The effect of Fiber Counts a preliminary study

Our plain-English summary of The effect of Fiber Counts a preliminary study, by A.A. Winer, M. Cossette (1979).

The study

• Authors: A.A. Winer, M. Cossette
• Published: Annals of the New York Academy of Sciences p661-672 (1979)
• Type of article: Journal article
• Our copy: from Johns-Manville's own medical library. The article is still under copyright, so we describe it here rather than post it.

In plain English

This was a preliminary round robin study of the NIOSH method for counting fibers with a phase contrast optical microscope. Two membrane filter samples from each of six asbestos industry plants (asbestos-cement, brake linings, mining mill and textile) were counted by government, research and industrial laboratories. The laboratories sorted fibers by length and by aspect ratio (the ratio of length to width: 3:1, 5:1 or 10:1). One laboratory also counted and identified fibers by transmission electron microscopy (TEM). The article's conclusion section reported that relative error fell from 93 to 42% when a 10:1 rather than a 3:1 aspect ratio was used, although about 45% fewer fibers were counted. The article's summary instead stated that 5:1 appeared better than 3:1. Practically all fibers seen by TEM were chrysotile. For fibers longer than 5 um, optical and TEM counts were correlated (correlation coefficient 0.978), with about 1 fiber counted for every 50 present. No such relationship was found for shorter fibers. The authors concluded that a higher aspect ratio should improve precision, to be checked in a planned second round.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Preliminary interlaboratory round robin study of fiber counting methods (phase contrast optical microscopy), with a parallel transmission electron microscopy comparison at one laboratory
• Who was studied: Twelve membrane filter samples, two from each of six plants, counted by participating laboratories. The introduction lists two government, three research and six industrial laboratories, mostly in Quebec and Ontario plus the U.S. Bureau of Mines. The TEM work was done at the Ontario Research Foundation. Dates of sampling are not given.
• Exposure: No human exposure was studied. Fibers on membrane filters taken from machines in asbestos-cement, brake lining (carding and press), mining mill and textile (twister and blender) operations. TEM identified the fibers as practically all chrysotile.
• Main results:
  • In the conclusion section, the authors reported that optical fiber count precision increased with a 10:1 instead of 3:1 aspect ratio criterion. Relative error decreased from 93 to 42%, while the number of fibers counted decreased by approximately 45%.
  • The article's summary table gave mean counts and coefficients of variation by counting criteria, as printed (some row labels are OCR-damaged): 3:1 with 5 um minimum length, mean count 100%, coefficient 0.43; 5:1 with 5 um, 80%, 0.38; 10:1 with 5 um, 53%, 0.54; 3:1 with 10 um, 38%, 0.49. The summary stated that 5:1 appeared better than 3:1, and that a 5 um minimum length gave better precision than 10 um.
  • Among the optical counts, the variability between counters for the aspect ratio groups was much greater than for the length groups.
  • Counters reported that many particles in the 3:1 and 5:1 aspect ratio groups were not chrysotile, judging by their prismatic appearance, yet all were counted as asbestos.
  • By TEM, practically all the fibers on the samples were chrysotile.
  • For fibers longer than 5 um, optical and TEM counts were related, with a correlation coefficient of 0.978.
  • For fibers longer than 5 um, the ratio of fibers counted optically to fibers present by TEM was about 1:50.
  • For fibers shorter than 5 um, no such relationship was found. In these shorter lengths, the average optical-to-TEM count ratio was 1:1000.
  • Of TEM-counted fibers longer than 5 um, the majority in every sample had aspect ratios greater than 10:1, ranging from about 60% to over 80% per sample.
• The authors' conclusion: The authors concluded that optical fiber count precision increased with a 10:1 rather than 3:1 aspect ratio. They suggested that raising the aspect ratio to greater than 5:1 should increase precision, which they said could be verified in the next, statistically designed round robin series.
• Limitations the authors state:
  • This was an initial series meant to identify problem areas. It was not the statistically designed round robin (allowing analysis of variance) that was planned to follow.
  • Some slides were too dense and some too light, causing counting difficulty.
  • Assigning aspect ratios to fibers 5 to 10 um long was difficult when fiber diameter was small.
  • Failure to adjust graticule marks to the exact length required could cause a bias that depends on both counter and microscope.
  • The optical microscope does not allow all fibers to be counted, particularly very short and thin ones.

Topics

Medical terms in this study, explained in our Lexicon: NIOSH, Asbestos-Cement, Brake Lining, Aspect Ratio (Length-to-diameter ratio), Transmission Electron Microscopy (TEM), Chrysotile.

Browse return to medical research.