Laboratory quality can be assessed through participation in external quality assessment (EQA) programs and laboratory accreditation, which involves on-site inspection. Since 1976, the Korean Association of External Quality Assessment Service (KEQAS) has been the leading authorized EQA institute for the standardization and quality management in clinical laboratories in Korea. In 2020, the number of EQA programs exceeded 70, covering over 850 test items [4, 5]. The Outstanding Laboratory Accreditation Program (OLAP), administered by the Laboratory Medicine Foundation (LMF), has been conducting inspection programs with clinical pathologists since 1999. Compared with unaccredited laboratories, OLAP-accredited laboratories have exhibited greater accuracy in various tests, including those for albumin, AST, ALT, and hemoglobin [6, 7]. This program is internationally certified by the International Society of Quality in Health Care External Evaluation Association [8] and play essential roles in the quality management of diagnostic tests. However, NCSP does not require participating institutions to engage in EQA or to obtain laboratory accreditation. In this study, we aim to emphasize the importance of EQA participation and OLAP accreditation within NCSP by assessing the participation rates of institutions in these programs and analyzing the effect of laboratory accreditation based on EQA results.

The AFP result data were provided by the National Health Insurance Service (NHIS) after a thorough review (NHIS-2024-1-414). Data on the participation of NCSP institutions in EQA were obtained from KEQAS, while information on laboratory accreditation was provided by LMF. The NHIS database from 2018 to 2020 revealed that, on average, 4,728 institutions participated in the HCC screening program every year. These institutions comprised an average of 345 general hospitals with >100 beds, 682 small hospitals with 30–99 beds, and 3,700 clinics with <30 beds. Institutions were categorized based on their participation in KEQAS, accreditation from OLAP, and whether they conducted AFP tests in their own laboratories. According to the guidelines for calculating test fees published by the Health Insurance Review and Assessment Service, 10% of the calculated amount is added as the outsourcing management fee when a test is outsourced to designated referral medical laboratories [9]. Therefore, institutions without additional bills for “outsourcing surcharge” in the insurance claims for patients’ AFP tests were presumed to be “performing AFP tests in their own laboratory” (Table 1).

From 2018 to 2020, the total number of institutions participating in the NCSP increased slightly, from 4,582 to 4,847. During this period, the percentage of general hospitals conducting AFP tests in their own laboratories rose from 86.8% to 91.2%, and the percentage of small hospitals also increased from 30.2% to 33.5%. Conversely, the percentage of clinics performing AFP tests decreased from 9.2% to 7.7% (Fig. 1). Among institutions conducting AFP tests within their own laboratories, an average of 98.9% of the general hospitals participated in KEQAS, and 75.2% were accredited by OLAP. For small hospitals and clinics, 68.5% and 24.8% participated in KEQAS, respectively; moreover, 1.4% of small hospitals and 0.3% of the clinics held OLAP accreditation (Table 1). These results were consistent with previous findings indicating low incidence of KEQAS participation among small hospitals and clinics [10] and their low-quality levels [11].

To evaluate the effect of OLAP accreditation on AFP result accuracy, the standard deviation index (SDI) was calculated using three-year data (2018–2020) from KEQAS (Table 2). The SDI was calculated by subtracting the average value of the group from the measured value of the institution, then dividing the result by the standard deviation of the group. The difference between the mean values of the absolute SDI (mSDI) for the OLAP-accredited and non-accredited laboratories was assessed using a t-test. A lower mSDI shows that the AFP results are closer to the average of participating laboratories, indicating the higher reliability of these results. Compared with non-accredited laboratories, OLAP-accredited institutions exhibited significantly lower mSDI values across all three years (P<0.0001). Furthermore, percentages of institutions with an absolute SDI greater than 2 in 2018, 2019, and 2020 were 0.0%, 0.0%, and 1.2%, for OLAP-accredited laboratories, and 7.2%, 8.3%, and 8.5% for non-accredited ones, respectively (P<0.001). For general hospitals specifically, mSDI values of OLAP- accredited institutions were significantly lower than those of non-accredited ones in 2018 and 2020 (Table 2). The higher P-value in 2019 may be attributed to a slight decrease in mSDI for non-accredited general hospitals from the previous year (P=0.2314) and a limited sample size. Overall, AFP results from OLAP-accredited laboratories were significantly closer to the consensus value, with results from OLAP-accredited general hospitals aligning more closely with consensus value in two of the three studied years.

While the observed differences in mSDIs suggest a potential impact of OLAP accreditation, other factors may also contribute to these variations. Our analysis controlled for the type of AFP test by focusing exclusively on quantitative assays, as less than 5% of institutions perform qualitative testing, which was therefore excluded. Additionally, we accounted for hospital size by focusing on general hospitals; however, further analysis was not feasible because of the limited number of secondary-level and smaller participating institutions. Future studies with a more controlled design and larger sample size could help to better isolate the effects of accreditation and address additional confounding factors.

By analyzing participation rates and comparing performance between accredited and non-accredited laboratories, this study revealed that OLAP-accredited institutions demonstrated higher consistency in AFP results. This is particularly meaningful for high-risk patients, as inaccurate AFP measurements can lead to unnecessary diagnostic procedures and increased healthcare costs [3]. Encouraging NCSP institutions to actively participate in KEQAS and acquire OLAP accreditation could enhance AFP testing quality, thereby improving the effectiveness of HCC surveillance and reducing the financial burden on patients.