Payment for medical services is shifting from paying for volume (more visits, tests, visits, days = more money) to paying for value (quality of care). Makes sense. But what does value and quality of care mean? It means that physicians get paid an incentive (more money) for certain results (outcomes, process, actions). An example is readmission rates. If a physician’s patients are readmitted to a hospital after discharge more than most physicians, they don’t get the extra payment. There are roughly 1,000 of such quality measures. These quality measures are very important to us – people at the center of care (patients, caregivers, parents, direct care clinicians and staff) – because measurement strongly influences people and organizations who get paid for medical services. Following the money doesn’t necessarily mean better medical care, better health for us, better relationships among our healthcare teams, or better work life for our health professional partners.
I was nominated to sit on a CMS (Center for Medicare and Medicaid Services)/Battelle Quality Measurement Development Technical Advisory Panel (TEP). The TEP had its first meeting in Baltimore last week. I was one of 19 Panel members (and one of two with expertise in all four of the selection criteria -Consumer Perspective, Clinical Content, Performance Measurement, Coding and Informatics). The TEP seeks to improve the process of developing measures. It isn’t trying to develop measures. The good news is that the TEP gelled as a team and the CMS/Battelle leaders seem open to, if not eager for, actionable advice. I am honored to have been asked to sit at this table.
As a Patient Activist and a change catalyst, I appreciate the formidable forces of inertia and the current business realities of the medical care industrial complex. What can little Danny van Leeuwen hope to accomplish? My goal in accepting this appointment is to find one lever that can move the Value-Based Measurement battleship three degrees toward value to people at the center of care. My superpower is to accept what is and go from there. After listening to my esteemed TEP colleagues, my perception of what is is:
- Measures serve to evaluate the performance of individual practitioners (not measure whether patients attain optimal health or how the team is functioning),
- Inertia is heading to further measure specificity by specialty and diagnosis (not toward the patient with more non-medical than medical determinants of health who is more than a sum of their diagnoses),
- Data for measurement exists primarily in claims, diagnostic systems, and Electronic Medical Records (much less patient-generated data and experience/perceptions of people at the center of care),
- Physicians bristle at the idea of being held accountable for anything they deem out of their control (rather than what can I do to contribute to improving whatever?),
- People at the center of care, insurers, and policymakers all feel ill at ease with uncertainty,
- Few, if any, incentives exist for data vendors to integrate their data (So patients, caregivers, and parents using the most health care dollars provide the bulk of communication at transitions in care, if they can do it at all),
- Testing measures in real-life seems to be an almost insurmountable challenge (so the link between measures and what they seek to measure and the link between measurement and value to patients is tenuous),
- Direct care clinicians are stressed and burning out – the proportion of time they spent documenting rather than caring is growing while they feel pressure to increase productivity (rather than technology helping to reverse those trends),
Jeesh. Houston, we have a problem.
At this early stage, I smell two opportunities to shift this battleship and support people at the center of care:
- Consider developing a process to measure clinical decision-making. Clinical decision-making is making informed choices along the medical journey. Informed decision-making by the patient’s lay and professional health team is a new area for clinical quality measurement, yet the body of evidence about Clinical Decision Support (CDS) is growing. Developing the process to measure clinical decision-making makes sense because it:
- Combines medical and non-medical determinants;
- Could make use of patient-generated data that could be integrated into EHRs and PHRs;
- Bridges home, community, and healthcare facilities;
- Includes medical outcomes, workflow and life flow components, and perceptions of people at the center of care;
- Shifts some burden from direct care clinicians to their patient, caregiver, and parent partners; and
- Empowers patients, caregivers, and parents.
- Build the infrastructure for testing measures in real-life. Testing measures in laboratories under narrow, restricted conditions using only claims, diagnostic, and electronic health record data does not reflect the diversity and variation in life. Until the infrastructure supports a shift toward patient-generated data, includes perceptions and experience of people at the center, and integrates the continuum of care, measurement will always do what measurement has always done – Maintain things as they are. There is no evidence without testing. No systemic change can happen without evidence.
I meant it when I said ‘smelled opportunities’. Clearly, these bones need meat. Am I on to something? What are the informing questions that need to be asked? What am I missing? I’d like to go to the next TEP meeting at the end of January/beginning of February with a more mature proposal. I find that groups find it easier to react than create. Please help me create. Use the comment feature of this blog. Thanks.
See my YouTube broadcast of a version of this post here
See some related posts:
What’s the Problem with the Experience of People at the Center of Care?
Hi Danny, Thanks for all the heavy duty thinking/writing it down so I can take it in and make sense of it–that you do. I just listened to an On Being (Krista Tipett) interview with Atul Gawande https://onbeing.org/programs/atul-gawande-what-matters-in-the-end-oct2017/ and what stood out for me is a study he talked about (done in Boston) which showed the value of making a Palliative care physician a part of the medical team from the moment of diagnosis instead of waiting for close to the end of life. The study was done with lung cancer patients who had little hope of living very long, but it was clear to me that ANYONE diagnosed with any serious illness — including all the chronic and or degenerative conditions that are out here, should have that kind of care–because it seemed to me to be clear that the Palliative care docs are very focused on all the issues that are important for the person at the center of care. I have a friend living with mucosal melanoma–who’s gone through immunotherapy drugs and radiation and lots of health crisis in the 2.5 years since he was diagnosed. He lives in Western NY but has been commuting to Dana Farber for treatments and love his Palliative Care doc who he began seeing months after his treatments began, because he didn’t realize he could have done it sooner. So that’s become my vision for what I want for myself, and I know we’re very far from it, especially out here in the boonies of NY State.
Hi Danny, Sue again, I messed up and didn’t give name etc.
Loving this discussion! Here’s a thought, based on a theme from a meeting I participated in last year with a broad group of stakeholders (patients, patient advocates, measure developers, clinicians, system leaders, researchers): One way to combat inertia is to find a common pain point, and identify a solution that will address that pain point for all involved. In particular, so many clinicians–and the hospitals and health systems they work within–feel a huge measurement burden. That’s an opportunity for change. Going about change in the same old ways, and with the same voices at the table, is a recipe for failure (ie, just landing back where we started). This is one of the reasons I’m so excited about the growing recognition that its important to engage patients–or people at the center of care–in measurement work. Their fresh perspective has the potential to be transformative, but only if we are open to their views and ideas and create a collaborative space that doesn’t assume a zero sum game of more patient-centeredness = more measurement burden for others. In that meeting last year, I was struck by how eager clinicians, in particular, were to partner with the patients in the room to think up innovative ways to make measurement more patient-centered. So here’s to working together towards a common goal: let’s make measurement more meaningful and less burdensome.
OMG, yes yes yes, this is fantastic.
On your last point, please note that at a recent meeting on the Quality Payment Program between patients and clinicians hosted by CMS, an idea was bantered about having CMMI be the platform to test PROMs and other patient-centered measures. You may have seen CMMI’s “new direction” Request for Information”.
In fact, my little break-out group talked about process measures – if you want to do a process measure then at least measure whether 1) providers ask patients what they value, i.e. ask them to name 3 things about themselves that are not medical – such as I like to ski, I like to cook, I like playing the piano – which then gives the provider an idea of the outcomes that matter. 2) whether providers planned care with patients to achieve outcomes that matter to them. And finally, have a core set of outcome measures that the patient gets to pick from a list so clinicians are measured for success on those measures that reflect what patients want.
What you say below is absolutely in line with that thinking. I love it.
It was great to meet you at the PC-CDS LN meeting and I am interpreting your blog entry as:
The TEP wants to improve the process of developing CQMs.
The current state of CQMs is they mostly measure performance (process measures) more than results (outcome measures), they mostly measure what is easy to record from available data rather than patient-generated data for the specific purpose of measuring quality, there is resistance to “additional work” to measure things beyond what is already done in care as it is already a high-strain environment, and direct testing of the impact of CQMs is generally not done. But the systems we have is where much of the attention is, so incremental adjustments is what could be considered for real-world change.
You see 2 opportunities to make progress (in terms of getting closer to what people care about for “clinical quality” in clinical quality measurement) within these parameters:
Consider developing a process to measure clinical decision making
Build the infrastructure for real-life testing of CQMs
I agree that what is measured in CQMs today is closer to convenience for measurement than what is established as most meaningful for people in terms of what they want from clinical care and how this would define “clinical quality”. We have developed appropriateness criteria for CQMs and are critically appraising CQMs and finding many that match your summary of the “current state”
I think your idea of developing a process to measure clinical decision making is a practical target inbetween the “process measures” (measuring what tests or treatments are done with assumptions that these surrogate measures are good at defining quality of care) and the “outcome measures” (measuring the outcomes patients ultimately care about but requiring different data systems and harder to connect to the care of individual patients in real-time marker of quality of care).
There can be extensive methods for measuring the process of clinical decision making and many outcomes one can measure in such a process – these could produce a lot of substance for researchers but would create a lot of resistance if trying to include in real-time workflow of a busy clinical practice.
My quick suggestion if moving down this pathway would be to make a single measure (the proportion of health care decisions made for which the patient’s wishes were documented) and not overly complicate how it is measured. People may disagree about what represents the patient’s wishes (values, preferences, informed preferences, selection after clarity of options provided, etc) and how such a thing should be obtained (formal vs. informal methods, direct dialog vs. completing forms, etc.). But simply expressing such a quality measure prominently could stimulate a lot of change to meet the measure. This concept is so poorly covered across clinical practice that it would have many get incrementally closer to it, even if poorly done by academic standards.
I have previously suggested higher levels of documentation with 5 elements for documenting a decision made in clinical practice (1. The options 2. The relevant observations about the patient data 3. The relevant evidence or guidance from community knowledge 4. The patient values 5. The decision made), keeping the details about how to a minimum, but as a CQM perhaps making it even simpler and just focusing on the documentation of the patient’s wishes would go a long way. This would define quality of care as health care decisions made with acknowledgement of the patient’s desires, and it seems like that is a key “quality problem” many people would associate with our healthcare system today, somewhere between what we call process measures and outcome measures today.
My commercial bias (developing a product that can address #3 above – providing relevant evidence or guidance from community knowledge, either for the healthcare professional as a point-of-care reference or for the patient as a customized patient decision aid) would push me to desire that get considered too (another area often missing in a system that focuses mostly on patient data observations without the rest), but to make the smallest incremental change to get the largest chance of real change toward a better state, perhaps only focusing on #4 will be better.
Brian
Thanks Danny for your insightful and provocative blog post (and the pointer to the LN website!) and thanks Brian for your likewise thoughtful response.
To me, the crux of the issue is that measurement (and related issues of reporting, payment, etc.) are currently focused mainly at the provider/process level and where we need to be moving is to a focus at the patient/outcome level. Using patient-centered/engaged decisions could be a useful next step in getting there, and in any case, I think greater attention (which you are spectacularly well suited to provide, Danny) on measuring/reporting/compensating things that matter to patients is critical. Realize that’s a huge challenge.
If the QM TEP does wind up taking up your focus on patient-centered/engaged decisions, then the the synergies with where the LN is trying to go will be tremendous. We could perhaps cultivate these using the target area we’ll be leveraging for next steps (i.e., opioid epidemic/pain management) as a use case for how enhancing and measuring decisions can be mutually supportive.
Thanks again. Please keep us posted!
Jerry