We are hoping to (finally) get rid of our card system for recording patient results.  Right now, and since the last millenium, we have a dual system in that we record results on a patient card and then transfer them to our BB HCLL system. 

 

Now when we check each other's work, we compare the card results to what is recorded on the computer.  For those of you who have gotten rid of this archaic 20th century paper-based system, how do you check each other's work when you only have the computer record?

 

Any tips would be greatly appreciated.

 

Thanks, Scott

Gel cards are labeled with barcoded specimen label. Gel card results are keyed directly into the computer.   A single  individual is wholly responsible for results entry.

^{I agree with Dansket.}  

Your current system is only comparing paper results. Why have you not just entered data directly into the BBIS? I have never understood copying paper results into the BBIS after the fact/at the end of shift/etc. I have seen major errors during inspections when this has been policy.

Your BBIS has truth tables and you should get an exception report and also be able to print out a copy of all work performed (not just the interps). Nothing will prevent someone from incorrect data entry, but your system will alert them (and you) if the truth tables are violated - should evoke some investigation before results are validated. Your system is your safeguard, that is why you validate its functionality before you go live.

Thanks for those responses.  I guess we are under the impression that some regulation, either FDA or, for us, CLIA/JCAHO requires us to document test review.  I think this is where we got into this overkill originally, years ago when we went from paper to paper/computer.  From what you are saying we do not have much to worry about. 

 

I admit that after a quick internet search, I cannot find a reg that would require us to review results at the intensity that we now do.

 

Scott

No, I don't believe there is a requirement to verify results in real time. Techs should result directly into the BBIS directly from the testing medium (tube, gel card, etc). Rarely our analyzer cannot export results and the techs have to print the results and transcribe them manually into the BBIS; if this is the case I require another tech to review it in real time and initial the paper. We only go this extra step because we had an error a few years ago with manual entry that ended up with a patient getting an incompatible unit.

I agree with David; transcription can lead to clerical error, so you want to limit it whenever possible.

We got rid of our card files a couple years ago and I have loved it ever since.

Hi Scott,

We report directly into our LIS from the gel card.  We label our Gel cards with barcode specimen labels.  When the tech results at the terminal, he/she scans the barcode and enters results directly from the gel card.  We also have a system in place to verify blood type on patients without a history in our system.  Those patients require a second type either from the same tube or a completely separate draw.

 

The following morning, myself or a designee reviews an automated report that contains all the patients tested the previous day.  That report is compared patient for patient to the actual gel cards tested the previous day.

 

Like David mentioned, your system should have rules or truth tables built into it.  For instance, ours is set up so that you can't crossmatch a unit on a patient until a second type has been done or a tech acknowledges a results field indicating a historical blood type is on file.

 

I hope this helps a little.

 

Michael

I agree with the other posters, no need to have someone double check an entry into a BB computer system.  I have worked several places within the same organization and we discontinued any form of manual entry about 15 years ago.

 

The truth tables that run behind the scenes act as a checking mechanism which is part of the patient safety features of the system.

 

Eliminating the paper step will greatly increase your efficiency and once you get past the first few time I bet very few if any techs ask for the paper to return.  

There is a CAP requirement TRM.40690 about verification of correct data entry AND documentation of computer system alerts. 

How are you documenting that the techs are verifying data entry?

That requirement says it is for computer crossmatches. 

Even if it is not just for computer entry - the tech has to verify their data entry before final acceptance. If they do not there is no (or should not be a) final result.

I agree with David.

Your current system is only comparing paper results. Why have you not just entered data directly into the BBIS? I have never understood copying paper results into the BBIS after the fact/at the end of shift/etc. I have seen major errors during inspections when this has been policy.

Your BBIS has truth tables and you should get an exception report and also be able to print out a copy of all work performed (not just the interps). Nothing will prevent someone from incorrect data entry, but your system will alert them (and you) if the truth tables are violated - should evoke some investigation before results are validated. Your system is your safeguard, that is why you validate its functionality before you go live.

CAP LAB GENERAL 7/29/2013

GEN.43825 Result Verification Phase II

Manual and automated result entries are verified before final acceptance and reporting by the computer.

NOTE: Data entered into the computer system either manually or by automated methods must be reviewed by an authorized individual who verifies the accuracy of the input data before final acceptance and reporting by the computer. An example of best practices for this step is checking the result against the reportable range and critical results for the test. Depending on the local environment, this may or may not require a second person. Verification procedures must generate an audit trail.

This checklist requirement does not apply to autoverification procedures (see below).

TRANS MED

TRM.40100 Test Result Recording Phase II

Observations of all test results are recorded properly at the time the test is performed.

NOTE: Test results must be recorded at the time the test is performed in order to reduce the risk of transcription errors from delayed recording.

REFERENCES

1) Department of Health and Human Services, Centers for Medicare & Medicaid Services. Clinical laboratory improvement amendments of 1988; final rule. Fed Register. 2003(Jan 24): [42CFR493.1291(a)]

TRM.40690 Data Entry Verification Phase II

If a serologic crossmatch is not performed, there is a method to verify correct computer data entry before issuing blood or blood components, and the computer alerts the user of any discrepancies.

NOTE: When a serologic crossmatch is not performed, patient safety must be ensured by requiring verification of proper data entry before issuing blood or blood components. The computer system must alert the user of any discrepancies of donor unit labeling, blood group confirmatory test interpretation, and to the existence of any ABO incompatibility.

Evidence of Compliance:

✓ Written policy requiring verification of correct data entry prior to release of blood/blood components AND

✓ Records for verification of correct data entry AND

✓ Documentation of computer system alerts used to prevent issuance of blood components when discrepancies exist

Joint Commission

Elements of Performance for QSA.02.11.01

The laboratory has written policies and procedures for surveillance activities that include a coordinated review of the following: A

- Patient test results

- Work records

- Equipment performance testing records

- Quality control results

(See also QSA.02.02.01, EP 5)

1.

The policies and procedures include criteria to determine acceptability of patient results before they are released. (See also A

QSA.02.02.01, EP 5)

2.

The general supervisor performs or delegates to technical staff the daily supervisory review of patient results. The supervisory C

review is documented.

Note: Technical staff performing the review use specific criteria or computer algorithms to identify outlier results for manual

review. Examples of criteria include the following:

- Unacceptable quality control results

- Test results that do not correlate with a patient‘s known condition, age, sex, diagnosis, or pertinent clinical data; distribution of

patient test results; and relationship with other test parameters

- Incongruent test results on one patient

- Abnormal test results

- Critical values

(See also LD.04.05.01, EP 1; QSA.02.02.01, EP 5)

3.

For high-complexity testing performed by trained high school graduates qualified under 42 CFR 493.1489( (5), the laboratory A

director, general supervisor, or technical supervisor reviews all results within 24 hours of patient testing. (See also QSA.02.02.01,

EP 5)

4.

The laboratory performs daily screening for errors in patient test results due to handwritten or manual data entry (for example, C

clerical errors). The daily screening is documented. (See also QSA.02.02.01, EP 5)

Note: Screening a sample of data is acceptable for compliance with this element of performance.

5.

The laboratory performs screening for errors (for example, electronic transmission errors, formatting errors) in electronic and C

printed patient test results at a frequency defined by the laboratory. The screening is documented. (See also QSA.02.02.01, EP 5)

6.

The laboratory performs review of other records (for example, work records, equipment records, quality control summaries) at a A

frequency defined by the laboratory, but at least monthly. The review is documented. (See also QSA.02.02.01, EP 5)

7.

Page

We use only a computer system and there is no check by somebody else. Our automated system exports results to the LIS, but our manual tube results are entered as they're performed and nobody double checks it. The LIS contains truth tables that will not allow incorrect entry of results. For example, if the reverse type doesn't match the front type an error box will come up saying there's an ABO mismatch. The tech then has to re-enter the results if they were entered incorrectly or if there is an ABO discrepancy to result an  Indeterminate for blood type. That ABO discrepancy result will appear on an exception report. As long as the techs are looking at the results before hitting verify (not just blindly hitting verify) and there are truth tables set up in the LIS, then you're in compliance with regulations.

Edited January 17, 201411 yr by Sophie1210