How real-world data could improve clinical decisions

Traditionally, physicians have relied heavily on clinical evidence—most notably from randomized and controlled trials—to guide their medical decision making.

Today, however, clinicians have access to a wealth of additional data generated from sources outside traditional research settings, such as remote monitoring devices, wearables, disease registries, electronic health records, and claims and billing activities.

While evidence-based data from clinical trials remains the bedrock for clinical decision making, clinicians now have the opportunity to enhance the process by incorporating findings from real-world evidence. A recent New England Journal of Medicine opinion column proposes that real-world evidence derived from an analysis of real-world data can effectively complement “the knowledge gained from ‘traditional’ clinical trials, whose well-known limitations make it difficult to generalize findings to larger, more inclusive populations of patients, providers and healthcare delivery systems or settings reflective of actual use in practice.”

An evidence-based approach to medical decision making emphasizes the use of evidence from well-designed and well-conducted research. While evidence-based practice is widely recognized as the “gold standard” for the delivery of safe and compassionate healthcare, clinical trials may not reflect regional variations in a population or may include a disproportionate percentage of subjects of a particular age, race or gender. Trials may also intentionally exclude subjects with comorbidities or individuals taking commonly prescribed medications.

Physicians are aware that patient populations do not always align well with clinical trial subjects and often adjust their prescribed therapies accordingly. For example, a medication that’s been proven effective for white male patients may not be the best therapy for female patients of Asian descent. A physician may not have a formal, structured study to prove this finding, but may have arrived at this conclusion based on observed outcomes within his or her patient population.

When individual observations can be aggregated with the help of an analytics engine, doctors can assess the benefit or risks of therapies not addressed in evidence-based guidelines

Although physicians routinely vary care based on their own observations, a better option would be to equip clinicians with technology that systematically considers evidence and variations based on both real-world data and data from structured research.

In this paradigm, EMR data, claims information, patient-generated health data, prescription refill patterns, and similar patient and population details could be combined with traditional evidence-based data from clinical trials. With the addition of an analytics engine, all the available information could be analyzed, and patient-specific therapy options could be generated.

Consider, for example, a patient who is experiencing heart failure. Evidence-based guidelines recommend a specific medication, such as a thiazide diuretic. However, pharmacy records indicate that the patient has stopped picking up his prescriptions. Individually, a patient may not be taking the medication based on a previously undisclosed side effect.

However, a real-world analytic engine can detect that a cluster of patients with similar characteristics (age, gender and location) is experiencing this side effect, and can alert both other providers, who can then prescribe a different medication to improve adherence among similar patients, and the pharmaceutical company, as part of a post-marketing surveillance program.

The aggregation of data from real-world therapies and outcomes can also help providers identify specific practices that may be more effective within a localized region. For example, an analysis of EHR and claims data might reveal that patients recovering from heart surgeries experience better long-term outcomes at a particular rehab facility, while patients recovering from orthopedic procedures have superior outcomes at an entirely different rehab facility.

Similarly, the analysis of aggregated data could help providers pinpoint specific lifestyle changes that have benefited patients with certain medical conditions, such as COPD. Armed with evidence of better outcomes, physicians are more likely to encourage other patients with COPD to make similar lifestyle changes to improve their health.

By giving physicians access to point-of-care technology that assesses real-world data with evidence-based data, clinicians can augment their decision-making abilities and prescribe therapies that are better tailored to the needs of individual patients and their localized patient populations. By leveraging insights from all available evidence, physicians are empowered to drive superior clinical outcomes and better population health.