Self-controlled case series design in vaccine safety: a systematic review

ABSTRACT

Background

There is limited comprehensive evaluation of the methodology and reporting quality of observational studies of vaccine safety.

Methods

Databases including Medline, Embase, Web of Science, Scopus, and Chinese databases were searched from inception to 31 May 2021. All observational studies regarding vaccine safety using an SCCS design were selected. Information regarding methodological elements were extracted. In addition, reporting quality was assessed using the REporting of studies Conducted using Observational Routinely collected health Data statement for PharmacoEpidemiology (RECORD-PE).

Results

: Of the 105 studies identified, administrative databases were the main data source for vaccination records and adverse events following immunization (AEFI). Twenty-eight articles (27%) used multiple designs to verify the association, and the results obtained with the SCCS design were robust. The top three AEFI studied were intussusception, Guillain-Barré syndrome, and convulsions. Only 21 studies (20%) reported the approach for case validation by chart review. The healthy vaccinee effect was considered by 51 studies (49%), with 16 of them (31%) using extended SCCS models to alleviate this effect. Overall, the reporting quality of included studies could be improved.

Conclusions

Administrative databases were the main data source for vaccination records and adverse events following immunization. Case validation, the validity of assumptions for standard SCCS, and quality of reporting should be given more importance in future research projects.

KEYWORDS:

• Self-controlled case-series design
• vaccine
• safety
• systematic review
• reporting quality
• RECORD-PE

Article highlights

• Administrative databases were the main data source for vaccination records and adverse events following immunization.

• The direction of association detected using SCCS designs were consistent with the results obtained using other designs, including cohort, case-control, case-crossover, and self-controlled risk interval designs.

• The healthy vaccinee effect should be considered in all studies about vaccine safety.

• The reporting quality of included studies using the SCCS design in vaccine safety could be improved.

Acknowledgments

We would like to thank the contributions of the research assistants (Xue Nie, Jieqiong Niu, Zhe Zhao, and Zhen Hu from School of Public Health, Peking University) to this study.

Author contributions

X.N., L.X., Z.L. and S.Z. designed the study. X.N., Z.L., Z.L., and Y.B. searched the articles. L.X., X.N., Y.B., and Z.L screened the included articles. X.N., L.X., and Y.B. designed the analytical strategy and helped to interpret the findings. X.N., and L.X. wrote the original draft. Z.L. and P.F. reviewed and edited the manuscript. S.Z. supervised the study’s implementation and reviewed the manuscript. S.Z. provided the funding support. All authors approved the final version of the manuscript and agreed to the published version of the manuscript.

Declaration of interests

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants, or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (Grant no. 91846303,81973146).