In the last couple of week’s blog posts, Should Reporting Adverse Events be Mandatory? and Is the Status Quo Killing Healthcare?, Brian raised very important questions that sparked a lot of discussion in various LinkedIn forums. Here are links to some of the conversations in case you want to join in:
(Also, as a shameless plug- we have launched our own LinkedIn group, POST APPROVAL DRUG MONITORING, and would love to invite our blog readers to join as the inaugural members)
We believe, like so many others who chimed in the blog discussions, that the health system must change dramatically, and that patient care and outcomes need to be the central focus. In order for this to happen, more data needs to be gathered, tracked and analyzed to ensure that the treatments and paths of care are effective and necessary. If data submission and gathering is not mandatory, then actionable steps need to be taken to ensure that there is a concerted effort to marshall resources to encourage the daily discovery and reporting of pertinent drug safety information.
How we propose to affect this change and increase drug side effect reporting:
1. Give healthcare providers access to the FDA Adverse Event Reporting System (FAERS) – the dataset that adverse events are reported into.
The FDA publishes FAERS via complicated data file downloads on its website and through the new OpenFDA initiative. However, both datasets are virtually unusable, especially by someone without relational database programming skills and the ability to create complex algorithms to clean and standardize data. And to worsen matters the data is old -- right now the most recent data published in the file downloads is through 3/31/14.
FDA should make public (really public… not how it currently is public) the data that it is using to make decisions, not the aforementioned complex, dirty data but the actual same FAERS records it is using to make regulatory decisions. It is unacceptable that one needs to file time consuming and resource draining Freedom of Information Act requests to see those data.
Healthcare providers care deeply about how medical decisions are made. They want to see the data. They want the ability to see exactly why actions are being taken. By seeing the full dataset they can know that at the very least their report didn’t just go into some black hole at FDA.
There are two ways to approach this problem. 1) Wait for the government to remedy the situation or 2) look to the private sector.
Although we may be biased, we believe that the benefits of looking to the private sector to address this issue outweigh the small economic costs to the healthcare system, and will lead not only to free market innovation but to improved outcomes and ultimately, a lower total cost burden.
2. Make the dataset easy to use, interpret, and make it actionable.
One of the main reasons that FAERS has not been used outside of the FDA and pharmacovigilance departments is that it has posed an interpretive nightmare for those that are not immersed in pharmacovigilance daily practice. A doctor, pharmacist, or any other healthcare provider on the front lines of patient care doesn’t have the time to pour over thousands of case reports and conduct statistical analyses on site. But this where the promise of Big Data and Big Data Analytics for the health industry has come to fruition.
Using logic, math, and software, the FAERS dataset can be liberated from its pharmacovigilance prison. So now, if a provider was unsure whether a drug/AE pair are related, she can quickly see trends, scores, and other statistics based on the reports of her peers. She can then be more comfortable that reporting the AE is worth her time and will know that she is contributing to a shared knowledge set that will advance patient care.
At the end of the day, data and analytics are only as good as the insight that they can provide. And even with the best math, logic, and software healthcare providers simply can’t arrive at all of the insight possible on their own. Best clinical practice is shared via peer reviewed research and other third party analysis that providers regularly rely on.
Our friends at the not-for-profit, ISMP publish a quarterly newsletter called Quarterwatch, which provides a deep dive into specific medications that they are tracking for safety concerns. It is a fantastic use of the data, and if you haven’t read it before I suggest that you explore a subscription.
But there are also numerous use cases where analytics can be used to help provide context. They can be used to compare all newly approved drugs to the post-approval experience of like drugs on the market, explore the evidence behind label changes, and help to support other research that is being done.
3. Quantify the monetary impact that reporting has on the healthcare system
For better or worse, in order to affect change it often takes more than just improved patient outcomes. You need to show economic benefit. It is this quest for ROI that has created health economics outcomes research (HEOR). Pharmaceutical manufactures spend millions of dollars every year sponsoring research that is designed to convince their customers (managed care organizations) to buy their drugs. They do this by demonstrating the economic superiority of those drugs to what is already on the market.
The problem with most outcomes research is the data that goes into it. It is manufacturer supplied. It is viewed as biased and it is not independently verified.
Post-approval safety data that are reported by healthcare decision makers can be perfectly suited to become a standard in calculating downstream medical costs of adverse events. Using the right data and the right analytics, a continuously updated, real world economic model can be supplied in almost real time, allowing for more informed real time decisions.
This type of insight should compel providers to report more, and with more reports by providers more insight can be gained.
As Brian mentioned in his last post, one of our commenters stated that drug AE reporting is the “urgent requirement of the decade,” while we agree with the sentiment, for us in the end, the only thing that will really change the healthcare industry significantly, is how the industry chooses to use this vital data.
Every day we work with innovative health plans, health systems, PBM’s and hospitals to begin to integrate these data and analytics into regular clinical workflow, if you’d like to know more about these efforts please contact us.
Executive Vice President