Skip to main content

Table 5 Assessment of the TIME trial using the PRECIS-2 tool

From: A mixed methods case study investigating how randomised controlled trials (RCTs) are reported, understood and interpreted in practice

DomainRatingSupport for judgement
Eligibility criteria3Included oesophageal and Siewert 1 tumours; pre-operative T1–3, N0–1; aged 18–75 years. Therefore, excluded Siewert 2, pre-operative T4 and N2–3, those aged > 75 years.
Recruitment path5‘The patient will be informed about the trial at the outpatient clinic.’
Setting4International study across 5 centres in 3 countries in Europe, including academic and non-academic units. Units performing > 20 oesophagectomies per year.
Organisation3No extra staffing. Surgeons in some centres had been proctored by lead centre, but not as part of the trial. Surgeons from these centres submitted videos to judge their experience and skill to be allowed to participate. Other centres ‘already well experienced’. Surgeons required a minimum of 10 MIO to participate.
Flex of experimental intervention - delivery3Operation was specified in protocol, but not exhaustively. For either open or MIO, the operation could be 2 or 3 stage. There was no monitoring to ensure compliance. Positioning and aspects of anaesthesia were specified, including left decubitus position and right lung block for open thoracotomy, with prone positioning and no lung block for MIO.
Flex of experimental intervention - adherenceAs per PRECIS-2 toolkit, not applicable when patients undergoing surgery.
Follow-up5Follow-up as per usual practice, at 6 weeks, 3 months, 6 months and 1 year, until 5 years postop.
Outcome3Primary outcome outlined in the protocol included all pulmonary infections, with 3 categories discussed. However, the trial report only presents one of these 3 categories, pneumonia (CXR/CT changes and positive sputum culture). No patient-centred outcomes, such as Patient Reported Outcome Measures. Outcomes were measured early in the patient’s recovery from the procedure.
Analysis5Intention-to-treat analysis of all randomised patients with no apparent loss to follow-up.
Median3.5 
\