Smarter Data, Simpler Tools: A New Path for Specialty Pharmacy Optimization

An expert shares a pragmatic approach to tech adoption in infusion centers—starting with simple, high-impact data tools like Power Query that deliver return on investment without the overhead of complex artificial intelligence (AI) systems.

Please introduce yourself by stating your name, title, and any relevant experience you’d like to share.

Chris Hilger: I'm Chris Hilger. I'm the CEO of SolisRx. I started out my career as an engineer, and I've always been really drawn to challenging problems.

As an engineer, I really like structure, but in health care, very few things are structured, and the workflows are very complex. That's what drew me to the space that we operate in now, which are these mid-size specialty practices, particularly ambulatory infusion centers and specialty pharmacy.

Over my 10-year journey in health care, including getting a master's in health data science at Harvard, I noticed we spent a lot of time trying to optimize machine learning models and generative artificial intelligence (AI) solutions, but oftentimes it's not really worth the time, effort, or money to do that. A simpler solution can get you most of the way there at a fraction of the cost.

Can you please provide an overview of your presentation at NICA 2025?

Hilger: The educational session was specific to how ambulatory infusion centers (AICs) or specialty pharmacies can walk before they run. Before they make a huge investment in generative AI technology, which has merit, you can get a lot of incremental value, including better return on investment (ROI) by starting with simpler things first. Look at your current workflows and claims analytics, and instead of using basic Excel, you might want to look at this other tool called Power Query, which now lives inside Excel and allows for the ability to perform data transformations like you would do in Microsoft SQL—but without needing to write actual code.

Excel came out in 1985. Power Query came out in 2016. It's a more contemporary tool, and there's a ton of value. It's a free license that comes in Excel today, and it's something that your team could pick up in a matter of a few weeks and begin driving immediate value.

I walked through the limitations of Excel and why Power Query is a better alternative. What makes it a better is the concept of getting all your data into a centralized tracker—a centralized source of truth—and that could literally be an Excel workbook, all enabled with Power Query. You're creating a data pipeline without writing code, one that can pull in fresh data and clean it up in a repeatable process.

If there are errors that occur, you can more easily document what's going on. There's tighter error handling and documentation that's associated with it. You can eliminate some of the errors that come from manual workflows or manual data entry.

How can infusion center administrators or clinicians, who may not have a technical background, start streamlining data workflows without heavy IT involvement?

Hilger: The beauty of Power Query is that it does not require any IT involvement because it's baked into excel today. It doesn't require any additional licensing or governance. A practice manager or a clinician within an AIC could start using it without needing any additional involvement from other teams.

There is a ton of great information on YouTube, and there are other training courses online that you can find. Even if you're not super data savvy, if you can do VLOOKUP, you can get 5 times more capability out of Power Query without bending over backward to make the data bend to your will. It's something that you could pick up in a week or two.

What kinds of ROI metrics can practices realistically expect to improve—whether operational efficiency, payer mix insights, or treatment-level profitability?

Hilger: Power Query could be used to gain visibility into all of those. Many of those metrics you mentioned require bringing in data from multiple data sets, and that's a real limitation of Excel. If you're trying to bring in multiple flat files or multiple CSV files, you have to join it together somehow, typically through the use of a VLOOKUP.

Then you run into all these edge-case errors where, say, for some reason, you have to join a patient name. Sometimes that patient might be Chris Hilger in your electronic health record (EHR), but in your practice management software, it's Christopher Hilger. In Excel, that's a tricky problem to work through.

In Power Query, there is a feature called fuzzy matching. You can set the threshold for how similar the names are and then use that to determine whether they need to be an exact match or can be, for example, 90% similar or 80% similar. It makes it a lot easier to deal with fragmented or messy data.

The key is being able to easily combine multiple data sources and then doing that on a daily or monthly basis. The trick is to have a centralized tracker, which is enabled by Power Query. In my demo, we did a breakout session analyzing claims data and looking across different payers to calculate the total number of visits and the average charge lags—the days between the date of service and when the charge first went out; the days between the date of service and when the first payment was received, for payment lag; and then the percent paid of net revenue, which is the payments divided by the net revenue after adjustments. We built these in a 20-minute session.

Once Power Query is set up, you can send that same file to your team and they don't even need to know Power Query. It results in a pivot table. If they wanted to slice it, not just by payer, but by CPT code or location or provider, that's all there in the Power Query data model. Historically, to have that kind of flexibility and granularity in the reporting, you would have to set up some type of data lake—or at the very least, Power BI. Now, you can do it within Power Query itself.

What’s one simple but high-impact change you recommend infusion centers implement the day after your session?

Hilger: Start learning Power Query and set up a centralized tracker—or centralized source of truth—to manage your claims data, pull in claims at the encounter level, and track your payments, adjustments, and charges by transaction date. That would be my strong recommendation.

We have a roadmap for how to think about becoming more sophisticated with data. The very first step before making any other IT investments involves getting a unified source of truth that is aggregating data across more than one system. For example, pulling data from your EHR billing software, practice management software, or accounting software—if you have it—so that everybody across the company has visibility into whatever the workflow is as well as high-level metrics. That way, no one says, "When I go into the EHR, I see 10 000 encounters, but I go into this other system, and I see 8700 encounters."

I would encourage audiences to lean into technology—but do so through the lens of understanding what business problem you want to solve first. A lot of people will receive a mandate from leadership, an investor, or someone else saying, "Hey, how are you using AI? How are you using technology?" It's this idea that AI is just going to solve all your problems.

Our strong recommendation is that technology is great—I’m a technologist myself—but you should take a more pragmatic approach and make incremental investments, starting with a unified source of truth or workflow tracker. Get visibility into that, get used to the data that's coming out of it, and then start to move up from there.

By following this process, you're going to get to action faster and achieve a higher ROI, rather than dumping a bunch of money into a very complex platform or model and then having to go back to the basics.