Cost analysis on reference lab testing

[kirkaw]

As I'm sure everyone else is doing, we are trying to find ways to cut costs. We are a medium sized hospital (<400 beds) in a rural setting without a trauma designation. Annually, I look at supplier contracts including reference lab testing. We currently do our own antigen typing and primary and secondary antibody ID panels but do not keep an enzyme panel nor do we do adsorptions, although we do perform elutions. Most of my staff are generalists and at least half of them are MLT's.

I am curious to know the size of your institution and how much, if any, of your serology you send to a reference lab. We are seeing more complex patients (in volume, not necessarily complexity) and I am considering bringing in enzymes and adsorptions. Please give me your thoughts. Thanks!

[David Saikin]

At my previous employer, ~24 beds, we did almost everything possible - abosrptions, elutions, Warm auto workups (unless recently transfused), enzyme, 0.2M DTT pretreatment, ag typings . . . all my techs were generalists.  I did the more complex abids, I had 3 panels to use, did my own enzyme pretreatment (papain) - don't like the purchased enzyme panels.  My current employer is ~300 beds and also includes a 40 and 20 bed insitutions.  Just do basic abids, sends everything else out.  That is one of the reasons I am here . . . to bring on board advanced serologies. . .

I had all the routine antisera except Lewis's but you may want to limit your stock as it is expensive. 

[kirkaw]

I like your thinking David. How long ago did you work at the smaller facility where you did all your own serology? I too, worked in a hospital where we did all that stuff, but it was the late 80's and early 90's before healthcare reimbursement went down the tubes. I really think I'm going to be asked to do a full scale cost analysis on this before attempting to bring all that testing in-house, and I was hoping someone else had done that in the past 5 years and could give me some ballpark estimates. 

[amym1586]

We are 165 beds.  I feel like we do just above the minimum.. We do type and screens/ antibody ID, antigen typing and Poly Dats.  We only send out about 5 samples a year to a reference lab.  I don't think it would be cost efficient for us to bring any of those tests in house. 

[Malcolm Needs ☆]

As a Reference Laboratory person - and in the UK - I'm staying well clear of this discussion!!!!!!!!

[goodchild]

>340 beds, urban, community teaching hospital. All techs on all shifts are designated blood bankers. Reagents we keep on hand: Resolve A & B 0.8%; Panocell 16 & 20; PEG; LO-ION; C, E ,c, e, K, Fya & b, Jka & b, Lea & b, A1, M, N, S, s antisera, and Elu Kit II. I've also considered bringing in Resolve C. We do >50 AB IDs per month.

We try to send very little to reference and screen our own inventory for antigen-negative units when feasible. The focus on in-house antigen typing intensified last fiscal year when our budget was cut a few thousand, and it was cut another 40% this year. I try to bring all the technologists into the discussion as motivation. As an example: for us four units of R1R1 blood from the reference lab costs the same as seven vials of E or c antisera.

Edited July 14, 2016 by goodchild
Added an example

[kirkaw]

Goodchild, we keep all the same reagents that you do except the Resolve panel C. Do you do adsorptions? We have a lot of warm-auto patients and that is the one thing we consistently have to send out. We keep an EGA kit too.

[goodchild]

There would be no way we could keep anyone competency assessed for adsorptions. We don't keep EGA, CDP, WARM, or RESt.

[David Saikin]

On ‎7‎/‎14‎/‎2016 at 5:43 AM, kirkaw said:

I wasa I like your thinking David. How long ago did you work at the smaller facility where you did all your own serology? I too, worked in a hospital where we did all that stuff, but it was the late 80's and early 90's before healthcare reimbursement went down the tubes. I really think I'm going to be asked to do a full scale cost analysis on this before attempting to bring all that testing in-house, and I was hoping someone else had done that in the past 5 years and could give me some ballpark estimates. 

I left there 3 weeks ago . . . never did a cost analysis.  I did workups for all the small hospitals around me.  I used to think it was cost effective, then found out later that my lab manager was only charging $40/workup to the other hospitals . . . anyway.  That was one of the things the Medical Director brought be there to do so the cost was irrelevant.  Patient TATs were improved by over 24 hrs in many cases.  Surgeries and transfusions weren't delayed, at least not significantly.  As usually happens with hospitals, every year there was a 2-4% markup.  The bean counters would tell me that costs were increasing, but my reagent prices weren't and I wasn't getting any salary increments . . . I always tried to stay a little bit cheaper than the reference lab.

[David Saikin]

2 hours ago, kirkaw said:

Goodchild, we keep all the same reagents that you do except the Resolve panel C. Do you do adsorptions? We have a lot of warm-auto patients and that is the one thing we consistently have to send out. We keep an EGA kit too.

I would do PeG autoabsorptions but could not do the allos (after pts were transfused).  There was no way to realistically type R1R1, R2R2 and rr untis to do these.

[Malcolm Needs ☆]

2 hours ago, David Saikin said:

. . . I always tried to stay a little bit cheaper than the reference lab.

I say, careful David; we have to pay mortgages too!!!!!!!!!!!!!!!!

[David Saikin]

On ‎7‎/‎15‎/‎2016 at 8:45 AM, Malcolm Needs said:

I say, careful David; we have to pay mortgages too!!!!!!!!!!!!!!!!

Me too! 

 

[AMcCord]

We are a 150 bed hospital, level 3 trauma designation and rural with a busy Oncology center and lots of Ortho surgical cases. We send as little as possible out to the reference lab. Sendouts require turn around times about 24 hours minimum, possibly more depending on the time of day we decide to send it out (its all about when the courier runs) and the complexity. Everyone but me is a generalist. Most of them do antibody IDs and antigen typing. I do the complex workups, elutions and  can do adsorptions. Once a patient with a warm auto has been recently transfused, I send them out. I stock a lot of antisera, though not anti-Le a or -Leb; LISS; PeG; 3 panels and Ready Screen 3 for the Echo; Panoscreen III and Panocell 20; WARM; ReST; and chloroquine. I used to stock Ficin, but stopped because I just didn't use it often enough to justify keeping it. I've played with EGA and liked it, but don't stock it because I don't work up enough warm autos to justify it.

[ANORRIS]

I may be in the wrong "blog", but my question is, how many of you XM units first then type the compatible ones for the necessary antigens as opposed to antigen typing the units first then cross matching the antigen negative units?

[Malcolm Needs ☆]

 

13 minutes ago, ANORRIS said:

I may be in the wrong "blog", but my question is, how many of you XM units first then type the compatible ones for the necessary antigens as opposed to antigen typing the units first then cross matching the antigen negative units?

From my point of view, as an ex-Reference Manager, it is a it of a chicken and egg question, in that we knew the "full" type of about 25% or more of our inventory (usually about 400 to 500 units of known type), so we would cross-match these units and, at the same time, just check the type serologically.  Also, it must be remembered that 100% of our units are typed for ABO and D (obviously!), but also for C, c, E, e and K.

[ANORRIS]

51 minutes ago, Malcolm Needs said:

 

From my point of view, as an ex-Reference Manager, it is a it of a chicken and egg question, in that we knew the "full" type of about 25% or more of our inventory (usually about 400 to 500 units of known type), so we would cross-match these units and, at the same time, just check the type serologically.  Also, it must be remembered that 100% of our units are typed for ABO and D (obviously!), but also for C, c, E, e and K.

I can see that from a reference lab,  thanks!

[goodchild]

3 hours ago, ANORRIS said:

I may be in the wrong "blog", but my question is, how many of you XM units first then type the compatible ones for the necessary antigens as opposed to antigen typing the units first then cross matching the antigen negative units?

That might happen depending on what antigens we're talking about and the strength of the patient's antibody, but generally no.

[AMcCord]

With some patients with stronger antibodies it could save on expensive antisera (anti-Jka, -Jkb, -Fya, -Fyb, -s), but quite honestly, unless its an antisera we use more frequently (anti-c, -E, -K), we are going to end up outdating some of it so we just use it. Might be a different answer if I had a lot of patients with antibodies. I do save patient plasma in our ultralow freezer if they have antibodies for which antisera is not available and use that to prescreen units for that patient.

[MAGNUM]

When I do an antigen typing on a patient that also has a crossmatch, I crossmatch the units at the same time as I do my Antigen typing