Population Health Vendors: What’s not to like?

A reader emailed me, “You have a large vocabulary.” “You ought to see the jar where I keep my adverbs,” I replied.  Clearly, I am not an imminent threat to win the Nobel Prize.

Suffering fools can be a synonym for flying. Today the term fit better than OJ Simpson’s glove, and by the time the plane was ready to depart I was ready to initiate a personal jihad.  There was no TSA line at security, so I had to shuffle through the cattle pen along with everyone else. The moving sidewalk between two of the terminals stopped moving at its midpoint.  The group in front of me, who had been riding the walkway as though it was a premium ride at 6-Flags clogged the way forward in much the same way an errant piece of plaque would have clogged an artery.

I learned that my assigned seat would not recline, but the USAIR attendant, who had the posture of a dislodged sock monkey and the look of someone who had forgotten to buy the radicchio at the supermarket, offered to sell me a reclining seat for only ninety dollars.  She smiled at me the way a fish smiles when it has been on ice all day. I sighed loudly and she said, “I’m not sure I like your attitude.”

“That’s okay,” I said. “I’m not selling it.”

I could hardly wait until we disembarked, eagerly anticipating that moment when the other passengers tried to rub two brain cells together to see if they could remember how to unlock their seatbelts, pull their luggage from the overhead compartments, and make their way to the front of the plane. I feel like I should award style points to anyone who manages to do two of the three tasks correctly.  Sometimes it seems it would be easier to teach sign language to a yak.

So, where were we?  Population Health and Population Health Applications Vendors.  Necessary and insufficient.

A word of advice. There is not a lot to be gained by losing sleep over which Population Health application to purchase.  That is not because the applications are all equal. It is because most hospitals will not get real value from their purchase. But before the vendors get all worked up over my remark, my perception of the lack of value of their product has little to do with their application.

May I try to explain?

Let’s assume a certain software vendor had written an application to predict how a baseball player would perform.  And let’s assume that a baseball team used the software.  The team tracked the performance of player A on the first game of the season, it tracked the performance of player B on game 57 and game 83. And it tracked the performance of player C on game 159.

For those who do not follow baseball, here are a few reference points.  Each team has twenty-five players, and each team plays 162 games each year.  Raise your hand if you see where this is headed.

One player makes up four percent of the team’s population.  One game represents .006% of the total number of games played in a year. A team that only has information about three of its players has no information about eighty-eight percent of its player population. Can a team that is missing information about eighty-eight percent of its population make smart and effective decisions to improve the performance of the entire team?

Can a team that only has information on three players, and is missing the information on those three players on 160-161 of the 162 games draw any meaningful conclusions about how those players will perform over the course of the season?  Can the team draw any inference from that limited data about how the performance of those players will impact the players for which they have no information?

Of course not.  Software without inclusive data that is representative of the population is worthless.

Now let’s make the following jump in logic.

Substitute Hospital for team, service population for 25 players, patients treated for the 3 players (A, B, C), and the days of collected health information for .006-.012% of the days for which data was collected on the three ballplayers.

Can a hospital who has no data on the health of so many of its stakeholders—former patients, discharged patients, and prospective patients (Consumers) draw any real inference about the health of the entire population whose health it is managing?

Can a hospital who only has data about the health of a patient for the days on which the patient was in the hospital draw any real inference about the health of that individual when it has no data for the other ninety-eight percent of the days in the year?

Of course not. Does having the best population health management application change that answer?  Of course not.  The value of every population health management application is only as good as the amount of data it has.  That application cannot make meaningful forecasts with an amount of data that is statistically insignificant.

For the application to be of any value, the hospital must have data from more of the population.  For the application to be of any value the hospital must have data on each person that includes the health of that individual over the course of the year.

The fly in the ointment is that using the current method the hospital only collects data on an individual when the individual is in the hospital. To get the data the hospital needs to effectively manage the health of the individual and the health of the population the hospital has two choices. It can either make every person come to the hospital every day, or it can create a way for all of the people to send their data to the hospital every day.

For population health management to be effective providers must find a way to make the collection and analysis of health and wellness data interactive.

What do you think?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s