POCT QA Bench marks

Hi Robert,

I am not sure of the National Benchmark, however it would be great to see how we compare.

Replying a year later, but hey, this is a relevant topic to my labs today:  to my knowledge, there are no National Benchmarks. However, many labs do have quality indicators that they track. I have my first full year of data and I am still trying to figure out what I have and what I want to achieve with it. 

I'd love to hear from folks who are collecting their data and it would be great to hear what they have achieved with it. 

I've not seen National Benchmarks either - probably reflects the notion of how POC in general is still relatively 'new' and also because many of the things typically used for lab QA don't really work for POC.  In my experience doing regulatory inspections, I've seen programs completely run the gamut from very detailed to barely anything, so it sure would be nice to see some benchmarks set.  I guess the advantage of not having any benchmarks  is it's hard to be 'wrong' with whatever you're doing, lol!

I include these to my POC reports:

1. Troponin turnaround time from Admit to Result

2. Patient ID error - preanalytical

3. Customer satisfaction- Physician complaints

4. Critical values for glucose - called to nurse in charge or physician

5. Competency audit  for non-waived tests compliance

6. Patient result audit from order to EMR- Connectivity audit

7. Comparison between POC and Main Lab

Can I get a clarification? 

Patient ID errors - are you asking about incorrectly entering the patient ID and it posts to the wrong patient's chart? Or those caught in the middleware system and there is no follow-up from the operator? 

The number of errors we have to correct because the wrong Patient ID was entered and it posted to the wrong chart, we average less than 10 per month with about a million test per year.  For those caught in the middleware without follow-up/correction from the person performing the testing it is about 20%.

Our official error rate is based on:

1. Result not entered in correct area in Cerner but found in provider’s notes.

2. Result not found in patient’s EMR at all

3. Result on patient log does not match EMR result

Our most recent rate based on total number of tests done is 3.0%

Rate based on number of audits performed (233) is 2.1%.

Sent via Groupsite Mobile.

I would like to know what rates others are getting as I have not found a national benchmark either.

Sent via Groupsite Mobile.

We are trying to get down to a zero error rate, but with over 320 staff performing POCT, its not really possible.

A staff member a year ago researched and found what she thought was a benchmark, that there was a 2.5% average error rate in Point of Care...but she did not share the info with the rest of us to know how that 2.5% was calculated/what metrics were used.

Our goal is less than 0.5% calculated error rate, with zero acceptable for QC errors (running patients without valid QC, wrong lot number info, etc). We audit 25% of patient results and 100% of QC info each month, then count up number of errors for each dept site, and divide by number of patients/testsx4. We used x4 because we have 8 possible errors which could occur with each test. Our EMR lets us print out a monthly report of patient results and test counts. The monthly paper log audit takes 40-80 hours monthly.

1) Patients resulted without valid QC

2) Wrong lot info documented for patient

3) QC out of range and not troubleshot/documented

4) QC info missing or incorrect for any result

5) No date and initials on log for result, cross out or comment.

6) Log to EMR data mismatch

7) Patients found in EMR but not on log

8) Patients on log but not in EMR

I teach my team that we don't measure things just to measure - ain't nobody got time for that!! If you have a less than 1% error rate on something, time to stop measuring, pretty sure you are good on that quality element, lol!

I am constantly pushing for where the actual problems are, then measure that. If you have an electronic process in place between your device and your operator, where say for example the QC is locked out, and they pretty much can't do it wrong - then we are definitely NOT measuring that element. 

Conversely, if you are constantly getting calls about cracked or broken glucometers - we definitely need to be measuring that. How long it takes to fix, get a replacement, downtime, etc.

Some of my fast metrics: glucose meter strip waste from flow errors, and iSTAT cartridge waste due to any cartridge error - I use less than 5% as green, 5-10% yellow and greater than 10% is red and I'm calling the manufacturer. That's whole house, but I also look at it at the nursing unit level as well, and adjust for my small low volume areas.

Patient ID errors is set at 0-5 errors per month is green, 5-10 errors is yellow and greater than 10 is red and I am doing an intervention with that nursing department - retraining, etc.

We also track proficiency testing error rate, patient ID scan rate into the devices, and we do a whole house monthly rounding where they are scored a 2 if all items are labeled with expiration dates and within expiration, 1 if some are good and some are not, and 0 if items are expired. 

I like the way you think, Silka!

We use broken/lost items as our first indicator.....ISTAT and glucose. We keep track of them in the software. We do a variance report on all of these that do not have a mechanical fault. (Dropped the ISTAT on the floor...liquid in the meter port etc.) We charge the units for ISTAT repair.

Since I have 5 hospitals, we do variance reports on Transfers ID between hospitals. Patient A is transferred from South Hospital to North Hospital....and South Hospital ID was scanned at North Hospital. This is usually in only a few units (ICU and Tele) but it is an awareness education item benchmark. Cerner and UniPOC stop these.

Secondly, I am radical on patient ID errors. I had one operator that made 2 in one shift (using all 1111111) because of laziness. Upon research the patient had an ID...and that operator came out of the system. My procedure is written that it is mine and the pathologist decision based on the situation involved. Our HR process requires a counseling for this. The CNA/PCT job description requires meter access. We usually leave them out until the counseling is done by the director. We do not have to see the counseling, just be told it was done. We usually track this by operator....as it does not happen often. Any ID errors without operator follow up have a variance report done. We do about 15 a month on about 1M tests per year.

We used to track ISTAT errors for cartridges but found that it was unreliable and misleading. One person could do 170 in a month and have no errors, then do 5 the next month and have 4 errors. The patient turned out to have a coagulopathy in that example.  

Our corporate standards do not allow hand entry of any POCT result. Physician complaints is included in my IQCP surveillance.

Yep...I agree....Silka is REALLY smart!!! Did all of you know she presented in front of the FDA this year?

Does anyone know how to track iSTAT cartridge waste?  I thought we were giving out more boxes of cartridges than we got results for.  Come to find out that if the nurse thinks they messed up and overfilled/underfilled the cartridge, they are just throwing it out so they don't waste the 2 minutes to find out they have a check code.  Thanks.

You could track how many patient tests were performed and documented in your interface vs. how much inventory that each department has each month.