Presentation on theme: "Dr Bill Bartlett Joint Clinical Director Diagnostics Group Biochemical Medicine Ninewells Hospital & Medical School NHS Tayside Scotland UK"— Presentation transcript:
Dr Bill Bartlett Joint Clinical Director Diagnostics Group Biochemical Medicine Ninewells Hospital & Medical School NHS Tayside Scotland UK Bill.Bartlett@nhs.net
Diagnosis Diagnosis Prognosis Prognosis Monitoring Monitoring Screening Screening Assessment of Risk Assessment of Risk
The metrology An understanding of its relativity to a point of reference Unusual Change
Biological Rhythms (time) Homeostasis Age Sex Ethnicity Pathology Response to Stimuli
eGFR > 60 in a 30 year old white female: Changing renal function?
Grasbeck & Saris 1969 Introduced the term “reference value”: The mode of generation of such values is known with respect to: - Selection of subjects Assessment of state of health Population characteristics, age, sex, Specimen collection and storage Analytical technique and performance characteristics Data handling techniques.
1. The Concept of Reference Values. 1987;25:337-342 2. The selection of Individuals for the Production of reference values. 1987;25:639-644 3. Preparation of individuals and collection of specimens for the production of reference intervals. 1988;26:593-598 4. Control of analytical variability in the production of reference values. 1991;29:531-535 5. Statistical treatment of collected reference limits. 1987;25:645-656 6. Presentation of observed values related to reference values. 1987;25:657-662 J Clin Chem Clin Biochem
This looks nice so far, but what is the use of biological variation data?
Analytical variance (CV A ). Within Subject biological variance (CV I ). Between Subject biological variance (CV G ).. Total = Analytical + Individual + Group
Setting of analytical goals (CV goal ). Quality specifications for : total allowable error (TE A ) Bias (B A ) Evaluating the significance of change in serial results (RCV). Assessing the utility of reference intervals (Index of Individuality). Assessing number of specimens required to estimate homeostatic set points. Choice of specimen type. Timing of specimens.
These fundamental data have many applications that under-pin our practice. We need to have confidence in the data and understand its limitations. Should we not have standards for their production and characterisation?
What is meant by the term biological variation in the context of clinical biochemistry? What is meant by the term biological variation in the context of clinical biochemistry? A component of the variance in biochemical measurements determined by the physiology of the subjects observed. A component of the variance in biochemical measurements determined by the physiology of the subjects observed.
Generation and Application of data on Biological Variation in Clinical Chemistry: - Fraser CG, Harris EK. Crit Rev Clin Lab Sci 1989:27,(5), 409-435. Optimal Conditions Precision.
Uncertainty Assay Characteristics Data Analysis Experimental Design
Purpose of study Experimental Design Characterisation of the methods Data analysis Confidence limits
What are the potential impacts of error in the data?
Biological Variation Database www.westgard.com/biodatabase1.htm CV I = 5.3% CV G = 14.2%
Desirable CV A < 0.5 x CV I B A < 0.25 x (CV I 2 + CVG 2 ) 0.5 Tea < 1.65 x 0.5 x CV I. + 0.25 x (CV I 2 + CV G 2 ) 0.5 Optimum CV A < 0.25 x CV I B A < 0.125 x (CV I 2 + CVG 2 ) 0.5 Tea < 1.65 x 0.5 x CV I. + 0.125 x (CV I 2 + CV G 2 ) 0.5 Minimum CV A < 0.75 x CV I B A < 0.0.345 x (CV I 2 + CVG 2 ) 0.5 Tea < 1.65 x 0.5 x CV I. + 0.375 x (CV I 2 + CV G 2 ) 0.5 www.westgard.com/biodatabase1.htm
n = [Z * (CV A 2 + CV I 2 )/D] 2 n = [Z * (CV A 2 + CV I 2 )/D] 2 D = % of closeness required
Biological variation data simulator. WWW.biologicalvariation.comWWW.biologicalvariation.com
CV I = 5.3 % CV G = 14.2%CV A =2.7%
CV I = 5.3 % CV G = 14.2%
Index of individuality = 0.4
Biological Variation Serum Creatinine: Average within subject (CVI) = 4.1% Gowans & Fraser. Ann Clin Biochem 1988:25:259-263
QuantityUnitsGroupMeanCV I CV G Index of Individuality Serum Creatinine µmol/LMale (7)22.214.171.124.54Fraser µmol/LFemale (8)71.44.911.80.41Fraser * µmol/L * Whole (15)126.96.36.199.29Fraser µmol/L??5.314.20.4BioV Site ** µmol/L ** N= 20 Male (7) Female(13) 774.714.40.33Reinhard et al * Jaffe ** Enzymatic
Upper Reference Limits: - Male = 106 µmol/L Female = 80 µmol/L RCV larger for men than for women.
If True: - Clinically important as disease progression needs to be monitored and appropriate actions taken (e.g. Acute on Chronic Kidney failure). Clinically important as disease progression needs to be monitored and appropriate actions taken (e.g. Acute on Chronic Kidney failure). Tighter analytical performance characteristics to be applied for females. Tighter analytical performance characteristics to be applied for females. Impact will be greater on eGFR Impact will be greater on eGFR
% Change at % Probability CV I 95%99% Rise in Creatinine 4.310.3%14.6% 5.312.6%17.8% Fall in eGFR 4.3 12.8% 12.8%15.4% 6.816.0%22.6% Assumes a CV A = 1%
Use eGFR for initial classification of CKD stage. Use creatinine to follow patients with RCV indicator flag? More Precise? Difficulty is that there is a suggestion that creatinine CV I is variable in disease. Therefore which CV I ?
State of Health CV I Number of Subjects Length of Studies (days) Number Samples/Sub Healthy Median? 4.3 4.3 CRF5.317218 Type 1 DM 5.927568 Impaired renal function 6.99211 Type 1 DM 6.511568 Post renal transplant 11.541908 Acute MI 13.420419.5 CKD children 13.0545409 Ricos et al Ann Clin Biochem 2007;44: 343-352
What is the uncertainty? What are the quality standards for BV Data? Assay Characteristics Data Analysis Experimental Design
40 years of data Do the data travel through timeDo the data travel through time Method developmentsMethod developmentsQuality Enough reported detail.Enough reported detail. Good Design?Good Design? Commutable Population demographics.Population demographics. Healthy?Healthy? DiseasedDiseased? Translated into databases Excellent ResourcesExcellent Resources Granular enough?Granular enough? Data archetype required?Data archetype required? The Literature 319 Constituents: 90 entries based on 1 Paper 90 entries based on 1 Paper
Longish history of evolving assay systems with differing analytical performance characteristics and specificities. 1970s – C-Terminal RIA Late 80s – Sandwich IRMA Assay 1990 – 98 Nichols IRMA assays dominate Late 1990s – variety of “intact” sandwich assays on a number of different analytical platforms. 2004 – Bioactive PTH assay Adapted from M Scott Focus 2010
Much evidence in the literature indicating that assays react to varying extents with the variety of PTH fragments present in Serum. M Scott Focus 2010
If clearance of fragments is not identical in all patients and non diseased patients the apparent biological variation will vary and be assay specific. Assay specificity an important BV qualifier?
Ankrah Tet et al. Ann Clin Biochem 2008;45:167-169 PTH = Nichols Advantage 4 Males 6 Females “Normals” Gardham et al. Clin J Am Soc Nephrol ePress May 24 th 2010 Abbot Architect Intact PTH Immunotopics Inc. Biointact PTH 1-84 12 “Normals” 22 Haemodialysis patients
Subjects nAssayPTH ng/L CV I CV G CV A RCV (%) N-Set* “Normal”10Nichols51.725.923.85.072.327 “Normal”12Abbott51.919.23.554.015 Immunotopics Bio-intact 1-84 27.523.84.267.022 Dialysis22Abbott303.025.63.672.026 Immunotopics Bio-intact 1-84 131.030.26.386.037 * Number of Specimens Required to estimate homeostatic point within 10% with a probability of 95%
Data in chronic stable disease “often can be considered constant over time and geography” “Same order of magnitude in disease and health”
66 quantities 34 disease with 45 references. “For the majority of quantities studied CV I of same order as diseased. “ Disease specific RCVs may be necessary in some cases. Effect of variability in variability not quantitatively studied. “Heterogeneity in study designs and methods compiled”
“Blood samples were taken at weekly intervals from 10 healthy subjects (4 men and 6 women, median age 21 years, range 19–27 years; mean body mass index 21.3, range 19.0–25.9) for six weeks at the same time of the day (between 12:30 and 14:30 h),” I’m healthy and normal ! I’m a biochemist!
Need to assess on a case by case basis. Need to assess on a case by case basis. Questions around uncertainty. Questions around uncertainty. What are the implications for their application? What are the implications for their application? Can the impact of uncertainty be quantified and reduced where necessary. Can the impact of uncertainty be quantified and reduced where necessary. Accepted standard needed for their production. Accepted standard needed for their production. Critical appraisal checklist required to enable veracity of existing and new publications. Critical appraisal checklist required to enable veracity of existing and new publications. Meta-analysis of data Meta-analysis of data Questions to be addressed by the EFCC biological Variation Working group
1. Define the purpose for which they are to be used. 2. Only meaningful and transferable if defined for the population or individual in terms of: - Inclusion and exclusion criteria Intake of food & drugs Physiological and environmental conditions Specimen collection criteria Performance characteristics of the analytical method The statistical methods used for estimation of the limits
3. State of health defined. WHO Defn: - “ a state of complete physical mental and social well being and not merely the absence of disease or infirmity” Disease is a state of health. Conceptually different in different countries. The term “Reference” should be accompanied or preceded by a word qualifying the state of health. E.g diabetic, hospitalised diabetic, ambulatory diabetic, Healthy laboratory worker?
The reference change value: a proposal to interpret laboratory reports in serial testing based on biological variation. C. RICO´ et al Scand J Clin Lab Invest 2004; 64: 175 – 184 The RCV data in this study are presented as a point of departure for a widely applicable objective guide to interpret changes in serial results.” “ The RCV data in this study are presented as a point of departure for a widely applicable objective guide to interpret changes in serial results.” HL7 recognised concept Requests for additional flags pending