin isn't the world we live in - except maybe when you're talking Blood Glucose Meters. These above examples may seem like an Apples & Oranges Comparison, but I don't see it that way. We all must live with accuracy and standards in this world, but some are not expected to be held to that same standard on a topic that literally means Life & Death for so many People With Diabetes. As well as those non-diabetics, who feel the ripple effects of our lives in numerous ways.
In my Quarter-Century of Living with Diabetes, I've seen mostly-cosmetic advances in BG Monitors - meters have gotten fancier, smaller, and come in multiple colors and sizes. But they're no more accurate than back in the 80s.

We must do better. Now. Not a decade or two from now, when it's more financially feasible or companies feel better about it. We're long-overdue for this improved accuracy.

Raise Your Voice on this important issue. Submit a public comment the FDA prior to the April 20, 2010 deadline. Here's how:
Go to the Regulations.gov page on Blood Glucose Meter Accuracy, which can be found by the Docket ID "FDA-2009-N-0604.” Questions about posted comments can be directed to the Dockets Management Public Room at (301) 827-6860.

For reference, here's my FDA comment:
I am a Type 1 diabetic who was diagnosed at the age of 5. At age 31, I represent the 10% or less who have this chronic condition form of diabetes. Regular blood glocuse testing 8-12 times a day is necessary to effectively and adequately manage my condition, so that it doesn't result in higher costs for prescriptions and complications that is most likely with inadequate control. When I began testing at home in the 1980s, the standard for error margin was roughly 20%. Despite our advances in diabetes management and technology in the quarter century since then, that standard remains a reality today. This inaccuracy could mean the difference for someone taking too much insulin for food they eat, resulting in a Low bloog glucose level that impacts their driving ability or puts them into a coma. Results that are actually higher than displayed also means that People With Diabetes are not able to accurately correct with insulin and bring their blood glucose levels down to regular ranges, meaning they stay at these elevated levels longer and ultimately are impacting their long-term health. Overall, this drives up the cost to society: police and ambulance services needed, any hospital services that might be needed, and the bigger cost for prescriptions and medical care needed as a result of unstable blood sugars. To classify this inaccuracy as "no big deal" is simply irresponsible, particularly at a time when we as a nation are discussing overall health care costs and reforms for the insurance industry. We must do better.

3 comments:

Wendy said...

GREAT POST!

Anonymous said...

This is a great blog. I also sent them a letter.

Another thing people can do is, as they go to expos and meetings where they run into reps from meter companies, be sure to ask about their meter. When they tell you, "It's pink and it's smaller!" ask them, "Does it have any better stats than plus or minus 20%?" When they say "No" ask "WHY NOT?" Word will get to the corporate honchos if enough people voice their displeasure, and maybe something will start to be done. Mom

jpnairn said...

Great post.
I commented on my blog about the fact that the FDA says that we are using the meters for tight glycemic control, and that's not what they're made for. Seriously.

http://t1d-runner.blogspot.com/2010/03/glucose-meter-accuracy-and-fda.html