Clarify that an intention-to-treatment analysis has been performed that explicitly describes how to address deviations from randomized allocation and no response While an ITT analysis aims to preserve initial randomization and avoid potential bias due to patient exclusion, the objective of a pro-protocol (PP) analysis is to identify a therapeutic effect that is suboptimal conditions would occur; that is, to answer the question: What is the effect when patients are fully compliant? Therefore, some patients (from the full set of analyses) should be excluded from the population (PP population) used for PP analysis. The ITT analysis requires that participants be involved, even if they have not fully adhered to the protocol. Participants who deviated from protocol (e.B. in case of non-compliance with the prescribed intervention or discontinuation of active treatment), should continue to be taken into account in the analysis. An extreme variation of this is participants who receive treatment from the group to which they were not assigned, who should be kept in their original group for analysis. This is not a problem, provided that you, as a systematic reviewer, can extract the appropriate data from the study reports. The rationale for this approach is that we primarily want to assess the impact of assigning an intervention in practice, not the impact in the subset of participants who adhere to it. About half of all published reports of randomised controlled trials indicated that treatment intent was used, but the treatment of deviations from randomised allocation varied considerably. Analysis of the data after a pro-protocol analysis would lead an investigator (or consumer of the medical literature) to wrongly conclude that the intervention (medical management + surgery) reduces the risk of death by 41% compared to conventional treatment (medical management) alone.

However, as already mentioned, we know that an operation in this example has absolutely no effect on the result (truth). This method of analysis would lead to a gross misinterpretation and an inaccurate (distorted) assessment of the effectiveness of the intervention. RCTs are the ideal design to assess the efficacy and safety of drugs. In an RCT, study participants are randomly assigned to one of the treatments studied after assessing relevance, but before performing the intervention. Randomization in clinical trials reduces bias. The RCT aims to ensure that groups differ only from the interventions to be compared. [8] In a study of endometrial resection or hysterectomy in menorrhagia, the authors excluded 26 (13%) women who withdrew after randomization but before the treatment surgery from the intention to treat analysis.22 The researchers contacted 10 of these women and found that „out of six who had received an endometrial resection, four had had a hysterectomy and two had a resection, while three out of four assigned hysterectomies chose endometrial resection and hysterectomy. In a study on folic acid supplementation, 17 (14%) women were excluded for non-compliance.23 The objective of this study was to predict the likely effect of food fortification that would not offer the same risk of non-compliance as tablet supplementation.

Therefore, the exclusion of women who did not comply was appropriate, but it should not have been described as an intention to process the analysis. Randomised controlled trials often suffer from two main complications, namely non-compliance and lack of results. One possible solution to this problem is a statistical concept called intention-to-treat analysis (ITT). The ITT analysis includes all subjects who are randomized after a randomized treatment assignment. It ignores non-compliance, protocol deviations, withdrawal and everything that happens after randomization. ITT analysis maintains the prognostic balance generated by the original random processing assignment. In ITT analysis, the estimate of the treatment effect is usually conservative. Better application of the ITT approach is possible when complete outcome data are available for all randomized subjects.

The protocol population is defined as a subset of the ITT population that completed the study without major protocol violations. A practical problem that investigators usually encounter at the RCT is that subjects do not always follow the instructions. In addition, in some studies, stopping subjects is a problem. Therefore, the RCT often suffers from two major complications, namely non-compliance and lack of results. One possible solution to this problem is a statistical concept called ITT analysis. The ITT analysis includes all subjects who are randomized after a randomized treatment assignment. It ignores non-compliance, protocol deviations, withdrawal and everything that happens after randomization. ITT analysis is generally described as „once randomized, always analyzed.“ However, in the itT analysis, the estimate of the treatment effect is generally conservative due to dilution due to non-compliance. Heterogeneity could also be introduced if non-compliant subjects, dropouts and compliant subjects are finally mixed. In addition, outcome data will differ significantly between non-conforming, discontinuous, and compliant subjects, and interpretation could become difficult if a large proportion of participants moved to opposite treatment arms. Better application of the ITT approach is possible when complete outcome data are available for all randomized subjects.

Always take care to minimize missing responses and continue to follow those who withdraw from treatment. Anyone who intelligently follows these principles and has a vision to minimize bias should stop worrying about „dealing with intent.“ Data on all patients randomly assigned. were analysed on the basis of intention to treat. This analysis did not take into account patients without hernias, those who withdrew their consent before surgery, those who turned out to be poor candidates for general anesthesia at the time of surgery, and those who did not undergo assigned surgery due to a misunderstanding that led to unplanned open or laparoscopic repair.21 Conclusions: The intention-to-treatment approach is often seen as insufficiently described and insufficiently applied. Authors should explicitly describe how to address deviations from randomized allocation and missing responses, and discuss the potential effects of a missing response. Readers should critically assess the validity of the reported treatment intent. One of the alternatives to ITT analysis is PP analysis. It is defined as a subset of the ITT population that completed the study without major protocol violations. [26] PP analyses exclude all protocol violators, including those who did not adhere to treatment, switch groups, or missed measurements. [27] TTI tends to make the two treatments similar, while PP eliminates patients who do not complete treatment and is better able to reflect differences in treatment.

[28] Single imputation approaches [e.B. Last Observation Carried Forward (LOCF), Best Case, Worst Case] generate a single complete set of data. LocF can be used in longitudinal studies; The most recently observed result data is transferred to replace the missing result. LOCF is not suitable for degenerative diseases where the outcome state is expected to decrease steadily over time. LOCF imputation may be more reliable at the extremes of the distribution of results. A subject who died after 3 months will, of course, also remain at 6 months, and a subject who has reached a GOSE of 3 months of good recovery will probably remain so even at 6 months. However, in the middle of the distribution, the 3 months between earnings evaluations can leave considerable room for improvement in the income statements. Alternatively, the best- and worst-case approaches assume missing outcomes with the best and worst possible outcomes, respectively. A combination of these elements is also possible, assuming the best possible outcome for subjects in the control group and the worst possible outcome for subjects in the treatment group, resulting in a conservative estimate of treatment effectiveness. Individual imputation approaches ignore the uncertainty associated with the imputation procedure, which leads to downward distorted P-values, i.e.

too often reject the hypothesis of a missing treatment effect.48 Thus, a high treatment effect leads to a successful study (i.e. proven efficacy). However, if you choose an overly optimistic method of analysis, that is. . .