Standards




I think Charles is looking for ways to improve the quality of the research - weed out the Dr. Lu's.


COLOSS, from my perspective, is an effort to impose standards.  As guidelines, they serve a purpose, but being assembled by committee's, I find two problems:


1) Whoever dominates the discussion 'sets' the guideline/standard. 
2) Lots of agencies and groups try to impose the COLOSS procedures on their researchers - rather than being used for guidance, it becomes the bible that 'thou shalt not disobey'.


     (I've heard the complaints long and loud from colleagues that I highly respect - the 'guidelines' part is forgotten, and their bosses are covering their behinds ('we follow COLOSS').  Personally, I find a number of COLOSS methods out-dated or poorly suited to the task.   It's a good starting point - it should not be the 'required' method(s).


GLP - Good Laboratory Practices


These are aimed at ensuring that there is an estimate of the accuracy and precision of each and every measurement, and when a project is audited, the researcher has to provide records that allow an auditor to assess these parameters.


1) It costs more time and money, but it at least provides a means of knowing whether the laboratory follows best practices.
2) It will not make up for a poor research design.
3) Anyone who says they can not do GLP level research shouldn't be allowed to publish.  
 (Full blown GLP is over kill for exploratory studies, but these studies should be performed in 'the spirit of GLP' - a vague term, but one used to indicate that the investigators should at least check their measurements)


SOP- Standard Operating Procedures


Properly done, these address some of Charles's concerns.  Before beginning any project, the investigator or investigative team sits down and outlines the experiment, agrees on how things are going to be done, lists materials and methods, etc.   It's like a contractor setting up to build a house - who are the suppliers, what materials are needed, when are they needed, etc.?  And like a general contractor, there are inspections to see that the job is done right.  In the contractor scenario, it's building and public works inspectors.  In a research setting, the SOP should specify who is responsible for seeing that the trials follow the SOPs.


(SOPs can be modified, but the modifications need to be added to the original plan with a date, identification of the person altering the method, and initials to show that the Principal Investigator or the designated person for monitoring adherence to the plan has approved the change.)


Everyone on the team should have a copy of the final, signed SOPs.   For complex experiments, there may be more than one.  In a mature lab, one can find an SOP for each and every common method - one that the lab developed, everyone agrees upon, and which is signed off (approved) BEFORE any measurements are made.  


Now, the common response is - I don't have time to develop SOPs.    Far as I am concerned, you can have your Pitty Party, but if you don't have time to properly plan an experiment, you shouldn't be pretending to be a scientist.


And yes, SOPs become more or less binding -- to the investigative team, but they are binding themselves to follow their own plan, and the signature of approval indicates that someone, whose job it is to know good procedures, has signed off on it.    


Our lab has a draw full of SOPs.  We keep a log book of each of them.  We modify or add SOPs for new projects, as appropriate.  New members of the team get up to speed quickly.  'I forgot' is not an excuse.


Still, none of this answers Charle's concerns.  For example, years ago, I was involved with the National Institute of Standards and Technology (aka, National Bureau of Standards) Specimen Banking Program.  The idea was to store frozen samples of things like common human diet components, bird eggs, liver tissues - in order that years down the line, if someone says - we're finding a lot more of chemical x in ....., one could go to the bank and retrieve a stored sample for comparison.  


This was an international effort, and the Germans insisted on Standard Specimens.  That was fine if one decided to collect and store eagle eggs or human liver samples (which was done), but it fell apart when the Germans insisted that the plant and animal specimens had to be specific species that were common to Germany, but at time didn't occur elsewhere.  Also, they liked clones.  It took three more meetings to agree upon a list of animal and plant species that were reasonably widely distributed around the world, and where, if not the exact species, at least a counterpart was available.


For example, the European honey bee would be a better choice than something like one of the seven native bees from HI that are now listed as endangered or threatened. 


Which brings me to what I think Charles may be looking for, a list of Standards or Guidelines to improve the quality of research.  Something that peer-review is supposed to do somewhat after the fact, but we've all seen that fail.


So, I'll give this a shot.   For pesticide investigations, I have a soft recommendation and a required recommendation:


Soft - the team should have a chemist involved - too many pesticide experiments are being conducted by people who jumped into this area, crossing over from other disciplines, with not real understanding of the chemistry.  Lots of recent graduate students have been poorly served by committees that don't have the proper background.  Ideally, someone on or accessible to the team should also have a background in toxicology.


Required - all investigations of 'new or newer' pesticides like the neonicotinoids or modern fungicides, that intend to address risks to bees, should include a positive pesticide control.  Remember Reed Johnson and his fungicide study - it wasn't the fungicide that was the problem.    And if the investigator intends to claim that the 'sky is falling' or that the product should be banned, I want to see how the chemical of concern compared in the same trial to a positive control chemical, one known to be harmful to bees, and this should be one of the alternative chemicals that would be used if the primary chemical being tested is banned or removed from the market.


All insecticides are poisons.  The dose is important - we all know that.   But, finding that neonics alter bee behavior isn't surprising.  What is surprising is that the idea that many other, possibly all other poisons, would also alter bee behavior is completely forgotten.  Especially, since many of the older pesticides are neural poisons.



















             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html