Pharmacovigilance Archives - Astrix

Case Intake – Key Considerations and Pitfalls that you Want to Avoid (Part 3)

Astrix — Thu, 09 Mar 2023 22:56:08 +0000

Today Pharmacovigilance has gone through what some would call a revolution. That’s due in large part to the technology that companies are implementing within their safety systems.

Many companies are assessing their case intake system. We at Astrix are assisting many companies with their projects and whether it’s helping them with their RFP’s, or with their implementations. The most common questions that we hear when assisting companies is:

What are the features and functionalities that we should have or expect the tool to be able to do and how’s it going to help me?
How is it going to make my processes more efficient?

We all know case intake is one of the most manual and time-consuming parts of the process. We at Astrix have had clients estimate that case intake is as high as 50% of the total time spent on a case.

One of the issues that companies have today, even with all the time they are spending on case intake, is that they still don’t believe they’re getting the accuracy that they desire (i.e., Inconsistencies like data missing or data not being put into the correct field). There’s a great deal of frustration with today’s process and the volume of cases that companies are getting are growing exponentially. Therefore, the workforce is extremely challenged to keep up with this increased volume. Companies are looking to invest in this area of case intake because they know there is a large return on their investment.

They are implementing intake tools that are highly automated using AI and ML automation. All the technologies that we’ve been hearing about for many years are finally coming to fruition and by using these technologies we’re seeing companies reduce the time that they’re spending on a case from what previously was 30% to 50% to maybe 20% to 0% depending on the type of case. They’re also seeing increased consistency with the data entry. For example, the AI tool they’re using is only going to spell hospitalization one way and don’t have to worry about different standards and people mistyping. They’re getting a higher degree of consistent data which is providing them with a tremendous amount of downstream efficiencies in regards to signal detection and many of the other downstream type of processes.

Companies are finding that they need less people for this part of the process by implementing these tools. This, however, doesn’t mean that they are reducing their workforce, but rather looking to reposition their employees to be more strategic and to work on other areas that really do use and require the human review or human mind. Of course, in the intake process, there’s still going to be many human interventions even with technology, however, we can also talk about a touchless situation where basically humans aren’t even looking at the case.

Now, we will discuss some of the key features that case intake systems should have and some of the pros and the cons of the systems, along with factors to be aware of when evaluating a case intake tool. We will discuss what is needed relative to the general steps of the case intake process. These case intake process steps are fairly standard although companies may have different or additional steps included or reorder them differently, however, we will discuss a general workflow that we see with clients.

The General Case Intake Process Steps

Data Acquisition – the company will acquire the data from some source.
Case Validity – next it will be checked that it’s a valid case.
Duplicate Check – They will see if it’s a duplicate or a follow-up.
Prioritization – If it’s not a duplicate, then it’s assigned a priority.
Translation – next it will go through a translation process for those cases that require it
Medical Coding – this is where the medical terminology specific to this case is added and checked.
Alert for Cases of Interest – if there is an alert that needs to go out for cases of special interest like a death case or a specific drug reaction combination.
Case Intake Review – then once all that is done the case is reviewed to ensure that all the data is entered correctly and handed off to the case processing step.

In the previous blog on this topic, we reviewed the Duplication Check and Prioritization of cases. In this blog we take a closer look at Translation and Medical Coding steps in more detail and the functionality required to ensure accuracy.

Translation

As we know, organizations receive many non-English cases and the volume of non-English cases are growing as well. It is therefore important to have a tool that can translate automatically. The historical process has always had human translation whether by leveraging translation servers or doing it through local affiliates. Today, vendors are integrating their systems with translation tools that are available on the market or leveraging those built in-house. These translation tools can translate to English from almost any language that is spoken. It is imperative that the translation tool you select can translate to the languages that are most common or prevalent within your business.

After the case has been translated, the system should present you with both the English and non-English values so you can confirm the translation (when needed) and all employees in your offices around the world can look at the case. There should also be appropriate scoring as to how confident the system is with the translation. If it had difficulty translating the text it should identify the fields or the sections that need your attention.

Having this automated translation is one of the most cost-effective areas that these systems are providing. The translations aren’t always 100% accurate and medical terminology can be challenging for these tools, however we’re seeing an increase in performance with them within the industry.

Medical Coding

After the system has extracted all the data and done all the other steps, you want the system to be able to code to MedDRA or WHODrug dictionaries or any other dictionaries which are relevant for your system. Having the system identify, translate, and code these terms from the reported term to the medical term is going to be extremely important as well as allow you to have that consistency that we are all looking for from the system.

The system automatically coding the case is one part of this step. The other part is having the capability to properly assign the code to the appropriate field. Is it an adverse event, is it a medical history, is it a suspect or concomitant medication.

You would think that this would be one of the challenges that these tools would have, however, that is not the case. If they see something as simple as a rash, they determine how the term rash is being used within the narrative or other non-structured field and surprisingly these systems do a very good job using Artificial Intelligence and Machine Learning. They’re looking at the context around the terminology and by doing so, they’re able to place these fields appropriately. At times, there still requires some human review from a dictionary or coding perspective. We also want to ensure that the system can utilize your IME as well as the DME and any other list that you have to help derive your seriousness or other assessment values for your case.

Summary

It is critically important, when looking to incorporate new technology into your business, that you ensure you have the right external organizations involved who can assist you. Formulating the appropriate strategy along with the proper processes, and technology is imperative. Astrix’s team of professionals have worked with many of the top life science organizations to assist them with respect to their business needs in these areas. As a technology-agnostic partner, without a preconceived preference for a specific supplier or product, we work closely with your team to ensure solutions are reviewed and incorporated into your business so that you succeed in realizing your vision and achieving your organizational goals.

About Astrix

Astrix is the unrivaled market-leader in creating & delivering innovative strategies, solutions, and people to the life science community. Through world-class people, process, and technology, Astrix works with clients to fundamentally improve business & scientific outcomes and the quality of life everywhere. Founded by scientists to solve the unique challenges of the life science community, Astrix offers a growing array of strategic, technical, and staffing services designed to deliver value to clients across t

The post Case Intake – Key Considerations and Pitfalls that you Want to Avoid (Part 3) appeared first on Astrix.

Case Intake – Key Considerations and Pitfalls that you Want to Avoid (Part 2)

Kevin Miller — Wed, 08 Mar 2023 17:17:43 +0000

Today Pharmacovigilance has gone through what some would call a revolution. That’s due in large part to the technology that companies are implementing within their safety systems.

What are the features and functionalities that we should have or expect the tool to be able to do and how’s it going to help me?
How is it going to make my processes more efficient?

We all know case intake is one of the most manual and time-consuming parts of the process. We at Astrix have had clients estimate that case intake is as high as 50% of the total time spent on a case.

The General Case Intake Process Steps

Data Acquisition – the company will acquire the data from some source.
Case Validity – next it will be checked that it’s a valid case.
Duplicate Check – They will see if it’s a duplicate or a follow-up.
Prioritization – If it’s not a duplicate, then it’s assigned a priority.
Translation – next it will go through a translation process for those cases that require it
Medical Coding – this is where the medical terminology specific to this case is added and checked.
Alert for Cases of Interest – if there is an alert that needs to go out for cases of special interest like a death case or a specific drug reaction combination.
Case Intake Review – then once all that is done the case is reviewed to ensure that all the data is entered correctly and handed off to the case processing step..

In the previous blog on this topic, we reviewed Data Acquisition and Case Validity. In this blog, we take a closer look at Duplication Check and Prioritization steps in more detail and the functionality required to ensure accuracy.

Duplicate Check

From a duplication check perspective, it’s imperative that your intake tool is looking at the data and recognizing if you’ve received this case before. The system is never going to be 100% accurate, however, the system should be capable of displaying all the cases that are potential duplicates, and most importantly, providing us with a score or probability of a match. Different systems will do it differently. Some may go from 0 to 1 or 0 to 100, with the high end representing the system is confident that it’s a duplicate.

Once we’re able to look at that case and confirm it may be a duplicate, we want to be able to display the potential duplicate case in a side-by-side view with the received case so that we can confirm that it is a duplicate or potentially a follow-up. If it is a follow-up, we want to be able to use the tool to indicate to the user what’s new, updated, or deleted.

One important point to keep in mind is that some of the vendors charge per transaction. That is whether it’s an initial case a follow up or a duplicate. Anything that’s received is a transaction, so they’ll charge you per transaction and therefore you want to try to see if there’s ways to minimize receiving duplicates or identifying them earlier in the process to help minimize this occurrence.

Prioritization

Once we have a case and we have determined it’s an initial case, we want the system to automatically assign a priority based on some automated rules. It’s extremely important for the system to be able to do this so that it pulls the user’s attention to the cases that are of the highest priority. Once we have this case list and its priority, we also want systems that are able to sort multiple columns within that case list. In many of the systems today, you can only sort on a single column. The ability to sort on multiple columns is extremely useful. The ability to have the system assign priority to a case will help draw their attention and work on the cases that need their consideration.

Summary

About Astrix

The post Case Intake – Key Considerations and Pitfalls that you Want to Avoid (Part 2) appeared first on Astrix.

Case Intake – Key Considerations and Pitfalls that you Want to Avoid (Part 1)

Astrix — Fri, 03 Mar 2023 19:55:25 +0000

Today Pharmacovigilance has gone through what some would call a revolution. That’s due in large part due to the technology that vendors are implementing within their safety systems.

Many companies are assessing their case intake system. We at Astrix are assisting many companies with their projects and whether it’s helping them with their RFP’s, or with their implementations, the most common questions that we hear when assisting companies are:

What are the features and functionalities that we should have or expect the tool to be able to do?
How is it going to make my processes more efficient?

We all know case intake/triage is one of the most manual and time-consuming parts of the process. We at Astrix have had clients estimate that case intake is as high as 50% of the total time spent on a case.

One of the issues that companies have today, even with all the time they are spending on case intake, is that they still don’t believe they’re getting the accuracy that they expect (e.g., Inconsistencies like data missing or it not being entered into the right field). There’s a great deal of frustration with today’s process and the volume of cases that companies are receiving are growing exponentially. Therefore, the workforce is extremely challenged to keep up with it. Companies are looking to invest in the area of case intake because, they know there is a large return on their investment.

Therefore, they are implementing these intake tools that are have incorporated technology like AI/ML and automation. All the technologies that that we’ve been hearing about for many years are finally coming to fruition and by using these technologies we’re seeing companies reduce the time that they’re spending on cases from what previous was 30% to 50% to maybe 20% or even 0% depending on the data received and the type of case. They’re also seeing consistent data entry. For example, the AI tool they’re using is only going to spell hospitalization one way. You don’t have to worry about different standards and people mistyping. They’re getting a higher degree of data quality which is providing them a tremendous amount of downstream efficiencies in regards to signaling and reporting and many of the other downstream type of processes.

What companies are finding is that they need less people for this part of the process given these tools. This, however, doesn’t mean that they are reducing their workforce, but rather looking to reposition their employees to be more strategic and to work on other areas that really do use and require human input. Of course, there’s still going to be many human interventions in the Case Intake process even with technology, however, we will also talk about a “touchless” processing where basically humans aren’t even looking at the case.

Now we will discuss some of the key features that case intake systems should have and some of the pros and the cons of the systems, along with factors to be aware of when evaluating a case intake tool. We will discuss what functionality is needed relative to the general steps of the case intake process. These case intake process steps are fairly standard although companies may have different or additional steps included or reorder them differently, however, we will discuss a general workflow that we see with clients.

The General Case Intake Process Steps

In the general case intake process

Data Acquisition – the company will acquire the data from some source.
Case Validity – next it will be checked that it’s a valid case.
Duplicate Check – They will see if it’s a duplicate or a follow-up.
Prioritization – If it’s not a duplicate, then it’s prioritized.
Translation – next, if needed, it will be translated to English.
Medical Coding – this is where the medical terminology specific to this case is added and checked.
Alert for Cases of Interest – if there is an alert that needs to go out for cases of special interest, for example, a death case or a specific drug reaction combination.
Case Intake Review – once all that is done the case is reviewed to ensure that all the data is entered and is correct before handing it off to Case Processing.

Let’s now take a closer look at Data Acquisition and Case Validity steps in more detail and the functionality required to ensure accuracy.

Data Acquisition

In the first step in the process, data acquisition, we’re getting information from all sorts of sources. Whether it’s from a spontaneous report, literature, a partner, a regulatory agency, or your Call Center.

We see them coming in from many different sources and through various channels. Companies are receiving e-mails, faxes, E2B messages as well as other electronic formats. We’re also starting to see companies leveraging social media to extract cases. Moreover, you may have a partner and there’s a shared file folder. There may be other sources that you have as well.

One of the key features that you want from your system in this area is the ability to not only receive all these types of information or reports, but also to receive them in their particular format. If you have an intake tool that can only bring in e-mail and electronic formats and you can’t receive a e-fax or a file folder, this can cause issues. You need to be able to receive all the cases in the various formats through the tool to be efficient. Because, if you’re having that mixed process of some pieces being automated and some are being manual, it becomes very cumbersome. We have seen companies miss cases due to this mixed process. So, the first thing that we need to ensure is that the tool will allow you to receive all of the type of reports that you get, in the different modes that you receive them.

Key Take-away – When a system can’t handle a particular source (e.g., Social Media), there are options that can be utilized to automate porting of data to the Intake tool.

Case Validity

With case validity, we need to ensure that the case has the main four elements (i.e., patient, reporter, adverse events, and product). It’s imperative that the tool can do this. The tool needs to be able to extract or look at the file that is received and can extract and recognize all four of these elements to ensure it’s a valid case. It needs to also ensure that if it doesn’t have all four of these elements that it brings that case to a different part in the workflow or to a case list. In this scenario, you may have some human review to ensure that there isn’t something that is missed. It’s important to note that different systems do it differently.

A classic example that shows the importance of this is the “cat fax case”. The story goes like this. The company received a fax of literally a picture of a cat. Somehow the system received it as a death case that had all four of these valid elements. Now of course, that system needs to be trained better, however, that’s more the anomaly. You do want the system to be extremely efficient at this and to ensure that all the cases that are valid are coming through into your workflow and processed appropriately.

Key Take-Away You’ll need to have a process in place to manage a receipt that doesn’t have the 4 elements.

Summary

About Astrix

The post Case Intake – Key Considerations and Pitfalls that you Want to Avoid (Part 1) appeared first on Astrix.

Key Considerations When Evaluating Pharmacovigilance System Providers (Part 2)

Astrix — Thu, 02 Feb 2023 17:01:07 +0000

Your safety program will be impacted by the system and provider you choose for years to come. Although system functionality frequently predominates in selection criteria, we have learned to consider key attributes of each candidate system vendor as early in the selection process as possible before selecting the best fit.

Astrix has direct experience working with clients to help them choose their system providers. This experience, along with our Pharmacovigilance System Providers neutrality and lack of financial interest policies, allows us to provide insights to help you throughout the entire process of choosing the business that will best serve the specific needs and preferences of your organization:

System Maintenance

Although it may be considered an issue for the end of a presentation, it is a high priority to know—up front—a potential Pharmacovigilance System Providers plan to support their product. It is wisest to know if they:

Provide a schedule for regular releases for upgrades and/or maintenance patches, or
Plan to maintain their system primarily via emergency fixes.

If a system provider doesn’t automatically, or cannot promptly, provide their schedule for anticipated upgrades and patches, it is crucial to obtain their documented process for identifying, capturing, and resolving operating defects you report to them.

Help Desk or Onsite Support

Does the Pharmacovigilance System Providers have a Help Desk that matches your operating hours when support is likeliest to be needed? What is the level of their Help Desk staff’s experience? Vendors sometimes view Help Desk duty as a training ground for new employees. If this is the case, you need to ascertain that policy—upfront—so that you are comfortable with the level of training that will be standing by to remedy a problem—or stop a crisis in your operation. Lastly, on this critical aspect of support, it’s essential to know the expected turnaround times for help desk requests so that you can assess their acceptability for your operational requirements.

System Administration

Since system maintenance is critical to maximizing uptime and minimizing interruptions to normal operations, it is important to know how candidate systems should be maintained. Based on your staffing and capabilities, only you will be able to assess the acceptability of having to do this internally, or by relying on the vendor to do this at a mutually convenient time. Some examples of administration include:

Adding a new lot to a product,
Obtaining, or resetting, a user password, or
Setting up a new agency for reporting.

The following questions are best posed before a system is demonstrated:

Who will perform critical system maintenance activities?
How are these critical system maintenance activities done?
How easily are critical system maintenance activities performed?

Technical Architecture

Although it is not uncommon for system vendors to promise seamless compliance with organizational preferences, it will save a lot of time, cost, and anxiety, to know their preferred environment for optimal installation and smooth operation. As part of the initial information gathering on compatibility, it is important to know:

Does the vendor’s technical architecture allow customization with your corporate IT management’s direction?
If your organization’s preference is to have systems in the cloud, you will want to ensure the candidate vendor provides robust support for cloud-based operation.
Does the vendor support your SSO and other integrations?
How easily can vendors’ systems integrate with other systems (e.g. Signal Detection software)?

Selecting a new system is a big decision and project for any company so it is important that you not only evaluate the system’s functionality but also consider the vendor that will be providing and supporting the software. It is important to feel confident that you can have a strong partnership with the selected vendor as well as their ability to support your company and regulatory requirements. Astrix’s unique position of being system agnostic allows us to help you determine the best vendor and system for your needs.

About Astrix

Astrix is the unrivaled market-leader in creating & delivering innovative strategies, solutions, and people to the life science community. Through world class people, process, and technology, Astrix works with clients to fundamentally improve business & scientific outcomes and the quality of life everywhere. Founded by scientists to solve the unique challenges of the life science community, Astrix offers a growing array of strategic, technical, and staffing services designed to deliver value to clients across their organizations.

The post Key Considerations When Evaluating Pharmacovigilance System Providers (Part 2) appeared first on Astrix.

Getting the Case Intake Process Right in Your Safety Systems

Astrix — Fri, 06 Jan 2023 15:33:54 +0000

The Intake process and its corresponding tools have been a point of focus for companies updating their safety systems because the Case Intake (Initial Triage) Workflow phase can take up to 50% of the overall case processing time. Current and new vendors are focusing on incorporating modern technology into their Intake tools to gain efficiency in this portion of the process. Sometimes, however, these modern technologies manifest themselves as features, functionalities, or post-implementation options that can distract, complicate, or unexpectedly impact budgets for new and upgraded safety systems.

While some companies are taking a “Big Bang” approach to evaluating (and replacing) safety systems, many look to maximize their investments by limiting their focus to improving or replacing their Intake tool. In either case, there are some key functionalities that companies should consider when evaluating the next generation of Intake systems:

Sources – Adverse events can be received from a variety of sources (e.g., email, fax, social media, company portal). An Intake tool should be able to receive potential cases from any of these sources.
Acknowledgement – Once a piece of information is received, it is essential for the tool to have the capability of replying to the sender with an acknowledgment. This can be in the form of an email, E2B acknowledgment, or another format.
Accuracy – Once an adverse event report is received from a source, data will be extracted and displayed to the user. An Intake tool should extract most (if not all) of the data with a high degree of accuracy.
Translation – Since a report may be generated in a non-English language, the tool should translate it prior to extraction of the case data.
User Interface – All tools will look different, but it is important to ensure that it is user-centric, intuitive to navigate, and easily read. It should allow staff to add, update, or delete extracted data when either inaccurate or incomplete. The visual interface should also allow the user to create a new case that was not received by a standard source (e.g., word of mouth).
Confidence Score – Tools should display to the user the confidence that it extracted the data accurately. It should clearly call out fields or sections for which it has low confidence so that the user can review the data to either confirm or correct data.
Duplicate Check – Post-extraction, it is important to determine whether the potential case is a duplicate or a follow-up to a case previously received. If it is a follow-up, the tool should indicate to the user what data has been updated, added, or deleted. If it is a duplicate, it should point the user to the previous case to confirm the potential duplication.

During the system selection process, it is important to bear in mind two points that will help ensure the best system is selected for the current and future state of the organization:

Define clear requirements prior to evaluating any system to avoid the selection bias of considering one system as either a gold standard or the favorite.
Evaluate any system’s additional capabilities separately, and in a second phase, after ensuring that required capabilities are both present and in a format that works well for your staff and processes.

Since Case Intake is the first step in the process, it is critical to have an efficient process that extracts all data sent with high accuracy to ensure that all the remaining steps in the case process continue, if not improve, their efficiency.

Why It Matters to You

Because the Case Intake (Initial Triage) Workflow phase can consume up to 50% of the total case processing time, organizations updating their systems need to focus on the Intake process and the accompanying tools. To improve the efficiency of this step of the process, both existing and new vendors are concentrating on integrating cutting-edge technology into their intake solutions. However, these contemporary technologies can occasionally appear as features, functions, or post-implementation alternatives that can be distracting, difficult to plan for, or have unanticipated financial effects on new and upgraded systems.

No matter if your organization is taking the “Big Bang” approach to evaluating (and replacing) safety systems or looking to maximize your investment by limiting the focus to improving or replacing the Intake tool, the key is to ensure the right functionality is available when evaluating the next generation of Intake systems.

In this blog we discussed:

Several of the key functionality areas to consider relative to Case Intake.
Why this functionality is important.
Key points to consider to ensure the best system is selected for the organization.

About Astrix

The post Getting the Case Intake Process Right in Your Safety Systems appeared first on Astrix.

The Future of Artificial Intelligence and Machine Learning in Pharmacovigilance

Astrix — Mon, 17 Oct 2022 00:38:09 +0000

Artificial Intelligence (“AI”) and Machine Learning (“ML”) are creating a buzz in the Pharmacovigilance (“PV”) space because of their transformative potential to provide new and interesting insights, by processing big data in a timely, accurate, and cost-efficient manner, and promises to enable product innovations to better protect patients and the world.

Defining Terms

Artificial Intelligence is a becoming challenging term to pin down because, increasingly, it’s applied across an exploding array of new approaches, applications, and products, but I like to think of it as, “The ability to train computers to do things that were, or are, being done by humans”.

To demystify this somewhat Orwellian description, let’s consider just two applications already in our daily life: cars with lane assist technology, and security measures involving facial recognition. AI is being employed to accelerate data-driven processes affecting our daily lives in ways that were nearly unthinkable (to anyone other than science fiction authors and NASA engineers) even a generation ago.

Pharmacovigilance can also benefit from AI—which promises to enhance human health via outcomes of product use—but it will require computers to deliver reliably better results than humans when using the same data. So, with this high of a bar, it makes sense that Artificial intelligence must rely on several subfields to actually facilitate true learning.

There are three subfields to know about right now:

Machine Learning (“ML”) uses data, algorithms, and processes to mimic how humans learn without being explicitly programmed to do so. As the computer receives more data, it gradually improves its accuracy to predict outcomes. Although we won’t cover them in detail here, there are 3 types of ML that are rapidly emerging and developing:
Supervised learning
Unsupervised learning
Reinforcement learning
Deep Learning (“DL”) is a subset of ML that reaches a higher level of sophistication via focusing on pattern recognition in the same way that the human brain does. Speech and facial recognition are two examples of Deep Learning.
Natural Language Processing (“NLP”) – is broadly defined as software that automatically manipulates natural language, like human speech and text. Language translation and analysis of unstructured data are two types of applications driven by NLP programs.

These three sub-areas of AI can—and often do— overlap as depicted in Fig 1

Given the critical nature of computations demanded of AI, and its sub-areas, there is intense focus on monitoring its performance in mission-critical applications. Professionals charged with developing and monitoring AI-supported programs use performance metrics to objectively assess AI processes for quality and consistency.

One popular performance metric used to evaluate how well ML is performing is the F1 score. The F1 score is a combination of precision and recall.

Precision is the number of True Positives divided by the number of Total Predicted Positives (True Positives + False Positives). Precision can be thought of as a measure of exactness. Therefore, low precision will indicate a large number of False Positives.

Recall is the number of True Positives divided by the number of real positive cases (True Positives + False Negatives). Recall can be thought of as a measure of completeness. Therefore, low recall indicates a large number of False Negatives.

The formula for the standard F1 Score is:

2*(Precision*Recall)

(Precision+Recall)

A strong PV system will be able to share or display its “confidence score” on each piece of extracted data, where it is located, and the algorithm(s) or approach(es) it used to obtain the confidence score. In addition, it is highly desirable for your PV system to be able to learn or improve as it receives more data and notifies someone within your organization when a new potential algorithm has been identified.

Future of AI and ML in Pharmacovigilance

AI is already having a profound impact on the performance of Pharmacovigilance programs, but we are still scratching the surface of AI’s potential role in PV program impact and efficiency. Here’s just a sample of the many areas in which AI can be applied:

Reference Safety Information (“RSI”) Evaluate – Using current AI techniques, your system can read through any of the RSI literature (e.g., Investigative Brochure, Product Insert, Summary of Product Characteristics) to determine and extract the terms that should be used to assess expectedness. As the RSI literature changes, a company can rerun its algorithms and identify changes that need to be applied to its system.
Causality Assessment – Some companies are already evaluating and piloting AI to identify the causality assessment. The problem that companies are seeing is that, in some cases, the F1 score is not high enough, simultaneously increasing manual reviews while decreasing confidence. As algorithms improve, and data increasingly runs through existing algorithms, this will continue to improve.
Touchless Workflows – For certain types of cases with a high enough F1 and confidence score, it is feasible that a user would never have to see the case data. The system would automatically extract the data, process the case and, if required, report it to appropriate agencies. While some companies are piloting this for very specific low-risk non-serious cases. As AI technology improves and PV communities feel more confident with the technology, the volume of cases that would be “Touchless” will increase. This would have one of, if not the biggest, impacts on both the financials and time for processing data.
Predictive Patient Risk – As regulatory agencies and companies receive more data AI can do one of the things it does best: analyze large amounts of data and identify significant patterns. AI could soon enable companies to identify which patients are at greatest risk for an adverse event to a prescribed medicine based on their genetic and biochemical characteristics.

Why it Matters to You

Even though AI is already having a significant impact on performance in the PV area, we have only begun to scratch the surface of AI’s potential impact on the effectiveness of PV programs.

In this blog we discussed:

Key definitions of AI related terminology.
A popular method to measure how well ML is performing.
Where AI and ML are being used in life sciences.
How AI and ML are impacting the evaluation of RSI Literature, Causality Assessments, Patient Risk.

To learn more about how AI and ML are impacting Pharmacovigilance you can download our white paper on this topic: Next Level Vigilance & Safety: Insights into the Transformative Potentials of AI & ML

About Astrix

The post The Future of Artificial Intelligence and Machine Learning in Pharmacovigilance appeared first on Astrix.

World Drug Safety Congress Conference Wrap Up

Astrix — Thu, 13 Oct 2022 15:09:12 +0000

Astrix had the opportunity on Oct 4-5 to participate as a Gold Sponsor of the World Drug Safety Congress conference in Boston, MA. This event had over 1000+ pharmacovigilance and safety experts in attendance.

The list included pharma and biotech companies, along with academic, government, regulators, and vendors who came together to discuss the key challenges being faced in the PV world. Participants from Astrix included individuals from our PV Delivery Services, Strategic Consulting Services, and Marketing organizations.

At this event, Astrix had the opportunity to meet many existing and new companies and introduce them to the Astrix PV practice and the domain-specific strategies, solutions, and people to assist their organizations.

Over this two-day event, Astrix participated in sessions on the following topics

AI & Machine Learning impact on improving PV processes
The next generation of tech and data
Strategies in data management and signal detection
Discussions on the challenges of creating PV strategies for advanced therapies
Medical device safety strategies
PV methods across the globe

Astrix Moderated Discussion on the Topic of AI and ML

Alan Summer, Astrix’s Vice President, Pharmacovigilance Services, moderated an important discussion at the conference where the topic of Navigating the Changing Landscape of PV Safety Databases was reviewed. In this session, there were panelists from Astellas Pharma and Moderna who with Alan discussed many different aspects of the changing PV Landscape including:

The influx of cases and how they are managing this increased volume.
The trend to move from an on-premise to an on cloud-based solution
The use of AI and ML in the PV area to improve the processes of their business.

Challenges they are seeing or keeping an eye on with the adoption of this new technology

About The Pharmacovigilance Practice at Astrix

Astrix provides strategic consulting services that enable the success of end-to-end Pharmacovigilance (“PV”) planning, design, and implementation as part of our comprehensive suite of services. Astrix experts know first-hand that the PV landscape is evolving and that our clients prefer a strategic IT partner (rather than an IT vendor) who understands and collaborates in solving challenges throughout the Requirement, Design, Implementation, Testing and Go-live phases of SDLC. Astrix achieves results using proven methodologies during guided and collaborative sessions led by highly experienced PV pros who keep our client’s needs at the forefront of every decision. Together we determine the optimal implementation strategy and business roadmap, as well as third-party software and hosting choices that represent the best fit for our clients.

The post World Drug Safety Congress Conference Wrap Up appeared first on Astrix.

Best Practices for a Data Migration Plan and Integration of a New Pharmacovigilance System

Kevin Miller — Wed, 17 Aug 2022 11:58:20 +0000

Ensure the Accuracy of the Data Migration Plan from the Legacy System

An effective data migration plan ensures your data is accurate and thorough during the critical transfer phase from the source platform to the destination platform.

Once your team is assembled, it is crucial that they focus first on mapping, selecting, preparing, extracting, transforming, and transferring to the new system data that’s of the proper form and quality. The team must focus on creating a thorough plan for data migration that factors in appropriate compliance requirements and corporate regulations.

The 7 Guarantors of a Successful Data Migration Plan:

Identify the format of the data and its location and sensitivity.
Identify the resources both internal and external needed to perform the migration.
Develop a map document that takes you from source to destination with needed transformations.
Plan the size and scope of the project and communicate to main stakeholders.
Back up your data in case you face the need to roll the process back and try again.
Engage the necessary resources both internal and external to execute your data migration plan.
Test and validate the new system to make sure it has no connectivity gaps.²

Map Integration Points and End-to-End Flow Completely

Integration mapping enables you to both execute and document your data migration plan step-by-step.

Effective system integration requires a thorough audit of all the parts to be joined in one interconnected infrastructure. Since this aspect of the process is typically fraught with oversights, neglected, and forgotten elements, it is essential that the team members conducting this phase of the project possess firsthand experience with the various touchpoints and data streams (by process, source, pathway, and destination).

Documenting the organization’s physical connections and data flow also provides a valuable opportunity to review and analyze the logic of each connection. A best practice, whenever you must connect three or more systems and your project involves multiple use cases with interdependencies, is to create an integration map.

When connecting three or more systems, an integration map will ensure that you’ve identified the full extent of physical and logical connections. This will substantially increase confidence in both the newly integrated system and data normalization.⁴

Once all parts (software and hardware components) are integrated and working as a whole you can expect the organization to enjoy these specific benefits:

Increased Productivity – When the systems are integrated, control can be consolidated over the daily activities. This, in turn, facilitates efficiency of the entire workflow: Employees can access all applications and all the data they require from one entry point.
Increased Data Accuracy and Trustworthiness – With data now able to be updated across all parts of the system simultaneously, all departments can see the same data contemporaneously.
Accelerated Decision-Making – Siloed data stored across multiple locations is eliminated. This allows the organization to perform analytics without the need to manually download and export data to a centralized repository. This holistic view of all information enables the organization to extract valuable insights that can simultaneously inform and accelerate enhanced decision making.
Reduced Costs – System integration, when done well, typically costs less than replacing disjointed parts with a new system. Moreover, it also mitigates the operational challenges and procedural disruptions inherent in implementing an entirely new computer infrastructure.³

Why It Matters to You

An effective data migration plan ensures your data is accurate and thorough during the critical transfer phase from the source platform to the destination platform. Moreover, Integration mapping enables you to both execute and document your project step-by-step.

In this blog we discussed:

The 7 keys to a Successful Data Migration Plan.
When to leverage an Integration Map.
Key benefits of Integrating systems.

To learn more, download our whitepaper 7 Techniques to Ensure Your PV Implementation Succeeds

About Astrix

For over 25 years, Astrix has been a market-leader in delivering innovative solutions through world class people, process, and technology that fundamentally improves scientific outcomes and quality of life everywhere. Founded by scientists to solve the unique challenges life sciences and other science-based business face, Astrix offers a growing array of strategic, technical, and staffing services designed to deliver value to clients across their organizations.

The post Best Practices for a Data Migration Plan and Integration of a New Pharmacovigilance System appeared first on Astrix.