Real-world data (RWD) and real-world evidence (RWE) are playing an increasing role in health care decisions:
- Regulators use RWD and RWE to monitor post-market safety and adverse events and to make licensing decisions.
- The NHS uses RWE to support adoption decisions and to develop guidelines and decision support tools for use in clinical practice.
- Medical technology and medicines developers can ensure that new treatments best service NHS patients by using RWD and RWE to support clinical trial designs and observational studies to generate innovative, new treatment approaches.
NHS Lothian has developed a Real World Data Analytical Service that works with public sector, academic or commercial partners to specify and conduct analysis of RWD that is part of routine care systems, within the NHS, under the jurisdiction of Caldicott Guardian approval. The Edinburgh Cancer Informatics (ECi) aspires to be a world leader in health care data for patients with cancer. Our main objective is to improve patient care via RWD and RWE. The Edinburgh Cancer Centre has collected data for over 40 years, which we are analysing to produce high impact research and support efficient health care decision making. We also provide support in the submission to government bodies allowing better and quicker access to medicine.
This article summarises many of the considerations necessary in using RWD for medicines evaluation.
Which data?
Our programme is the result of a close collaboration between the NHS and the University of Edinburgh. Additional information around the data sources can be found here. The following information is available in a coded format:
- Demographics
- Diagnosis and pathology details
- Localised Treatment Details
- Systemic Therapy Details
- Biomarkers
- Outcomes and Follow-up
- Healthcare utilisation
Patient-level data remains within the NHS Lothian secure environment throughout the study, and all analysis takes place by NHS Lothian staff. The data is de-identified prior to analysis to further protect patient privacy. The requestor does not access patient data at any point. Aggregate reports where small numbers are hidden will be released from NHS Lothian, subject to censoring for NHS disclosure control.
You can find our projects in the blog or project section of this website. We support studies investigating: time to treatment analysis, treatment patterns, survival, incidence, prevalence… Feel free to contact us for more information.
How does it work?
Our experienced team of medical oncologists, analysts, data managers, coders and research managers provide high quality work based on clinical and research expertise. We will support you through each step of the analysis, as described below:
1. Feasibility – data exploration
This is an interactive process where we would share our expertise (both clinical and analytical) and support you in the development of your study concept and feasibility. Get in touch with information about the cohort you are interested in, and we can investigate if we have enough numbers and variables to power the analysis.
2. Finance/legal work
If you approve of the above, a quote will be provided based on a cost-recovery model, with an anticipated delivery time-frame. Once approved, a contract will be put in place. Both are supported by Edinburgh Innovation, who responds to any enquiry within four days.
Information governance approval will be requested by the ECi team operating under the ‘Dataloch‘. This process is necessary to maintain patient data privacy, and is expected to last about 8-12 weeks, however times are expected to reduce in the coming year. Additional information about our unique and streamlined governance framework can be found here.
3. Analysis
A protocol, created with the requestor, can be developed. We would organise 2-3 rounds of review, and can use the requestor’s template if easier. The analysis is then performed by an NHS analyst, on an NHS computer, on the NHS IT environment. The data does not leave this set up. All our staff are trained for data protection (both NHS and MRC) as well as for the NHS disclosure policy. The requestor will not have access to the patient data.
A draft of the output (aggregated data only), will be submitted to the NHS Lothian Caldicott Guardian, via the Dataloch team, for final disclosure control as per NHS policy. This is to ensure that the data shared does not allow for patient identification, and that small numbers are hidden. The output is then shared with the requestor.
The output is a product of collaborative work and is published in the public domain for use as part of an evidence based technology development and adoption process.
Data quality assurance
With real-world studies, missing or inaccurate data are expected in line with the nature of routine electronic health records. Critical variables are identified during protocol development and data completeness will be reported for these variables.
Manual data curation can be undertaken to complete critical missing values and to quality check accuracy using a sample-based approach. Where critical missing data cannot be completed patients will be excluded from the analysis.
Our studies are conducted in accordance with the ethical principles that have their origin in the Declaration of Helsinki and that are consistent with Good Pharmacoepidemiology Practices (GPPs) and applicable laws and regulations of the country or countries where the study is being conducted, as appropriate.
If you are interested in commissioning an analysis please contact maheva.vallet@ed.ac.uk. Examples from previous analyses can be found under our blog/project section.