|
A transfusion medicine physician at a 2000 bed hospital in India reports that they are working on an electronic medical record for patients. Their 'Computer Team' has been working on different modules for about a year. They have successfully tested modules that allow a physician to do discharge summaries, to view lab reports (including serial results with graphical 'dot-plot' displays, view all radiological images, see a patient's inpatient bill, schedule appointments made for an outpatient and drug ordering. One of the spin-offs of this medical records redesign is that electronic requests would do away with paper requests, including paper test requisitions. The ordering physician (or clerical assistants) would log into a computer to make the requests, samples would be collected, collection time would be logged, and samples sent to the lab would bear the patients Hospital ID Number (a unique 7 digit alpha numeric identifier). Currently they use a paper request (which bears the unique identifier as well as the patient's name) against which the label on the specimen tube is verified. Theoretically, once paper requisitions are eliminated, they will have no independent means of verification of the accuracy of the specimen label. In fact, their blood bank samples are currently handled manually, but they are currently 60% of the way through computerizing the Blood bank with a customized ISBT 128 based software. So they will eventually need to address the specimen label verification concern. They wonder if anybody has experience or a practical suggestion as to how best to deal with this situation. Is it even a problem?
The following comments have been received.
ADDENDA Aug. 12, 2005
1. A colleague in Massachusetts asked for clarification and requested additional information from colleague in India. Below are the questions and the responses:
Q.Details provided by the colleague from India need further clarification. For example, patient blood tubes should have two independent identifiers - such as medical record number plus full name - rather than just a number. Please clarify the situation at your hospital. Is the plan for the tubes to be labeled only with a number?
A.The hospital in India does use the patient's name and the hospital number, and will continue to do so with their new system.
Q.In addition, a record of 'who' collected the sample is valuable for error evaluations.
A.With their new system, they will know who has collected the sample.
Q.Samples need the date of draw as well.
A.Already compliant
Q.What is the plan in regards to identifying the phlebotomist and documenting the time/date of specimen collection?
A.The phlebotomist will create the request after samples have been drawn. There is a procedure for documenting the Employee ID number for this purpose. The collection date will be written on the sample label. The collection time will be entered on the e-req. With the new system, they will know who has collected the sample and when it was collected.
Q.Regardless of the degree of computerization, the blood specimen tube needs to be fully and properly labeled. If specimen tube labels are generated by an electronic printer, the system should be careful NOT to print these labels at a distance from the patient or to apply them to the tube in advance of collecting the specimen. What is the status at your hospital with regards to label printers and their proximity to the patient?
A.We plan to use pre-printed labels.
Q.Will pre-printed labels be employed, since pre-labeled specimens invite mis-collection errors? The data used to label the tube is best taken from the patient as a source (e.g., the wristband) rather than taken from an electronic database. Systems which scan a patient barcoded wristband and then generate a specimen label have a favorable design.
A.As of now they have only eye readable 'Name + number' on the patient's wrist bands.
Q.Will you be using barcoded wristbands to generate the specimen labels?
A.Not at this point in time.
Q.Electronic requisitions can still be sent to paper printout. A physician might enter an electronic order and that order could be printed on the patient's floor or could be printed inside the laboratory depending on the site. While a printed order is not required for safety, the laboratory may simply find it more convenient in order to assign work flow. What are your plans with regards to printing a lab copy of a requisition?
A.They plan to generate one printed copy of the request along with the results in their lab for use within the lab, until they sort out some 'glitches'.
Q.Whether printed or not, the "requisition" needs to be matched to the sample tube and the data compared for the integrity of the patient's identity, before the sample is accessioned into the laboratory. What are your plans to accomplish this matching?
A.This was the issue which triggered the question to the e-Network Forum for discussion. For now they are working on the premise that the number on the sample matches that on the e-request. That is, when the sample arrives in the lab, the barcode is scanned to call up the e-req attached to that particular sample. The name and number on the sample are compared to the one on the screen. However, they do realize that this will be biased in the sense that the number on the screen will need to match the scanned number. Only if they match will the request be called up. If there is no match, they will get an alarm message.
|
Printable PDF of these pages