A Decade of Preventing Harm.

Medical errors are a significant source of morbidity and mortality, and while focused efforts to prevent harm have been made, sustaining reductions across multiple categories of patient harm remains a challenge. In 2008 BJC HealthCare initiated a systemwide program to eliminate all major causes of preventable harm and mortality over a five-year period with a goal of sustaining these reductions over the subsequent five years. METHODS: Areas of focus included pressure ulcers, adverse drug events, falls with injury, health care-associated infections, and venous thromboembolism. Initial efforts involved building system-level multidisciplinary teams, utilizing standardized project management methods, and establishing standard surveillance methods. Evidence-based interventions were deployed across the system; core standards were established while allowing for flexibility in local implementation. Improvements were tracked using actual numbers of events rather than rates to increase meaning and interpretability by patients and frontline staff. RESULTS: Over the course of the five-year intervention period, total harm events were reduced by 51.6% (10,371 events in 2009 to 5,018 events in 2012). Continued improvement efforts over the subsequent five years led to additional harm reduction (2,605 events in 2017; a 74.9% reduction since 2009). CONCLUSION: A combination of project management discipline, rigorous surveillance, and focused interventions, along with system-level support of local hospital improvement efforts, led to dramatic reductions in preventable harm and long-term sustainment of progress.

Decision Analysis for Metric Selection on a Clinical Quality Scorecard.

Clinical quality scorecards are used by health care institutions to monitor clinical performance and drive quality improvement. Because of the rapid proliferation of quality metrics in health care, BJC HealthCare found it increasingly difficult to select the most impactful scorecard metrics while still monitoring metrics for regulatory purposes. A 7-step measure selection process was implemented incorporating Kepner-Tregoe Decision Analysis, which is a systematic process that considers key criteria that must be satisfied in order to make the best decision. The decision analysis process evaluates what metrics will most appropriately fulfill these criteria, as well as identifies potential risks associated with a particular metric in order to identify threats to its implementation. Using this process, a list of 750 potential metrics was narrowed to 25 that were selected for scorecard inclusion. This decision analysis process created a more transparent, reproducible approach for selecting quality metrics for clinical quality scorecards.

Multifaceted approach to reducing occurrence of severe hypoglycemia in a large healthcare system.

PURPOSE: Substantial reductions in inpatient episodes of severe hypoglycemia achieved by a large healthcare system through enhanced use of technology and sustained quality-improvement initiatives are described. SUMMARY: After internal data collection and analysis revealed that severe hypoglycemia accounted for 75% of all systematically monitored adverse drug events in its hospital network, St. Louis-based BJC HealthCare designed and executed a multifaceted approach to reducing hypoglycemia events. Initiated by a pharmacist-led task force, the project entailed (1) automated event detection and creation of dashboards for comparing hypoglycemia rates among at-risk patients at 11 BJC facilities, (2) implementation of evidence-based and internal best practices in use at BJC's top-performing hospitals, (3) development of an online "Hypoglycemic Event Analysis Tool" (HEAT) to support event investigation and collection of data on causative factors, and (4) the assembly of targeted interventions at a "Hypoglycemia Facility Strategy Tracking" (H-FaST) intranet site. As a result of the launch of the HEAT and H-FaST tools and associated provider education activities, the systemwide rate of hypoglycemia events in the specified at-risk patient population declined from 6.45 per 1000 patient-days during a preimplementation baseline period (July-December 2009) to 1.32 per 1000 patient-days during a designated postimplementation period (January-June 2014), an 80% overall reduction in hypoglycemia (p < 0.01); reductions in severe hypoglycemia events ranging from 70% to 100% were observed at all 11 hospitals. CONCLUSION: A multifaceted, evidence-based, data-driven approach enabled a large healthcare system to markedly reduce the frequency of severe hypoglycemia events.

The cost of serious fall-related injuries at three Midwestern hospitals.

BACKGROUND: Consequences of fall-related injuries can be both physically and financially costly, yet without current data, hospitals cannot completely determine the financial cost. As part of the analysis for an initiative to minimize falls with injury, the cost and length of stay attributable to serious fall injury were estimated at three hospitals in a Midwestern health care system METHODS: In a retrospective case-control study, 57 hospital inpatients discharged between January 1, 2004, and October 16, 2006, who sustained a serious fall-related injury (fracture, subdural hematoma, any injury resulting in surgical intervention, or death) were identified through the incident reporting system and matched to nonfaller inpatient controls by hospital, age within five years, year of discharge, and diagnosis-related group (DRG). RESULTS: Multivariate analyses indicated that operational costs for fallers with serious injury, as compared with controls, were $13,316 more (p < .01; 95% confidence interval [CI], $1,395-$35,561) and that fallers stayed 6.3 days longer than nonfallers (p < .001; 95% CI, 2.4-14.9). Univariate analyses indicated they were also significantly more likely to have diabetes with organ damage, moderate to severe renal disease, and a higher mean score on the Charlson Comorbidity Index. In optimal bipartite matching (OBM) analyses, fallers with serious injury cost $13,806 more (p < .001; 95% CI, $5,808-$29,450) and stayed 6.9 days longer (p < .001; 95% CI, 2.8-14.9). CONCLUSIONS: Hospital inpatients who sustained a serious fall-related injury had higher total operational costs and longer lengths of stay than nonfallers. Despite possible limitations regarding the cost allocation methods, the analysis included data from three different hospitals, and supplemental multivariate analyses adjusting for academic hospital status did not meaningfully affect the results.

Inpatients' willingness to recommend: a multilevel analysis.

BACKGROUND: Satisfaction with health care is one of the most widely assessed measures of hospital care quality, yet studies that account for clustering effects are uncommon. We constructed a multilevel model to identify predictors of willingness to recommend while controlling for clustering effects due to hospital and care unit. We also examined differences in predictors by care unit. PURPOSE: The aim of this study was to identify factors that both influence patient perceptions of care and are potentially modifiable by the hospital delivering care. METHODOLOGY: Our sample includes Hospital Consumer Assessment of Healthcare Providers and Systems survey data collected between July 1, 2007, and June 30, 2008, for 131 hospitals and 33,445 patients. The primary outcome was willingness to recommend the hospital to family and friends. Variables were collected at three levels: patient (Hospital Consumer Assessment of Healthcare Providers and Systems survey item responses and demographics), care unit, and hospital. Data were analyzed using multilevel modeling. We also ran a series of two-level models to explore differences in predictors by care type. FINDINGS: The strongest predictors of willingness to recommend, controlling for clustering effects, were items that generally reflected interpersonal aspects of care such as nursing and physician behaviors. In the two-level models, predictors of willingness to recommend overlapped across care units, but important differences were noted. PRACTICE IMPLICATIONS: Our results suggest that hospitals that wish to improve their performance would benefit most from focusing on interpersonal aspects of care. Hospitals that focus resources on improving in these areas, that assess care units separately, and that investigate the meaning and context of survey responses will be most likely to see improvements in satisfaction scores.

Clinical validation of the AHRQ postoperative venous thromboembolism patient safety indicator.

BACKGROUND: The Agency for Healthcare Research and Quality (AHRQ) patient safety indicators (PSIs) screen for potentially preventable complications in hospitalized patients using hospital administrative data. The PSI for postoperative venous thromboembolism (VTE) relies on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes for deep vein thrombosis (DVT) or pulmonary embolism (PE) in secondary diagnoses fields. In a clinical validation study of the PSI for postoperative VTE, natural language processing (NLP), supplemented by pharmacy and billing data, was used to identify VTE events missed by medical records coders. METHODS: In a retrospective review of postsurgical discharges, charts were processed using the AHRQ PSI software. Cases were identified as possible false negatives by flagging charts for possible VTEs using pharmacy and billing data to identify all patients who were therapeutically anticoagulated or had placement of an inferior vena caval filter. All charts were reviewed by a physician blinded to screening results. Physician interpretation was considered the gold standard for VTE classification. RESULTS: The AHRQ PSI had a positive predictive value (PPV) of .545 (95% confidence interval [CI], .453-.634) and a negative predictive value (NPV) of .997 (95% CI, .995-.999). Sensitivity was .87 and specificity was .98. Secondary coding review suggested that all 9 false-negative results were miscoded; if they had been properly coded, the sensitivity would increase to 1.00. Most false-positive cases resulted from superficial venous clots identified by the PSI due to coding ambiguity. DISCUSSION: The VTE PSI performed well as a screening tool but generated a significant number of false-positive cases, a problem that could be substantially reduced with improved coding methods.

Computerized surveillance for adverse drug events in a pediatric hospital.

Adverse drug event (ADE) surveillance is needed to inform processes and methods for prevention. Voluntary reporting and manual chart review have limitations. Automated surveillance systems are gaining recognition and provide useful information to supplement the other methods. Preliminary evaluation of a pediatric automated adverse drug event application shows a positive predictive value of 13%, discovering events with harm in 1.3% of inpatient admissions.

Improving adherence to dyslipidemia medication guidelines in hospitalized diabetic patients using a technology-assisted pharmacist intervention.

We tested whether a technology-assisted pharmacist intervention improved physician adherence to guidelines for lipid-lowering therapy in diabetic patients. Computerized alerts identified diabetic patients above LDL-Cholesterol (LDL-C) goal. During Period 1 prescribing behavior was observed in both control and intervention physician groups without intervening. In Period 2, pharmacists conducted academic detailing with intervention group physicians. Control group physicians were observed. The intervention significantly improved the proportion of diabetic patients discharged on statin therapy.

Improving healthcare-associated infection surveillance at a multi-hospital institution using an existing data repository.

The accurate and timely reporting of healthcare-associated infections is an essential infection control practice. Rates provide benchmarks for detecting trends and can help facilities identify intervention opportunities. Standardizing how hospitals within an organization calculate these rates is critical if the rates are to be compared among hospitals and to national standards. We describe a new web-based tool to improve how infection control practitioners identify and report healthcare-associated infections at BJC HealthCare.

Comparison of mortality risk adjustment using a clinical data algorithm (American College of Surgeons National Surgical Quality Improvement Program) and an administrative data algorithm (Solucient) at the case level within a single institution.

BACKGROUND: There is great interest in efficiently evaluating health care quality, but there is controversy over the use of administrative versus clinical data methods. We sought to compare actual mortality with risk-adjusted expected mortality in a sample population calculated by two different methods; one based on preexisting administrative records and one based on chart reviews. STUDY DESIGN: We examined a sample of patients (n = 1,234) undergoing surgical procedures at an academic teaching hospital during 1 year. The first risk-adjustment method was that used by the National Surgical Quality Improvement Program, which is based on dedicated medical record review. The second method was that used by Solucient, LLC, which is based on preexisting administrative records. RESULTS: The ratio of observed to expected mortality for this population set was higher using the National Surgical Quality Improvement Program algorithm (1.1; 95% CI, 0.8 to 1.5) than using the Solucient algorithm (0.9; 95% CI, 0.6 to 1.2) but neither estimate was notably different from 1.0. Similarly, when observed to expected mortality ratios were calculated separately for each quartile of mortality, there were no marked differences within quartiles, although minor differences with potential importance were noted. Fit was comparable by age categories, gender, and American Society of Anesthesiologists' categories. A number of actual deaths had higher predicted mortality scores using the Solucient algorithm. CONCLUSIONS: Risk-adjusted mortality estimates were comparable using administrative or clinical data. Minor performance differences might still have implications. Because of the potential lower cost of using administrative data, this type of algorithm can be an efficient alternative and should continue to be investigated.

Circumstances of patient falls and injuries in 9 hospitals in a midwestern healthcare system.

OBJECTIVE: Preventing hospital falls and injuries requires knowledge of fall and injury circumstances. Our objectives were to determine whether reported fall circumstances differ among hospitals and to identify predictors of fall-related injury. DESIGN: Retrospective cohort study. Adverse event data on falls were compared according to hospital characteristics. Logistic regression was used to determine adjusted odds ratios (aORs) with 95% confidence intervals (CIs) for risk factors for fall-related injury. SETTING: Nine hospitals in a Midwestern healthcare system. PATIENTS: Inpatients who fell during 2001-2003. RESULTS: The 9 hospitals reported 8,974 falls that occurred in patient care areas, involving 7,082 patients; 7,082 falls were included in our analysis. Assisted falls (which accounted for 13.3% of falls in the academic hospital and 9.8% of falls in the nonacademic hospitals; P<.001) and serious fall-related injuries (which accounted for 3.7% of fall-related injuries in the academic hospital and 2.2% of fall-related injuries in the nonacademic hospitals; P<.001) differed by hospital type. In multivariate analysis for the academic hospital, increased age (aOR, 1.006 [95% CI, 1.000-1.012]), falls in locations other than patient rooms (aOR, 1.53 [95% CI, 1.03-2.27]), and unassisted falls (aOR, 1.70 [95% CI, 1.23-2.36]) were associated with increased injury risk. Altered mental status was associated with a decreased injury risk (aOR, 0.72 [95% CI, 0.58-0.89]). In multivariate analysis for the nonacademic hospitals, increased age (aOR, 1.007 [95% CI, 1.002-1.013]), falls in the bathroom (aOR, 1.46 [95% CI, 1.06-2.01]), and unassisted falls (aOR, 1.83 [95% CI, 1.37-2.43]) were associated with injury. Female sex (aOR, 0.83 [95% CI, 0.71-0.97]) was associated with a decreased risk of injury. CONCLUSION: Some fall characteristics differed by hospital type. Further research is necessary to determine whether differences reflect true differences or merely differences in reporting practices. Fall prevention programs should target falls involving older patients, unassisted falls, and falls that occur in the patient's bathroom and in patient care areas outside of the patient's room to reduce injuries.

Reporting and disclosing medical errors: pediatricians' attitudes and behaviors.

OBJECTIVE: To characterize pediatricians' attitudes and experiences regarding communicating about errors with the hospital and patients' families. DESIGN: Cross-sectional survey. SETTING: St Louis, Mo, and Seattle, Wash. PARTICIPANTS: University-affiliated hospital and community pediatricians and pediatric residents. MAIN EXPOSURE: Anonymous 68-item survey (paper or Web-based) administered between July 2003 and March 2004. MAIN OUTCOME MEASURES: Physician attitudes and experiences about error communication. RESULTS: Four hundred thirty-nine pediatric attending physicians and 118 residents participated (62% response rate). Most respondents had been involved in an error (39%, serious; 72%, minor; 61%, near miss; 7%, none). Respondents endorsed reporting errors to the hospital (97%, serious; 90%, minor; 82%, near miss), but only 39% thought that current error reporting systems were adequate. Most pediatricians had used a formal error reporting mechanism, such as an incident report (65%), but many also used informal reporting mechanisms, such as telling a supervisor (47%) or senior physician (38%), and discussed errors with colleagues (72%). Respondents endorsed disclosing errors to patients' families (99%, serious; 90%, minor; 39%, near miss), and many had done so (36%, serious; 52%, minor). Residents were more likely than attending physicians to believe that disclosing a serious error would be difficult (96% vs 86%; P = .004) and to want disclosure training (69% vs 56%; P = .03). CONCLUSIONS: Pediatricians are willing to report errors to hospitals and disclose errors to patients' families but believe current reporting systems are inadequate and struggle with error disclosure. Improving error reporting systems and encouraging physicians to report near misses, as well as providing training in error disclosure, could help prevent future errors and increase patient trust.

Long-term effectiveness of an automated guideline adherence monitor for secondary prevention of acute myocardial infarction.

An automated guideline monitor for secondary prevention of acute myocardial infarction (AMI) has been favorably tested at an academic medical center using a randomized controlled trial. Subsequently, the monitor was implemented in a production mode and has been in production use for five years. Statistical process control analysis shows a generally sustainable effect of the intervention.

Physicians' knowledge and attitudes about coronary heart disease secondary prevention.

We administered a knowledge and attitudes questionnaire regarding a technology assisted pharmacist mediated academic detailing intervention designed to improve physician adherence to coronary heart disease (CHD) secondary-prevention guidelines. Questionnaires were administered in two settings: an academic hospital and a community hospital. More knowledgeable physicians reported being more likely to prescribe a pharmacists' recommended medication and to agree that lipid profiles should be automatically performed for inpatients with elevated troponin.

Natural language processing to identify venous thromboembolic events.

The Agency for Healthcare Research and Quality (AHRQ) has promulgated patient safety indicators to identify potentially preventable adverse safety events, including venous thromboembolism (VTE). Identification of these events for quality reporting is commonly done with AHRQ-defined ICD9-CM codes. We tested a natural language processing service (NLP) as an alternative method of identification.

Automated clinical data collection for national quality measurement reporting.

Collecting data for the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) ORYX Core Measurement Reporting can be automated using an object-oriented, client-developed program that extracts data from a clinical data repository and utilizes an MHA vendor upload process. The process eliminated 39% of the manual data collection efforts.

Migrating toward a next-generation clinical decision support application: the BJC HealthCare experience.

The next-generation model outlined in the AMIA Roadmap for National Action on Clinical Decision Support (CDS) is aimed to optimize the effectiveness of CDS interventions, and to achieve widespread adoption. BJC HealthCare re-engineered its existing CDS system in alignment with the AMIA roadmap and plans to use it for guidance on further enhancements. We present our experience and discuss an incremental approach to migrate towards the next generation of CDS applications from the viewpoint of a healthcare institution. Specifically, a CDS rule engine service with a standards-based rule representation format was built to simplify maintenance and deployment. Rules were separated from execution code and made customizable for multi-facility deployment. Those changes resulted in system improvement in the short term while aligning with long-term strategic objectives.

A technology-assisted approach for discontinuing contact isolation.

While contact isolation can be an effective tool in reducing the transmission of nosocomial methicillin-resistant Staphylococcus aureus (MRSA), it can increase costs and may decrease the quality of patient care. Therefore, it is important to insure that only patients who need contact precautions are isolated. We describe a simple automated report to help infection control practitioners pro-actively identify patients that may no longer need isolation.

Automated dose checking and intervention for bariatric patients.

According to a recent Center for Disease Control survey, 33% of the US population is obese. Because labeled dosing guidelines are based upon non-obese individuals, under dosing of antibiotics may be problematic in this population. We developed an automated dose checking tool to efficiently detect potentially inappropriate dosing of antibiotics in bariatric (morbidly obese patients).

Improving adherence to coronary heart disease secondary prevention medication guidelines at a community hospital.

Using a two-period group randomized study, we tested whether a technology assisted pharmacist intervention improved physician adherence to coronary heart disease (CHD) secondary prevention medication guidelines. After an observation period, physician practices were randomized to intervention or control arms. In the intervention arm, alerts prompted a pharmacist to communicate with the responsible physician about secondary prevention medications. The intervention significantly improved the proportion of patients discharged on appropriate secondary prevention medications.