Why They Fail The Control Test
Don’t blame yourself for all those bad readings if your meter continues to omit data analysis.
Although blood glucose meters have improved tremendously in ease of use, most people with diabetes don’t realize how helpful their meters could be in their effort to control blood sugars. Meter companies have not really made a concerted effort to present blood sugar data in ways that would really be useful.
Today’s meter business generates billions of dollars in sales and results in hundreds of millions of dollars in profits from insurance and health care charges. Meters are viewed as revenue generation devices, not as devices for real BG control by meter companies, and to a large extent by users as well. Meter companies are on a profitable treadmill of giving away meters, selling test strips at a large profit, giving away meters, selling strips at a large profit, giving away meters… Yet little is done to make this technology improve health and health care outcomes.
Meter company advocates say that health care outcomes are not their job, and, actually, they are correct. So if health care outcomes mean getting blood sugars under control, whose job is it? It is you and your physician’s job. But your job is made much harder today because the information given by your meter is presented so ineffectively that few consumers or health providers use it. Your blood sugar readings are presented as a disjointed, meaningless series of numbers that are not connected to the things affecting it: food, exercise, insulin injections, or medications. No attempt is made to reveal patterns that can be used to change therapy and improve your blood sugars.
As the ability to do blood sugar monitoring on a continuous basis nears, it is critical that this central problem of data presentation be recognized and fixed. Think how great it would be if you and your doctor could look at the numbers stored in your meter’s memory, receive meaningful advice about changes that could be made to improve control, and use this information to actually change your insulin doses or other therapy?
EKGs Analyze Data, Why Not Your Meter?
The way in which data is collected by meters is nearly useless for identifying blood sugar patterns and making meaningful therapeutic interventions. Barry Ginsberg of Becton Dickinson Technologies refers to the problem of identifying patterns in blood sugars as a classic Point-Vector Problem, similar to the one seen in the early days of EKGs. He means that most people, physicians included, look at the points or trees in an EKG, rather than the vector or forest. In EKGs’ early days when only the EKG was recorded, few physicians could correctly identify EKG patterns. And if physicians could not do this, they could not separate an inferior wall MI from a normal heartbeat.
EKG companies recognized this problem many years ago. Rather than having to look for a deep Q-S segment and inverted T wave in leads 3 and AVF to identify a heart attack in the bottom wall of the heart, a doctor can simply look at the computer analysis written on the EKG printout, which says in plain English “inferior wall MI”. This enables the physician to make quick and appropriate clinical decisions. Although a doctor would want to verify the EKG data, this quick analysis of the pattern contained in the EKG can literally save someone’s life.
In a similar fashion, identifying patterns in blood sugars comes from looking at the whole picture, not at individual readings. Currently identifying ways to change therapy for better readings can only occur after someone, either you or your doctor sees these patterns in your blood sugars. Wouldn’t it be easier, if instead of presenting the jumble of numbers, your meter wrote out: “You are having too many low blood sugars. Discuss immediately with your physician.” or “After the last 3 low blood sugars, your next reading is averaging 425 mg/dl. Have you considered you may be over-treating these lows?”
Currently, pattern analysis is done by a few individual users or doctors. But, on the whole, most of the data collected from today’s meters, at a great out-of-pocket expense to you, does not lead to any real change in therapy. How your meter company is presenting your data to you is a major handicap in your search for control. Let’s look at how this could change if your meter company simply decided to provide you and your doctor with useful information.
|Problems With Current Meters:|
|Blood sugar, time of day, memory of same, and global averages (ie, your blood sugar averaged 150 mg/dl over the last 14 days).||This is classic point analysis where individual data points are presented rather than a larger picture.|
Even the “average blood sugar” has limited use, ie, an average of 150 mg/dl can be found in one person swinging wildly between 30 and 300, and in another person who is almost always between 140 and 160.
|How Tomorrow’s Meters Could Deliver:|
|Average BG & standard deviation before and after each meal||Provides much more detailed information, such as whether your breakfast reading tends to run higher or lower than your lunch, and how much the blood sugar varies at different times of the day.|
|Link glucose reading to real events (premeal, postmeal, exercise, stress)||Lets you know if your reading was before or after a meal, and can be linked to particular meals, i.e, “You were 367 mg/dl after your Krispy Kreme breakfast, well above your normal average of 156 after breakfast.” Let’s you know to eat just one next time.|
|Frequency and severity of lows & highs in last week||“Six readings below 40 mg/dl in the last week” lets you and your doctor quickly recognize that an adjustment is needed in the diet, insulin doses, or sulfonylurea medication.|
|Give time of day when lows or highs are happening||Speeds up insulin, diet and medication adjustments|
|Show if BG patterns are improving or worsening, trends over any selected time interval||If you are having fewer readings or more readings above or below your target range this week, wouldn’t it be nice to know this?|
|Inform whether current BGs suggest over or under treatment||How many times have you gone into your doctor’s office after 3 or 6 months of 200+ readings (or 60- readings) and have them say “Have you considered using more (or less) insulin?” Wouldn’t it be more convenient if your meter lets you know every week or two that your readings are above or below the target range you and your doctor have selected?|
|Show if highs follow lows or lows follow highs||Lets you and your doctor quickly realize you may be over-treating lows with carbs or injecting too much insulin for your highs.|
|Show how much residual Humalog is active from recent doses||Let’s say you took 10 units of Humalog for a late dinner at 8 p.m., and your meter reads 100 mg/dl at 10 p.m. on your way to bed. You may be thinking “Wow, what a great reading”, but your meter says “Warning: 4 units of residual Humalog have not yet acted on your blood sugar.” Wouldn’t this advice be helpful in avoiding the nighttime low that is otherwise likely?|
|Do 500 Rule and 1800 Rule calculations for you||This is a suggestion from Jose Riberio who hates to manually calculate the carb coverage on his pump that his doctor has given him. Instead, he’d like to enter the grams of carb he desires to eat in a meal into his meter, and let his meter do the calculation for him based on his carb ratio. And if his blood sugar is high or low, he also wants a device to work out a suggested insulin dose to cover his high blood sugar, based on his anticipated point drop given to him by his physician.|
|Avoid placing repeated BGs done in a short period of time into the average on meter.||Current meters average all blood sugars. Prevents enterprising teenagers and others from doing repeated consecutive tests during lows to improve their “average” blood sugar. It also lets you test repeatedly during a high or low, when you want to track readings more carefully, without it creating a discouragingly high overall average.|
Current Meter Company Mindset Is A Problem
Why haven’t meter companies provided you with easy ways to identify your blood sugar patterns and problems? Largely because they are making so much money for so little effort. Meter companies rightly claim they cannot give medical advice (always a good excuse for inactivity), but don’t be fooled. Presenting data in a meaningful format is NOT medical advice. Presenting blood sugar data in clearer ways that allow you and your doctor to decide on what changes you need to make to improve your readings is not medical advice.
Blood sugar patterns are present within your meter’s data, and these patterns are obvious to those who have an eye for spotting them and are willing to dig through the data. Bless your physician if he or she is dedicated and does this for you. Wouldn’t this job be easier if your meter actually helped to identify unwanted patterns?
The best meters today that collect extra data from which patterns and suggestions could be developed have problems — the One Touch Profile is a brutal system for recording carb intake or insulin doses as “events” tied to your blood sugar, and no one uses this feature because of the difficulty. The Accuchek Complete meter does a better job of data analysis but does no recording of events. These meters were never really designed for data collection. No current meter does analysis in any meaningful way.
Excuses, such as “no one wants it”, allow meter companies to keep reasonable data collection and analysis as an unreachable goal. Both users and insurers have a vested interest in improving health through fast and accurate analysis of blood sugar patterns, carb intake, insulin doses, and other factors that impact health. Meter companies need to be encouraged to create well designed, user-friendly systems that actually improve control today, as well as health outcomes later. Meter users and clinicians would love to have data collection and blood sugar analysis, but when meter companies do not provide an easy way to do it and provide no meaningful way to analyze the results, few are going to take on this task on their own, similar to physicians in the early days of EKGs.
Meter companies also claim that data analysis is too complicated for people with diabetes. This is certainly true for most people who use today’s meters. But why make it harder, when analysis of blood sugars could be done in a small, dedicated computer system? Wouldn’t it be nicer to have your meter tell you, “In the last 7 days, we find your average blood sugar is 246 mg/dl and your lowest reading was 137 mg/dl? Please discuss this pattern of high readings with your physician to identify any reason for this rise and whether additional insulin or medication is needed to reduce your risk of complications.” or “In the last 3 days, your average pre-lunch blood sugar has risen to 217 mg/dl from the previous 132 mg/dl. Has your breakfast or something else changed?” You could stop ordering two muffins at the new Starbucks near work and get just one, or discuss with your physician how to increase your breakfast Humalog for the extra carbs.
Easy data collection and blood sugar analysis can now be done. Presenting blood sugar data in new ways will take some planning and redesign of meters, plus attention to user-friendly designs not present in any of today’s BG recording devices. It will require the use of a relational database or a remote connection to a relational data server like IBM’s DB2 or Oracle, a portable operating system such as Linux (which has relational databases available), a keyboard for data entry into the meter or an accompanying Handspring/Palm type device (could even be tied into today’s cell phone technology), with associated XML or Bluetooth or IR or Java enhancements. All of the technology necessary for blood sugar data collection and analysis is available today.
Incentive For Meter Companies To Change
Most people with diabetes do not test as often as their doctors recommend. Half of the people with Type 1 diabetes, where testing is critical, apparently do not test at all. Among those with Type 2 diabetes, the numbers are even higher. And why should they test if neither they nor their doctor can make sense of the numbers without a great deal of effort? The numbers of test strips sold would likely skyrocket if people saw that they were being helped by testing.
Which meter would you rather use? One like your meter today, or one that can tell you in plain English (or French, Spanish, Japanese, German…) what your blood sugars are doing so you could improve them?
If improvements in data management are begun now, it will ease the transition into continuous monitoring where tremendous amounts of blood sugar data will suddenly become available. Even with blood sugars being tested every 5 to 15 minutes, this improved frequency will have little impact if only a few can make sense of the data. Meter companies have kept below the radar of the FDA, insurance payers, and users. They make lots of profit but have had relatively little impact on improving the health of most of their customers.
Because of tremendous profits from a device that impacts millions of lives, meter companies should be held to a simple outcome standard, such as “What is the average HbA1c of 100 random users of brand X meter?” Random users could be obtained by the FDA from strip prescription data randomly selected from pharmacies. These current outcomes would be published and available to everyone, especially users.
Each manufacturer would then be given a reasonable time, such as 18 months, to demonstrate a real improvement in the average HbA1c of these same users through improvements in meter design and data presentation. Let users choose which system is delivering on a reasonable goal: “Does this device really help me improve my blood sugars?”
Demand an easier path to good control! Call your meter company now. Request “enhanced data management” from the manufacturer of every meter you own:
- Lifescan (One Touch, Profile, FastTake, SureStep, Ultra) at (800) 227-8862
- Roche (Accu-Chek Advantage, Complete, Easy) at (800) 440-3638
- Medisense (Precision QID, Xtra, Card Sensor) at (800) 527-3339
- Bayer (Dex, Elite) at (800) 445-5901
- Therasense (Freestyle) at (888) 522-5226
- MiniMed (CGMS, a new continuous monitoring system) at (800) MINIMED
Call the FDA’s Division of Clinical Lab Devices directly at (301) 594-3084 and request attention to this matter.