ABSTRACT
The term 'external quality assessment (EQA)' has been recommended by WHO consultative meeting in 1980. It has been using to run the retrospective and com-parative statistic tests' reliability among the EQA participants of their organizers. EQA schemes in all fields of laboratory pathology should provide objective information as tools to assist laboratory analysts ensuring the test result quality and on the contrary deciding error-eradicated actions, or more or less improving national test performance quality and standard. Scoring assessment has been designed in the EQA schemes to assist monitoring, and encouraging quality and standard of laboratory performance in the long run of consecutive trials. The Faculty of Medical Technology of Mahidol University now as a collaborating WHO:SEARO EQA organizers, has provided wider EQA schemes, servicing participants all around Thailand, including 400 labs in clinical chemistry of EQAC scheme, 150 labs in clinical hormone of EQAH scheme, 200 labs in clinical microscopy of EQAM scheme, and 190 labs in clinical immunology and serology of EQAI schemes. Regular EQA newsletters, and yearly summer-time seminar and training prove to be the essential education tools promoting both laboratory quality and standard intensively.
Subject(s)
Chemistry, Clinical/standards , Humans , Laboratories/standards , Clinical Laboratory Techniques/standards , Quality Assurance, Health Care , Quality Control , ThailandABSTRACT
Tested serum of donors from the Blood Bank unit of Siriraj Hospital was collected. Only negative hepatitis B-antigen determined by RPHA (reverse passive haemagglutination) method was selected. The sera were pooled as one pooled serum. Filtrations were performed after adjusting some biochemical constituents. To obtain lyophilized serum, the pooled serum was subjected to freeze-dry process. It was divided into aliquot of 5.0 ml in each vial, then quick freezed and dried in the lyophilizer. Precision of between-vial filling volume was studied. We found that the coefficient if variation was 0.01%. Reconstitution of the lyophilized serum before use was performed by adding 5.0 ml of distilled water to each vial. The coefficient variation of between-vial biochemical constituents was found to be less than 0.5%. The precision of 19 routine biochemical tests in these Home-Made lyophilized human serum were compared to those of 3 batches of well-known commercial control materials (ORTHO, MONITROL, VALIDATE). By F-test analysis, the coefficient variation of most biochemical tests in the Home-Made lyophilized human serum was not significant different from those of the commercial ones (P>0.05). The long-term consistency of the biochemical constituents was studied. In a duration of one year, from March 1985 to February 1986, three-month periods means and standard deviations were analyzed. The results showed no significant difference for these period of study (P>0.05). Since 1985, the Home-Made lyophilized serum has been used as internal control material in our Central Clinical Chemistry Laboratory, Faculty of Medicine Technology, Mahidol University in place of imported control serum. The external control evaluation was successfully reported by the International External Quality Control Assessment Scheme. (Birmingham UK) and the improvement of Overall Mean Running Variance Index Score (OMRVIS) is obviously seen in the graphic curve.
ABSTRACT
Two batches of lyophilized control serum, LYOPHIL (M.Tech. Diagnostics, Thailand) and VALIDATE (General Diagnostics, USA) were distributed weekly as blind internal quality control for our central clinical chemistry laboratory. The precision of 19 biochemical tests was studied. We found that the routine condition variance (RCV) of most tests was not greater than the RCV claimed by Wilding at al except for the test of chloride and amylase. Chisquare test and Youden plot were used to analyze the two batches of control serum. The results show good correlation for both in and out control except for creatinine and phosphate. Consistency of the biochemical tests in both batches was evaluated by analysis of variance (ANOVA) of sets of consecutive 10-week period means for 1 year. Most of the mean values of each test in LYOPHIL and VALIDATE were not significantly different except those of bilirubin and creatinine (p>0.05).