The question I am interested in exploring is when does POCT turn into a hot lab? Most common definitions of POCT involve testing near the patient, hence the earlier of Near Patient Testing.  Now something like urine dipsticks would be described within the traditional realms of POCT.  Blood glucose monitoring has also been a staple of traditional POCT. The boundary seems to start when you begin to consider blood gas analysis.  It is an accepted POCT test when performed within a hospital environment but very unusual when you begin to consider performing it in a GP surgery environment.

Blood gas analysis tends to be performed in a highly critical setting e.g. ITU, Casualty etc. I would suggest that these environments tend to be high pressure and are staffed by members of staff who might be more willing to embrace newer ways of doing things.  As a consequence, the analytical requirements for a good sample for analysis are, hopefully, well known.  When you consider putting a blood gas machine into a location like a GP surgery, you would likely have staff not used to the running of such equipment and may not be prepared to embrace change.

When you start to place equipment for D-dimer, Troponin, blood gas, amylase, CRP and other similar tests, you certainly start to leave the comfort zone of some members of staff. I consider that at this level of technical analysis, things have moved beyond POCT into replicating the laboratory. I think the line is has been crossed with this setup due to the technical demands of the tests rather than the tests themselves.  If these tests could be performed with the same level of ease and with appropriate analytical performance, this could still be described as a POCT setup.  I feel that the technical nature of these tests tips the balance towards hot lab.  This is why the local laboratory has had such input in the running of the setup.

From a regulatory point of view, it is vital that these kinds of setups don’t just have oversight from a laboratory but are setup as a partnership with equal input from both sides. Over the years, laboratories have learnt the lessons from mistakes in treating a patient based on the wrong unit.  Rigorous analysis of IQC and EQA has increased the performance of the analytical equipment and laboratory procedures. When a GP surgery starts to run in effect an in-house laboratory service, they could easily repeat the mistakes laboratories have already learnt from.  This is why the service needs to be a partnership, in order to provide insight and advice when required.  The lab should never lose sight of the fact that they are providing knowledge and possibly staff but ultimately the final decision rests within the surgery.

Here is a link to a Journal discussing Near Patient testing:

Journal of Near Patient Testing & Technology

From my experience of working in a variety of labs, Levey-Jennings charts are widely used to view QC performance.  Westgard, from my experience, seems to be less formally used in labs.  Though a lab may not mention Westgard within their SOP’s, people are still using some of the rules to assess their QC performance from the L-J charts.  Sometimes people use the 10x rule without realising they are actually using Westgard rules.  When they view the chart they notice a bias above or below the mean and then take action.

I have used a variety of instruments over the years and the vast majority have the ability to produce L-J charts. The quality of the information provided by the charts has generally improved over the years but I still feel that manufacturers can provide L-J charts with more information without compromising clarity.  For me, the perfect L-J chart would have a variety of attributes.  The firs would be the ability to plot multiple QC plots on the same chart.  To ensure the user can still track the performance of a single QC level, the plot would either be in a different colour for each level or some other distinctive feature.  The chart should also be able display when there has been reagent or calibrator lot changes.  I have seen different ways of conveying when reagent lots have changed.  I have seen breaks in the QC plot to indicate reagent lot change or changes in QC plot point colour.  Personally, I have no preference in how the reagent or the calibrator lot change is indicated, I feel that this should be conveyed somehow in the chart without having to look elsewhere for the information.  Having this information displayed within the chart would be able to show whether a step change in QC is due to reagent lot change or some external event e.g.  blocked probe.  Without this information, a step change in QC performance might be wrongly interpreted as a reagent lot change when the cause is something else.

I would also like to see calibration events recorded automatically in the chart.  Several instruments have the ability to log a comment against a QC point. I don’t think this is sufficient as a way of automatically recording cal events.  My problem with using comment boxes is that the comment function should be used to record a variety of comments e.g. dregs of QC.  If cal events were systematically recoded via the comment box, events like dregs of QC bottle could then wrongly seen as a cal event. I understand that the use of the comment box for recording cal events could be desirable if the instrument has no way of easily recording this information.  In a perfect world, cal events and comments would be separate to allow the recording of extra information to be distinct from cal events.

I am not sure if QC lot changes should be plotted on the same chart or not.  To allow various QC’s to be plotted on the same chart, the y axis of the L-J chart would have to be SD rather than concentration.  This would allow QC lot changes to plotted without any adverse effects from changes in analyte concentration between lot changes.  Is there much benefit from comparing two different lots of QC? Any performance differences could be due to poor workup of the new lot rather than genuine performance problems.

During my career, I feel fortunate to have come into contact with a variety of people who have shaped my and my outlook on the profession.  I suppose I prefer teaching students compared to showing experience members of staff who to use new equipment, as it allows me to try and provide the knowledge to get them to fully understand why the SOP asks them to perform certain tests.  With the march towards  ever increasing automation, I fear we may forget to ask why in the quest for ever increasing throughput.

I hope that this blog can go some way to reinvigorating my belief in teaching and possibly actually help someone.

I have made a start on the first page.  I’m not sure if I should just list the links or include the links in a essay like format.  Nothing particularly formal, but certainly more readable than a bunch of links.  I’m thinking towards the essay format but might change my mind if the scale of the task begins to daunt me.

This concern about a graduate level profession got me thinking about the change within biomedical science.  Has the change to a degree level profession meant people were unwilling to do the routine tasks?  In my experience, I have not met anyone who felt that certain tasks were beneath them.  But I get to start thinking are were now producing more biomedical scientists than available posts?

Considering the number of Universitys that run a BMS degree, there must be a signifiicant number of people entering the profession each year.  But labs still seem to be short of staff.  Is this mainly due to purse strings being drawn ever tighter or a lack of suitable candidates? From a candidates point of view, I would be reluctant to apply for a job which I have seen repeatedly advertised.  Likewise, a lab would not want the extra cost of advertising month after month.  I do wonder what will happen to labs once the highly skilled staff start to retire if there truly is a recruitment problem.

Please feel free to let me know about your thoughts about staff ing in Pathology.