Intention-to-Treat Analysis: Implications for Quantitative and Qualitative Research

Abstract

Intention-to-treat analysis is an important aspect of randomized controlled trials of health care interventions. The concept is now widely accepted in theory, but not always implemented in practice. Failure to analyse by intention-to-treat can give misleading and indeed life-threatening interpretations. In some studies, a case is put for estimating the effect that would have been observed if all patients had received the allocated treatment. Situations where this is valid are rare, but an example is given of such an exceptional study. The relevance of the intention-to-treat concept is not always taken into account in qualitative research. Interviews with new mothers who delivered their babies at home in a hypothetical controlled trial of home versus hospital confinement would provide fascinating accounts of the pleasure of successful delivery at home. But by definition the interviews would exclude the hazard and drama of necessary transfers to hospital due to complications in late pregnancy and early labour. The intention-to-treat approach would avoid this bias.

Comparable non-relapse mortality and survival after HLA-identical sibling blood stem cell transplantation with reduced or conventional-intensity preparative regimens for high-risk myelodysplasia or acute myeloid leukemia in first remission

“…19,21 Results All patients had initial donor-derived hematological recovery. The number of days with neutrophils below 0.5 Â 10 9 /l and platelets below 20 Â 10 9 /l were shorter in the FB-RIC with respect to the CTCD34 þ group (15 vs 8 days for neutrophils and 14 vs 4 days for platelets, respectively; Po0.05 for both comparisons).…”

“…In the context of a single-center consecutive patient cohort, 'age randomization' can be a suitable way of comparing two transplant strategies, while eliminating potentially unknown selection biases. 21 In addition, we analyzed survival according to the ITT principle to avoid misleading interpretations and biased treatment effects, 19 since reasonable doubts have arisen as to the true impact of RIC allo-HSCT to elderly patients due to the various biases that lead to selection of only very fit older patients. [23][24][25] The similar 4-year OS and DFS in this ITT analysis suggest that RIC alloHSCT may be proposed for elderly patients with AML/ MDS.…”

“…Young patients (those with 50 years of age or younger) were conditioned with cyclophosphamide plus 19 ITT ¼ intention to treat; m-sGVHD ¼ moderate-to-severe GVHD; MDS ¼ myelodysplastic syndrome; NT ¼ not tested; Re ¼ recipient; UF ¼ unfractionated. a Refers to all patients who had an HLA-identical sibling and were assigned to receive one of the two transplant strategies according to their age (i.e., intention-to-treat analysis).…”

The role of reduced intensity conditioning allogeneic stem cell transplantation in patients with acute myeloid leukemia: a donor vs no donor comparison

“…21 As we did in this analysis, one important aspect is to analyze the results according to the 'intention-to-treat' principle in order to avoid misleading interpretations and biased treatment effects. 22 Indeed, lack of compliance with the assigned treatment reflects one of the obstacles intrinsic to this type of analysis. In our study, only 71% of patients with a matched sibling donor actually received the RIC-allo-SCT, in accordance with previous data from the myeloablative allo-SCT setting.…”

Statistical Models for Causation

“…Heckman (2000) considers the role of potential responses in economics, and the limitations of statistical models for causation. The intention-to-treat principle goes back to Bradford Hill (1961, 259); for additional discussion, see Newell (1992).…”