The Westgard Rules

FDA to ISO 15197: Not good enough

Posted by Sten Westgard, MS

Here's an interesting quotation from the recent FDA draft guideline on blood glucose meters:

"Sections A-F below indicate the types of device performance information that you should include in a 510(k) submission for a BGMS device. Although many manufacturers design their BGMS validation studies based on the International Standards Organizations document 15197, FDA believes that the criteria set forth in the ISO 15197 standard do not adequately protect patients using BGMS devices in professional settings, and does not recommend using these criteria for BGMS devices." [emphasis added]

Now, note that there is a difference between the ISO 15197 standard for glucose meters that came out in 2003 and the more recent one from 2013, which has tighter guidelines. Is the FDA saying even the most recent guideline isn't good enough?

You don't see that very often.

NY: Glucose Meters Going Off-Label?

Posted by Sten Westgard, MS

Back in 2010, FDA hosted a series of public meetings about problems with glucose meters. 4 years later, we're beginning to see significant repercussions.

The New York State Department of Health has sent a letter to Laboratoy Directors informing them that many common uses of glucose meters will henceforth be considered "off-label":

"To date, FDA has approved (or cleared) glucose meters for the quantitative
measurement of whole blood glucose for use by healthcare professionals or lay users
as an aid in monitoring the effectiveness of a diabetes control program....

"The FDA has reminded clinical laboratory regulatory agencies that “Intended Use” in the
package insert or product manual does not include diagnosis of, or screening for,
diabetes. Other off-label uses of glucose meters include the monitoring of glycemic
control of non-diabetic patients in hospitals; use on critically ill patients; and use in
health fairs and other community events to screen the public for diabetes.

"Laboratories that use glucose meters for purposes or in populations beyond the
Intended Use in the package insert or device manual are considered to be engaging in
off-label use. The Centers for Medicare and Medicaid Services (CMS) have instructed
us that the laboratory must be informed that in the event of such off-label use, the
glucose meter defaults to high complexity and the laboratory must meet the CLIA
requirements for high complexity testing. In New York State, this means that the testing
would require a New York State clinical laboratory permit in the category of Clinical
Chemistry and only personnel licensed by the New York State Education Department
would be eligible to perform the testing. In addition, the laboratory would be required to
fully establish the analytic and clinical performance specifications (i.e. validate) of all
such devices for any change from Intended Use in their patient population. Until such
requirements have been met, the use of glucose meters in health fairs, other community screening events, and/or critical care settings must be discontinued."

If a glucose meter goes "off-label" - it is no longer a waived device, but instead it becomes HIGH COMPLEXITY, which requires many more studies before it can be validated for use.

This is a very big deal, but since NY is one of the few states that has its own regulations, most other labs will be waiting to see what happens with CMS regulations, accreditation guidelines, etc.

One way to look at this: in the four year gap, manufacturers had the opportunity to improve analytical performance of glucose meters, but obviously chose not to. This is the consequence.

Stay tuned, this will be a bumpy ride!

 

Bio-Rad's Bold Advice on Risk QC

Posted by Sten Westgard, MS

As we noted earlier, the big news from Houston was that CMS will soon be issuing it's IQCP implementation regulations.

In advance of that release, Bio-Rad has issued a White Paper : Looking Ahead to Patient Risk Management.

So, how many controls does Bio-Rad want you to run?  More after the jump...

New Book: The Poor Lab's Guide to the Regulations

Posted by Sten Westgard, MS

Declare your Independence from the Fear of Inspection:

By Dr. Sharon S. Ehrmeyer, Ph.D.

In 9 concise chapters, Dr. Sharon S. Ehrmeyer, Ph.D., describes a common-sense path to compliance:

•  An overview of regulations from CLIA, CAP, TJC, and COLA
•  Proficiency Testing
•  Procedure Manuals
•  Method Verification, Calibration and Calibration Verification
•  Quality Control for the small laboratory
•  Quality Assurance
•  Personnel Competency Requirements
•  Inspection Tips, Tools for Self-preparedness
•  Point-of-Care testing requirements, EQC and the future

For decades, Dr. Ehrmeyer has been guiding laboratories in their compliance efforts. She speaks frequently on CLIA, CAP, The Joint Commission, and COLA. Her down-to-earth advice has made her a favorite at conferences around the world. Now all of her recommendations (and more) are available in this reference manual.

 More than 250 years ago, Founding Father Benjamin Franklin published "Poor's Richard's Almanack," a compilation of practical tips and words to the wise. His bestselling book spoke to the growing spirit of self-reliance and independence in America. Continuing in this tradition, the Poor Lab's Guide is an essential collection of pragmatic wisdom for the American laboratory.

Try the Guide
Find out more and Download the Foreword, Preface, TOC, Index and sample excerpts at
http://www.westgard.com/guide-sample.htm


Introductory Special:
 Save $15 off the cost of this manual with the online coupon: LABFREEDOM.
 Order at http://www.westgard.com/the-poor-labs-guide.htm store before July 31st to take advantage of this discount.

Pop Quiz: HbA1c quality

 Posted by Sten Westgard, MS

Time for a few questions!

I was just reviewing some of the poster abstracts from the 2011 AACB (That's Australasian Clinical Biochemists) 49th annual conference and came across three posters in a row that described new HbA1c methods.

See if you can tell which is which.

1. Match the method data (A,B,C) to the abstract's conclusion (X, Y or Z):

Data	Matches to?	Study conclusion
A. CV of 0.6%, bias at 6.5 of 3.5%		X. "The [method] gives acceptable assay precision."
B. CV of 3.5%, bias of 11.8%		Y. "Both the precision and bias are within the... allowable limits of performance."
C. CV of 2.6%, bias of 1.6%		Z. "[A]nalytical performance is good and satisfies essential performance criteria."

Want to see the answers? (And more questions?) After the jump...

More Error Rates at POC

Posted by Sten Westgard, MS

In the December 2011 issue of Point of Care journal, an interesting study was published:

Preanalytical Errors in Point-of-Care Testing: Auditing Error of Patient Identification in the Use of Blood Gas Analyzers, Natalie A Smith, David G Housley, Danielle B. Freedman, Point of Care, Volume 10: Number 4, December 2011.

The study looked at patient identification errors on a blood gas analyzer in various departments in a hospital. Bearing in mind that this is just one type of pre-analytical error, what do you think the rate was? Given around 100,000 tests, what would you guess as the number of defects?

An Advisory on Blood Glucose Meters

Posted by Sten Westgard, MS

[Hat-tip to the AACC Point-of-Care listserve, which first posted a notice about this article]

The Pennsylvania Patient Safety Advisory has a regular electronic newsletter highlighting new science and studies about healthcare safety. Their December 2011 issue has a particularly interesting article for laboratory testing:

Point-of-Care Technology: Glucose Meter's Role in Patient Care, Lea Anne Gardner, PhD, RN, Senior Patient Safety Analyst, Pennsylvania Patient Safety Authority.

This review examined more than 1,300 reports of glucose-meter problems from the Pennsyvlania reporting system database from 2004 to 2011. Of those reports, 71 near-miss or adverse event reports occurred. Most intriguing are the report excerpts directly quoted in the study. Of those reports, 72% of the near-miss or adverse events occurred with high-blood glucose results. That is, where the glucose meter had a sudden high value that may or may not have been reflective of the actual patient's clinical state. For example:

"A patient's blood sugar was checked using a [glucose meter]. The lunchtime result was 517. A [blood glucose test] was [immediately] retaken to check for accuracy, and the result was greater than 600. A blood [laboratory] test was conducted per protocol, and the [lab] glucose [result] was 136..."

What do you think happened next?

Error Rates at the POC

A new study in Clinical Chemistry investigated the errors rates for Point-of-Care (POC) devices:

Quality Error Rates in Point-of-Care Testing, O'Kane MJ, McManus P, McGowan N, Lynch PL, Clin Chem 2011 Sep;57(9):1267-1.

Can you guess what the error rates were?

Glucose meter vs. Lab result: What would you do?

Posted by Sten Westgard, MS

Here's a scenario for you:

Dec. 21, 2010: Patient admitted

Jan. 1, 2011: Physician orders patient to be treated for high blood sugar.

Jan. 2, 11 pm: Physician orders insulin drip to be increased.

Jan. 3: Beside tests show elevated blood sugar; lab tests show very low level

Jan. 3, 4 am: [You become aware of the testing discrepancy.]

What would you do?

Don't forget the performance of the HbA1c assay

Posted by Sten Westgard, MS

Glycated hemoglobin is becoming a hotter topic in medical laboratories. Not only is there a movement to use this test as a way to diagnosis diabetes, but there is also increasing scrutiny of assay performance by National Glycohemoglobin Standardization Program (NGSP).

While NGSP is tightening the requirements for performance, the use of the test seems to be outpacing the performance of the assays. A recent letter to the editor highlights this issue:

Glycated haemoglobin A1c (HbA1c) in the diagnosis of diabetes mellitus: don't forget the performance of the HbA1c assay, RK Schindheim, E Lenters-Westra and RJ Slingerland, Diabet Med. 27, 1214-1215 (2010)

Some excerpts of that letter:

"...healthcare professionals should be aware of the inherent analytical variability of the HbA1c laboratory assays. Although the analytical performance has improved over the last 5 years, all methods still do not concur with the primary reference methods, which may contribute to bias in epidemiological studies and lead to misclassification of patients in the diagnosis of diabetes mellitus. Indeeed, the results of the College of American Pathologists 2009 GH2-B Survey Data on HbA1c...demonstrated that only two of 25 methods have an inter-laboratory coefficient of variation of <2% with the sample with a reference value of 6.6%....the analytical bias ranged from -0.19 to 0.27% HbA1c....

"The clinical biochemist can play a valuable role in this matter and should be encouraged to use HbA1c methods with optimal analytical performance (no bias and a total coefficient of variation of <2%). Given the insufficient analytical performance of most HbA1c point-of-care testing devices, caution should be exercised when applying these devices in the diagnosis of diabetes mellitus."

A short article with an important point. The potential of the HbA1c assay is great, but only if we assure that we've got the quality we need first.