Abstract
BACKGROUND: Assessing the efficacy of chemical disinfectants requires precise and reproducible test procedures. In Germany, the Commission for Infection Prevention and Hygiene in Healthcare and Nursing (KRINKO) at the Robert Koch Institute, the federal government public health institute, recommends that two test reports from independent accredited laboratories, accompanied by expert opinions, to assess the efficacy of disinfectants for use in settings which harbor a special risk for acquiring infections. This requirement, also endorsed by the Association of Applied Hygiene (VAH) for the certification of disinfectants, is based on the inherent statistical variance of microbiological testing methods. METHODS: This paper presents the analysis of data from four VAH-organized ring trials to underline the statistical relevance of the two-test stipulation. Specifically, two quantitative suspension tests were evaluated, one testing efficacy of a test product against Staphylococcus aureus according to EN 13727 and one testing efficacy against Candida albicans according to EN 13624. In addition, two phase 2 step 2 tests to evaluate the bactericidal efficacy of test products against Enterococcus hirae using the quantitative test on non-porous surfaces without mechanical action according to EN 17387 and mycobactericidal efficacy against Mycobacterium terrae using the quantitative carrier test for instruments according to EN 14563 were analyzed. A one-way analysis of variance (ANOVA) was performed to assess interlaboratory variations and a Monte-Carlo simulation study was conducted to evaluate the impact of including results from a second laboratory on the reliability of results. RESULTS: The ANOVA revealed a significant interlaboratory variation. In addition, descriptive analysis indicated substantial intralaboratory variability across trials. The simulation analysis indicates that requiring matching results from two laboratories provides a more conservative and consistent assessment of efficacy. The most marked effects occurred at intermediate concentration-time ratios, where the implementation of the two-laboratory stipulation resulted in a 69-89% reduction in the number of efficacy classifications, thereby significantly mitigating the risk of overly optimistic conclusions. Within the clearly effective or ineffective ranges, however, the differences between the two approaches (one-laboratory rule vs. two-laboratory rule) were minimal. CONCLUSION: The study demonstrates considerable interlaboratory variability and illustrates that including results from a second independent laboratory can reduce inconsistencies in efficacy assessments. These findings support the rationale behind requiring two independent test reports. The analysis underlines the importance of ring trials as a key element of quality assurance and method validation in disinfectant efficacy testing.