By Fred Jarrett and Ron Sims
Too often when the conversation turns to health care cost and quality, there’s a lot of nodding and wringing of hands. We know there are huge problems with the health care system, the skyrocketing costs, overuse and unnecessary treatments, and disparities, just to name a few. Yet despite our collective awareness that something must be done, we’ve been stalled in neutral, unable to make systemic changes.
But here in Washington state, we have access to the kind of information that can help us get into gear and move forward. We sit on the board of the Washington Health Alliance, founded 15 years ago by Ron when he was King County Executive, motivated by his passion to help health care purchasers like employers, union trusts, and government understand their health care expenditures, and thus, be able to take action to improve them.
The Alliance pioneered the idea of a voluntary All Payer Claims Database (APCD) here, a statewide database of medical and pharmacy claims, eligibility, and provider files, collected from 24 health plans, self-insured purchasers and union trusts, and the Washington State Health Care Authority. This APCD provides objective insight into the performance of our healthcare system. At the Alliance, we’ve worked steadily with one purpose in mind—to use our APCD data to support evidence-based improvements across the state. And, it’s the collaborative nature of the Alliance: 185 organizations, representing all aspects of our region’s healthcare system, providers, plans, purchasers, government agencies and other allies, enabling this accountability for moving us from aspiration to concrete outcomes.
They say you can’t improve what you don’t measure and along that line, we’re proud to announce that the Alliance’s 14th Community Checkup report is being released today, with results for 1,978 clinics, 327 medical groups, 106 hospitals, and 16 health plans for claims from July 1, 2018 to June 30, 2019. This release is the first time the Alliance is using its new Quality Composite Score to compare medical groups and clinics, using a combination of up to 29 health care measures that includes prevention and screening, chronic disease care, avoiding unnecessary hospital visits, and whether evidence-based practices are used.
Here’s how the chart below works.
The composite score is a weighted average of all of the results from the health care measures that are included in this report. A composite score of 0.0 indicates that the medical group’s performance is the same as the state average. A positive score indicates better performance than the state average with a negative score indicating performance below the state average. The higher the score is (positive or negative), indicates the degree to which the entity’s performance is better or worse than the state average.
The composite percentile ranges between 0 and 100%. It represents the distribution of the composite scores and reports on each medical group or clinic’s performance relative to all others. For example, a medical group at the 80th percentile means we estimate its performance would exceed 80% of all others.
UPDATED CHARTS (See explanation for this correction at the end of this story)
This kind of variation for medical groups, from -.25 to .82, is a perfect illustration of our point: the door you go into really does make a difference in the quality of care you receive. It’s time that we all recognize the undeniable truth that not all medical groups and clinics are equal when it comes to consistently delivering high quality, evidence-based care. We wouldn’t dream of buying a car without doing our research on makes and models, customer satisfaction, and repair rates, yet we are reluctant to do the same kind of research when it comes to our health care? Let information from the Washington Health Alliance help guide your choices.
The Alliance’s Community Checkup provides people with the information they need for some of the most important decisions of their lives—where to get their primary care, have a baby, or go for a knee or hip replacement. And it’s not that hard. With one click, you can see all of the results for health plans, hospitals, medical groups, and clinics across the state on more than 100 health care measures going back six years. But wait, that’s not all:
- people with diabetes, heart disease, or low-back pain can see how providers compare to others for that type of treatment;
- guidance is available if there is a baby in your future, you are looking to change pediatricians, or need surgery; and
- you can see in the 2020 Medical Group Highlight how each medical group did on the different types of care measured (prevention and screening, chronic disease care, coordinated, cost-effective care, and appropriate, cost-effective care).
The Alliance not only reports on quality, but cost and patient experience our measure of the value delivered by our healthcare systems. All three, quality, cost and patient experience, define the concrete outcomes we desire.
You can see the Community Checkup and all of the Alliance’s reports here, get consumer-friendly guidance on the Alliance’s Own Your Health website and if you’re a health care purchaser, provider, payer, or run a health-related business, we invite you to join us.
It’s time for us collectively to take the wheel and get going in the right direction.
UPDATE EXPLANATION:
The Alliance’s new Quality Composite Score enables us to simplify and improve reporting by combining 29 different quality measures into one score. Last Friday, the Alliance team identified a miscalculation in five of the measures included in the new Quality Composite Score (QCS) used for medical group comparisons in the 2020 Community Checkup results released last week, and previously in the report released on Accountable Communities of Health and counties. All individual measures included in the Community Checkup report are accurate; it is only in the QCS calculation that a problem occurred. Here are the five measures that were miscalculated in the QCS:
- potentially avoidable ER visits;
- hospital readmissions within 30 days;
- avoiding antibiotics for adults with acute bronchitis;
- avoiding antibiotics for children with upper respiratory infection; and
- avoiding x-ray, MRI and CT scans for low-back pain.
Each of these measures describe how often an undesirable service occurs. This means a lower rate is better. By mistake, a higher rate resulted in a positive score in our first analysis.
The results have been re-calculated and confirmed as of noon today, and they will be posted on the Community Checkup website tomorrow. We will also update the pdf report tomorrow. A summary of detailed changes is included in the attached document. Pay close attention to the color changes of which there are relatively few. Generally, for those organizations with results near the state average, there was little change, but for those further from the state average, either above or below, there could be significant effects.
We own the mistake we made and we regret any confusion caused by our error. We have initiated additional steps in our internal quality improvement processes to ensure they do not reoccur. We take the business of transparency and accuracy very seriously, and therefore felt it important to clarify what we learned, and to correct it as quickly as possible.
Ron Sims and Fred Jarrett are members of the Washington Health Alliance’s Board of Directors.
Ron Sims chairs the Washington Health Benefit Exchange Board, the board charged with implementing the Affordable Care Act in Washington State. He is also on the Board of Regents of Washington State University, the university’s governing body. Sims was appointed by President Barack Obama and served as Deputy Secretary of the US Department of Housing and Urban Development from 2009 to 2011. Prior to that, Sims served for 12 years as the elected Executive of King County, where he convened a Health Advisory Task Force which led to the creation of the Washington Health Alliance.
Fred Jarrett recently retired following nearly a decade as the Senior Deputy County Executive where he led the executive branch of King County government, the King County Executive Leadership Team, the Executive’s Best Run Government Initiative and improvements to local and contract public service delivery. Fred was awarded the King County Executive’s first leadership award for his work. Previously a state senator, he also served four terms in the Washington State House of Representatives after stints as Mercer Island Mayor, City Council and School Board member, and had a 35-year career at The Boeing Company.
Thank you. This is helpful information. Could you follow up with sample size information and statistical variance information for each provider?
Thank you for your inquiry.
Each measure for each medical group used to calculate a composite score has unique populations. These measure-level populations can be viewed on a measure-by-measure basis on the Community Checkup website (wacommunitycheckup.org) for each entity through the Organization Detail. The Organization Detail appears by hovering over the name of the entity on the Quality Composite Score chart.
The actual statistical variations underlying the Quality Composite Score are considered the Washington Health Alliance’s intellectual property and are not currently available for release.
Where does the Health Alliance describe its “Quality Composite Score” – or is that ALSO private intellectual property? Does the Quality Composite Score take sample size (by diagnosis or diagnosis category) into account? Sample size and statistical variation are the numbers which people use to decide whether they can trust ANY statistical report as a decision-making tool. Why should anyone take these (undocumented) statistics seriously?
This animated chart published by the Health Alliance allows you to break out scores by clinic versus medical center. You can also break out scores by commercial insurance versus Medicaid. Yes, sample size and statistical significance is not available. However, if the Alliance deemed the sample size significant ENOUGH, you can break out scores by clinic.
There is A LOT of variance by clinic, so see what you get when your check that option. https://www.wacommunitycheckup.org/highlights/2020-quality-composite-scores-for-medical-groups-and-clinics/.
More information about the Alliance’s Quality Composite Score is available at https://www.wacommunitycheckup.org/highlights/2020-quality-composite-scores-for-medical-groups-and-clinics/ in the Quality Composite Score Overview document located at the end of the Overview section, including the names of all 29 measures used in the four domains and a brief description of how the results are calculated.
By clicking on the Score icon for a particular measure, you will see the result type, confidence interval, number of patients, and state average.