In 2007, the Joint Commission created Ongoing Professional Practice Evaluation (OPPE) as a standard. It seemed fairly straightforward at the time: Using an objective process and data, hospitals need to evaluate the competency of all practitioners to exercise their granted privileges and this must be done every six to nine months. Typically, this role has been delegated to the organized medical staff as part of their peer review and credentialing responsibilities. Hospitals and medical staffs across the country have struggled to create a robust and effective process to fulfill this requirement. Some hospitals have specifically changed their accrediting bodies based on these types of requirements. This article will examine why it has been so difficult and suggest some general strategies to overcome this hurdle.
Why So Tough?
When quality and medical staff service departments as well as physicians are queried as to the causes of their OPPE implementation problem, the answers generally fall into two categories.
First, most point to problems in knowing what to measure. They note there are certain obvious performance measures that are already being collected and reported, such as core measures. How do you know what to measure for each of the specific clinical specialties? There are no standard measures out there for a hospital or medical staff to adopt and creating them from scratch is almost impossible. One person interviewed put it this way: “Early in the experience, just measuring a few performance items seemed to be OK. Now, they expect us to measure each individual’s performance across all six general competencies and for all delineated privileges. It’s an impossible task, especially with our limited resources. So, we piece together bits and pieces and create a hodgepodge OPPE or reappointment performance report”
The second area that the groups identified as problematic is the difficulty in collecting and reporting the data. Most of the present data collection is manual and decentralized meaning that it is collected in many different areas by many different groups. There is no way to collate all these data sources to allow for a reasonable report. There is no or little automation in the process.
Addressing the Performance Indicator Challenge
When questioned further, many hospital support staff said that aforementioned ‘hodgepodge’ report came about because there was no standardized list of data points to be collected. Most places would just gather what they could – a sort of cherry picking process at best.
The starting point would be to begin with the end in mind – what would an ideal OPPE report look like? It should have at least four major components:
- Clinical Activity – the descriptive data on admissions, discharges, procedures, consultations, etc.
- General Clinical Performance Measures – this category includes performance measures that apply to practitioners across the board such as medical record keeping, patient satisfaction scores, incidents of inappropriate behavior, compliance with blood transfusion criteria, etc. They tend to be the rules based measures of performance.
- Specialty Specific Performance Measures – these are the specific measures related to each specialty practice and involve mostly aggregate rate outcomes such as mortality rates, complication rates, appropriateness of medical therapy, etc.
- Case Review Results – for individual instances of potentially egregious care or complex care that still requires the labor intensive case review by a peer or group of peers.
Usually, the indicators in groups one, two and four can be created by a multidisciplinary task force of the medical staff since they generally apply to large groups of practitioners. Group three is more challenging since there are so many different specialty areas.
One approach taken by a small hospital system in Northern California seemed to work well. They had representatives from every specialty attend a dinner where there was a short presentation on developing indicators. After that, the physicians had to develop three to five indicators specific to their particular specialty before they were allowed to leave the room (excepting bathroom breaks, naturally). When they finished, the hospital system had a large collection of indicators covering all specialties that they used as a basis for going forward with their OPPE development.
A quick, general point, no one needs to write indicators completely from scratch. The Agency for Healthcare Research and Quality, the research arm of the Centers for Medicare & Medicaid Services, has a national warehouse of performance indicators that one can access through the Internet at www.qualitymeasures.ahrq.gov/.
The other great challenge is in collecting accurate and supportable data regarding the indicators chosen to measure performance. In addition to the general scattering of the data repositories, the number of different people tasked with this data collection within a given organization can be enormous.
One way to begin to get a handle on this one is to do an inventory of all data sources of performance measure that exist in your hospital. The big ones, of course, may be your basic hospital data system that often has the coded data as well as modules that may contain various portions of the measures, the electronic medical records, the operating room system as well as any systems supporting specific areas like the intensive care unit, the delivery room, pharmacy, etc. Often, there are many homegrown databases on individual computers which are storing some of this data as well.
The trick will be to first decide what will be the centralized recipient of the data which creates a database of performance that becomes the source of reports generated for OPPE and other competency assessments. You will need to work with your information technology (IT) department to effect this process successfully.
Many organizations now are adopting outside software to help this data collection along. No matter what decision your organization may make regarding this IT support, it is clear that the organizations that do this well are the ones who can automate the process to a significant extent. The more the dependence on manual data storage, the less likely it is that you can get to a robust OPPE. So get automated today.