Prompt lab response to an emergency alert.
So we thought. Then a two-car accident sent five unconscious and unidentified female patients of about the same age to the ER, and the laboratory's emergency response system ran into problems. Here's what went wrong:
* Under a long-standing disaster plan, the alert was sounded by a "code green" message paged around the hospital. That was the laboratory's cue to dispatch two phlebotomists to the emergency room to collect blood specimens from the accident victims.
Unfortunately, the alert did not reach everyone in the laboratory: many were unaware of an urgent testing need. While the phlebotomists rushed off, a technologist finished dismantling a hematology analyzer and a blood gas analyser for regular maintenance. Thus we had to rely on much slower methods, such as the manual spun hematocrit and a non-automated backup blood gas analyzer during the emergency.
* Because the victims were unconscious, it took longer than usual for the admitting department to collect patient information, enter it into the hospital computer, and assign patient numbers. ER physicians and nurses used Doe-Room 1, Doe-Room 2, etc., as IDs in the meantime, and that's how specimens and test requests were labeled. But upon delivering the results, we discovered the patients had been moved to the floors for further medical treatment. This transfer stripped them of their original IDs. The only way to match results to patients was to gather phlebotomists, ER physicians, and nurses at bedside and reconstruct the circumstances surrounding each draw.
* Our phlebotomists needed on-the-scene supervision, despite their experience. We expected them to go to the ER, draw blood, and return without delays of any kind. Instead, they had to cope with blood draw requests from several physicians at once, all demanding first priority. The phlebotomists grew confused as physicians seemed to be getting upset with them for not following orders. Precious minutes were also lost because the phlebotomists had arrived in the ER with their everyday trays, stocked with supplies inadequate for the emergency.
It took up to 20 mintes to get some specimens to the laboratory. In one case, the emergency room called the lab for results before we had received the speciments.
Eventually, all patients were properly identified, and their lab results were placed on the charts. After the excitement subsided, a critique was held at a meeting of ER physicians, nurses, and assistant department heads from radiology and the laboratory. Everyone's major concern was the lack of positive patient identification. Working with the admitting department, the laboratory and the ER devised a procedure for identifying unconscious, unidentified patients in an emergency/disaster situation.
The new ID procedure is separate but compatible with the hospital computer system. When an unknown patient is brought to the emergency room, admitting or ER personnel attach a prenumbered "disaster" tag to a wrist or ankle. This unique number is used immediately by all departments as positive ID. Multipart self-stick labels bearing the number are affixed to laboratory specimens.
If time permits computer entry, the number goes into the patient name field--for example, Doe, Jane 5602. It remains there for at least one hour to enable the laboratory to match it to the victim's specimens. When the hospital learns the patient's real name, that becomes the main ID along with an assigned patient history number. The disaster number is removed from the computer, but it stays on the chart permanently.
The laboratory, now jolted from its complacement attitude toward emergency response, formed a committee under the guidance of the lab director. Representatives from all areas, including a technical assistant from the phlebotomy team, met to design a program that would address all the problems encountered during the real disaster.
Within six months, we established a system called Emergency Alert. It starts with improved communication.
The emergency room telephones the laboratory in advance, whenever possible, that patients in critical condition are on the way into the hospital. The laboratory paging system, which now reaches all sections, then calls the lab administrator or the supervisor in charge to the frong desk. Informed of the emergency, the administrator sends two phlebotomists to the ER. She also goes there to asses the situation. If she determines that a full-scale response is required, the message is broadcast throughout the laboratory: Emergency Alert-1 or 2 or 3 (the number indicates how many patients are involved).
Everyone in the lab clears the decks and follows the emergency plan. Routine instrument maintenance is delayed or, if it is in progress, backup methods are immediately set up and controls run so patient specimens can be analyzed without delay. Technologists have an opportunity to consolidate their workload if possible, prioritize remaining test orders, and request additional staffing if necessary before emergency victims arrive.
The lab administrator may ask for additional personnel in the ER, including a pathologist--or may decide that only one phlebotomist is needed. She may appoint a supervisor as coordinator or do the job herself. She may also transport speciments to the lab or assign this task as the situation warrants.
All laboratorians in the ER now wear "LAB" armbands. In emergencies, many different departments respond. Without clear identificiaion, lab staff members may be mistaken for other kinds of personnel and be asked to perform duties for which they are not qualified. The armbands and the presence of a lab supervisor in the ER have ended the confusion that our phlebotomists experienced.
Emergency phlebotomy supplies are carried in a bright red, plastic fishing tackle box. This kit contains all the blood collection apparatus needed, including coagulation tubes and a freshly refigerated tube for fibrin split products not normally carried on a phlebotomy tray. There is equipment for pediatric draws, an arterial blood gas kit, and blood culture bottles. A daily check insures that the box is always ready.
Now when an emergency is in progress, all lab results are reported on manual request forms and delivered to the physician in the emergency room. The results are also entered in the computer at the first available opportunity.
After Emergency Alert was thoroughly explained to all labortory personnel, the protocol was added to our laboratory policy and procedure manual. There are two versions, one for the day shift and another for the evening, midnight, and weekend shifts, as shown in Figure I.
If anything, Emergency Alert probably gets used a little too often. We launch into it as soon as the ER calls, and sometimes the laboratory administrator finds that a full-scale response would be inappropriate.
But since our problems with the five Jane Does, there have been several similar instances of unconscious critical patients arriving simultaneously in the emergency room, and each time the laboratory responded very well. Our improved performance has won praise from ER physicians and staff.