Sunday, March 27, 2011

Critical Incidents in the Research Laboratory

My university has a course for undergraduates which leads to a degree called the Bachelor in Medical Laboratory Sciences (BMLSc).  It was originally designed as a program for medical laboratory technologists to re-enter the academic stream, however in the last 10 years I suspect there has not been a single RT that has enrolled.  Most of the students are from from a science stream with no background in clinical laboratories.

It is a typical 4 year bachelors degree program.  Some of the graduates continue through a MSc or PhD, or apply to Medical School.  Many do a reverse entry and go on to take RT training to find work in the clinical laboratory.  Some  find work as research technologists either in the university or private sector.
 
I provide them with a lectures in Quality and Laboratory Safety. The laboratory safety lecture is straight forward, but I regularly have a dilemma about how to cover in the Quality presentation.  It is a subject about which they have little to no knowledge, or experience, or a frame of reference.   It is relatively easy to give a context for those interested in ending up in a clinical laboratory, but there are no quality management standards or even best practice guidelines actively in play in the research laboratory setting.  Today the only quality requirements that they might see are those associated with  US FDA and pharma research.

The approach that I plan to take this year is to frame Quality Management concepts as the only organized approach to prevent critical incidents.

In the clinical laboratory we define a Critical Incident as a result that is associated with inconvenience or harm to a patient that results from a faulty or significantly delayed report.  If we start from the perspective of harm, then perhaps an equivalent definition for a Critical Incident in the research laboratory is an event that results in the production of non-reproducible results that contributes to damage to the laboratory finance or reputation or credibility.

Personally I think this is a good starting point because it provides a basis for introducing Shewhart (Quality Control) and Deming and (PDSA  and Continual Improvement) and ISO 9001 (Training and SOPs, Document Control).   It also creates a basis for implementing programs for Cost of Poor Quality, since the Principal Investigator (PI) is responsible for monitoring grant funding.   And most PIs would consider it a sad point in their career  if the only jounal that would accept their data was the Journal of Irreproducible Results ©.

Phillip Crosby and DIRFT probably works better as a research laboratory concept than in either manufacturing or in service or in the clinical laboratory because the the environment and task-at-hand are easier to define and control, and the expectation of doing it right the first time is easier to implement.  One might argue that DIRFT has no place in the  trial and error process which is the fundamental raison d’etre of the research laboratory, if the procedure is executed properly as planned, then one has confidence that the result is credible, even if the result is negative.  If the process was done right the first time, the outcome can be both negative and positive at the same time.

And finally, it is critically important for this group of budding scientists to understand that quality is a performance issue, not an accreditation issue.  It would be reasonable for them to at least know there are ISO standards that can help their laboratories implement some form of cobbled quality system.

So that is the plan.  I will be interested in seeing how the students respond.

Friday, March 25, 2011

Quality is never an accident...

I am giving a number of presentations in the next little while, so I thought I would look for some quotations that might stimulate some juices.   One that comes up frequently is “Quality is never an accident; it is always the result of high intention, sincere effort, intelligent direction and skillful execution; it represents the wise choice of many alternatives” ascribed to a William A. Foster.    I suspect that all of you have heard this one at one time or another.

My first thought was it was typical a most feel good quotations in that it contains some grains of relevancy, and is based on good intentions, but is pretty much self indulgent.  I agree about quality never being an accident and the part about taking effort and direction, but the rest is more than a little excessive.   It might be something that Deming might have thought, but never would have written (too pedantic).  It is the sort of thing that folks quote, but rarely if ever actually say.

What interested me was that I have never heard of William A. Foster in the historical trail of quality; so I turned first to Wikipedia to at see what I could find.  What interested me was that the only person with that name cited is a fellow who earned a Medal of Honor for falling on a grenade in the Battle at Okinowa in 1945 when he was 28 years old.  An admirable act to say the least, and perhaps an action that someone with grand thoughts on quality might say, but unlikely considering he worked as a factory laborer prior to his stint in the Marines in WWII.    Perhaps the quote was not so much BY William A Foster, but more ABOUT him.  That being said someone was convinced that this William A Forster wrote the quote and linked his picture to the quotation.

Turns out that I was not the first person ask the question.  One research found an advertisement in the New York Times for an advertising company with the exact quote in 1939.  At that point William A was 21 years old, so that nailed it for me that whoever the writer was, is was not William A.

With a little more searching I found another possible and more probable candidate.  Turns out there was another William Foster, born in 1879 with the middle name Trufant.  Trufant was probably his mother’s French family name.  Born in Boston, educated at Harvard (1901), Professor in English and Argumentation (Debating), PhD from Columbia in Economics (1911), and an author of a series of  books on Money, Profits, and Business, most of which were written in 1928, which was when he was 49, and about 11 years before the aforementioned advertisement.

If I had to chose between William A,  a factory worker and Marine who died valiantly  in a war at a very early age, or William T, a pedantic type guy with a background in English studies and debating and economics,  as the likely author of a grandiose view of quality, I would personally guess the latter.  

But I could be wrong.

Sunday, March 20, 2011

Dealing with complaints in the medical laboratory

There is a lot to like about Quality in the laboratory setting.  Like a lot of the laboratory process, it has some elements strongly grounded in evidence.  PDSA cycle as expressed by Deming and practiced widely is a written in stone as a proven, effective, indeed definitional approach to Quality Improvement.    
 
There is other stuff, a level down in which the concept may not be as definitive, but the probability of an established linkage is high or very high.  Working with Quality Partners is an example.  Working the partners is a powerful idea, and likely contributes to improved Quality and improve laboratory care, but this not established.



This entry is about the a concept one step lower;  Organizational Jidoka.

Jidoka is a part of Lean which automatically raises one of two probable reactions; some will automatically buy in, and others will reject almost out of hand.  But hold on.  
Jidoka has two parts, one derived from Poke Yoke, which means error-proofing.  To the extent possible, error prevention and error detection should be built into the hardware to either make it impossible to put a part in backwards, or to signal as soon as the bad thing is either about to happen, or more typically as soon as it is detected.  That is the easy part.  The second part is that the person responsible for that part of the manufacturing process is also charged with the responsibility to sort out what went wrong and fix it immediately, or to stop the system and go get help.

Recently I was reading in the Quality Digest an article by Norman Bodek and Jeremy Green (Zenjidoka, Solving Toyota's Quality Problems: A simple tool for a complex problem - Quality Insider 02/14/20) who suggest that Toyota would have been well served if every employee been given jidoka authority and responsibility to pull the red cord and stop everything.  The argument goes, that had everything been stopped when the first accelerator accident was reported,  than most of the others  would not have happened and Toyota would not have incurred all the costs associated with accidents, injuries and litigation.

This sounds like an interesting  idea, and I suspect it is probably both effective and efficient in the manufacturing setting.  But outside of the manufacturing line, this sounds like a truly scary, maybe even a dangerous concept.

With medical laboratory, it is not uncommon for automated analyzers to falter.  Usually this will trigger an alarm, just as with the manufacturing station, and at that point usually the technologist has the authority to halt function until the problem is solved.  The biggest problem in the examination phase is not when the alarm goes off; it is when the problem occurs, but the alarm does not go off.  We see this all the time with splashes, carryover, and all sorts of problems that cause screwy results.  Often equipment errors are transient and situational and don’t get detected until the clinician calls back and says the results make no sense.  In other words, most complaints are received way downstream from the causative examination phase event.

Another example can be seen when the error occurs in the pre-examination phase when the person entering accessioning information is aware of a typographical problem, they stop, correct the information and go on.  Jidoka works here too. 

But what would happen if you gave people in client services  who receive complaints the same authority.  Client complaints occur after-the-fact and most are diffuse in nature.  Many of them result, not from an analytic error, but from misinterpretation or misunderstanding.  Most resolve early and easily.  Some need a process corrective action, but only a very very few need something definitive done now.   

But the person receiving the call is not in a position to make that judgement.  If they were given the responsibility  to make  stop-action decisions it would be a lose-lose almost every time.  The first loss would be the production process would be constantly halted, and the second loss would be the impact on the person making the decision.  Everyone would be unhappy all the time.  

If there is a bigger point here it is that all too often, organizations in general, and medical laboratories in specific have no constructive approach to gathering and addressing complaints in a  rapid fashion.  Most get a phone call solution without any documentation, or collation, or evaluation or Management Review.  In our business this is our traditional standard practice but it doesn’t make it appropriate. 
If complaints were fed forward to an authorized and responsible person in a timely fashion, who could make a thoughtful and focused review the lose-lose becomes a win-win.  If someone is going to be called to account, it should be management.

So there is a message here, but it is not to adopt the concept of Jidoka as an approach to complaint resolution, just because it might be a useful component of Lean strategy in another sector.  In the medical laboratory we should not be looking to  solve problems by assigning a line-person to do management’s work.  We  solve problems by making change and requiring  management to make management decisions.

Thursday, March 17, 2011

Certificates and Certification: Recognition that is fit for purpose

The other day I saw an interesting promo in my daily perusals.  The chair of a large Certification Board was trying to point out that the certificates are handed out by their board were  a form of recognition of achievement, having participated or completed a course.. In their context certification was a form of recognition of “mastery of a body of knowledge”.  Yes, that makes sense, but to get there it seemed necessary for him to take a shot at organizations that provide certificates based solely on participation.  The message was clear, they would never provide a certificate only for participation.

The discussion struck me on a variety of levels, in part, because it is the area in which I live all the time.  In our proficiency testing program we provide a Certificate of Participation so that the laboratories can indicate to their respective health authority or accreditation body that they are participating in proficiency testing.  The certificate does not guarantee that the laboratory is proficient, but does indicate that they are engaged in external quality assessment.  In our Certificate Course for Laboratory Quality Management we provide another kind of certificate, but in this case a Certificate of Achievement which is based on participation in discussion, completion of assignments, successful grades on quizzes, and a final examination.  We know by the end of the course that folks are engaged, interested and knowledgeable.    With CMPT being certified each year to ISO9001:2008 by assessment we are thoroughly visited and inspected and demonstrate our compliance with the standard, not only on paper, but in action.We receive our Certificate of Assessment each year as one of conformance and achievement.

So I understand the complexities of the process inside and out.  What got me going was the message that certification was not only different from receiving a certificate of participation, but rather, that it was better and proof of a higher level of quality.  And that is an attitude that I have heard all too often.  It is disappointing that it is a message received from inside an organization that promotes itself as  committed to Quality.  

The point that got missed along the way is that we have a variety of ways to provide acknowledgement or recognition.  We can provide Certificates for participation, Certification by exam. Certification by assessment. and finally, Accreditation by assessment. It is not that one is higher and the others lower, or that one is superior and the other inferior.  Recognition has a variety of tools and all are different and all need to be  fit for purpose.  

Somewhere along the way folks have developed a notion of assessment that is more about competing interests than being about acknowledging achievement and recognition.  I understand that the chair of a certification board perceives that what they do is important.  I got that and I agree that it is.  But is his certification meeting client needs or is it designed to fit the needs of the certification board?  
There are lots of people or organizations, for example,  whose need is to be able to demonstrate that they attended a meeting or participated in a program.  That is the sole point of the exercise.  For them a piece of paper with a gilded frame and a title Certificate of Participation is exactly what they need and want.  We at CMPT have been asked in the past if we would consider creating two certificates, one for participation and another for excellence if the laboratory got all its challenges correct.  But  upping the ante to a Certificate of Excellence would not only exceed the need, but may damage the value.  In proficiency testing for example, it is pretty clear that the more expectations that are put on the line, the more likely organizations participate in “gaming” or deception or special procedures to increase the likelihood of a correct  answer.  So the process become less about improving quality and more about getting a good grade.   Its like being in Grade 4 all over again.  

For years I have listened to accrediting bodies drone on about how accreditation provides a  higher level of assurance of quality because accreditation bodies address competency while certification does not.  There is a term for this; bunk.  I have seen and experienced the work of really good accreditation bodies and also seen the effects of really bad ones.  Recently, for example, a colleague told me of a laboratory he visited that was very proud of its 15189 accreditation.  What they had was a room off an office with a shelf with their quality manual, and procedure books.  Not much evidence for quality there.

Now I am a pretty competitive type guy, and I am pretty comfortable with the concept of excellence and recognizing achievement.  Let’s just make sure that the process that we develop is actually about meeting the needs of the client.