OIG Report: Hospitals Value PSO Participation, But Data Aggregation Lags

The SR PSO has provided our members with a safe place to collaborate and discuss quality and safety improvement issues and events. Groups that actively participate and attend meetings report a deeper commitment to identifying opportunities to improve quality of care and services provided to their patients.

Lisa Mead, RN, MS, CPHQ, CHPC
Executive Director, SR–PSO
November 20, 2019

A report on the value, challenges, and prevalence of PSO participation among acute care hospitals issued by the Office of Inspector General in September reveals that virtually all hospitals that reported participating in a PSO value the relationship, with 63% saying that PSO participation has made a measurable difference in the quality of their care. 

The agency also reported a challenge that will resonate with other health-care entities that struggle to identify meaningful data fields in quality assurance and map them across disparate organizations. Among the 74 PSOs that responded to the survey, fewer than half (31 of 74) accept data in a form that would enable them to contribute to the Network of Patient Safety Databases (NPSD) established by the Agency for Healthcare Research and Quality PSO program. According to the OIG, AHRQ's standard method for reporting patient safety data is configured for just three care settings and not applicable to the full range of patient safety events that occur in those settings.

The OIG undertook the analysis on the 20th anniversary of the landmark report, To Err Is Human: Building a Safer Health System, which found that between 44,000 and 98,000 preventable deaths due to medical error occurred annually. Since then, other research has indicated that the toll of medical error may be twice the high estimate. The report called for a nonpunitive reporting and learning system for the U.S. health care providers.

A Short History

The PSO origin story provides helpful context for understanding the recent report, but also illustrates how slowly the wheels of government turn when it comes to operationalizing legislation. Furthermore, the experience to date bolsters the case for an active patient safety program for U.S. health care and reveals the challenges of anonymizing and standardizing personally identifiable data for the purpose of research.

More than 5 years after the seminal To Err Is Human report was published, the Patient Safety and Quality Improvement Act of 2005 created a voluntary Patient Safety Organization program, establishing PSOs as entities to collect, aggregate, and analyze patient safety information contributed by providers. To address provider concerns that such information could be used against them, the Act also created the first and only comprehensive and nationwide confidentiality and privilege protections for certain patient safety information submitted to PSOs, known as patient safety work product (PSWP). At the same time, the Act required a national NPSD be established for the purpose of aggregating and analyzing nonidentifiable patient safety data and making it available for research.

In 2010, the OIG issued its own report that found 27% of hospitalized Medicare beneficiaries experienced harm because of medical care, recommending that the AHRQ be charged with encouraging hospitals to participate in the Patient Safety Organization program. The following year, HHS delegated most authority under the 2005 Patient Safety Act to the AHRQ. Responsibility for interpreting and enforcing the legal protections of the Patient Safety Act were delegated to the Office for Civil Rights, and HHS published a final rule implementing the Act in 2008. 

Methods and Findings

The OIG sent surveys to all 82 PSOs listed in 2018 and a nationally representative sample of 600 acute care hospitals, receiving a 90% response rate from the PSOs and a 70% response rate from the hospitals. The OIG also conducted stakeholder surveys of PSOs and acute-care hospitals and solicited data from the AHRQ on its oversight of the PSO program through 2017.                         

Since 2008, AHRQ has approved 164 listings of PSOs, an average of 16 per year. During that time, the agency conducted 13 compliance reviews, three of which resulted in corrective action and two resulted in expedited revocation of the listing. A total of 79 PSOs have been delisted since 2008, 69 of them voluntarily, five were due to expired listing, and five to corrective action.

OIG found that:

  • 97% of hospitals that work with PSOs find it valuable to work with a PSO
  • 95% of PSOs offer aggregate analysis of PSWP across providers
  • 92% of PSOs help cultivate a culture of safety
  • 89% of PSOs offer at least one learning-based service, e.g. web seminars, in-person training, technical assistance
  • 73% of PSOs offer either safety huddles or safe tables
  • 63% of hospitals that work with PSOs report measurable improvement in care quality
  • 59% of general acute care hospitals work with a PSO

The Strategic Radiology-Patient Safety Organization provides all of the above services for its members, facilitating quality discussions, collecting and aggregating information on numerous topics, and supporting the development and enhancement of member practices’ quality programs.

“The SR PSO has provided our members with a safe place to collaborate and discuss quality and safety improvement issues and events,” said Lisa Mead, RN, MS, CPHQ, CHPC, executive director, SR–PSO. “Groups that actively participate and attend meetings report a deeper commitment to identifying opportunities to improve quality of care and services provided to their patients." 

Hospitals overwhelmingly (97%) found value in PSO participation, and the OIG report established that 59% of all acute care hospitals that bill Medicare are current participants.  More than half of those hospitals rated participation as very valuable. Participation helped prevent patient safety events for 80% of hospitals and 72% said that feedback and analysis of patient safety events enabled them to understand the cause of events. Nearly all (95%) aggregate analysis of PSWP across providers. 

Challenges in Data Standardization

Data anonymization and standardization have challenged the AHRQ in its efforts to aggregate data and provide meaningful feedback to PSOs. In order to activate the national database known as the NPSD and to begin receiving data from providers, AHRQ created the PSO Privacy Protection Center (PSOPPC) to render the data nonidentifiable.  According to the report, the process of de-identifying data had the effect of limiting the usefulness of the data and delaying the release of the data to the NPSD. AHRQ finally launched a public-facing NPSD website on June 21, 2019, configured to accept data from acute-care hospitals, community pharmacies, and skilled nursing facilities.  Both the NPSD and the PSOPPC are operated by a contractor for the AHRQ.

The Common Formats allow for event reporting in just three settings—hospital, skilled nursing facilities, and community pharmacies. For instance, the Strategic Radiology–Patient Safety Organization (SR–PSO) includes data from freestanding outpatient imaging settings, necessitating a different, standardized format.

“PSOs are required to collect and analyze data in a standardized format,” explained Mead. “When data is collected, the fields and metric definitions are discussed so that we clarify what we are collecting and how we will aggregate and provide benchmarks. It is also beneficial to have open comment and text fields when analyzing events, so these fields are included to promote learning.”

Nonetheless, because of the wide range of patient safety events that occur in even just the three targeted settings, the Common Formats cannot accept useful data for all patient safety events. One PSO reported that up to 80% of the PSWP data collected could not be accommodated by the Common Formats.  Examples of limitations of the Common Formats are the inability to collect useful data on anesthesia-related events and contextual information that would be useful for specialty hospitals, such as a fall occurring during a routine physical therapy session.

The OIG report also identified another factor that may inhibit PSO providers from contributing data to the NPSD. Court actions have created some uncertainty as to what is protected under the Act, with both the Florida and the Kentucky Supreme Courts coming to the conclusion that data collected to comply with state regulations is not protected under the PSO provisions, HHS subsequently issued clarification that “information prepared for reasons other than reporting to a PSO is not PSWP.

The OIG has seen progress in the few short years that the NPSD has been active. “In the past few years, AHRQ has made progress in getting PSOs to submit data to the NPSD, with the number of records growing from 740,000 in 2017 to 1.8 million in 2019,” the authors wrote. “According to AHRQ, 18 PSOs have submitted data to the NSPD, with three PSOs submitting the bulk (87%) of the records. Because PSOs vary in the numbers of and types of providers they serve, some are likely to submit more data than others. In any case, the number of records the NPSD has received from a limited number of PSOs shows the potential for data aggregation if more PSOs submitted data.”

Praise and Recommendations

The OIG ended the report with praise for a landmark program that is the first to provide the all-important protections to providers intent on improving patient safety and care quality.

“The PSO program has the potential to improve health care,” the authors wrote.  “Indeed, this review shows that the program has made progress in its first decade. AHRQ has invested in developing and revising the Common Formats, and in creating the NPSD. Over half of hospitals work with a PSO; those hospitals find their participation valuable, with many reporting measurable improvement in patient safety. The number of records in the NPSD is growing, and AHRQ has launched a public-facing website for sharing NPSD data.”

The OIG report made three recommendations, suggesting that AHRQ develop and execute a communications strategy to increase nonparticipating hospitals' awareness of the PSO program and the program's value to participants; update guidance for PSOs on processes for listing PSOs; and take steps to encourage participation in the NPSD, including accepting data into the NPSD in other formats in addition to the Common Formats. 

Mead agreed that a campaign to increase PSO awareness would be beneficial. “There are many benefits for all providers not just hospitals to participate in a PSO, and a strong communication strategy from AHRQ to increase awareness of the PSO program and its value would provide more focus on the programs and support learning and promotion of high reliability organizations,” she said.

As for the SR-PSO, Mead is deep into some new, front-burner projects within the SR-PSO. They include promoting peer learning through the creation of a Peer Collaborative that will develop a process to evaluate care and services in a non-punitive manner with a focus on learning. Also, eight SR practices are collaborating on a project to improve the communication of incidental findings in their hospital emergency departments by communicating directly with patients through a grant from Coverys Community Healthcare Foundation.

Click to access the complete report.


Subscribe to Hub
News from around the coalition and beyond

Hub is the monthly newsletter published for the membership of Strategic Radiology practices. It includes coalition and practice news as well as news and commentary of interest to radiology professionals.

If you want to know more about Strategic Radiology, you are invited to subscribe to our monthly newsletter. Your email will not be shared.