I Learned A Lot From PAM (Patient Activation Measure)

A few months back I posted a piece on how to improve HCP and patient engagement titled “Patient & Physician as HC Partners: The New Black in Healthcare“. The premise was ‘The question becomes how can we foster and accelerate this relationship? How can the HCP create a nutrient enriched environment where the 15-minute office visit presents greater productivity? One tool is to determine the problems the patient is seeking to solve. What are his or her healthcare issues, what do they need/want to understand, what can the HCP help guide the patient through?’

The key point supporting this was the recommendation that a simple inventory be done with each patient to determine their learning inventory. It was overly simple but at the time I thought it had value in a rudimentary way. 

These past couple of days I have been researching patient physician engagement in order to outline a study I think may have some value in mental health. Well guess what I found? Go on guess. There exists a validated tool that is reliable called the Patient Activation Measure (PAM). It was created by Judith H. Hibbard and has been widely used to determine what it means to be activated with your HCP as a patient. It showed that activation of patient engagement involves four stages 

1. Believing the patient role is important

2. Having confidence to take action

3. Actually taking action to maintain and improve health

4. Staying with it under stress

This is a very interesting tool and one that appears simple to administer but provides a great deal of information for both the HCP and the patient. My take away when finding this was “Oh darn I am so behind the times” But I now see this as a tool that can be adapted and worked with to really determine how patients learn and who would engage more readily and who would less and what can be done move those with lower desire or knowledge to activate. 

There is always something new to learn and this must be what patients face as they navigate an illness or caregiving. That is why I still believe and support the need for patient HCP engagement and measure.

Learner First and Foremost, Patient Second

I’m a staunch believer in adult learning and how when the theory is put into healthcare practice it can improve patient care and create durable outcomes for the patient, aid the HCP in improving patient management, and help lower utilization costs.  

This weekend I read with rapped attention Jim Rutenberg’s article in the New York Times Magazine “Data You Can Believe In” and last week I listened to Jonathan Alter’s interview on Fresh Air about the Obama reelection. Both spoke in great detail about how the analysis and use of data was the difference in victory for the Obama win the 2012 election.

What does this have to do with healthcare outcomes? I was struck by how the Obama campaign accessed Facebook data, identified people who supported the President, and were able to have those supporters reach out to friends on the fence or not active become active. Further they were able to identify better tools to find and reach uncommitted voters by comparing TV cable box data with lists of uncommitted voters in order to change their behavior.

In healthcare we have been striving to improve physician patient engagement while recognizing that more and more patients and caregivers are searching the WWW to learn about their health. All the while providers and HCP are moving toward EMR. This is creating one the richest databases in healthcare.

The questions becomes; how can we analyze current patient files within a provider system (I would submit that is being done), and take subsets of that data to identify areas where learning would yield the greatest improvement in patient care, and finally how do we identify (think set top box) who would be the most active learners and least active? How can using data as they did in the Obama campaign improve patient physician engagement?

We can look at data within the provider system to determine which patients are yielding the best outcomes with the lowest utilization cost. And as we move further away from best outcomes to not so good outcomes within the same CD9 code we can identify what the differences are in age, gender, socioeconomic data etc. This will yield a picture on who is doing well and who is not while hinting at why and what are the differences between great and not great. We have a map per disease of behaviors and a model relative to outcomes identifying key demographics.

I don’t believe we will learn what learning behavior or motivation is present from this analysis. What we need to add to patient EMR is data on the patient as an active learner, how, why, where, etc. This is a simple and easy to administer inventory which becomes our set top box of user behavior around learning. It tells us who is learning and where. Are they active learners or not. It’s that extra bit of knowledge that can be used to intervene in disease management and its progress. Matching learning behavior with outcomes with patients would be powerful tool to know where we want to apply pressure to foster and drive healthcare change at the patient level.

Keep in mind I’ve lead this post with the patient first and added patient as a learner. Now let’s reverse that to lead with the learner as a patient. It would be easy to use this data to identify the characteristics of learners who wants to know more about their disease, think active learners.

Now consider a provider or HCP offering this well identified and targeted group improved knowledge access and uptake. Not just better care but becoming learning partner with the patient. If providers can target learners better and either bring new patients into their system or better anchor current patients to improve outcomes and lower cost. Isn’t that what we want from healthcare, a long-term productive and learning driven relationship? Healthcare is not passive; it is an active learning relationship between peers with different skills sharing decision-making. 

The proviso for all of the above, any data analysis would work within HIPAA guidelines etc.