The April 1 issue of Healthcare Informatics contained an article titled,  ‘Study: 75% of Adults Will Use Personal Health Records By 2020.’

We’re not in Whoville anymore. Maybe we should have stayed there when we had the chance.

I don’t believe everything I read. Given its publication date, at first, I wondered if the article was intended as an April Fools joke.  I have a ‘Show me,’ attitude. So you might believe me when I write to say that I almost fell out of my chair when I read the results of a recent study projecting that 75% of adults will use a personal health record (PHR) by 2020.

My first reaction to the article for the title to have any relevance was, “Well, somebody had better define ‘use‘ and what is meant by ‘Personal Health Record.’

The information available about someone’s health tells how somebody was. Very little is known about how they are.  Knowing that someone had their gall bladder removed three years ago–based on data contained in the EMR–says nothing about what other health conditions the person had during that period–are they showing signs of hypertension, diabetes, heart disease? Also, the EMR has no information about how that person is today.

And we know nothing about the health of anyone who is not currently under treatment.

The one health service nobody can pay for today is to manage their wellness; not from their provider, not from their payer, and not from their retail pharmacy. Nobody in an individual’s healthcare food chain can tell an individual how they are today, and that is because nobody has any information about how an individual is.

And what is most ironic about this issue is that current and habitual health data exists for millions and millions of people.  The problem is that nobody other than the individual who is collecting that data even knows that it exists.  That data is collected daily through health apps and wearable devices. But nobody in a clinical organization can see it, assess it, or respond to it.

All of that data could be turned into information.  It could be used to populate a PHR.  Individuals cannot ‘use’ the data.  If I have a heart attack during a run, RunKeeper won’t tell me to sit down and dial 911.  It will tell me I am still 2,000 steps short of my daily goal.

For a PHR to be effective it has to be relevant.  For it to be relevant it has to be current, and it has to contain habitual data. It has to be linked to a cognitive tool that can provide proactive information like, “Based on the data we collected about you over the last 4 months, we’ve noticed that your resting heart rate has increased by 30 beats per minute, as had your blood pressure. We think you need to be examined for hypertension.”

Without that, people are simply collecting numbers.