Patient Safety: Taking a “Pulse”

The New England Journal of Medicine published a piece on February 17th entitled, “Health Care Safety during the Pandemic and Beyond — Building a System That Ensures Resilience.” Despite some improvement in the last two decades, it’s clear that patient safety is under pressure.

The article was co-authored by Dr. Lee Fleisher, Chief Medical Officer and Director at CMS; Dr. Michelle Schreiber, Deputy Director for Quality and Value at CMS; Dr. Denise Cardo, Director of the Division of Healthcare Quality Promotion for the CDC’s National Center for Emerging and Zoonotic Infectious Diseases; and Dr. Arjun Srinivasan, Associate Director for Healthcare-associated Infection Prevention Programs in the Division of Healthcare Quality Promotion at CDC’s National Center for Emerging and Zoonotic Infectious Diseases.

An interview with Dr. Fleisher is also available accompanying the NEJM article.

The joint CMS-CDC analysis points to the decline in patient safety, suggesting the potential impact of COVID-19, and calls for a new approach to patient safety, as well as an imperative for improvement.

Where We Are

The co-authors begin by citing evidence showing that patient safety has improved over the last two decades, specifically healthcare-associated infections and “other complications of care.” However, they point to indicators showing that the trend may have almost reversed.

In general, the healthcare industry has struggled to track the state of patient safety and quality in a timely manner, as national data often lags what is being measured by several quarters to a year.

For example, central line-associated bloodstream infections (CLABSIs) had “decreased by 31% in the five years preceding the pandemic; this promising trend was almost totally reversed by a 28% increase in the second quarter of 2020.” The CDC published a report in September 2021 that found healthcare-associated infections increased significantly in 2020 after years of steady decline.

Increases were also noted with respect to catheter-associated urinary tract infections, ventilator-associated events, and staph infections. Post-acute care did not fare better, registering increases in rates of falls by 41% and pressure ulcers by 41.8%.

The CMS and CDC leaders frame their argument as follows: “The fact that the pandemic degraded patient safety so quickly and severely suggests that our health care system lacks a sufficiently resilient safety culture and infrastructure. We believe the pandemic and the breakdown it has caused present an opportunity and an obligation to reevaluate health care safety with an eye toward building a more resilient health care delivery system, capable not only of achieving safer routine care but also of maintaining high safety levels in times of crisis.”

How We Got Here, Tactically

The co-authors focus on a few explanations:

  • Patient Influx. The pandemic delivered a vast number of very ill patients to the health system, resulting in under-staffing and supply shortages. Notwithstanding heroic efforts, many clinicians burned out, sometimes exiting the field altogether and exacerbating the problem.
  • Safety Practices. The strains of patient volumes and under-resourcing led to traditional patient safety practices not being followed. Clinicians barely surviving had to prioritize critical activities over safety rounds, safety audits, and error reporting. Supply chain disruptions resulted in inadequate personal protective equipment (PPE) and other vital supplies. Checklists proved inadequate. And the safety practices that were followed were likely unequitably distributed.

What Happened, Systemically?

The NEJM piece steps back and diagnoses the patient safety problem. The deeper issue, it explains, is that over time the U.S. healthcare sector has addressed “various pieces of the safety-assessment-and-improvement puzzle,” but it has not implemented a comprehensive system of safety that “reaches from the boardroom to the front lines….”

The co-authors argue that the U.S. deserves:

“breakthrough thinking about systems built on foundational principles of safety, akin to those used in other industries in which safety is embedded in every step of a process, with clear metrics that are aggregated, assessed, and acted on.”


“We also need renewed national goals of harm elimination throughout the health care system and a core safety strategy that includes promoting radical transparency, addressing workforce shortages, and continuing to strive for safety while being sensitive to such trade-offs as reporting burden and costs. This effort should extend across the continuum of care, beyond the traditional hospital-based safety indicators, and include attention to diagnostic errors and outpatient care.”

New Policy From Here

The regulatory leaders suggest where they would like to lead policy by saying, “As we emerge from this public health emergency, we at CMS and the Centers for Disease Control and Prevention (CDC) are committed to a renewed focus on patient safety.” They seek to:

  • Review patient safety practices and seek better and more deeply embedded solutions that also address health disparities, “since there is no true health care quality and safety without equity”
  • Expand the collection and use of data on safety indicators in our programs, which is already in progress
  • Also continue to work to develop safety metrics that draw on the rich clinical data captured digitally in electronic medical records, which incorporate information from all health care payers.

On this last point, they state that some electronic clinical quality measures are already being considered for inclusion in patient-safety monitoring in the CMS Quality Payment Program.

Pascal Comment

Pascal lauds CMS and CDC leadership for:

  1. Identifying that patient safety is a problem in healthcare, has worsened due to COVID-19, and is unlikely to improve without a more material change in approach.
  2. Affirming the necessity to embrace “breakthrough thinking” and “developing new approaches” “beyond the traditional hospital based safety indicators” that will “permit the delivery of safe and equitable care across the continuum of care during both normal and extraordinary times.”
  3. Committing to “developing safety metrics that draw on the rich clinical data captured digitally in electronic medical records, which incorporate information from all health care payers”.

This is the right approach. That said, Pascal respectfully submits to CMS and CDC are three facts to bear in mind as they consider moving to a new policy:

  1. Shift from tactical harm targeting to systemic all-cause harm measurement and management. Both peer-reviewed and real-world evidence shows that CLABSIs and hospital-acquired conditions as measured by the regulators (and what many compliance-consumed patient safety programs are naturally focused upon) represent only a fraction of adverse events. It’s time to move beyond what Dr. Don Berwick elsewhere has termed the preoccupation in traditional patient safety programs with the “particles of harm” instead of measuring and reducing all-cause harm. Resilience is strongest when systemic; identifying the “weakest links” requires comprehensive measurement and subsequent management.
  2. Recalibrate the role of safety culture from being the solution for reducing adverse events to a foundation for operational methods and technologies supporting#1 above. As the old saying goes, “When all you have is a hammer, everything looks like a nail.” Historically, the only tool that most healthcare organizations have had – absent a cost-effective way to generate and use AE Outcomes – is assessing culture, generating safety culture data, and providing a system to support data-driven cultural interventions at the unit level. Pascal was the first to pioneer this approach and to do so at scale in the late 2000s at Johns Hopkins, Ascension Health, Adventist Health System, Kaiser Permanente, Mayo Clinic, Partners Healthcare (now MassGen Brigham), LifePoint, the NHS, and many other healthcare organizations. Establishing a culture of safety is foundational and essential for achieving a high reliability organization (HRO), but the idea that improving culture alone without proactive adverse event surveillance and subsequent intervention and improvement will one day – if it doesn’t already – be concluded to be, at best, optimistic and destined to fail.
  3. The statutory and regulatory framework for Patient Safety Organizations (PSO) may offer support in this shift. The first and one of the only to do so successfully to date at scale, Pascal has demonstrated unequivocally that, by using an evidence-based “trigger-driven” adverse event measurement method, health systems can find over 10x the level of serious harm as compared to the traditional approach of voluntary event reporting, i.e. “See something, Say something.” Therefore, PSOs are important when finding this level of harm to protect the “culture of safety” and provide a safe harbor for health systems who, to their credit, are investing to find harm and, understandably, have limited resources that require prioritizing which adverse events to address first, second, third, and so forth. When health systems have never measured accurately before, it should come as no surprise that much will be found. That happens in every field measuring for the first time. The first health systems to do this will not only gain and retain the trust of patients, they will deliver better care more broadly. [NB: Pascal still holds that event reporting should be done, especially to identify harm or near misses not identifiable in electronic health record data, as well as promoting a culture of safety – but not be relied upon for measurement, and certainly not at the prices vendors traditionally have charged.]
  4. Continued development and promulgation of eCQMs does not have to be mutually exclusive with regulation that requires that some adverse events must be clinically validated. Just as NHSN has required for many years with infection-related harm, sound epidemiology, peer-reviewed publication, and real-world evidence in the patient safety domain affirm the necessity of clinical adjudication of adverse events in order to generate scientifically validated and clinically credible adverse event outcomes data. Pascal has demonstrated not only that this can be done at scale but that doing so can create CFO-grade financial value as well – paying for itself and generating a return. By all means, require machine-only generated eCQMs for certain measures, but there is no reason to jettison the important role of clinical adjudication and validation. Doing so only will only compromise the credibility and utility of compliance-based measures and generate more cost – deepening the divide between compliance activity and that activity that results in better care at lower cost in a world where value-based incentives are beginning to knock at the door of patient safety.
  5. The most successful U.S. policies across technology-related domains have been the result of public-private partnerships, and patient safety is no exception. The most powerful levers in the policy toolkit include not only regulatory requirements but also convening power. Convening power can be used to identify methods and technologies that are working in the real world in the private sector. Once these are validated, the regulators can use their power to establish standards – accelerating the path to value as well as compliance. The private sector is able to allocate capital more quickly, and is far better positioned to develop technologies and methods that are viable and valuable in the real world. Contrariwise, private actors should not be setting policy due to conflict of interest, and public agencies have failed miserably in developing technology – and the record of contracting with large federal contractors (service companies) to build products is littered with failures (e.g. most recently, and see pieces by Brookings, the Economist, an illustrative blog here, and many more).

The reality is that there are healthcare organizations which have succeeded in implementing #1, #2, and #3 above, and adoption of an outcomes-based solution using evidence-based clinical methodology and advanced technology is happening in the field with some of the largest health systems in the United States and internationally.

Both CMS and CDC would be well served by using its convening power to learn what is happening and use these insights to establish and to accelerate the establishment of standards, e.g. common definitions of adverse events, one of the most significant impediments to progress in patient safety historically.

Moving from the regulators to the quasi-public side, Pascal calls on accreditation agencies and entities to move from relying on the “See something, Say something” approach of event reporting to assess patient safety at the hospitals they are charged to accredit; failing to do so will imperil their credibility with patients and regulators alike as leading health systems move beyond the sophistication of the ostensible monitors. And when an accrediting agency finds a hospital that needs help, the evidence suggests that measuring the results of a mitigation plan with “See something, Say something” data is grossly inadequate.

Finally, on the private side, the hospital ratings agencies would be well served to require that, if a hospital wants to be considered a top hospital in the United States, that hospital – which has spent far more funds on implementing and maintaining an electronic healthcare record – should invest what it takes to use a proven evidence-based solution proactively to look for all harm all the time across all patients – also eliminating the racial identification bias inherent in the Social Determinants of Safety (SDOS).