Back to Research News Research News and Information

2275 Swallow Hill Road - Building 2500 - Pittsburgh, PA 15220
(412) 429-0673 Fax: (412) 279-1367

March, 1996

An open letter to stockholders and diabetics:

On February 26, 1996, an FDA Advisory Panel reviewed the marketability of the Diasensor 1000 (TM), the world's first noninvasive glucose sensor developed by Biocontrol and the only noninvasive glucose sensor to ever be submitted to the FDA. Following this, much misleading, even outrightly erroneous, information about the research data and the panel review results appeared.

We'd like to get the record straight.

First, the Diasensor 1000 is not yet the perfect noninvasive glucose sensor. It is the world's first such technology, and as such will definitely need continued research and development. However, the research data submitted to the FDA more than adequately demonstrates that this sensor does safely and accurately measure glucose for a certain percentage of the diabetic population.

Why didn't the FDA Advisory Panel think so? There are many factors at play here, only one of which centered around the fact that the data was gathered and presented using the Clarke Error Grid graphical method of assessing clinical performance of blood glucose monitors. Devised in 1987, the method has been gaining acceptance by clinicians who believe it does a better job of evaluating clinical consequences of home-use blood glucose monitor errors than the conventional analytical methods which better represent laboratory reference devices. Many of the panel stated that they were not familiar enough with the Clarke Error Grid analysis method to be comfortable with it. To fairly evaluate Biocontrol's data submission, this error grid analysis method must be understood. The panel members should have been required to be familiar with it.

Second, eight patients were enough to gather sufficient data on the efficacy and safety of the Diasensor 1000 monitor. It was enough because for those eight patients, 263 data points (Diasensor reading compared to an industry standard reading) were submitted to the FDA - that's an average of 32 data points per patient. Currently used finger stick technology frims only submit an average of 1 data point per patient for devices they are attempting to get cleared. That means 100 data points submitted equals 100 patients studied. Therefore, 263 data points submitted for the Diasensor 1000 is equal to having tested 263 patients - a substantial test size.

Consider also that it should not matter to the FDA whether the Diasensor 1000 sensor can help 8, 200, 2000, or 2 million people. If it works for only one diabetic in the United States, that one person should be allowed to have the machine. In that regard, what makes this product different from an FDA-approved drug like Penicillin which definitely cannot help everyone yet it is allowed on the market to help those who can use it.

To get to the eight patients, BICO began with 85 original patients. Of those 85 patients, 22 were eliminated due to a detector (critical part of the machine) that failed at one test site and two were eliminated because their glucose levels did not vary sufficiently to calibrate the machine to them. Of the remaining 61 patients, 47 were successfully calibrated (necessary algorithms were obtained) and Biocontrol chose, and the FDA did not object, to follow 23 of these 47 patients to allow for a more rapid accumulation and reporting of the data. At the time, the Company did not have sufficient funds to purchase additional computers and software that would have allowed for rapid processing of the test data. What it can process in a day or two now took Biocontrol weeks to process at that time.

What BICO ended up with was a noninvasive sensor that worked well for eight of the 23 patients calibrated, verified, tested, and followed for 30 days. For 35% of the diabetics who could calibrate to the sensor, the machine did its job of measuring glucose noninvasively.

Third, how do we know it measured glucose and measured it effectively? Each patient's glucose levels were also measured by the industry standard YSI (Yellow Springs Instrument), which is, incidentally, the same measurement reference device used to test and compare currently used invasive monitors. Therefore, if these invasive monitors are held up as effectively measuring glucose by parallel comparison to the YSI, then it must be accepted that the Diasensor 1000 noninvasive sensor is also measuring glucose and measuring it comparative to these currently used monitors.

Fourth, following each patient for 30 days was considered entirely acceptable as the physician's manual submitted to the FDA requires the sensor to be brought back to the calibration centers every 30 days. Therefore, if the sensor functioned accurately and safely for that period of time, it would be rechecked and would then be used for another 30-day period.

Fifth, many features are built into the design and marketing plans of the Diasensor 1000 noninvasive sensor to insure patient safety in use. The device is to be available only by physician prescription. No one will buy the sensor then take it home and pray that it functions properly. When the device is prescribed, the patient will then be evaluated to determine calibration potential. If it is possible to calibrate to the patient, it will be done before the patient is permitted to take home the sensor. Most importantly, the Diasensor 1000 must initially be used concurrently with an invasive monitor and any inconsistencies by the Diasensor must be reported to the physician. Only after successful 30-day use in the home and recheck at the calibration center, will the patient purchase the machine. And even then, the pattern of 30-day checks will be continued.

Yes, the sensor at this time only gives a reading 50% of the time because its abilities to give a reading can be obstructed by hair, dirt, oil, etc. This will get better and keep in mind that the sensor doesn't give an erroneous report because of these interferences - it just doesn't read at all. When that happens, the patient can simply take another reading. Remember, for every ten tries, five readings will be obtained; for every 20 attempts, ten will be obtained, etc. Again, at this time, the Diasensor 1000 gives a reading 50% of the time it is asked to do so. Of course, if the patient wishes, an invasive monitor can be used if a reading is not obtained by the Diasensor 1000 sensor.

When the sensor does give a reading, it is as accurate as currently used sensors. We have the data to prove it. Actually, the Diasensor 1000 can be more accurate in the home setting than currently used invasive meters because there is very little technique needed and, therefore, much less chance for human error that causes inaccurate readings with the finger stick method. And many of you with diabetes are "painfully" aware that invasive glucose meters give erroneous readings, as several diabetics loudly pointed out to the members of the FDA Advisory Panel on February 26. This fact was driven home again recently by the father of a diabetic child, Ronald Brenners, who called the company to relate such an incident that just happened a few days ago. While Mr. Brenners' son, Ben, was playing baseball, his father checked his glucose and got a reading of 50. Since Ben didn't show any signs of hypoglycemia (low blood glucose), Mr. Brenners thought it wise to re-test him immediately. This time, with the same meter, he received a reading of 300! Evidently, FDA-approved invasive meters are not perfect either.

Sixth, it is true that the noninvasive sensor is not yet fully tested at the hypoglycemic range. That is not a safety issue, however. When initially calibrated and evaluated, a patient will be given his range of calibration. The patient will be instructed to use his finger stick method of evaluation if readings outside of his calibration range are given by the Diasensor 1000. In upcoming testing the Diasensor 1000 will be tested in the hypoglycemic range.

Lastly, the FDA did not reject the Diasensor 1000. Actually, to date, the FDA has not reacted to the Advisory Panel at all - the Company has not yet heard their decision or recommendations based on the Panel Review. We can assure you, however, even though we have reservations about this Panel Review, Biocontrol will do whatever is necessary to get the Diasensor 1000 noninvasive glucose sensor to market.

For every 100 people with diabetes, about 78 can be identified as potential users of the Diasensor 1000 and about 27 of those 78 will be "compatible" with the sensor and therefore able to purchase and use it. That's on the research model - the production model Diasensor 1000, on which no data was permitted by the FDA Advisory Panel, has demonstrated up to a 30-fold improvement in data collection efficiency in bench tests.

To put the above information in perspective, consider that the Juvenile Diabetes Foundation (JDF) estimates that worldwide there are about 110 million people with diabetes. By the testing on our research model, approximately 78% (or 85,800,000) of those people win be identified as potential users of the Diasensor 1000 and about 35% (or 30,030,000) of them will be "compatible" with the machine and able to purchase and use it. That is a lot of people who can be helped.

No, the world's first noninvasive glucose sensor is not perfect, but as a father of a diabetic recently put it, if we don't start somewhere, we won't get anywhere. The Diasensor 1000 can safely and accurately work for a percentage of the diabetic population NOW, while continued R & D makes it ever better.

Which of you with diabetes does not want the chance to see if you fall into that 35% for whom the Diasensor 1000 will eliminate many of those finger pricks? Also, when asked, 80-90% of you with diabetes or who care for someone with the disease, have emphatically said that a US$170 monthly payment for the sensor would not be too much and it doesn't matter if the sensor is not yet small enough to go everywhere. The message here seems to be that size and cost, in this case, are not the issues. Improvement to the quality of life is. Those of you who do not have your lives touched by diabetes, should-ask any person with diabetes or any parent of a child with diabetes about these issues.

Since Biocontrol developed the Diasensor 1000 noninvasive glucose sensor for them, the diabetics of the US should have a voice regarding its use. Many of them voiced their opinions on February 26th. They were not heard. Frankly, in my opinion, they are truly the ones that count.


Fred E. Cooper
Chief Executive Officer

  Back to Research News Return to the Top of This Page

Last Updated: Thursday February 27, 2014 19:28:21
This Internet site provides information of a general nature and is designed for educational purposes only. If you have any concerns about your own health or the health of your child, you should always consult with a physician or other health care professional.

This site is published by T-1 Today, Inc. (d/b/a Children with Diabetes), a 501c3 not-for-profit organization, which is responsible for its contents. Our mission is to provide education and support to families living with type 1 diabetes.
By using this site, you agree to our Terms of Use, Legal Notice, and Privacy Policy.
© Children with Diabetes, Inc. 1995-2018. Comments and Feedback.