Statistical considerations for real world studies supporting new drug registrations

doi:10.19803/j.1672-8629.2022.01.06

Abstract

Abstract: Objective To review statistics-related problems in the process of obtaining real world evidence (RWE) from real world research (RWS). Methods In regard to real world data (RWD), causal inference methods for RWS and approaches to RWE evaluation were reviewed while the existing problems were analyzed and discussed. Results Obtaining reliable RWE that could be used to support regulatory decisions about medical products was an important goal of RWS. An appropriate standard for RWD data was a prerequisite for quality evaluation of the data. A standardized process of data governance and a perfect system for data quality evaluation underlay high-quality RWS. Standardized implementation of causal inference analytical methods could ensure the quality of RWE. Clinically explicable and innovative conclusions were critical to RWE. Objectivecriteria for evaluation of RWE were also needed in practice. The verifiability of results of RWS was the most important feature of RWE. Conclusion A single clinical trial is often unable to meet the criteria for determining causality. Instead, multiple studies are required to verify causality from different perspectives. Therefore, the combination of randomized controlled trials (RCT) and RWS is a good approach during drug development.

Key words: real world study, real world data, real world evidence, causal inference, data quality

CLC Number:

R911

HUANG Lihong, CHEN Feng. Statistical considerations for real world studies supporting new drug registrations[J]. Chinese Journal of Pharmacovigilance, 2022, 19(1): 27-31.

References

[1] BLACK N.Why we need observational studies to evaluate the effectiveness of health care[J]. BMJ, 1996, 312(7040): 1215-1218.
[2] LI M, SHI JP, YU HH.Relationship between the“Real World” research, randomized controlled trial and number of one randomized controlled trial in clinical therapeutic study[J]. Chinese Journal of Epidemiology(中华流行病学杂志), 2012, 33(3): 342-345.
[3] Food and Drug Administration. Use of real world evidence to support regulatory decision marking for medical devices[EB/OL]. (2017-08-31)[2021-08-14]. https://www.fda.gov/media/99447/download.
[4] National Medical Products Administration. Guidelines for drug development and review supported by real-world evidence (trial document)[EB/OL]. (2020-01-03)[2021-08-14]. https://www.nmpa.gov.cn/xxgk/ggtg/qtggtg/20200107151901190.html.
[5] National Medical Products Administration. Real-world data guidelines for generating real-world evidence (Trial document)[EB/OL]. (2021-04-15)[2021-08-14] https://www.cde.org.cn/main/news/viewInfoCommon/2a1c437ed54e7b838a7e86f4ac21c539.
[6] BOOTH CM, KARIM S, MACKILLOP WJ.Real-world data: towards achieving the achievable in cancer care[J]. Nat Rev Clin Oncol, 2019, 16(5): 312-325.
[7] CHEN DF, LIU H.Methods and applications of medical big data mining(医学大数据挖掘方法与应用)[M]. Beijing: Peking University Medical Press, 2020.
[8] MAKADY A, HAM RT, DE BOER A, et al.Policies for use of real-world data in health technology assessment (HTA): a comparative study of six HTA agencies[J]. Value Health, 2017, 20(4): 520-532.
[9] KAHN MG, CALLAHAN TJ, BARNARD J, et al.A harmonized data quality assessment terminology and framework for the secondary use of electronic health record data[J]. EGEMS (Wash DC), 2016, 4(1): 1244.
[10] Food and Drug Administration. Standardization and querying of data quality metrics and characteristics for electronic health data-data quality metrics system final report [EB/OL]. (2019-12-31)
[ 2021-08-14]. https://dataquality.healthdatacollaboration.net/.
[11] PEARL J, GLYMOUR M, JEWELL NP.Causal inference in statistics: a primer[M]. John Wiley, 2016.
[12] HERNÁN MA, ROBINS JM. Causal inference: what if[M]. Boca Raton: Chapman & Hall/CRC, 2020.
[13] QIU JP.Practical measurement methods for causal inference(因果推断实用计量的方法)[M]. Shanghai: Shanghai University of Finance and Economics Press, 2020.
[14] HUANG LH, WEI YY, CHEN F.Confounder adjustment in observational comparative effectiveness researches II: statistical adjustment approaches for unmeasured confounders[J]. Chinese Journal of Epidemiology(中华流行病学杂志), 2019, 40(11): 1450-1455.
[15] STERNE JA, HERNÁN MA, REEVES BC, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions[J]. BMJ, 2016, 355: i4919.
[16] SHEN HB.Epidemiology (Vol. 3)[流行病学(第3卷)][M]. Beijing: People's Medical Publishing House, 2014.
[17] TANG JL, YANG ZY.Observation versus experiment, efficacy versus effectiveness[J]. Chinese Journal of Epidemiology(中华流行病学杂志), 2014, 35(3): 221-222.