Example Microbiology Measurement Uncertainty Calculations [PDF]

Citation preview

Example Contributors to Measurement Uncertainty – Direct Air Environmental Microbiology Analyses (Spore Trap) (representative list - may not include of all contributors) (QC sample types in this list are typical of those utilized in AIHA-LAP, LLC laboratories) See Example Calculations (to the right of the table) and tabbed sheets for additional examples



Contributors to Uncertainty



Representative and Applicable QC Data



Comments to Clarify Contributor Effects



Temperature, Storage, Handling:



NA NA NA



No impact on direct air exam samples



Not applicable to direct air exam samples



sample size



NA NA NA



slides & coverslip contamination



MB



With proper care there should be no contamination of daily blanks; therefore, no impact



mounting medium



MB



With proper care there should be no contamination of daily blanks; therefore, no impact



MB



Samples are not exposed to air for any length of time; therefore there should be no impact



shipping time, container & temperature lab storage time, conditions & temperature contamination in lab storage areas



No impact on direct air exam samples No impact on direct air exam samples



Laboratory Subsampling: sample nonhomogeneity blending techniques



Not applicable to direct air exam samples Not applicable to direct air exam samples



Sample Preparation:



Lab Environmental Conditions: seasonal background spore variances Analysts:



RS RS RS



Reference slides analyzed by multiple analysts



microscope magnification level used



RS



Reference slides analyzed with multiple microscopes



eye piece graticule & field of view calibration



RS



Reference slides analyzed with multiple microscopes Varies by analyst



interferences



RS RS RS



ranges (high, medium, low)



RS



Uncertainty may be concentration dependent. Lab should evaluate this as part of method validation.



RS RS



Manual, spreadsheet, LIMS, etc



NA



Typically provided by the customer. This is not part of analytical uncertainty, but must be considered by labs providing sampling and providing combined sampling and analytical uncertainty.



different analysts analyst training level & experience data interpretation by analyst



Reference slides analyzed by multiple analysts Reference slides analyzed by multiple analysts



Measuring Instruments:



Test Procedure Variations: portion and fields of sample analyzed microbial density



High concentrations or clumps of spores may impact results Debris level and resolution of spores in field of view



Data Manipulation: reading, interpreting and reporting results Accuracy of calculations



area or air volume sampled



MB = Daily method blank RS = Daily reference slides



Please note that the original column I (CV of the pair) of the “culturable analyses” tabbed worksheet had a formula incorrectly entered. The worksheet has been corrected and any affected values have been highlighted in yellow.



Total Spores/m3 of air by Direct Examination Air (Spore Trap) Example Using Pooled CV Data from Three Reference Slides Reference Slide Number Slide 1 Reading Number 1 2 3 4 5 6 7 8 9 10



Mean std dev CV



Slide 2



Slide 3



Spores/m3 Spores/m3 Spores/m3 813 1640 1213 773 1473 1353 687 1507 1467 707 1420 1460 733 1273 1440 847 1247 1173 760 973 1487 793 1747 1373 973 1720 1487 693 1653 1273



11



960



1940



1453



12 13



587 807



1787 1693



1327 1500



14 15 16 17 18 19 20



1000 940 853 940 793 793 747



1847 1760 1320 1447 1920 1807 1893



1447 1487 1053 1520 1627 1160 1127



21 22 23 24 25



840 820 873 733 833



1860 1667 2107 1887 1853



1227 1427 1180 1233 973



26 27



853 753



1407 1787



1133 1280



28 29



900 647



1313 1900



1220 1480



30



887 811 98.9 0.122



1453 1643 261.6 0.159



1213 1326 162.1 0.122



CV pooled = √[((0.122)2 + (0.159)2 + (0.122)2)/3] = 0.135 Expanded MU @ 95% C.L. (k=2) equals CV pooled X 2 = 0.27 Bias cannot be determined. No quantitative reference material available



(27%



Example analytical uncertainty for air sample with 890 spores/m3: Expanded analytical uncertainty = 890 spores/m3 X 0.27 = 240 spores/m3 Example of reporting for air sample with 890 spores/m3: 890 spores/m3 with an analytical uncertainty of +/- 240 spores/m3 at the 95% confidence level



RSD)



es/m3: 40 spores/m3



res/m3 at the 95%



Example Contributors to Measurement Uncertainty – Culturable Environmental Microbiology Analyses (representative list - may not include of all contributors) (QC sample types in this list are typical of those utilized in AIHA-LAP, LLC laboratories) See Example Calculations (to the right of the table)



Contributors to Uncertainty



Representative and Applicable QC Data



Comments to Clarify Contributor Effects



Temperature, Storage, Handling: shipping time, container & temperature



lab storage time, conditions & temperature contamination in lab storage areas



NA NA NA



May affect growth but not under control of lab so no impact on analytical uncertainty Affects growth rates of organisms in samples. Study samples included in some method validation studies may reflect these contributors. As long as samples are contained there should be no impact



Laboratory Subsampling: Area of bulk samples selected for analysis



sample size



DUP DUP DUP



contamination during preparation



BLK



Sterility of materials used/autoclave operation



sample nonhomogeneity blending techniques



Extraction/vortexing non-homogeniety of samples



Sample Preparation:



sample homogenization/subsampling sample dilution (pipettes, etc) balance plating technique composition of sample



DUP DUP DUP DUP DUP



Contributions from this and the rest of Sample Prep contributors measured only if duplicate samples are prepared; not captured by inter-analyst readings of same sample prep Uncertainty related to method used to make serial dilutions Balance error is often insignificant compared to other MU sources How distributed on media Organism competition on media



Lab Environmental Conditions:



BLK



Should have minimal impact when proper aseptic techniques are used



different analysts



DUP



Must use inter-analyst instead of intra-analyst repeat analyses.



analyst training level & experience



DUP



Must use inter-analyst instead of intra-analyst repeat analyses.



data interpretation by analyst



DUP



Must use inter-analyst instead of intra-analyst repeat analyses.



microscope magnification level used



DUP



Determines level of detail of organisms that can be easily seen



eye piece graticule calibration data logger



DUP DUP



media used



DUP



Different media show preference for different organisms.



incubation conditions



DUP



Affects rate of growth but cannot be monitored unless using known concentration control cultures or do time/analysis study



seasonal background organism variances Analysts:



Measuring Instruments:



Proper calibration required to facilitate spore identification based in part on size When used to assist in organism identification, e.g. bacteria



Test Procedure Variations:



identification techniques used microbial density interferences



DUP DUP DUP



Especially for bacteria (chemical tests vs. data logger, etc) Organism competition (inhibition, overgrowth, etc) Presence of organisms other than those of interest



Data Manipulation: reading, interpreting and reporting results Accuracy of calculations



DUP DUP



Manual, spreadsheet, LIMS, etc



BLK = Lab Blank prepared when samples are prepared DUP = Inter-analyst duplicate preparation and analysis



Total Fungal CFU/swab on MEA media Example Using Pooled Duplicate CV Data



number of pair



1st 2nd CV of analyst analyst Std Dev of pair CFU/swab CFU/Swab pair (s/mean) CV2 1 230 140 63.6 0.344 0.118 2



8



6



1.4



0.202



0.041



3 4 5 6 7 8 9 10 11



69 220 150 155 94 22 17 9 33



54 200 155 130 125 16 17 7 41



10.6 14.1 3.5 17.7 21.9 4.2 0.0 1.4 5.7



0.172 0.067 0.023 0.124 0.200 0.223 0.000 0.177 0.153



0.030 0.005 0.001 0.015 0.040 0.050 0.000 0.031 0.023



12 13



32 110



30 120



1.4 7.1



0.046 0.061



0.002 0.004



14 15 16 17



110 270 17 160



100 290 29 170



7.1 14.1 8.5 7.1



0.067 0.051 0.369 0.043



0.005 0.003 0.136 0.002



18 19



300 6



330 8



21.2 1.4



0.067 0.202



0.005 0.041



20



190



260



49.5



0.220



0.048



21



6



9



2.1



0.283



0.080



22 23



19 42



11 40



5.7 1.4



0.377 0.034



0.142 0.001



24



42



49



4.9



0.109



0.012



25 26 27 28



690 170 28 180



590 240 38 160



70.7 49.5 7.1 14.1



0.110 0.241 0.214 0.083



0.012 0.058 0.046 0.007



29



61



68



4.9



0.077



0.006



30



240



300



42.4



0.157 ∑ CV2 CV pooled = √ (∑CV2/30) =



0.025 0.988 0.181



Expanded MU @ 95% C.L. (k=2) equals CV pooled X 2 =



0.36



36%



Example analytical uncertainty for swab sample with Bias cannot be determined. No quantitative reference material available 440 CFU/swab: 440 CFU/swab Expanded analytical uncertainty = 440 X 0.36 =



160



Example of reporting for swab sample with 440 CFU/swab: 440 CFU/swab with an analytical uncertainty of +/- 160 CFU/swab at the 95% confidence level 440 CFU/swab with an analytical uncertainty of +/160 CFU/swab at the 95% confidence level



CFU/swab



b at the 95% confidence level