CAPA / Deviation

[RR1]

If anyone has tips/ docs on how they are managing their CAPA system - or requires info, would be good to post it here.

My main problems are trying to close the incidents in a timely way- it does become quite a challenge. it would also be interesting to know what databases/ systems folk are using for this function. I have had to produce an excel spreadsheet for this, but extracting the information required for the annual Blood Compliance Report- has been difficult....obviously I now need to improve the system ready for next years BCR!.

[TStec]

We use a form called Episode Report and I created an Access Database to input the data and categorize by systems. These were previously tracked with a work document where our manager would log in all reports after completion and review. I presented at ASFA annual meeting http://www.apheresis.org/%7EDOCUMENTS/Fri_230_Stec_Galleon_II_&_III.pdf hopefully will link you to the presentation. Our process was on AABB Commendable Practices several years ago, it has been updated since then. Our hospital subscribes to UHC Patient Safety Network for event reporting. The transfusion portion was not good but I have helped develop that with the categories from the database I created. If you need more information e-mail me theresa.stec@baystatehealth.org

[jcoburn]

Great information, Thank you.

[RR1]

Thanks Theresa.....i'm sure your presentation will be a great help.

I want to create an access database for my incidents/ deviations too (once I figure out how to use access). I am currently placing these on an excel spreadsheet- in a workbook format on re-design this will include my audits on a separate tab and reference numbers to any incidents I may have escalated to senior management for actioning.

I have attached a blank version + SOP + reporting template if of use to anyone.

capa proc 100.doc

QI Excel database.xls

Quality incident report form BT-F-BT-002 v21.doc

[pluto]

excellent docs

like the incident form , we have pathology wide one and was going to make a more transfusion specific one and this is good, would probably include time as well

Some staff are touchy regarding names of who made error do you anomise this or do you not find it a problem

[RR1]

When I began at my current job, the system was anonymised- but i'm a firm believer that all mistakes and errors should be transparent. We now use names....and there are some forms with my name on them too ( the staff love it when they can name me as the cause!)

We all make mistakes, the CAPA system is about learning from problems and improving things- not about blaming.

[pluto]

I agree with you , no one is infallible

everyone will at some point make a mistake - hopefully systems in place will prevent it being a big one

[scubasammy]

If your looking for a quality management system to manage incidents, investigations, capa and reports, you may want to look at trackwise. its being used at 1/2 dozen blood centers in the states and is widely known around the life sciences world www.sparta-systems.com

[bmsjbatt]

Rashmi thanks for your very useful info it has helped me understand better what I need to do, will share my other stuff

regards

Jo

[RR1]

Thanks Jo, nice to have you on the forum. i'm not saying any of my docs are correct- i'm trying to build an understanding of and improve my systems too ! Be great if you could share your stuff, and spread the word about the quality aspects of this forum to others.

[Nicki]

Thanks for the docs Rashimi. We use an excel spreadsheet for recording our corrective actions - I have recently got our IT guy to make the colors change to yellow 1 month before it is due and red after it is due so it is easier to see at a glance what is out of time (last years database had so many entries and counting them was proving a nightmare). We discuss each incident each week at our TMG meeting. We add incidents to it from audit and an access database we run in transfusion to catch the information. Unfortunately it is too big to post. I am trying to roll it out to the rest of haematology but they are a bit resistant! We are trialling an incident form similar to the one Rashimi has posted, with the idea that the person involved does their own investigation in the hope they will have an understanding of how much is involved in root cause analysis.

Nicki

[RR1]

Hi Nicki,

Thanks for the info...how did your IT guy make the colours change with time period ? I would really like to use this function. Also are you ensuring your database is password protected to stop unauthorised alteration of reports? I gather that with Excel you can only protect each spreadsheet separately (if these are together in a 'workbook' format with different sheets on different tabs).

Haematology folk are always a wee bit resistant.....you just have to persevere!

Nice to have you on the forum- spread the word, all the information we can collectively gather will help everyone improve practices.

Edited May 13, 2009 by RR1
.

[Nicki]

Hi,

If you go to conditional formatting in the format menu, and the condtions in the cell for the date completed are:

Condition 1 cell value is between =TODAY()+31 and TODAY()+14

Condition 2 cell value is less than=TODAY()+14

Then you just select the appropriate colour - we have orange for condition 1 and red for condition 2.

Ours isn't password protected as we are trying to encourage its use so staff can see we are doing something with the non-conformities they find with their audits.

Nicki

[RR1]

Thanks Nicki- you're a star!!!!!!...going away to play with my database now-how sad is that?

[pluto]

conditional formatting is fun once you have cracked how to do it

spent hours one day conguring a spreadsheet for mandatory training , green if trained in last 12 months and red if greater than 12 months and amber if on month 11. you can also set it to compare to current date ie date you open sheet

If memory serves me right you can have three rules whcih means three colours per column / row

google conditional formatting and you will find examples of how to do it

[pluto]

Grading of risks for incident forms Low Moderate High

What would you consider would be good examples of each

? Low : not signing fridge chart daily as stated in sop, not entering unit into stock which is found later, missed sample rejection found by second checker

? Moderate :significant errors found in lab but not gone out of lab ie antibody misidentification found on checking. XM labels transposed but found before issue- same patient but wrong donation No

Leaving expired units in stock fridge

? High : , wrong blood issued, red cell antibody not detected, any reported misgrouping, leaving expired units in issue fridge available for collection

I think I am ok with Lows and highs ? but more moderate examples needed

[RR1]

Looks good, the thought process is a bit more complex I suppose when you are trying to teach staff to assess on this basis.

moderates :

Missed training deadlines

Delayed equipment maintenance

Delayed document reviews

Delay in testing/ blood provision (depending on the situation)

i'll look through my QI folder tomorrow- and get back.

[Malcolm Needs ☆]

Hi Rashmi,

In the NBS (NHSBT) we use something called QPulse5. It is excellent, but I would not be the correct person to explain it.

Perhaps if you give a call to the Quality Department at Tooting, they would be able to give you more, and more useful information.

Edited May 23, 2009 by Malcolm Needs
I can't spell!

[RR1]

Hi Malcolm, QPulse is meant to be wonderful for all of this...but i'm not allowed to have it!!!- and have to put up with a system that is vastly inferior and time consuming.......... one day things WILL get better- I live in hope, but for the time being i'm even having to trade services to buy pipettes in my lab !!

Don't even ask about the Typenex wristbands i've been trying to buy in for the last 5 months- trying to spend £150 is a nightmare......so if anyone wants me to map their fridges in exchange for lab equipment-feel free to contact me.

[RR1]

Hi Pluto,

Have looked through my data and it seems the 'medium' errors are around checking type of problems, errors that had a potential to have caused a 'delay' type of problem, GMP type errors, identified missing SOPs.

[RR1]

I get the feeling that folks are closing their CAPA at different stages of the process. Currently we review our incidents and hopefully the initial actions (remedial) will have been completed within a few days (or sooner) depending on severity. At review, further corrective actions are assigned if needed, some of which may take many weeks to complete.

Is it acceptable to close a CAPA after the initial corrective actions have been performed and call the further actions needed 'quality improvements' which can then be assigned a different time line?

I would really appreciate ideas and suggestions of how your lab manages this process,

many thanks.

Edited February 7, 2010 by RR1

[scubasammy]

Rashmi,

Our software is used by the top pharma, medical device, CROs and blood centers around the world for amongst other areas CAPA. In almost all cases the process is 1) corrective action 2) preventative action and 3) effectiveness check. CAs are short term immediate fixes, while an effetiveness check may not happen for weeks or months. Our recommended 'best practice' is to close the corrective action once completed, open a seperate preventative action referencing the original incident(s) and open an effectiveness check on both the CA and the PA. Some companies will allow the re-opening of a CAPA, but this should be only used when the original information (investigation results, root cause, etc.) needs to be ammended or changed. The agencies and management will be looking for how you corrected the problem, how long it took during each step (investigate, approve, implement) and how effective was the correction. If you leave a CAPA open for an extended period while awaiting an effectiveness check the metrics will be skewed.

[RR1]

Thanks scubasammy that makes a lot of sense, will definitely help with my Quality metrics....and possibly my sanity!!

Are other folk doing this too?

Thanks

[Malcolm Needs ☆]

Thanks scubasammy that makes a lot of sense, will definitely help with my Quality metrics....and possibly my sanity!!

Are other folk doing this too?

Thanks

Far, far too late!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

:rolleyes::rolleyes:

[RR1]

Having further thoughts about this, surely CAs and PAs are all really corrective actions so should the deviation not remain open until both are complete. Could these then be closed and trended for the effectiveness check? -I am sure there are many ways we could do this, i'm just trying to find a simple one that works for my lab.

I think that our regulators expect CAPA ( unsure if this is just corrective actions or preventive too???) to be completed within 30days ( or sooner for more critical incidents/ later for more complex ones as long as this is justified).

So as Malcolm has mentioned...this may well be too late to save me from insanity!

Edited February 11, 2010 by RR1