Researchers are often challenged with assessing the impact of an intervention on an outcome of interest in situations where the intervention is nonrandomised, the intervention is only applied to one or few units, the intervention is binary, and outcome measurements are available at multiple time points. In this paper, we review existing methods for causal inference in these situations. We detail the assumptions underlying each method, emphasize connections between the different approaches and provide guidelines regarding their practical implementation. Several open problems are identified thus highlighting the need for future research.
Bibliographical noteFunding Information:
We acknowledge funding and support from NIHR Health Protection Unit on Evaluation of Interventions ((PS, MH, DDA), Medical Research Council grants MC_UU_00002/10 (SRS) and MC_UU_00002/11 (DDA, AMP), Public Health England (DDA), and NIHR PGfAR RP-PG-0616-20008 (EPIToPe, PS, MH, DDA). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.
© Institute of Mathematical Statistics, 2019.
- Causal impact
- Causal inference
- Intervention evaluation
- Latent factor models
- Panel data
- Synthetic controls