Biological Variation Of Hematology Tests Based On The
Number 54 n July 12, 2012Biological Variation of Hematology Tests Basedon the 1999–2002 National Health andNutrition Examination Surveyby David A. Lacher, M.D.; Janet Barletta, Ph.D.; and Jeffery P. Hughes, M.P.H.,Division of Health and Nutrition Examination SurveysAbstractIntroductionObjective—Biological variation consists of between-person (BP) andwithin-person (WP) variation. Estimates of WP coefficients of variation (CVw)and BP coefficients of variation (CVg) for hematology laboratory tests wereestimated from the 1999–2002 National Health and Nutrition ExaminationSurvey (NHANES).Methods—NHANES is a survey of the civilian noninstitutionalized U.S.population that uses a stratified, multistage probability design. Between- andwithin-person variations were estimated for 18 hematology tests. For WPvariation, a nonrandom sample was obtained with a median of 17 days betweentwo test measurements. Between-person variation was estimated from the WPsample and additional participants were matched for age group, gender, and raceand ethnicity to the WP sample.Results—The BP and WP variations were estimated on as many as 2,496and 852 sample participants, respectively. Mean corpuscular hemoglobinconcentration had the lowest CVg (2.25% for men and 2.40% for women), andmean corpuscular volume had the lowest CVw (0.31% for men and 0.37% forwomen). The index of individuality (CVw /CVg) ranged from 0.06 for meancorpuscular volume for men and women to 0.62 for segmented neutrophilnumber for men, and 0.55 for segmented neutrophil percent for women. Womenhad higher CVw compared with men for hematocrit, hemoglobin, meancorpuscular volume, red blood cell count, and red blood cell distribution width.Several hematology tests’ CVw also differed by age group, including meancorpuscular volume; eosinophil, lymphocyte and segmented neutrophil percent;monocyte and segmented neutrophil number; white blood cell count; and redblood cell distribution width.Laboratory analytes for individualsare subject to several sources ofvariation, including biological,preanalytical (specimen collection),analytical (bias and imprecision), andpostanalytical (reporting of results).Analytical bias is the closeness of ananalyte result to the ‘‘true value’’ of theresult. Precision is the repeatability ofan analyte result if the same sample istested many times. Biological variationconsists of within-person (WP) andbetween-person (BP) variation. Thesecomponents of biological variation areused to set analytical goals for bias andimprecision, evaluate changes for aperson’s tests using delta checks, andassess the clinical utility of populationbased reference ranges (1).The goals of imprecision and biasfor a laboratory analyte are differentdepending on the intended use inscreening, diagnosis, or monitoring thecourse of diseases in patients. When theWP variation is much smaller than theBP variation, the individual results staywithin a narrow range compared withthe population-based reference interval(range). Hence, the WP variation wouldbe used to monitor serial changes ofKeywords: within-person variation between-person variation laboratory testsU.S. DEPARTMENT OF HEALTH AND HUMAN SERVICESCenters for Disease Control and PreventionNational Center for Health Statistics
Page 2laboratory values in a person. Contrarily,when the WP variation is similar to theBP variation, the person’s results overtime have a wide range comparable withthe population-based reference interval.In this situation, the BP variation(population reference range) is used tomonitor serial change of the laboratoryvalues in a person. Ideally, desirablegoals for imprecision (I) and bias (B)have been related to the WP coefficientsof variation (CVw) and the BPcoefficients of variation (CVg) oflaboratory analytes (1–3).Buttarello discussed sources ofvariation of hematology analytes (4).Preanalytical sources of variationinclude type of anticoagulant used,specimen storage temperatures, andstability. Postanalytical sources ofvariation include delta checks(differences between consecutivelaboratory values of a test), limit checks(laboratory values when exceededrequiring further investigation), andreports of unusual cell morphology.Analytical variation sources ofimprecision and bias of hematology testshave been characterized by monitoringquality controls, proficiency testing, andcomparing automated instrumentmeasurements with manual methods.BP variation of hematology testshas been studied extensively fordemographic characteristics includingage and gender. The WP source ofbiological variation has been evaluatedfor chemistry tests, but rarely forhematology tests. Most hematologystudies of biological variation had fewsubjects. Statland examined the meanhourly, daily, and weekly intra individual variation of hematology testsin 20 adults (5). Costongs examineddaily, weekly, and 6-month intra individual variations of hematology testsfor 62 adults aged 18–57 (6). Costongsalso examined the critical differences(delta checks) using WP data. Costongsnoted that the WP coefficients ofvariation were indirectly related to thelife span of cells, with red blood cells(life span 120 days) having the lowestCVw and white blood cells (life span6–8 hours) having the highest CVw.Fraser examined WP and BP variationof hematology tests in 24 elderlyNational Health Statistics Reports n Number 54 n July 12, 2012persons (mean age 75) over 20 weeks at14-day intervals (7). Ricos (2) andFraser (1) have compiled lists of WPand BP variation for laboratory tests,including hematology tests.The WP and BP variation oflaboratory tests has been examined inseveral NHANES surveys. Lookerexamined hematology and biochemicalmarkers of iron status in the HispanicHealth and Nutrition ExaminationSurvey conducted from 1982 through1984 for 80 persons (8). The effect ofincreased WP variation onoverestimation of prevalence ofhematologic disorders was examined.WP and BP variation have been reportedfor selected analytes from the thirdNational Health and NutritionExamination Survey conducted from1988 through 1994 (9). The BP and WPvariation for general biochemical,nutritional, and environmental analyteswas analyzed for the 1999–2002National Health and NutritionExamination Survey (10). In this report,BP and WP variations of hematologytests are presented for NHANES1999–2002, and gender and age groupsare compared for WP variation.Methods andProceduresEstimates of CVw and CVg forlaboratory tests were calculated from the1999–2002 NHANES (11,12), across-sectional survey that collected dataon the civilian noninstitutionalized U.S.population through questionnaires andmedical examinations, includinglaboratory tests. NHANES 1999–2002used a stratified, multistage probabilitydesign to collect a nationallyrepresentative sample.The hematology tests were collectedin EDTA tubes as part of a completeblood and five-part differential cellcount profile and were analyzed on theBeckman Coulter MAXM analyzer(Beckman Coulter Corporation, Brea,Calif.). Details of the laboratorymethods have been described (13) andthe 18 hematology tests are listed inTables 1–5.The MAXM instrument is alaser-based flow cytometer that usesimpedance, conductivity, and lightscatter to directly measure some of thehematology tests. Other hematologytests are calculated from the directlymeasured analytes. The methodanalytical CV (CVa ) used to calculatethe CVw was derived from imprecisionCV using bench quality controls orimprecision based on manufacturerinformation.The BP and WP means, standarddeviations, and coefficients of variationfor hematology tests are shown for menin Table 1 and for women in Table 2.The BP sample was generated from theWP sample, and two additionalparticipants were selected for every oneWP sample participant. The additionalBP participants were selected bymatching for gender, race and ethnicity,and age group. BP participants werematched in 3-year age groups (e.g.,16–18, 19–21, 22–24, 25–27, etc.) tohave more participants to compare with3-year age groups in the WP sample.The BP variations were estimated on asmany as 2,496 sample participants fromNHANES 1999–2002. The WPvariations were estimated from aconvenience sample of 852 personsbased on NHANES 2000–2002 data.The WP sample participants wererecruited for a second test measurement.The WP participants were not selectedrandomly but recruited according toseveral criteria, including selectingapproximately equal proportions of menand women with an approximatelyuniform age distribution of 16–69 years.Participants were recruited to obtainabout equal numbers for race andethnicity of Mexican-American,non-Hispanic black, non-Hispanic white,and other persons. The target size of theWP sample was 5% of those participantswho had a venipuncture during theinitial visit to the NHANES mobileexamination center. The WP participantswere asked to return for a secondphlebotomy no sooner than 8 days aftertheir initial blood draw (1.8% ofparticipants had second phlebotomiesbefore 8 days). Because the BP samplewas matched for age and gender to theWP sample, no differences inproportions by gender were seen for theage groups (16–29, 30–49, and 50–69).
National Health Statistics Reports n Number 54 n July 12, 2012A chi-square test showed no significantdifferences between age groups bygender for the WP ( p 0.71) and BP( p 0.44) sample.The WP variation was estimatedfrom a nonrandom, unweighted samplewith a median of 17 days (25thpercentile: 13 days, 75th percentile: 23days, range: 3–51 days) between twotest measurements. The analyticalvariation includes the imprecision andchanges in bias (for example, changes inmethod calibration) that are usuallynegligible. Hence, the CVa is estimatedby the method imprecision CV (CVi).The total coefficient of variation (CVt)of a laboratory analyte can be estimatedassuming that all sources of error aremeasured at the same analyte mean andthat preanalytical and postanalyticalsources of variation are negligible. TheCVt is calculated as [(CVa)2 (CVw)2]1/2(1). Hence, the CVw was calculated as[(CVt)2 – (CVa)2]1/2. The CVg wascalculated as SD/mean for the BPsample. The distributions of severalhematology tests were nongaussian, andextreme outliers were excluded to obtainan approximately gaussian distributionwith more stable estimates of variation.Outliers were eliminated by use ofTukey’s method, which defines outliersas three interquartile ranges below the25th percentile or above the 75thpercentile (14). The basophil andeosinophil number had extremenongaussian distributions and wereexcluded. Gender and age group WPvariations were compared. Ninety-fivepercent confidence intervals wereestimated for the CVw. A likelihoodratio test was performed to determine ifthe CVw for gender or age group for alaboratory analyte were equal (15).Statistical analyses were carried out withSAS for Windows software (SASInstitute, Cary, N.C.).Results and DiscussionThe BP and WP means, standarddeviations, and coefficients of variationfor 18 hematology tests are shown formen in Table 1 and for women inTable 2. The mean of hematology testsof the BP and WP sample for men andfor women, respectively, were similar.As expected, the BP mean red bloodcell-related hematology tests(hemoglobin, hematocrit, and red bloodcell count) were higher for mencompared with women. The CVgexceeded the laboratory CVi for allhematology tests. The CVw exceededthe laboratory CVi for 15 of 18hematology tests. However, the CVwwas less than the laboratory CVi forbasophil percent, mean corpuscularhemoglobin, and mean corpuscularhemoglobin concentration. The CVwcannot be estimated for thesehematology tests because the CVw iscalculated as [(CVt2) – (CVa)2]1/2 andCVa, as estimated by CVi, exceeded theCVt. The imprecision of the hematologytests were estimated from between-runbench quality controls that are usedmost commonly to estimate imprecisionof tests for biological variation. Becausethe WP CV is estimated over severalruns, the between-run imprecision wasused.Analytical goals for imprecision andbias can be judged using the CVw andCVg. Imprecision should ideally be lessthan one-half of the CVw, and biasshould be less than 0.25 [(CVw)2 (CVg)2]1/2 (1). The total error of alaboratory measurement reflects theunderlying bias and imprecision of ananalyte. The goal for total error shouldbe less than kI B, where k 1.65 forα 0.05 (1). For example, the observedhemoglobin imprecision of 1.1% wasless than the imprecision goal ofone-half of CVw (2.46%), or 1.23% inmen (Table 1). The bias for hemoglobinin men should ideally be less than0.25[(CVw)2 (CVg)2]1/2, or0.25[(0.0246)2 (0.0785)2]1/2, or 8.2%.The total error is estimated as B 1.65(I), or 8.2% 1.65(1.1%), or10.0%. Thus, the total error forhemoglobin in men estimated at the BPmean of 15.3 g/dL (Table 1) was 1.53g/dL (15.3 g/dL multiplied by 0.10). Thetotal error of 10.0% for malehemoglobin exceeded the ClinicalLaboratory Improvement Amendmentsacceptable performance for total errorfor hemoglobin of 7% (16).The mean corpuscular hemoglobinconcentration (MCHC) CVg of 2.25%was lowest among all the hematologyPage 3tests in men (Table 1) and 2.40% forwomen (Table 2). Compared with allother hematology tests, the MCHCbetween-person CV was lowest in allage groups, with 2.33% for participantsaged 16–29 (Table 3), 2.39% forparticipants aged 30–49 (Table 4), and2.29% for participants aged 50–69(Table 5). The mean corpuscular volume(MCV) CVw was lowest amonghematology tests in men (0.31%) andwomen (0.37%). The MCV withinperson CV was also lowest amonghematology tests in all age groups with0.28% for ages 16–29, 0.51% for ages30–49, and 0.18% for ages 50–69. Thelow within-person CV for MCHC canbe seen for men and women (Figure 1).High CVg was seen in eosinophiland basophil percent for men andwomen (range 52.4%–64.4%) and in allage groups (range 50.8%–68.6%), whichreflects very low analyte values. Otherhematology analytes also had relativelyhigh CVg (greater than 25%) includingwhite blood count, segmentedneutrophil, lymphocyte and monocytenumber, and lymphocyte and monocytepercent. High CVw was seen ineosinophil percent and segmentedneutrophil number in men and women(range: 21.5%–25.2%) and in all groups(range: 19.1%–26.5%). High withinperson CV reflects individual variationdue to gender, age, diurnal or cyclicalvariation, disease, or very low analytevalues. The high within-person CV forsegmented neutrophil number can beseen for men and women in Figure 2.The ratio of CVw to CVg, alsoknown as the index of individuality, isimportant in determining the use ofpopulation-based reference (normal)intervals in detecting changes of diseasestatus in individuals (17,18). When theindex of individuality is low ( 0.5), theindividual results stay within a narrowrange compared with the populationbased reference interval. Hence, a lowindex suggests the utility of evaluatingserial changes in analyte values in anindividual, whereas population-basedreference intervals would be of limiteduse. A high index ( 0.5) suggests thatthe population-based reference intervalis appropriate when interpreting aperson’s laboratory analyte value. The
National Health Statistics Reports n Number 54 n July 12, 2012Page 4110Men10090Second-day mean cell volume (fL)80700070758085909510010511080859095First-day mean cell volume (fL)100105110110Women100908070007075NOTE: fL is femtoliter.SOURCE: CDC/NCHS, National Health and Nutrition Examination Survey, 1999–2002.Figure 1. First-day versus second-day cell volume, by genderindex of individuality ranged from 0.06for mean corpuscular volume for menand women, to 0.62 for segmentedneutrophil number for men and 0.55 forsegmented neutrophil percent for women(Tables 1 and 2). The index ofindividuality was lowest for meancorpuscular volume for all age groups(0.05 for ages 16–29, 0.09 for ages30–39, and 0.03 for ages 50–69)(Tables 3–5). The index of individualitywas highest for segmented neutrophilpercent (0.63) for ages 16–29. For ages30–49, segmented neutrophil andlymphocyte percent had the highestindex of individuality (0.52); and forages 50–69, segmented neutrophilnumber and percent had the highestindex of individuality (0.53).The BP and WP variations wereanalyzed by gender. The sample size,means, standard deviations, andcoefficients of variation for hematologyanalytes are presented for men (Table 1)and women (Table 2). Several laboratoryanalytes had significant differences (p 0.001) in the CVw when males andfemales were compared. Women hadhigher CVw compared with men fortests associated with red blood cellsincluding, hematocrit, hemoglobin, meancorpuscular volume, red blood cellcount, and red blood cell distributionwidth. For example, the red blood cellcount CVw in women was 3.45%compared with 2.53% in men. Femalesmay have more within-person variationdue to blood loss during menstruation orincreased iron utilization duringpregnancy.The BP and WP variations ofhematology analytes were also analyzedby age group. The sample size, means,standard deviations, and coefficients ofvariation of hematology analytes arepresented for age groups 16–29(Table 3), 30–49 (Table 4), and 50–69(Table 5). Several hematology tests alsodiffered (p 0.05) by age group whilecontrolling for gender. Participants aged16–29 had higher CVw than those aged30–49 and 50–69 for eosinophil,lymphocyte, and segmented neutrophilpercent, and segmented neutrophilnumber. Also, participants aged 16–29had higher CVw than those aged 50–69for monocyte number, red blood celldistribution width, and white bloodcount. Participants aged 30–49 hadhigher CVw than those aged 50–69 formonocyte number, red blood celldistribution width, and white bloodcount. Mean corpuscular volume CVwwas highest in the middle age group,with 0.28% for those aged 16–29,0.51% for those aged 30–49, and 0.18%for those aged 50–69 (Figure 3).In this report, BP and WP estimatesof coefficients of variation wereobtained for 18 hematology analytes.The literature on WP hematologyvariation is very limited, and this reportadds information on within-personcoefficients of variation. NHANES1999–2002 provides a better estimate ofBP variation than other locallyrepresentative studies because theNHANES sample was nationallyrepresentative and had a larger samplesize. The WP variation estimate waslimited by the nonrandom, self-selecteddesign and reflected a median of 17days between two measurements. Inaddition, the CVw and CVg estimates inNHANES were based on a relativelyhealthy sample of the population. TheCVw and CVg would be increased in asample of unhealthy individuals becauseof changes in disease status andtreatment. The BP and WP sample
National Health Statistics Reports n Number 54 n July 12, 201212Page 5Men10Second-day segmented neutrophil number (1,000 cells/µL)864200241012246810First-day segmented neutrophil number (1,000 cells/µL)121268Women10864200NOTE: µL is microliter.SOURCE: CDC/NCHS, National Health and Nutrition Examination Survey, 1999–2002.Figure 2. First-day versus second-day segmented neutrophil number, by genderparticipants were restricted to those aged16–69. The estimate of CVw could beimproved by use of a stratified,multistage probability design overdifferent time periods.
National Health Statistics Reports n Number 54 n July 12, 2012Page 6References11016–29 859095First-day mean cell volume (fL)100105110Second-day mean cell volume (fL)11030–49 years10090807000707580859011050–69 years100908070007075NOTE: fL is femtoliter.SOURCE: CDC/NCHS, National Health and Nutrition Examination Survey, 1999–2002.Figure 3. First-day versus second-day mean cell volume, by age group1. Fraser CG. Biological variation:From principles to practice.Washington, DC: AACC Press;p 151. 2001.2. Ricos C, Alvarez V, Cava F, GarciaLario JV, Hernandez A, Jimenez CV,et al. Current databases on biologicalvariation: Pros, cons and progress.Scand J Clin Lab Invest 59:491–500.1999.3. Fraser CG, Harris EK. Generationand application of data on biologicalvariation in clinical chemistry. CritRev Clin Lab Sci 27:409–37. 1989.4. Buttarello M. Quality specification inhematology: the automated blood cellcount. Clin Chim Acta 346:45–54.2004.5. Statland BE, Winkel P, Harris SC,Burdsall MJ, Saunders AM.Evaluation of biologic sources ofleukocyte counts and otherhematologic quantities using veryprecise automated analyzers. Am JClin Pathol 69:48–54. 1978.6. Costongs GM, Janson PC, Bas BM,Hermans J, Van Wersch JW,Brombacher, PJ. Short-term andlong-term intra-individual variationsand critical differences ofhematological laboratory parameters.J Clin Chem Clin Biochem 23(2):69–76. 1985.7. Fraser CG, Wilkinson SP, NevilleRG, Knox JD, King JF, MacWalterRS. Biologic variation of commonhematologic laboratory quantities inthe elderly. Am J Clin Pathol92:465–70. 1989.8. Looker AC, Sempos CT, Liu K,Johnson CL, Gunter EW. Withinperson variance in biochemicalindicators of iron status: effects onprevalence estimates. Am J Clin Nutr52:541–7. 1990.9. Lacher DA, Hughes JP, Carroll MD.Estimate of biological variation oflaboratory analytes based on the thirdNational Health and NutritionExamination Survey. Clin Chem59:450–2. 2005.10. Lacher DA, Hughes JP, Carroll MD.Biological variation of laboratoryanalytes based on the 1999–2002National Health and NutritionExamination Survey. National healthstatistics reports; no 21. Hyattsville,MD: National Center for HealthStatistics. 2010.
National Health Statistics Reports n Number 54 n July 12, 201211. National Center for Health Statistics.NHANES 1999–2000 public datarelease. Available from: nes99 00.htm.Accessed December 28, 2011.12. National Center for Health Statistics.NHANES 2001–2002 public datarelease. Available from: nes01 02.htm.Accessed December 28, 2011.13. National Center for Health Statistics.NHANES 2001–2002 laboratoryprocedural manual: Complete bloodcount (CBC) with five-partdifferential. Available from: http://www.cdc.gov/nchs/data/nhanes/nhanes 01 02/l25 b met completeblood count.pdf. Accessed December28, 2011.14. Tukey JW. Exploratory data analysis.Reading, MA: Addison-WesleyPublishing Co.; p 44. 1977.15. Verrill SP, Johnson RA. Confidencebounds and hypothesis tests fornormal distribution coefficients ofvariation [research paper FPL RP638]. Madison, WI: U.S.Department of Agriculture, ForestService, Forest Products Laboratory;p 57. 2007.16. U.S. Department of Health andHuman Services. Medicare, Medicaidand CLIA programs: Regulationsimplementing the Clinical LaboratoryImprovement Amendments of 1988(CLIA) [final rule]. Fed Regist57(40):7002–186. 1992.17. Fraser CG. Inherent biologicalvariation and reference values. ClinChem Lab Med 42(7):758–64. 2004.18. Harris EK. Effects of intra- andinterindividual variation on theappropriate use of normal ranges.Clin Chem 20(12):1535–42. 1974.Page 7
National Health Statistics Reports n Number 54 n July 12, 2012Page 8Table 1. Between-person (CVg), within-person (CVw), and method (CVa) coefficients of variation for menBetween-personAnalyte (units)Basophil percent (percent) . . . . . . . . . . .Eosinophil percent (percent) . . . . . . . . . .Hematocrit (percent) . . . . . . . . . . . . . . .Hemoglobin (g/dL) . . . . . . . . . . . . . . . .Lymphocyte number (103/µL) . . . . . . . . .Lymphocyte percent (percent) . . . . . . . . .Mean corpuscular hemoglobin (pg) . . . . . .Mean corpuscular volume (fL) . . . . . . . . .Mean corpuscular hemoglobin concentrationMean platelet volume (fL) . . . . . . . . . . . .Monocyte number (103/µL) . . . . . . . . . . .Monocyte percent (percent) . . . . . . . . . .Platelet count (103/µL) . . . . . . . . . . . . . .Red blood cell count (106/µL) . . . . . . . . .Red blood cell distribution width (percent) . .Segmented neutrophil number (103/µL) . . .Segmented neutrophil percent (percent) . . .White blood cell count (103/µL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .(g/dL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . nSDCVw(percent)CVa(percent)CVw 40.620.590.55Method1Index2† CVw could not be calculated because the CVt (total coefficient of variation) was less than the CVa.††p 0.001 where the CVw for men is equivalent to the CVw for women.1Method analytical CV is the laboratory method precision assuming no method bias exists.2Index of individuality.NOTE: CV is coefficient of variation, n is the size of the sample, and SD is standard deviation; g/dL is grams per deciliter, µL is microliter, pg is picogram, and fL is femtoliter.Table 2. Between-person (CVg), within-person (CVw), and method (CVa) coefficients of variation for womenBetween-personAnalyte (units)Basophil percent (percent) . . . . . . . . . . .Eosinophil percent (percent) . . . . . . . . . .Hematocrit (percent) . . . . . . . . . . . . . . .Hemoglobin (g/dL) . . . . . . . . . . . . . . . .Lymphocyte number (103/µL) . . . . . . . . .Lymphocyte percent (percent) . . . . . . . . .Mean corpuscular hemoglobin (pg) . . . . . .Mean corpuscular volume (fL) . . . . . . . . .Mean corpuscular hemoglobin concentrationMean platelet volume (fL) . . . . . . . . . . . .Monocyte number (103/µL) . . . . . . . . . . .Monocyte percent (percent) . . . . . . . . . .Platelet count (103/µL) . . . . . . . . . . . . . .Red blood cell count (106/µL) . . . . . . . . .Red blood cell distribution width (percent) . .Segmented neutrophil number (103/µL) . . .Segmented neutrophil percent (percent) . . .White blood cell count (103/µL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .(g/dL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . t)CVa(percent)CVw 60.520.550.50†CVw could not be calculated because the CVt (total coefficient of variation) was less than the CVa.††p 0.001 where the CVw for women is equivalent to the CVw for men.1Method analytical CV is the laboratory method precision assuming no method bias exists.2Index of individuality.NOTE: CV is coefficient of variation, n is the size of the sample, and SD is standard deviation; g/dL is grams per deciliter, µL is microliter, pg is picogram, and fL is femtoliter.
National Health Statistics Reports n Number 54 n July 12, 2012Page 9Table 3. Between-person (CVg), within-person (CVw), and method (CVa) coefficients of variation for ages 16–29Between-personAnalyte (units)Basophil percent (percent) . . . . . . . . . . .Eosinophil percent (percent) . . . . . . . . . .Hematocrit (percent) . . . . . . . . . . . . . . .Hemoglobin (g/dL) . . . . . . . . . . . . . . . .Lymphocyte number (103/µL) . . . . . . . . .Lymphocyte percent (percent) . . . . . . . . .Mean corpuscular hemoglobin (pg) . . . . . .Mean corpuscular volume (fL) . . . . . . . . .Mean corpuscular hemoglobin concentrationMean platelet volume (fL) . . . . . . . . . . . .Monocyte number (103/µL) . . . . . . . . . . .Monocyte percent (percent) . . . . . . . . . .Platelet count (103/µL) . . . . . . . . . . . . . .Red blood cell count (106/µL) . . . . . . . . .Red blood cell distribution width (percent) . .Segmented neutrophil number (103/µL) . . .Segmented neutrophil percent (percent) . . .White blood cell count (103/µL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .(g/dL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Method1Index2CVw(percent)CVa(percent)CVw 56Method1Index2CVw(percent)CVa(percent)CVw 947948942938947948938
for 18 hematology tests are shown for men in. Table 1 and for women in Table 2. The mean of hematology tests of the BP and WP sample for men and for women, respectively, were similar. As expected, the BP mean red blood cell-related hematology tests (hemoglobin, hematocrit, and red blood cell
Liquichek Hematology-16 Control A hematology reference control used in monitoring determinations of blood cell values on Sysmex and Abbott 3 part differential hematology analyzers. 105 day shelf life at 2-8 C 14 day open-vial stability at 2-8 C Liquichek Hematology-16T Control Liquichek Reticulocyte Control (S)
Mendelssohn: Variations sérieuses, Op. 54 cm Andante sostenuto 0:53 cn Variation 1 0:40 Variation 2 – Un poco più animato 0:31 cp Variation 3 – Più animato 0:22 cq Variation 4 0:23 cr Variation 5 – Agitato 0:26 cs Variation 6 – a tempo 0:27 ct Variation 7 0:25 cu Variation 8
NFP121 5 Sommaire 1) Tests et tests unitaires - Outil : junit www.junit.org une présentation Tests d'une application - Une pile et son IHM Tests unitaires de la pile Tests de plusieurs implémentations de piles Tests d'une IHM Tests de sources java Invariant et fonction d'abstraction comme tests - Tests en boîte noire - Tests en boîte blanche
Variation – Direct Variation, Inverse Variation and Joint Variation Chapter 5.1 Sometimes the equation that relates two or more variables can be described in words by the idea of “variation”. There are three types. Dire
5. Learn the scope and practices of “Special Hematology” testing and how it is utilized to diagnose hematologic disorders. 6. Complete general/special hematology experience checklist. 4. Clinical experience in Hematology/Performance of Bone Marrows (with Hematology/Oncology Division) (1 day, Wednesday of first rotation week)
Hematology, 12th ed., by J.P. Greer et al (Eds), Lippincott Williams & Wilkins, 2008 (two volumes). This is an encyclopedic clinical reference book in hematology. The large multi-authored work, Williams Hematology, 8th ed, MA Lichtman et al (Eds), McGraw-Hill, 2006 i
3.3.2 Hematology and Blood Chemistry Examination The results of the CTVT blood hematology and chemical examination in domestic dogs are shown in Table 1. Table 1. Hematology examination Examination Result Normal range Hematology White blood cell (103/ L) 47.1 16.0 -17.0 Red blood cell (106/ L) 1.12 5.5 -8.5
Aberdeenshire Council Local Transport Strategy 2012 6 / 28 key partner in the North Sea Commission’s Transport Group. Our successes to date have been recognised externally with the Council receiving National Transport Awards for specific projects, and the accolade of ‘Transport Local Authority of the Year’, in both 2008 and 2009, while