Turning real world data into evidence

RWD is an overarching term for data not collected from randomised controlled trials (RCTs). RWD can be collected both prospectively and retrospectively from observations of routine clinical practice. Data collected may include clinical and economic outcomes, patient-reported outcomes and health-related quality of life.

Real world evidence (RWE) is the evidence derived from the analysis and/or synthesis of RWD.

The use and benefits of RWE are many, but first and foremost RWE allows analyses of how health care interventions work outside a controlled setting of a randomised clinical trial. RWE studies are conducted in the real clinical practice and include all the noise seen in real life. This makes the data collected a very good source of data on how an intervention performs under the stresses and complexities of normal clinical practice. In contrast to RCTs, patients and their clinicians choose treatments on the basis of the patient’s clinical characteristics and preferences. However, since the factors that influence treatment choice in clinical practice may also influence clinical outcomes, RWE studies generally cannot yield definitive causal conclusions about the effects of treatment. In contrast to RCTs, patients and their clinicians choose treatments on the basis of the patient’s clinical characteristics and preferences. However, since the factors that influence treatment choice in clinical practice may also influence clinical outcomes, RWE studies generally cannot yield definitive causal conclusions about the effects of treatment.

RWD can be sourced, collected, and analysed in many different ways. The growing digitalisation has expanded the recording of RWD resulting in more complete and detailed datasets being available for RWE studies. Before big databases and big data processing capabilities, the collection of RWD from paper medical records etc. was a very labour-intensive process. In addition, it was a lot more difficult to obtain access to these data since researchers could not avoid being exposed to sensitive personal information. Today most RWD can be accessed or generated in a completely anonymous way.

Real world data sources

RWD can be sourced from various sources including:

Patient registries: Prospectively collected data with a predetermined scientific, clinical or public health purpose. People included often have a particular condition, a certain treatment or using a health-related service.

Medical records: Data sourced from routine clinical practice in primary or secondary care, incl. clinical and laboratory data entered by the healthcare provider.

Pharmacy and health insurance claims: Collected primarily for administration and billing purposes these data include information on the use of inpatient, outpatient, emergency room and pharmacy services.

Surveys: Data collected by asking respondents directly in a standardised way by means of structured questionnaires, phone interviews, emails etc. to understand e.g. treatment satisfaction, impact on quality of life.

Social media: Sourcing from social media platforms allows an accelerated and real-time access to patient experience, adverse events signals.

Real world evidence study designs

Deciding on an appropriate design for your RWE study is a crucial step as the quality and credibility of a study are likely to have an impact on the reported result. Different research questions require different study designs.

Cross-sectional study

Data collected from a population or a representative subset of a population at one specific point of time (or over a short period of time) to examine current health outcomes. Often used for prevalence of disease and traits as well as attitudes and knowledge among patients and health care professionals. As there is no time dimension in a cross-sectional study interpretation of exposure and outcome are not easily made unless the exposure is assumed to be constant over time.

Cohort study

A cohort study follows a group of individuals over a period of time to measure various health outcomes. Well-suited for rare exposures, but difficult to identify an appropriate comparison group. Cohort studies can be both prospective, where individuals are enrolled before they develop the disease or outcome, and retrospective, enrolling individuals who already have the disease or experienced the outcome in question.

Case-control study

A study that examines associations between outcomes and prior exposures by comparing people with an outcome or exposure of interest to those without the outcome or exposure. Comparisons between groups can be influenced by confounding factors which needs to be adjusted by appropriate statistical methods.

Pragmatic trial

Pragmatic trials are essentially combining real-world evidence and randomisation. Pragmatic trials aim to measure relative effectiveness of treatment strategies in the real world clinical practice.They provide evidence of the added value of a treatment strategy in routine clinical practice, while maintaining the strength of comparisons based in randomised controlled trials.