Our formal consultation process is over, thank you to those who have provided feedback. We are currently reviewing responses and are commencing revisions of the final guidance text. If you have particular comments please do send in to goodtrials@wellcome.org and we will consider those as we now move towards finalisation of the guidance.

1

Draft Guidance for Good Randomized Clinical Trials

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

11

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

168

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

Introduction

Randomized controlled clinical trials (RCTs) play a central role in generating the evidence needed to inform the development and implementation of health interventions.

However, useful evidence from good RCTs is often lacking. This can be because the RCTs were never done, because those that were done failed to produce scientifically robust and clinically relevant answers, or because the results were never published. This can result in failure to identify and use effective interventions or the continuing use of ineffective or hazardous interventions. These issues waste resources, cause unnecessary harm or suffering, and reduce trust in those who develop or use healthcare interventions. It must be made easier to do good RCTs to inform future care.

A range of guidelines for clinical trials exist, but largely fail to provide guidance on the underpinning principles of RCTs necessary to generate reliable results safely and ethically, regardless of context. Many guidelines focus narrowly on clinical trials (including non-randomized studies) that are intended to generate data for regulatory submission to support a new drug license. There is an unmet need for guidance to promote the unique benefits of RCTs across all contexts and which focuses on the unique strengths of randomization.

The Good Clinical Trials Collaborative (GCTC) was established to develop and promote the adoption of new guidance to address this issue. The GCTC has brought together a wide range of individuals and organizations with an interest and role to play in the design, delivery, analysis and reporting of RCTs, and in implementing the results. This includes those who fund, regulate, design, deliver, or are responsible for RCTs, those who provide quality assurance, audit or inspection functions, research organizations, ethicists, clinicians, participants, and lay health advocates. It includes those from a wide variety of sectors (industry, academia, government, charitable, non-governmental organizations, participant and public groups) and settings (including higher and lower income countries around the world).

This new guidance has been developed to be:

  • Based on key scientific and ethical principles and focused on issues that materially matter to the well-being of trial participants and the reliability of RCT results;

  • Clear, concise, consistent and proportionate to the context and setting in which RCTs are conducted, recognising that there are risks associated with both usual clinical practice and a lack of reliable evidence on the effects of an intervention;

  • Forward looking, fostering innovation in health interventions and trial methods, including the appropriate use of routine healthcare data, technologies, and designs; and

  • Flexible, widely-applicable, utilisable & durable across disease areas, intervention types, development phases, trial designs, geographies and time.

 
The objective of this guidance is to establish the key principles of RCTs: what makes an RCT good in its design and analysis, as well as ethical and social value; and why this is so. This guidance aims to enable those involved in RCTs to work out for themselves how to design and deliver their particular RCT in their particular setting.

The guidance focuses on randomized controlled trials. Most healthcare interventions have small or moderate effects on health and disease. However, even moderate improvements in health can be important to those they benefit, provided any benefits are not substantially offset by detrimental effects. To establish reliably whether a healthcare intervention has any modest effect requires that any moderate biases or random errors inherent in the study design are both small with respect to the expected treatment effect. This requires the combination of:

  • Randomization without foreknowledge of intervention allocation: to guarantee the avoidance of moderate biases by making sure that the groups being compared are as similar as possible in all ways other than the interventions being compared. Thus, any differences in health outcomes between the groups will either be due to the effect of the study intervention or to the play of chance; and

  • Adequate sample size: to guarantee the avoidance of moderate random errors (i.e., the play of chance);

  • Unbiased assessment of outcomes: i.e. not influenced by knowledge of intervention allocation; and

  • Intention-to-treat analyses: to compare outcomes according to the intervention to arm to which participants were allocated and without emphasis on data-derived subgroups.

 
This guidance is intended to support all individuals and organizations involved in the planning, conduct, analysis, oversight, interpretation, funding, and oversight of all trials in which randomization is used to assess the effects of any health intervention for any purpose in any setting. The remit includes, for example:

Any design: including comparisons of two or more interventions (one of which may be to provide no additional active intervention on top of usual practice); blinded or not; parallel, cluster, crossover or other design.

Any health intervention: pharmaceutical and biological therapies; medical devices; surgical procedures; vaccines; nutritional measures; cognitive, behavioural and psychological interventions; digital and public health approaches.

Any purpose: intended to support determination of the safety and efficacy of new and existing interventions; regulatory submissions; health technology assessments; and public health strategies.

Any setting: include any geographic, economic or societal context; and any context including RCTs based in hospital, primary care or community settings; or delivered direct to participant.

Any role: include researchers and clinicians; patient and public groups (including trial participants); regulators and other government bodies; ethics committees and institutional review boards; funders; trial sponsors (academic and pharmaceutical); the health intervention industry and those who regulate or provide audit and quality assurance functions.

How to use this guidance

This document provides the underpinning principles of good RCTs. A supplemental glossary is in development to explain terms commonly used in this guidance. Illustrative examples of the principles are being prepared to enable a better understanding of how this guidance can be used in practice.

Principles of Good Randomized Controlled Trials

In this guidance, ‘good’ should be taken to mean: informative, ethical and efficient. The following principles, taken together, capture the qualities of a well-planned, well-run, and clinically relevant trial. The methods and approaches necessary to achieve these qualities will differ in small or large ways from trial to trial but their validity is universal.


1.   Good RCTs are designed to produce scientifically sound answers to relevant questions


RCTs should help to resolve important uncertainties about effects of health interventions. Depending on the context, the results may be needed to determine whether to proceed with development or further evaluation of the intervention; or to directly inform regulatory licensing, clinical guidelines, and/or health policy. In each case, any uncertainties that remain at the end of the RCT should be sufficiently small to allow meaningful decisions to be made.


Good RCTs include the following features: 


Appropriate trial population


Key Message: The eligibility criteria should be tailored to the question the RCT sets out to answer. Inclusion criteria should define a population for whom the intervention being assessed is relevant and should not be unduly restrictive. Exclusion criteria should be focused on identifying individuals for whom participation would place them at undue risk by comparison with any potential benefits (e.g. based on their medical history or concomitant medication) or for whom the benefits are already well established. Efforts should be made to include a broad and varied population (e.g. with appropriate sex, age, ethnic and socioeconomic diversity), unless there is strong medical or scientific justification for doing otherwise.


Why this is important: Inclusive eligibility criteria increase the relevance of the findings and allow assessment of whether there is good evidence that the effects (beneficial or adverse), and/or acceptability of an intervention or its delivery, differ in any particular subgroup. By including a wide range of participants but recording key characteristics for them (e.g. specific genetic, demographic, or health characteristics), it will be possible to assess whether there are material differences in the effect of an intervention between such groups.


Relevant outcomes


Key Message: RCT outcomes need to be relevant to the participants under study and the question being addressed. Such outcomes should reliably measure or assess what they are intended to measure (i.e. be validated for their intended use). Outcomes may include a wide range of assessment types including physiological measures, symptomatic scores, patient-reported outcomes, functional status, clinical events, or healthcare utilization.


Why this is important: Trial outcomes should be sensitive to the anticipated effects of the intervention, appropriate to the study question, and applicable (generalizable) and meaningful to the relevant population. The choice of outcomes may vary depending on the extent of prior knowledge (e.g. assessing first the effects on imaging and laboratory markers and later the effects on clinical outcomes).


Robust intervention allocation


Key Message: Proper randomization requires both generation of an unpredictable allocation schedule and concealment of intervention allocation before the point of randomization. It should be impossible to predict in advance the study intervention to which an individual trial participant or individual cluster (e.g. hospital or city in a cluster RCT) is likely to be allocated, so that investigators, health care providers involved, and potential participants are not aware of the upcoming intervention assignment.


Why this is important: Proper randomization allows for like with like comparisons and avoids moderate biases, by making sure that groups being compared in a study are as similar as possible in all ways other than the interventions being compared. This means that subsequent differences in health outcomes between the groups are due either to the play of chance or are due causally to differences in the study intervention. Additional measures such as stratification or minimization may be used to reduce the size of random differences between intervention groups, provided that they do not result in predictability of intervention allocation. The absence of adequate allocation concealment prior to randomization can lead to selection bias (i.e. the decision to enter a particular participant in a trial can be influenced by knowledge of which intervention they are likely to be assigned to).


Proportionate, efficient and reliable capture of data


Key Message: Data collection should focus on those aspects needed to assess and interpret the trial results as specified in the protocol and should not be excessive. The choice of data collection approach may be influenced by considerations such as availability, suitability, and usability as well as the extent to which such information is sufficiently comprehensive, detailed, and timely.


Tools and methods for data collection, storage, exchange, and access should enable the RCT to be conducted as designed, support privacy and security, and enable reliable and consistent analyses. Digital technology and routine healthcare data can provide alternative or complementary means to record information about participants and their health at study entry, during the initial intervention and follow-up period, and for many years beyond.


Why this is important: Disproportionate data collection wastes time and resource and detracts from the objective of the RCT. The choice of method used for data collection can have an important bearing on trial reliability and feasibility. Use of data standards can help ensure data quality and data integrity. Use of digital technology and routine healthcare data can improve the relevance and completeness of information collected (e.g. reducing loss to follow-up).


Adequate RCT size


Key Message: An RCT should be sufficiently large and statistically powered to robustly assess the question it sets out to address.


Why this is important: For moderate effects of healthcare interventions to be reliably detected or reliably refuted, then in addition to randomization (to minimise moderate biases), random errors must be small by comparison with the anticipated size of the effect of the intervention. The only way to guarantee the avoidance of moderate random errors is to study sufficiently large numbers. One of the main techniques for obtaining appropriately large-scale randomized evidence is to make RCTs as simple as possible through use of streamlined methodology. However, there are scenarios for which it is inappropriate or challenging to randomize large numbers of participants, such as trials assessing interventions in rare diseases. For such trials, it may be helpful to select a relevant outcome for which the effect size is expected to be larger (e.g. a physiological or imaging biomarker). Alternatively, it may be possible to reduce random errors through the statistical analyses that are done (e.g., analyses of a continuous outcome adjusted for baseline values of that outcome would typically increase statistical power compared with an analysis of either mean follow-up levels or an analysis of mean changes in levels).


Adherence to allocated trial intervention


Key Message: Efforts should be made to facilitate and encourage adherence to the allocated intervention(s).


Why this is important: If trial participants allocated to active intervention do not receive it as planned, or if those allocated to the control group (e.g. placebo or usual care) start to receive the active intervention, then the contrast between the two study groups is less. As a consequence the statistical power to assess any differences (beneficial or hazardous) between interventions is reduced (and it is more likely to falsely conclude that there is no meaningful difference between the interventions when in fact there is one).


Blinding and masking of allocated trial intervention


Key Message: Knowledge of the allocated trial intervention may influence the behaviour of participants, those who care for them, or those assessing study outcomes, leading to unreliable or false conclusions about the impact of the intervention. Masking (or blinding) participants, investigators, health care providers, or those assessing outcomes to the assigned intervention can help prevent such issues.


Why this is important: In some RCTs, knowledge of the allocated intervention can influence the nature and intensity of clinical management, the reporting of symptoms, or the assessment of functional status or clinical outcomes (particularly if they are not objective such as death). This can be avoided through use of placebo medications or dummy interventions or by ensuring that those individuals or systems responsible for assessing participant outcomes are unaware of the treatment allocation. The latter is particularly important for trials in which blinding of the allocated intervention is not feasible or desirable.


Completeness of follow-up


Key Message: Participant outcomes should be ascertained over the full duration of the RCT, regardless of whether trial participants continue to receive the allocated intervention or cease to do so (e.g. because of perceived or real adverse effects of the intervention). In some cases, it may also be appropriate to continue follow-up for many years beyond the end of the trial.


Why this is important: Continued follow-up of all randomized participants (even if some stop taking their assigned intervention) maintains the like-with-like comparison produced by the randomization process. Premature cessation of follow-up or post-randomization exclusion of participants should therefore be avoided since it may introduce systematic bias, particularly if the type of patients excluded from one intervention group differs from those excluded from another. Incomplete follow-up may reduce the statistical power of an RCT (i.e. the ability to distinguish any differences between the interventions) and underestimate the true effects (benefits or hazards) of the intervention.


Extended follow-up can allow for detection of beneficial or harmful effects of the study intervention that may persist or emerge months or years after the initial randomized comparison. 


Ascertainment of relevant outcomes 


Key Message: Processes for ascertaining study outcomes should be the same in all randomized groups. This includes the frequency and intensity of assessments. Particular care should be taken to ensure that the people used to assess, clarify, and adjudicate study outcomes are not influenced by knowledge of the allocated intervention (i.e. blinded or masked outcome assessment). Equally, the methods for acquiring, processing, and combining sources of information (e.g. to define participant characteristics or clinical outcomes) should be designed and operated without access to the intervention allocation for individual participants or knowledge of the unblinded trial results.


Why this is important: If the methods used to assess, clarify or classify outcomes differ between the assigned interventions, the results may be biased in one direction or other leading to inappropriate conclusions about the true effect of the intervention. Therefore, the approach used to assess what happens to participants should be the same regardless of the assigned intervention and those making judgements about the occurrence or nature of these outcomes should be unaware of the assigned intervention (or features, such as laboratory assays, that would make it easier to guess the assignment) for each participant.


Statistical analysis and reporting


Key Message: Trial results should be analysed in accordance with the protocol and pre-specified statistical analysis plan, with any post-hoc analyses being clearly identified as such. The main analyses should follow the intention-to-treat principle, meaning that outcomes should be compared according to the intervention arm to which the participants were originally allocated at randomization regardless of whether some of those participants subsequently received some or none of the intended intervention, and regardless of the extent to which the post randomization follow-up procedures were completed.


Subgroup analyses should be interpreted cautiously, especially if they are not pre-specified or multiple in number (whether pre-specified or not). In general, any prognostic features that are to be used in analyses of intervention effects in RCTs should be irreversibly recorded (or sample collected) before randomization.


Why this is important: High quality statistical analyses are vital to reliability of RCT results. Analyses should be pre-specified so that they are not prompted or biased by knowledge of the RCT results.


Analysing all participants according to the intervention to which they were originally allocated (‘intention-to-treat’ analysis) is important because even in a properly randomized trial, bias can be inadvertently introduced by the post-randomization exclusion of certain patients (such as those who are non-adherent with study treatment).


Additional analyses can also be reported, for example, in describing the frequency of a specific side effect it may be justifiable to record its incidence only among those who received the active intervention, because strictly randomized comparisons may not be needed to assess large effects. However, in assessing moderate effects on the main outcome of interest, ‘on-treatment’ or ‘per protocol’ analyses can be misleading, and ‘intention-to-treat’ analyses are generally more trustworthy as to whether there is any real difference between the allocated trial interventions in

their effects.


One of the most important sources of bias in the analysis is undue concentration on just part of the evidence (i.e. data-derived subgroup analyses). This is because apparent differences between the therapeutic effects in different subgroups of study participants can often be produced solely by the play of chance. Subgroups therefore need to be relevant, limited and pre-specified. Analysis of results in sub-groups determined by post-randomization features should be avoided because if the recorded value of some feature is (or could be) affected by the trial intervention, then comparisons within subgroups that are defined by that factor might be biased. If sub-groups are not interpreted with caution it can lead to people being treated inappropriately (given an intervention that is ineffective or harmful) or untreated inappropriately (not being given an intervention that would benefit them).


Monitoring emerging data on safety and efficacy


Key Message: An independent Data Monitoring Committee (DMC) provides a robust means to evaluate safety and efficacy data from an ongoing RCT.


Why this is important: All those involved in the design, conduct and oversight of an ongoing RCT should remain blind to the emerging data until after the study conclusion so as not to introduce bias into the conclusions (e.g. by stopping the trial early when the results happen by chance to look favourable). A DMC should include members with relevant skills to analyse and interpret the emerging data. A DMC can review both efficacy and safety data and advise the RCT organisers when there is clear evidence to suggest a change in the protocol or procedures or cessation of one or more aspects of the trial. This may be due to evidence of benefit or harm or futility (where continuing the trial is unlikely to provide any meaningful new information). In making such recommendations, a DMC should take account both of the unblinded analyses of the RCT and also of information available from other sources (including publications from other trials). Some trials may not require a DMC (e.g. if the intervention is well-known to have minimal potential for harm or the trial is short-term and would not be modified regardless of interim data).

2.   Good RCTs respect the rights and well-being of participants


All those involved in RCTs share responsibility for building and sustaining the trust of collaborating partner organizations and clinical communities, participants, and the wider public. Trust is undermined when RCTs are not sufficiently relevant, fair, transparent, and respectful of the rights, interests, concerns, and values of all involved (especially those people who participate in them or whose care will be influenced by the results).


Working in partnership with people and communities


Key Message: Potential participants and/or members of the relevant community provide valuable contributions to the design, execution and interpretation of RCTs.


Why this is important: Patient and public involvement can play a key role in refining and prioritising research questions; assessing RCT acceptability and feasibility; selection of outcomes that are relevant and meaningful to the intended population; developing the RCT design and procedures; optimising the nature and delivery of information; and encouraging dialogue about access to healthcare interventions that prove effective. Working in partnership with people and communities is likely to increase trust and confidence, while significantly decreasing the risk of important groups being excluded, or the needs of local populations or sectors being overlooked or misunderstood. 


 

Appropriate participant communication


Key Message: At all stages of an RCT (before, during and after), relevant, easily understandable information should be shared with trial participants, carefully balancing the duty to inform against the risk of information saturation and taking account of the clinical context. Information should be provided in a clear manner and in a suitable language and format for the intended audience.


Why this is important: Providing timely and relevant information to participants during a trial facilitates ethical research with benefits to both the participants and the quality of the trial results. It is essential that potential or recruited trial participants are appropriately informed but presenting excessive or exhaustive detail can work against this objective by overwhelming, confusing or disconcerting potential participants. Care should be taken to communicate effectively and enable relevant discussion. The exact approach may be influenced by clinical or other issues.


Relevant consent


Key Message: The trial consent process should clearly explain to potential trial participants the reasons why the trial is being done, the questions it is seeking to answer, what is involved for them, and the potential benefits and risks of participation. The extent, nature and timing of information provided to inform consent decisions should be guided by the level of additional risks and commitment that participation in the RCT would involve by comparison with the risks posed by the usual clinical care or circumstances that the same person would normally receive. The information provided should prioritize the needs and expectations of the participant rather than of the organization or individuals conducting the RCT. Models and methods of obtaining and maintaining ongoing consent and communication should be relevant to RCT it relates to.


Why this is important: Valid consent should be informed, voluntary, and competently given. Note however, that for some trials and in some individual situations, consent may be unnecessary. In such cases, risks should be minimal and there should be no additional burdens to participation. The proposed trial should be contextualized through comparison with the usual care a prospective participant might receive outside the trial. Consent documentation that is excessively lengthy or laden with legalistic or other technical language should be avoided.


Modifying consent


Key Message: Participants should be free to withdraw or modify the nature of their consent, but care should be taken to determine the intended meaning of such individual decisions.


Why this is important: The term ‘withdrawal’ can mean different things to different people, ranging from participants wishing to stop receiving the study intervention, to stopping attending study visits in person (but perhaps be happy to be contacted by telephone or for information about their health outcomes to be collected from their regular doctors or from routine health data systems), to their biological samples no longer being assayed or stored, or their data no longer being processed or shared. If the reasons for withdrawal are not properly explored, and the ‘withdrawal’ is interpreted with prejudice to mean complete removal from the study, trial participants may be unnecessarily and inadvertently lost to full or partial follow-up.


Implications of changing consent


Key Message: The rights of an individual participant to withdraw consent for use of trial data that has already been collected should be balanced with scientific and ethical requirements.


Why this is important: It can be appropriate to make data which has already been collected available for analysis in order to demonstrate or preserve research integrity. Those involved in a trial and those whose care is influenced by its results should be able to be assured that the data are valid, and that they have not been modified through inadvertent, deliberate, or malicious means. Removing data can result in unreliable or inconclusive findings, with ethical and clinical safety consequences for both participants continuing in the trial, and the care of future patients.

3.   Good RCTs are collaborative and transparent


Collaboration among organizations


Key Message: It is important that interactions between individuals in different organisations, including those in resource-rich and resource-poor settings and among commercial, academic and healthcare sectors are fair and respectful of the interests, concerns and values of all involved, including trial participants and the communities they come from.


Why this is important: Collaborative working shares ideas and expertise and can maximise use of resources and increase efficiency.


Transparency and trust


Key Message: Clinical trials should be registered from the outset on a publicly available trials database. Making other trial information (including the trial protocol) public is strongly encouraged. Once the RCT is completed, trial reports should be publicly available and should describe the study design, methods, and results in a clear and transparent manner. It can be helpful for such reports to be available in formats that permit both professional and lay readers to understand and interpret the results. Reporting results to participants and to the public requires different approaches to reporting results to the clinical and scientific community. Data sharing should be enabled at a suitable time if ethical, feasible, and scientifically appropriate.


Why this is important: Transparency and sharing of knowledge about healthcare interventions helps generate further knowledge, build and maintain trust, and gives confidence both to those involved in the RCT and to those who are not. Communication of trial results (regardless of what those findings are) is vital to guide future research and reduce unnecessary duplication of effort (which wastes resources). Good communication can also support wider efforts to foster potential collaborations and increase informed participation in RCTs.

4.   Good RCTs are designed to be feasible for their context 


Setting and context


Key Message: The design and implementation of RCTs should recognize and be shaped by the characteristics of the settings in which they take place, including the health needs and preferences of communities, and their understanding of clinical trials, as identified through appropriate patient and public involvement.


Why this is important: Relevant and accessible RCTs are more likely to recruit a sufficient number of trial participants. Good patient and public involvement and education across the relevant communities help shape successful recruitment and subsequent adoption of the results.


Use of existing resources 


Key Message: RCTs should be tailored to be practicable given the available infrastructure in relevant settings. This includes making optimal use of pre-existing resources and facilities, including utilising any expertise, skills, professional standards, and quality oversight mechanisms associated with routine healthcare practice. While all individuals involved in performing an RCT should be qualified by education, training or experience to perform their respective task(s), it should be recognized that there are many aspects of delivering a clinical trial that are in line with routine care and should not require additional training, procedures or checks.


Why this is important: RCTs should not be wasteful of staff and participants’ time, use of interventional or other medical supplies, energy, or environmental resources. Where there are strengths and safeguards in routine systems, these should not be duplicated or altered without careful justification. The closer trial processes are to routine practice (for participants and staff), the more efficiently and effectively they are likely to be delivered, the fewer mistakes they are likely to make, resulting in improved quality.

5.   Good RCTs manage quality effectively and efficiently


Planning for success and focussing on issues that matter


Key Message: Good quality should be prospectively built into the design and delivery of RCTs, rather than relying on retrospectively trying to detect issues after they have occurred (when often they cannot be rectified).


RCTs should be described in a well-articulated, concise, and operationally viable protocol which is tailored to be practicable given the available infrastructure in relevant settings.


Why this is important: Rather than trying to avoid all possible issues, the aim should be to identify the key issues that would have a meaningful impact on participant well-being and safety or on decision-making based on the trial results. Efforts can then be focused on minimizing, mitigating, and monitoring those issues. Such an assessment should consider the context of the RCT and what is additional or special about it by comparison with routine care. Broadly, these considerations come under four headings:

 

  • factors associated with the intervention (e.g. known and potential adverse effects; comorbidities or concomitant medications that might impact safety; special requirements for administering the intervention)

  • factors associated with evaluations required to answer the study objective that would not be expected in usual care (e.g. additional invasive investigations) 

  • resource implications (e.g. need for specialist imaging or laboratory assays; unfamiliar or novel procedures requiring additional training)

  • ethical implications (e.g. access to medical records and sharing of health information with pharmaceutical companies, researchers, or regulators)

Such an assessment process can then be used to guide the development of error mitigation approaches such as standard operating procedures, training, and trial monitoring. Trial processes that add scientific or ethical value to RCTs should be prioritized, and those that do not, or where the additional complexity outweighs the benefit should be avoided.


Rational monitoring


The purpose of monitoring is to identify issues that matter (important deviations from the protocol or unanticipated issues that threaten to undermine the quality of the trial) and to provide an opportunity to further improve quality (e.g. through modifications to the protocol and procedures, training and mentoring staff, or provision of information to participants). The nature and frequency of any trial monitoring plan should be proportionate to any identified risks.


Regulatory, auditing or inspection requirements should be proportionate and sensitive to the scientific and ethical qualities and objectives of an RCT and should recognise the opportunity-cost of and avoid discouraging RCT activity by setting irrelevant or disproportionate requirements.


Why this is important: Rational monitoring focuses on the issues that will make a material difference to the participants in the trial and the reliability of the results (e.g. trial recruitment, adherence to allocated intervention, blinding, and completeness of follow-up). It informs corrective actions, supports staff, and enables improvements. It is import not to confuse more documentation for better quality. Excessive monitoring and focus on details that do not have a material impact on trial quality wastes resources and creates distraction.

6.   Good RCTs have appropriate trial governance 


Key Message: RCTs should be subject to sufficient scrutiny to support delivery of an informative, ethical and efficient study, and to avoid, correct, or mitigate problems. The integrity of the trial results should be protected by ensuring that decisions about trial design, delivery and analysis are not influenced by premature access to unblinded information about the emerging results.


Why this is important: Effective and efficient governance helps to maintain the scientific and ethical integrity of a trial and advise on appropriate courses of action. Trial governance and management should build on the diverse strengths and capabilities of those involved. It should be structured to enable a reasonably consistent approach to trial delivery and effective response to issues that may arise, particularly when multiple organizations are involved.


Governance approaches should seek balance between activities that improve the trial and protect those involved with activities that might impede participants and communities from benefiting from an effective intervention or prolong the time an ineffective or hazardous intervention is used. Prolonged or excessive governance activities, which drive up unnecessary costs, deter trial designs of sufficient size or duration, or discourage clinicians and participants from being involved should be avoided.


Membership of trial governance structures should reflect the expertise necessary to scrutinise key roles, responsibilities and risks. The need for a member or a component of the governance structure to have independence from trial sponsorship and management should be determined in by assessing the risk that judgement and advice could be materially influenced by the relationship.



7.   Good RCTs use a proportionate approach to clinical safety


Taking a proportionate approach


Key Message: RCTs should consider not only the safety of individual participants in the trial, but also the safety of other parties receiving the intervention now or in the future. Potential safety issues should be considered alongside potential benefits of the intervention and in the context of the clinical circumstances.


Why this is important: All health interventions, including those used in routine practice, have the potential to cause unwanted effects. Not every unwanted health event that happens in a trial is due to one of the interventions; all individuals involved in a trial may suffer unwanted health events that have nothing to do with the trial or the interventions being studied. Many trials deliberately study individuals with disease - the less healthy the participants, the more likely that any unwanted health event is related to factors other than the intervention.


The safety and benefits of health interventions may have different impact or frequency, may have different time courses, and may occur in particular groups of individuals. For example, any real or potential early hazards of an intervention (e.g. chemotherapy, surgery) should be considered in the context of any potential benefits (e.g. symptoms, survival). It should also be recognised that for many interventions, the benefits may not be obvious on an individual patient basis (you never see the cancer, stroke, or infection that was prevented). Monitoring and evaluation of clinical safety (for both individual participants and overall) should therefore be tailored to the trial population and to what is already known about the effects of the interventions. Such approaches may be modified as new information emerges (e.g. from other trials or clinical studies in the relevant population).


Information on hazards of the intervention learned from the clinical trial both while it is ongoing and when it is completed, may inform the way it is used and safety is monitored. By tailoring the protocol to account for events occurring in a trial population which would be expected to occur based on what is already known about the interventions and the population involved, it is possible to specify not only events which might require collection, rapid safety assessment, and expedited communication to regulators and others, but also those that would not. Excessive reporting creates distraction from genuine signals of potential hazards.


Managing the safety of individual participants in the RCT


Key Message: Management of important safety issues for trial participants requires proportionate processes for identifying relevant events, evaluating and intervening, and assessing the evolution and outcome. In some circumstances it may be appropriate to exclude certain types of participant from a trial if the likely risk is excessive (compared with potential gain) and cannot be mitigated by reasonable clinical monitoring strategies.


Why this is important: The procedures used to detect, investigate, and respond to unwanted health events should be shaped by what is already known about the effects of the intervention from previous research or usage, as well as the background epidemiological and clinical features of the intended trial population (e.g. their demographics, comorbidities, and concomitant intervention).

 
Evaluation of safety signals from within a RCT


Key Message: RCTs provide a fair test of whether signals seen among those allocated to receive a health intervention are significantly more frequent than in the control group. Individual event reports (on their own) are only informative if the event is rare and the probability that it is causally related to the intervention is large.


Why this is important: The principles of good RCT design and analysis (see Principle #1) apply to evaluation of safety as to evaluation of efficacy. The strength of RCTs is that there is a randomized control group with which to compare the incidence of all health events. As a consequence, it is possible to distinguish those events that are causally impacted by allocation to the intervention versus those that are part of the background health of the participants. To detect moderate hazards of an intervention on common events (which typically represent the bulk of trial event data), safety evaluation should involve comparison of events (regardless of whether they are believed to be related to the intervention) among those who were randomly allocated to the intervention versus those who were randomly allocated to control. A moderate adverse effect on a common outcome may well be more important from a public health perspective than a big effect on a rare outcome.


Single event reports in isolation are rarely informative due to the lack of a comparison of the incidence of the event in treated and untreated subjects. The exceptions are events that are rare in the types of people involved in the trial but known to be potentially strongly associated with particular interventions (examples include anaphylaxis, bone marrow failure, and Stevens-Johnson Syndrome in association with drugs).


It is critical to balance the nature of safety data collection against the potential value. The extent to which clinical events are detected and recorded should be tailored to the trial question and the level of existing knowledge about the background health condition and the intervention being studied. Likewise the means by and level of detail with which this is done. The risks of over-burdening those involved in RCTs with collection of excessive detail (e.g. narrative description) for events is that it distracts attention from those aspects that have greatest consequence for the individuals in the trial and reduces the scale of what is achievable (so that the randomised comparisons are more likely to miss important safety issues that would influence the use of the intervention in some or all future individuals).


The use of an independent Data Monitoring Committee allows randomised comparisons of potential safety issues to be made during an ongoing RCT and considered alongside emerging data on efficacy without prematurely unblinding any others involved in the design, conduct, or governance of the trial. Thus the DMC can judge the relevance of safety information without damaging the integrity of the trial results (which inform the ongoing care of all trial participants and of future patients).


Evaluating and responding to potential external safety signals 


Key Message: During an RCT, information may emerge about the possible safety and efficacy of the interventions. Examples of sources of new information include updates to documentation for a particular drug or intervention (e.g. Summary of Product Characteristics or Investigator Brochure), information from scientific publications and scientific meetings, or communications from external parties. An assessment should be made to determine whether the new information:

 

  • Alters the overall risk-benefit balance of the trial;

  • Alters the standard of care;

  • Affects all trial participants or only certain participant sub-groups;

  • Raises any concerns regarding other therapies or procedures used in the trial or by the trial population (e.g. drug-drug interactions)

 

Depending on the findings of this assessment, necessary action may range from no changes being made to the trial, to a change to the trial protocol or other trial procedures (such as refinement of trial eligibility criteria or introduction of additional safety blood monitoring), cessation of intervention in particular individuals, or to the need to temporarily suspend or permanently cease the RCT. The DMC should be informed of any significant new information and any consequent proposed action. 


Why this is important: The evaluation of safety of an intervention should take into account all available sources of information and take proportionate action. Information from external sources may be stronger or weaker than the evidence from within the RCT itself and may need to be judged in the context of emerging information of benefits. Hence important new safety information should be provided to the Data Monitoring Committee (if there is one) for consideration in the broader context of the RCT data. 


Communicating new safety information


Key Message: The timing and nature of provision of new safety information (whether generated from within the trial or from external sources) should be determined by the nature of the safety issues and the action required. Individual reports of previously unrecognised fatal or life-threatening issues that are highly likely to be related to the intervention should be rapidly communicated to all concerned parties. Other information may be amalgamated into periodic reports (e.g. an annual report) with an evaluation by a Data Monitoring Committee or may be included in the final study report where it can be set in the context of information on any benefits from the intervention. Reports should be distributed to relevant parties (e.g. participants, trialists, and regulators) in a format and timing that is commensurate with the action that is likely to be needed.

Why this is important: The purpose of providing safety updates to others is to avoid or minimise potential harm to participants in this or other RCTs and the wider patient group or public. Hence reports should be informative, timely and actionable. Excessive, irrelevant or uninformative reports (particularly of individual cases) distract attention from those that require action. It is often more helpful to produce and circulate more comprehensive contextualized periodic updates to make them maximally informative and keeping the focus on safety issues that matter.

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Glossary (coming soon)

Appendix - Guidance Development

 

Good Clinical Trials Collaborative Secretariat

  • Prof Sir Martin Landray | Senior Lead

  • Nick Medhurst | Policy & Advocacy Manager

  • Charlie Rowley | Project Manager

  • Nana Shimosako | Project Manager

  • Alice Barney | Coordinator 

  • Aisha Mazhar | Officer

  • Isabel Smith | Officer

with special thanks to

  • Dr Christina Reith | Nuffield Department of Population Health, University of Oxford

Good Clinical Trials Collaborative Steering Group

  • Dr Amrit Ray| Bain Capital Life Sciences

  • Dr Dorcas Kamuya| KEMRI-Wellcome Trust Research Programme

  • Dr Ian Hudson| Bill & Melinda Gates Foundation

  • Sir Jeremy Farrar| | Wellcome Trust

  • Jim Elliott | UK Health Research Authority

  • Prof Kathryn Maitland | KEMRI-Wellcome Trust Research Programme

  • Mimi Darko | Food And Drugs Authority, Ghana

  • Nancy Roach | Fight Colorectal Cancer

  • Dr Rob Califf | Verily Life Sciences and Google Health

  • Dr Soumya Swaminathan | World Health Organization

  • Dr Fergus Sweeney (Observer) | European Medicines Agency

  • Janet Woodcock M.D (Observer) | U.S. Food & Drug Administration

Acknowledgements

In acknowledging the contributions of the individuals named below, we do not claim their endorsement of the draft or final guidance but wish to express our gratitude for having volunteered their expertise and time towards its development.

  • Adam Cohen MD PhD | Leiden University Medical Centre

  • Prof Alan G Fraser | Cardiff University

  • Dr Amalee McCoy | Independent

  • Prof Anders Bjartell | Skåne University Hospital and Lund University

  • Andrew J. Farmer | NIHR Health Technology Assessment Programme

  • Ann Meeker-O’Connell | formerly Vertex Pharmaceuticals

  • Anne S Lindblad PhD | formerly The Emmes Company LLC

  • Barbara E. Bierer MD | Brigham and Women's Hospital, Harvard Medical School

  • Barbara Farrell | UK Trial Managers' Network

  • Briggs W. Morrison MD |Independent

  • Mrs Carol Knott |Nuffield Department of Population Health, University of Oxford

  • Catherine Wandie | Matchboxology 

  • Mr Chris Hayward | Peninsula Clinical Trials Unit, University of Plymouth

  • Prof Christian Gisselbrecht MD | European Hematology Association

  • Prof Christoph Wanner | European Renal Association-European Dialysis and Transplantation Association

  • Daniel I. Sessler | Michael Cudahy Professor & Chair, Department of Outcomes Research, Cleveland Clinic

  • Dean A. Fergusson | Ottawa Hospital Research institute

  • Dr Deepa Pandit | Centre for Mental Health Law and Policy, Indian Law Society, Pune

  • Della Ogunleye | Independent Lay Adviser

  • Denis Lacombe MD | European Organisation for Research and Treatment of Cancer

  • Devi Leena Bose | International AIDS Vaccine Initiative (IAVI)

  • Evelyne Kestelyn | Oxford University Clinical Research Unit (OUCRU)

  • Ezekiel J. Emanuel MD PhD | Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania

  • Dr Frank E. Baiden | Department of Clinical Research, London School of Hygiene and Tropical Medicine

  • Georgina McMasters | Independent Lay Adviser

  • Mrs Gillian Booth | University of Leeds

  • Hamid Moradi MD FASN | Associate Clinical Professor

  • Prof Henry Mwandumba | Malawi Liverpool Wellcome Clinical Research Programme

  • Ian Roberts | LSHTM

  • Prof Janet Darbyshire | University College London

  • Janet Frost OBE | Independent Adviser

  • Dr Janet Messer | Health Research Authority

  • Janet Wittes, PhD | WCG Statistics Collaborative

  • Dr Jennifer Harris | Association of the British Pharmaceutical Industry

  • Jeremy Sugarman MD MPH MA | Harvey M. Meyerhoff Professor of Bioethics and Medicine, Berman Institute of Bioethics and School of Medicine, Johns Hopkins University

  • John H. Alexander MD | Duke Clinical Research Institute, Duke University

  • Prof John Eikelboom | David Braley Cardiac Vascular and Stroke Research Institute, Hamilton Health Sciences, McMaster University 

  • John Roberts | Independent Lay Adviser

  • Dr J.M. den Heijer | Centre for Human Drug Research, Leiden

  • Prof Jonathan Emberson | Nuffield Department of Population Health, University of Oxford

  • Mr Joseph Skhosana | Independent

  • Dr Kannan Subramaniam Viatris 

  • Dr Kaustubh Joag | Centre for Mental Health Law & Policy, ILS, Pune

  • M. Khair ElZarrad PhD MPH (Observer) | CDER - U.S. Food & Drug Administration

  • Maisie McKenzie | Patient and Public Involvement and Engagement Adviser

  • Marc Buyse, ScD | International Drug Development Institute (IDDI), Louvain-la-Neuve, Belgium

  • Dr Marcela Fajardo-Moser | ERA-EDTA

  • Marina Cardozo Vasco, PhD | Flutter Innovation

  • Martin Dreyling | LMU University Hospital, Department of Medicine III

  • Martin O'Kane | MHRA

  • Prof Matthew R Sydes | MRC Clinical Trials Unit at UCL, Institute of Clinical Trials and Methodology & BHF Data Science Centre with HDR UK

  • Prof Dr Michel Satya Naslavsky | Human Genome and Stem Cell Research Center, Biosciences Institute, University of São Paulo

  • Prof Michael Parker | Ethox Centre, Nuffield Department of Population Health, University of Oxford

  • Prof Morenike Oluwatoyin Folayan | Faculty of Health Science, Obafemi Awolowo University, Ile-Ife, Nigeria

  • Mrs Neetha Shagan Morar | HIV Prevention Research Unit, South African Medical Research Council (HPRU, SAMRC)

  • Nichapa Choengsamor | DSIL Global

  • Prof Nicholas J White | Mahidol Oxford Tropical Medicine Research Unit, Faculty of Tropical Medicine, Mahidol University, Bangkok

  • Nicola Perrin | Independent

  • Nittaya Phanuphak | Institute of HIV Research and Innovation (IHRI)

  • Pamela Tenaerts MD MBA | Medable (formerly Clinical Trials Transformation Initiative)

  • Patricia Kingori | Nuffield Department of Population Health, University of Oxford

  • Pichet Kamolrungsantisuk | Medical Research Network for Social, Medical Research Foundation Thailand

  • PJ Devereaux MD PhD FRCPC | McMaster University

  • Rachit Shah | Quicksand Design Studio Pvt. Ltd

  • Richard Haynes | MRC Population Health Research Unit and Clinical Trial Service Unit, Nuffield Department of Population Health, University of Oxford

  • Roger J. Lewis MD PhD | Berry Consultants, LLC, Austin, Texas

  • Prof Sir Rory Collins | Clinical Trial Service Unit & Epidemiological Studies Unit (CTSU), University of Oxford

  • Sabrina Comic-Savic | Trialspark

  • Sara Román Galdrán | European Hematology Association

  • Sarah Collen | European Association of Urology

  • Scott Williams | Independent

  • Prof Shaun Treweek | Health Services Research Unit, University of Aberdeen

  • Dr Sheuli Porkess | Actaros

  • Dr Sunita Sheel Bandewar | Vidhayak Trust;  Health, Ethics and Law Institute of FMES

  • Susana Almeida PhD |Medicines for Europe

  • Tanyaporn Wansom MD PhD | Independent

  • Ms Trishna Bharadia BA MFPM | Independent Patient Advocate

  • Prof Trudie Lang | The Global Health Network, University of Oxford

  • Prof Usha Menon | MRC Clinical Trials Unit, Institute of Clinical Trials & Methodology, University College London

  • Prof Wojciech Szczeklik MD PhD | Jagiellonian University Medical College, Krakow