intention to treat
An analysis principle in randomized trials where you compare everyone assigned to the treatment group against everyone in the control group based on assignment alone, regardless of whether they actually received the treatment, preserving the integrity of randomization.
As defined in Jim Manzi on the Oregon Medicaid Study, Experimental Evidence, and Causality
Other definitions
A medical research approach that studies the effect of offering a treatment to people rather than the effect of treatment actually received, capturing real-world imperfect compliance.
Megan McArdle on Catastrophes and the Pandemic
Where it comes up · 2
- Jim Manzi on the Oregon Medicaid Study, Experimental Evidence, and Causality3 claims
- Megan McArdle on Catastrophes and the Pandemic1 claim
Most important claims · 4 of 4Search all claims for “intention to treat” →
- Medical research distinguishes treating people from merely offering treatment (intention-to-treat), since real-world policies resemble the latter—most patients stop taking hypertension meds as directed within nine months—so lockdowns and mask mandates would be highly effective if perfectly followed, but the right critique is that the actual legal policies do less than expected because compliance is imperfect, not that the interventions themselves don't work.Megan McArdle on Catastrophes and the Pandemic
- If the lottery winners who enrolled in Medicaid are more prudent, conscientious, and self-disciplined than those who didn't enroll, then comparing the enrolled (all prudent) against a control group that is a mix of prudent and imprudent people will mis-ascribe improvements actually caused by prudence to the effect of Medicaid coverage.Jim Manzi on the Oregon Medicaid Study, Experimental Evidence, and Causality
- Once you stop analyzing the full randomized groups (intention-to-treat) and instead analyze only the subset of lottery winners who actually obtained Medicaid, you reintroduce the very selection-bias problem randomization was meant to eliminate, because you no longer know which losers are comparable to the self-selected enrollees; Manzi calls this introducing 'the bacillus of econometrics.'Jim Manzi on the Oregon Medicaid Study, Experimental Evidence, and Causality
- Because only about 25-30% of lottery winners actually received Medicaid, the intention-to-treat effect (the causal effect of winning the lottery) is roughly one quarter of the reported treatment effect; you can divide the headline per-person effects by four, and Manzi's critique of the study only strengthens under this stricter analysis.Jim Manzi on the Oregon Medicaid Study, Experimental Evidence, and Causality