Real World Evidence Flashcards
Real World Data (RWD)
Healthcare data which is routinely collected.
Comes from a variety of sources, including electronic health systems
Types and Sources of Real-World Data
Clinical Data:
-electronic health records
- case report forms (eCRFS)
Patient-generated data
-health and treatment history
- biometric data
-patient-reported outcomes (PROs)
Cost and Utilisation data
-claims datasets
-public datasets such as CMS and AHRA
Public health data
-government data sources
-national networks and centres
Real World Evidence (RWE)
Real world data collected outside of clinical trials
It can be used to contextualise randomised trials, to estimate effects of interventions in the absence of trials, or to complement trials to answer a broader range of questions about the impacts of interventions in routine settings.
Includes electronic health records, patient registries and administrative health care claims - through routine clinical practice
RWE can be used to complement evidence from controlled clinical trials.
Why do we need RCTs and RWE
Patients in RCTs are highly selected and have a lower risk profile than real-world populations, with the frequent exclusion of elderly patients and patients with co-morbidities.
Supplementing RCT evidence with data from observational settings can also improve the external validity of oncology drug trials, such that physicians treating patients in real-world setting have the appropriate evidence on which to base their clinical decisions.
Clinical Trial VS Real World
clinical trial: includes very specific patients (i.e might exclude patients with additional health conditions)
real world: includes a much wider range of patients!
NICE (National Institute for Healthcare and Excellence) –> RWE in practice
NICE Real World Evidence Framework published June 2022.
“Real-world data can improve our understanding of health and social care delivery, patient health and experiences, and the effects of interventions on patient and system outcomes in routine settings”.
“The framework aims to improve the quality of real-world evidence informing our guidance”
What points to recognising and avoid bias in clinical trials
1- Patient selection, wrong endpoints and biases that limit application of clinical trials
2- Spin the reporting of clinical trials
3- Under-reporting of harm
4- Lies, damned lies and statistics
5- External validity of clinical trials
1- Patient selection, biases that limit application of clinical trials
Patient selection
- exclusion of high morbidity patients
Broadening RCT inclusion and exclusion criteria
- collection outside trial
- observational studies
- development of large patient registries in specific disease areas
Women in clinical trials
Are women in clinical trials?
- Yes. Women are already in clinical trials. However, women from diverse backgrounds still need to participate. Women of all ages, racial and ethnic groups, and women with disabilities or chronic health conditions should think about being in a clinical trial.
Why should women participate?
- Medical products can affect men and women differently. Sometimes women have different side effects. It is important that women participate to show if products are safe and work well in both men and women.
Wrong endpoints
In clinical trials, endpoint refers to an event or outcome that can be measured objectively to determine whether the intervention being studied is beneficial. Usually included in the study objectives.
Examples:
survival,
improvement in quality of life,
relief of symptoms
disappearance of the tumour
See slide 31 for advantages and disadvantages.
Wrong endpoint:
Not well-defined, relevant, or powered
Clinical versus surrogate
Composite versus single
Patient versus tumour
Setting: Low risk vs high risk disease
Sources of bias within clinical trials
Types of bias:
1 selection bias
2 performance bias
3 detection bias
4 attrition bias
5 reporting bias
Systematic differences between:
1 baseline characteristics of the groups that are compared
2 groups in the provided care, or in exposure to factors other than the interventions(s)
3 groups in how outcomes are determined
4 groups in withdrawals from a study
5 reported and unreported findings
2 Spin the reporting of clinical trials
The dissemination of research through publications in peer-reviewed journals is essential.
Reporting bias leads to the manipulation or embellishment of data.
Spin in publications: a way of reporting to convince the reader that the beneficial effect of the experimental treatment (efficacy, safety) is higher than shown by the results.
Distorted interpretation of study results spin
Interpreting study results inaccurately can lead to biased representations. It’s important for data to speak for themselves.
Scientists or sponsors often have vested interests in trial outcomes, influencing factors such as publication timing, treatment adoption, and personal gains.
Authors enjoy considerable freedom in crafting articles, which can introduce bias into how results are presented and interpreted
3 Under-reporting of harm
In contemporary clinical trials harm is under detected and underreported by investigators
Because clinical trials are becoming more selective about who can join, it’s making it harder to use the results of those trials in everyday medical practice.
To better serve patients, oncologists, along with journal editors and professional societies, must implement measures to ensure comprehensive reporting of treatment toxicity.
4 “Lies, damn lies and statistics” (Mark Twain)
Statistics: to measure it to know; a language of making statements that is as vulnerable as any other language
The complexity:
- medical experimentation in humans is very difficult
-performance of the trial by (many) humans in ‘real life’ conditions
- bad conditions for objective assessment
- huge variability at the level of the disease, the host, the genes
- pressure: disease (human need), commercial, “professional vs lay”
- relatively poor understanding of the variability
- multidisciplinary
