Loading...
Publication
AEQ na quantificação da hemoglobina: desempenho analítico a longo prazo de trinta laboratórios portugueses
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 668.1 KB | Adobe PDF |
Advisor(s)
Abstract(s)
É importante para os laboratórios clínicos avaliarem o desempenho dos seus
métodos ao longo do tempo. O principal objetivo deste estudo foi a aplicação
de um modelo de regressão linear para avaliação a longo prazo, do desempenho
dos laboratórios na quantificação da hemoglobina, comparando os resultados
individuais do laboratório com a média de consenso de cada ensaio,
após a exclusão de outliers. Para avaliar o desempenho do laboratório foram
calculados o coeficiente de variação analítico a longo prazo (LCVa) e o Bias
total analítico. O modelo de regressão linear foi aplicado aos resultados quantitativos
da hemoglobina para avaliar o desempenho analítico a longo prazo de
trinta laboratórios que participam no programa de avaliação externa da qualidade
PNAEQ (2014-2016), utilizando os resultados de doze amostras controlo
de sangue em EDTA, com diferentes concentrações de hemoglobina. Os
participantes foram selecionados aleatoriamente, sendo incluídos laboratórios
hospitalares e ambulatórios. As variáveis introduzidas para avaliar o desempenho
a longo prazo neste modelo foram o CV analítico a longo prazo (LCVa) e o
Bias total analítico. Dois laboratórios foram excluídos da análise. Foi determinado
o número de laboratórios que cumpriram os objetivos de desempenho
analítico definido com base na variação biológica. A mediana do LCVa foi de
1,4% (variação de 0,4% -3,4%). O LCVa foi 0,58 vezes inferior à variação biológica
total (teste de diagnóstico) para todos os laboratórios (100%) e foi 0,5
vezes inferior à variação biológica intraindividual (teste de monitorização) em
62% dos laboratórios. A mediana do Bias total foi de 1,1% (variação de 0,1%
-3,3%), sendo que 83% dos laboratórios apresentaram um Bias total inferior
a 0,25 da variação biológica total. A participação em programas de avaliação
externa da qualidade é de extrema importância, uma vez que fornece informações
ao laboratório sobre o desempenho do seu método tanto numa única
análise, bem como ao longo do tempo e permite a avaliação da necessidade
de melhoria.
It is impor tant for medical laboratories to know about the stability and performance of their methods over the time. The main objective of this study was to apply a linear regression model for long-term evaluation of the performance of laboratories in haemoglobin quantification by comparing the laborator y individual results with the consensus mean of each round, af ter outliers exclusion. To evaluate the per formance of the laborator y, the long term analytical CV (LCVa) and the total analytical bias were established. A linear regression model was applied to quantitative haemoglobin results to evaluate the long-term analytical per formance of thir ty laboratories that par ticipate in the PNAEQ external quality assessment programme (2014- 2016), using the results of twelve blood EDTA samples with dif ferent hemoglobin concentrations. Par ticipants were randomly selected concerning laboratories from hospital and ambulator y. The variables introduced to define the long-term per formance in this model were the long-term analytical CV (LCVa) and total analytical Bias. Two laboratories were excluded from the analysis (one outlier and one that uses point-of-care equipment). We evaluate also the number of laboratories that fulfill the widely accepted analytical per formance goals based on the biological variation. The median LCVa was 1.4% (range 0.4%-3.4%). The LCVa was less than 0.58 times the total biological variation (diagnostic testing) for all laboratories (100%) and was less than 0.5 times the within biological variation (monitoring testing) in 62 % of the laboratories. The median Total Bias was 1.1% (range 0.1%- 3.3%). 83% of the laboratories had a total bias less than 0.25 of the total biological variation. Par ticipation in External Quality Assessment schemes is of extreme impor tance, as it provides information to the laborator y about its own method per formance both in a single sur vey as well as over time and allows the evaluation of the needs for improvements.
It is impor tant for medical laboratories to know about the stability and performance of their methods over the time. The main objective of this study was to apply a linear regression model for long-term evaluation of the performance of laboratories in haemoglobin quantification by comparing the laborator y individual results with the consensus mean of each round, af ter outliers exclusion. To evaluate the per formance of the laborator y, the long term analytical CV (LCVa) and the total analytical bias were established. A linear regression model was applied to quantitative haemoglobin results to evaluate the long-term analytical per formance of thir ty laboratories that par ticipate in the PNAEQ external quality assessment programme (2014- 2016), using the results of twelve blood EDTA samples with dif ferent hemoglobin concentrations. Par ticipants were randomly selected concerning laboratories from hospital and ambulator y. The variables introduced to define the long-term per formance in this model were the long-term analytical CV (LCVa) and total analytical Bias. Two laboratories were excluded from the analysis (one outlier and one that uses point-of-care equipment). We evaluate also the number of laboratories that fulfill the widely accepted analytical per formance goals based on the biological variation. The median LCVa was 1.4% (range 0.4%-3.4%). The LCVa was less than 0.58 times the total biological variation (diagnostic testing) for all laboratories (100%) and was less than 0.5 times the within biological variation (monitoring testing) in 62 % of the laboratories. The median Total Bias was 1.1% (range 0.1%- 3.3%). 83% of the laboratories had a total bias less than 0.25 of the total biological variation. Par ticipation in External Quality Assessment schemes is of extreme impor tance, as it provides information to the laborator y about its own method per formance both in a single sur vey as well as over time and allows the evaluation of the needs for improvements.
Description
Keywords
Avaliação Externa da Qualidade Hemoglobina Laboratórios Portugueses Programa Nacional de Avaliação Externa da Qualidade PNAEQ Saúde Pública Portugal
Pedagogical Context
Citation
Boletim Epidemiológico Observações. 2016 setembro-dezembro;5(17):29-31
Publisher
Instituto Nacional de Saúde Doutor Ricardo Jorge, IP