Does Motivational Interviewing Improve Osteoporosis Treatment? Does Motivational Interviewing Deserve Better Research?

In some of its Grand Rounds and other conference presentations, the Disease Management Care Blog likes to show a slide portraying a basic disease management business model: a nurse call center with patient calls going in the front end and piles of cash coming out the back end. That's built on a widespread assumption that if you educate persons about their health, insurance savings will accumulate faster than PCPs at a weekend "here's how you can make money with botox" seminar.

The veteran disease and population health management providers know it's not that easy. But that doesn't mean that the outbound phone model isn't still worth trying. That's why the DMCB was interested in this just released 'Online First' article by Daniel Solomon and colleagues on the "Osteoporosis Telephonic Intervention to Improve Medication Adherence" (or "OPTIMA").

The study was conducted at physicians at Boston's Partners Healthcare. The participating patients were enrolled in an unnamed Pennsylvania pharmacy benefits program. At the time of entry into the study, they were independently living, community-dwelling and low income elders. Patients who received a new prescription for any osteoporosis drug were contacted by telephone and asked about participating in the study.

If they didn't "opt out," patients were randomly assigned to one of two protocols. The intervention group (1046 participants) received telephonic coaching sessions that were based in "motivational interviewing" that reviewed attitudes about taking the osteoporosis medicine and addressed barriers to its long term use. They also received health promotion mailings. The control group (1041 participants) got just the mailings.

Both the intervention and control groups were quite elderly (average 80 years), mostly white (approximately 90%), widowed (about 67%) and already taking a lot of pills (10 medications). About 63% were on weekly oral biphosphonates, while the rest received a smattering of other types of medications.

One year later, based on a "medication possession ratio" (MPR) statistic, there was no evidence that the intervention group did a better job of taking their osteoporosis medication.

The MPR is the ratio of days that are covered by a purchased prescription versus the total number of days. The intervention group had a median MPR of 49% (in other words, they were in possession of their purchased medicine a little less than half of the eligible time) vs. 41% in the control group. While the intervention group had a higher MPR, the difference of 8% was not statistically significant. By the way, the telephonic motivational interviewing also had no impact on the self-reported rate of bone fractures (about 11% in both groups) or falls (about 37%).

To the DMCB, this looked like a solid study. The authors recruited an impressive number of patients. Their motivational interviewing and support was subjected to a lot of internal quality control. Yet, despite a seemingly strong approach, it didn't translate into patients being more compliant with their medications.

The DMCB can't say it's surprised. Based on its own experience and what it has learned in other settings, it knows that remote telephony, even if it's provided by dedicated nurses using the most modern behavioral techniques, is not going to have much of an impact if it's not accompanied by other program supports.

DMCB readers know that those supports include predictive modeling (to spot the patients at greatest risk of not taking their prescriptions), a spectrum of interventions based on that risk (including home visits, and high frequency telephonic outreach for patients most unlikely to forgo their medicines), patient incentives (including meaningful limits to any out of pocket costs) and close coordination with the patients' physicians. 

That multi-dimensional and synergistic approach - which is typically used in modern vendor-based telephonic disease management - was not tested here.

Motivational interviewing and osteoporosis treatment deserves better.  As previously noted by Don Berwick, dealing with complex patients involves non-linear social constructs that escape the confines of randomized trials with unidimensional interventions. They are ultimately designed to ask the unsatisfying question "what," instead of the far more important one of "how."

Last but not least, how about that fancy "OPTIMA" acronym?  The DMCB suggests this one: AFFRONT: Another Fabulous Failure of Research (to address) Osteoporosis Needing Treatment

Image from Wikipedia

CODA:  While doing the background research on this blog posting, the DMCB came across the FRAX tool to assess osteoporosis risk.  Looks like a worthy option for any population health management program that needs a guide on osteoporosis risk assessment and treatment.

A Primer on Motivational Interviewing: The Definition and a Review of the Supporting Science for Disease Management

First, I'll express empathy....

The Disease Management Care Blog typically doesn't don the peer-review swimmies for a jump into the pool of nursing journals, but the title "Motivational Interviewing: A Useful Approach to Improving Cardiovascular Health" was too much to resist.  While "motivational interviewing" is one of those widely quoted catch-phrases that sound good in care management marketing collaterals, what's the science behind the concept and does it really work?

Your handy DMCB investigates by summarizing the article for its thousands of readers.

According to the authors, motivational interviewing (MI) can be defined as "a client-centered, directive method for enhancing intrinsic motivation to change by exploring and resolving ambivalence."  Four underling principles of MI are 1) expressing empathy, 2) developing discrepancies between current and desired behaviors, 3) 'rolling' with any client resistance by finding a different way to respond and 4) supporting self-efficacy by building on the client's choosing and carrying out ways of changing. 

It turns out there have been 29 randomized trials that compare MI to usual care, many of which involve substance as well as tobacco abuse, HIV risk behaviors, healthy eating and exercise.  They generally show a positive effect, a dose-response relationship and some evidence that it works even during brief encounters.  That being said, in the area of cardiovascular risk management, the science is still limited by only 6 studies with small sample sizes, varied outcomes, bias, confounding and limited generalizability. 

What's more, we still don't know which elements of MI are essential, which patients are most likely to respond, how it can be made more culturally appropriate, how long it takes to train a health professional (for example, is a one day workshop enough?), how well it addresses underlying patient misconceptions and the best fit in high-volume or fast-paced clinical settings.  It also turns out that there are times when simple advice and nothing more is all that is needed, so MI is not the answer to every patient counseling need.

This was a very helpful review and will be tucked away in the DMCB intelligence files.  It also stands as a good reminder that this corner of disease and population-based care management is still in its infancy and more work needs to be done in better understanding the role of MI.  As population health management continues to evolve, look to the industry to perform additional studies to gain additional insights on how it can be used to drive quality and reduce unnecessary costs. 

Last but not least, the DMCB points out that, based on its reading, MI remains very much a nurse-led intervention: few physicians have the skills or the interest to provide this particular kind of counseling.