Data Quality in Public Health Standards

Steve Emrick (00:04):
Thanks everyone. We are here to talk about data quality and public health standards from the US Pharmacopia Convention in Rockville, Maryland. So full disclosure, I’m a former USP employee, so I’m excited to have our presenter here, Nakia Eldridge. So Nakia Eldridge leads activities, science, marketing and educational activities within the Healthcare Quality and Safety Center of Excellence at the US Pharmacopia Convention. Nakia is a doctor of pharmacy with a master’s degree in business administration. Nakia is a pharmacist with a demonstrated history of working in the hospital and healthcare industry, skilled in pediatrics, hematology, oncology, informatics, medication safety and hospital administration in the Washington, DC, Maryland, and Virginia area. Prior to joining USP in August of 2020, Nakia served several roles in the hospital setting for over 16 years. At USP, Nakia is the Director of the Healthcare Patient Safety Information Team, enabling USP to deliver quality standards and solutions that meet the needs of healthcare professionals and improve quality and patient safety. Nakia has direct reporting lines for three expert committees, including nomenclature and labeling, healthcare information and technology, healthcare safety and quality. Nakia also serves as the secretariat to the National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP) and a member of various pharmacy associations. So without further todo, Nakia, welcome.

Nakia Eldridge, PharmD, MBA (01:48):
Hello. Thank you everybody for having me here today. I am excited to really talk to you about the US Pharmacopia role in really taking public health standards and applying them into the digital environment. I’m going to share a little bit about USP, about how we work and what we’ve done and some highlights in getting our data quality and information into the digital world. So I want to first start by saying that USP started in 1820, was founded by 11 physicians and they really came together to address the issues at that time with poor quality of medicines. And those physicians, what they ended up starting was the US Pharmacopia, also known as USP, which is an independent nonprofit science organization. And our branding is really about quality and trust in the public health space. And so we’ve worked really hard to improve the global health through public standards and related programs that ensure quality, safety, and benefit of medicines and food.

(02:59):
USP pretty much is a standard solution setting organization and we actually have three standards that we provide. We provide written documentary standards and also reference standards to help generic medications meet quality. And we also provide monographs and recipes around medications and how to make these quality products on the market. We have about 9,000 standards that are quality benchmarks across the supply chain. And again, we touch on medicines, excipients, dietary supplements, food ingredients, biologics, and healthcare quality and safety. And so what I’d like to share today is really how our division in USP, which is called Healthcare Quality and Safety Center of Excellence, also known as HQS, is a very specific group within USP. And what we focus on are healthcare professionals in making sure that we provide those professionals with quality, patient safety and access to medications through standards and solutions. And what we do in that group is we actually have three buckets.

(04:11):
We have the personalized medicine, which really works around compounding in a lot of healthcare settings. They know we do compounding standards around sterile and non-sterile medications, hazardous medications, but we also have Medicare model guidelines and drug classifications that we provide. But the big thing that we’ve been working on is healthcare information and technology. And with that it’s really about getting data sets and turning those data sets into digital environments. So we’re really working over the last two years on how do we take these standards, which were at that time written in Excel and move them into a more digital format in order to share. And so what we’ve done is we’ve taken the time to really take two years back and look at how can we look at the data sets we had at the time. And the two that we had in that time were the USP Drug classification, which is a classification system that started in 2006.

(05:14):
It came off of the Medicare model guidelines that we work with CMS to provide over Medicare Part D drugs. And this classification actually not only has those drugs, but other drugs in it. And in those hundreds of drugs, they’re classified and they’re put into categories and they’re utilized by outpatient non-acute settings in regards for payers systems to figure out how to tier medications. And at the time when we created this in 2006, it was an Excel program and multiple people across the teams had the Excel. And so what you found is that we had issues where we had to transcribe information prior to publishing, and we had a lot of transcription errors, we had manual processes, and basically it took a lot of resources and a lot of time and a lot of data cleansing to get these together. And we realized that we weren’t providing a complete clean, consistent data set, and we really wanted to figure out what we needed to do to get to a place where we could have higher quality data sets.

(06:21):
At the time, we also had what we call hazardous handling standards. And these are standards around NIOSH drugs that are considered hazardous. And what we would do is we have a dataset that really maps those drugs with the NIOSH classification, the activity and the guidance recommended for handling it. Again, at that time, those were also Excel spreadsheets managed by different people that weren’t complete, weren’t as clean, and weren’t as consistent in their data quality as we would like them to get to. So we started looking at how do we get to a place where we could have a higher data quality, what sort of platform would provide that to us? But not only for our current state, but for our future state. And our future state we knew at that time two years ago that we wanted to start doing data model building, and that was something very new for us.

(07:14):
We knew at the time that our expertise was in compounding and that we would like to create a data model around compounding preparations, particularly for the pharmacist. And what we wanted to do in that model is to figure out how we could use our monographs and get all of the information needed for a pharmacist to prepare a high quality product for a patient, pediatrics, geriatrics, those who really can’t handle a manufactured product. And with that data model, we really wanted something that could take a lot of attributes into consideration. How long is the medication good for? What sort of instructions does the patient need to have? Should they refrigerate it? Do they need to do some extra things with it? So we really knew we needed a platform that could take about eight or ten different attributes so that we could build this. We also know that in personalized medicine, we were going to be and are starting to venture into pharmacogenomics.

(08:12):
We know personalized medicine is the world today, and there’s a lot of information with pharmacogenomics, but that information isn’t mapped and it is not mapped or integrated into the electronic health record for a clinician to make the decisions that they need around a patient’s medication treatment. So we took all of this into consideration as we looked at what sort of platform would give us cleaner data, complete data, consistent data, and also allow us to scale up in our data models as we start that off. And so what we found in this journey is that basically we’ve actually found Symedical with Clinical Architecture to be quite honest. And we found that this was a great platform for us in three different efforts as we’ve shared. One thing that we know we needed was terminology management. We know we needed a source of truth. We wanted to get from multiple Excels to just one platform and Symedical allowed us to do that.

(09:14):
And on that one platform, we have some data governance and we have a data quality where we have one person who owns it. We have the fields that are defined for everyone to use. It’s organized, and it also allows us to do terminology retrieval. And so what we knew with our data model is that we wanted to access some standard terminologies, RxNorm, RXQEs, NDAs, things like that. And we knew we could do that through this platform. And it also allowed us to collaborate more across the different divisions internally. We also knew that we wanted to have semantic interoperability. So we knew that as we created these data models around things that are not standardized now, we were going to need to create codes. We were going to need to normalize and create a coding for the data exchange because we do want this information to go from, let’s say, an inpatient setting to an outpatient setting.

(10:12):
And we know that current systems aren’t as standardized coding. So we wanted something that would allow us to create a code that we could attach to our attributes for it to be interoperable. We wanted our information to be actionable, we needed it to be usable in the electronic health records, and we needed to have a structured message to go across one platform to another platform. And so we’ve been able to start that model in Symedical. The final thing we wanted is we learned from the Excel spreadsheets that our team did not like using it. They avoided it because it was so complex because it wasn’t clean. And so they really didn’t touch it until the absolute last moment, and they didn’t interact with the data as much as they needed to. And so what we have found with Symedical is that now our workflow is optimized.

(11:02):
We found that the data quality has influenced our efficiency. We found that we’re able to have the staff look at it more, work with it more, and that’s decreased errors, that’s increases productivity and has really put us in a good place where we can scale up this information. And so we are very excited to really use this platform to get our information across platforms like Electronic Health Records to really help healthcare practitioners get the information they need to better do their job. So I know I’ve said a lot and I thank you for letting me share some of this information and I’ll open it up to any questions or any thoughts.

Steve Emrick (11:47):
Thank you, Nikia. So I’m going to ask the first question. So it’s for those that know USP, USP has been around, like Nikia said, for over 200 years. The main purpose of USP is they’re a standards development organization and they produce standards that generic drug manufacturers have to adhere to ensure the strength, stability, and quality of drugs on the market. What’s a little bit lesser known is some of the things that are in the drug classification space for outpatient formularies and ensuring a minimum standard of drugs for a given patient population. So Nakia, maybe you can describe a little bit more things like the Medicare Model Guidelines, the USP drug classification, who uses them, and then kind of the risk of poor data quality. If a drug is misclassified, what does that look like? Right. Just briefly.

Nakia Eldridge, PharmD, MBA (12:43):
So USP started about 20 years ago to work on the Medicare Model Guidelines. And this came off of, actually, it was written into law that we would play a role with CMS to classify medications with Medicare Part D. And what we’ve done is we really have guiding principles around that on how we classify these drugs and the ideas about balancing the classification system so there are options for folks. You don’t want to have one drug in a classification because that’s not going to give you equitable access, that’s not going to give you options, and that can lead to higher cost, that can lead to a lot of negative patient outcomes in regards to access. So we have about 20 experts who go through every newly approved FDA drug and they classify it in this system and they really, we work with CMS and we do that every three years only for Medicare Part D.

(13:43):
What happened is that when it got to that stage, we heard from stakeholders that there are new drugs on the market every day. Three years is a long time to wait on knowing where those should be classified. We need something more frequent. And so we took that up to say, okay, we’ll create something, not Medicare Model Guidelines, absent of CMS, we will create our own guideline system. And that system has the same guiding principles, so there has to be two medications, at least two in each classification. But we also thought, hey, we should be looking at other drugs, Medicare part B, we should be looking at drugs that like anti-obesity drugs, drugs that the states are needing to classify that don’t have any classification system around it in this setting, in the state setting, reporting, in the federal setting for CMS. And so we have now been doing that every year, and it’s been very interesting on who’s interested.

(14:43):
We get reached out to by a lot of manufacturers on where to place their drugs. But we also hear back from advocacy groups, I like to say the hemophilia group, it’s like, hey, there’s shortages, there’s issues. You need to make sure we have options for our patient population. So we do reach out and have people reach out to us on a regular basis on how we classify things, where we can classify it, and what is the balance that we can help provide with options. We do find that the payers do look at this list. And so we do have folks that find that valuable as they look at where will this drug maybe be placed or what are my competitors or what are the things I need to look at from not just a pharmacological perspective, but also from an availability access perspective. And so we’ve been doing that for quite a bit and we’ve actually heard from folks that there’s more information that would be helpful. So we just launched USP DC PLUS. It’s been very interesting to find that not only do researchers look at it for their classification and understanding, but we also have folks that are looking at it for pricing information. So packaging information, how do I tell some of that information? So we’ve added some of that into our data set so that it will help them use it for a variety of reasons, maybe not just for placement on formularies. Helpful?

Steve Emrick (16:12):
Thanks Nakia. One more question from me and then we’ll open it up to the audience here, I promise. So you had a slide in there that referenced some things in the future. One of them was compounded drugs, which is kind of a form of personalized medicine. So maybe explain to the audience that might not be familiar, what is a compounded drug, how it’s different than a traditional FDA approved drug on the market, and what is USP doing there in terms of digitalizing that knowledge?

Nakia Eldridge, PharmD, MBA (16:43):
That’s a loaded question. So USP has for a very long time been in this space about providing alternative medications. So when a manufactured medication is not available, how do we make it available? What can we do for the patient? They need it. They need it. And so we created compounding standards for those situations where you have to compound a medication, be it a medication you take orally, which we call non-sterile, or be it a medication that has to be injected, which we call sterile. We’ve spent a long time updating these standards to say that if you are someone who’s going to make these medications, you need to make sure the environment is sterile, you’re gowning up, that you’re doing what’s safe for you, but also safe for the patient. USP sets these standards, but we do not enforce these standards. So what you have found is that Joint Commission, Boards of Pharmacy, other entities actually uphold these standards in certain settings because we do want to make sure that the medication is of the highest quality, sterility, and safety for the patient.

(17:56):
What has happened is those standards just got updated last November, and there’s a lot in them that has to be considered. The FDA is of a mindset, if I can say, that you should use manufactured products as much as possible and they will say, okay, this is not available, we’ll allow compounding in this situation. And you’ll then find, some organizations will say, follow the USP guidelines on how you make those medications. And that’s kind of the ecosystem around the compounding. What has happened is, it became a questionable thing of how do I make it, okay, now I know I can, but what’s the ingredients? What do I put together? Some people mix certain things. I consider it like cooking at home, you throw different ingredients in, and so it may not last as long depending on what you put into it. And so we actually test stability to make sure that these five ingredients you put in there means that this product is going to be good for that patient for 14 days, 24 days, whatever timeframe.

(18:59):
And we have a recipe that you can purchase that’ll tell you how to make it, make sure you use these ingredients and this is how long it’s good for, and these are the conditions you should keep it in. So at home, keep it in your refrigerator or keep it at room temperature, all of that information so that it’ll stay stable for the amount of time. What we have found in this project is that because compounded medications are so unique, they don’t have an NDC, they don’t have a code. So when we build them, it may not be as transparent what ingredients went in. It may not be clear. So you may go to Walgreens and they may make it one way. You may go to CVS, they may make it another way. They may make it a different concentration of different strength. And if you take it the same with incorrect information, you may overtreat or undertreat, and we think there should be a standard, right?

(19:53):
So no matter where you go, it should be made the same way and it should be made with the same consideration and ingredients so that it’s consistent. Some people are allergic to certain excipients. Red dye may make someone allergic. You may get allergic when one person makes it, but not when another, how do we capture that information and make sure we’re protecting you? And so with this project, what we’re looking to do is create our own code and basically make a compound almost like a manufactured product. It’ll have a compounded formulation ID that will make it real, make it official, but also have the ingredients there for the prescriber to put it through and for the pharmacist to make it the same way every time for the patient. So that’s what we’re excited about.

Steve Emrick (20:41):
Thank you, Nakia. I like that one because compounding goes back to the roots of USP all the way back to 1820 where if you got a compounded preparation in Pennsylvania versus New York versus Vermont, it might be made very, very, very differently between those three states. And that’s why USP was founded over 200 years ago. And I think one of the big takeaways is this kind of digital journey that USP has been on. The benefits go beyond just the operational efficiencies. It’s really kind of maximizing the impact of the important work in terms of public health standards development and making sure those things are adoptable by USP stakeholders. So thank you Nakia. Open it up to any questions in the audience, does anybody have any questions?

Audience Speaker 1 (21:34):
So Nakia, that new compounded drug identifier, where will it live? Will it live with USP or are there ideas of promoting it up to some standards body or anything like that?

Nakia Eldridge, PharmD, MBA (21:50):
Very nice question. So yes, USP will be the one to assign that coded formulation ID. And so we’re working on making sure that every time we make a preparation and create that data model, it has that ID. And what has been beautiful with Symedical is truly allowing us to build that model where that ID can have its own designated space and for it to go over. I will say that we’re at a place now to proof of concept this, because what you find is that the outpatient pharmacies have their own homemade data systems, databases, your inpatients, we’re Epics, we’re Cerners, we’re Meditech, but outpatient works off of a different system. So it’s going to be very interesting to see how that CFID will move across systems, but at least it’ll have that identifier that’s the same no matter where that you’re able to utilize. And we have other national databases that have been willing to build that into their system so that you can find it if you’re someone who’s making a formulary in any setting. So we are excited.

Steve Emrick (23:00):
Thank you. Great question. Any other questions from the audience? Very well. Well, Nakia, thank you so much again. This has been very enlightening. Thank you all for attending and I think that’s the last session of the day, so enjoy the rest of the day. Thanks!