Leaderboard

I am retired now and will never forget the sound they make when they are dropped on the floor.

My greatest apology for leaving you all hanging. We've been so incredibly busy and short-staffed that I could not even think about anything else but trying to finish up my daily duties. Anyway, seems that the patient also had an impella device that had to be "adjusted" and I believe that corrected the hemolysis. I'm only reading the responses today (2 weeks later)☹️so hats off to you all who thought mechanical causes. I would have not thought that the device would be that far-off to cause the gross hemolysis we saw. We do see slight hemolysis with impella devices but not like this one. I guess never say never. Thank you all for your responses.

Thank you so much! I've been perusing the site for a few years and thought, hey, why don't I sign up! LOL I'm the BB Lead at a smaller hospital in AZ and have 30 years experience. Looking forward to learning more here and, hopefully, even providing a few answers along the way.

Happy Friday everyone! First off, Thank you Malcolm 😊 for that excellent PPT lecture! 👌 With your permission, I'd love to share it with my students; it's exactly the kind of content that helps bring these complex concepts to life (and mildly melt brains in the best way possible). Speaking of melted brains… let's talk Anti-G differentiation, shall we? While I agree that differentiation isn’t necessary for transfusion purposes because in cases where anti-G is suspected, the recommendation is the same (D and C antigen-negative PRBCs); in the US Reference Lab world, it's a whole different dance. Differentiating anti-D + anti-C from anti-G is essential in alloimmunized patients with anti-D + anti-C for Rh Immune Globulin (RhIg) prophylaxis administration indication. Medical teams want to know if a patient is a candidate for RhIG, and if it's still indicated, especially to avoid future medical and legal complications. And so begins the beautiful (read: painstaking) double adsorption and elution process. 😫 The presence/development of a real anti-D indicates the patient does not require RhIg administration, whereas the presence of anti-G indicates the need for RhIg prophylaxis. That way, RhIG is avoided in patients with a real anti-D (although clinical correlation is recommended to guide every decision). In this case, we saw what looked/reacted like anti-D and anti-C (1). Adsorption with R2R2 cells showed anti-C in the adsorbed serum (2) and anti-D and/or G in the adsorbing R2R2 cell eluate (3). Usually, we would stop and call anti-C and anti-D, if we have no reactions on C Ag Pos cells, but since suspected anti-G was in the mix, we routinely differentiate/separate it. The anti-D was confirmed in (4) by adsorbing the eluate containing the suspected (anti-D + anti-G) with an r'r unit. So far, so good. However, when we tried to separate the anti-G by elution of the adsorbing r′r cells, expecting a reaction on D and C antigen-positive cells (5), we got a negative result (Cue dramatic music). I know, I know...at this point you’re probably wondering if we’re still in the Blood Bank or if we’ve accidentally wandered into theoretical physics. Trust me, my brain also wobbles when explaining G differentiation. It's the kind of thing that makes you rethink your career choices... for about 5 minutes... and then roll up your sleeves and start prepping another adsorption. 😅 A couple of housekeeping: We ficin treat our adsorbing cells, and sometimes use PEG in adsorption for efficiency. After briefly considering a career in something less chaotic, we retraced our steps. No PEG this time, 60-minute incubation, followed up with a DAT after each adsorption to check for antibody coating... and voilà: finally got the positive reaction we were looking for. The patient has anti-D, anti-C, and anti-G. I can only think of a possible weak DAT strength pending elution, or PEG interference when used in adsorption. As always, this is Blood Bank, we know a million things can go wrong, and often do. But in the end, science (and a lot of perseverance) wins. Thanks again to everyone for your input, patience, your brains, and your sense of humor. And again, Malcolm, thanks for your lecture, teaching, dedication and inspiration! I wish everyone a calm weekend. May your panels be clear, your DATs negative, and your eluates informative. 😉

In the UK, we would test serum/plasma samples from pregnant patients to see if there was an anti-C + Anti-G, or an anti-G on its own, but if the tests showed an anti-D+C, we didn't go any further to see if there was an anti-G there as well. I mean, what for? What difference does it make? I attach a PowerPoint lecture on the subject I wrote some years ago, but I think it is still pertinent. The G Antigen and Anti G.pptx

Your policy (to draw and test your own specimen) is the strictest interpretation of the standards (and one that most blood banks likely follow), but I do believe you'll find systems that share MRNs, share LIS, share policy & procedure (may even share technical staff) that will use pretransfusion results performed in blood bank A to transfuse the patient at location B by technical staff in blood bank B and are doing so within AABB and CAP standards. All that is to say, I don't think you'll find a standard that explicitly says you have to draw and test your own specimen at your own physical location. We come to that conclusion because of all the other standards.

Today is Biomedical Scientist Day 2025 UK. I would like to wish best wishes to all laboratory staff throughout the world.

Was the patient transfused? If not, hyper hemolysis is less likely. Sounds like mechanical causes are most likely.

You can also contact the FDA and ask, I have always found them to be very helpful. Worst case, report it (again, I don't think it's reportable) and they will reject it and let you know why. That is safer than not reporting it. I understand a lot of facilities are reluctant to report, but I came from a large facility and we reported about 50 - 75 events a year and never got in "trouble" from the FDA.

Hi, I realize you are not un the US; however, AABB has a lot of terrific resources available, for free. https://www.aabb.org/news-resources/resources/donor-history-questionnaires/blood-donor-history-questionnaires There is a questionnaire that you can administer, which will uncover the conditions you are looking at, and they also have guidelines on how to proceed should you receive an unexpected answer. These should be reviewed with your medical director to ensure the suggestions are what they would like implemented. In the last blood center where I worked, we pretty much followed them exactly. Why reinvent the wheel

Thank you for your exceptionally kind words DLabGirl, and by all means use the lecture as you wish. The same goes for anyone else who might want to use it, with the proviso that you a) realise that it is a bit "long in the tooth", and b) I did mean that we, in the UK, would still look to ensure that there really is an anti-D present, before we do not recommend giving anti-D immunoglobulin.

In our lab, we do 30 patients ABO typing daily in average. In those tests we will find out forward and reverse typing mismatch at least once daily. Maybe because we tested patients' sample, the incidence is higher than donors', but just as Malcolm said it is definitely necessary to do forward and reverse typing and make sure they are matching.

We have seen this phenomenon from time to time, albeit rarely. We use R1wR1 red cells with all of our antibody identifications, purely because it is always cell 1 in the panel! We also, however, use two R1R1 panel cells and an r'r panel cell. I'm not too sure why we get the odd sample that reacts like that, because the C antigen on the R1wR1 panel cell is not that much weaker than the C antigen on the R1R1 panel cells, and certainly no weaker than that on the r'r panel cells. Remember that the "w" of "Cw" stands for "Willis", and not "weak", as it was named after the donor who caused the immunisation to the antigen in the first example of anti-Cw described (something I forgot a few years back in an article I wrote, much to my embarrassment, if that is any solace to you) and that RH8 (RHCw) is allelic to both RH9 (RHCx) and RH51 (MAR), and not to either RH2 (RHC) or RH4 (RHc). Therefore, any weakening must be due to steric hinderance, or something similar. :confuse::confuse:

Definitely!

I actually didn't have to google the answer for this one! I knew that Landsteiner discovered ABO typing at the beginning of the 20th century :)

In our facilities, it also has to do with the Cost Center. Even if they transfer with the same MRN/FIN, our "Mother" hospital cuts off the band and repeats testing.

(Slightly paraphrased from the Guidance to the 34th Edition of AABB Standards for Blood Banks and Transfusion Services) AABB Standard 5.14.8 requires that there be 2 determinations of a recipients ABO group..... the first determination must be on a current sample and the second determination may be determined by one of the following methods: 1) comparison with previous records 2) testing a second sample collected at a time different from the first sample 3) retesting the same sample if the patient identification was verified at the time of sample collection using an electronic identification system. CAP standards (12/2024) Changes to Standards clarifies the requirement even further: "In the transfusion medicine checklist, TRM.40300 Historical Record Check says ABO, Rh, and antibody screen test results must be compared with results of the same tests recorded previously to detect discrepancies and identify patients requiring specially selected units. “New language was added to this existing requirement to clarify what is acceptable ABO and Rh historical records,” says Matthew Karafin, MD, MS, chair of the CAP Transfusion, Apheresis, and Cellular Therapy Committee and clinical professor of transfusion medicine services at the University of North Carolina at Chapel Hill. The revised requirement now says the acceptable ABO and Rh historical records for transfusion purposes are only those generated or entered by laboratory personnel into the health system’s laboratory information system and performed by an accredited lab or certified by the relevant government agency in its jurisdiction. “We learned that assessors have encountered facilities that were using blood types from other institutions, such as from Epic Care Everywhere,” Dr. Karafin says. “Moreover, we were made aware that blood typing information from these other institutions was sometimes being added by non-blood bank personnel, so we had concerns about the reliability of this information.” For patient safety, the CAP wants to ensure that the ABO and Rh used for transfusion are the ones that laboratory personnel have entered into an LIS. This “implies that a transfusion service reviewed the information and was responsible for its data entry,” Dr. Karafin says."

I don't think there is a requirement to share between hospitals, only that a search of previous records was performed for ABO/Rh and antibody screen. The lab needs to define how they will perform the search. The reasoning that labs use 2 types is a. to limit mistransfusion events and b. to use electronic crossmatch. You don't need 2 types is you aren't using electronic crossmatch and if you use other ways to reduce mistransfusion risks- you can always use immediate spin to allocate type specific blood. To use a type from another hospital, personally, I would want the testing/results to show up in my LIS automatically, and know that my lab and the other lab follow the same policies for patient ID, labeling, testing, etc. TRM.40300 Historical Record Check TRM.40670 ABO Group and Rh(D) Type Verification TRM.30550 Misidentification and Mistransfusion Risk Monitoring TRM.30575 Mistransfusion Risk Reduction

I think this is the key: Is initial testing at that ED is under one patient identifier/armband and the testing at your facility under a completely different patient identifier/armband? It's less about the testing, more about the patient identification. From CAP: TRM.40230 Specimen Labeling for Pretransfusion Testing Phase II All blood samples used for pretransfusion testing are labeled at the time of specimen collection in the presence of the patient with: 1. Patient's first and last name 2. Unique identification number 3. Date of collection 4. A method to identify the individual collecting the specimen. NOTE: Blood specimens collected for pretransfusion testing must be positively and completely identified and labeled before leaving the patient. Acceptable practices for positive identification of patient and blood specimen labels must be defined in the procedure manual and may include visual inspection and/or an electronic system to read the identifying information contained in bar codes or radio-frequency identification (RFID) microchips or the patient's wristband. Acceptable practices for generating specimen labels must be defined in the procedure manual (refer to GEN.40490) and may include electronic devices utilizing information encoded in bar codes or RFID microchips. There must be a dependable method to identify the individual who collected the blood specimen, such as initials or another identifier on the tube, or an electronic record. Evidence of Compliance ✓ Properly labeled blood specimens AND ✓ Records identifying the individual collecting pretransfusion testing specimens

In my opinion, if you have: A cold antibody (not pathological), and Have done the workup to show that there is no other antibody present (i.e. a clinically significant antibody such as anti-E), and Your LIS dictionary set up so that your reported antibody is not deemed as clinically significant You can set it up so that the computer (electronic) crossmatch can be used. If it is a pathological antibody, we use incompatible (i.e. Least incompatible in Meditech)

At our facility, we have to do a cold adsorption IF the cold auto interferes with the reverse and/or ISXM. Since testing with "neat" plasma is our standard of practice, we would still report the XM as incompatible and send it out with a release stating the incompatibility was due to an autoantibody. We are not "allowed" to prewarm away IS reactions. 🥶 IF we prewarm - it's only if the cold has a high thermal amplitude and causes interference after 37deg incubation / antiglobulin phase...........but that's just us.........

Are you doing the immediate spin for a crossmatch or antibody screen? Well, we use the prewarm method for Cold agglutinins. We enter the results to the prewarm technique which is Negative, so the computer doesn't have a cow. Then we write in the test comments that the prewarm technique was used due to cold agglutinin, or history of cold agglutinin.

The last time we saw this at our institution, the ECMO team found a crack in the cannula!!! I agree with Yanxia - the clinical and surgical team needs to check the ECMO circuit for causes of mechanical hemolysis and lines. Our team had to redo the cannulas and circuit in this case and the hemolysis went away as soon as that was completed.

Yes, we needed to reduce the expiration to 28 days from the date of collection. We issued about 100 units of RBCs a day, we didn't outdate a lot. We had a huge cancer population, we just felt it was safer to irradiate everything. We also adopted 100% leukoreduction much earlier than most facilities.

I just answered this question. My Score PASS  

I just answered this question. My Score PASS  

I leave that for the physicians to respond to, but I suspect that since there will be lingering lymphocytes, no, they would not be protected. I, too, was part of a large level 1 trauma center. Our inventory was typically 700 - 1,000 RBCs. We were 100% irradiated. Once they were received, they went right into the irradiator room (we had a double door fridge in there. We usually had someone in irradiating most of the first and second shift, we had two old cesium irradiators. It was very hard to keep up. We looked into getting two x-ray irradiators, but it was cost-prohibitive, even with the government program where they would take the cesium ones away and pay for half the cost of the new ones.

I do not see how a genuine massive protocol can be supported with irradiated products. This is based on experience supporting patients during many massive transfusion protocol in a level 1 hospital, and also working in/with two facilities with three different types of irradiator. The irradiating process is just not that fast. I imagine the practical solution would be to give irradiated blood once the patient's bleeding is getting under control. I wonder if the patients stressed systems would prevent GVHD response?

I think there are a lot of variables that you, and your medical director, need to consider. The biggest is your patient population. Do you have a lot of cancer patients taking monoclonal antibody drugs? Do you have a lot of maternity patients? Do you have a lot of antibody positive patients? The only correct answer is to test 100% of the samples to be 100% sure. That is not practical, and no one would ever suggest that. Scan the FDA guidance documents to see if you find any relevant ones. Here are a couple of unrelated ones: https://www.fda.gov/files/drugs/published/Process-Validation--General-Principles-and-Practices.pdf https://www.fda.gov/files/drugs/published/Bioanalytical-Method-Validation-Guidance-for-Industry.pdf For a NEW process that we were unfamiliar with, I always went with 385, regardless of the population size. We migrated a patient database with millions of records from our homegrown system to a commercial system, we tested different scenarios 385 times. As time went on, and we grew more confident with that particular software system, and continuously found no errors, we reduced it, slowly, over each validation. The key point is that you and your medical director need to have a sound, and justified plan. Your primary goal is to protect your patients. Your secondary goal is not to get closed down. Should you test every sample, no. Should you test 20 samples for a process you have never performed before, no. I suggest looking at some statistics sites and determining what works best for you. Here's just one, there are MANY: https://www.calculator.net/sample-size-calculator.html?type=1&cl=95&ci=5&pp=50&ps=&x=Calculate

I have the guide from the vendor, but they said it up to us how many and what type of scenarios to test with the reader. I was just curious if someone had done it before and had some insight.

Patient should be monitored for a delayed hemolytic transfusion reaction for about 10-14 days, not necessarily in hospital. Most common signs are fever and progressive anemia. sometimes dark urine or jaundice. Patient education before discharge if earlier than this is essential.

I would say as long as the rule out was done per policy and appropriately then there is nothing to report.

No, not an error.

It's been a while for me, but my initial instinct would be to say no. The safety, purity, and potency of the product were not compromised due to an error (intentional or not) on your part. Is it your policy to always check to ensure that every patient has not been previously transfused at any hospital and does not have an antibody? I'm not being facetious, but if it is, then yes, you did not follow your policy. You can try to find a code for it here, but I don't think you will. https://www.fda.gov/media/161919/download?attachment

Some of the worst hemolysis I have seen was in clostridium septicemia. Both cases were fatal.

Welcome to this SUPREMELY WONDERFUL site MaeveQ. ENJOY!

Not yet available. Being developed by Grifols. Probably months to a year away from FDA approval. You can contact them about becoming a testing site for licensure I'd guess. Until it's licensed you won't be able to use it in patient care, just research/validation.

My first recommendation is to ask the vendor. They will often provide the skeleton of a validation plan, that you can then adapt to what you have outlined in your SOPs.

In the UK a unit of blood would NEVER be sent out to a hospital without a full (and matching) ABO type - both forward and reverse. We would also do everything possible to ensure that, if the forward and reverse ABO types of any patient do not match, we find out why before transfusion. An ABO mismatch is probably the most common cause of fatal haemolytic transfusion reactions (although, thank goodness, they are NOT common), and this is why we will always go "the extra mile" to try to prevent any such situation.

The FDA Guidance on Computer Crossmatching calls out that patients with an ABO typing discrepancy should not be allowed to qualify for computer crossmatches. " If ABO typing discrepancies exist, you should not rely on a computer crossmatch. This is particularly important if there is mixed field red cell reactivity, missing serum reactivity, or apparent change in blood type following hematopoietic stem cell transplantation. Under those circumstances, your procedures should provide for compatibility testing using serologic crossmatch techniques." I wrote our policy to include any non-straightforward ABO types for any reason will be required to get an IS or IAT crossmatch.

This may be a little outdated, it's from a prior facility. We did a tremendous amount of transplant infusion, and this evolved over the decades I was there. Management of Hematopoietic Progenitor Cell Transplant Recipients.docx

If you issue blood via a pneumatic tube system, this introduces some additional processes.

It is most unlikely to be anti-f, as this specificity will not react with either R1R1 or R2R2 red cells (I presume that you are using both antibody screening cells and antibody identification panel cells that have these phenotypes represented). Anti-f reacts with red cells where the c and e antigens are the result of an RHCE*ce haplotype where, for want of a better way of putting it, the c and e antigens are the result of the RHc and RHe genes being in the cis position - NOTE THAT THIS TERMINOLOGY IS (KNOWINGLY) WRONG! Like all Rh antibodies, anti-f reacts most strongly with its cognate antigen on red cells that have been treated with a proteolytic enzyme (such as papain or ficin), but will very often react with untreated red cells by IAT.

Technically, any sample you don't know the answer to is "blind" to you, so any regular patient with no history can be used for a blind blood type for example. Yes, for DAT and FMH it's harder, but we typically use the CAP samples as Bet'naSBB said, rather than try to make up samples that are not quite right.

We use our CAP samples AFTER the results have been submitted and results have been received from CAP. We just finished assigning a BUNCH of "Internal Assessments" and "Method Comparisons" using our first batch of CAPs that we'd already received our results for. All these count as "blinds" for the staff. Instead of making 1 tech do the whole survey, we give each assignee one sample to do and then compare their results with those expected by CAP. works great! For FMH, we get two CAP "TMCAF" surveys per year. 1/2 the staff does the first and the other 1/2 the second so everyone gets a blind for FMH.

General Thoughts First, following these tips will not ensure citations are not issued. This guide is designed to help you survive your first, or hundredth, inspection. Feel free to leave your tips in the comments below. I have led or been part of a team that has hosted dozens of inspections over more than two decades. Below are some tips on how to prepare for your first inspection, respond to your inspector, and survive with your sanity intact. I may be glib at times in this post, but I do take this process very seriously. You may lose a prestigious accreditation, or worse, the ability to perform a critical test. Occasionally substantial fines can be involved. Rarely, prison sentences are imposed. Keep good notes. document everything and keep copies of everything – more below. Notifications Keep people apprised. I suggest creating a group email address now, not on the day of the inspection, but now. It should include your leaders, leaders of other areas that may be impacted – Nursing, Admissions…, your direct reports (unless that is line staff), and others that may need to know. I promise you will forget someone, that’s OK, add them to the list and move on. On the arrival of the inspectors, send a quick note to that list letting them know basic details: agency, areas they are likely to review, expected length of inspection, time / location of the opening meeting (if there is one). Create an email to send to staff, do this now, again not on the day of the inspection. Save it somewhere as a Word document so you can quickly copy and paste the information into an email. It should include items such as, only answer what you are asked, ensure you have your institution provided ID worn as required, say “I don’t know, I would ask my manager…”. At the end of each day, send out a summary to the leaders list. If there are findings, let the leaders know. Also notify line staff. This is an inspection of their work; they need to feel proud of what they have done. You may want to hold back on any citations (if any) until you have developed a fix. Types of Inspections The terms below are a few of the more frequent organizations you may encounter during an inspection. There are several terms used for the process, for the purpose of this post, I will use “inspection” to represent the different terms, and “agency” to capture all inspection organizations. For anything that goes awry and is documented as needing improvement, we’ll use “citation”. FDA Inspection AABB Assessment TJC Survey CAP Inspection FACT Inspection ISO Conformity assessment GCP Inspection Announcement Inspections are designed to be a surprise. Some agencies will give you a window of when they are likely to arrive, others do not provide any information at all. This is by design. It is the expectation of the agency that you are always ready for an inspection. Continuous Readiness You are always ready to perform testing on a patient specimen, correct? The same should be true for an inspection. You need solid processes in place for all the items an inspector will look for. These are important for inspections, but more important for patient testing. A short list of activities you need to have procedures for, and most important, documentation, follow: Correlation studies Transfusion Reaction Investigations Competency Assessment Training (this is not competency) Equipment maintenance records – as required by the manufacturer Quality control (signed by an approved person) Personnel records – evidence of an annual review, qualification for their position How to write an SOP Organizational structure - this does not need to be a beautiful chart, it can be as simple as a listing of what position reports to whom, and their general duties Speak When Spoken To Think of an inspection as a first date. You will be nervous, regardless of this being your first inspection or 20th. There will be many times of extended quiet – sometimes an hour or more. Bring your laptop, cellphone, or something else private for you to work on in the presence of the inspector. During these quiet times, you may get extra nervous. The inspector may take notes, may mumble to themselves while reviewing an SOP, may furrow their brow when looking over quality control records… do not be tempted to ask if they need anything. Trust me, they will ask. Do not be belligerent but be close. When asked a question, answer the question honestly and fully, but only to the point of answering the question. For example, you’re asked “Do you perform quality control on reagent xyz?”. Answer according to your SOP, then stop talking. Do not respond with “Yes, we perform daily quality control on reagent xyz; however, we feel we could improve how we document this information and have begun a process improvement project to address this.”. This will lead to more questions, and a possible citation. Sadly, we’ve all had the inspector who will respond, “No worries, it’s not an actual rule, but it’s they I’ve always done it.”. Do your best to smile (at least inside) and do not respond. Be Honest You know where your weaknesses are. Please do not point them out (see the section above). If an inspector fails to uncover a process you plan to improve, no harm is done. If an inspector asks about a process you are not performing correctly, answer the question directly. For example (US based) “I was reviewing employee 123’s training record. I see you’ve done a terrific job training them on Process ABC. I really like your training documents, great work. I also see they have documented independent performance of test ABC one week after they were trained. Please show me the documentation of their competency assessment.”. You do not have this documentation. Uh oh. You’ve been caught, own up to it. A possible response: “Thank you for recognizing the hard work we put into training our staff, we really are proud of it. As you noticed, employee 123 does not have documented initial competency, this is something we are working on. Can I show you our project plan for bringing this into compliance?”. This is a serious concern for inspectors, and you may be cited, you may not, based on your project plan. The crucial point is to not argue with them. Do Not Argue Should an inspector find something they don’t find quite right, ask for clarification. Something that I found helpful is, “Thank you for pointing that deficiency out. We take our work very seriously. As you can imagine, process change is a critical step in our continuous quality improvement and needs to be agreed upon by senior management. Can you please let me know what regulation I am not fulfilling – specifically with the rule number.”. If you are going to be cited, you are going to be cited. Do not get defensive, do not argue, do not try to negotiate. There is time to do that in your response. Ask for Specifics Let’s say the inspector say, “Please show me quality Control.”. That’s a big request. Ask for clarification. You: “I would be happy to provide you quality control information, can you please be more specific?”. Inspector: “Yes, I want to see reagent quality control.”. You: “Can you please provide a date range for the records you would like to see.”. Inspector: “Please give me last month, and the same month for the prior year.”. You: “Is there a particular reagent or process you would like to see?”. You get the point. You are not trying to be uncooperative or unhelpful. You only want to provide exactly what they need. More is not better. Do not attempt to bury them in data, they will find every unchecked box and unreviewed document you have. Document What They See Always provide an escort for your inspector(s). Never let them walk through the laboratory on their own, even to go to the restroom. Clearly do not follow them in; however, observe them as they head out of the laboratory toward the restroom, then stay within sight of where they will return. They will be observing what is going on as they walk through, and if something piques their interest, they may stop and ask questions. You need to be present for this, so you have a firsthand account of what transpired. Many times, an inspector will ask for documents to be provided as a copy. Depending on the agency, you may not be allowed to redact. One example in the US is the FDA. Unless you are given permission to redact, you must provide anything they ask for, including private / protected information. Always make a copy for the inspector and an identical copy for yourself. Keep these documents in your folder for that inspection and maintain them. Do a Gap Analysis Performing a gap analysis is an amazing opportunity. Pick the most critical set of rules you’re required to follow. It’s wonderful when these rules are available for download. Don’t make the process overly complicated. Add the rule to a column in a spreadsheet. In the next column document the exact location in the SOP where you are fulfilling this rule. Some document management systems allow for the recording of this, but if you do not have this available to you, again, a simple spreadsheet will suffice. I assure you, you’ll learn a lot doing this exercise, most importantly, that there will be items you assumed were in your procedures that are not. This is a fun process, but it takes resources. If you are doing this on your own it could take months, maybe an entire year. Ask your staff for help. Give them clear instructions on what you need them to do, give them small chunks at a time to work on. I promise they will enjoy the education and chance to help improve your overall quality. There is zero chance you will complete this task with no processes to improve. This is OK. Now you know where your weaknesses are and can prioritize working on them. Response to Inspection Most agencies tell you how they want you to respond, and when. Examples: AABB has a format they prefer you to follow when responding – use their document. TJC has timelines for when items must be corrected, and more timelines for evidence of these corrections - meet these timelines. FDA (unless it’s a Warning Letter) does not expect a response. The key point, the inspecting agency has a proposed format, use it. Never be late with your response. If circumstances do not allow you to meet the deadline, well in advance of when your response is due, ask for an extension. The agency will often grant one. If the agency does not require a response, respond anyhow. For instance, if the FDA finds one item that needs attention, write to them thanking them for their time, let them know you appreciate the thoroughness of their investigation, and describe specifically how you have corrected the deficiency. They will likely mention this the next time they visit, and it will set a positive tone for that inspection. If you are fortunate enough to survive the inspection without a citation, still write a letter thanking the agency for their time. Closing Notes This may be your first inspection, but it will not be your last. It is part of doing business in a laboratory. Accept that fact. Develop a procedure for inspections: do you provide parking; do you offer lunch… Learn from missteps you had during each inspection. Most important, don’t take a citation personally. Learn from what your inspector provided you and get better.