Intent-to-Treat

Definitions #

A clinical trial method where all participants are analyzed in the group they were originally assigned to, regardless of treatment received.

Technical The principle that outcome data from all randomized subjects, including dropouts and crossovers, are analyzed according to initial randomization to preserve comparability.

Example Per the intent-to-treat protocol, the patient who never received the drug was still counted in the treatment arm's results.

Etymology #

From clinical trial methodology; the term emerged in the 1960s–1970s as the 'intention to treat' approach advocated by ICH E9 guidelines.

Collocations #

ITT analysis 意向性分析（保持随机化，分析所有分配对象）
ITT principle ICH E9 中的意向性治疗原则
ITT vs per-protocol 对比：ITT 用全部随机对象，PP 仅用完成方案者

Real Business Examples #

The ITT analysis included all 1,200 randomized patients, even those who withdrew after the first visit.
Regulatory agencies prefer an intent-to-treat approach as the primary analysis to avoid bias.

Register & Variants #

Register: formal
BrE/AmE: 全球通用

Synonyms & Antonyms #

Antonyms: per-protocol analysis

Common Errors #

将 ITT 与分析集（mITT）混淆，mITT 排除未接受治疗者 ITT 包括所有随机者；mITT（修改 ITT）只包括接受至少一次治疗且有基线数据的患者
在 ITT 中排除失访者但未解释 ITT 要求纳入所有随机者，失访者需采用合适方法（如敏感性分析）处理，不能简单剔除

Confusable Terms #

Per-protocol analysis vs Intent-to-Treat: PP 仅分析按方案完成者，可能高估疗效；ITT 保留随机化优势，避免选择偏倚
Modified ITT (mITT) vs Intent-to-Treat: mITT 从 ITT 中剔除未接受任何研究治疗的受试者，比 ITT 窄

Frequently Asked Questions

为什么监管机构首选 ITT？?

ITT 保持随机化带来的基线可比性，反映真实临床效果（effectiveness），而非解释性疗效（efficacy）

ITT 是否会低估疗效？?

可能，但更保守，不易偏袒试验药，因此被接受为主要分析

Published: 2026-09-22 Category: Clinical