pharmacovigilance Articles / Blogs / Perficient

Agentic AI for Real‑Time Pharmacovigilance on Databricks

David Callaghan — Wed, 01 Oct 2025 18:12:02 +0000

Adverse drug reaction (ADR) detection is a primary regulatory and patient-safety priority for life sciences and health systems. Traditional pharmacovigilance methods often depend on delayed signal detection from siloed data sources and require extensive manual evidence collection. This legacy approach is time-consuming, increases the risk of patient harm, and creates significant regulatory friction. For solution architects and engineers in healthcare and finance, optimizing data infrastructure to meet these challenges is a critical objective and a real headache.

Combining the Databricks Lakehouse Platform with Agentic AI presents a transformative path forward. This approach enables a closed-loop pharmacovigilance system that detects high-quality safety signals in near-real time, autonomously collects corroborating evidence, and routes validated alerts to clinicians and safety teams with complete auditability. By unifying data and AI on a single platform through Unity Catalog, organizations can reduce time-to-signal, increase signal precision, and provide the comprehensive data lineage that regulators demand. This integrated model offers a clear advantage over fragmented data warehouses or generic cloud stacks.

The Challenges in Modern Pharmacovigilance

To build an effective pharmacovigilance system, engineers must integrate a wide variety of data types. This includes structured electronic health records (EHR) in formats like FHIR, unstructured clinical notes, insurance claims, device telemetry from wearables, lab results, genomics, and patient-reported outcomes. This process presents several technical hurdles:

Data Heterogeneity and Velocity: The system must handle high-velocity streams from devices and patient apps alongside periodic updates from claims and EHR systems. Managing these disparate data types and speeds without creating bottlenecks is a significant challenge.
Sparse and Noisy Signals: ADR mentions can be buried in unstructured notes, timestamps may conflict across sources, and confounding variables like comorbidities or polypharmacy can obscure true signals.
Manual Evidence Collection: When a potential signal is flagged, safety teams often must manually re-query various systems and request patient charts, a process that delays signal confirmation and response.
Regulatory Traceability: Every step, from detection to escalation, must be reproducible. This requires clear, auditable provenance for both the data and the models used in the analysis.

The Databricks and Agentic AI Workflow

An agentic AI framework running on the Databricks Lakehouse provides a structured, scalable solution to these problems. This system uses modular, autonomous agents that work together to implement a continuous pharmacovigilance workflow. Each agent has a specific function, from ingesting data to escalating validated signals.

Step 1: Ingest and Normalize Data

The foundation of the workflow is a unified data layer built on Delta Lake. Ingestion & Normalization Agents are responsible for continuously pulling data from various sources into the Lakehouse.

Continuous Ingestion: Using Lakeflow Declarative Pipelines and Spark Structured Streaming, these agents ingest real-time data from EHRs (FHIR), claims, device telemetry, and patient reports. Data can be streamed from sources like Kafka or Azure Event Hubs directly into Delta tables.
Data Normalization: As data is ingested, agents perform crucial normalization tasks. This includes mapping medical codes to standards like RxNorm, SNOMED, and LOINC. They also resolve patient identities across different datasets using both deterministic and probabilistic linking methods, creating a canonical event timeline for each patient. This unified view is essential for accurate signal detection.

Step 2: Detect Signals with Multimodal AI

Once the data is clean and unified, Signal Detection Agents apply a suite of advanced models to identify potential ADRs. This multimodal approach significantly improves precision.

Multimodal Detectors: The system runs several types of detectors in parallel. Clinical Large Language Models (LLMs) and fine-tuned transformers extract relevant entities and context from unstructured clinical notes. Time-series anomaly detectors monitor device telemetry for unusual patterns, such as spikes in heart rate from a wearable.
Causal Inference: To distinguish true causality from mere correlation, statistical and counterfactual causal engines analyze the data to assess the strength of the association between a drug and a potential adverse event.
Scoring and Provenance: Each potential ADR is scored with an uncertainty estimate. Crucially, the system also attaches provenance pointers that link the signal back to the specific data and model version used for detection, ensuring full traceability.

Step 3: Collect Evidence Autonomously

When a candidate signal crosses a predefined confidence threshold, an Evidence Collection Agent is activated. This agent automates what is typically a manual and time-consuming process.

Automated Assembly: The agent automatically assembles a complete evidence package. It extracts relevant sections from patient charts, re-runs queries for lab trends, fetches associated genomics variants, and pulls specific windows of device telemetry data.
Targeted Data Pulls: If the initial evidence is incomplete, the agent can plan and execute targeted data pulls. For example, it could order a specific lab test, request a clinician chart review through an integrated system, or trigger a patient survey via a connected app to gather more information on symptoms and dosing adherence.

Step 4: Triage and Escalate Signals

With the evidence gathered, a Triage & Escalation Agent takes over. This agent applies business logic and risk models to determine the appropriate next step.

Composite Scoring: The agent aggregates all collected evidence and computes a composite risk and confidence score for the signal. It applies configurable business rules based on factors like event severity and regulatory reporting timelines.
Intelligent Escalation: For high-risk or ambiguous signals, the agent automatically escalates the issue to human safety teams by creating tickets in systems like Jira or ServiceNow. For clear, high-confidence signals that pose a lower operational risk, the system can be configured to auto-generate regulatory reports, such as 15-day expedited submissions, where permitted.

Step 5: Enable Continuous Learning

The final agent in the workflow closes the loop, ensuring the system improves over time. The Continuous Learning Agent uses feedback from human experts to refine the AI models.

Feedback Integration: Outcomes from chart reviews, follow-up labs, and final regulatory adjudications are fed back into the system’s training pipelines.
Model Retraining and Versioning: This new data is used to retrain and refine the signal detectors and causal models. MLflow tracks these updates, versioning the new models and linking them to the training data snapshot. This creates a fully auditable and continuously improving system that meets strict regulatory standards for model governance.

The Technical Architecture on Databricks

The power of this workflow comes from the tightly integrated components of the Databricks Lakehouse Platform.

Data Layer: Delta Lake serves as the single source of truth, storing versioned tables for all data types. Unity Catalog manages fine-grained access policies, including row-level masking, to protect sensitive patient information.
Continuous ETL & Feature Store: Delta Live Tables provide schema-aware pipelines for all data engineering tasks, while the integrated Feature Store offers managed feature views for models, ensuring consistency between training and inference.
Detection & Inference: Databricks provides integrated GPU clusters for training and fine-tuning clinical LLMs and other complex models. MLflow tracks experiments, registers model versions, and manages deployment metadata.
Agent Orchestration: Lakeflow Jobs coordinate the execution of all agent tasks, handling scheduling, retries, and dependencies. The agents themselves can be lightweight microservices or notebooks that interact with Databricks APIs.
Serving & Integrations: The platform offers low-latency model serving endpoints for real-time scoring. It can integrate with clinician portals via SMART-on-FHIR, ticketing systems, and messaging services to facilitate human-in-the-loop workflows.

Why This Approach Outperforms Alternatives

Architectures centered on traditional data warehouses like Snowflake often struggle with this use case because they separate storage from heavy ML compute. Tasks like LLM inference and streaming feature engineering require external GPU clusters and complex orchestration, which introduces latency, increases operational overhead, and fractures data lineage across systems. Similarly, a generic cloud stack requires significant integration effort to achieve the same level of data and model governance.

The Databricks Lakehouse co-locates multimodal data, continuous pipelines, GPU-enabled model lifecycles, and governed orchestration on a single, unified platform. This integration dramatically reduces friction and provides a practical, auditable, and scalable path to real-time pharmacovigilance. For solution architects and engineers, this means a faster, more reliable way to unlock real-time insights from complex healthcare data, ultimately improving patient safety and ensuring regulatory compliance.

Conclusion

By harnessing Databricks’ unified Lakehouse architecture and agentic AI, organizations can transform pharmacovigilance from a reactive, manual process into a proactive, intelligent system. This workflow not only accelerates adverse drug reaction detection but also streamlines evidence collection and triage, empowering teams to respond swiftly and accurately. The platform’s end-to-end traceability, scalable automation, and robust data governance support stringent regulatory demands while driving operational efficiency. Ultimately, implementing this modern approach leads to better patient outcomes, reduced risk, and a future-ready foundation for safety monitoring in life sciences.

Perficient is a Databricks Elite Partner. Contact us to learn more about how to empower your teams with the right tools, processes, and training to unlock your data’s full potential across your enterprise.

Regulatory Science and PV: What I learned from Humpty Dumpty

Kari Blaho-Owens — Wed, 07 Sep 2022 14:50:12 +0000

For as long as I can remember, Humpty Dumpty had a great fall that put him in a tragic state. This rhyme has always been one of my favorites. The story first appeared in 1870, in James William Elliott‘s National Nursery Rhymes and Nursery Songs. Humpty Dumpty is a memorable and versatile teaching lesson, even for adults.

In my previous post, we looked at what the Three Little Pigs taught me about risk management. Humpty Dumpty also taught me something quite important to pharmacovigilance (PV) and regulatory compliance. (Truly, I’m not stuck in childhood rhymes, but I am amazed at how applicable they are to any stage of life.)

The first question that stands out for me is, “Why is he sitting on a wall?” His proportion wasn’t correct for sitting on the wall and fell into the category of “horrible parent-based outcomes.” In today’s regulatory environment, governing pharmacovigilance and drug safety, it’s easy to see how we can become our own Humpty Dumpty.

Choosing a PV or regulatory workflow isn’t stable and allocating our resources disproportionally can be a disaster. What happens when the system fails or becomes overrun? We take Humpty’s fall.

As the rhyme goes, sometimes all the kings’ soldiers and all the kings’ men can’t put your PV/Regulatory system back together again.

So, how do we stabilize in pharmacovigilance and regulatory practices in this world of constant movement and change?

First and foremost, what is your regulatory environment and your geographical footprint? While there are commonalities; not all are a fit. The United States Food and Drug Administration has different centers for different drug categories; EMA and local country affiliates operate differently. Just because one authority has a particular guideline does not mean it will have an iteration from each authority. If you work across regulatory areas, these categories include:

Human drugs
Tobacco
Animal drugs
Animal vaccines
Medical devices
Biologics
Animal parasiticides

Regulatory affairs is infinitely more complicated. Recently, all the 2013 GxP documents – which was the first move from Volume 9A in EMA in 2013 – were updated. The Middle East, South America and APAC are all evolving at record speed. This should give us the message that the wall isn’t stable.

We have moved from a PV system that was based on retrospective analysis to one that expects proactive pharmacovigilance.

This requires faster, more efficient, and more intimate knowledge of our products and the risk benefit paradigm. That also means understanding and complying with the regulations for the full life cycle of our products.

How do you do that without falling off the wall?

None of us at three years of age, while hearing of Humpty’s woes, brilliantly announce to our parents that we are going to be “regulatory affairs or pharmacovigilance” subject matter experts. With the complexity of the new animal health regulations and the updates to the EMA human regulations, regulatory affairs has become more of a specialized regulatory science. That means understanding more, keeping abreast of country based specific regulations, making sure that our PV systems and organizations are compliant, and using the regulations in an efficient manner to develop a regulatory strategy.

We can either ask for an army-sized budget or embrace technology solutions to help. What is available? What is a good fit? What are the end-goals of introducing technology into regulatory affairs and PV?

Faster and more accurate collection of regulatory information
Separating relevant vs. non-relevant information
Approval and distribution of PV regulatory information
Distribution of information needed for regulatory strategy and design of submission planning
Ready access to our PV data to rapidly identify potential signals, opportunities for expansion and risk mitigation

The reality for most is that they don’t have enough resources to manage ever-changing regulatory documents – to find, digest, and distribute relevant regulatory information. Each organization wants and needs optimal compliance within the regulatory environment in which they operate. We know that the strategy (both before and after marketing) is unique to the product, indication, geography, and license. The technology should augment and accelerate marketing and compliance. Use technology to automate regulatory documents from global agencies to maximize the use of data, comparative analytics, and automate parts of regulatory submissions.

At Perficient, those technology solutions are our sweet spot. We get it; keeping up with regulations to protect currently marketed assets is difficult and putting together a regulatory strategy to get a drug or device to market is even more complicated. There is also the PV data and analytics.

Technology solutions don’t have to be painful or overly expensive; we believe in fit for purpose.

Check out the diagrams below, contact us for more information (you can also just give us a holler – we all work remotely, so we’ll hear you). Let us show you how to simplify regulatory affairs into a science and survey your PV data based on real time data that is relevant to your unique area. Whether it be animal or human health, we can make it simpler, more accurate, and help you get more with less using technology solutions.

Moreover, your regulatory and PV group will love you for it.

Technology Automation for Surveying Global Regulations and PV Impact:

Kari Blaho-Owens, EMT, Ph.D. is the Director of PV and RA for Healthcare at Perficient. She lives and works in Montana. Kari is a firm believer that finding workable solutions to tough Regulatory, MI, and PV solutions can be found at the end of her fly line. She loves fly fishing, donating her spare time to serve others as a volunteer EMT, and exploring the vast beauty of the state… with the goal of not being eaten by a Grizzly bear.

For PV, MI, call center, and RA conundrums, contact us for workable solutions.

Pharmacovigilance and Medical Information Call Centers by “I Love Lucy”

Kari Blaho-Owens — Tue, 08 Mar 2022 18:08:50 +0000

The medical information (MI) call center is part of a complex pharmacovigilance system. It may be the only opportunity to gather adverse and product quality data. It’s imperative that it operate efficiently. With a growing number of calls, it’s easy for a center to overrun. That can lead to team burn out, long wait times, dropped calls, and decreased overall quality of your PV system.

This isn’t an isolated scenario. Many companies are struggling to meet the growing demand from patients and HCP. So, after a day of heavy caseloads and difficult calls, late night re-runs are comforting. I became a fan of “I Love Lucy” reruns, trying to close the day with a bit of humor.

One of my favorite episodes is pictured above. Lucy and Ethel decide to take jobs in a candy manufacturing company on the assembly line, wrapping each piece before the chocolates get to the box. At first, it’s manageable, but then the pace picks up as does the volume. They soon become over-run. Sound familiar? As a stop gap, they have a great solution – eat the chocolates or stuff them down their uniforms. That is only temporary; something must give or change. Just like in a runaway MI call center, we must change our paradigm.

A holistic pharmacovigilance system in many ways reminds me of this episode. We’ve all felt the pressure of too many calls and too many serious cases. We’ve all been overwhelmed with emails, calls, and IMs, while trying to give the best service to callers. So, let’s focus on the chocolate assembly of a call center. Whether it’s for a patient/consumer question or for a health care provider, the call center in Pharma requires patience, good training, and current drug information.

The call center is the face of your company. How the call center functions makes immediate impact. No one wants calls backed up or put on hold; just like no one wants to see chocolate that never makes it to the box. It’s not a sustainable business model.

Options

To keep medical information (MI), or any call center, flexible and providing the same or a higher level of quality, we turn to our scary friend – technology. Why is technology scary? Because most of us don’t have a robust background that gives us a deep understanding of what technology can do in a call center setting.

We now have tools such as artificial intelligence (AI), natural language processing (NLP), and automation that have matured and can be layered over your current workflow.

See the below configuration for a call center workflow, and the addition of adding multiple contact methods and a human like empathetic virtual agent who can take voice to text, text to text, and triage calls to the proper front-line agents.

The data collected can then be parsed, AEs, concomitant medications, and other case data collected, formulated into an E2B XML file, and uploaded directly for processing in your AE database. The call itself (audio or text or other) can be uploaded as the source document. The beauty of this model is that the NLP and AI train against your data, using your procedures, your coding conventions, and gradually can take more calls over time.

Customer Interaction Flow

If you don’t have the resources internally (including a team to validate against your data); choose a partner who understands the pharmacovigilance and MI needs, has a strong technical team, will support the solution for you in a cloud-based environment, and will manage and monitor the solution for you; including providing a hosting service where users can call in with “how-to” questions. The ideal partner will have a team that speaks on everyone’s level, so that there is no chance for technology and PV/MI speak to be misinterpreted. Choose a partner who listens to your users.

A good partner will start with a Phase 0 to understand how you are currently working, what your goals are, and the ideal user experience. An effective partnership interfaces with existing technology and workflow practices (unless you want to change it).

A partner should be able to communicate in a meaningful way identifying pain points, has deep experience with various technology solutions, and listens to your requirements.

We start with developing a road map. Will the MI center see immediate relief? Is it adaptable? Is it compliant with regulatory guidance, such as FDA-2019-N-1185, FDA Guidance for Industry No. 0910-0856?

Sample Roadmap

Then we fit for purpose. Does the solution for your call center maintain the user experience? Is it omni-channel capable, flexible, scalable? Does it have down-stream benefits in your PV/PQC workflow? Does it meet the guiding principle of implementing computer systems validation and the tenants for good machine learning and artificial intelligence?

Will your partner support you during implementation and after? No one wants a “load and go” partner, one who disappears once things are completed. A good partner will “stay and play” by maintaining your system, keeping you apprised of other improvements, and monitoring to mitigate technical issues.

What does the return-on-investment look like in efficiency and ability to absorb higher call volumes without a corresponding increase in staff?

Return Calculations

It’s not just the technology. A call center that utilizes these tools is not designed to replace staff, but to augment their ability to provide quality service to all types of callers. For and MI call center, the reach of benefits extends far beyond the call center itself. Parsing data and having it uploaded directly into the database saves on data entry. Depending on what AE database you are using, there are many more points where automation can be introduced. It’s a crawl, walk, run fly approach.

PV Challenges And Rewards

Perficient has an uncompromised commitment to delivering custom solutions for complex problems, especially in the pharmacovigilance and regulatory science space. Please feel free to reach out directly for a run through of our demo MI platform and how it can help optimize your MI center now and grow into the future. kari.owens@perficient.com

Kari Blaho-Owens, Ph.D., is the Director of PV/RA and Safety within Perficient’s Healthcare business unit. She was born and raised in the southern US (hence her southern humor) and now lives with her family in northwestern Montana.

10 Questions & Answers About Using Virtual Agents for Medical Information Sharing

Madeline McDermott — Fri, 21 Jan 2022 20:03:03 +0000

In our recent webinar, “Improve Medical Product Information Sharing With Virtual Agents,” Prabha Ranganathan, Kari Blaho-Owens, and Nico Frantzen discussed the ways that implementing virtual agents in life sciences companies’ call center operations can help to promote pharmacovigilance and more effectively service customers.

Below are some of the questions we answered during the webinar and some that we didn’t have a chance to answer live.

1. What are the challenges with speech-to-text capabilities?

“Speech-to-text is all around us nowadays — in our phones, at home with your Alexa or Google Home devices, etc. It is a part of AI that has made tremendous progress, thanks to Deep Learning models’ abilities to process complex signals like the human voice. The challenge still remains in specialized domains or applications where uncommon or complex vocabulary, or jargon is used. In these domains, specialized and custom training is often required.” – Nico Frantzen, Director, Artificial Intelligence

2. What do you see as the challenges with different languages and dialects?

“There are still some significant challenges around models adapting to particular dialects and accents. Therefore, vendors and data science teams must continuously monitor, test, and validate their models against various large datasets that have to be regularly augmented with new sample audio.” – Nico Frantzen, Director, Artificial Intelligence

3. Does it make sense to integrate a virtual agent with WHODrug/MedDRA/VedDRA dictionary?

“Absolutely. The FDA has come out with guidance about AI/NLP/ML parsing data and coding AEs with MedDRA or VedDRA. It lends harmonization to the organization but does require revisiting with dictionary updates. The bonus is the downstream in the PV workflow; this task is already done and only requires review.” – Kari-Blaho-Owens, Ph.D., Director, Safety, Pharmacovigilance, and Regulatory Affairs

4. What types of things can be done to improve call drops? Are there any solutions for reaching back out to call drops? How do you ensure call continuation across channels?

“There are a number of techniques that can be used and implemented; for example, programs can be designed to detect a call drop and send an automated SMS to follow-up with the user to resume their call, or continue the conversation via chat or SMS. This is why choosing the right omni-channel platform is just as important as the choice of a conversational AI platform. The ability to switch seamlessly between channels is a very important capability to enable in your solution.” – Nico Frantzen, Director, Artificial Intelligence

5. We have a large volume of scientific resource documents. How do we make conversations friendly?

“The scientific resource documents need to be parsed and stored in other formats (like knowledge graphs) so that it is easier for AI/ML models to search and retrieve the right information for the interactions. The format and structure of the existing scientific resource documents should be examined to see how they can be converted to graph format, and this step will be automated to the fullest extent possible so that a large number of resource documents can be converted with less manual intervention.” – Prabha Ranganathan, Life Sciences Strategist

6. What serves as your source documents?

“The AI/NLP/ML can take voice and convert to text, or any other avenue of communication and produce an accurate source document that can be incorporated into an E2B XML and uploaded directly into the AE database.” – Kari-Blaho-Owens, Ph.D., Director, Safety, Pharmacovigilance, and Regulatory Affairs

And how do you make sure that AEs with source documents get into your safety database?

“As the case is updated after follow-up, the solution records outbound and inbound communication, source documents, etc.”

7. How do you maintain MI knowledge into Virtual Agents?

“A combination of scientific resource documents in the format of knowledge graph and training of AI/ML models to understand the conversation will enable the virtual agents to search/retrieve/maintain MI knowledge. A business rules engine combined with product knowledge from scientific resource documents will enable the Virtual Agent to have an intelligent human-like conversation with the customer. The primary source of MI knowledge will be the knowledge repository.” – Prabha Ranganathan, Life Sciences Strategist

8. Are we aware of validation requirements – described in multiple FDA guidances?

Yes, most of the validation requirements are identical to the guidances for computerized systems validation. There are other guidances published by global regulatory authorities that set the precedence for expectations for the validation of solutions that utilize AI, NLP, and ML. – Kari-Blaho-Owens, Ph.D., Director, Safety, Pharmacovigilance, and Regulatory Affairs

9. Do disaster recovery (DR) and business continuity (BC) plans come standard with such a solution (since this is cloud-hosted), or do we interface with the company’s DR?

“Perficient is capable of hosting this solution and maintaining it for our clients; we have a DR/BC plan in place to minimize the impact of an unlikely production outage. This is standard in our hosting environment. We encourage all of our partners to have a separate DR environment because it limits potential data loss and loss of communication.” – Kari-Blaho-Owens, Ph.D., Director, Safety, Pharmacovigilance, and Regulatory Affairs

10. Would Perficient provide a list of SOPs to go with the solution (as there will be a need for new SOPs), or is it up to the customer?

“As part of developing a customized roadmap and gathering requirements, Perficient is happy to author SOPs. Standard with any solution, we offer training, support after go-live, and user guides. Perficient will also partner with you in the event of a regulatory audit.” – Kari-Blaho-Owens, Ph.D., Director, Safety, Pharmacovigilance, and Regulatory Affairs

Curious to learn more? Watch the webinar recording here or below.

5 Takeaways From Our “Improve Medical Product Information Sharing With Virtual Agents” Webinar

Madeline McDermott — Mon, 17 Jan 2022 17:39:59 +0000

Our life sciences and technology experts recently delivered the webinar, “Improve Medical Product Information Sharing With Virtual Agents,” where they discussed how automation and artificial intelligence can be used to optimize call center operations and improve end-user experiences.

Here are five key takeaways:

Just because a system is powered by AI doesn’t mean you have to sacrifice empathy. Systems can be designed to have a natural conversation flow and recognize whether a message is negative or positive. For example, suppose someone calls to inquire whether they should give their puppy more medicine because it vomited shortly after taking its dose. In that case, an AI system can be programmed to compassionately respond with, “I’m sorry to hear that,” before searching its database for the accurate medical advice to relay.
AI isn’t meant to replace the call agent, but rather, its purpose is to allow them to focus on higher-value escalations. The severity of adverse reactions and urgent HCP questions can vary significantly from case to case. Conversations with live agents can be prioritized and triaged to quickly route those calls to the best call center resource.
AI can help address specific call center challenges. Many pharmaceutical and medical call centers experience extremely high call volumes, especially at certain times of the year (i.e., flu season), resulting in unacceptably long wait times and excessive dropped calls. Call center agents are also challenged with having to meticulously record their interactions, which only exacerbates these issues. AI can use NLP to help address these issues and record interactions. The great thing is that it interfaces with current systems, works in any regulatory area (e.g., biologic, device, animal health, human health, dietary supplements), and can be deployed globally to overcome language barriers.
AI virtual agents can be fine-tuned to effectively handle various types of callers. For example, a virtual agent that speaks with doctors may be programmed to communicate using more medical vocabulary than a virtual agent designed to speak with patients.
An AI virtual agent delivered in a truly omnichannel capacity will greatly reduce frictions in the user experience. Suppose someone is experiencing an adverse reaction to a medication that gives them a sore throat. In that case, a texting method of communication may be better suited for them than a phone call, and AI systems can equip users with such options. If someone is at the point of sale and has questions about an OTC product or their prescription, they could easily text with questions or concerns.

Curious to learn more? Watch the recording here or below.

Industry Slant // Kari Blaho-Owens on Montana, Aviation, and Pharmacovigilance

Madeline McDermott — Fri, 07 Jan 2022 14:40:44 +0000

Kari Blaho-Owens, Ph.D., is Perficient’s Director of Pharmacovigilance, Safety, and Regulatory Affairs. She leads a team that provides PV, regulatory, risk management knowledge, and consulting to pharma and CROs. Perficient also hosts PV-related applications, configure, validate and maintain PV adverse event databases, and reporting solutions. At the end of the day, we contribute to patient safety by providing the means to ensure that drugs are safe and effective. Kari has more than 25 years of experience building and supporting global safety and regulatory initiatives for pharmaceutical, medical device, animal health, and clinical research organizations (CROs).

Kari Blaho-Owens joins Madeline McDermott and Eugene Sefanov for a Microsoft Teams Chat

“Huntin’, Fishin’, and Lovin’ Every Day”

Madeline McDermott: First things first. Where were you born?

Kari Owens: I was born in Coral Gables, Florida.

MM: Did you spend most of your childhood there?

KO: My family moved pretty often when I was little. I lived in Mississippi and Texas, and when I was about 10 or 11, we settled in a small, rural town in Arkansas. There wasn’t even a McDonald’s nearby until I was in high school. That’s where I grew up and consider home.

MM: Where do you live currently?

Kari taking a photo in Sunday Falls, in Stryker, Montana

KO: Eureka, Montana. It’s such a beautiful mix of people, land, culture. It’s awesome!

MM: What’s the population of Eureka?

KO: There are around 1,300 people – but it’s not as rural as you’d think. We’re just very spread out.

MM: Where did you live before Montana?

KO: Before moving to Montana, I lived in Bennett, North Carolina. North Carolina really morphed in a short period of time because a lot of technology and pharma moved there.

MM: Why did you end up moving away from North Carolina?

KO: We were looking for a place that was different. I wanted to live on a river where I could fly fish, hunt, and take pictures of the amazing scenery. We were fortunate to find a place on the Tobacco River. There is a lot of history here, and the people are lovely. And it wouldn’t be possible to live here if it weren’t for my role at Perficient. The company and role allow me to work anywhere.

Kari catches a little red trout

Kari when she brought home her golden retriever puppy, Trip

MM: You must really love nature.

KO: I think that’s the thing that makes living here so awesome; you can look out at any time of the day and see all this wildlife.

We had a group of deer that came through last week. I was on my porch giving out some dry corn to the birds when I saw them. I held out my hand super still, and a doe came up and ate some of the corn out of my hand. And then she put her head on my shoulder. It was just amazing.

***

From White-Knuckle Flying to Volunteering

MM: I see that the sign behind you says, “Shelby Aviation.” Tell me about that.

KO: When I was in academia and traveling quite a bit, I was a white-knuckle flier. Let’s just say physics and math are unpleasant for me. To confront this, I thought I’d go to ground school to get a better understanding of the physics and math behind flying and how a 250,000-pound metal tube stays in the air. I thought it would make me more comfortable.

I then thought I’d buy three hours of airtime with an instructor to see if I couldn’t “kiss the dragon.” I spent the first 45 minutes sitting on my hands. My instructor finally had me take the yolk for a bit.

Something happens while flying…it’s like when you’re in a boat and you hit a backwash. You can just feel it. Once you’ve done that in the air, it’s like a sickness. There was no stopping after that.

I started flying with these guys at Shelby Aviation in Millington, Tennessee, where I had a Navy instructor who taught me more, let’s just say, new vocabulary, than I had ever heard in my life. He was an extraordinary pilot.

Kari and her air instructor at the world’s largest airshow and in front of a P51 Mustang airplane

I went on to get my private, instrument rating, then my multi-engine commercial industry rating. Next, I would like to earn my seaplane rating because there are lots of lakes in the mountains here. I’d also like to try flying a helicopter.

It’s never been just about flying for the — what they used to call the “$100 hamburger.” By the way, it’s more like the “$400 hamburger” now, when you consider inflation.

I got involved volunteering with Veterans Airlift Command, an organization that provides free air transportation to those wounded in combat, as well as their families, for medical care. I also flew with Angel Flight, an organization that flies patients to points of care for medical treatment. There was also a third group that I helped support, too, that was formed right after 9/11, and its mission was to recruit and train general aviation pilots to fly supplies or whatever else was needed in case we encountered that type of crisis again.

MM: Did you do a lot of those types of flights over the years?

KO: I did as much as time allowed. While I lived in Atlanta, we flew veterans with disabilities to Shepherd Center, a patient rehabilitation hospital for the military wounded.

When we lived in Tennessee, we flew a lot of kids to St. Jude Children’s Research Hospital.

During Hurricane Katrina, we flew a lot of supplies down from Memphis as far down south as we could to get them to the National Guard.

It was a real blessing to be able to do this. These groups opened a lot of opportunities, and through them, I have met so many incredible people.

There is a lot you can do with general aviation, and I’d encourage anyone who dreams of flying to try it. Like many things in life, it may not be what you think it is, but it has been an extraordinary privilege and blessing.

***

“No Three-Year-Old Wakes Up and Thinks ‘I want to work in pharmacovigilance.’”

MM: What was your first job?

KO: My first job or my worst job?

MM: Both!

KO: My first job was nannying and housekeeping for a family in my hometown in Arkansas. The shortest job I ever had was a waitress – it’s one of the most difficult jobs in the world – I lasted four days.

MM: What was your first corporate job?

KO: My first corporate job was with Schering-Plough in the consumer and OTC health division. Right after they approved Claritin to be sold over the counter, the company needed a pharmacovigilance group that could integrate with its medical information and global PV group. I was hired to build a team that could support Claritin and manage the legacy products in the rest of the Schering-Plough line. The company had everything from wart removers to laxatives. You can imagine the number of cases we got.

MM: How did you end up getting into the pharmacovigilance space?

KO: No three-year-old wakes up and thinks, “I think I want to work in pharmacovigilance.” I worked in the inner-city emergency department at the Elvis Presley Memorial Trauma Center as the research director in the ER. It’s a level 1 trauma center that takes patients from a four-state area. We had a research program, and that’s when I got involved working on clinical studies around toxicology, envenomations, and overdose. I also worked closely with a forensics team that we built from scratch.

There is this vast amount of data available to you when you’re right there at the point of patient care. For example, sometimes patients would come in because of out-of-control blood pressure, and we didn’t immediately know the reason for that. We started collecting data on this and realized that they had back pain and were taking nonsteroidals, which we now know interact with the disease and hypertension medications.

Observing this data and identifying these trends in this position led me to pharmacovigilance.

Pharmacovigilance is not just about observing, although it used to be. Today, it’s about being proactive and recognizing patterns that might indicate something is unusual and worth investigating further.

The faster we can get information out there, in a digestible format, to the people who are caring for patients, the better patient care that we will be able to provide.

I mean, look at the miraculous way pharma has been able to roll out the COVID-19 vaccines. In record time, we have three safe and effective vaccines. Twenty years ago, this wouldn’t have been possible.

It’s hard to remember this when you’re in pharmacovigilance. You’re usually not there to see a patient walking out of the ER or out of the hospital. I always try to remember that, in PV, we’re touching lives in ways we don’t always see.

MM: You mentioned that you worked with forensics. Can you talk a little about that?

KO: Since we were an inner-city emergency department, we saw a lot of severe reactions from drug abuse, mostly to cocaine. I even accompanied the Shelby County DEA on drug raids and acquired samples, so my colleagues and I could study the difference in the cocaine from different locations. What’s interesting is that we found there is no dose-response effect in cocaine. This kind of data helped the police better recognize when someone in their custody needed to be seen to minimize the risk of in-custody deaths.

MM: That must have been tough to observe. Did you get burned out from that kind of work?

KO: You see a lot of patients who do the same thing time and time again. I did get burned out. Too many late nights. It does take a toll on you. Hats off to those healthcare workers who work those long hours in acute care settings. It can be heart-wrenching.

MM: Let’s talk about what you’re doing day-to-day. What is the last project you completed?

KO: Well, there’s never just one project starting or completing. We recently just finished a safety system upgrade for a client, and we created custom reports for them, so the data could be used in a more meaningful way. We also recently helped another client with their intake data flows and processes.

MM: What was the client’s intake process like, and what did we recommend?

KO: We essentially were able to simplify and expedite the flow of data and enhance compliance. We helped the client eliminate multiple redundant steps in its intake process. These steps were confusing, took too long, and had too many people involved at every step of the way. Rather than going through a more direct line from A to B, all these detours were occurring. We took away the barriers and said we would get to the same endpoint easier, faster, and more accurately.

MM: Is there anything we’re doing right now that you want readers to know?

KO: Many people, even some of our clients, don’t realize the breadth of our PV team. I’ll share two examples:

We look at the pharmacovigilance ecosystem holistically. That starts with the intake of the adverse event or at the point where medical information is distributed, and it ends with the data analysis. We build and deploy intelligent processes and solutions that save people time and help them get to valuable and digestible data quickly to do their jobs more efficiently. At the moment, we’re helping organizations understand how they can use AI and natural language processing to support their medical information and call center operations. Things like virtual agents and conversational chatbots are very hot right now.

This brings me to my next point. A lot of what we do at Perficient is education. If I’m at a pharma, device, CRO, or animal health company and work a challenging 50-60 hours a week just to keep up with data, when do I have time to think about improving how my company or I work?

Unfortunately, most life sciences companies don’t have time to focus on improvements. But it’s important to sit down and talk through a problem and expand conversations into areas that feed into the pharmacovigilance ecosystem. There is a great need to better understand the landscape, the rules, and the pain points so that we can fix things practically and logically. The fact that there are so many talented people at Perficient has really helped make such conversations conducive. Fortunately, we have people here who can look at things from various perspectives. I believe it helps set us apart from our competitors.

MM: What interests you most about the industry?

KO: I get excited about working through the complexity of the drug safety regulatory network, keeping abreast of the challenges of data, and offering common-sense solutions. I’m a nerdy scientist at heart; I love keeping up with new therapeutic areas and how PV impacts the quality of patients’ lives, and how much we depend on technology. At the end of the day, we contribute to patient safety by providing the means to ensure that drugs are safe and effective.

MM: What impact will technology have in the PV space?

KO: One example is how technology helps us to understand and analyze AE data. We will need to work smarter, not harder. While there are many technology solutions available, the key to success is pairing the right solution as a fit-for-purpose agile solution. One of the biggest wake-up calls has been the value of PV data during the COVID-19 pandemic. Many of the early treatments tried in COVID-19 patients were based on historical knowledge, gained in part from legacy pharmacovigilance data.

Vicky Green, Prabha Ranganathan, and Kari at a business dinner

***

Rapid-Fire Questions

MM: What are your favorite brands or companies?

KO: I am very picky about where something is made. There are things that I will not buy if it is made in a certain geographic area.

My favorite brand of everyday cabernet is from BV Rutherford out of Napa, and for special occasions, Rothschild from the Bordeaux region of France.

My favorite brand of aircraft is Beechcraft; they make some of the most awesome innovative aircraft.

My favorite truck brands are Ford and Toyota.

My favorite perfume brand is Bond No. 9 New York.

Kari with a VTail Beech Bananza Airplane

MM: What about your favorite genre of music?

KO: I love country music, and I absolutely love Opera – Puccini is my absolute favorite composer for Opera and Mozart for classical music. And I feel happy listening to Bluegrass like Sweet Potato Pie.

MM: Coffee or tea?

KO: Definitely tea. Iced. Unsweet. Unsweet iced tea.

MM: What is the closest chain business near you?

KO: It would be a grocery store in Whitefish, MT. There are no fast-food places here, though, just local places, and they’re phenomenal. You wouldn’t believe the kind of meat we have.

SEE MORE PEOPLE OF PERFICIENT

It’s no secret our success is because of our people. No matter the technology or time zone, our colleagues are committed to delivering innovative, end-to-end digital solutions for the world’s biggest brands, and we bring a collaborative spirit to every interaction. We’re always seeking the best and brightest to work with us. Join our team and experience a culture that challenges, champions, and celebrates our people.

Visit our Careers page to see career opportunities and more!

Go inside Life at Perficient and connect with us on LinkedIn, YouTube, Twitter, Facebook, and Instagram.

Is the Clinical Development Platform Approach Right for You?

Param Singh — Tue, 13 Apr 2021 12:27:16 +0000

Life sciences organizations typically manage their IT infrastructure and applications in silos. They implement disparate applications and manage the content and process within those applications separately. They then deploy integrations between those applications to streamline processes and share content and data with different business areas. Much of this work, including the development of integration tools and accelerators, is accomplished with the support of system integrators and consulting partners. However, the cost and complexity of clinical and safety integrations can be quite prohibitive.

Software vendors have tried to address the challenges associated with operating point solutions by creating cloud-based platforms that can replace a multitude of clinical R&D applications. These clinical development platforms, which include the likes of Oracle Clinical One and Veeva Development Cloud, are designed to encompass capabilities from traditional systems, including clinical trial management, electronic data capture, safety, electronic trial master file, and regulatory information management. The main difference by implementing platforms is that companies can optimize the collection, sharing, and tracking of data and content. They remove many manual processes by reducing data entry, reconciliation, validation, training, and user support.

This platform strategy can be extremely beneficial; however, it may not be a practical approach for all organizations to pursue. In particular, organizations with established collaborations and partnerships might find it incredibly challenging to adopt a clinical development platform in its entirety and gain all of its benefits due to how the different collaborating companies share data with each other.

Before organizations evaluate clinical development platforms, they should carefully analyze their current clinical technology landscape, map out their requirements, and develop a clear strategy that will result in a clear return on investment.

Example Evolution of an Organization Adopting a Clinical Platform Strategy

Stage 1

An organization leverages several modules/components of a clinical platform instead of implementing all functions within a unified platform. In fact, the organization doesn’t realize its clinical operations and safety/pharmacovigilance groups are using related systems from the same vendor. The organization understands processes can be streamlined by moving to a single platform, even if it’s not ready to replace all of its systems at one time.

Stage 2

In a step toward adopting key components of a clinical development platform, the organization teams up with a services provider, such as Perficient, for advisory consulting services. Working together, it is determined that the organization can immediately adopt and combine CTMS, EDC, and safety/pharmacovigilance functions by leveraging one unified platform. This will eliminate the need to replicate shared data, such as Studies, Site, Account, Contact, Addresses, etc. across systems, as well as remove the need for integrations between these systems.

Stage 3

Once the organization adopts several components of its new clinical development platform, it determines that additional clinical platform components can be adopted. This continues to reduce the need for other complex system integrations.

In-house and third-party clinical systems will continue to be part of an organization’s overall IT landscape. Ensuring these work in conjunction with the platform will be a key part of the strategy.

Contact Perficient to start your journey

At Perficient, we have the experience, skillset, and expertise to help you determine whether a clinical development platform is right for you, and if it is, create a path to move forward. Our goal is to help your organization take advantage of a comprehensive, clear strategy and the technologies that accompany it. That could be a clinical development platform in its entirety or a portion of its capabilities, with the continued use of system integrations. It could also be that this is not the right time for your company to move to a platform. Our goal is to ensure that your decisions make sense for your business.

The Clinical Development Platform Dilemma

Interest in clinical development platforms is picking up steam. And for the right reasons. These platforms substitute multiple disparate point applications – such as electronic data capture, trials management, randomization, and pharmacovigilance systems – with one unified solution.

But is a clinical development platform right for you, and if so, which one?

Download our perspective to learn the:

Brief history of clinical platforms
Potential challenges
Questions to ask before and during your search
How Perficient can help

Argus Safety, Oracle Clinical & RDC Release Notes [April 2021]

Jane Hamilton — Mon, 05 Apr 2021 13:39:38 +0000