VigiRank prediction model of Uppsala monitoring center

doi:10.19803/j.1672-8629.2022.03.08

Abstract

Abstract: Objective To outline the principles, processes and applications of the vigiRank signal detection method in Uppsala Monitoring Centre (UMC) so as to provide reference for the establishment and improvement of pharmacovigilance signal detection statistics in China. Methods Scopus, Web of Science, PubMed and other databases were searched for the purpose of literature research, and the UMC official website was consulted to analyze the workflow and application of the vigiRank prediction model. Results VigiRank,a signal detection method developed by UMC, combined multiple strength-of-evidence aspects of adverse drug reaction (ADR) reports, including Informative reports (INF), Narrative (NAR), Dechallenge (DCH), Rechallenge (RCH), Causality assessment (CAU and CAU+), Time-to-onset (TTO), Sole reporting (SOL), Multiple reporting elements (MUL), Resultsreporting (REC), Disproportional reporting (DIS), Geographic spread (GEO), and Time trend (TRE). According to the lasso-logistic model, five predictive variables were included in the vigiRank score system and the vigiRank score of a drug-ADR pair was calculated. An expert team conducted a preliminary evaluation in the descending order of scores, and chose drug-ADR pairs that needed to be further analyzed before they conducted an in-depth evaluation and finally screened out risk signals. Conclusion VigiRank has taken into account the number and quality of ADR reports, and is able to identify early signs of harm from drugs, which is of great referential value for improving the detection methods of ADR signals in China.

Key words: adverse drug reaction, signal detection, Uppsala Monitoring Center, vigiRank

CLC Number:

R994.1

CHEN Chao, ZHAO Yunfei, WANG Changzhi, GONG Lixiong, LIU Wei. VigiRank prediction model of Uppsala monitoring center[J]. Chinese Journal of Pharmacovigilance, 2022, 19(3): 270-274.

References

[1] CIOMS Working Group VIII. Practical aspects of signal detection in pharmacovigilance[R]. Geneva: Cioms, 2010.
[2] National Medical Products Administration. Annual report for national adverse drug reaction monitoring (2020)[EB/OL].(2021-03-26) [2021-04-28]. https://www.nmpa.gov.cn/xxgk/fgwj/gzwj/gzwjyp/20210325170127199.html.
[3] STRANDELL J, CASTER O, HOPSTADIUS J, et al.The development and evaluation of triage algorithms for early discovery of adverse drug interactions[J]. Drug Safety, 2013, 36(5): 371-388.
[4] ZHANG W, WEI FQ, YANG Y.Study of vigibase database of uppsala monitoring centre[J]. Chinese Journal of Pharmacovigilance(中国药物警戒), 2020, 17(10): 676-680.
[5] Uppsala Monitoring Centre. Signal detection at UMC[EB/OL] (2020-09-16)[2021-04-28]. https://www.who-umc.org/research-scientific-development/signal-detection/signal-detection-at-umc/.
[6] CASTER O, JUHLIN K, WATSON S, et al.Improved statistical signal detection in pharmacovigilance by combining multiple strength-of-evidence aspects in Vigirank[J]. Drug Safety, 2014, 37(8): 617-628.
[7] Uppsala Monitoring Centre. Signal detection at UMC[EB/OL].(2020-09-16) [2021-04-28]. https://www.who-umc.org/research-scientific-development/signal-detection/signal-detection-at-umc/.
[8] BERGVALL T, NORÉN G, LINDQUIST M. Vigigrade: a tool to identify well-documented individual case reports and highlight systematic data quality issues[J]. Drug Safety, 2013, 37(1): 65-77.
[9] ICH.E2B(R2) ICSR specification and related files[EB/OL]. (2020-09-16)[2021-04-28]. https://ich.org/page/e2br2-icsr-specification-and-related-files.
[10] Uppsala Monitoring Centre. What is a signal?[EB/OL]. (2020-09-16) [2021-04-28]. https://www.who-umc.org/research-scientific-development/signal-detection/what-is-a-signal/.
[11] NOREN GN, HOPSTADIUS J, BATE A.Shrinkage observed-toexpected ratios for robust and transparent large-scale pattern discovery[J]. Statistical Methods in Medical Research, 2013, 22(1): 57-69.
[12] FRIEDMAN JH, HASTIE T, TIBSHIRANI R.Regularization paths for generalized linear models via coordinate descent[J]. Journal of Statistical Software, 2010, 33(1): 1-22.
[13] CASTER O, SANDBERG L, BERGVALL T, et al.Vigirank for statistical signal detection in pharmacovigilance: first results from prospective real-world use[J]. Pharmacoepidemiology & Drug Safety, 2017, 26(8): 1006-1010.
[14] WATSON S, CHANDLER RE, TAAVOLA H, et al.Safety concerns reported by patients identified in a collaborative signal detection workshop using vigibase: results and reflections from lareb and uppsala monitoring centre[J]. Drug Safety, 2018,41(2):203-212.
[15] STAR K, SANDBERG L, BERGVALL T, et al.Paediatric safety signals identified in vigibase: methods and results from uppsala monitoring centre[J]. Pharmacoepidemiology and Drug Safety, 2019,28(5): 680-689.
[16] YUE YQ. Indapamide-induced (acute hypokalaemic) rhabdomyolysis[EB/OL]. (2020-09-16) [2021-04-28]. https://www.who-umc.org/media/165201/indapamide-inducedfinalweb.pdf.
[17] Uppsala Monitoring Centre. Communicating signals[EB/OL]. (2020-09-16) [2021-04-28]. https://www.who-umc.org/research-scientific-development/signal-detection/communicating-signals/.
[18] HOU YF, LIU CL, SONG HB, et al.Optimized signal detection model in adverse drug reaction monitoring system[J]. Chinese Journal of Pharmacovigilance(中国药物警戒), 2016, 13(7): 408-409, 421.
[19] GONG LX, CHEN C, CHEN XB, et al.Study on quality assessment method of adverse reactions reports of cosmetics[J]. China Pharma-ceuticals(中国药业), 2019, 28(7): 76-78.