Student or educator? Sign up with your school email and get 6 months of Pro free.
Cited by 235

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

Intention to treat – who should use ITT?

Open access
Lewis, JA, Machin, D·1993·British Journal of Cancer

Abstract

In the clinical researcher's perfect world every patient entered into a randomised controlled clinical trial (RCT) would satisfy all entry criteria, would complete their allocated treatment as described in the protocol, and would contribute data records which were complete in all respects. In practice it is doubtful if this ideal is ever achieved, and hence strategies have been developed for the analysis of RCTs which seek to protect their inferential basis from consequent biases. Such a strategy is 'intention to treat' (ITT) which is frequently advocated as the preferred approach to analysis. This is particularly so for major trials designed to establish definitively the efficacy and safety of a new medication or medical procedure. Indeed ITT has been endorsed in regulatory guidelines in Europe and the USA (Nordic Council on Medicines, 1989; Food & Drug Administration, 1988). But is ITT always the preferred approach? Can it always be satisfactorily applied? Is it a well-defined and well understood concept? The subject of ITT regularly arouses debate and controversy. In particular the idea that a patient who has been randomised to treatment A, but actually receives treatment B, should be assigned to the group who received A for analysis purposes is totally incomprehensible to some. This controversy is fuelled by the fact that there are a large number of clinical trials, particularly in the early phases of research, for which ITT would indeed be inappropriate. Further fuel is added by the different attitudes to ITT in different research environments it is a constant concern in the later phases of drug regulation, for example, whereas research workers in the Medical Research Council generally show greater flexibility, probably because they are less concerned with satisfying conservative regulatory requirements. It has even been argued that ITT encourages sloppiness 'Whatever we do, give the treatment or not, it is OK since the analysis is by ITT!'. ITT came to prominence in connection with long-term RCTs with mortality as their major end-point. Peto and co-workers laid out the general principles for the design and analysis of such studies in their comprehensive and highly influential publications in this journal (Peto et al., 1976, 1977). They did not use the term 'intention to treat' but did advocate that 'even patients who do not get the proper treatment must not be withdrawn from the analysis'. Their guidance is detailed: the various ways in which patients may fail to complete the study as per protocol are all discussed, together with the possibility that the patients who so fail may differ in characteristics, or numbers, from treatment arm to treatment arm. Such differences are the source of the bias which ITT attempts to avoid any analysis which omits patients is open to bias because it no longer compares the groups as randomised. However, although Peto and coworkers do recommend that all properly randomised patients should be included in the analysis regardless of protocol deviation, they do leave open the possibility for some inappropriately randomised patients to be excluded, and thus stop short of espousing the inclusion of all randomised patients.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/bjc.1993.402

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 1 references

Cited by

See all 235

Metadata

Authors
Lewis, J., & Machin, D.
Venue
British Journal of Cancer
Published
1993-10-01
Volume
68
Issue
4
Pages
647-650
ISSN
  • 0007-0920 (ISSN-L)
  • 1532-1827
OpenAlex
W2044763202
Cited by
235
References
1
Type
journal article
Language
en
Open access
bronze
Licence

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Intention to treat – who should use ITT? https://app.cochat.ai/papers/10.1038/bjc.1993.402
The paper
Lewis, J., & Machin, D. (1993). Intention to treat – who should use ITT? British Journal of Cancer. https://doi.org/10.1038/bjc.1993.402

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.