Dashboard-Driven Population Health Management in Northwest Ohio

A year ago, our rate of diabetic patients whose hemoglobin a1c was controlled to below the level of 8 was at 34 percent; as of December 13, we ended the year with a 51-percent rate, a huge improvement. And hemoglobin a1c, or glycosylated hemoglobin, doesn’t change rapidly, so typically, you only measure it once every three months; so this is a huge improvement within one year. With regard to our patients with controlled hypertension, we went from 54 percent to 65 percent. For tobacco cessation, per counseling, we went from 22 percent to 41 percent.

Meanwhile, our rate of breast cancer screening rose from 51 percent to 59 percent, while our pneumonia vaccination rate for patients over 65 went from 23 percent to 31 percent. So we did show some significant improvement in those areas. But keep in mind that, at the beginning of 2012, we didn’t have any measures at all. So we put in an EHR, and needed to put in a dashboard that was meaningful and easy to use. And Explorys is very simple to use and meaningful. I can train someone in a half-hour.

So we went from not even knowing how we’re doing to getting the data, and then the practices started doing things to move those numbers. What happened was that, for the first time, our physicians knew which patients needed breast cancer screening or pneumonia vaccination, and were able to send out letters and contact patients to get those things done. We were finally starting to do some actual population health management.

This is a journey of a thousand miles, isn’t it?

Absolutely, and it’s one that can only be done one step at a time. And that’s why we didn’t take on 20 different measures all at once for our patient-centered medical home practices. And in terms of breast cancer screening for women ages 40-69, you could argue that younger women whose mothers and sisters had breast cancer should be screened sooner; and that’s true. But we felt we should start by working with our general screening rate, which started at 21 percent.

What we’ve found is that it’s better to move the bar significantly on only a few measures at first. It’s like anything in life; you can’t do everything at once. You absolutely have to focus. And we’ll continue to focus on areas we haven’t yet done well on; and we’re expanding our focus this year. For example, with the pneumonia vaccination, we’re kind of modeling our work after what one of our other regions did. So we won’t stop what we’re doing on these patient-centered medical homes and six measures. We won’t stop; but some of it will be a little more automatic going forward, and we’ll now go into doing the D5.

So by the end of 2013, in the patient-centered medical home, we won’t be looking at just six measures anymore, we’ll be looking at 11. But you absolutely have to take it a bite at a time or a step at a time. But you have to have the data, it has to be easy to use, it has to be fresh data, which is why Explorys has been so nice, because it’s updated every night.

I always like to say, you measure what you get, and you get what you measure. Before we started measuring diabetes care, I thought I was doing a pretty good job on my patients with diabetes; and I probably was, with my patients who were coming in every three months. But the other thing that Explorys does is that it brings me data on patients I’m not seeing regularly and probably didn’t know about before. And that’s one of the underlying principles of value-based purchasing; that you’re responsible for a whole population, and now I can reach out to patients and focus in on their diabetes, without having to wait for them to come in for care.

Do you have any particular advice for non-clinician CIOs with regard to this work?

I think these initiatives need to be led by clinicians. And the data needs to be up-to-date, clear, easy to use, and I might even go so far as to say that a simple, more basic system that’s easy to use and gets used, is better than a more complex system that doesn’t get used.