The trusted source for
healthcare information and
Like it or not, the Joint Commission on Accreditation of Healthcare Organizations’ (JCAHO) plan for using patient safety management profiles to rate hospitals is moving forward. Some health care providers say the plan will create unfair comparisons and increase liability exposure.
Healthcare Risk Management has learned that you will be subject to some sort of safety rating before long. In recent months, health care providers swamped the Joint Commission’s board of directors with objections and concerns, leading the committee to slow down implementation of the plan, says Ken Shull, FACHE, president of the South Carolina Hospital Association in West Columbia. Shull also chairs the Joint Commission’s Accreditation Process Improvement Implementation Task Force. The board sent the plan to his committee for further review.
The board had been considering whether to proceed with a plan that would score hospitals and other providers according to how well they complied with certain standards and best practices considered key to providing a safe environment for patients.
The board decided that the health care community had some legitimate concerns and decided to send the plan back for another look. The committee has met once to consider the plan and likely is to meet once more on the same topic.
"They wanted field input to make the plan worthwhile, meaningful, doable, and not have it lead into more liability than necessary," Shull says. "It’s considered a very important issue and it’s on a fast time frame."
Shull says that his committee hopes to present its recommendations to the Joint Commission board in November and that the plan might be implemented soon after. Even though the proposal has generated so much criticism, Shull says the Joint Commission is determined to enact some version of it without delay. He acknowledges that much of the concern is legitimate and says he hopes his committee’s work with overcome some of the problems.
"Disclosure of anything is a touchy issue for health care," he says. "Concern about liability is a top priority. I think the general feeling is that the we need some release of data, but we have to be careful about it and make sure it’s accurate, fair, and presented in a way that people can understand, a way that is relevant to how people obtain health care."
Some of the complaints came from the Illinois Hospital and HealthSystems Association (IHHA), which recently asked the board of the Joint Commission to hold off on the plan for patient safety management profiles until it could be significantly revised. The plan to use Sentinel Event Alerts as a compliance measure was a particular concern to the association.
In a strongly worded letter based on comments from IHHA’s Quality Measurement Advisory Group and accredited member hospitals, IHHA vice president Pat Merryweather said that both the proposed profiles and the Sentinel Event Alert assessment structure were "flawed technically, scientifically, and conceptually."
Though the patient safety management profile is supposed to reflect the hospital’s progress in implementing patient safety processes, IHHA contends that the standards have not been empirically demonstrated to bear any relationship to safety outcomes. Nor does the profile directly measure the safety of the hospital or the care it delivers, according to IHHA. The American Hospital Association also has asked the Joint Commission to withdraw the profiles.
"These standards have not been empirically demonstrated to bear any relationship to safety outcome as the new index of standards has not been empirically demonstrated to bear any relationship to safety outcomes," Merryweather stated in the letter to the Joint Commission. "Providers do not have feedback on their individual scores when surveyed. Hospitals will not be able to determine if their scores are accurate or not. There are no checks and balances built into this approach. Since surveyor scores are subjective and the variance among two provider scores can be due to the surveyor subjectivity, JCAHO needs to have reliable measurements for the basis of any composite profile.
"Since variations in score can be due to surveyor subjectivity, JCAHO needs to apply statistical measures to reach acceptable confidence levels and control for these variations. A single score based upon an average does not control for variance and allows for no confidence in the measurement. JCAHO needs to make clear as to how scores will be updated based upon corrective actions taken by providers that have received Type I recommendations. A change in a score should be reflected in the profile so that it accurately reflects hospital performance. JCAHO needs to define the overall average score that a provider is being compared to including the comparative organizations and for what time period." (See "Is patient safety profile plan dangerously flawed?," for more of the IHHA concerns.)
Merryweather says in order to allow for continuous innovations in care and treatment, providers will always develop new and improved ways to address diseases and patient safety care issues. She says the Joint Commission needs to fully train surveyors about alternative approaches and needs to be prepared to update the recommendations in prior alerts based upon information gathered from the field. The Joint Commission might need to build in expiration dates because new literature is emerging constantly, and it might need to be prepared to change the recommendations of any historical sentinel events based upon information gathered from the providers at survey time.
"Finally, this approach is completely unnecessary because the new Patient Safety standards already require that providers develop mechanisms to incorporate new recommendations from Sentinel Event Alerts and other reliable sources," Merryweather wrote. "Thus, compliance with the alerts, in aggregate, is already in included in the current JCAHO standards. Even with this approach, assessment and surveyor training needs to be examined."
Shull says Merryweather’s comments were typical of those the board received and forwarded to his committee. He says the committee is considering two main issues: What information should be included in the profile, and how to publicly report those data. As proposed, the patient safety management profile would be part of the Joint Com-mission survey process, with each organization getting a report card on how it manages hospital safety. Each hospital would receive a score with quantitative numbers that could theoretically be used to compare providers. That is one of the biggest concerns.
"The information would be live and in color on the web, accessible to anyone. There’s no easy way to display data like that and make sure people understand them in the way you intended," Shull says. "I don’t care how many disclaimers you have on something; if there is a graph, chart, or picture, people are going to look at that and forget all the words. They will make assumptions that you may not have intended, and that may not be an accurate assumption."
The committee is reassessing one of the provisions that troubled Merryweather and the others who expressed concern: The plan to use the Joint Commission’s Sentinel Event Alerts as a way to calculate the hospital’s patient safety management score. The Sentinel Event Alerts are periodically published by the Joint Commission as a way of highlighting sentinel events and bringing attention to the type of dangers involved, plus the lessons learned by the health care providers. Shull explains that the original idea was for the alerts to be used as criteria for determining how well a hospital has addressed patient safety, in effect considering each one a lesson and then seeing how accredited hospitals have put those lessons to use. But there has been criticism that the plan to use the Sentinel Event Alerts is too complex.
Karen Reeves, vice president of professional services with the South Carolina Hospital Association in Columbia, has been monitoring the situation since the Joint Commission first proposed the plan. Though Shull is her boss, Reeves is not shy about voicing her opinion of the project he is trying to improve.
"Thank God that didn’t fly," she says. "The Joint Commission wants to develop a methodology for a grid that would show a numerical score like 90%, with that number used as an indicator of patient safety. But it’s a black box methodology. It’s not been disclosed how you would calculate that numerical score, so there’s no reason to think it’s valid or reliable."
The plan to use the Sentinel Event Alerts as a measure of patient safety causes particular concern for Reeves. She is concerned that the Joint Commission would cavalierly throw too many of the Sentinel Event Alerts at hospitals and not realize how much work is required to comply with them.
"They wanted to tell hospitals in October of each year that you have to show compliance with these 10 alerts for next year. But if you tell me in October that you’re focusing on these 10 things, there is no way that in January I can have a good process in place for doing that," she says. "It would be much better for them to say here are 20 alerts, and you need to pick a couple that involved concerns at your hospital, implement them, and then explain to the surveyor why you chose those."
Shull has heard similar concerns from many other health care providers and observers. He says that, despite some serious misgivings by different parties, the patient safety management profile will be a reality within a matter of months. The actual implementation date may come after the grace period built into most new Joint Commission standards and procedures.
"We’re going to have to click this together pretty quickly," Shull says. "It’s on a fast track because it’s important. The public, employers, and insurers are all looking for information like this. It’s all part of the increased awareness and emphasis on medical errors and patient safety."