A transfusion service medical director oversees the transfusion services at two different licensed acute care hospitals. The two hospitals are 12 miles apart, but in addition to having the same medical director, they share a common set of transfusion service SOPs, the same computer (Cerner) system, and the same database. Recently, the smaller of the two hospitals (hospital S) reports that it began to transfer patients, (but not units of blood - see previous e-Network discussion) to the larger hospital (hospital L), because hospital L has a trauma center, while hospital S does not. Almost always, when a patient is transferred from hospital S to hospital L for a higher level of care, the patient arrives with a blood bank armband already in place, and a tube of blood that was drawn and tested at hospital S for ABO, Rh and compatibility testing. Each hospital has the same requirements for patient ID band placement and blood specimen labeling for blood bank testing, including ABO, Rh, antibody screen and crossmatch. The medical director has inquired of the e-network forum if hospital L would be violating any AABB, state, or federal regulations/standards if the ABO, Rh and antibody screen results entered into the computer by hospital S were relied upon for issue of group-specific blood at hospital L. Further, if hospital L performed crossmatches using the inter-hospital transferred sample, would that practice violate any regulation or standard? The medical director reminds us that his two transfusion services act as if they were one service, since both hospitals use the same SOPs and the same computer system, the blood bank testing results entered at hospital S appear immediately on the computer system at hospital L. No additional data entry is necessary at hospital L, other than to indicate that the patient was transferred. Currently, the results of the inter-hospital transferred sample are used by hospital L, but only after they have been verified by direct testing in hospital L's transfusion service, since AABB Standards imply that the ABO/Rh must be "performed" at the transfusing "facility".

The following comments were received in response to the above scenario:

1. The Director of Standards and International Affairs at the American Association of Blood Banks responded on behalf of the AABB. He polled the Director of Quality Assurance and Accreditation at the AABB National Office and the Chair of the Standards Program Unit that writes BB/TS Standards. According to the AABB, the key to whether or not the practice in the scenario is compliant with AABB Standards depends on the definition of a facility/organization, but unfortunately, this definition is not made clear in the glossary of the 20th edition of Standards. It appears that if a facility operates under one continuous quality system, regardless of whether all parts of it are physically located in the same building, then the practice described in the scenario would be in sync with AABB Standards. However, the AABB cautions that the FDA may have a more narrow definition of "facility".

EDITOR's NOTE: Based on the foregoing information, it would seem prudent for the institutions described in the scenario to write to the AABB directly, and to obtain a written statement that they are in compliance with the Standards.

2. An AABB assessor commented that since the computer system worked from the same database, the armband system of identification processes were identical at both facilities, the armband matched the unit / the specimen / the patient, she does not know that she would require additional testing. She thinks that she would recommend the facilities request a variance to Standards, based on being one 'functional' facility, with "two" ancillary sites, and see what the Committee says. The strategy would be to refer back to the Centralized Transfusion Service concept. If she were assessing this scenario, she would want to see the transfusion protocols, patient identification protocols, blood administration protocols, specimen receipt protocols, etc., and she would want to verify that each are "mirror images." She would also want to see the computer validation documentation ensuring that the system works (what do they do in "downtime?"), and to investigate what happens when they get multiple patients with the same name, etc. If the hospitals in the scenario could show all of that, she does not think that any additional testing would be necessary.

ADDENDUM Aug. 1, 2001

3. A member wrote a cautionary note, as follows: "My experience has been that even though they try to be identical and attempt to use identical practices, it never happens. There are human beings involved in the equation and that introduces variables in many ways. Also just the wording in the SOPs alone requires that the SOP reflect the situation in the environment in which it is used. In the example given, one hospital was larger and one smaller. If I were the medical director responsible for this operation, I would perform an on-site assessment to validate that what they say they are doing is actually what is being done."

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator