Are Safety-Culture Surveys a Window into Patient Safety?

Few hospitals even disclose that they use such surveys, let alone reveal the result.


High-quality patient care hinges just as much on keeping patients safe as much as on treating them correctly and skillfully. Medication errors, deadly infections and poor communication persist 14 years after the Institute of Medicine concluded in the headline-generating "To Err is Human" report that mistakes kill as many as 98,000 hospital patients a year. The IOM issued a followup prescription, "Crossing the Quality Chasm: A New Health System for the 21st Century," a few years later. But even optimists say progress has been measured at best.

A promising sign, however, is that more hospitals are conducting internal surveys to determine the extent to which the organizational culture helps or hinders patient safety. Some of these safety-culture surveys are crude and homegrown. Others, however, have been carefully assembled by consensus organizations and are sufficiently robust, given an adequate response rate, to allow analysis at the level of individual units within a hospital, such as the cardiac ICU or the oncology service.

The Safety Attitudes Questionnaire (SAQ), the first of its kind and widely used, is a direct descendant of aviation's Flight Management Attitudes Questionnaire. Several of the original questions, in fact, such as "Fatigue impairs my work in critical situations" and "When my workload becomes excessive, my performance is impaired," have been retained, says its developer, J. Bryan Sexton, director of the Duke University Health System Patient Safety Center. "We borrowed an entire domain from one industry and transferred it to another."

Another instrument, the Hospital Survey on Patient Safety Culture, is offered for free to hospitals by the federal Agency for Healthcare Research and Quality (AHRQ). Aggregated survey results for 2012 from about 570,000 individuals at more than 1,200 hospitals are displayed online.

Such surveys measure aspects of patient safety culture that include caregivers' perception of the institutional support for patient safety, frequency of adverse events, quality of handoffs and transitions, comfort in reporting a potential problem or error, and level of teamwork within hospital units and the organization as a whole.

Safety culture and Best Hospitals. Safety culture is an appealing metric for evaluating hospital performance. U.S. News is currently consulting with experts about the possibility of building it into both the Best Hospitals ranking methodology and our pending evaluation of individual hospitals' performance in high-volume conditions and procedures.

The Keystone ICU Project is a striking demonstration of how a well-done survey can enhance care, says Allan Frankel, who helped pioneer the use of patient safety culture surveys and co-founded Safe & Reliable Healthcare, a consulting group that helps hospitals improve their clinical quality and patient safety. The project's goal was to determine whether catheter-related bloodstream infections (CLABSI) could be reduced in 103 intensive-care units at 67 Michigan hospitals, starting in October 2003.

The plan involved improving the safety culture – starting with the use of the SAQ to determine individual ICU results and inform a comprehensive improvement program – and making a few basic changes such as introducing checklists to insure adherence to clinical guidelines, handwashing, using full-barrier precautions and disinfecting the skin.

It was a resounding success. The hospitals' median infection rate dropped from 2.7 per 1,000 patients to 0 and stayed there through 18 months of follow-up. Teams with the best safety and teamwork climate also demonstrated the most rapid improvement in CLABSI rates.

Overall, Michigan hospitals reduced the incidence of CLABSI from 2004 to 2012 by more than 50 percent and reduced the incidence of ventilator pneumonia by 60 percent, according to the Michigan Hospital Association's Keystone Center.

The message from Keystone is clear to the project's lead investigator, Peter Pronovost, director of the Quality and Safety Research Group at Johns Hopkins Hospital: "Safety culture is measurable, it seems to correlate with outcomes, and, as our study shows, it's responsive to interventions."

The use of safety culture surveys has been boosted by a mandate from the principal U.S. hospital accrediting body, the Joint Commission, that each hospital survey its staffers at least every three years to determine how well it encourages blame-free reporting of errors and unsafe conditions. The standard requires hospital leaders to "prioritize and implement" solutions to problems identified by the survey and to develop a code of conduct that defines acceptable behaviors and behaviors that undermine safety.

What patient, or prospective employee, for that matter, wouldn't want to know whether a given unit's staff agreed or disagreed with the statement: "I would feel safe being treated here as a patient"? Unfortunately, survey results are unavailable to the public or U.S. News. We know of only one hospital, the University of Michigan Hospitals and Health Centers, willing to display survey results on its website; we'd like to hear from others that do the same. AHRQ's data-use agreement obligates the agency to keep hospital results confidential. AHRQ declines even to release the names of hospitals that participate in the survey process. Understandably, consultants like Pascal Metrics, which supports both the SAQ and AHRQ surveys in some 700 hospitals, do not release customer lists or survey results.

Perhaps U.S. News could develop a metric in the absence of hospital-specific results. One possibility would be to give credit to hospitals that affirm to U.S. News that they actively participate in a validated survey, conduct it at the unit level, achieve a sufficient response rate and have an action plan to take the results and transform them into real improvement. Yes-no credit of this kind would only scratch the surface. Full disclosure of survey results would be better. But it would be a start.

More from Second Opinion:

  • Broadening the Mission of Best Hospitals
  • What the Best Hospital Rankings Are – and Aren't
  • When Hospitals Market 'Lifesaving' Tests