Randomized clinical trials (RCTs) have proven over time to be the gold standard for assessing the safety and efficacy of new drugs. However, there are a number of known deficiencies of them such as not enrolling certain types of patients (e.g., pregnant women) and their high cost (see: Computer-Simulated Tests Eyed at FDA to Cut Drug Approval Costs). One alternative to RCTs for post-market drug testing is the use of EHR databases which would seem to be an ideal tool because they provide information about drugs as well as patients' clinical signs and symptoms. Even the fact that much of the relevant EHR data is in textual format is no longer a major barrier with the recent rapid improvements in natural language processing (NLP) (see: What is the Role of Natural Language Processing in Healthcare?).
However, EHR data for drug assessment has not been shown to be practical because such systems were designed to support physicians and nurses in the delivery of healthcare and not as research tools or for post-market drug surveillance. Even those health systems with clinical data repositories or data warehouses that are designed for analytics have not been harnessed for drug surveillance. In response to this need, a new approach has been suggested called using real world evidence (RWE) to assess drug performance and safety (see: Real world evidence and pharma). Below is an excerpt from the article:
Randomized...[clinical] trials (RCTs) have long been the gold standard for assessing the efficacy and safety of new drugs, but there is an increasing desire among stakeholders for evidence of how drugs work in the real-world, not just within the confines of a highly controlled experimental environment. No longer is Food and Drug Administration (FDA) approval the ultimate goalpost. As drug prices rise, so resistance becomes more entrenched from government, payer organizations and the patients themselves. Stakeholders now want real world evidence (RWE) of a drug’s true performance and effectiveness before they make final decisions about pricing and payment. ....[R]eal world data is any data outside the constraints of conventional RCTs to evaluate routine clinical practices. This information, harvested organically in the real world, can then be assessed in terms of its organization, analysis, and interpretation to enable collectors to generate RWE....One advantage to RWE is that it can be collected in different ways.The most common way is to analyze databases and electronic patient record platforms....Another way of collecting RWE is to undertake ad hoc or tracker surveys with patients from online patient panels....These questionnaires can also provide valuable insights to pharma into particular diseases and how patients react to their medications, all of which can assist with interventions to improve the effectiveness in chronic disease self-management through better adherence to medications.
I believe that the use of online patient panels with data input using smart phones may have great utility in the future for RWE analysis of post-market analysis of drugs. The most interesting point in the excerpt above is the suggestion that FDA approval may no longer the "ultimate goalpost" for a drug. Payers and patients are demanding additional evidence in support of the value of the new, and always pricey, drugs. On paper, RWE would seem to be one way to provide concrete evidence of a drug's value in addition to clinical trials. However, the idea of what constitutes persuasive RWE evidence has not yet been established. Moreover, I have a concern that the term "real world evidence" would have great marketing appeal for pharmaceutical companies and that they might use this technique in a way that serves their interests but without proven scientific validity.