Measuring Patient Safety: The Medicare Patient Safety Monitoring System (Past, Present, and Future).

ABSTRACT: The explicit declaration in the landmark 1999 Institute of Medicine report "To Err Is Human" that, in the United States, 44,000 to 98,000 patients die each year as a consequence of "medical errors" gave widespread validation to the magnitude of the patient safety problem and catalyzed a number of U.S. federal government programs to measure and improve the safety of the national healthcare system. After more than 10 years, one of those federal programs, the Medicare Patient Safety Monitoring System (MPSMS), has reached a level of maturity and stability that has made it useful for the consistent measurement of the safety of inpatient care. The MPSMS is a chart review-based national patient safety surveillance system that provides rates of 21 specific hospital inpatient adverse event measures, which have been divided into 4 clinical domains (general, hospital-acquired infections, postprocedure adverse events, and adverse drug events) for analysis. The 2014 MPSMS national sample was drawn from 1109 hospitals and includes approximately 20,000 medical records of patients admitted to the hospital (all payors) for at least 1 of the 4 conditions of congestive heart failure, acute myocardial infarction, pneumonia, and major surgical procedures as defined by the Centers for Medicare and Medicaid Services Surgical Care Improvement Project. The MPSMS is now going through a major transformation to capture additional types of adverse events and is being redeveloped as the Quality and Safety Review System (QSRS). As an example of this transformation, QSRS will electronically import electronic data, which are standardized according to the Centers for Medicare and Medicaid Services billing definitions and will be updated and evolve over time to incorporate expanded standardized data available from electronic health records. This article reviews the development of MPSMS, the strengths and limitations of MPSMS, and expected future directions in patient safety measurement, focusing on those issues that are informing the development and implementation of QSRS.

A National Implementation Project to Prevent Catheter-Associated Urinary Tract Infection in Nursing Home Residents.

Importance: Catheter-associated urinary tract infection (UTI) in nursing home residents is a common cause of sepsis, hospital admission, and antimicrobial use leading to colonization with multidrug-resistant organisms. Objective: To develop, implement, and evaluate an intervention to reduce catheter-associated UTI. Design, Setting, and Participants: A large-scale prospective implementation project was conducted in community-based nursing homes participating in the Agency for Healthcare Research and Quality Safety Program for Long-Term Care. Nursing homes across 48 states, Washington DC, and Puerto Rico participated. Implementation of the project was conducted between March 1, 2014, and August 31, 2016. Interventions: The project was implemented over 12-month cohorts and included a technical bundle: catheter removal, aseptic insertion, using regular assessments, training for catheter care, and incontinence care planning, as well as a socioadaptive bundle emphasizing leadership, resident and family engagement, and effective communication. Main Outcomes and Measures: Urinary catheter use and catheter-associated UTI rates using National Healthcare Safety Network definitions were collected. Facility-level urine culture order rates were also obtained. Random-effects negative binomial regression models were used to examine changes in catheter-associated UTI, catheter utilization, and urine cultures and adjusted for covariates including ownership, bed size, provision of subacute care, 5-star rating, presence of an infection control committee, and an infection preventionist. Results: In 4 cohorts over 30 months, 568 community-based nursing homes were recruited; 404 met inclusion criteria for analysis. The unadjusted catheter-associated UTI rates decreased from 6.78 to 2.63 infections per 1000 catheter-days. With use of the regression model and adjustment for facility characteristics, the rates decreased from 6.42 to 3.33 (incidence rate ratio [IRR], 0.46; 95% CI, 0.36-0.58; P < .001). Catheter utilization was 4.5% at baseline and 4.9% at the end of the project. Catheter utilization remained unchanged (4.50 at baseline, 4.45 at conclusion of project; IRR, 0.95; 95% CI, 0.88-1.03; P = .26) in adjusted analyses. The number of urine cultures ordered for all residents decreased from 3.49 per 1000 resident-days to 3.08 per 1000 resident-days. Similarly, after adjustment, the rates were shown to decrease from 3.52 to 3.09 (IRR, 0.85; 95% CI, 0.77-0.94; P = .001). Conclusions and Relevance: In a large-scale, national implementation project involving community-based nursing homes, combined technical and socioadaptive catheter-associated UTI prevention interventions successfully reduced the incidence of catheter-associated UTIs.

Evaluating the Accuracy of Sampling Strategies for Estimation of Compliance Rate for Ventilator-Associated Pneumonia Process Measures.

BACKGROUND Measuring processes of care performance rates is an invaluable tool for quality improvement; however, collecting daily process measure data is time-consuming and burdensome. OBJECTIVE To evaluate the accuracy of sampling strategies to estimate monthly compliance rates with ventilator-associated pneumonia prevention measures. SETTING AND PARTICIPANTS A total of 37 intensive care units affiliated with 29 hospitals participating in a 2-state 35-month ventilator-associated pneumonia prevention collaborative. Analysis was limited to 325 unit-months with complete data entry rates. METHODS We calculated unit-month level actual and sample monthly compliance rates for 6 ventilator-associated pneumonia prevention measures, using 4 sampling strategies: sample 1 day per month, sample 1 day per week, sample 7 consecutive days per month, and sample 7 consecutive days per month plus additional consecutive days as necessary to obtain at least 30 ventilator-days for that month whenever possible. We compared sample versus actual rates using paired t test and χ2 test. RESULTS Mean sampling accuracy ranged 84%-97% for 1 day per month, 91%-98% for 1 day per week, 92%-98% for 7 consecutive days per month, and 96%-99% for 7 consecutive days with at least 30 days per month if possible. The most accurate sampling strategy was to sample 7 consecutive days with at least 30 ventilator-days per month if possible. With this strategy, sample rates were within 10% of actual rates in 88%-99% of unit-months and within 5% of actual rates in 74%-97% of unit-months. CONCLUSION Sampling process measures intermittently rather than continually can yield accurate estimates of process measure performance rates. Infect Control Hosp Epidemiol 2016;37:1037-1043.

A Program to Prevent Catheter-Associated Urinary Tract Infection in Acute Care.

BACKGROUND: Catheter-associated urinary tract infection (UTI) is a common device-associated infection in hospitals. Both technical factors--appropriate catheter use, aseptic insertion, and proper maintenance--and socioadaptive factors, such as cultural and behavioral changes in hospital units, are important in preventing catheter-associated UTI. METHODS: The national Comprehensive Unit-based Safety Program, funded by the Agency for Healthcare Research and Quality, aimed to reduce catheter-associated UTI in intensive care units (ICUs) and non-ICUs. The main program features were dissemination of information to sponsor organizations and hospitals, data collection, and guidance on key technical and socioadaptive factors in the prevention of catheter-associated UTI. Data on catheter use and catheter-associated UTI rates were collected during three phases: baseline (3 months), implementation (2 months), and sustainability (12 months). Multilevel negative binomial models were used to assess changes in catheter use and catheter-associated UTI rates. RESULTS: Data were obtained from 926 units (59.7% were non-ICUs, and 40.3% were ICUs) in 603 hospitals in 32 states, the District of Columbia, and Puerto Rico. The unadjusted catheter-associated UTI rate decreased overall from 2.82 to 2.19 infections per 1000 catheter-days. In an adjusted analysis, catheter-associated UTI rates decreased from 2.40 to 2.05 infections per 1000 catheter-days (incidence rate ratio, 0.86; 95% confidence interval [CI], 0.76 to 0.96; P=0.009). Among non-ICUs, catheter use decreased from 20.1% to 18.8% (incidence rate ratio, 0.93; 95% CI, 0.90 to 0.96; P<0.001) and catheter-associated UTI rates decreased from 2.28 to 1.54 infections per 1000 catheter-days (incidence rate ratio, 0.68; 95% CI, 0.56 to 0.82; P<0.001). Catheter use and catheter-associated UTI rates were largely unchanged in ICUs. Tests for heterogeneity (ICU vs. non-ICU) were significant for catheter use (P=0.004) and catheter-associated UTI rates (P=0.001). CONCLUSIONS: A national prevention program appears to reduce catheter use and catheter-associated UTI rates in non-ICUs. (Funded by the Agency for Healthcare Research and Quality.).

Engaging health care workers to prevent catheter-associated urinary tract infection and avert patient harm.

Preventing catheter-associated urinary tract infection (CAUTI) remains a significant challenge for US hospitals. The "On the CUSP: Stop CAUTI" initiative represents the single largest national effort (involving >950 hospitals) to mitigate urinary catheter risk. The program brings together key organizations to assist state hospital associations and hospitals by providing education and coaching support, addressing both the technical aspects of preventing CAUTI and CAUTI-specific socio-adaptive challenges. At the local level, engaging health care workers, from physicians and nurses to other ancillary services, is critical. This includes (1) making the importance of addressing CAUTI stakeholder specific, (2) ensuring support from leaders of essential disciplines, (3) underscoring the importance of the collaborative nature of CAUTI prevention, and (4) identifying champions within the organization to lead and be accountable for the work. Sustainability is ensured by integrating the process into the health care worker's daily routine activities.

Introduction: Taking national action to prevent and eliminate healthcare-associated infections.

BACKGROUND: The widespread prevalence and enormous cost of healthcare-associated infections (HAIs) constitute a major public health problem and patient safety concern. OBJECTIVES: In 2009, IMPAQ International and the RAND Corporation initiated an independent, outside evaluation of Health and Human Services' HAI prevention efforts as guided and driven by the Action Plan. The 3-year evaluation, whose findings are presented in this special issue, was intended to assess the outcomes of the US Department of Health and Human Services' (HHS's) past efforts, and also to provide ongoing, formative feedback to Action Plan leadership to guide their efforts. RESEARCH DESIGN: This special issue presents results from the evaluation of the Action Plan, along with related articles intended to examine the issue of HAIs from many angles. RESULTS: To address the national epidemic of HAIs, in 2009 HHS released the HHS National Action Plan to Prevent Healthcare-associated Infections, which was updated and expanded in 2012. The Action Plan established national goals for HAI prevention and identified key actions needed to reduce, prevent, and eventually eliminate the burden posed by HAIs. CONCLUSIONS: Broad lessons learned from the Action Plan evaluation document changes in structures, processes, and outcomes pertinent to eradicating HAIs, and identify lessons that are applicable to other large federal implementation efforts.

From research to nationwide implementation: the impact of AHRQ's HAI prevention program.

INTRODUCTION: The Agency for Healthcare Research and Quality (AHRQ's) Patient Safety Program is responsive to AHRQ's mission of quality improvement in healthcare. As part of this program, AHRQ has invested in projects to prevent healthcare-associated infections (HAIs), and funding has increased significantly over the last decade. AHRQ-funded projects have focused on generating new knowledge and promoting the nationwide implementation of proven HAI prevention measures in diverse healthcare settings. OBJECTIVES: To provide insight to AHRQ's HAI prevention strategies by: first, discussing the context and structure of AHRQ's HAI research portfolio and funding decisions; secondly, describing the process of prevention practice implementation and lessons learned; and third, explaining the outcomes and national impact of the AHRQ program. RESULTS AND CONCLUSIONS: In the early 2000s, AHRQ identified HAIs as an important and preventable public health threat and built their HAI-prevention portfolio based on National Action Plan priorities, available resources, advice from experts, and the state of science. This paper describes major contributions that have emerged from AHRQ-funded HAI projects. The projects examined, many of which focus on implementation of HAI prevention practices, yield useful lessons learned for future implementation and research endeavors and show significant impact of AHRQ's program in reducing HAIs.

National trends in patient safety for four common conditions, 2005-2011.

BACKGROUND: Changes in adverse-event rates among Medicare patients with common medical conditions and conditions requiring surgery remain largely unknown. METHODS: We used Medicare Patient Safety Monitoring System data abstracted from medical records on 21 adverse events in patients hospitalized in the United States between 2005 and 2011 for acute myocardial infarction, congestive heart failure, pneumonia, or conditions requiring surgery. We estimated trends in the rate of occurrence of adverse events for which patients were at risk, the proportion of patients with one or more adverse events, and the number of adverse events per 1000 hospitalizations. RESULTS: The study included 61,523 patients hospitalized for acute myocardial infarction (19%), congestive heart failure (25%), pneumonia (30%), and conditions requiring surgery (27%). From 2005 through 2011, among patients with acute myocardial infarction, the rate of occurrence of adverse events declined from 5.0% to 3.7% (difference, 1.3 percentage points; 95% confidence interval [CI], 0.7 to 1.9), the proportion of patients with one or more adverse events declined from 26.0% to 19.4% (difference, 6.6 percentage points; 95% CI, 3.3 to 10.2), and the number of adverse events per 1000 hospitalizations declined from 401.9 to 262.2 (difference, 139.7; 95% CI, 90.6 to 189.0). Among patients with congestive heart failure, the rate of occurrence of adverse events declined from 3.7% to 2.7% (difference, 1.0 percentage points; 95% CI, 0.5 to 1.4), the proportion of patients with one or more adverse events declined from 17.5% to 14.2% (difference, 3.3 percentage points; 95% CI, 1.0 to 5.5), and the number of adverse events per 1000 hospitalizations declined from 235.2 to 166.9 (difference, 68.3; 95% CI, 39.9 to 96.7). Patients with pneumonia and those with conditions requiring surgery had no significant declines in adverse-event rates. CONCLUSIONS: From 2005 through 2011, adverse-event rates declined substantially among patients hospitalized for acute myocardial infarction or congestive heart failure but not among those hospitalized for pneumonia or conditions requiring surgery. (Funded by the Agency for Healthcare Research and Quality and others.).

Eliminating central line-associated bloodstream infections: a national patient safety imperative.

BACKGROUND: Several studies demonstrating that central line-associated bloodstream infections (CLABSIs) are preventable prompted a national initiative to reduce the incidence of these infections. METHODS: We conducted a collaborative cohort study to evaluate the impact of the national "On the CUSP: Stop BSI" program on CLABSI rates among participating adult intensive care units (ICUs). The program goal was to achieve a unit-level mean CLABSI rate of less than 1 case per 1,000 catheter-days using standardized definitions from the National Healthcare Safety Network. Multilevel Poisson regression modeling compared infection rates before, during, and up to 18 months after the intervention was implemented. RESULTS: A total of 1,071 ICUs from 44 states, the District of Columbia, and Puerto Rico, reporting 27,153 ICU-months and 4,454,324 catheter-days of data, were included in the analysis. The overall mean CLABSI rate significantly decreased from 1.96 cases per 1,000 catheter-days at baseline to 1.15 at 16-18 months after implementation. CLABSI rates decreased during all observation periods compared with baseline, with adjusted incidence rate ratios steadily decreasing to 0.57 (95% confidence intervals, 0.50-0.65) at 16-18 months after implementation. CONCLUSION: Coincident with the implementation of the national "On the CUSP: Stop BSI" program was a significant and sustained decrease in CLABSIs among a large and diverse cohort of ICUs, demonstrating an overall 43% decrease and suggesting the majority of ICUs in the United States can achieve additional reductions in CLABSI rates.

How event reporting by US hospitals has changed from 2005 to 2009.

CONTEXT: Information is needed on the performance of hospitals' adverse-event reporting systems and the effects of national patient-safety initiatives, including the Patient Safety and Quality Improvement Act (PSQIA) of 2005. Results are presented of a 2009 survey of a sample of non-federal US hospitals and changes between 2005 and 2009 are examined. METHODS: The Adverse Event Reporting System survey was fielded in 2005 and 2009 using a mixed-mode design with stratified random samples of non-federal US hospitals; risk managers were respondents. Response rates were 81% in 2005 and 79% in 2009. RESULTS: Virtually all hospitals reported they had centralised adverse-event-reporting systems. However, scores on four performance indexes suggested that hospitals have not effectively implemented key components of reporting systems. Average index scores improved somewhat between 2005 and 2009 for supportive environment (0.7 increase; p<0.05) and types of staff reporting (0.08 increase; p<0.001). Average scores did not change for timely distribution of event reports or discussion with key departments and committees. Some within-hospital inconsistencies in responses between 2005 and 2009 were found. These self-reported responses may be optimistic assessments of hospital performance. CONCLUSIONS: The 2009 survey confirmed improvement needs identified by the 2005 survey for hospitals' event reporting processes, while finding signs of progress. Optimising the use of surveys to assess the effects of national patient-safety initiatives such as PSQIA will require decreasing within-hospital variations in reporting rates.

Establishing a global learning community for incident-reporting systems.

BACKGROUND: Incident-reporting systems (IRS) collect snapshots of hazards, mistakes and system failures occurring in healthcare. These data repositories are a cornerstone of patient safety improvement. Compared with systems in other high-risk industries, healthcare IRS are fragmented and isolated, and have not established best practices for implementation and utilisation. DISCUSSION: Patient safety experts from eight countries convened in 2008 to establish a global community to advance the science of learning from mistakes. This convenience sample of experts all had experience managing large incident-reporting systems. This article offers guidance through a presentation of expert discussions about methods to identify, analyse and prioritise incidents, mitigate hazards and evaluate risk reduction.

Learning from patients' experience: enhancing patient safety event reporting systems.

UNLABELLED: BACKGGROUND: everyone has a personal story of an incident in which the healthcare system has caused harm to a family member, friend, or work colleague. In 2004, one in three Americans (34%) said that they or a family member had experienced a preventable medical error; among them, 21% said the error caused "serious health consequences" such as death (8%), long-term disability (11%) or severe pain (16%). DISCUSSION: the information patients give is important and can be part of a strategic model to make systemic changes to improve health outcomes and patient safety. It has been identified that one of these shortcomings is that patients' complaints are not considered able to judge technical quality in their experience with care. We argue for an approach which should actively engage patients and their caregivers in contemplating and describing their experiences as a means to gather evidence about risks and hazards in the healthcare setting. CONCLUSIONS: patients ought to be viewed as partners with health care providers to improve patient safety; self-reports on adverse events can provide useful information that may be incorporated into patient safety event. Data obtained from this strategy should be useful to improve general changes in health care and a better clinical practice based on evidence.

Evaluation of the AHRQ patient safety initiative: framework and approach.

OBJECTIVE: Describe the evaluation performed of the patient safety initiative operated by the Agency for Healthcare Research and Quality (AHRQ). AHRQ PATIENT SAFETY INITIATIVE When patient safety became a national priority in 2000, Congress charged and funded AHRQ to improve health care safety. Over the next 6 years, AHRQ funded more than 300 research projects and other activities, addressing diverse patient safety issues and practices. THE PATIENT SAFETY EVALUATION: AHRQ contracted with RAND in 2002 to perform a 4-year evaluation of the initiative, which was completed in 2006. This formative evaluation used the CIPP program evaluation model, which emphasizes multiple stakeholders' interests (e.g., patients, providers, funded researchers). We monitored the progress of the patient safety initiative and provided AHRQ annual feedback that assessed each year's activities, identifying issues and offering suggestions for actions by AHRQ. Given the size and complexity of the initiative, the evaluation needed to examine key individual components and synthesize results across them, and it also had to be responsive to changes in the initiative over time. We used a conceptual framework to bring together the disparate pieces to synthesize overall findings. The remaining articles in this issue describe selected results from this evaluation.

The validity of ICD-9-CM codes in identifying postoperative deep vein thrombosis and pulmonary embolism.

BACKGROUND: Deep vein thrombosis and pulmonary embolism (DVT/PE) are common complications after surgery and are associated with substantial excess mortality and length of stay. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes recorded in hospital claims have been used to identify and study DVT/PE, but the validity of this method is not well studied. METHODS: Identification of postoperative DVT/PE events were compared using ICD-9-CM codes and medical record abstraction in random samples of hospital discharges of Medicare beneficiaries in 2002-2004. RESULTS: Among 20,868 eligible surgical hospitalizations, 232 DVT cases and 95 PE cases were identified by ICD-9-CM codes; 108 DVT cases and 31 PE cases by medical record abstraction; 72 DVT cases and 23 PE cases by both methods. The resulting estimates of PPV of ICD9-CM coding were 31% (72/232 cases) for DVT, 24% (23/95) for PE, and 29% (90/308) for DVT/PE combined. The resulting sensitivity estimates were 67% (72/108 cases) for DVT, 74% (23/31) for PE, and 68% (90/133) for DVT/PE combined. DISCUSSION: ICD-9-CM codes in Medicare claims are sensitive but have limited predictive validity in identifying postoperative DVT/PE. Improvements in the validity are needed before the indicator can be used for safety performance assessment.

Probabilistic risk assessment of accidental ABO-incompatible thoracic organ transplantation before and after 2003.

BACKGROUND: A widely reported ABO-mismatch accident in March of 2003 raised concerns about the reliability of the transplantation system. Because this type of failure is rare and significant, we performed a probabilistic risk assessment (PRA) of the donor-recipient matching processes for thoracic organ transplantation. METHODS: A probabilistic risk assessment was performed. RESULTS: The likelihood of accidental incompatible implantation was already low in 2003. The PRA model indicates that the likelihood of such an event was 1.38x10 per donated organ. This estimate correlates closely with the observed rate of these accidents. Based on this model, process changes put in place shortly after the accident reduced the probability to approximately 3.08x10 and changes put in place in October 2004 further reduced the probability to approximately 2.22x10 per organ donated. CONCLUSIONS: The observed and predicted likelihoods of accidental incompatible thoracic organ transplantation are comparable. These likelihoods are several orders of magnitude smaller than other hazards associated with solid organ transplantation. The PRA model indicates that changes that followed the March 2003 accident further reduced the likelihood of accidental incompatible implantation by roughly two orders of magnitude. Quantitative estimates from PRA can be used to assess risks in healthcare and to gauge the impact of system changes on these risks.