I promise this is the last time...

[Malcolm Needs ☆]

however blood bank is the only department that routinely checks for completeness.

And therein lies one of the problems. Some other Departments have lower standards, and people tend to go for the "lowest common denominator".

I agree; stick to your guns.

:(:(

[Brenda K Hutson]

In the UK, we only accept hand-written specimen tubes for blood transfusion, unless the sticker is produced at the bedside (or wherever the specimen is drawn) and the sticker is produced by bar-code reading the patient's wristband.

This was following two papers that showed real problems with the use of pre-printed addressograph labels.

Cummins D, Sharp S, Vartanian M, Dawson D, Amin S, Halil O. The BCSH guideline on addressograph labels: experience at a cardiothoracic unit and findings of a telephone survey. Transfusion Medicine 2000; 10: 117-120.

Sharp S, Cummins D. The BCSH guideline on addressograph stickers: an update. Transfusion Medicine 2001; 11: 221.

:):)

Thanks for the references. I have to say, I am shocked that totally hand-written is preferred over a label that prints out (since I would also assume that a label that prints out, is using the same information that is in the database for that patient which created the armband and the order). So, I look forward to reading those.

Brenda

[Brenda K Hutson]

Brenda,

At our facility we encounter the same situation. We allow for the first name to be trucated but the last name, MR#, and DOB are required in addition to the phlebotomist intials, date, and time of collection. We will reject the specimen if any of this info is missing or illegible. Also, according to AABB standards the specimen sumitted for BB testing must contain three unique indentifiers; these do not change. Additionally, the same strictness with patient indentification becomes more lenient as patients become more distant from the hospital. Floor specimens require three unique indentifiers while outpatient specimens require first and last name with the exception of BB.

Hmmmm...the AABB Standards I am looking at (26th edition) states 2 identifiers; not 3??

Brenda

[LaraT23]

Funny that this thread is posted today! I had an incident yesterday in which a phlebotomist filled out a blood bank armband using the computer labels as a guide. She then asked a nurse to verify that she had filled it out correctly ( mind you she had not set eyes on the patient yet) and proceeded to enter the wrong room and band and draw another patient.

It is just so important being that we are the only department in the lab to dispense something that goes in to a patient that all identification be done by the book with no exceptions. I ask that the staff just write smaller ( ours are hand written) for longer names.

The patient listed above was A neg, but typed Opos at first. She had no previous history with us so there were no red flags. Now I have an incident report and retraining and a lovely blood type change error log. Great!!!

So after all that, yes do stick to your guns. We have gone one better and asked to use three identifiers, one more than required. Yes, we have had two Bob Jones with the same DOB and sex and even room number( not at the same time!), which is why I required name, DOB, and social security or medical record number.

[Eagle Eye]

We have hand written labels for all patient requiring transfusion. I am too scared to use printed label. Even though we need two people to verify patient before the draw I do get wrong spelling and wrong MR# etc. Before two signature we use to have WBIT.

[Malcolm Needs ☆]

Thanks for the references. I have to say, I am shocked that totally hand-written is preferred over a label that prints out (since I would also assume that a label that prints out, is using the same information that is in the database for that patient which created the armband and the order). So, I look forward to reading those.

Brenda

The reason is, if the labels are pre-printed, as opposed to printed at the draw, it is not unusual for another patient's pre-printed label to be stuck on to the sample tube.

Makes you go all cold doesn't it!!!!!!!!!!!!!!!!!!??????????????????

:eek::eek:

[rravkin@aol.com]

Brenda,

Alright, 2 identifiers. But I think the issue with hand writen BB specimens and pre-printed lables lies with the idea that there is some assurence that the specimen is being labled at the bed side or immediately after it is aquired. Correct me if I am wrong. But in reality there realy is no assurence that the specimen is being labled immediately after being drawn weather it is hand writen or using a pre-printed lable, outside of cameras maybe. And why is this labeling of the specimen so important immediately after the draw; because specimens get placed in other areas and can potentially get mixed up and mislabled as you know. I have recieved many hand writen lables and more often then not they are legible but still a good percentage are not. But the idea behind the hand writen lable is that it is being done at the bed side or immediately after the draw assuring that a misslabled specimen is not submitted to the BB. We practice at our facility the collection of a second specimen if there is no patient history and perform an ABO/RH. This practice gives us some assurence that the initial specimen was correctly labled along with other assurences. Does anyone know of any literature that studied the habits of personel handling the specimen and specifically labling it immediately after the draw? Personly, I prefer a printed lable; they are so much easier to read and check and there is no desyfering and if the first name is trucated then you are still left with some of the letters of the first name and other none-changing identifiers which, apparently only 2 are needed; all the better.

Edited March 17, 2010 by rravkin@aol.com

[Brenda K Hutson]

The reason is, if the labels are pre-printed, as opposed to printed at the draw, it is not unusual for another patient's pre-printed label to be stuck on to the sample tube.

Makes you go all cold doesn't it!!!!!!!!!!!!!!!!!!??????????????????

:eek::eek:

The system we previously used, printed the name labels out in the phlebotomy area and they then took them to the rooms. Problem was, sometimes it would insert a label from a 2nd patient into the middle of labels printing on another patient. That is an accident waiting to happen (and it did many times; though had they followed protocol and checked each specimen label against the armband......). Now they use a hand held barcode scnanner which I am told, prints out the labels at the bedside.

Brenda Hutson

[Malcolm Needs ☆]

The system we previously used, printed the name labels out in the phlebotomy area and they then took them to the rooms. Problem was, sometimes it would insert a label from a 2nd patient into the middle of labels printing on another patient. That is an accident waiting to happen (and it did many times; though had they followed protocol and checked each specimen label against the armband......). Now they use a hand held barcode scnanner which I am told, prints out the labels at the bedside.

Brenda Hutson

Ah, now those are the ones I mean we are quite happy to accept.

:D:D:D:D

[Mabel Adams]

Just some ammo for how "unique" first names are as identifiers. Maybe it will help you require all the letters of them on your tubes.

This is from the Social Security Baby Names website. (Reposted as an attachment since the table lost its formatting in the original post.)

What percentage of all names are represented in the top 1000 names?

For U.S. births in 2008, the top 1000 names represent about 74 percent of all names.

How unique is a first name.docx

Edited March 18, 2010 by Mabel Adams
to fix the table

[Brenda K Hutson]

Ah, now those are the ones I mean we are quite happy to accept.

:D:D:D:D

What, the barcode scanned labels? And if yes, would you then accept them if part of the first name was cut off (because that happened recently)

Thanks

[Brenda K Hutson]

Just some ammo for how "unique" first names are as identifiers. Maybe it will help you require all the letters of them on your tubes.

This is from the Social Security Baby Names website. (Reposted as an attachment since the table lost its formatting in the original post.)

What percentage of all names are represented in the top 1000 names?

For U.S. births in 2008, the top 1000 names represent about 74 percent of all names.

So couldn't they also then use that to prove "their" point; that how helpful (and therefore, significant) is the first name since so many people have the same first name (or did I miss your point)?

Thanks,

Brenda

[Malcolm Needs ☆]

What, the barcode scanned labels? And if yes, would you then accept them if part of the first name was cut off (because that happened recently)

Thanks

Yes, the barcode scanned labels, but we have to have written confirmation (once, not with every sample) that they are using this method of producing the labels before we will accept them.

Obviously, if the situation is dire (i.e. the patient is about to bleed to death, or the sample is irreplaceable - a pre-transfusion sample or a cord sample) we will allow addressograph labels (with medical concession), but, if a cross-match is concerned, we will only cross-match group O, HT-, rr blood (unless there is an anti-c, anti-e or anti-f present).

We would not accept them if part of the first name is missing (except in the dire situations given above) as it goes against both BCSH and NHSBT Guidelines. In such a case, the label would have to be hand written.

:redface:

[Griffin L]

We had a similer problem, were the first names were cut off on armband/patient labels. We found the problem lied in the program used to make these labels. The fix was to change the format of the labels so the last name prints on the first line and the first (and middle initial) print below it on the second line. While the problem was being fixed we insisted on both the billing # and the MRN along with the patients name be on all samples for blood bank. I find it regrettable that all areas of the lab are not required to use as strict patient ID as BB but the powers that be require samples for transfusion to be strictly labeled. Maybe you should find the Blood Bank standards that apply and take them to your meeting. I am called anal about proper identification and labeling of samples for blood bank use. Don't give in on this, it could cost a life!

[Brenda K Hutson]

We had a similer problem, were the first names were cut off on armband/patient labels. We found the problem lied in the program used to make these labels. The fix was to change the format of the labels so the last name prints on the first line and the first (and middle initial) print below it on the second line. While the problem was being fixed we insisted on both the billing # and the MRN along with the patients name be on all samples for blood bank. I find it regrettable that all areas of the lab are not required to use as strict patient ID as BB but the powers that be require samples for transfusion to be strictly labeled. Maybe you should find the Blood Bank standards that apply and take them to your meeting. I am called anal about proper identification and labeling of samples for blood bank use. Don't give in on this, it could cost a life!

Well, the Standards state 2 unique identifiers and that all information must be complete, legible and accurate.

However, that does leave a gray area. There are a number of things that could be used as a unique identifier

(ie. Name, MR#, Account#, D.O.B., Typenex Armband#) and each Institution has to decide for themselves which 2 (or more) they will require; and we have (Name and MR# for Inpatients and Name and D.O.B. for Outpatients). The gray area is this; what if something is missing from one of your Institutions required identifiers (not erroneous, but just incomplete) but they have other unique identifiers on the label (i.e. D.O.B. and/or Account#, and/or Armband#); do you accept it because it does have at least 2 "acceptable" identifiers, even though 1 of them might not be the one established in the SOP for your Institution, or do you dig your heals in and say NO, it must be complete and accurate for the identifiers in our SOPs, every time; no exceptions!

I too am described as you stated above in your next to last sentence; but am trying to look at things a little more open minded and be "reasonable" when the reality is, I have plenty of identifiers on that specimen to feel totally comfortable that it is the correct patient. And again, we are talking rare exceptions here, not a standard of practice; that I would not accept. But one of my recent rejects was an Outpatient who then had to drive back 1 hour for a redraw, all because there were some letters missing from the first name because the last name was really long; but there was still a D.O.B. and Red Armband Label.

Thanks,

Brenda Hutson, CLS(ASC)SBB

[adiescast]

I suppose you could modify your SOP to allow for a different unique identifier to be used in the case of one of your primary identifiers being incomplete?

[Brenda K Hutson]

So, I had another meeting with the 2 phlebotomy managers today (in response to one of you, yes, they are both former Techs.). I can't say I made much progress.

While it is possible to change the font on the labels, it would be costly (not for the hand-held lattice that phlebotomy uses, but for labels made by the Core Lab area computer system). Anyway, even if they would agree to that, doesn't sound like it is the short-term solution. As far as part of the name being cut off, their response: If you have a patient with 30 characters in the name and you got 23, you got "most of them; that should be good enough!"

And no one (including QA) liked the idea of hand writing the missing information.

I had not taken the time to really learn much about lattice and how it worked, so I asked to see, thinking that perhaps that would ease my mind. But here is my issue with that: when the phlebotomist scans the patient armband, the entire name of the patient can be viewed in lattice; however, the label that prints, by virtue of the way it was set up and the space limitations, may not have all of the letters. So, my thought then (and what I tried to counter) was that if you are supposed to compare the specimens to the Order and/or patient armband, it would not be a complete match (in some situations). Again, that was met with "most of it is there." They feel that the use of the lattice system is technology and a move in the forward direction, vs. having them hand write missing letters, etc.

At this point, after reading all of your responses (which I cut and paste as a handout for the meeting; left out a few words here and there when we made reference to the managers), I am not toally closed to keeping our SOPs with the defined 2 identifiers (name and MR# for Inpatients and name and D. O.B. for Outpatients), with the addendum that in those "rare" instances where the name was so long that it was truncated, the 2 identifiers for inpatients would then be the MR# and D.O.B. and for outpatients, the D.O.B. and red typenex band#. But hearing things like (23/30 letters should be good enough; the account# is another unique identifier, sex is an identifier; etc.) makes it harder for me to accept their "alternatives" because those comments alone show a lack of understanding of the nature and concerns of the Blood Bank. The Managers made the statement that prior to my coming 2 years ago, they were never told they had to do this (hand write the missing letters); I countered with "prior to my coming 2 years ago, people sent back incomplete and/or erroneously labeled specimens for correction to, and I would NEVER do those things).

Anyway, we agreed to meet again in 1 week. What I would like to hear more about from any of you who use barcoded armbands, are how you handle labels, and what if the label truncates, etc. In other words, does the barcode armband increase security and identification such that some of my stubborness is aniquated? Go ahead and lay it on me, I can take it....

Thanks

Brenda Hutson, CLS(ASCP)SBB

[Mabel Adams]

We just started using a barcode banding system last summer with our new HIS & LIS. It had a lot of start up problems. ICU was supposed to use it but couldn't figure it out so now the phlebs print off the labels for them and leave them in ICU for when they draw the specimens. So much for barcoded wristbands and labels printed at the time of draw! I would follow phlebs around and see how the system is really being used before deciding to trust it. We still use a BB armband system and will continue until this high-tech process gets tightened up. Problem is, once work-arounds are used it will be hard to convince everyone it matters to use the system as intended.

I am about ready to refer BBer's "unique identifier" problem to a mathematician I know to see if we can actually get a real picture of the risk in various sized hospitals. The odds of having two patients with the same name and DOB in a hospital with a given bed size. I sometimes feel like we don't know what the real risk is. MRNs are completely unique and if we could get them right 100% of the time, then no problem. Unfortunately, we can't. So what is best to use to help us catch errors in MRNs or to notice differences between 2 that are very similar? Sounds like a good probability problem for some student of math or statistics.

[John C. Staley]

Time to step back and re-evaluate what we are trying to accomplish with "absolute" patient identification.

1st: Is the patient in the bed the one we have orders for.

2nd: is the armband on the patient actually the patient it is attached to.

3rd: is the blood in the tube from the patient wearing the armaband.

4th: is the patient in our records the same patient in the bed.

5th: is the blood we crossmatched going into the same patient the blood in the tube came out of.

What is the minimum and I do mean minimum, that you need to safely and routinely accompish the 5 steps above. Is 80% of the name, along with one other acceptable identifier enough to adequately guarantee safety in your process?

Remember, complicating a process does not make it better.

[galvania]

John - I loved that last post. It's true, it's so easy to get lost in the details and forget the reasons!

Just to frighten you all, I live in Switzerland and many hospitals here don't even use armbands!! In fact, some hopitals just put your name (no Dob) on a card above (not attatched to) the bed.........

A question for Malcolm. I read last week that for the next census in the UK, they are going to include a third gender which they will call something like intersex. Is the health service going to do the same?

[Brenda K Hutson]

We just started using a barcode banding system last summer with our new HIS & LIS. It had a lot of start up problems. ICU was supposed to use it but couldn't figure it out so now the phlebs print off the labels for them and leave them in ICU for when they draw the specimens. So much for barcoded wristbands and labels printed at the time of draw! I would follow phlebs around and see how the system is really being used before deciding to trust it. We still use a BB armband system and will continue until this high-tech process gets tightened up. Problem is, once work-arounds are used it will be hard to convince everyone it matters to use the system as intended.

I am about ready to refer BBer's "unique identifier" problem to a mathematician I know to see if we can actually get a real picture of the risk in various sized hospitals. The odds of having two patients with the same name and DOB in a hospital with a given bed size. I sometimes feel like we don't know what the real risk is. MRNs are completely unique and if we could get them right 100% of the time, then no problem. Unfortunately, we can't. So what is best to use to help us catch errors in MRNs or to notice differences between 2 that are very similar? Sounds like a good probability problem for some student of math or statistics.

Another concern I have with accepting truncated names is that of course within families, the last name will be the same. Then I have heard of families that like to start all of their children's first name with the same initial, etc. And then we have the issue of children who have the same name as the parent but with a Jr after it; what if Registration picks the incorrect one?

Anyway, I am trying to "think of" all of the possible pros and cons with truncated names, unique identifiers that are used, the new technology of barcoded armbands, etc. I think I am having "brain freeze" with regard to possible scenarios of problems that could occur with the barcode system. So, I would welcome thoughts from any of you who can process this (especially those of you who have used these systems for awhile).

Thanks,

Brenda Hutson, CLS(ASCP)SBB

[Brenda K Hutson]

Time to step back and re-evaluate what we are trying to accomplish with "absolute" patient identification.

1st: Is the patient in the bed the one we have orders for.

2nd: is the armband on the patient actually the patient it is attached to.

3rd: is the blood in the tube from the patient wearing the armaband.

4th: is the patient in our records the same patient in the bed.

5th: is the blood we crossmatched going into the same patient the blood in the tube came out of.

What is the minimum and I do mean minimum, that you need to safely and routinely accompish the 5 steps above. Is 80% of the name, along with one other acceptable identifier enough to adequately guarantee safety in your process?

Remember, complicating a process does not make it better.

I agree with your last line; that is why I am trying to keep an "open mind" with this and appreciate all of your comments. I also agree with all of your points but my unfamiliarity with all of the pros and cons as well as potential problems a barcode system could have are still leaving me unable to answer all of those questions.

Thanks,

Brenda Hutson

[Malcolm Needs ☆]

A question for Malcolm. I read last week that for the next census in the UK, they are going to include a third gender which they will call something like intersex. Is the health service going to do the same?

I have not heard this, but it would come as no surprise whatsoever.

Whilst the UK definitely needed some Laws to stop prejudice of all kinds (it used to be rampant), it has now got to the stage where political correctness has not gone so much mad, as completely and utterly insane.

It has got to the stage that conversations between friends have become quite stilted if there is a chance of being overheard, just in case an innocent remark can be regarded as racist, sexist, oldist (you name it, we've got an "ist" for it in our Laws now - except smokist - you are actively encouraged to insult smokers) and they are reported to the authorities.

In the old days, however, and still this is true of most of our Laws, you are presumed innocent until proved guilty, but with some some of these "ist" Laws it is just the opposite - and trying to prove innocence is almost impossible. I've known completely innocent peoples' lives ruined by false accusations.

It is like living inside George Orwell's book "1984" sometimes.

The really daft thing about it is that the Government, of whatever political persuation, fail to see that interpreting the Laws in such a way (or allowing the Laws to be interpreted in such a way), plays straight into the hands of right wing extremist parties.

Anyway, end of rant, (sorry about that) and, no, I haven't heard that it is going to be introduced into the Health Service (but watch this space, as they say).

:mad::mad::mad::mad::mad:

Edited March 24, 2010 by Malcolm Needs

[Mary Ann]

Ok, my apologies for bringing up this subject "again," but just left a rather tense meeting with the Inpatient and Outpatient Phlebotomy Managers and we will pick this up again later this week to "make a final decision." I could use ALL of the input out there; regardless of your stance.

With relation to labeling of patient specimens for the Transfusion Service:

I have always been very strict on this with little room for exceptions (and that is in 6 different Institutions). In the past 2 weeks, we received specimens on 1 outpatient and 1 inpatient in which the last names were so long that part (or all) of the first names were truncated. The BB Specimen Labeling SOP as well as the Lab Guide (just for BB) state that for inpatients, there must be a complete first and last name and a MR#. The 2 identifiers defined for outpatients are the first and last name and D.O.B. However, the Managers believe we can/should make exceptions for patients with names so long that part of the first name is cut off. They state that there are still other pieces of documentation that identify the patient (for inpatients, the D.O.B. is also on the specimen labels; for outpatients, though the MR# may be on the specimen, it is not used as an identifier since the patient does not have a Hospital band on (just a red typnex band). The Outpatient Manager stated that you had other things to assist in identifying the patient; i.e. most of the name, D.O.B. and sex! I cannot believe she would suggest the sex, which has a 50/50 chance of being right should be acceptable to assist!

The Managers also believe that since the other areas of the Lab are more lenient than this, that we should above all, be consistent and that I should not require stricter requirements for the BB; that all Lab results are cricitical.

Anyway, I would love to hear from as many people as possible on:

1. The specimen requirements of your BB

2. Do you make any exceptions? If yes, in what areas?

3. Do the same strict requirements go for ALL of the other areas of the Lab in your hospital?

Need responses by Thursday!

Thanks so much,

Brenda Hutson, CLS(ASCP)SBB

The hardest part of my job is documenting and calling the floor for recollects,wrong identifier-typenex id torn from sample and patient's deciding on what name to use.I dislike hyphenated last names-keep it simple folks.OK, sometimes the name is like a novel and will not even fit on their hospital band.Blood bank must be labeled by the bedside taken from patient's hospital band.So if the band has Jane Plain-thomason... that's whats on the label with mr number specific for that patient with typenex id.

2. Hospital band is required-with typenex for proper identification in administering blood products.BLOOD.NO exceptions-in fact DNA testing labs will charge a fee for each incorrect identifiers.If govt wanted to fine for mislabeling-your manager would improve the process.Core lab has generated test labels.Remind your manager clerical error is the number one cause of transfusion reactions,sounds like they think blood bank is just too picky,what's there to it.I bet they have never worked bbk.SAD-we will continue being the safety police to cause no harm.

[Bill]

Brenda--When you enter the barcode number in the BB computer, what comes up one the screen? If entire name comes up on the screen just as it does on the handheld then your barcode becomes a unique identifier for the name.