Value-based payment (VBP) models have the dual aim of reducing costs and improving quality of care. A typical VBP approach identifies potential savings by establishing baseline health care costs for specific conditions or populations. The model assumes that the potential for sharing savings generated against the baseline costs will incentivize providers within the network to use more effective treatments and improve outcomes.
A key challenge is to document that providers are in fact offering better quality of care and are being paid for value, as opposed to generating shared savings by limiting access to services. To do this, VBP arrangements incorporate performance measures that document health care processes and outcomes, with the proviso that providers can share savings only if specific performance measure thresholds are met.
One area that will be key to successful health care transformation and payment reform efforts is behavioral health conditions, which are among the costliest in the United States, accounting for $201 billion in health spending in 2013. Yet, the current state of behavioral health performance measurement presents major obstacles to incentivizing change and ensuring quality. There are few endorsed measures of key behavioral health processes and outcomes such as: control of core symptoms of psychiatric illnesses; engagement and retention in care; establishment of family/community supports; access to recovery-oriented services; and personal and public safety. The majority of currently endorsed behavioral health measures address quality of care for medical co-morbidities—such as diabetes and cardiovascular disease or medical readmissions—seen in a minority of individuals with behavioral health conditions. And few capture processes or outcomes directly related to care for the behavioral health condition itself.
What does this mean for psychiatrists, behavioral health provider agencies, and the value-based provider networks they are considering joining? Three serious challenges arise for VBP networks that go without robust behavioral health process and outcome measures: 1) There is little incentive for the network to support care specifically for behavioral health conditions; 2) It becomes difficult to justify what, if any, shared savings should be available to participating behavioral health providers; and 3) There are no quality signals to alert the network when individuals may in fact be receiving substandard behavioral health care.
We believe there is a way forward for behavioral health performance measurement strategies that psychiatrists, agency directors, payers, and providers should consider as they enter into VBP arrangements.
Measure Treatment Response
The paucity of measures reinforces the misconception that treatments for behavioral health conditions lack precision. Consider, for example, management of depression in primary care. Depression regularly co-occurs with a host of other chronic medical conditions, adding significant fiscal and clinical costs, yet routine screening and measurement is rarely implemented. Evidence-based standards for management of depression in primary care settings exist and should be used in VBP arrangements targeting primary care.
For example, the PHQ-9 is a simple questionnaire that takes from two to five minutes to complete and can provide a quick and accurate summary of depression severity. It also serves to monitor symptoms and treatment response. A number of additional endorsed and newly proposed performance measures document initiation and maintenance of antidepressant medication therapy, depression remission, and identification and treatment of substance use disorders. These measures should also be included in VBP arrangements targeting primary care.
Measure Acute Behavioral Health Service Use
A primary value opportunity in behavioral health care involves hospital inpatient and emergency department (ED) care for behavioral health conditions. The Agency for Healthcare Research and Quality (AHRQ) has developed measures of potentially avoidable hospital admissions that require a primary diagnosis of a condition that could otherwise have been managed in the ambulatory care setting. The list of these "ambulatory care sensitive conditions" does not include behavioral health conditions, so the measures will be of limited use in incentivizing value for behavioral health populations.
The 3M measure of potentially preventable readmissions is more relevant as it includes behavioral health readmissions within 30 days as well as readmissions in which either the first or subsequent admission is to a behavioral health setting. This measure allows VBP arrangements to break out potentially preventable readmissions by behavioral health conditions and settings to adequately capture the impact of inpatient behavioral health interventions.
Endorsed measures for seven- and 30-day follow-up after hospitalization for mental illness also present significant value opportunities. Seven-day follow-up rates in both public and commercially insured populations typically hover between 30-50 percent. Value-based arrangements should require that these measures are used to incentivize not only hospital providers but the community-based behavioral health providers to ensure successful care transitions.
For Critical but Rare Outcomes, Measure Key Treatment Processes
In certain circumstances, process measures may be more relevant and useful than outcome measures. One example involves suicide: over 40,000 individuals commit suicide in the United States each year, but the frequency of suicides in smaller populations of patients attributed to VBP arrangements will likely be too low to allow for opportunities to demonstrate improvement.
Another example involves the increasing focus on the relationship between individuals with behavioral health conditions and public safety. There is an ongoing public debate about whether and to what degree the presence of a serious mental illness increases the risk of an individual committing violent acts, but it is neither feasible nor reasonable to create measures of rates of violence among this population. Like suicide, violent episodes involving individuals with behavioral health conditions are too rare to allow for identification of meaningful performance improvement opportunities. Calculating and publishing violence rates in behavioral health populations will also reinforce stigmas and false stereotypes that serve to further limit access to needed services for highly vulnerable individuals.
Nevertheless, research has shown that individuals with behavioral health conditions are less likely to be involved in violent episodes when they are engaged in care. Additionally, adherence to mood stabilizing and antipsychotic medications has been shown to reduce rates of violence. For populations of individuals with serious mental illness and/or substance use disorders, process measures of engagement and retention in care should therefore be employed until more relevant outcome measures are available. Examples include measures of timeliness and frequency of visits with community-based behavioral health providers following discharge from an inpatient program or following identification of a new substance use disorder.
Develop Social and Functional Outcomes Measures
Many behavioral health conditions contribute directly to functional deficits in employment and education status, social connectedness, quality of life, and independent living skills. Yet there are no currently endorsed outcome measures for these domains, which are the targets for evidence-based behavioral health interventions such as motivational interviewing and supported employment. Health care leaders increasingly acknowledge the impact of social determinants on cost and outcomes, and behavioral health providers participating in VBP arrangements should emphasize that many social determinants are in fact targets for behavioral health interventions that should be incentivized.
Claims data will never yield functional outcome measures, so experts strongly endorse standardizing data elements in provider electronic health records (EHRs) to document social determinants and functional outcomes. Providers and stakeholders participating in VBP arrangements should advocate that VBP network providers incorporate standard elements in their EHRs defining key social determinants and functional capabilities to create a source of reliable information and also to generate the data necessary to validate measures of social and functional outcomes. Until such measures are validated and endorsed, VBP arrangements should include pay-for-reporting rules to incentivize providers to incorporate these data elements into their EHRs.
Currently, mature payment schemes such as VBP are being implemented in an immature behavioral health performance measurement environment, creating unique challenges for providers and stakeholders participating in VBP arrangements. But there is a clear path forward to overcome those challenges. The performance measurement activities we've described here will allow behavioral health providers to become valued and active partners any VBP arrangement and ultimately, to improve outcomes for individuals with behavioral health conditions.
from Health Affairs BlogHealth Affairs Blog http://ift.tt/2gCjY4m
No comments:
Post a Comment