Trending Incidents

[RR1]

Please could you tell me what categories you use to for trend analysis of your deviations/incidents/complaints

Thanks

[bmsjbatt]

Hi Rashmi,

Recent Inspection, we were commended on the groups we have been using so far, but were also recommended to clearly separate the incidents identified internally and those identified external to the department.

our groups so far are.

CodeTrend analysis code 2010

A Personnel

B Reagent

C Equipment

D Training- lab staff

E Process/ Failure to follow Trust Guidelines

F Communication

G Concessionary Release / deviation

H DelayIRecall

J Wrong patient bled/ details written on

K Training- porters/ nurses/ drs/ drivers

L Traceability

M Hospital patient numbering problems

N SLA Hospital

O NBS errors

P Product

Q Winpath

R Complaints

S Cold Chain

T Inappropriate product issued/transfused

U None assigned/ not applicable

V Clerical Errors

Hope these help.

Jo

[Eoin]

Hi Rashmi,

Main categories used here are

A = General Equipment Failure

B = Sampling Error

C = Date Entry / Clerical Error

D = Reagent / Material Problem

E = IT Problem

F = Test Method Failure; QC Failure

G = Miscellaneous Failure (Specify)

H = Product Labelling

I = Inadequate Stock Levels (Consumables)

J = Administration Service Failure

K = Documentation Control Failure

L = Security Issues

M = External Failures

N = Planned Deviation

O = Cleaning Issues

P = Non Adherence to Procedure

R = Test Result Reporting Issues

S = Inadequate Training of Staff

T = EQA Issues

U = Proficiency Testing Failure

V = Environment for Testing Failing

Sometimes needs further breakdown.

Cheers

Eoin

[RR1]

Hi Jo and Eoin, thanks for sharing your trend codes, certainly helps with further developing our systems.

Here are the categories I currently use:

A=Equipment failure (all)

B=Cold Chain failure

C=Traceability Failure

D=Process failure (Lab areas)

E=Process Failure (Clinical areas)

F=Complaints

G=Recalls

H=Concessionary testing/release

I=Transfusion Reactions

J=Quality System Failures

K=Quality Improvements

L=Audit Non-conformances.

Some of my categories tend to overlap a bit- so I may need to further specify. Thanks for the ideas!

[ABIDer]

What do you do with your information once you have it? Do you follow-up on each one and do a root cause or do you set a threshold and then once passed do a more in-depth look?

[RR1]

That's a very good point lucylou,I personally get a lot of satisfaction just looking at the nice colourful graphs! But seriously, at the moment we just review some of the reports of a particular trend to see if there are patterns.

Recently, some of our process failures indicated that patient demographic errors were repeat events, and when we looked at each of these a bit more there was a theme of either clinicians incorrectly identifying the correct number of digits in the hospital record number when completing forms and labelling samples,or the lab entering these onto the LIMS incorrectly

We have subsequently re-designed our blood request forms to make it easier for everyone to identify whether the number used is a local hospital number or a new NHS number (different number of digits) by just sequencing the boxes on the form in groups of 3 with a final single box for the 10 digit NHS number being more widely used in the UK.

I suppose we should write a summary to explain tends for each category - which should reduce if the corrective actions were effective. There obviously has to be a risk based approach to doing this due to time constraints.

I would find it very helpful to know how other handle their trending data too.

Edited November 10, 2010 by RR1

[EMB4879]

Thanks so much...I recently entered a position and am in charge of this, and there was no real system to follow, kinda was on a whim. I am revamping it and this is great!

[Malcolm Needs ☆]

What worries me a bit Rashmi, is that your list only goes from A to L.................!!!!!!!!

:haha::devilish::haha:

[RR1]

What worries me a bit Rashmi, is that your list only goes from A to L.................!!!!!!!!

:haha::devilish::haha:

Ha, Ha, Ha....I didn't think it would be appropriate to include M-Z as they were all various categories for NBS problems, with M trend = "Malcolm"!!!!!!!!!

:p:p:p:p

[Malcolm Needs ☆]

Ha, Ha, Ha....I didn't think it would be appropriate to include M-Z as they were all various categories for NBS problems, with M trend = "Malcolm"!!!!!!!!!

:p:p:p:p

TOUCHE!!!!!!!!!!!!!!!!

:highfive::highfive:

[RR1]

Further to this trending stuff, I am trying to learn about performing a Pareto analysis of our problems using the trending data. As far as I can tell, individual trends should be placed in % order of the total number (highest to lowest), and the incident categories that make up 80% of recorded problems are further analysed to direct more specific actions.

Just wondering if anyone else uses a formal Pareto analysis approach to this...any advice on getting this right would be appreciated!!!

Thanks