COMPARATIVE EFFECTIVENESS RESEARCH: DAVID SHERN, PAST PRESIDENT AND CEO OF MENTAL HEALTH AMERICA
By Jessie Gruman
April 2, 2014
This is the fourteenth in a series of interviews between CFAH President and Founder Jessie Gruman and patient and consumer group leaders about their experiences with and attitudes toward comparative effectiveness research (CER).
Gruman: Tell me about yourself and your organization.
David Shern: Mental Health America is the nation’s oldest (104 years) and largest (by total budget) advocacy organization concerned with all aspect of mental health, mental illnesses and addictions. We have 240 affiliates in 41 states. Some provide services to people with mental illnesses but all provide public education and advocacy.
We are distinguished by a public health perspective with overarching goals of building mentally healthy communities, reducing the rate of mental illness and increasing the positive psychological features that help people fully engage in their lives. With this public health focus, we have traditionally worked on prevention as well as treatment and on representing the voice of the patients. This approach is in contrast to other advocacy organizations that are more disease specific or that have traditionally represented the perspectives of family members of persons with severe mental illnesses, like the National Alliance for Mental Illness.
I just recently stepped down after serving as the President and CEO of Mental Health America for six years to spend more time with my family.
Gruman: Where do you see comparative effectiveness research (CER) fitting in the effort to improve the effectiveness of health care?
David Shern: Mental Health America was the founder of the National Working Group on Evidence-Based Health Care. This coalition was founded in 2005 in response to fears among patient groups that the evidence based practice movement could significantly restrict access to needed treatments. The working group represents over 40 voluntary health organizations, and the initial orientation of that group has persevered, that is that the ultimate decisions about clinical care should be made by patients and clinicians who are appropriately informed about options and potential consequences.
We believed then, as now, that it is important to give people information that is responsive to their questions and concerns and that the average effects documented by randomized controlled trials (RCTs) tell us very little about what any individual can expect will happen to them. And so our initial push-back on CER was that using average between group differences in (often single) outcome variables as the basis for payment policy was inappropriate since it did not appreciate heterogeneity of response to treatment.
I don’t believe it will be long before we reflect back on our concentration on RCT data and consider our use of it to be primitive and naive ‘ to make important treatment decisions based on an average between group differences. Now we are on to the next question: How to unpack heterogeneity of response? How to design studies that will result in treatments that are personalized and thus more effective?
Gruman: Tell me how Mental Health America views the relationship between CER and patient-centered outcomes research (PCOR).
David Shern: When we started focusing on CER, the big concern was the head-to-head trials of drugs and devices and the naïve application of their findings to insurance reimbursement policies. Our ultimate fear was that access to medications would be restricted, so we pushed back on this. We were very active in advocating as a part of the National Working Group for specific prohibitions against CER being naively used in cost containment efforts, and we saw the language prohibiting PCORI from focusing on differential cost effectiveness as a victory to help assure that their research wasn’t used in federal insurance coverage decisions.
We are impressed with PCORI trying to understand what it means to be patient-centered. If old CER was head-to-head trials, PCOR includes trials but also tries to incorporate patient decision processes and outcomes and to operationalize them. I’m very encouraged by this effort. However, it’s one thing to say you are committed and quite another to do it. The Devil is always in the details.
Gruman: Can you give me an example of how your constituents have been affected by CER?
David Shern: In the late 90s NIMH launched a number of CER trials, one of which was the CATIE trial that compared a 2nd generation antipsychotic medication with a 1st generation one. It was very controversial and rife with methodological concerns. Major conclusions of the study were that there were few differences between the 2nd and 1st generation drugs in symptom management or time to discontinuation. Since the 2nd generation drugs are much more expensive than the 1st generation drugs, these results were used to argue for using the older agents. However, from our perspective, the major finding from the study was that none of the drugs were particularly well tolerated ‘ there was great heterogeneity in time to discontinuation. It was only if persons were randomly assigned to the drugs that they were taking prior to random assignment that they were likely to remain on the randomly assigned drug.
If there is no difference in outcomes between cheap versus expensive medications, the policy will be to only cover the less expensive drugs.
We pushed back on these potential policy implications of a study and used the study as an example of the importance of heterogeneity of response. This is an example of how, if the results of CER are carelessly applied, it can be harmful to individuals.
We have been encouraged by the degree to which there is openness at PCORI and in the methodology community to trying to model heterogeneity and designing studies to uncover differential effectiveness.
Gruman: How do you see the increased attention to CER affecting the care your constituents will receive in the future?
David Shern: I hope we do it right, that it will result in shared decision making technologies that include what’s known about subgroup responses. This will require large data sets to model heterogeneity and user friendly technology to support clinicians and patients to make better, personalized treatment decisions. All of this is conceptually within the patient centered focus, this increased accessibility and availability of information about differential treatment effects.
Gruman: Some professionals believe that patients are opposed to comparative effectiveness research. Do you think this is accurate?
David Shern: Our issue has always been that CER could result in restricted access to treatments. This concern continues as cost is a front and center policy concern. Patient advocacy groups will continue to be vigilant around the inappropriate application of CER findings.
The ACA has provisions that push quality and efficiency. We see solid, informed decision making as key to those aims.
Right now we don’t do very well in supporting people making good decisions. But I’m optimistic that the implementation of the ACA will help.
Gruman: What are your fears and hopes for CER?
David Shern: The danger is that CER becomes a blunt force policy instrument that plays into the cynicism of modern medicine: cheaper is better and that fewer choices are better than many.
Meaningfully engaging people in the research process is difficult. And it is especially complicated for those with a disability that is life-long and intertwined with their identity, such as some mental illnesses. In the case of mental illness particularly, it is sometimes hard not to confuse the person with the illness.
The question is how do you get patients involved in research in a meaningful way?
I think that the extent to which we can fully engage patients ‘ all different kinds of patients ‘ in the design and conduct of research, will allow us to make great inroads and live up to rhetoric of patient centeredness. This means, though, that we will need to craft participation that makes sense for patients. It’s critical that patients be part of research in order to fully understand their concerns: they are the ones who will use the results of CER as the basis of decisions about their health care.
Original blog post by Jessie Gruman. Updated by the GW Cancer Institute June 2016.