Randy L. Thomas

Healthcare IT is at the heart of healthcare reform.

As the debate around healthcare reform continues in Washington, a few consistent themes are emerging around the issue of modifying care delivery and reimbursement models. Value-based purchasing (or pay for performance), bundled payments, and accountable care organizations (ACOs) are different variations on the theme of reimbursing for results rather than transactions. And although there are clearly differences between these approaches to changing how care is provided and paid for, there is a common need to be able to track and measure patient outcomes.

Data Rich-but Information Poor

Achieving quality outcomes-doing the right thing at the right time for the right patient-is becoming the driver behind reimbursement. Consider the following payment models.

Value-based purchasing. This payment model rewards physicians, hospitals, medical groups, and other healthcare providers for meeting certain performance measures for quality and efficiency. Disincentives, such as eliminating payments for negative consequences of care (termed never events by the Centers for Medicare & Medicaid Services [CMS]), can also be included.

Bundled payments. This approach refers to one payment per case for all services associated with an acute care episode across silos of care and time (30, 60, or 90 days after discharge). For instance, a single rate could be paid for both physician and hospital services related to an acute episode, while associated postacute care is paid under the respective prospective payment systems (www.prometheuspayment.org).

Accountable care organizations (ACOs). An ACO is a local network of providers that can manage the full continuum of care for patients assigned to them based on their historical use of the network for the preponderance of their care. In exchange for this accountability, the ACO would receive additional payments based on the associated savings to the payer. An ACO could consist of multispecialty group practices, independent practice associations, physician-hospital organizations, and integrated health systems such as academic medical centers (www.medpac.gov/chapters/Jun09_Ch02.pdf).

Together with CMS's refusal to pay for never events and proposed changes in reimbursement for readmissions that meet certain criteria, a tone is set that focuses on quality and results rather than volumes. This is not new news.

Regardless of the exact forms new payment and delivery mechanisms might take, what is clear is the need for information that tracks patient outcomes across the continuum of care in a timely manner. This need, however, underscores the truth underlying the widely held perception that healthcare organizations are "data rich and information poor."

Will Your Organization Have What It Takes to Succeed?

Setting aside for the moment that far fewer than 50 percent of all hospitals and even fewer ambulatory settings (e.g., physician offices, home health agencies, rehabilitation facilities, long-term care) have adopted electronic health records (EHRs)-a situation that will hopefully be addressed by the incentive reimbursement established in the Health Information Technology for Economic and Clinical Health Act (HITECH) provision of the American Recovery and Re-investment Act-there are several areas that now warrant attention if an organization wishes to readily succeed under these new payment models.

Access to data. All of the potential changes in delivery models and reimbursement methods require that an organization have access to data about patient care delivery. These data are needed to measure care delivery outcomes and provide feedback to clinicians. Of course, if your organization has an EHR completely implemented, you have all the data captured in the process of documenting each patient's care. Even if your organization doesn't have an EHR, some of the data (e.g., patient demographics, lab results, medications, imaging orders) will be available in the various ancillary systems in your organization.

But having the data is just the first step toward being able to utilize these data. The challenge is extracting these data from the transaction systems and moving them into a separate repository designed to support measurement, analytics, and reporting. And that's just for the data contained within your systems. As you look to measure and manage care delivery across the continuum of care, you will need to determine mechanisms to access these data as well.

Transaction systems are designed to create, capture, or display information about one patient at a time. To understand patient outcomes both within the hospital and across the continuum of care, you need to extract the data from these systems into a format amenable to population analysis, so you can look for patterns and trends and track progress toward your quality outcome targets.

Although it may seem like a straightforward task, current adoption of interoperability standards is low. And it is particularly low for semantic interoperability (i.e., adoption of a common vocabulary so that every test, result, procedure, medication, or other element is called the same thing in each data system in every hospital and ambulatory setting), which means, for example, that a "CBC" in one system may be a "blood count" in another. Although interoperability standards adoption will increase over the next five to seven years (spurred by the requirements of the EHR meaningful-use criteria associated with HITECH), hospitals have a more immediate need for these data. Healthcare finance professionals-particularly CFOs-should actively participate in developing the extraction and standardizations mechanisms for their organizations so they can access these data in a usable form.

Ability to link patient information. Your organization probably has a master patient index (MPI) capability to identify the correct patient for any given transaction (e.g., admission, order, or result). This capability ensures that all transactions are correctly associated with the right patient for both clinical and financial reasons-for instance, to make certain the correct lab test shows up for the right patient and that you are billing correctly. The process of linking patient information within your organization, although not necessarily flawless, is aided by a common medical record number.

In the case of bundled payments or ACOs, however, you may need to link together different pieces of information about a patient's care in situations where some of that care occurred outside of your direct facilities. As your organization forms alliances for bundled payments or forms an ACO, in its arrangements with independent physicians, home health providers, and/or other nonacute facilities, it may assume responsibility for tracking patient outcomes across episodes of care. In such an instance, you will need to be able to link patient information for care delivered at these different sites. The data themselves may not need to be in the same repository, but you must be able to track, for each patient, what care was received to be able to measure outcomes across the continuum of care.

There are a number of commercially available tools to link patient information. None of them are foolproof-some manual intervention is typically required to resolve incomplete matches-but such tools are an essential element of measuring outcomes for an episode of care.

Measurement and feedback. Once you have access to these data and a process in place to link patient information from multiple sites of care, you will have the foundation on which you can apply outcomes measurement techniques. Some of the things you will measure will be fairly straightforward, such as repeat tests and duplicate medications, but others will require more sophisticated analytics. For example, if you are trying to measure how long it takes hip replacement patients to return to their regular activities of daily living, you will need to risk-adjust the hip replacement population to facilitate  comparing apples to apples. 

Your particular reimbursement methodology. Each type of reimbursement methodology-value-based purchasing, bundled payments, and ACOs-will likely dictate some specific metrics that indicate good performance, much like the quality indicators currently reported to CMS via Hospital Compare (www.hospitalcompare.hhs.gov). The specific metrics will dictate the data that are needed, and you may need to supplement the data you already have with manually abstracted data, as some data required for measurement may only be available in text notes and therefore require human interpretation (or the adoption of natural language processing technology).

To help keep organizational focus on achieving and sustaining quality outcomes and reimbursement targets, it is advisable to publish your organization's results across the organization while maintaining appropriate privacy safeguards. It can also be quite useful to allow physicians and clinical teams to have access to analytic tools that allow them to evaluate their own performance against peers and discover how they might innovate and improve.

Trading Dollars for Data

If your organization is like most, it has already invested heavily in healthcare IT and is likely to continue that investment. As we move toward reimbursement mechanisms based on outcomes, you can leverage that HIT investment to provide the data, measurement, and feedback mechanisms needed to support performance improvement. You can "trade" your data for dollars!

Randy L. Thomas, FHIMSS, is a vice president, Premier healthcare alliance, Charlotte, N.C. (randy_thomas@premierinc.com).

Publication Date: Sunday, November 01, 2009

Login Required

If you are an existing member, please log in below. Username and password are required.



Forgot User Name?
Forgot Password?

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:


   Become an HFMA member instead