Diabetes is a chronic, long term condition that affects many people’s lives around the world. Despite effective pharmacotherapy and national programmes designed to tackle this condition, such as the UK’s NHS Long Term Plan, defining and navigating a successful care pathway remains a challenge for many patients. In that context, how could real world data underpin research and improve patient outcomes?

Diabetes: a public health burden

In 2019, an estimated 3.8 million people were living with diabetes in the UK, more than dementia and cancer combined. Approximately 90% of individuals with diabetes are diagnosed with Type 2, wherein a person becomes resistant to insulin or a has a reduced capacity to produce insulin and is, therefore, unable to lower blood glucose levels naturally. The remaining 10% of individuals have Type 1 diabetes; a genetic disorder affecting the ability of the pancreas to produce insulin. Diabetes UK reported that 500 people with diabetes die prematurely each week. This puts diabetes at the forefront of public health.

Successful care remains a challenge

People with Type 1 diabetes require exogenous insulin. Although Type 2 diabetes can be partially managed through lifestyle changes, most people will require pharmacological therapy at some point. Metformin has provided first line therapy for diabetes since 1958 , but some people will be prescribed additional or alternative pharmacologic intervention to control blood glucose. Some of these therapies stimulate the pancreas to produce more insulin (e.g., sulphonylureas and glucagon-like peptide-1 [GLP1] receptor agonists), some allow for prolonged duration of insulin in the blood (e.g., dipeptidyl peptidase-4 [DPP-4] inhibitors), some improve insulin sensitivity (e.g., thiazolidinediones) while others reduce the amount of sugar in the blood (e.g., sodium-glucose cotransporter-2 [SGLT2] Inhibitors). Some people will need exogenous insulin if their body cannot produce enough. The NHS spend approximately £40 million/week on drugs to manage diabetes.

High blood glucose affects multiple organs; leading to not only complications of diabetes but also exacerbating other conditions, the costs for which are significant. Individuals with diabetes are known to be at a higher risk of numerous other diseases including skin infections, eye problems (e.g. glaucoma), neuropathy and gastrointestinal disorders, as well as being more likely to experience complications of, and more severe outcomes to, other diseases such as infections or cancers. The cost of medications for complications is estimated to be around 3 to 4 times the cost of medication for diabetes alone and people with Type 2 diabetes were found to incur higher lifetime costs than people with obesity. Due to the high cost to the NHS of escalating treatment and complications of unmanaged Type 2 diabetes, it is important to focus on early interventions, as highlighted by the NHS diabetes prevention plan which aims to refer those at high risk of Type 2 diabetes to behaviour change programmes.

Tackling diabetes: the promise of insights from real world data…

Identifying people who would benefit most from earlier interventions – behavioural and therapeutic - and observing the long-term (cost) effectiveness of these (e.g. effectiveness of metformin vs SGLT2 on cardiovascular outcomes) on outcomes that matter to people with diabetes, people at risk of diabetes and healthcare professionals, could be used to inform patient management, eventually reducing the overall burden of diabetes. Such data is difficult to reliably capture in a traditional clinical trial due to the administrative and patient burden of protocol-required site visits and relatively short-term follow-up. Similarly, the restricted population of clinical trials is not always representative of the nearly 4 million people living with diabetes in the UK. Real-world data (RWD), exploring patterns of prescribing, treatment persistence and treatment outcomes can provide important insights.

...bolstered by the UK’s unique health data ecosystem

The UK presents a particularly useful landscape for conducting studies using real world data. The National Health Service (NHS) provides healthcare available to all, free at the point of use. The entry-point for a person to receive healthcare (beyond accidents and emergencies) is through primary care, i.e., a general practitioner (GP) allocated to the person depending on their area of residence. GPs both diagnose and manage diabetes with their patients, and since 2004, the Quality Outcomes Framework has provided incentives to ensure that GPs “maintain a register of all patients aged 17 or over with diabetes mellitus, which specifies the type of diabetes where a diagnosis has been confirmed” . This means that clinically diagnosed individuals (usually through glucose evaluation) are recorded in their electronic health records, as are their repeated blood glucose test results. The data collated allows for generalisable and representative population studies.

IQVIA Medical Research Data incorporating data from THIN, a Cegedim Database, includes non-identified electronic patient health record data from over 18 million patients collected from UK GP Practices, including the data collected for patients with diabetes. A 2017 study using IQVIA Medical Research Data successfully demonstrated that liraglutide (once-a-day use) was more effective than lixisenatide (twice daily use) at helping patients with Type 2 diabetes achieve glycaemic control. More recently, IQVIA Medical Research Data has been utilised in the ATTAIN study which demonstrated that therapy intensification significantly reduced HbA1c levels after 6 months. This is of particular importance since almost half of individuals with Type 2 diabetes fail to reach the recommended HbA1c levels and require therapy intensification.

While clinical trials are still considered the ‘gold-standard’ of medical research; particularly in identifying causal relationships, they can over-estimate the effects of therapies when compared to their use in the wider population and may not include information pertinent to patient and healthcare professional decision-making. Persistence and adherence to newer classes of Type 2 diabetes treatments in clinical practice has been lower than anticipated from clinical trials and data from sources such as IQVIA Medical Research Data can help us understand why this may be. Multiple aspects of care are known to be important to people with Type 2 diabetes, including: number of healthcare professional visits, risk of hypoglycaemia and other adverse treatment effects, amount of engagement in self-management activities, and quality of life. Evidence suggests that these are, in potentially multiple ways, associated with diabetes medication adherence. Some of these data are readily available in real world data, others can be collected, but should be explored to understand the patients’ experience with treatment, the holistic benefits of treatment on outcomes beyond glycaemic control, and to inform clinical care decisions.

In conclusion

Real world data can help us understand many aspects of diabetes care and early interventions that cannot be understood from regular clinical trials, including intervention decisions/patterns, long-term outcomes, adherence and persistence, rare adverse treatment effects, and interactions with comorbidities/concomitant medications. Further, the size and scope of real world databases like IQVIA’s IQVIA Medical Research Data allow exploration of large, heterogenous groups of people with diabetes to understand systematic differences among populations, many of which will have been actively excluded from clinical trials. To better understand the true effects of behavioural and therapeutic interventions for diabetes, we need both clinical trials and real-world observations to provide evidence from which to derive policy and healthcare decisions.

For further information, please contact james.philpott@iqvia.com