How to mine your own business in the realm of data


[See also: Data-centric security a first step for physicians' mobile device strategies]

 

625s34a027f7e99t34y7910.

Safety in numbers -- a common comfort and, according to Nate Moore MBA, CPA, CMPE of Moore Solutions, a common theme when it comes to quantifying and correlating data in a practice management system.

"The days of being able to kind of eyeball it and guess -- if they ever were around -- boy, they're ending, and what we want to do is have safety and data in numbers and actual facts rather than just trying to guess how our practice is doing," Moore said in the beginning statements of his MGMA conference session "Getting Your Data Out of a Mine
 and Singing Like a Canary."

"What I want to do is talk about the big picture: what you ought to be doing with your data. To get your data out of wherever it is -- buried in the mine or wherever -- and get it singing about what's important to your organization," Moore continued, channeling the guise and gusto of a garish Jersey mob man.

But it wasn't long before Moore's gangster was lassoed by a southern declaration: "Dang it, your data is an asset."

And indeed, when positioned and organized as it should be, the bellow of data -- country twanged or city smug -- can travel to all corners of a given practice and potentially beyond. Unfortunately, most practice managers and other practice constituents today aren't getting that sweet talk from either their data or the systems hording it.

"There's a lot of frustration out there trying to get the data that you want out of your PM system," Moore said. He utilized the results of a recent FACMPE survey to bolster that point: "More than half [of practice managers] are trying to go some place else to get data out of their PM system to get safety in numbers to run their practice. If you ask the same question about the EMR, you're going to see the same kind of thing -- where again, a little less than half think it's okay; about half are either not very satisfied or not at all satisfied with what they get out of their EMR, but they don't have a dashboard because there isn't one. They've struggled to get data and trying to make actual decisions about what's going on with their data and their practice."

Moore's elixir for this prominent PM vexation? Mine your own business, of course.

The art of data mining is defined by wordnetweb as such: Data processing using sophisticated data search capabilities and statistical algorithms to discover patterns and correlations in large preexisting databases; a way to discover new meaning in data.

While the complexities there seem overwhelming at first glance, it's important to break down the mining process of a PM system branch by branch, Moore said -- only then can one appreciate the tree in its entirety.

"So where's your low hanging fruit? Typically it's charges, payments and adjustments -- the basic financial data that's in your PM system. If you haven't started with data mining, that's probably the place to start."

Once the financial aspects have been thoroughly considered, the second data mining source doesn't fall far from the tree.

"When you have that [financial] data," Moore said, "you might look at more PM data. I'm talking about things like appointments and, around that, things like no-shows, or marketing data or patient demographics or things like that. That's something you might think about mining out of your practice management system."

Further up the trunk, other system-specific tools begin to see the light of day: "As we kind of move up the tree here, EMR data, clinical data [come into play]. What kind of protocols or procedures or side effects or medication doses -- what can I do to understand what's going on with the clinical side of my practice. Along with that, account and cost data. Nobody seems to have their accounting data expenses in the same computer system as their revenue data. Very few other industries in America have that problem. We've go to data mine if we're going to get those together."

Lastly, the leaves of census, phone and third party data should have their moments in the sun. Thereafter, it's important to break down each data set section by function and account for/ask the appropriate questions in a fashion similar to that practiced by Moore below:

  • Financial: Collections, billed charges, accounts receivable, accounting software/costs.
  • Marketing:  Where are new patients coming from? How much revenue does my practice earn per new patient? Is the practice referring out enough procedures to start providing the service in-house?
  • Contracting: Analyze actual reimbursement vs. contracted reimbursement; if a payer has approved every pre-auth for 2 years can we simplify the process? Can my outcomes data prove superior quality to justify a better contract?
  • Compliance/Audit: Benchmark E&M Coding levels by provider by subspecialty, cash controls/reconciliation, HIPAA.
  • Staffing: Can I use appointment data to schedule support staff by location? How much more efficient are doctors with midlevels than doctors without midlevels?
  • Clinical: Which providers are meeting meaningful use requirements? To what extent are doctors complying with practice treatment protocols? Conservative care guidelines?
  • Balanced scorecard: Quality/Patient satisfaction surveys; Calculate practice value Employee turnover.
  • Forecasting: Can I use appointment and/or billed charge data to forecast future procedures? Can I use historical payer mix and reimbursement data to project future cash flows? Can I analyze next appointment data to project demand for new providers? How can I forecast demand for new or expanded locations?

Moore also shared the following results practice managers should glean from the data mining process:

  • Data to be mined is the back end of your application, accessed with and stored in SQL (Structured Query Language: Programming language for managing data in relational databases).
  • Use ETL (Extract, Transform, and Load) to get your data into the data warehouse.
  • Join/Combine data with common fields.
  • Achieve tradeoff between more data and more speed.
  • Start small.

And always remember your data is an asset, dang it -- leverage accordingly.