Is the TIME study on antihypertensive therapy and chronotropic effects on cardiovascular outcomes a pragmatic trial?

Pragmatic trials are those aiming to help prescribers and decision-makers in choosing between interventions (or dosing schedules) and that are conducted mimicking usual clinical practice; conversely, explanatory trials are conducted in ideal conditions aiming to assess causal mechanisms of action.¹ A typical explanatory trial is a Phase 3 double-blind randomized controlled trial (RCT) assessing the efficacy of a new drug vs. placebo. A comparative effectiveness RCT conducted on marketed drugs prescribed following the summary of product characteristics and managed as normal clinical practice is a typical pragmatic trial.

The term pragmatic is increasingly used to describe RCTs. Pragmatic conveys the message that the trial provides robust external validity: The results are generalizable to other populations and settings. However, many investigators who have streamlined one or a few trial processes are erroneously labelling their RCTs as pragmatic.² Furthermore, recent research has shown that in only 68% of 415 RCTs, the label pragmatic was justified to at least one design trial element.³

The PRECIS-2 tool was designed to assess if the investigators’ intention is translated into appropriate trial design features.⁴ PRECIS-2 comprises the evaluation of nine trial domains. The assessment of each domain is scored between 1 (very explanatory) and 5 (very pragmatic). There is a continuum between the two extremes, the explanatory and the pragmatic extremes.⁴ Almost all trials have pragmatic and explanatory features. The objective is to know how many of these are pragmatic and how many are explanatory, so that investigators at the trial design stage can make appropriate decisions to tailor the design to their intent. PRECIS-2 is also used to retrospectively assess the pragmatic/explanatory characteristics of RCTs described in published reports by researchers that were not involved in them. Yet, articles describing trial results provide insufficient information to conduct a throughout assessment: All other available sources of information should be used in the retrospective assessment.

The TIME study is a prospective, parallel-group, open-label, blinded-endpoint, RCT conducted in 21 104 UK patients that showed that evening dosing of usual antihypertensive medications was not different from morning dosing in terms of a composite endpoint (vascular death or hospitalization for non-fatal myocardial infarction or non-fatal stroke).⁵ In the article where the results were reported, the investigators described it as pragmatic. But can it really be labelled as such? This was a decentralized 5-year RCT, started in 2011, which did not require for in-person study visits: Consent, randomization, and follow-up were done through an online study portal and by email.⁵ Many readers might think that all these features are far from usual clinical practice. A thorough assessment of the information available will shed light on whether the TIME study could be reasonably labelled as pragmatic.

After checking that the term pragmatic was not included in the likely last protocol version (v12.0, 10 February 2021) provided by the investigators with the trial report,⁵ or in two reports describing the methods of the trial⁶ and the analysis of recruitment, follow-up and retention rates post-recruitment,⁷ or in the register (ISRCTN18157641), we should conclude that the term pragmatic seems to have been added at the very last minute in the article.⁵

The scores given––and the reasons supporting them––to the nine PRECIS-2 domains after checking the information provided by the three published reports,^5–7 are shown in the supplementary material. There are only two domains (primary outcome and primary analysis) that are clearly pragmatic. Conversely, six [eligibility, recruitment, setting, flexibility (delivery and adherence), and follow-up] are clearly in the explanatory side of the continuum. Of critical importance were the recruitment strategies used to reach potential participants, far from usual clinical practice. Among these, it should be noted that GP practices sent 594 289 letters to patients on antihypertensive therapy,⁷ but only 21 104 (3.6%) technologically literate were included in the primary analysis.⁵ It is also very relevant the need of regular completion of online follow-up questionnaires, and the reminder follow-up emails in case of no answer––something unusual in clinical practice. These and other trial features call into question the pragmatism of the TIME study.

We conclude that although the TIME study provides very useful chronotropic information on the clinical effects of antihypertensive medications on major cardiovascular outcomes, it should probably not be labelled as pragmatic. The use of this term by the TIME investigators makes it (erroneously) used by other authors,⁸ which favours its original meaning¹ to be lost and gives reasons to those who believe that the use of this term has become uninformative.³

Funding

Not required.

Conflict of interest: None declare.

Data availability statement

No new data were generated or analysed in support of this research.

References