![]()
14
| 3 | 2008
Klinisk Biokemi i Norden
(Fortsat fra side 13)
individual EQA results should deviate less than 0.4%
HbA1c units from the assigned target value (fig 5).
The target value is assumed to have an uncertainty of
approximately 0.1% HbA1c unit.
It is not only for the individual patient we need
high quality HbA1c values. When HbA1c results are
used as variables in registers, for example the Swedish
National Diabetes Register (NDR), any bias in measu-
rement will influence the fraction of patients having
desirable HbAc values. A small negative bias of 0.2%
HbA1c will result in a larger fraction of patients
having desirable HbA1c values.
The precision within laboratories and between
laboratories using the same method has steadily
improved in recent years and is now 2 – 3% CV for
the best methods. If also the current bias between the
different methods is eliminated, the overall repro-
ducibility for HbA1c measurements will be greatly
improved.
From now on until the end of 2009 there will be
an ongoing transition towards IFCC standardization
of all HbA1c methods. Some producers, e.g. Roche,
have already now IFCC traceable values assigned to
their calibrators, and use the Master Equations to
express results either in alignment with NGSP or with
Mono S. Other producers will follow.
During a period laboratories will have to report
IFCC figures together with NGSP figures (or Mono
S figures in Sweden). The exact routines for the
transition will have to be agreed upon with diabe-
tologists, other health care professionals and the
Fig 5. Individual result report
from EQUALIS HbA1c survey.
The upper graph shows the
deviation from target value
for all results from the current
survey in ranked order. The
two lower graphs show the
consecutive deviations and the
deviation in relation to target
since one year for the parti-
cipant. The Swedish quality
goal is a deviation less than
0.4% HbA1c for an individual
result, indicated by red lines.
The mean deviation (bias) for
6 surveys should be less than
0.2%.
Laboratorielista - Avvikelse
Hemoglobin A1c
Analysomgång: 2008-35
Labkod: 530-1
Klinisk kemi
Universitetssjukhuset MAS
205 02 MALMÖ
B-HbA1c ( % )
Resultat aktuell omgång:
Aktuell metod: HPLC-Mono S
Ert resultat:
6,55
%
Åsatt värde:
6,38
%
Er avvikelse:
0,17
%
Resultat tidigare omgångar:
0,1
%
Er genomsnittliga avvikelse från
åsatt värde de senaste 6 EQUALIS
utskicken:
För optimal kvalitet bör
genomsnittsavvikelsen vara < +/- 0,20 %
enheter
För optimal kvalitet bör det enskilda resultatet
avvika < +/- 0,40 % enheter
"Åsatt värde" utgörs av medelvärdet från
ett antal utvalda laboratorier (HPLC-
system) i Sverige.
För optimal kvalitet bör det enskilda resultatet
avvika <+/- 0,40 % enheter
-0,8
-0,4
0
0,4
0,8
Samtliga deltagares avvikelse i stigande ordning.
Eget resultat inringat.
Avvikelse från åsatt värde (%)
-0,6
-0,4
-0,2
0
0,2
0,4
0,6
0
2007-37
2007-41
2007-45
2007-50
2008-03
2008-07
2008-11
2008-16
2008-20
2008-24
2008-35
Avvikelse från åsatt värde senaste året
Avvikelse från åsatt värde (%)
Er avvikelse från åsatt värde i förhållande till nivå
-0,6
-0,4
-0,2
0
0,2
0,4
0,6
4
5
6
7
8
9
Nivå
Avvikelse från åsatt värde (%)
Beräknad: 2008-09-03 11:13:57
EQUALIS AB, SE-751 09 Uppsala
Sida 2
Dokumentet innehåller upphovsrättskyddat material.
Mångfaldigande, utan medgivande från EQUALIS AB, är inte tillåtet.
