“The early lesson is that data is what makes [accountable care] possible.”
Elliot Fisher, M.D., Dartmouth Institute for Health Policy and Clinical Practice
As we begin to absorb the public results announcements and comments for CMS MSSP entities that have been operating for over a year, and for the hundreds of commercial and other Accountable Care Organization (ACO) arrangements during the first quarter of 2014, a number of data points in the form of survey results, interviews and press releases are emerging that paint a picture of significant data-related barriers to success.
ACO Data – The Crux of the Problem
In a recently released survey by the National Association of ACOs (NAACOS) of MSSPs operating for over a year, 40 percent identified data among the top problems that they faced. Throw in the 11 percent separately citing quality reporting (which of course in large part is data related) and close to half cite significant data-related struggles. Here is the first “how can this be?”
Setting aside for the moment the widely publicized CMS problems around file lag times and data quality, MSSPs specifically mentioned other troubling areas:
- Translating the data into actionable information for care managers and providers
- New skill sets that are needed to analyze the data
- Finding suitable software to manage and report what the data means
- Meeting implementation schedules
- Slow stand-up of the IT system
These survey findings are surprising at some level in light of the throng of seasoned consultants, industry analysts, and various reports from Brookings-Dartmouth ACO Learning Network , Commonwealth/Premiere and others over the past several years communicating ‘upper classman’ accountable care lessons learned by seasoned organizations and emerging best practices. We even had a sobering warning from Commonwealth’s December 2012 ACO readiness assessment results which, in theory, should have been a wake-up call (even though it came during the operational evaluation cycle for the freshman class). Thus, here is my second “How can this be?”
It cannot be overstated that data are the foundation of ACOs and their associated success. However, ACOs cannot enable the required transformation to population management workflow and behavior change without that data effectively and efficiently being turned into information, knowledge and, ultimately, wisdom that can be leveraged.
ACO Data-Related Problem Root Causes
Speaking with a variety of industry experts regarding the “How can this be?” questions above, the answers related to these data problems can primarily be divided into these buckets:
- Overconfidence in the magnitude and orchestration of the effort required for successful accountable care transformation – We previously covered this point of view when the first Pioneer ACO results came out in our Accountable Care 2.0: It’s a Journey Not a Program posting. Even for up-side only MSSP participants, it is fundamentally important that the leadership team understands that this early stage government process is one step toward value-based benefits and care delivery that the countries’ purchasers (Federal and State governments, employers, risk-bearing health plans, and consumers) are driving toward.
- All data is not created and understood equally – To date, very few health systems and physician groups (including those with their own health plan) understand and have experience with the context, processing, and management of post-adjudicated claims data, let alone merging it with clinical and other data required for population management (white paper). Add on top of that, the strategies required to analyze this integrated information and turn it into knowledge and wisdom to drive: 1) prioritized actionable intervention initiatives; and 2) value-based decision support before, at and after the ‘moment’ (not just a physical point, but virtual as well) of care.
- The health care industry to-date has not been good at robust data scrubbing and sharing practices –Extra-organizational transparency and associated closed-loop data quality inspection/correction along its route by associated stakeholders (including the patient) is something all other major industries take for granted. Accountable care requires extensive bidirectional exchange and management of external data and knowledge, yet relatively few ACO stakeholders possess the required expertise and experience in doing it.
- The health care industry’s historical provincial nature – There is a broad lack of standard core competencies that are naturally developed in other, more market-based industries. Those competencies are a result of leveraging lessons learned and the resulting best practices that have been developed. Then, that’s when the real transformational models can be developed. Many in the field hear things like, “We are aware of the other initiatives and their challenges but we are different.” In other industries, you have to make a compelling case about why a particular approach requires an exception process before proceeding, and not just declare it.
It’s no surprise then that in a recent survey of hospital CEOs, they place population health management, information technology and creating an ACO in the top 11 most concerning issues facing them.
ACO Data-Related Recommendations
Solving the associated data-related ACO challenges requires the following tactical and strategic action-item recommendations:
- Review and internalize relevant lessons learned and best practices. Don’t limit your prism to your particular type of ACO(s). For example, all accountable care entities can gain important take-aways from the experience of the at-risk Pioneer ACOs as they take on more risk over time as they journeyed toward value-based care. As one, large Pioneer ACO commented last year:
“As the risks will be higher, the need for strong IT strategies is especially essential to positively impact quality measures. We’re starting to see that data are everything. Data and IT systems are probably going to be the most strategic parts of any ACO. Success is within your ability to manage data, and understand population management shifts and risk stratification.”
Given the explosion in the number and types of variations of ACO models, scanning the horizon needs to be an ongoing process for the ACO team.
- Utilize a step-by-step process to implement effective population management infrastructure. MSSPs, like other forms of value-based and accountable care, require a fundamentally different IT approach to enable population management vs. fee-for-service (FFS) workflow and decision-making. Industry analyst group IDC Health Insights recently released a report highlighting the IT components essential for ACO success (get the report here). The entities that have seen meaningful results from the onset of behavior change recognize that this is not an incremental program, but a transformational journey requiring substantial people, process and technology changes.
Back-office analytics must move from supporting FFS to supporting value-based clinician and physician-patient shared decision-making at the ‘moment’ of care where education has the greatest impact on patient outcome and experience. Finally, integrated provider-based care management provides the population-based ‘engine’ to deliver ACO results.
- Leverage a collaborative payer relationship – Much of the confusion around gathering, integrating and leveraging claims-based data can be minimized by taking advantage of the decades of experience of enlightened and transformed payer-partners who, through shared savings contracts, are aligned with providers and retooled to support their success. If your provider organization is currently involved in the MSSP program, reach out to your commercial payers so see if they offer ACO programs that meet your requirements. Learnings from that experience can be very helpful in your MSSP efforts.
- Develop a future state vision for your health delivery organization and how your data strategy supports it. Major transformation journeys such as the move to value-based health care require that strategies and tactics be evaluated in the context of a shared vision. How do you know you are being successful if you don’t know where you are going? Review and consider our Accountable Delivery System Institute’s future state vision, adapted from the Institute of Medicine’s Chasm report, as a starting place for your organization.
ACO Data-Related Policy Recommendations for CMS
Circling back to the CMS-related data challenges, the feedback from many Pioneer and MSSP participants is that the challenges they initially faced have markedly improved steadily over the last two years. However, they communicated that these early struggles may have limited their ability to maximize success in the program.
Having run a ‘learning lab’ Medicare Advantage collaborative-payer organization in St. Louis, Essence Healthcare for nearly 10 years, our Accountable Delivery System Institute has repeatedly seen superior quality and cost-outcome results from a highly integrated, closed-loop data flow that supports population health and the journey to full provider risk. From the onset, claims data were used as input into the provider’s population management platform, claims are processed every seven days and then loaded for use after that (immediately to two days). This short data cycle provides critical claims-sourced (to supplement clinical-sourced) data to all stakeholders in the ACO and the information required to closely track readmissions and become aware of out-of-network care.
We offer two recommendations related to our experience:
- Transparent claims-sourced data availability benchmarks and tracking. The 2013 AMA National Health Insurance Report Card provides baseline benchmarks for claims processing times – we encourage the AMA moving forward to include the time mark that data is made available for population management in this report. This would not only help with Medicare data-enabled information, but for commercial and other ACO data availability as well.
- Within five years, CMS should develop improved availability through claims-sourced data. Based on our years of experience, we encourage Pioneer and MSSP entities to ask that CMS develop a roadmap to provide claims data closer to a nine-day as opposed to the current monthly feed. Subsequent milestones for CMS (and the industry) should ratchet down the time to process the claim and make available the data to be as close as the required availability of clinical data for transitions of care under the CMS Meaningful Use program.
ACOs and Data: The Road Ahead
The great news is that it’s not too late for ACO entities, MSSP program participants and others to learn from the industry’s missteps and actively retool their HIT-enabled population management strategy. This, by necessity, can drive a more robust and effective supportive data approach. However, this window of opportunity for action is beginning to close given the natural cycle time it takes for the required transformational changes to be reflected within improved Triple Aim-based outcomes.