Intervention and comparator delivery

Item 24a: Intervention and comparator as they were actually administered (eg, where appropriate, who delivered the intervention/comparator, whether participants adhered, whether they were delivered as intended (fidelity))

Examples

“Patients were randomly assigned to the P2Y12 inhibitor monotherapy group (aspirin plus a P2Y12 inhibitor for 3 months and thereafter a P2Y12 inhibitor alone) or to the DAPT group (aspirin plus a P2Y12 inhibitor for at least 12 months) in a 1:1 ratio . . . Overall adherence to the study protocol was 79.3% in the P2Y12 inhibitor monotherapy group and 95.2% in the DAPT group . . . The rates of P2Y12 inhibitor use were similar in both groups: 96.4% at 6 months and 95.0% at 12 months in the P2Y12 inhibitor monotherapy group and 98.1% at 6 months and 96.6% at 12 months in the DAPT group. The median duration of aspirin was 96 days (interquartile range, 88-118 days) in the P2Y12 inhibitor monotherapy group and 365 days (interquartile range, 363-365) in the DAPT group. The proportion of patients receiving aspirin beyond 3 months in the P2Y12 inhibitor monotherapy group was 14.4% at 6 months and 8.9% at 12 months [440]."

“Most participants received treatment as allocated [table 8]. Across intervention groups, high protocol adherence was achieved in terms of the delivery, type, and content for the injection, progressive exercise, and best practice advice interventions. 53 physiotherapists delivered corticosteroid injections to 329 (97%) participants and three doctors to ten (3%) participants. Progressive exercise was delivered by 104 physiotherapists to 339 participants and best practice advice was delivered by 83 physiotherapists to 324 participants. Two physiotherapists swapped groups during the trial because of staffing issues and delivered both interventions. We found no difference in attendance rates between those receiving progressive exercise or best practice advice and those who received the intervention in conjunction with corticosteroid injection [table 8] [114]."

Explanation

This new item has been added to the CONSORT 2025 checklist to address the poor reporting of the intervention and comparator in randomised trials [195, 198-201, 214]. For example, in a review of 102 randomised trials evaluating bariatric surgery, only 14% reported the intervention as implemented [441]. A review of 192 randomised trials assessing pharmacological treatments in six major chronic diseases published in journals with high impact factors showed that adherence to medication was reported in only one third of the publications [442]. A review of 100 randomised trial reports published in general medical journals with high impact factors highlighted that only 11% of the trials assessing long term interventions and 38% of those assessing short term interventions adequately reported treatment initiation and completeness of treatment [209]. A review of 111 randomised trials reports showed that only 46% reported adherence results [443]. A review of 94 placebo/sham controlled randomised trials published in high impact journals showed that only 54% reported actual adherence or fidelity [214].

There is frequently a gap between the intervention/comparator as planned and described in the trial protocol and how the intervention/comparator were actually administered. This gap could be related to poor fidelity, which can be driven by the extent to which the intervention/comparator are implemented as planned in the protocol by practitioners, and/or poor adherence to treatment, defined as the extent to which participants comply with the care providers’ recommendations (eg, taking a drug, placebo, behavioural change, doing exercises) [209, 444] This gap could also be related to the expected diversity in the implementation of the intervention/comparator in clinical practice particularly for complex interventions.

The gap between the intervention/comparator as planned and as delivered also depends on how the trial was planned. In explanatory trials, the aim is to estimate treatment effect under ideal circumstances. The intervention/comparator are usually highly standardised with close monitoring of fidelity and adherence to interventions and strategies to increase them. Intensive efforts to maximise fidelity and adherence in early phase trials or explanatory trials may lead to unrealistic, inflated estimates of treatment benefit that cannot be reproduced under real life circumstances [445, 446]. Reporting the results of this monitoring is essential to allow readers to interpret the study results.

In contrast, pragmatic trials aim to determine treatment effect in clinical conditions. The intervention and comparator are usually highly flexible, and measurement of fidelity and adherence is unobstructive with no strategies to maintain or improve them.

Reporting how the intervention/comparator were actually administered is nevertheless crucial to allow readers to accurately interpret the trial results. For example, in a large international randomised trial comparing endarterectomy to medical management for patients with symptomatic carotid artery stenosis, there were important differences in the delay in receiving the surgical procedure which impacted the outcomes [179].

Authors should provide details on who actually delivered the intervention/comparator (number and expertise), how the intervention/comparator were delivered, what was actually administered, participants’ adherence to treatment, and the caregiver’s fidelity to the intervention/comparator protocol where appropriate. Reporting fidelity and adherence can be complex and vary according to the type of intervention or comparator (eg, one-off, short term repeated, long term repeated). Various deviations to the protocol can occur. Participants might initiate the intervention/comparator but then discontinue the intervention/comparator permanently and completely after a specific period of time, discontinue temporarily, reduce the dose, or modify the schedule.

More detailed information is available in TIDieR23 and the CONSORT extension for non-pharmacological treatments [22] (item 13).