High-Impact RCTs without Prospective Informed Consent: A Systematic Review

Abstract

The prevalence of randomized controlled trials (RCTs) performed without fully informed prospective consent from subjects is unknown. We performed this study to estimate the prevalence of high-impact RCTs performed without informed consent from all subjects and examine whether such trials are becoming more prevalent. We performed a systematic review of English-language RCTs published from 2014 through 2018 identified in Scopus and sorted to identify the top 100 most highly cited RCTs each year. Text search of title and abstract included terms randomized controlled or clinical trial and spelling variants thereof, and excluded metaanalyses and systematic reviews. We independently identified the most highly cited RCTs based on predefined criteria and negotiated to agreement, then independently performed keyword searches, read, abstracted and coded information regarding informed consent from each paper and again negotiated to agreement. No quality indicators were assessed. We planned descriptive qualitative analysis and appropriate quantitative analysis to examine the prevalence and characteristics of trials enrolling subjects with other than fully informed prospective consent. We find that 44 (8.8%, binomial exact 95% CI 6.5% to 11.6%) of 500 high-impact RCTs did not secure informed consent from at least some subjects. The prevalence of such trials did not change over the 5 years (OR=1.09, z=0.78, p=0.44). A majority (66%) of the trials involved emergency situations, and 40 of 44 (90.9%) of the trials involved emergency interventions, pragmatic designs, were cluster randomized, or a combination of these factors. A qualitative analysis explores the methods of and justifications for waiving informed consent in our sample of RCTs.

Keywords

biomedical research emergency service hospital public policy research research design

Introduction

Securing informed consent for clinical trials is an international ethical requirement. Nonetheless, ethical codes such as the Declaration of Helsinki1 and the Council for International Organizations of Medical Sciences Guidelines,2 as well as laws of many countries, recognize circumstances in which prospective informed consent may be unnecessary. The standards justifying the waiver of the general requirement for informed consent vary widely, as do the rules for carrying out waivers.3 Typically, exceptions have been recognized for research in emergency situations where patients are at risk of having diminished capacity and time may be of the essence. One systematic review examined trials in acute stroke, reporting waivers of written informed consent in 9 of 36 trials (25%).4

Other justifications for waiving consent typically focus on very low levels of risk and the impracticability of securing consent (or the impracticability of carrying out the research if consent is required).5 6 For example, it is often argued that research using cluster randomization makes consent difficult because, depending on the specific design, there may be little interaction between subjects and researchers, and subjects may have limited choice or ability to refuse because the intervention is determined for them by their inclusion in the cluster.7 A recent review of a random sample of 39 randomized controlled trials (RCTs) conducted across the world using cluster randomization testing individual-level healthcare interventions found that more than 20% of the trials waived individual informed consent.8

Most recently, numerous bioethicists and researchers have been advocating the concept of a learning healthcare system that would strive for continuous improvement in the delivery of care.9 10 This strategy relies on greater use of pragmatic and comparative effectiveness trials, which attempt to embed trial procedures as much as possible into routine clinical practice and often are focused on resolving latent uncertainties about optimal care. Many authors have proposed ways for streamlining or simplifying the processes of informed consent,11 -17 and in the limit, propose waiving consent from subjects.18 A recent systematic review examined pragmatic and comparative effectiveness RCTs performed at least in part in the USA and first published in 2014 and 2017, finding that 23 (22%) of 103 trials were done with a waiver of individual informed consent (LY Lin, N Jochym and JF Merz, unpublished data, 2020).

While the rates of waiving consent in these reviews seem high, there are no data for comparison. Thus, we sought to identify a background rate of published, high-impact RCTs that do not secure prospective fully informed consent from patients or subjects (or their family members or legally authorized representatives (LAR)), to examine the prevalence of trials involving emergency research, pragmatic or comparative effectiveness designs, or cluster randomization, and to describe the range of methods used for enrolling subjects without consent and the justifications for not securing consent. We focused our inquiry on highly cited RCTs, conceiving that the prevalence of foregoing or modifying informed consent in important, well-designed, thoroughly performed, and influential research, if not representative of the greater universe of clinical trials, may reflect state-of-the-art methods and presage or foretell where trial methodology may be heading.

Materials and methods

We performed a systematic review of RCTs focusing on the top 100 most cited trials published each year from 2014 through 2018. We performed a search of the literature using Scopus (https://www.scopus.com/). Scopus was chosen because it enables sorting of publications by number of citations. We searched titles and abstracts for ‘randomized controlled trial’, ‘randomized clinical trial’, ‘randomised controlled trial’ and ‘randomised clinical trial’ and excluded ‘systematic review’, ‘metaanalysis’ and ‘meta-analysis’ to identify all RCTs published in English-language journals (see the online supplemental materials). We constrained results by publication year (2014, 2015, 2016, 2017, 2018) and then sorted annual results by number of citations received. We narrowly defined our search for RCTs to minimize the burden of manual selection, believing that papers not using the specific language would be less likely to be RCT reports. The abstracts of the top 500 trials from each year were exported into Excel.

Abstracts were reviewed independently by the 2 authors (RD, JFM). We included all RCTs in which people were subjected to randomly assigned interventions for evaluation of their effects on biomedical or behavioral outcomes. Trials were excluded from the sample if they were: (1) uncontrolled or non-randomized trials; (2) retrospective studies; (3) consensus/committee statements, guidelines, commentary, editorials, and letters; (4) secondary analyses or long-term follow-ups of RCT data, including embedded non-randomized studies of RCT subjects or subgroup analyses of RCTs; (5) reviews, including systematic reviews and meta-analyses; (6) conceptual or methods papers; (7) non-health papers, including those with no health outcomes, such as educational interventions carried out in schools; (8) interventions with healthcare providers that did not assess patient outcomes; (9) pilot, feasibility, explorative, and proof-of-concept trials, as self-characterized by authors; (10) unpublished reports and conference abstracts that were not later peer reviewed and published; or (11) protocol, rationale, and designs of trials. Each author identified between 110 and 120 RCTs, and the authors negotiated to agreement to identify the top 100 most cited RCTs from each year for full-text review.

Both authors independently performed keyword searches to locate relevant language (eg, consent, inform, written, waiver, ethic, brief, retrospective consent, delayed consent, deferred consent, Zelen,19 cluster randomized, exception from informed consent (EFIC), and mental health act), examined every paper, and summarized approaches to informed consent for the full set of 500 trials. Geographic location of trial sites was coded for all trials into the following categories: USA, North America (but not USA), Central and South America, Europe, Africa, Asia, and Australasia. Protocols and supplemental materials were examined as necessary to understand the role of consent in each of the trials. Differences were negotiated to agreement.

We identified trials for which fully informed, prospective consent was not secured from all patients or subjects. This includes trials in which prior consent is not secured, as well as trials in which prior consent is secured but any information that would normally be disclosed is modified or withheld, including studies involving deception or those using limited disclosures. We did not code oral consent processes as waivers unless the information disclosed was stated by authors to be abbreviated or brief. That is, we did not consider waivers of written documentation of informed consent to be waivers of informed consent. Further, we do not consider trials proceeding with prospective parental consent for children or surrogate consent from incapacitated patients’ LAR or family members to involve waivers.

Our intent is to describe the journal and year of publication, the locations of study sites, the populations studied, trial methods including pragmatic, comparative effectiveness, and cluster randomized, the disease or condition of interest, the intervention, a description of the approach to informed consent, and justifications for not securing prospective consent. Appropriate descriptive statistics were planned, including contingency tables and tests of proportions. All analyses were run in Stata V.12.1 (2012; StataCorp, College Station, Texas).

Authors were emailed to obtain clarifications or additional information for all trials in which consent processes were unclear. Querying authors for details about their published trials is considered non-human subjects research by the University of Pennsylvania Institutional Review Board. This systematic review was done without patient or public involvement.

Results

We performed the Scopus search on October 29, 2019. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses diagram is shown in figure 1.20 Our initial literature search identified a total of 70,458 papers. We sorted results in each year by the number of citations each paper had received, then reviewed 998 titles and abstracts to cull out at least 110 RCTs for each year, ensuring that we would agree on a set of 100 for each year. Overall, the authors were highly concordant in title and abstract review (94% agreement, kappa=0.90), with very high agreement identifying RCTs (97% agreement, kappa=0.95). We attempted to contact 34 authors for clarification and further details about their trials, of whom 26 (76%) replied.

Figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram. RCTs, randomized controlled trials.

Our review of all 500 papers yielded 44 trials (8.8%, binomial exact 95% CI 6.5% to 11.6%) in which prospective informed consent was not sought from at least some subjects. There were 8 such trials from 2014, 9 from 2015, 9 from 2016, 4 from 2017, and 14 from 2018. We ran a logistic regression to look for trend in likelihood of waiver by year, which shows no evidence of trend in our sample (OR=1.09, z=0.78, p=0.44). A complete list of all 500 RCTs is available from the authors.

We coded studies if the authors characterized the study as pragmatic trials (or involving pragmatic methods) or comparative effectiveness trials and if the trials used cluster randomization. For trials that used waivers, we coded whether authors noted emergent circumstances (eg, need for rapid treatment, emergency, and so on) as justifications for using waivers (see the online supplemental materials).

There were 52 author-described pragmatic trials or trials using pragmatic methods, 15 of which did not secure prospective consent from subjects (28.8%, binomial exact 95% CI 17.1% to 43.1%). There were 7 author-described comparative effectiveness trials, including 1 also described as pragmatic, and all secured prospective consent. Pragmatic trials in which fully informed consent was not secured are summarized in table 1.

Table 1

Pragmatic trials that did not secure prospective informed consent from all subjects

Study	Description of RCT
IMPROVE Trial Investigators27	IMPROVE RCT—pragmatic, multicenter (UK and Canada) RCT of endovascular versus open repair of ruptured abdominal aortic or aortoiliac aneurysm; brief written or verbal consent or waiver, with reconsent in recovery period.
Shahzad et al28	HEAT-PPCI trial—pragmatic, open-label, single-center (Liverpool, England) RCT of heparin versus bivalirudin in patients undergoing emergency angiography; delayed consent; deferred consent criticized in Dickert et al.29
Holcomb et al22	PROPPR trial—pragmatic, multicenter (USA and Canada) RCT comparing 2 standard transfusion ratio interventions (1:1:1 vs 1:1:2 plasma/platelets/red cells) in patients admitted to emergency department with massive truncal hemorrhage; EFIC waiver.
Perkins et al30	PARAMEDIC trial—a pragmatic, cluster-randomized RCT (4 ambulance services in the UK) comparing manual cardiopulmonary resuscitation (CPR) to LUCAS-2 mechanical CPR in adults with non-traumatic cardiac arrest; informed consent waived, with consent for 3 and 12 mo follow-up.
Stub et al31	AVOID study—pragmatic, multicenter (Melbourne, Australia) RCT comparing supplemental oxygen therapy with no oxygen therapy in normoxic patients with ST-segment elevation myocardial infarction; delayed consent (brief information and opportunity to opt out) with consent sought after stabilization.
Gonzalez et al32	Pragmatic, single-center (Denver, Colorado, USA) RCT examining massive transfusion protocol (MTP) goal directed by the viscoelastic assay thrombelastography compared with an MTP guided by conventional coagulation assays in trauma victims; EFIC waiver.
Heddle et al33	INFORM trial—pragmatic multinational (Canada, USA, Australia, and Israel) RCT examining the difference in mortality between hospitalized patients who required a red cell transfusion and were given the freshest stored cells versus the oldest available cells; informed consent waived.
Hofmann et al34 35	DETO2X-AMI trial—pragmatic, multicenter (35 acute cardiac care clinics in Sweden) RCT compared routine supplemental oxygen therapy versus ambient air in patients with suspected myocardial infarction; abbreviated oral consent from subjects followed by written consent within 24 h.
Muir et al36	PISTE RCT—pragmatic, multicenter (UK) RCT comparing intravenous thrombolysis (IVT) alone with IVT and adjunctive intra-arterial mechanical thrombectomy in acute ischemic stroke; consent secured from patient or surrogate if possible, with option in England and Wales for consent by ‘nominated consultee’ (under Mental Capacity Act), with follow-up consent from patient or family.
Benger et al37 38	AIRWAYS-2 RCT—pragmatic, cluster-randomized (paramedics from 4 EMS services in England) trial comparing a supraglottic airway device versus tracheal intubation for initial advanced airway management strategy in treating non-traumatic cardiac arrest; waiver of informed consent, followed by written consent of patient or surrogate for follow-up.
Franklin et al39 40	PARIS study—pragmatic, multicenter (Australia and New Zealand) RCT comparing respiratory support and oxygen delivery via high-flow nasal cannula versus standard subnasal oxygen therapy in infants less than 1 y of age admitted to hospital with bronchiolitis; 16 of 17 sites waived informed consent (deferred consent), and 1 site required prospective parental consent.
Lamb et al41	DAPA trial—pragmatic, multicenter (England) RCT comparing a 4 mo exercise training program to usual care in people with mild to moderate dementia; consent from patients and family secured if possible, or ‘nominated consultee’ under Mental Capacity Act with follow-up consent from patient or family.
Shah et al42	High-STEACS trial—pragmatic, stepped-wedge, cluster-randomized (10 hospitals in Scotland) RCT examining whether the introduction of a high-sensitivity cardiac troponin I assay with a sex-specific 99th centile diagnostic threshold would reduce subsequent myocardial infarction or cardiovascular death in patients with suspected acute coronary syndrome; waiver of informed consent.
Steinhubl et al43	mSToPS trial—pragmatic RCT (members of large national health plan in the USA) comparing a self-applied continuous ECG monitoring patch used at home during routine activities for up to 4 wk, initiated either immediately after enrolling or delayed for 4 mo after enrollment to a matched observational cohort (control) in a population at high risk of undiagnosed atrial fibrillation drawn from commercial and Medicare Advantage members of a large health insurer; digital consent from randomized subjects, and no consent from controls.
Wang et al44	A pragmatic, multicenter, cluster-randomized (27 EMS agencies in the USA) RCT comparing laryngeal tube insertion versus endotracheal intubation in adults with out-of-hospital cardiac arrest; EFIC waiver.

EFIC, exception from informed consent; EMS, Emergency Medical Services; RCT, randomized controlled trial.

Cluster randomization was used in 27 trials in the sample of 500, 11 of which did not secure consent from subjects (40.7%, binomial exact 95% CI 22.4% to 61.2%). Thirteen of the 27 cluster-randomized trials were published in 2018, suggesting that these methods might be becoming more popular. Notably, 6 of the trials completely waived consent, with no attempt to provide notice before, during, or after participation. Six self-described pragmatic trials used cluster randomization, 4 of which (66.7%, binomial exact 95% CI 22.3% to 95.7%) did not secure prospective consent from subjects. Non-pragmatic trials using cluster randomization are summarized in table 2.

Table 2

Non-pragmatic trials that used cluster randomization

Study	Description of RCT
Zimmermann et al45	Cluster-randomized, single cancer center (Toronto, CA) RCT comparing consultation and follow-up (at least monthly) by a palliative care team to standard cancer care in patients with advanced cancers; subjects consented to take part in their assigned group without being informed of the comparator arm of the trial.
Garg et al46	Cluster-randomized RCT (8 urban community health centers in Boston, MA) testing the effect of a clinic-based screening and referral system (WE CARE) in the first year of a child's life; parents approached for consent before research procedures, consenting to the arm of the trial in which their center was assigned.
Young et al47	SPLIT RCT—cluster-randomized, multicenter (New Zealand) trial comparing buffered crystalloid to saline on renal complications in patients admitted to the intensive care unit (ICU); adopted an ‘opt-out consent’ process allowing patients or families to refuse use of their data.
Brewer et al24	Cluster-randomized (30 pediatric clinics in North Carolina, USA) RCT examining effect of clinician training for vaccine announcement versus conversation versus usual care on rates of HPV vaccination among adolescent patients, assessing aggregated outcomes collected by state vaccine registry; no consent.
Dempsey et al48	Multicenter, cluster-randomized RCT (16 primary care practices in Denver, Colorado, USA) compared a 5-component healthcare professional HPV vaccine communication intervention versus usual care (control) on rates of adolescent HPV vaccination; waiver of informed consent.
Martin et al49 50	CAP trial—cluster-randomized RCT (primary care practices in England and Wales) compared a low-intensity, single invitation PSA test and standardized diagnostic pathway to recommended NHS care (control); men in the intervention group consented to PSA testing and study participation, while men in control arm could opt out in response to notices posted in clinics.
Parshuram et al26 51	EPOCH RCT—cluster-randomized RCT (Canada) comparing the effect of the Bedside Paediatric Early Warning System (BedsidePEWS) with standard care (no severity of illness score) in pediatric ICUs; waiver of patient and clinician consent.

HPV, human papillomavirus; NHS, National Health Service; PSA, prostate-specific antigen; RCT, randomized controlled trial.

For the set of 427 trials that were neither pragmatic nor used cluster randomization, 22 (5.2%, binomial exact 95% CI 3.2% to 7.7%) did not secure fully informed consent from all subjects, and 18 of these involved emergency interventions. These trials are summarized in table 3. Of the 44 trials using waivers, 29 (66%) involved emergency circumstances, 11 of which were characterized as pragmatic trials, including 3 that were cluster randomized. Eighteen of the 29 were neither pragmatic nor cluster randomized. Overall, 40 of 44 (90.9%) RCTs in which fully informed prospective consent was not secured from all subjects were pragmatic, cluster-randomized, emergency research, or some combination of these.

Table 3

Trials that were neither pragmatic nor used cluster randomization

Study	Description of RCT
Ebinger et al52	PHANTOM-S RCT—multicenter (Berlin, Germany) trial comparing prehospital initiation of thrombolysis to usual care in patients who had an ischemic stroke; waiver for patients unable to consent, and informed consent secured for follow-up data collection.
Kim et al23	Multicenter (Washington State, USA) RCT comparing standard treatment with or without prehospital-induced hypothermia in patients with cardiac arrest; EFIC waiver.
Robertson et al53 54	Epo Severe TBI trial—multicenter (Houston, Texas, USA) RCT examining erythropoietin and transfusion thresholds in patients with traumatic brain injury; surrogate consent required for 20 mo, and an EFIC waiver was secured for 50 mo.
Rubertsson et al55	LINC study—multicenter (Sweden, The Netherlands, and Britain) RCT comparing mechanical chest compressions with defibrillation versus manual cardiopulmonary resuscitation in patients with cardiac arrest; waiver of consent for intervention, patient or surrogate consent for follow-up.
Skolnick et al56	SYNAPSE trial—multinational phase III, placebo-controlled RCT examining progesterone in patients with traumatic brain injury; consent secured from LAR with follow-up from subjects who regain capacity, and consent from independent physician where legally permissible.
Banerjee et al57	Single-center (Rochester, Minnesota, USA) RCT examining rapid multiplex PCR (rmPCR) detection of bacteria, fungi, and resistance genes with templated reporting of results in a population of children and adults; informed consent waived.
Ringh et al58	Community-based (Stockholm, Sweden) RCT examining efficacy of use of phone-based positioning system to identify and contact trained volunteers located within 500 m of suspected out-of-hospital patients with cardiac arrest; informed consent waived.
Steenbergen et al59	A placebo-controlled RCT in Leiden, Netherlands, studying a multispecies probiotic in healthy young students; subjects were not informed about the use of placebo.
Stéphan et al60	biPAP study—multicenter (6 ICUs in France) RCT comparing high-flow nasal oxygen therapy delivered continuously through a nasal cannula versus bilevel positive airway pressure in surgical patients who were at risk for acute respiratory failure; consent waived, and written and oral information provided to patient or family.
Fivez et al61 62	PEPaNIC trial—multinational (Belgium, The Netherlands, Canada) RCT comparing early versus late initiation of parenteral nutrition in critically ill children admitted to the PICU; consent from parents or guardian if admission planned, and deferred consent used if unplanned.
Girardis et al63	Oxygen-ICU trial—a single-center (Modena, Italy) RCT of conservative arterial oxygenation SpO₂ target (94%–98%) versus usual care (97%–100%) in population of patients admitted to ICU with expected length of stay of 72 h or more; informed consent from patient or legal surrogate or use of deferred consent.
Gordon et al64	VANISH trial—multicenter (18 adult ICUs in the UK) RCT comparing early vasopressin use to norepinephrine in patients suffering septic shock; waiver of consent, though prospective consent sought from legal representative if available.
Goyal et al65	SWIFT PRIME trial—multinational RCT (USA, Germany, France, Switzerland, Austria, Denmark, Spain) studying the benefit of Solitaire stent retriever thrombectomy added to intravenous tissue plasminogen activator (tPA) administration in patients who had an acute stroke; some sites secured consent while others used some form of waiver.
Reade et al66	DahLIA study—multicenter (Australia and New Zealand) RCT of dexmedetomidine versus saline for sedation of intubated ICU patients with agitated delirium; use of surrogate or retrospective consent or physician ‘best interest’ decision.
Schönenberger et al67	SIESTA trial—single-center (Heidelberg, Germany) RCT comparing conscious sedation without intubation versus general anesthesia with intubation during thrombectomy in patients with acute ischemic stroke; consent or surrogate consent sought if possible, or deferred consent (within 72 h of surgery).
Zarbock et al68	ELAIN trial—single-center (Muenster, Germany) RCT examining early initiation of renal replacement therapy in critically ill patients with acute kidney injury; consent from patient or LAR or deferred consent (approval by independent consultant physician) used in cases of emergency.
Lascarrou et al69	MACMAN trial—multicenter (7 ICUs in France) RCT compared use of video laryngoscope to the Macintosh direct laryngoscope for orotracheal intubation of patients in the ICU; consent waived if patient was incapable and no next of kin available, with consent requested if patient regained capacity.
Barbar et al70	IDEAL-ICU trial—multicenter (Dijon, France) RCT comparing an early strategy (within 12 h) versus a delayed strategy (after 48 h) for the initiation of renal replacement therapy in the management of patients in the initial phase of septic shock; patient or surrogate consent before randomization or deferred consent.
Göcze et al71	BigpAK study—single-center (Regensburg, Germany) RCT comparing a biomarker-triggered Kidney Disease Improving Global Outcome care bundle versus standard ICU care in patients at increased risk of acute kidney injury after major abdominal surgery; consent secured from patient or surrogate if possible or deferred consent.
Jabre et al72	CAAM study—multicenter (Belgium and France) RCT comparing the efficacy of bag-mask ventilation versus endotracheal intubation for advanced airway management in patients with out-of-hospital cardiorespiratory arrest; waiver of consent, with deferred written consent from patient or surrogate.
Simonsen et al73	GOLIATH trial—single-center (Denmark) RCT comparing conscious sedation to general anesthesia on infarct growth in patients undergoing endovascular therapy for acute ischemic stroke; waiver of informed consent followed by written consent from patient or surrogate to remain in the trial.
van den Boogaard et al74	REDUCE study—multicenter (21 ICUs in The Netherlands) RCT comparing 2 different doses of haloperidol given as prophylactic agent compared with placebo in ICU patients at risk for delirium; consent from patient or surrogate, or if consent not possible at time of ICU admission, deferred consent.

EFIC, exception from informed consent; ICU, intensive care unit; LAR, legally authorized representative; PICU, pediatric intensive care unit; RCT, randomized controlled trial.

Table 4 shows the journals in which the 500 RCTs were published, as well as the countries in which trials were performed. Not surprisingly, our sample drew heavily from journals having high impact factors. Overall, papers were published in 96 different journals, with 52 journals publishing only 1 of the trials and 18 journals publishing 5 or more of the trials. Table 4 also presents the number of trials published in those journals having more than 5% of the total sample (ie, 25 or more papers). The fraction of RCTs foregoing prospective consent to those securing consent published in the Journal of the American Medical Association (JAMA) and the New England Journal of Medicine (NEJM) is significantly higher than in the remainder of the sample by test of proportions (χ²=16.60 with 1 df, p<0.001).

Table 4

Summary of number of trials and those that did not secure prospective fully informed consent

Journals	Trials (n)	Trials without consent, n (%)
JAMA	132	20 (15.2)
Lancet	82	4 (4.9)
Lancet Oncol	27	0 (0)
NEJM	32	7 (21.9)
All others	227	13 (5.7)
Regions *
USA	207	13 (6.3)
Other North America	113	6 (5.3)
South and Central Americas	43	1 (2.3)
Europe	295	28 (9.5)
Africa	38	0 (0)
Asia	116	2 (1.7)
Australasia	74	6 (8.1)

*These values are superadditive because of cross-region multinational trials.

A majority of RCTs in this review were performed with at least 1 site in Europe, followed by large pluralities in the USA, other North American countries, and Asia. Ninety-five trials were conducted solely in the USA.

Finally, we examined the explanations and justifications for not securing prospective consent either provided by authors or presumed based on trial details and regulatory requirements (details are provided in the online supplemental materials). Figure 2 maps these justifications against the methods of approaching consent, classifying the 44 trials. The figure shows that emergency research is the most common reason for foregoing consent (29 trials, 66%). The approaches ranged from completely waived consent with no notice or opportunity to opt out to, at the other end of the spectrum, attempts to secure consent from a patient if capable, or a family member or LAR if not, and otherwise deferring consent until some time after enrollment.

Figure 2

Reasons for not securing prospective consent from at least some subjects and methods of addressing consent in randomized controlled trials (RCTs). Numbers are reference numbers. When multiple references are provided for a trial, only the first reference appears in the figure. Note that trials may appear more than once.

Discussion

This is the first study to estimate the prevalence of published, high-impact RCTs that do not secure prospective informed consent from all subjects. Over the 5-year period, in a sample of the most highly cited RCTs published in the English language, 8.8% of 500 RCTs met our criteria for not securing prospective fully informed consent from all persons involved in the research. The majority of these trials (66%) involved emergent treatment, where patients often are incapacitated, time is of the essence, and family members who might consent may not be readily available. Nearly all of the trials (90.9%) in which prospective consent was not secured involved emergency situations, used pragmatic designs, or used cluster randomization, or a combination of these factors. This finding shows that the prior systematic reviews sensibly focused on emergency settings or trial design features that necessitated or were enhanced by provisions for waiving informed consent from at least some subjects. Nonetheless, a third of the trials were not emergency research, meaning that subjects from whom consent was waived had capacity and would be capable of providing consent, if informed and asked. The justifications for waiving or modifying consent vary widely, reflecting different national standards and accepted scientific practices.

The predominant approaches used in these trials involved either a waiver followed by securing patient or LAR consent after enrollment, or securing consent from a patient or LAR before enrollment, but if unable to do so, then waiving consent and securing consent when capable (or a LAR becomes available, where permitted). Some trials used a brief or abbreviated verbal or written consent prior to enrollment, followed by a fully informed consent after enrollment. The US Food and Drug Administration EFIC rule technically falls in the middle, requiring consent of a patient or LAR if capable and time permits or a waiver followed by notice and an opportunity to opt out.21 Several US trials using EFIC waivers in our sample appear to have gone further than required and secured follow-up consent when possible.22 23

Methods of approaching waivers of consent varied broadly in this sample of trials. Eleven trials involved a complete waiver of informed consent, with no notice provided to subjects that the study was being done and no provision of information afterwards about their participation. Most of these trials were cluster randomized, where institutional structures such as hospitals, intensive care units (ICU), clinical practices or healthcare providers, or time periods are the units of randomization, with all patients receiving the treatment assigned by the trial protocol to the cluster in which they appear for medical care. The justifications for using cluster randomization, the nature of the interventions, and the ability of individual subjects to opt out, vary widely.7 One trial, which did not mention consent at all, tested training of providers about human papillomavirus vaccination communication, and collected practice-level data on rates of vaccination of adolescent patients from a state registry.24 Thus, the intervention was directed to altering patient behaviors and outcomes, but did not fulfill the US definition of research involving human subjects.25 A different trial implemented a risk assessment tool in pediatric ICUs as part of clinical practice, with waivers of consent of patients’ parents as well as clinicians for collection of follow-up data.26

Practices vary around the need to follow-up with subjects to either give them an opportunity to opt out, or to consent either to continued participation or to collection of follow-up outcome information. Some studies sought consent, others provided information by posting notices or by verbal or written disclosure, and some did nothing. It seems there is no broad ethical consensus or standard for when it is appropriate to tell people they have been in research and how to do so.

Finally, many of the trials in our sample used multiple approaches to consent or waiver, driven either by national regulatory requirements or institutional ethics review requirements. Seventeen trials (39%) sought consent from patients when capable, or surrogates when not, at least at some sites or some of the time, while other trials secured consent from some subjects but not from others (eg, controls). To what extent enrolling vulnerable patients without prospective consent is justified when it is feasible to run trials with populations capable of consent (directly or by surrogates) is worthy of further research.

The main limitation on our review is that we focused on highly cited papers, which drew a large number of trials from high-impact journals (NEJM, Lancet, and JAMA). While our sample drew from 96 different journals, we find the odds of a highly cited paper published in JAMA or NEJM foregoing informed consent to be significantly higher than in other journals. We are unable to correct for the volume of RCTs published in these journals over the 5 years of this study, which might paint a different picture. Determining whether the rate of performance of RCTs without prospective consent is the same for a broader set of less influential trials requires further research.

In conclusion, we find that almost 9% of high-impact, international RCTs published from 2014 through 2018 did not secure informed consent from at least some subjects. Nearly all of the trials using waiver involved emergency interventions, pragmatic designs, were cluster randomized, or a combination of these factors.

Footnotes

Acknowledgements

The authors thank the study investigators who responded to our inquiries and provided detailed information about their trials, Maylene Qiu for assistance with the systematic search, and several anonymous reviewers for helpful comments.

Contributors

JFM conceived the idea for the study and performed the data analysis; the authors contributed equally to the design, search, reading, coding, checking, and interpretation of data, and the editing of the manuscript. RD wrote the first version of the paper, and subsequent edits were managed by JFM. Responsibility for the study is solely that of the authors.

Funding

The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Competing interests

JFM reports personal fees from Merck Sharp & Dohme and the American College of Radiology Imaging Network and a grant from Public Responsibility in Medicine and Research, all outside the submitted work.

Patient consent for publication

Not required.

Provenance and peer review

Not commissioned; externally peer reviewed.

References

World Medical Association . Declaration of Helsinki – ethical principles for medical research involving human subjects. Available: https://www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects/ [Accessed 11 May 2020].

Council for International Organizations of Medical Sciences . International ethical guidelines for health-related research involving humans. Geneva, 2016.

Rebers

, Aaronson

, van Leeuwen

, . Exceptions to the rule of informed consent for research with an intervention. BMC Med Ethics 2016;17:9. doi:10.1186/s12910-016-0092-6http://www.ncbi.nlm.nih.gov/pubmed/26852412

Feldman

, Kim

, Josephson

, . Effect of waivers of consent on recruitment in acute stroke trials: a systematic review. Neurology 2016;86:1543–51.doi:10.1212/WNL.0000000000002587http://www.ncbi.nlm.nih.gov/pubmed/27009262

National Health and Medical Research Council and Australian Research Council . National standards on ethical conduct in human research. Canberra, 2018.

Department of Health and Human Services . Protection of human subjects. code of federal regulations title 45, part 46.116(f)(3)(ii, 2019.

Eldridge

, Ashby

, Feder

. Informed patient consent to participation in cluster randomized trials: an empirical exploration of trials in primary care. Clin Trials 2005;2:91–8.doi:10.1191/1740774505cn070oahttp://www.ncbi.nlm.nih.gov/pubmed/16279130

Taljaard

, Goldstein

, Giraudeau

, . Cluster over individual randomization: are study design choices appropriately justified? review of a random sample of trials. Clin Trials 2020;17:253–63.doi:10.1177/1740774519896799http://www.ncbi.nlm.nih.gov/pubmed/32367741

Faden

, Kass

, Goodman

, . An ethics framework for a learning health care system: a departure from traditional research ethics and clinical ethics. Hastings Cent Rep 2013;Suppl:S16–27.doi:10.1002/hast.134http://www.ncbi.nlm.nih.gov/pubmed/23315888

10.

Faden

, Beauchamp

, Kass

. Informed consent, comparative effectiveness, and learning health care. N Engl J Med 2014;370:766–8.doi:10.1056/NEJMhle1313674http://www.ncbi.nlm.nih.gov/pubmed/24552325

11.

McKinney

, Beskow

, Ford

, . Use of altered informed consent in pragmatic clinical research. Clin Trials 2015;12:494–502.doi:10.1177/1740774515597688http://www.ncbi.nlm.nih.gov/pubmed/26374677

12.

Wendler

. “Targeted” consent for pragmatic clinical trials. J Gen Intern Med 2015;30:679–82.doi:10.1007/s11606-014-3169-2http://www.ncbi.nlm.nih.gov/pubmed/25586870

13.

Kim

. Clinical trials without consent? Perspect Biol Med 2016;59:132–46.doi:10.1353/pbm.2016.0023http://www.ncbi.nlm.nih.gov/pubmed/27499490

14.

Kim

SYH

, Miller

. Waivers and alterations to consent in pragmatic clinical trials: respecting the principle of respect for persons. IRB 2016;38:1–5.http://www.ncbi.nlm.nih.gov/pubmed/26964403

15.

Wendler

, Dickert

, Silbergleit

, . Targeted consent for research on standard of care interventions in the emergency setting. Crit Care Med 2017;45:e105–10.doi:10.1097/CCM.0000000000002023http://www.ncbi.nlm.nih.gov/pubmed/27509386

16.

Dal-Ré

, Avendaño-Solà

, Bloechl-Daum

, . Low risk pragmatic trials do not always require participants’ informed consent. BMJ 2019;1:l1092.doi:10.1136/bmj.l1092

17.

Dal-Ré

, Avendaño-Solà

, de Boer

, . A limited number of medicines pragmatic trials had potential for waived informed consent following the 2016 CIOMS ethical guidelines. J Clin Epidemiol 2019;114:60–71.doi:10.1016/j.jclinepi.2019.06.007http://www.ncbi.nlm.nih.gov/pubmed/31212001

18.

Rich

, Auten

, Gantz

, . Antenatal consent in the support trial: challenges, costs, and representative enrollment. Pediatrics 2010;126:e215–21.doi:10.1542/peds.2009-3353http://www.ncbi.nlm.nih.gov/pubmed/20587676

19.

Zelen

. A new design for randomized clinical trials. N Engl J Med 1979;300:1242–5.doi:10.1056/NEJM197905313002203http://www.ncbi.nlm.nih.gov/pubmed/431682

20.

Moher

, Liberati

, Tetzlaff

, . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009;62:1006–12.doi:10.1016/j.jclinepi.2009.06.005http://www.ncbi.nlm.nih.gov/pubmed/19631508

21.

U.S. Food and Drug Administration . Exception from informed consent requirements for emergency research. code of federal regulations Title 21, Section 50.24, 2018.

22.

Holcomb

, Tilley

, Baraniuk

, . Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma: the PROPPR randomized clinical trial. JAMA 2015;313:471–82.doi:10.1001/jama.2015.12http://www.ncbi.nlm.nih.gov/pubmed/25647203

23.

Kim

, Nichol

, Maynard

, . Effect of prehospital induction of mild hypothermia on survival and neurological status among adults with cardiac arrest: a randomized clinical trial. JAMA 2014;311:45–52.doi:10.1001/jama.2013.282173http://www.ncbi.nlm.nih.gov/pubmed/24240712

24.

Brewer

, Hall

, Malo

, . Announcements versus conversations to improve HPV vaccination coverage: a randomized trial. Pediatrics 2017;139:e20161764. doi:10.1542/peds.2016-1764http://www.ncbi.nlm.nih.gov/pubmed/27940512

25.

U.S. Department of Health and Human Services . Federal policy for the protection of human subjects. code of federal regulations, 2018.

26.

Parshuram

, Dryden-Palmer

, Farrell

, . Effect of a pediatric early warning system on all-cause mortality in hospitalized pediatric patients: the epoch randomized clinical trial. JAMA 2018;319:1002–12.doi:10.1001/jama.2018.0948http://www.ncbi.nlm.nih.gov/pubmed/29486493

27.

IMPROVE Trial Investigators, Powell

, Sweeting

, . Endovascular or open repair strategy for ruptured abdominal aortic aneurysm: 30 day outcomes from IMPROVE randomised trial. BMJ 2014;348:f7661. doi:10.1136/bmj.f7661http://www.ncbi.nlm.nih.gov/pubmed/24418950

28.

Shahzad

, Kemp

, Mars

, . Unfractionated heparin versus bivalirudin in primary percutaneous coronary intervention (HEAT-PPCI): an open-label, single centre, randomised controlled trial. Lancet 2014;384:1849–58.doi:10.1016/S0140-6736(14)60924-7http://www.ncbi.nlm.nih.gov/pubmed/25002178

29.

Dickert

, Wendler

, Devireddy

, . Consent for pragmatic trials in acute myocardial infarction. J Am Coll Cardiol 2018;71:1051–3.doi:10.1016/j.jacc.2017.12.043http://www.ncbi.nlm.nih.gov/pubmed/29495987

30.

Perkins

, Lall

, Quinn

, . Mechanical versus manual chest compression for out-of-hospital cardiac arrest (PARAMEDIC): a pragmatic, cluster randomised controlled trial. Lancet 2015;385:947–55.doi:10.1016/S0140-6736(14)61886-9http://www.ncbi.nlm.nih.gov/pubmed/25467566

31.

Stub

, Smith

, Bernard

, . Air versus oxygen in ST-segment-elevation myocardial infarction. Circulation 2015;131:2143–50.doi:10.1161/CIRCULATIONAHA.114.014494http://www.ncbi.nlm.nih.gov/pubmed/26002889

32.

Gonzalez

, Moore

, . Goal-directed hemostatic resuscitation of trauma-induced coagulopathy: a pragmatic randomized clinical trial comparing a viscoelastic assay to conventional coagulation assays. Ann Surg 2016;263:1051–9.doi:10.1097/SLA.0000000000001608http://www.ncbi.nlm.nih.gov/pubmed/26720428

33.

Heddle

, Cook

, Arnold

, . Effect of short-term vs. long-term blood storage on mortality after transfusion. N Engl J Med 2016;375:1937–45.doi:10.1056/NEJMoa1609014http://www.ncbi.nlm.nih.gov/pubmed/27775503

34.

Hofmann

, James

, Jernberg

, . Oxygen therapy in suspected acute myocardial infarction. N Engl J Med 2017;377:1240–9.doi:10.1056/NEJMoa1706222http://www.ncbi.nlm.nih.gov/pubmed/28844200

35.

Hofmann

, James

, Jernberg

, . Supplementary appendix. Available: https://www.nejm.org/doi/full/10.1056/NEJMoa1706222 [Accessed 15 May 2020].

36.

Muir

, Ford

, Messow

C-M

, . Endovascular therapy for acute ischaemic stroke: the pragmatic ischaemic stroke thrombectomy evaluation (PISTE) randomised, controlled trial. J Neurol Neurosurg Psychiatry 2017;88:38–44.doi:10.1136/jnnp-2016-314117http://www.ncbi.nlm.nih.gov/pubmed/27756804

37.

Benger

, Kirby

, Black

, . Effect of a strategy of a supraglottic airway device vs tracheal intubation during out-of-hospital cardiac arrest on functional outcome: the AIRWAYS-2 randomized clinical trial. JAMA 2018;320:779–91.doi:10.1001/jama.2018.11597http://www.ncbi.nlm.nih.gov/pubmed/30167701

38.

Taylor

, Black

, Brett

S J

, . Design and implementation of the AIRWAYS-2 trial: a multi-centre cluster randomised controlled trial of the clinical and cost effectiveness of the i-gel supraglottic airway device versus tracheal intubation in the initial airway management of out of hospital cardiac arrest. Resuscitation 2016;109:25–32.doi:10.1016/j.resuscitation.2016.09.016http://www.ncbi.nlm.nih.gov/pubmed/27697605

39.

Franklin

, Babl

, Schlapbach

, . A randomized trial of high-flow oxygen therapy in infants with bronchiolitis. N Engl J Med 2018;378:1121–31.doi:10.1056/NEJMoa1714855http://www.ncbi.nlm.nih.gov/pubmed/29562151

40.

Franklin

, Dalziel

, Schlapbach

, . Early high flow nasal cannula therapy in bronchiolitis, a prospective randomised control trial (protocol): a paediatric acute respiratory intervention study (Paris). BMC Pediatr 2015;15:183. doi:10.1186/s12887-015-0501-xhttp://www.ncbi.nlm.nih.gov/pubmed/26572729

41.

Lamb

, Sheehan

, Atherton

, . Dementia and physical activity (DAPA) trial of moderate to high intensity exercise training for people with dementia: randomised controlled trial. BMJ 2018;361:k1675.doi:10.1136/bmj.k1675http://www.ncbi.nlm.nih.gov/pubmed/29769247

42.

Shah

ASV

, Anand

, Strachan

, . High-sensitivity troponin in the evaluation of patients with suspected acute coronary syndrome: a stepped-wedge, cluster-randomised controlled trial. Lancet 2018;392:919–28.doi:10.1016/S0140-6736(18)31923-8http://www.ncbi.nlm.nih.gov/pubmed/30170853

43.

Steinhubl

, Waalen

, Edwards

, . Effect of a home-based wearable continuous ECG monitoring patch on detection of undiagnosed atrial fibrillation: the mSToPS randomized clinical trial. JAMA 2018;320:146–55.doi:10.1001/jama.2018.8102http://www.ncbi.nlm.nih.gov/pubmed/29998336

44.

Wang

, Schmicker

, Daya MR , . Effect of a strategy of initial laryngeal tube insertion vs endotracheal intubation on 72-hour survival in adults with out-of-hospital cardiac arrest: a randomized clinical trial. JAMA 2018;320:769–78.doi:10.1001/jama.2018.7044http://www.ncbi.nlm.nih.gov/pubmed/30167699

45.

Zimmermann

, Swami

, Krzyzanowska

, . Early palliative care for patients with advanced cancer: a cluster-randomised controlled trial. Lancet 2014;383:1721–30.doi:10.1016/S0140-6736(13)62416-2http://www.ncbi.nlm.nih.gov/pubmed/24559581

46.

Garg

, Toy

, Tripodis

, . Addressing social determinants of health at well child care visits: a cluster RCT. Pediatrics 2015;135:e296–304.doi:10.1542/peds.2014-2888http://www.ncbi.nlm.nih.gov/pubmed/25560448

47.

Young

, Bailey

, Beasley

, . Effect of a buffered crystalloid solution vs saline on acute kidney injury among patients in the intensive care unit: the split randomized clinical trial. JAMA 2015;314:1701–10.doi:10.1001/jama.2015.12334http://www.ncbi.nlm.nih.gov/pubmed/26444692

48.

Dempsey

, Pyrznawoski

, Lockhart

, . Effect of a health care professional communication training intervention on adolescent human papillomavirus vaccination: a cluster randomized clinical trial. JAMA Pediatr 2018;172:e180016. doi:10.1001/jamapediatrics.2018.0016http://www.ncbi.nlm.nih.gov/pubmed/29507952

49.

Martin

, Donovan

, Turner

, . Effect of a low-intensity PSA-based screening intervention on prostate cancer mortality: the cap randomized clinical trial. JAMA 2018;319:883–95.doi:10.1001/jama.2018.0154http://www.ncbi.nlm.nih.gov/pubmed/29509864

50.

Donovan

, Martin

, Hamdy

, . CAP study: protocol v3, 2005. Available: https://jamanetwork.com/journals/jama/fullarticle/2673968 [Accessed 15 May 2020].

51.

Parshuram

, Dryden-Palmer

, Farrell

, . Evaluating processes of care and outcomes of children in hospital (epoch): study protocol for a randomized controlled trial. Trials 2015;16:245. doi:10.1186/s13063-015-0712-3http://www.ncbi.nlm.nih.gov/pubmed/26033094

52.

Ebinger

, Winter

, Wendt

, . Effect of the use of ambulance-based thrombolysis on time to thrombolysis in acute ischemic stroke: a randomized clinical trial. JAMA 2014;311:1622–31.doi:10.1001/jama.2014.2850http://www.ncbi.nlm.nih.gov/pubmed/24756512

53.

Robertson

, Hannay

, Yamal

J-M

, . Effect of erythropoietin and transfusion threshold on neurological recovery after traumatic brain injury: a randomized clinical trial. JAMA 2014;312:36–47.doi:10.1001/jama.2014.6490http://www.ncbi.nlm.nih.gov/pubmed/25058216

54.

Yamal

J-M

, Robertson

, Rubin

, . Enrollment of racially/ethnically diverse participants in traumatic brain injury trials: effect of availability of exception from informed consent. Clin Trials 2014;11:187–94.doi:10.1177/1740774514522560http://www.ncbi.nlm.nih.gov/pubmed/24686108

55.

Rubertsson

, Lindgren

, Smekal

, . Mechanical chest compressions and simultaneous defibrillation vs conventional cardiopulmonary resuscitation in out-of-hospital cardiac arrest: the LiNc randomized trial. JAMA 2014;311:53–61.doi:10.1001/jama.2013.282538http://www.ncbi.nlm.nih.gov/pubmed/24240611

56.

Skolnick

, Maas

, Narayan

, . A clinical trial of progesterone for severe traumatic brain injury. N Engl J Med 2014;371:2467–76.doi:10.1056/NEJMoa1411090http://www.ncbi.nlm.nih.gov/pubmed/25493978

57.

Banerjee

, Teng

, Cunningham

, . Randomized trial of rapid multiplex polymerase chain reaction-based blood culture identification and susceptibility testing. Clin Infect Dis 2015;61:1071–80.doi:10.1093/cid/civ447http://www.ncbi.nlm.nih.gov/pubmed/26197846

58.

Ringh

, Rosenqvist

, Hollenberg

, . Mobile-phone dispatch of laypersons for CPR in out-of-hospital cardiac arrest. N Engl J Med 2015;372:2316–25.doi:10.1056/NEJMoa1406038http://www.ncbi.nlm.nih.gov/pubmed/26061836

59.

Steenbergen

, Sellaro

, van Hemert

, . A randomized controlled trial to test the effect of multispecies probiotics on cognitive reactivity to sad mood. Brain Behav Immun 2015;48:258–64.doi:10.1016/j.bbi.2015.04.003http://www.ncbi.nlm.nih.gov/pubmed/25862297

60.

Stéphan

, Barrucand

, Petit

, . High-flow nasal oxygen vs noninvasive positive airway pressure in hypoxemic patients after cardiothoracic surgery: a randomized clinical trial. JAMA 2015;313:2331–9.doi:10.1001/jama.2015.5213http://www.ncbi.nlm.nih.gov/pubmed/25980660

61.

Fivez

, Kerklaan

, Mesotten

, . Early versus late parenteral nutrition in critically ill children. N Engl J Med 2016;374:1111–22.doi:10.1056/NEJMoa1514762http://www.ncbi.nlm.nih.gov/pubmed/26975590

62.

Fivez

, Kerklaan

, Verbruggen

, . Impact of withholding early parenteral nutrition completing enteral nutrition in pediatric critically ill patients (PEPaNIC trial): study protocol for a randomized controlled trial. Trials 2015;16:202. doi:10.1186/s13063-015-0728-8http://www.ncbi.nlm.nih.gov/pubmed/25927936

63.

Girardis

, Busani

, Damiani

, . Effect of conservative vs conventional oxygen therapy on mortality among patients in an intensive care unit: the Oxygen-ICU randomized clinical trial. JAMA 2016;316:1583–9.doi:10.1001/jama.2016.11993http://www.ncbi.nlm.nih.gov/pubmed/27706466

64.

Gordon

, Mason

, Thirunavukkarasu

, . Effect of early vasopressin vs norepinephrine on kidney failure in patients with septic shock: the VANISH randomized clinical trial. JAMA 2016;316:509–18.doi:10.1001/jama.2016.10485http://www.ncbi.nlm.nih.gov/pubmed/27483065

65.

Goyal

, Jadhav

, Bonafe

, . Analysis of workflow and time to treatment and the effects on outcome in endovascular treatment of acute ischemic stroke: results from the swift prime randomized controlled trial. Radiology 2016;279:888–97.doi:10.1148/radiol.2016160204http://www.ncbi.nlm.nih.gov/pubmed/27092472

66.

Reade

, Eastwood

, Bellomo

, . Effect of dexmedetomidine added to standard care on ventilator-free time in patients with agitated delirium: a randomized clinical trial. JAMA 2016;315:1460–8.doi:10.1001/jama.2016.2707http://www.ncbi.nlm.nih.gov/pubmed/26975647

67.

Schönenberger

, Uhlmann

, Hacke

, . Effect of conscious sedation vs general anesthesia on early neurological improvement among patients with ischemic stroke undergoing endovascular thrombectomy: a randomized clinical trial. JAMA 2016;316:1986–96.doi:10.1001/jama.2016.16623http://www.ncbi.nlm.nih.gov/pubmed/27785516

68.

Zarbock

, Kellum

, Schmidt

, . Effect of early vs delayed initiation of renal replacement therapy on mortality in critically ill patients with acute kidney injury: the ELAIN randomized clinical trial. JAMA 2016;315:2190–9.doi:10.1001/jama.2016.5828http://www.ncbi.nlm.nih.gov/pubmed/27209269

69.

Lascarrou

, Boisrame-Helms

, Bailly

, . Video laryngoscopy vs direct laryngoscopy on successful first-pass Orotracheal intubation among ICU patients: a randomized clinical trial. JAMA 2017;317:483–93.doi:10.1001/jama.2016.20603http://www.ncbi.nlm.nih.gov/pubmed/28118659

70.

Barbar

, Clere-Jehl

, Bourredjem

, . Timing of renal-replacement therapy in patients with acute kidney injury and sepsis. N Engl J Med 2018;379:1431–42.doi:10.1056/NEJMoa1803213http://www.ncbi.nlm.nih.gov/pubmed/30304656

71.

Göcze

, Jauch

, Götz

, . Biomarker-guided intervention to prevent acute kidney injury after major surgery: the prospective randomized BigpAK study. Ann Surg 2018;267:1013–20.doi:10.1097/SLA.0000000000002485http://www.ncbi.nlm.nih.gov/pubmed/28857811

72.

Jabre

, Penaloza

, Pinero

, . Effect of bag-mask ventilation vs endotracheal intubation during cardiopulmonary resuscitation on neurological outcome after out-of-hospital cardiorespiratory arrest: a randomized clinical trial. JAMA 2018;319:779–87.doi:10.1001/jama.2018.0156http://www.ncbi.nlm.nih.gov/pubmed/29486039

73.

Simonsen

, Yoo

, Sørensen

, . Effect of general anesthesia and conscious sedation during endovascular therapy on infarct growth and clinical outcomes in acute ischemic stroke: a randomized clinical trial. JAMA Neurol 2018;75:470–7.doi:10.1001/jamaneurol.2017.4474http://www.ncbi.nlm.nih.gov/pubmed/29340574

74.

van den Boogaard

, Slooter

AJC

, Brüggemann

RJM

, . Effect of haloperidol on survival among critically ill adults with a high risk of delirium: the reduce randomized clinical trial. JAMA 2018;319:680–91.doi:10.1001/jama.2018.0160http://www.ncbi.nlm.nih.gov/pubmed/29466591