Causal inference methods for evaluating comparative effectiveness: target trial emulation

Abstract

Causal inference is grounded in contrasts between potential outcomes under alternative interventions. Randomized trials are the reference standard for estimating average causal effects because randomization renders treatment assignment independent of potential outcomes, eradicating confounding by design. However, many clinically important questions in rheumatology cannot feasibly be addressed through randomized experiments due to practical, ethical, or temporal constraints. In such settings, observational data can inform decisions. This paper argues that principles derived from randomized trials should continue to anchor causal reasoning and proposes the target trial framework as a structured approach to strengthen causal inference from observational data. By explicitly specifying the protocol of the hypothetical randomized trial that would answer the question and emulating it using real-world (observational) data, investigators can clarify eligibility criteria, treatment strategies, time zero, outcomes, causal contrasts, and identifying assumptions. By describing how each part of the target trial is emulated in observational data, they increase transparency, clarify the causal estimand, and make assumptions, limitations, and sources of bias explicit. This design-based perspective helps prevent common biases, including immortal time bias, selection of prevalent users, and inappropriate conditioning on post-treatment variables, and aligns reporting with clearly defined causal contrasts and effect measures. Ultimately, credibility of causal inference from observational data depends on whether the assignment mechanism is plausibly reconstructed, design choices are made without access to outcome data, and analyses target explicitly defined causal contrasts under stated assumptions, all of which are explicitly addressed within the target trial emulation framework.

Keywords

causal inference / target trial emulation / real-world data / observational studies / rheumatology

References

[1] Little RJ, Rubin DB. Causal Effects in Clinical and Epidemiological Studies Via Potential Outcomes: Concepts and Analytical Approaches. Annu Rev Public Health. 2000; 21: 121-45.

[2] Rubin DB. The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trials. Stat Med. 2006; 26: 20-36.

[3] Hernán MA, Robins JM. Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available: Table 1. Am J Epidemiol. 2016; 183: 758-64.

[4] Hernán MA, Wang W, Leaf DE. Target Trial Emulation. JAMA. 2022; 328: 2446-7.

[5] Djulbegovic B, Guyatt GH. Progress in evidence-based medicine: a quarter century on. Lancet. 2017; 390: 415-23.

[6] Murad MH, Asi N, Alsawas M, Alahdab F. New evidence pyramid. Evid Based Med. 2016; 21: 125-7.

[7] Zhang Y, Akl EA, Schünemann HJ. Using systematic reviews in guideline development: The GRADE approach. Res Synth Methods. 2018; 10: 312-29.

[8] Schünemann HJ, Tugwell P, Reeves BC, Akl EA, Santesso N, Spencer FA, et al. Non-randomized studies as a source of complementary, sequential or replacement evidence for randomized controlled trials in systematic reviews on the effects of interventions. Res Synth Methods. 2013; 4: 49-62.

[9] Hernán MA. Methods of Public Health Research - Strengthening Causal Inference from Observational Data. N Engl J Med. 2021; 385: 1345-8.

[10] Hernán MA, Dahabreh IJ, Dickerman BA, Swanson SA. The Target Trial Framework for Causal Inference From Observational Data: Why and When Is It Helpful? Ann Intern Med. 2025; 178: 402-7.

[11] Agoritsas T, Merglen A, Shah ND, O'Donnell M, Guyatt GH. Adjusted Analyses in Studies Addressing Therapy and Harm. JAMA. 2017; 317: 748-59.

[12] Cashin AG, Hansford HJ, Hernán MA, Swanson SA, Lee H, Jones MD, et al. Transparent reporting of observational studies emulating a target trial: the TARGET Statement. BMJ. 2025; 390: e087179.

[13] Hernán MA, Robins JM. Causal Inference: What If? Boca Raton: Chapman & Hall/CRC; 2020.

[14] Taubman SL, Robins JM, Mittleman MA, Hernán MA. Intervening on risk factors for coronary heart disease: an application of the parametric g-formula. Int J Epidemiol. 2009; 38: 1599-611.

[15] Robins JM, Hernán MA, Brumback B. Marginal Structural Models and Causal Inference in Epidemiology. Epidemiology. 2000; 11: 550-60.

[16] Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983; 70: 41-55.

[17] Smith LH, García-Albéniz X, Chan JM, Zhao S, Cowan JE, Broering JM, et al. Emulation of a target trial with sustained treatment strategies: an application to prostate cancer using both inverse probability weighting and the g-formula. Eur J Epidemiol. 2022; 37: 1205-13.

[18] Skoetz N, Goldkuhle M, van Dalen EC, Akl EA, Trivella M, Mustafa RA, et al. GRADE guidelines 27: how to calculate absolute effects for time-to-event outcomes in summary of findings tables and Evidence Profiles. J Clin Epidemiol. 2020; 118: 124-31.

[19] Hetland ML, Christensen IJ, Tarp U, Dreyer L, Hansen A, Hansen IT, et al. Direct comparison of treatment responses, remission rates, and drug adherence in patients with rheumatoid arthritis treated with adalimumab, etanercept, or infliximab: Results from eight years of surveillance of clinical practice in the nationwide Danish DANBIO registry. Arthritis Rheum. 2009; 62: 22-32.

[20] Amris K, Christensen R, Wæhrens EE, Duhn PH, Henriksen M. Impact of enhanced pain knowledge on core outcomes in fibromyalgia patients with high self-reported pain education needs: a target trial emulation using the DANFIB registry. RMD Open. 2025; 11: e005576.

[21] Rasmussen LE, Hoffmann AG, Blanche P, Espersen F, Justesen TF, Rasmussen LE, et al. Surgeon Training and Revision Rates After Patellofemoral Arthroplasty. JAMA Netw Open. 2025; 8: e2517825.

[22] Christensen R, Ranstam J, Overgaard S, Wagner P. Guidelines for a structured manuscript: Statistical methods and reporting in biomedical research journals. Acta Orthop. 2023; 94: 243-9.