More Data on the Value Proposition
Financing health care is dominating current news headlines, but there is nothing new in the arguments. The challenge is still how many people to cover, at what cost, and who is going to pay. As part of that debate, all agree with the notion that people should get good value for their money. And, in fact, we are in the midst of a great experiment with payment systems driven by the search for value. A recent study suggests that despite the sound and fury, the changes have produced little real increase in value.
Ryan and associates examined the question of changes in hospital quality associated with “value-based purchasing,” the phrase used to specify a host of changes such as readmission penalties and quality standards for common illness. To do this, they looked at the HospitalCompare database. The most recent release of this database includes information from 4799 hospitals. The authors “cleaned” the database of Veterans Administration, psychiatric, children’s, and specialty hospitals and all of those in the state of Maryland (because of a state-wide exemption,) and those that did not have data in all study periods from 2008 through 2015. This left them with 2842 hospitals in the clinical-process analysis, 3247 in the patient-experience analysis, and 2195, 3256, and 3525 hospitals in the analysis of mortality for the first three clinical conditions: acute myocardial infarction, heart failure, and pneumonia.
As would be expected from such a huge data set, they had to make some simplifying assumptions. For clinical processes, they constructed a composite score consisting of the seven measures used in the first three years of the value-based purchasing initiative and standardized it against the group, so that the expected score was 1. They did a similar process to the eight measures in the patient experience category. For hospital mortality, they evaluated hospital-level 30 day risk-standardized mortality for acute myocardial infarction, heart failure, and pneumonia as separate outcomes.
As has been shown in other studies, the composite scores of the clinical process indicators and the patient experience indicators improved after an intitial delay following the start of the program. However, 30 day mortality improved only for acute myocardial infarction. There was no real change in either heart failure or pneumonia mortality.
The unique aspect of this study is they created a comparator group of “critical access” hospitals, who were exempt from the financial penalties associated with value-based purchasing, and their clinical process and patient-experience composite scores increased in parallel fashion to those at financial risk. Their mortality experience was also not different from the “value-based purchasing” hospitals. This suggests something other than financial penalties may have been in play.
The authors note their results are not as significant as those seen with the readmission penalty improvements, and suggest the financial penalties may have been insufficient. They also note that the publicly reported process measures have attainment rates in the high 90’s, so further improvement was (and is) improbable. So they suggest for the program to have real impact, further program refinements are needed.
As a clinician, these data suggest several other possibilities which don’t seem to be part of the current conversation. First, maybe the process indicators don’t really have much to do with clinical outcomes. I have looked at the “quality-paradox” at some length, and it is quite possible that chasing the measures has caused organizations and individuals to take their eye off the small things that impact outcomes in a way that offsets any mortality gains from application of the big things.
Second, patient experience, as currently done by the government in all areas of health care, is a statistical construct designed to “de-personalize” the data. While there are good reasons to do this, one practical effect is to make it nearly impossible for individual practitioners or organizations to translate lower than desired scores into actions, particularly ones likely to improve clinical outcomes. Most programs focus on “guest excellence,” which is certainly useful, but not likely to impact clinical results in a major way.
Third, payment changes already in place have, at least in my area, resulted in the slow death of physician-delivered primary care, closure of small county hospitals, and transfer of more and more patients for hospitalization at our regional referral center. Many of these patients are not receiving consistent care between admissions, and each admission is done by a different team, resulting in loss of continuity of care. Thus, we are seeing an increase in sicker patients, a trend I suspect is also happening elsewhere.
Finally, the current program is focused on the hospital admission as an isolated event. Yet if we want to improve clinical outcomes, we have to improve chronic care management and see hospitalization as an “episode of care.” This means we need better networks of community-based practitioners, hospitals, and chronic care facilities, and a payment mechanism that makes it possible to provide the right care in the right place at the right time. Presently, we do not do this. There are some preliminary efforts underway to work on this, but they are weak and it is too soon to know if they are going to work. (And the forces for maintaining the status quo are strong.) Of course lack of effectiveness is not likely to stop the intitiatives—after all clawbacks and penalties are a form of “savings” if looked at in the peculiar ways of the Federal Government.
27 June 2017
 Ryan AM, Krinsky S, Maurer KA, Dimick JB. Changes in Hospital Quality Associated with Hospital Value-Based Purchasing. N Engl J Med 2017; 376(24):2358-2366. doi. 10.1056/NEJMsa1613412.
Another Look at the Value Proposition
A review of published data show pay for performance programs have not impacted either cost of care or health outcomes.
Confronting The Quality Paradox - Part 1
CMS is in its third year of applying penalties to hospitals with excess readmissions, but almost 75% of hospitals were penalized this year. Is this a quality improvement or a surtax?
Why the Value Proposition is Not Selling
Why is there resistance to the value proposition in health care?