Generate real-world evidence

Conventional randomised controlled trials (RCTs) alone may not provide sufficient evidence of relative effectiveness to support reimbursement decision-making. An estimation of how well a medicine may work in the real world can be estimated from analyses of the existing RCTs. However, it may be possible to generate ‘earlier’ estimates of the relative effectiveness of the new medicine of interest in time to inform reimbursement decision-making by analysing existing real-world data sources or by conducting new studies to generate real-world evidence (RWE). For more information about the limitations of RCTs to estimate relative effectiveness see here and here, for an overview of methods for predicting relative effectiveness in the real world using RCT data see here, and for more information about existing real-world data sources see here.

Some experimental and observational study designs that could provide RWE are summarised in the table below. While some study designs may provide evidence on relative effectiveness, some ‘more epidemiological’ observational studies may not be able to provide evidence of relative effectiveness. However, they may be useful to define the disease area and understand the natural disease progression, or provide information about a relevant comparator if there is no comparative data.

Since the quality and credibility of a study may have a significant impact on the reported effect of a medicine and its interpretation, it is crucial to assess each study individually, whether or not it includes an element of randomisation. For more information about assuring quality and credibility of RWE, see here.

Table. Study designs that may provide RWE

Experimental study designs
Pragmatic RCT A pragmatic trial aims to measure the relative effectiveness of treatments in real-world clinical practice. It combines the strength of RCTs with evidence of the added value of a treatment strategy in routine clinical practice. Read more
Population enrichment RCT A population enrichment RCT includes patients typically excluded from RCTs combined with predictive modelling techniques to better predict relative effectiveness in a real-world population. Read more
Cohort multiple RCT (cmRCT) (also known as or trials within cohorts) cmRCTs are a type of pragmatic RCT that use a large cohort of patients as a source of participants for multiple RCTs, providing a more generalisable study sample. Read more
Comprehensive cohort study (CCS) CCS is a type of pragmatic RCT that includes participants who do not consent to be randomised to the treatment group. This reduces selection bias and improves generalisability. Read more
Cluster RCT Cluster RCTs randomise groups or clusters rather than individual participants as in traditional RCTs. Read more
Non-randomised controlled trial Any experimental study allocating participants to different treatments using a method other than randomisation, such as clinician or patient preference.
Observational study designs
Cohort A cohort study follows a group of individuals over a period of time to consider associations between interventions received and outcomes.
Case-control A study that examines associations between outcomes and prior exposures by comparing people with an outcome of interest to those without the outcome. These are not often used for interventions.
Cross-sectional In a cross-sectional study, data are collected from a population or a representative subset of a population at one specific point of time or over a short period to examine associations between the outcomes and exposure to interventions.
Controlled before-and-after Similar to a case series, in which observations are recorded on a series of individuals before and after receiving an intervention, but this study design includes a control group.
Case series, interrupted time-series or before-and-after Recorded observations on a series of individuals before and after receiving an intervention. Does not typically provide evidence of relative effectiveness.
Case report A detailed report on an individual patient, typically describing symptoms an unusual or new occurrence, including outcomes after a treatment. Does not provide evidence of relative effectiveness.