18 References
18.1 Introduction
This document summarizes the correspondence between the international guidelines (the CLSI EP series) for each analysis module of ivdtools and the main methodological references. The correspondences are used to assist understanding and analysis-plan design; they do not constitute a regulatory determination; for registration, review, or accreditation matters, implement according to the applicable standard in force at the relevant time.
About domestic (Chinese) standards: Domestic standards fall into two categories — WS/T are health-industry standards (mainly aimed at clinical-laboratory practice), and YY are pharmaceutical/medical-industry standards (aimed at medical devices/in vitro diagnostic products; YY/T is recommended). For the same topic there may be both a laboratory-practice standard and a product-performance standard; when choosing, first clarify the applicable object (laboratory workflow or product evaluation).
18.2 Module-to-CLSI-Guideline Mapping
The correspondence between the modules of ivdtools and the guidelines of CLSI (Clinical and Laboratory Standards Institute) is as follows:
| Module | Main entry | CLSI guideline | Main functions |
|---|---|---|---|
| Method comparison | mcr() |
EP09 | regression, correlation, Bland–Altman, bias, outliers |
| Precision | precision() |
EP05 | variance components, confidence intervals, Sadler profile |
| Qualitative analysis | raw_to_table() etc. |
EP12 | four-fold table, diagnostic metrics, Kappa, McNemar |
| Reference intervals | reference_interval() |
EP28 | percentile/parametric/robust methods |
| Stability | stability_*(), arrhenius(), mkt() |
EP25 | bias, regression, time-to-limit, MKT, Arrhenius |
| Linearity/curve fitting | fit_equation() series |
EP06 | linear/polynomial/4PL/5PL fitting |
| Analytical sensitivity/detection limit | lob_lod_loq() |
EP17 | LoB/LoQ, Sadler variance model |
| Bottle/lot ANOVA | bottle_anova() |
EP15 | one-way/nested ANOVA, Tukey |
| ROC | roc() series |
EP24 (reference) | AUC, cutoff, multivariate regression |
| Quality control | qc_chart(), youden_plot() |
Westgard rules | Levey–Jennings, Westgard, Youden |
| Sample size | sample_size_*() |
General | agreement/proportion/one-arm test sample sizes |
| Outliers and normality | outliers_test(), normal_test() |
General (pre-analysis steps) | Grubbs/ESD/Dixon/IQR, normality tests |
Before actual application, be sure to verify the officially published text and, if necessary, consult the competent authority or a qualified assessment body.
18.3 Methodological References
The main statistical methods of ivdtools refer to the following publications (included in the package DESCRIPTION):
- Bland J M, Altman D G. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, 1986, 327(8476): 307–310. doi:10.1016/S0140-6736(86)90837-8
- Passing H, Bablok W. A new biometrical procedure for testing the equality of measurements from two different analytical methods. Journal of Clinical Chemistry and Clinical Biochemistry, 1983, 21(11): 709–720. doi:10.1515/cclm.1983.21.11.709
- Linnet K. Evaluation of regression procedures for methods comparison studies. Clinical Chemistry, 1993, 39(3): 424–432. doi:10.1093/clinchem/39.3.424
- Hanley J A, McNeil B J. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology, 1982, 143(1): 29–36. doi:10.1148/radiology.143.1.7063747
- Horn P S, Pesce A J, Copeland B E. A robust approach to reference interval estimation and evaluation. Clinical Chemistry, 1998, 44(3): 622–631. doi:10.1093/clinchem/44.3.622
- Westgard J O, Barry P L, Hunt M R, et al. A multi-rule Shewhart chart for quality control in clinical chemistry. Clinical Chemistry, 1981, 27(3): 493–501. doi:10.1093/clinchem/27.3.493
- Lu M J, Zhong W H, Liu Y X, et al. Sample size for assessing agreement between two methods of measurement by Bland–Altman method. The International Journal of Biostatistics, 2016, 12(2). doi:10.1515/ijb-2015-0039
18.4 Usage Suggestions
- Defer to the applicable standard: Standards differ across countries/regions and applicable objects (laboratory vs product); first determine the applicable framework.
- Versions and supersession: Standards are updated or superseded; cite the version and implementation date when referencing.
- Separate statistics from compliance: This package provides statistical computation; whether all requirements of a given standard are met is determined by the complete validation/evaluation plan.
- Record and verify: Write the standard list, versions, and DOI/number into the analysis plan for review and audit.