University of Birmingham · FACULTY OF PSYCHOLOGY

13344 Chap.9 Trial Phases, CONSORT and Statistics in RCTs

- one subject, every graph, every model, every mark
6 Chapters3-page Bible
Our own words - no uploaded lecturer files
Updated for this semester
Chapter 9 of 10 · 13344

Trial Phases, CONSORT and Statistics in RCTs

From first dose to market

The chapter sets out the four phases of pharmaceutical trials, the features of complex interventions and the trial designs the module names, from crossover to stepped wedge.

Reporting and numbers

It explains the CONSORT 2010 checklist and four-stage flow diagram, the inputs to a sample size calculation, how the module reads p-values and confidence intervals, and how to compute and interpret the number needed to treat.

Calculations to practise

A worked example computes a number needed to treat from remission rates, and practice items read a published hazard ratio and its confidence interval and identify what reduces the sample size a trial needs.

In this chapter

What this chapter covers

  • 01

    Phases I to IV

  • 02

    Complex interventions

  • 03

    Crossover, factorial, cluster and stepped wedge designs

  • 04

    CONSORT checklist and flow diagram

  • 05

    Power and sample size

  • 06

    Number needed to treat

Worked example · free

Match four studies to their trial phase

Q [4 marks]. AskSia-authored practice. Assign each study to a phase: (a) monitoring adverse effects in people prescribed a licensed antidepressant; (b) first doses of a new compound in volunteers to set a safe dose; (c) a comparison of the new compound against an established drug; (d) a first test of whether the compound has any beneficial effect. The marks shown are an AskSia study allocation, not a University marking scheme.
Four phases as a rising staircase.Four phases as a rising staircase. Each step is one phase from left to right, labelled with the question it answers; a drug only climbs to the next step once the earlier question has a satisfactory answer, and monitoring continues after it reaches the market. Phase I dose, side effects Phase II first efficacy Phase III against placebo or drug Phase IV after marketing each step builds on the last
Four phases as a rising staircase. Each step is one phase from left to right, labelled with the question it answers; a drug only climbs to the next step once the earlier question has a satisfactory answer, and monitoring continues after it reaches the market.
  • 1(b) is Phase I: first use in people, to establish dosing and side effects.
  • 1(d) is Phase II: initial evaluation of efficacy.
  • 1(c) is Phase III: the treatment effect compared with placebo or an existing treatment.
  • 1(a) is Phase IV: post-marketing surveillance of adverse effects.
In order of the staircase, the studies run b, d, c, a. Each phase answers a narrower question than the one before, so a proposal should state which question it addresses and choose its sample and outcome to match. Placing a new trial on the staircase before writing anything else fixes its sample, comparator and outcome.
Sia tip — Ask whether the study is about dose, any benefit, comparative benefit or long-term safety.
Glossary

Key terms

Stepped Wedge Design
A Stepped Wedge Design gives every unit the intervention eventually, in a randomised, staggered order.
Statistical Power
Statistical Power is the chance that a trial detects a true effect of the target size.
CONSORT
CONSORT is an evidence-based reporting standard for randomised trials, made up of a checklist and a flow diagram.
Crossover Design
A Crossover Design gives each participant more than one condition in turn, so that each acts as their own control.
FAQ

Trial Phases, CONSORT and Statistics in RCTs FAQ

What are the four stages of a CONSORT flow diagram?

Enrolment, allocation, follow-up and analysis. Each stage reports numbers and reasons: how many were assessed and excluded, allocated and treated, lost to follow-up, and analysed or excluded from analysis, for every arm of the trial.

What goes into a sample size calculation?

CONSORT lists the clinically important target difference, the alpha or type I error level, the power or type II error level and, for continuous outcomes, the standard deviation. The module gives standard power as 80%, and alpha is usually set at 0.05.

How do I interpret a number needed to treat?

It is the number of people who must receive the intervention for one additional person to benefit, calculated as one divided by the risk difference and rounded up. Smaller values mean a more effective intervention, and a negative value signals harm.

What is a complex intervention?

An intervention with several interacting components, such as different people delivering it, several organisations or levels, people with complex needs and a range of outcomes. Evaluating it means testing across sites, using varied outcomes and asking who it works for, since a null result may reflect poor implementation.

What does a confidence interval that includes 1 mean for a hazard ratio?

It means the data are compatible with no difference between the arms, so the trial has not shown that one is better. That is not the same as showing they are equal, which would need a trial designed to test equivalence.

What trial designs does the module name besides individually randomised trials?

Crossover designs, where each person acts as their own control; factorial designs, which test interventions alone and combined; community or cluster designs, which randomise units such as GP practices or schools; and stepped wedge designs, where every unit eventually receives the intervention in a randomised order.

Study strategy

Assessment move

Redraw a CONSORT flow diagram from a published trial, then compute an NNT from its main result. Learn the four phases as four questions rather than four labels. Practise reading the results section of a trial: find the primary outcome, its effect size and confidence interval, and the absolute measure if one is given.

Working through Trial Phases, CONSORT and Statistics in RCTs in 13344? Sia is AskSia’s AI Psychology tutor — ask any 13344 Trial Phases, CONSORT and Statistics in RCTs question and get a clear, step-by-step explanation grounded in how 13344 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.

A+Everything unlocked
Unlocks this Bible + all 17 of your subjects at University of Birmingham, plus 1,000+ more Bibles across the AskSia library.
Sia - your 13344 tutor, unlimited, worked the way the exam marks it
The full 3-page Bible + practice bank with worked solutions
Chrome extension - sync your LMS so Sia knows your deadlines
Bilingual EN / Chinese on every Bible and every Sia answer
$0.99 Trial
30-day money-back · cancel in one tap · how it works