Making Healthcare Safer III: Executive Summary

ES. 1 Background/Introduction

The Making Health Care Safer reports from the Agency for Healthcare Research and Quality (AHRQ) have had an important role in reducing harm and improving the safety and quality of care for patients. The reports—providing an analysis of the evidence for various patient safety practices (PSPs)—have served as a source of information for multiple stakeholders, including healthcare providers, health system administrators, researchers, and government agencies. The reports have also identified contextual factors that contribute to successful PSP implementation. The reports have helped to shape national discussion regarding patient safety issues on which providers, payers, policymakers, and patients and families should focus attention.^1,2

Since the second report was published in 2013, there have been many improvements in patient safety. Building on the success of PSPs in inpatient settings, AHRQ is seeking to support a culture of safety across the healthcare continuum, including in nursing homes, home care, outpatient, and ambulatory settings, and during care transitions. Making Healthcare Safer III has made strides in transitioning from a predominantly acute care PSP review to include PSPs from other settings and during transitions. The scope of this report has also expanded to match emerging themes and strategic goals championed by the U.S. Department of Health and Human Services, including addressing the opioid crisis and emerging health risks (e.g., multidrug-resistant organisms), and overall directives to “put patients first” and to reduce provider burden and burnout.

The Making Healthcare Safer III report project team, composed of Abt Associates and IMPAQ International, began its work by developing a new conceptual framework that does the following: (1) puts the patient in the center; (2) acknowledges that patients are constantly exposed to potential harms; and (3) proposes patient safety approaches that mitigate patients’ past and future vulnerabilities. We have thus taken an approach that is both holistic (considering the whole patient through the continuum of care) and targeted (focusing on what harms are relevant to a particular patient at a particular point in care). Additionally, by following the patient, this framework includes harms during movement between settings and harm risks from existing vulnerabilities and disparities. Starting from the new conceptual framework, we organized the report by “harm areas.” This is intended to make the report easier to access for all patient safety stakeholders, who will be able to quickly locate topics of interest and importance to their particular needs and circumstances.

ES. 2 Objectives

The purpose of the Making Healthcare Safer III report is to create a source of information on practices that can improve patient safety across a variety of settings and stakeholders. For this report, patient safety practices are defined as “discrete and clearly recognizable structures or processes used for the provision of care that are intended to reduce the likelihood and/or severity of harm due to systems, processes, or environments of care.”³ A PSP may have varying degrees of evidence to support its ability to prevent or mitigate harm or its use in specific contexts.

ES.3 Methods

The methods by which the full project team—including AHRQ and the patient safety and clinical experts on the Advisory Group (AG) and Technical Expert Panel (TEP)—completed the report are outlined in Table 1.

Table 1: Six-Step Process to Developing the Making Health Care Safer III Report

	Steps
1.	Development of Conceptual Framework
2.	Identification, Selection, and Prioritization of Harm Area Topics
3.	Identification, Selection, and Prioritization of Patient Safety Practices
4.	Literature Searches
5.	Review of the Evidence
6.	Report Development

ES.3.1 Step 1. Development of Conceptual Framework

ES.3.1.1 Description of Framework

To help guide the development and content of this report, the project team created a patient-centric framework of safety. The framework focuses on the experience of the individual as they interact with the health care system throughout various phases of health and in different settings (Figure 1).

Figure 1: Framework for Making Healthcare Safer III Report

Figure 1 outlines the role of patient safety practices to mitigate threats to safety and pre-existing contextual factors as an individual moves through various settings of care and health phases.

The underlying state of the individual is wellness, in which the patient may be receiving intermittent preventive care, such as surveillance for diseases or immunizations, or receiving regular care to maintain stability of chronic conditions. As the patient moves from state to state, they interact with different providers in different settings and the resources, tools, culture, and environments specific to those settings. Threats to safety including medication management issues, healthcare-associated infections, nursing-sensitive events, procedural events, and diagnostic errors can be present during these interactions and patient safety practices (PSP), which are the focus of this report, are used during the provision of care to mitigate the effects of these threats.

ES.3.2 Step 2. Identification, Selection, and Prioritization of Harm Area Topics

The project team conducted an environmental scan of patient safety resources to identify existing and potentially new harm areas. Sources reviewed included AHRQ’s PSNet website, the National Quality Strategy, the Joint Commission’s National Patient Safety Goals, the National Quality Forum’s 2015 Patient Safety Report, the Centers for Medicare & Medicaid Services Hospital Value-Based Purchasing Program and Partnership for Patients, ECRI Institute’s 2017 and 2016 Top 10 Patient Safety Issues briefs, and Becker’s Hospital Review 10 Top Patient Safety Issues for 2018.^4-12 This scan identified 8 broad categories of harm (e.g., healthcare-associated infections) and 74 specific harm area topics (e.g., Clostridium difficile infection).

The AG performed an initial review of the identified harm areas and topics, resulting in the exclusion of seven topics deemed outside the scope of this report. In order to determine which of the remaining 67 topics should be included, the TEP and AG prioritized the topics on a scale of 1 to 5 (1 = low priority; 5 = high priority) and provided feedback through an electronic survey. The results were presented to AHRQ and, after several iterative rounds, which included adding several topics not among the initial 67 harm area topics (shown in Table 2 of the report's Methods section), a total of 17 harm area topics (shown in Table 3 of the report’s Methods section) were identified for inclusion in the report.

ES.3.3 Step 3. Identification, Selection, and Prioritization of Patient Safety Practices

The project used guidelines and systematic reviews to identify PSPs for the 17 harm area topics. The inclusion of a PSP in the report was based on how many TEP and AG members selected a specific PSP as a high priority for inclusion. After a comprehensive review of the results by AHRQ, a master list of 47 PSPs composed of 40 core PSPs and 7 additional AHRQ-identified PSPs was generated.

ES.3.4 Step 4. Literature Searches

The authors identified PSP-specific search terms and the team librarian ran the search terms for every PSP in the MEDLINE and CINHAL databases while also filtering for English-language publications only between the years 2008 and 2018. The individual studies for some PSPs, such as Patient and Family Engagement and Cultural Competency, were limited; therefore, the project team followed the approach outlined by Whitlock et al. (2008), which was to search for systematic reviews first and decide if the primary literature was of a determined level of adequate quality.¹³ The MeaSurement Tool to Assess systematic Reviews (AMSTAR) was then applied to determine systematic review quality.¹⁴ The studies were screened based on the exclusion criteria established using the population, interventions, comparators, outcomes, and study designs (PICOS) criteria.¹⁵ For example, we excluded studies with specialized populations such as armed forces and pilot study designs, particularly when the PSP was considered new or developing. Other exclusion criteria included lack of rigor, small sample size), lack of intervention or protocol description, PSP not universally applicable to most settings and populations, and study found to be out of scope.

ES.3.5 Step 5. Review of the Evidence

Across the PSPs examined there was wide variation in the rigor of studies included in the evidence reviews, and individual authors were permitted to decide the minimum threshold of quality for including specific studies given the state of the field for each PSP. Similar to the previous report (Making Health Care Safer II), we aimed to apply the criteria drawn from the AHRQ Guides on strength of evidence derived from Grading of Recommendations, Assessment, Development, and Evaluations (GRADE), a framework for developing and presenting summaries of evidence.^2,16-19 To the extent possible, authors for each review indicated the strength of evidence by practice, outcome, and/or setting.

ES.3.6 Step 6. Report Development

For this report, the project team has made an effort to synthesize the evidence through a research lens, while placing it in a format that allows for practical application for the end user. Each chapter, as written by the project team with final review from AHRQ, represents a specific threat to safety (i.e., the harm) that can occur to a patient when exposed to healthcare and includes the targeted PSPs selected for review.

Given the wide variation in study availability and quality, as well as the variation in strength of evidence sometimes by setting and specific outcomes, the project team decided against providing a single determination of strength of evidence by PSP (as was done in the previous report). Instead, the project team—with expert input from AHRQ, the TEP, and the AG—determined it was most appropriate to provide tables that summarize for each harm area, by PSP, the following: key safety outcomes: Table 2 summarizes these points for each PSP to provide a user-friendly overview of what stakeholders will find in each specific review.

ES.4 Results

The results of the 47 PSPs grouped according to the 17 topics are summarized in Table 2. Instead of including strength of evidence for each PSP, the project team decided to include key takeaways as a way to allow the user to determine if the PSP is of interest or beneficial.

Table 2: Patient Safety Practice Summary Table

Patient Safety Practice	Key Safety Outcomes	Studies and Systematic Reviews (#)	Settings	Key Takeaways/Points
Diagnostic Error: Clinical Decision Support (CDS)	Diagnostic accuracy	31 studies; 3 systematic reviews; 1 meta-review; 1 meta-analysis	Inpatient and outpatient (primary care and specialty care), emergency department (ED)	CDS has been shown to improve diagnosis in exploratory and validation studies, but the systems need to be fully implemented and tested in a clinical setting. They are best used as adjuncts to the clinician’s decision-making process and not as replacements.
Diagnostic Error: Peer Review	Diagnostic errors Diagnostic discrepancy rates	14 studies; 2 systematic reviews	Inpatient and outpatient (radiology and pathology)	There is a lack of evidence to show that traditional random peer review and feedback mechanisms improve diagnostic quality over time or prevent diagnostic errors from reaching the patient. Nonrandom peer review appears to be more effective at identifying diagnostic errors than random peer review. When nonrandom peer review is conducted prospectively, there is an opportunity to identify and remediate the diagnostic error before it reaches the patient.
Diagnostic Error: Result Notification Systems (RNS)	Acknowledgement of result receipt Timeliness of result receipt Timeliness of action taken on test result Documented action	15 studies; 2 systematic reviews	Inpatient and outpatient (primary care and specialty care), ED	Results varied by type of test result, setting, synchronous vs. asynchronous communication, and manual vs. automated alerting mechanisms. For both critical and noncritical clinically significant test results of radiologic studies, lab studies and tests pending at discharge, the use of RNS showed mixed results in the timeliness of receipt and action on the test results.
Diagnostic Error: Education and Training	Diagnostic accuracy Diagnostic errors Cognitive biases	19 studies; 2 systematic reviews; 1 meta-analysis	Classroom, online training, inpatient and outpatient (primary care, specialty care)	Although there are a limited number of studies, training on metacognitive skills may improve diagnostic accuracy, particularly as clinical experience increases. Online training, either didactic or via simulation, can be successfully used as a mode of delivery for educational interventions targeting clinical reasoning and diagnostic safety.
Failure To Rescue: Patient Monitoring Systems (PMS)	Number of rescue events (rapid response team [RRT] or code blue calls) Time to collect vital signs Mortality Hospital length of stay (LOS) Intensive care unit (ICU) transfer ICU length of stay (LOS)	8 studies; 3 systematic reviews	Hospital (medical/surgical units)	There was moderate evidence for a reduction in rescue events following implementation of a PMS with continuous monitoring (CM), but studies were inconsistent. PMS with CM showed no significant effect on mortality, while PMS with intermittent vital sign input had a moderate and inconsistent association with mortality rates. There was moderate evidence for improvement in hospital LOS but low evidence for improvement in other outcome measures (ICU LOS, ICU transfers).
Failure To Rescue: Rapid Response Teams	Hospital mortality Cardiac arrest rate ICU transfer rate	4 studies; 3 systematic reviews; 3 meta-analyses	Acute care hospitals	There is inconclusive evidence as to whether RRT implementation is associated with decreased overall hospital mortality or ICU transfer rates. There is moderate evidence that decreased non-ICU cardiac arrest rates are associated with implementation of RRT. Recognition of the benefits of RRT implementation often takes a long time.
Sepsis: Screening Tools	Time to Surviving Sepsis Campaign (SSC) bundle administration Mortality ICU transfer ICU LOS	26 studies; 2 systematic reviews	Hospital, pre-hospital (emergency medical services [EMS]), and nursing homes	Performance of screening tools varied widely, especially in the pre-hospital setting. More research is needed to determine the optimal variables and thresholds for a screening tool. There was moderate evidence of process measure improvement in the hospital setting, including time to antibiotics. Pre-hospital evidence was sparse but showed improvement as well. Evidence for outcome measures (e.g., mortality, ICU LOS, ICU transfer) was sparse but showed a trend toward improvement, although the improvement was not always significant.
Sepsis: Patient Monitoring Systems	Time to SSC bundle administration Mortality Hospital LOS ICU transfer ICU LOS	15 studies; 4 systematic reviews	Hospital (ICU, ED, general unit, telemetry, multiple units)	There was moderate evidence of process measure improvement across multiple types of hospital units, and evidence was most consistent outside of the ICU. Evidence for outcome measures (e.g., mortality, ICU LOS, ICU transfer) was mixed, but over half of the studies showed a significant improvement, and several showed an absolute improvement that did not reach statistical significance.
Clostridium difficile: Antimicrobial Stewardship	Clostridium difficile infection (CDI) rates Amount of prescribed high-risk antimicrobials	17 studies; 3 meta-analyses; 2 systematic reviews	Inpatient (hospitals, and long-term care facilities [LTCFs]/nursing homes)	The majority of studies showed reductions in CDI following a period of antimicrobial stewardship (both statistically significant and statistically non-significant reductions). In the reviewed studies, significant reductions in CDI were associated with higher baseline CDI rates/outbreaks, antimicrobial stewardship programs (ASPs) developed specifically to reduce CDI (as opposed to ASPs focused on other clinical and microbiological outcomes), and ASPs that included restrictions of high-risk antimicrobials and/or a pre-authorization component. ASPs require staffing, technological resources, and provider buy-in.
Clostridium difficile: Testing	Sensitivity Specificity Predictive value (e.g., c-statistic) Time it takes to get results	25 studies; 7 systematic reviews	Inpatient (hospitals), ED	Screening and isolating asymptomatic carriers can prevent CDI transmission but are resource intensive. Nucleic acid amplification tests (NAATs) of unformed stool have relatively accurate sensitivity and specificity. Concerns with NAATs include that they detect toxigenic C. difficile genes, not the actual damaging toxins, and may capture colonized patients in addition to those infected with C. difficile. Certain multi-step test algorithms that include a test for C. difficile and for CDI toxins perform as well or better than NAATs but take longer. Tools that identify patient risk for CDI could be useful in preventing CDI.
Clostridium difficile: Surveillance	Prospective outbreak and cluster identification Surveillance case definition accuracy Speed of case identification	16 studies; 2 systematic reviews	Inpatient (hospitals and LTCFs), outpatient, and regional	Studies showed that automated surveillance systems are generally accurate, and save time and resources compared to manual case review. Automated laboratory alerts have been shown to help expedite contact precautions for CDI patients. Classifying CDI cases using standard case definitions is important, although some researchers have found that the current definitions over-represent the number of nosocomial cases. Case studies show that genotyping provides detail about differences in C. difficile virulence and has helped to identify transmission pathways and outbreaks.
Clostridium difficile: Hand Hygiene	CDI rates Staff compliance Hand contamination	11 studies; 1 systematic review	Inpatient (hospitals and LTCFs)	In vitro evidence supports the use of gloves and hand washing with soap and water for C. difficile prevention; multiple experimental studies show alcohol-based hand rubs are not effective for eliminating C. difficile spores. Studies that measured hand hygiene and CDI patient outcomes were quasi-experimental, and showed large and mostly statistically insignificant decreases in CDI following implementation of hand hygiene programs that targeted multiple healthcare-associated infections (HAIs) (statistical significance was impacted by small sample sizes).
Clostridium difficile: Environmental Cleaning and Decontamination	CDI rates Performance of cleaning staff Environmental contamination Time to clean and decontaminate	18 studies; 3 systematic reviews	Inpatient (hospitals and LTCFs)	Studies supported daily and/or discharge cleaning with chlorine-based agents for CDI-occupied rooms. In many studies, the addition of hydrogen peroxide decontamination or ultraviolet light decontamination to standard cleaning was associated with significant reductions in facility-level CDI rates.
Clostridium difficile: Multicomponent Interventions	CDI rates Staff compliance	8 studies; 3 systematic reviews	Inpatient (hospitals, LTCFs)	Multicomponent interventions to prevent CDI were associated with significant decreases in CDI rates. The most common component was environmental cleaning, followed by hand hygiene, and patient isolation practice,; antimicrobial stewardship, and contact precaution,; and CDI testing, and surveillance. There was no single CDI prevention toolkit used across studies.
Controlling Multidrug-Resistant Organisms (MDROs) and Preventing MDRO-Related Infection: Hand Hygiene for Controlling MDROs	MDRO acquisition (methicillin-resistant Staphylococcus aureus [MRSA], carbapenem-resistant Enterobacteriaceae [CRE], vancomycin-resistant Enterobacteriaceae [VRE], multidrug-resistant gram-negative bacteria [MDR-GNB]) MDRO infection (including HAIs) Environmental contamination Hand hygiene compliance	13 studies; 3 systematic reviews; 1 meta-analysis	Hospitals (including intensive care transplant, dialysis, and general care units), LTCFs	Hand hygiene is indispensable for infection control, and hand hygiene compliance reinforces compliance with other practices. The World Health Organization’s “My Five Moments for Hand Hygiene” was frequently recommended for improving hand hygiene compliance, but there are many effective campaign materials to choose from. Existing campaigns can be made even more effective by having staff personalize the implementation through creating educational/promotional materials and supporting each other in observing hand hygiene. The biggest barriers to hand hygiene compliance are: (1) awareness that an opportunity for hand hygiene is occurring; and (2) remembering to complete hand hygiene protocol consistently at every opportunity. Education can help with the first, and direct observation with immediate feedback helps improve the second.
Controlling Multidrug-Resistant Organisms (MDROs) and Preventing MDRO-Related Infection: Surveillance for Controlling MDROs	MDRO acquisition (MRSA, CRE, VRE, MDR-GNB) MDRO infection (including HAIs) Results reporting completeness and accuracy Compliance with other patient safety practices (PSPs) (such as contact precautions)	20 studies; 2 systematic reviews; 1 meta-analysis	Hospitals (including intensive care, neonatal intensive care, hematology/ oncology, and general care units)	Targeted active surveillance performs as well as universal active surveillance for many MDROs and uses fewer resources. In places where universal active surveillance is already in place, screening for other MRDOs using the same sample may be cost-effective, due to shared risk factors. Some consensus exists for screening high-risk patients (those with a history of MDROs or risk factors associated with MDRO colonization/ infection) on admission, but any screening approach will require compliance with infection prevention protocols when a positive culture result is found. Surveillance may improve compliance with other PSPs when it is part of a multi-component intervention, but more research is needed on the mechanisms and circumstances of this association, as it can be confounded by the co-implementation of other, bundled practices.
Controlling Multidrug-Resistant Organisms (MDROs) and Preventing MDRO-Related Infection: Minimizing Exposure to Invasive Devices and Reducing Device-Associated Risks	Incidence of infections (e.g., bloodstream infections, pneumonia) Measures of antimicrobial resistance (e.g., minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC)	11 studies; 5 systematic reviews; 1 meta-analysis	Inpatient settings (e.g., hospitals, LTCFs), outpatient settings (e.g., dialysis), and patient homes	Using devices minimally and appropriately, and practicing hygiene and infection control precautions when inserting them are basic steps that can be taken to reduce device-associated infections. Antimicrobial resistance has not been eliminated as a concern when using antibiotics in antibiotic lock therapy (ABLs), impregnated catheters, or prophylactic treatment to prevent infections. Ongoing implementation education, monitoring, feedback for medical staff, patients, and caregivers are recommended for improving adherence to recommended PSPs.
Controlling Multidrug-Resistant Organisms (MDROs) and Preventing MDRO-Related Infection: Chlorhexidine Bathing for Controlling MDROs	MDRO acquisition (MRSA, CRE, VRE, MDR-GNB) MDRO infection (including HAIs) Measures of chlorhexidine resistance (e.g., MIC and MBC) Adverse reactions to chlorhexidine application	38 studies; 4 systematic reviews	General healthcare settings, community settings, hospital settings (including intensive care, pediatric intensive care, transplant, and general care), LTCFs, and laboratory studies (for resistance)	Chlorhexidine bathing is effective at reducing colonization, particularly by multi-drug resistant (MDR) gram-positive bacteria; evidence is mixed about its effectiveness in reducing MDR-related infections (beyond skin preparation for central line insertion). As an intervention, chlorhexidine is low-cost to implement (especially if routine bathing is already in place) and generally well-received by staff, but compliance with bathing can wane over time. Chlorhexidine resistance is a potential problem, and while there are no clinical impacts described in the literature to date, this should continue to be monitored.
Controlling Multidrug-Resistant Organisms (MDROs) and Preventing MDRO-Related Infection: Communication of Patient’s MDRO Status	Percent compliance with guidelines Adverse patient outcomes associated with lack of or incorrect communication	12 studies; 1 systematic review	General healthcare settings, especially transfer of patients from one setting to another	Communication failures have been linked to poor patient outcomes, especially in the field of organ transplantation. Multimodal and redundant communication policies can improve communication compliance in settings with complex communication (e.g., organ donation) or with multiple care providers (i.e., transfers). Modes of communication can include checklists, brightly colored leaflets attached to medical records, and electronic or automated communication. Revisiting policies to ensure they are meeting a facilities’ needs, performing ongoing monitoring and feedback of policy compliance, and involving staff from multiple disciplines in policymaking are all important for improving patient status communication.
Controlling Multidrug-Resistant Organisms (MDROs) and Preventing MDRO-Related Infection: Environmental Cleaning and Disinfection	Transmission rates Proportion of deactivated microbial cultures Reduction in environmental contamination markers (e.g., fluorescent gel, UV detectable powder, or adenosine triphosphate (ATP) markers) Measures of antimicrobial resistance (e.g., MIC and MBC)	54 studies; 2 systematic reviews; 2 meta-analyses	General healthcare settings (e.g., hospitals and LTCFs)	There is a need for more studies in clinical settings that examine the different cleaning and disinfecting agents individually, as opposed to as part of a multicomponent intervention. No-touch disinfection technologies are promising additions to disinfection practices, but must be further studied to determine the most efficacious and cost-effective options. Environmental screening is a useful tool for auditing and monitoring ongoing cleaning practices, and for identifying highly contaminated surfaces for targeted cleaning during outbreak scenarios.
Carbapenem-Resistant Enterobacteriaceae: Contact Precautions To Prevent CRE	CRE carrier prevalence CRE incidence Compliance rate	18 studies; 3 systematic reviews	General healthcare settings (e.g., hospitals and LTCFs)	Contact precautions have been shown to reduce transmission of CRE as part of infection control bundles in a variety of healthcare settings, including long-term care facilities and acute care facilities. Active surveillance is recommended in outbreak scenarios, in highly endemic regions, and in healthcare facilities or units with ongoing transmission. Additional research is needed to determine whether there is an appropriate time to discontinue contact precautions, based on duration of CRE carriage.
Anticoagulants: Anticoagulation Management Service	Time to therapeutic range (TTR) Bleeding events Thrombotic events	5 studies; 6 systematic reviews	Ambulatory home-bound	There are a range of models; most are pharmacist-led but some are led by nurse practitioners, physician assistants, or nurses. Some models examined different modes of providing these services (e.g., telephone). Overall quality of studies (individual and within reviews) is moderate to high given number randomized controlled trials (RCTs), non-RCTs with comparison groups, or pre/post designs. Evidence of the effect of anticoagulant management services is moderately positive on time to therapeutic range (TTR), low and mixed on bleeding events and thromboembolic events.
Anticoagulants: Nomograms/Protocols for Novel Oral Anticoagulants	Protocol adherence Mean time to activated partial thromboplastin time (aPTT) stabilization	4 studies	Group practice, academic medical center, community hospital	Standard and simplified dosing nomograms were successfully implemented by nurses. Strength of evidence is low; no control groups were used in the four studies and all reported small sample sizes. Cases of major bleeding were reported in some of the intervention protocols.
Anticoagulants: Anticoagulant Care in Transitions Between Hospital and Home	Hemorrhagic events Readmission ED LOS Death Medication adherence TTR	5 studies	Discharge from ED	The strength of evidence for anticoagulant care transition from hospital to home was low to moderate, as small sample sizes and single sites were cited. Most studies reported no statistically significant differences for outcomes (i.e., recurrent venous thromboembolism (VTE), death, and readmission).
Diabetic Agents: Diabetes Protocol for Reducing Hypoglycemia	Incidence of hypoglycemia Frequency of hypoglycemia LOS Blood glucose levels	11 studies	ED, ICU	Although glycemic outcomes usually improved, a statistically significant difference was rare. Implementation of study protocols were usually implemented by nurses.
Diabetic Agents: Teach-Back	HbA1c levels Diabetes knowledge score Health literacy scores	4 studies	Federally Qualified Health Center (FQHC), academic medical center, outpatient clinic	Each study used a different model to facilitate teach-back. Interventions need to address various levels of health literacy to be successful. Health literacy scores improved, but only for a limited time.
Reducing Adverse Drug Events in Older Adults: Deprescribing	Drug burden index scores Adverse drug reactions Prescribing Good health status rating Healthcare utilization	13 studies	Residential care facility, community pharmacy, day center for senior citizens, skilled nursing facility, LTCF	Intervention models were specifically designed to reduce harm in older adults. Intervention models were primarily led by a pharmacist or physician. Interventions led to a decrease in the number of medications prescribed and a decrease in medication-related costs.
Reducing Adverse Drug Events in Older Adults: Using the Screening Tool of Older Persons’ Potentially inappropriate Prescriptions (STOPP) Criteria	Prescribing Potentially inappropriate medications Potentially inappropriate prescriptions Adverse drug reaction incidence	14 studies	Primary healthcare center, geriatric psychiatry admission unit, acute care admission, LTC, nursing homes, geriatrics outpatient clinic	STOPP criteria specifically target older adults to reduce avoidable adverse drug events. STOPP criteria are particularly effective when combined with another tool. Implementation of the STOPP criteria was either led by a pharmacist or physician. Studies showed improved prescribing appropriateness but no statistically significant differences in admission or in-patient death rates.
Opioids: Opioid Stewardship	Opioid prescribing Opioid dosage Potential misuse Overdose Recommended risk mitigation practices (urine screen)	14 studies; 1 systematic review	Primary care, health system ED, hospital, specialty	Majority of studies examined multicomponent interventions consisting of clinical interventions and implementation strategies. Six studies had control groups (2 were randomized). Post-intervention periods ranged from months to years. The strength of evidence for opioid stewardship producing significant reduction in opioid dosages was moderate.
Opioids: Medication-Assisted Treatment (MAT) Initiation	Opioid dependence Illicit drug use Treatment retention rates	25 studies; 1 systematic review	ED, community practice, clinic for the homeless, primary care clinic, outpatient substance use disorder treatment center, FQHC	MAT can be initiated and provided safely in a variety of healthcare settings. Initiation of MAT in the ED, primary care setting, or outpatient clinics may result in faster access to care and longer retention in or adherence to treatment. The majority of studies were focused on one component of MAT, the initiation of medications, in a few specific settings.
Patient Identification Errors: Patient Identification Errors in the Operating Room	Compliance audit Incidence of wrong-site surgery	8 studies; 1 systematic review	Hospital operating room and theaters	Drawing meaningful statistical comparisons is difficult because wrong-site surgeries are rare.
Infusion Pumps/Medication Error: Structured Process Change and Workflow Redesign	Medication administration errors Procedural errors Process outcomes	6 studies	Hospitals	Multiple studies identified standardization and streamlining of processes and workflows as main facilitators of optimal infusion pump use. Integration and alignment of technology and workflows, and engaging multiple members of the care team were also found to be important facilitators.
Infusion Pumps/Medication Error: Staff Education and Training	Adverse drug events Nurse adherence	5 studies	Hospitals	Five studies identified the type and content of education provided as facilitators. One study noted that time and energy constraints on nurse educators can be barriers to implementing large hospital-wide education programs. More research is needed to understand why clinicians commonly bypass smart pump safety technology and what type of training should be implemented to limit medical errors.
Alarm Fatigue: Safety Culture	Number of alarms Noise level	10 studies	Hospitals; ICU; progressive care unit; neonatal intensive care unit; or telemetry, step-down, transplant cardiology, surgical, or surgical orthopedic units	Current literature on this PSP is primarily quality improvement initiatives and case studies; higher quality studies could help to better understand the impact of implementing elements of safety culture to address alarm fatigue.
Alarm Fatigue: Risk Assessment	Number of alarms Number of false alarms	8 studies	Hospitals; ICU; progressive care unit; neonatal intensive care unit; step-down, transplant cardiology, surgical, or surgical orthopedic units	Studies reviewed focused on one hospital or specific unit and thus have limited generalizability. The decision to engage a team and conduct a risk assessment was often in response to a specific adverse patient event or external influence.
Delirium: Screening and Assessment	Recognition Prevention Management of delirium to prevent possible harms	13 studies; 2 systematic reviews; 2 non-systematic reviews	All settings were included	Evidence identified for this review is markedly heterogeneous. The tools most frequently used and evaluated were the Confusion Assessment Method (CAM) and the Confusion Assessment Method-Intensive Care Unit (CAM-ICU). These tools have been tested singly and in comparison with other tools to determine concordance.
Delirium: Staff Education and Training	Recognition Prevention Management of delirium to prevent possible harms	16 studies	All settings were included	Studies reviewed identified a need for much more education and training in the identification of individuals at risk for developing delirium, the contributing factors to delirium in a variety of care settings, and strategies to appropriately manage delirium Consideration should be given to implementing the Acute Care for the Elderly (ACE) Model Education and training utilizing a variety of modalities—including partnering ACE units with non-ACE units, e-learning, combining didactic course work with either simulation or supervised clinical practice with feedback from experts—have shown promise.
Delirium: Nonpharmacological Interventions To Prevent Delirium	Delirium incidence Delirium duration	8 studies; 4 systematic reviews; 1 non-systematic review	ICU	Multicomponent nonpharmacological interventions are effective for reduction of delirium among intensive care patients, although the quality of the evidence is low to moderate. Reproducibility and scalability are hindered by a lack of evidence regarding which components of many are required to achieve the desired effect. In addition, specific details of implementation required for replication and level of adherence to protocols are not often reported.
Care Transitions:Transition of Care Models
BOOST: Better Outcomes for Older Adults through Safe Transitions	Readmission rates	3 studies	Large hospitals	Model contributes to reduction of 30-day re-hospitalization rates, and using the 8 Point assessment tool accurately predicts 90% of readmissions.
CTI: Care Transition Intervention	Readmission rates	7 studies	Acute care hospitals	Model contributes to reductions in hospital readmission at 30, 60, and 180 days. Model contributes to significant reductions in healthcare costs.
TCM: Transitional Care Model	Readmission rates	3 studies; 1 systematic review	Community/outpatient, academic health system	Model reduces rates of readmission and costs for healthcare systems. Measuring changes in key outcome categories is important for benchmarking evidence of the impact of the TCM.
Venous Thromboembolism: Use of aspirin for VTE prophylaxis	Deep vein thrombosis Pulmonary embolism Operative site bleeding and other major bleeding	27 studies; 6 systematic reviews	Hospital, post-surgical care, tertiary care orthopedic referral centers; countries included USA, UK, China, Canada, and Korea	Use of aspirin following major orthopedic surgery was generally found to be of similar effectiveness as other agents. An overwhelming majority of studies concluded that aspirin has a lower bleeding risk rate than other pharmacologic agents which, combined with its lower cost, makes it an appealing option for VTE prophylaxis, particularly in low-risk patients. More prospective RCTs are needed to directly compare the effectiveness of aspirin to other prophylactic methods across patient risk levels.
Cross-Cutting Factors: Patient and Family Engagement (PFE)	Patient and provider perception and attitude to PFE	1 study; 2 systematic reviews	Hospital	Studies revealed a lack of understanding about the effects of PFE on patient safety among healthcare providers, patients, and families. PFE implemented through an educational intervention was linked to positive perceptions and attitudes about PFE among healthcare providers. More studies are needed to measure the direct outcomes of patient and family engagement.
Cross-Cutting Factors: Safety Culture
Leadership WalkRounds	Adverse events Perceptions of safety climate	4 studies; 1 systematic review	Hospital	A previous systematic review reported that the use of WalkRounds was associated with a reduction in adverse events. The single studies reviewed found that WalkRounds were associated with improvements in various aspects of safety climate.
Team Training	Perceptions of safety climate	8 studies; 2 systematic reviews	Hospital, VA facilities, rehabilitation unit in long-term care facility	The second Making Healthcare Safer report found improved care processes after team training was introduced to improve safety culture. The single studies reviewed found that team training interventions were associated with improvements in various aspects of safety climate. In the second Making Healthcare Safer report, team training was associated with reduced errors and safety events.
Comprehensive unit-based safety program (CUSP)	Surgical site infection rates Central line-associated blood stream infection (CLASBI) rates Perceptions of safety culture	6 studies; 1 systematic review	Hospital	CUSP interventions were associated with improved perceptions of various aspects of safety culture. The introduction of CUSP was associated with patient outcomes such as decreased CLASBI rates and decreased surgical site infections.
Multiple Interventions	Perceptions of safety culture	1 study	Hospital	The bundle interventions were associated with improved perceptions of various aspects of safety culture, although some scores decreased in post-intervention period.
Cross-Cutting Factors: Clinical Decision Support (CDS)	NA	26 studies; 1 systematic review	All settings included in searches	CDS is widely believed to have the potential to positively impact patient safety; this belief has face validity. The most consistent impact of CDS in the literature reviewed was on improving medication safety. While some results are promising, more evidence is needed to clearly establish the significant role CDS could play in increasing patient safety.
Cross-Cutting Factors: Cultural Competency	Preventable hospital readmissions Medication adherence LOS Advance care planning and informed consent	7 studies; 4 systematic reviews	Inpatient, outpatient, and home health	Most of the small group of reviewed studies found language services were associated with improved patient safety. There is a need for studies that explore associations between a range of cultural competency interventions and patient safety outcomes. Interventions that are framed around cultural competency and aim to improve patient health and indicators of health have mostly positive outcomes.
Cross-Cutting Factors: Monitoring, Audit, and Feedback	NA	28 studies; 3 systematic reviews; 1 non-systematic review	All settings included in searches	Audit and feedback is a somewhat common strategy for improving compliance with patient safety processes. Audit and feedback appear to be most effective when both written and verbal feedback are used. Studies show more significant improvements when performance was lower at baseline. Research on audit and feedback predominantly focuses on process improvement, and more research is needed to measure the impact of audit and feedback on patient outcomes.
Cross-Cutting Factors: Teamwork and Team Training
Crew Resource Management (CRM)	Adverse outcome index Standards of care Participant reactions Knowledge about teamwork	6 studies; 1 systematic review; 1 meta-analysis	Hospital	Participants had greater knowledge about teamwork and felt more confident in their ability to use teamwork skills following CRM training. Increases in participants’ confidence, increased use of teamwork skills, and improved clinical processes were sustained for 6 to 9 months following training. None of the individual studies collected clinical outcomes. Systematic review and meta-analysis reported improved patient outcomes. Participants had positive reactions to CRM training.
Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS®)	Infection rate LOS Participant reactions Teamwork skills	6 studies; 1 systematic review; 1 meta-analysis	Hospital, ED, psychiatric hospital	Participants had positive reactions to TeamSTEPPS training. Participants had greater knowledge about teamwork and demonstrated more teamwork skills immediately following TeamSTEPPS training. Some improvements in teamwork skills were sustained for up to 12 months following the training, and more accurate decisions about care were noted for up to 3 months following training. Positive perceptions of safety culture, decreased infection rates, and decreased LOS (nonsignificant) were noted as outcomes associated with the training.
Veterans Health Administration Medical Team Training	Surgical morbidity rates Attitudes toward safety Delays in care	2 studies; 1 systematic review; 1 meta-analysis.	VA facilities	Positive attitudes toward teamwork were found 12 to 17 months after the training. Fewer case delays were reported for up to 24 months following training. Decreases in surgical morbidity were noted in one study for facilities that had participated in the training.
Team Simulation	Adverse outcome score Use of teamwork skills Confidence in emergencies Decision-making Workload management	6 studies; 3 systematic reviews; 1 meta-analysis	Tertiary care medical center, teaching hospital, VA facility	Participants were more confident in their ability to handle emergencies following simulation training. Participants demonstrated greater teamwork skills following simulation training, with some longer term sustainment reported. Improved neonatal outcomes and a reduction of postpartum hemorrhage cases were associated with the simulation interventions.
Brief/Debrief	Survival to discharge Survival with favorable neurological outcomes Debrief quality	3 studies	ED, surgical department, ICU	The use of briefings was associated with improved clinical processes. The use of a coach to facilitate briefings was associated with increased quality of briefings, which was sustained on the job. A debriefing intervention was associated with increased patient outcomes (i.e., survival to discharge rate, survival with favorable neurological outcomes).
Handoff	Adverse events Checklist compliance Information sharing	3 studies	Hospital (surgical units, neurointerventional suite)	Participants reported favorable reactions to the checklist introduced. Compliance with checklist use and completion of checklist items increased in the post-intervention period and were sustained for up to 18 months. Use of the checklist was associated with a decrease in adverse events.
Cross-Cutting Factors: Staff Education and Training
Simulation-Based Medical Education for Residents and Fellows	Complication rates Frequency and time of successful procedures	5 studies; 3 systematic reviews; 1 meta-analysis	Teaching hospital, tertiary teaching hospital	Simulation-based medical education curriculums were associated with decreased complication rates, fewer errors, reduced CLASBI rates, improved pain management of patients, and a lower proportion of adverse events across studies. Cost savings were associated with reductions in central-line infections, overnight hospital days, or additional hospital days. Improved procedural skills were reported for participants, such as increased rates of successful first attempts to intubate patients, fewer needle passes for central venous catheter insertion, and increased compliance with recommended guidelines and protocols.
Simulation as Part of Continuing Education of Nurses	Infection rates Medication administration errors Compliance with guidelines and protocols	2 studies	Children’s hospital, teaching hospital	A decrease in medication administration errors was reported after nurses completed simulation exercises and participated in debriefs on their performance. Simulation training provided to nurses on sterilization techniques was associated with decreased infection rates. Knowledge of and adherence to recommended protocols increased following simulation training.

ES.5 Discussion

This report covers 47 PSPs chosen for the high-impact harms they address and interest in the status of their use. The harms include diagnostic errors, failure to rescue, sepsis, infections due to multi-drug resistant organisms, adverse drug events, and nursing-sensitive conditions. While going through the process of selecting PSPs to address specific harm areas, it became evident that several commonly recommended practices should also be reviewed. These cross-cutting practices are: improving safety culture, teamwork and team training, clinical decision support, patient and family engagement, cultural competency, staff education and training, and monitoring, audit, and feedback. All of the harm-specific PSPs and cross-cutting PSPs included in this report underwent focused systematic reviews to establish the current evidence base for their use.

The most significant harms patients face continue to be found in higher acuity settings, such as the ED and ICU, and the research is biased toward those settings. One “setting” that poses a unique threat to patients is the transition between one setting and another: the hospital to the outpatient setting, in particular. As we move out from the silos required in setting-specific research, the research needs to address these gaps.

Regardless of setting, several themes emerged from the report:

More than one PSP can be used to reduce a given harm. The PSPs presented in the report are those that the project TEP and AG felt were ready for a fresh review of the literature or that were relatively new and needed to have an evidence base established. The PSPs in the report are not intended to be an inclusive list.
Selecting a particular PSP should be based on the root cause of the harm. If a facility is experiencing an increase in sepsis mortality, the root cause may be a lack of recognition of patients with sepsis arriving to the ED. In another facility, it may be due to lack of monitoring of patients who are experiencing deterioration on a medical-surgical unit.
When using a specific PSP, consideration must be given to potential new harms that can be introduced. For example, strategies to improve anticoagulation-related events must be balanced with strategies used to reduce venous thromboembolism.
PSPs are not implemented in isolation and are often part of a broader safety strategy. The strategy often relies on a strong safety culture, teamwork, communication, and involvement of the patient and family. These cross-cutting practices are the foundation for success.
The context in which a PSP is implemented determines success. Understanding the impact of context through rigorous, large-scale research studies is difficult. It is extremely difficult, and sometimes may be impossible, to design a study that takes into consideration all potential contextual factors, such as staffing, other PSPs in place, safety culture, and leadership engagement, and to control for those factors across enough sites to make the findings generalizable.

ES.6 Limitations

There are several limitations to conducting such a broad review of the literature as found in this report.

As our understanding of patient safety expands, there is an increasing amount of published research, with most showing positive effects of the intervention under question (i.e., publication bias). With the paucity of recent randomized controlled trials (RCTs) in the literature and the reliance on pre-post studies and observational studies, it is difficult to assess the impact of the biases introduced by study design on the findings.

The low number of RCTs in the patient safety literature is also a limitation in conducting focused systematic reviews such as those found in this report. Many of the PSPs under examination have been implemented in some form. Staff are aware of the implementation of the PSPs, so being blinded to the intervention is not possible. Since PSPs are typically implemented across an entire facility rather than a single unit, finding a control group for comparison can be difficult.

Lastly, some of the PSPs addressed were introduced as part of multicomponent interventions (e.g., strategies to reduce C. difficile infections). When a specific PSP is part of a multicomponent intervention, it is difficult to ascertain which of the components is the driver for success.

ES.7 Conclusions

This report presents reviews of 47 different PSPs covering a wide variety of harms across multiple settings. The amount of research in patient safety has exponentially grown since the second report was published. PSPs that are more well-established are now being investigated in light of emerging harms, such as PSPs related to infection prevention to address multi-drug resistant organisms. Similarly, emerging PSPs are being investigated for use to address well-established harms, such as the use of clinical decision support to reduce diagnostic errors.

It is clear that many factors impact the success of any PSP on reducing harm. Patient safety culture, teamwork and communication, person and family engagement, providing culturally competent care, reinforcing good practice with education and training, and learning from data are all necessary to ensure success.

ES.8 Future Research Needs

It is clear from the reviews of these PSPs that the importance of context for implementation cannot be overstated. Context plays a large role in the successful uptake and use of a PSP. Setting, safety culture, staffing, and other organizational factors often contribute to harm reduction as much as a PSP itself. More implementation research needs to be conducted across all of the PSPs to understand and work within real-world constraints, rather than conducting studies that may be rigorous but are stripped of that context. We often know what to do, and in these cases, the challenge now is to implement PSPs into a specific facility or setting and have them succeed.

References

Shojania KG, Duncan BW, McDonald KM, Wachter RM, eds. Making Health Care Safer: A Critical Analysis of Patient Safety Practices. Evidence Report/Technology Assessment No. 43 (Prepared by the University of California at San Francisco–Stanford Evidence-based Practice Center under Contract No.290-97-0013). AHRQ Publication No.01-E058. Rockville, MD: Agency for Healthcare Research and Quality. July 2001. https://archive.ahrq.gov/clinic/ptsafety/pdf/ptsafety.pdf.
Shekelle PG, Wachter RM, Pronovost PJ, Schoelles K, McDonald KM, Dy SM, Shojania K, Reston J, Berger Z, Johnsen B, Larkin JW, Lucas S, Martinez K, Motala A, Newberry SJ, Noble M, Pfoh E, Ranji SR, Rennke S, Schmidt E, Shanman R, Sullivan N, Sun F, Tipton K, Treadwell JR, Tsou A, Vaiana ME, Weaver SJ, Wilson R, Winters BD. Making Health Care Safer II: An Updated Critical Analysis of the Evidence for Patient Safety Practices. Evidence Report/Technology Assessment No. 211. (Prepared by the Southern California-RAND Evidence-based Practice Center under Contract No. 290-2007-10062-I.) AHRQ Publication No.13-E001-EF. Rockville, MD: Agency for Healthcare Research and Quality. March 2013. https://www.ahrq.gov/research/findings/evidence-based-reports/ptsafetyuptp.html#Report.
National Quality Forum (NQF). Safe Practices for Better Healthcare–2009 Update. Washington, DC; 2009. https://www.qualityforum.org/Publications/2009/03/Safe_Practices_for_Better_Healthcare %E2%80%932009_Update.aspx.
Agency for Healthcare Research and Quality. Patient Safety Network. https://psnet.ahrq.gov/. Accessed January 23, 2020.
Agency for Healthcare Research and Quality. The National Quality Strategy: Fact Sheet. https://www.ahrq.gov/workingforquality/about/nqs-fact-sheets/fact-sheet.html. Accessed January 23, 2020.
The Joint Comission. National Patient Safety Goals. https://www.jointcommission.org/standards/national-patient-safety-goals/. Accessed January 23, 2020.
National Quality Forum (NQF). Patient Safety 2015 Final Report. Washington, DC; 2015.
Centers for Medicare & Medicaid Services. The Hospital Value-Based Purchasing (VBP) Program https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment- Instruments/Value-Based-Programs/HVBP/Hospital-Value-Based-Purchasing. Accessed January 23, 2020.
ECRI Institute. Executive Brief Top 10 Patient Safety Concerns for Healthcare Organizations 2017. Plymouth Meeting, PA; 2017. https://www.ecri.org/EmailResources/PSRQ/Top10/2017_PSTop10_ExecutiveBrief.pdf.
ECRI Institute. Executive Brief Top 10 Patient Safety Concerns for Healthcare Organizations 2016. Plymouth Meeting, PA; 2016. https://www.ecri.org/EmailResources/PSRQ/Top10/2016_Top10_ExecutiveBrief_final.pdf.
Vaidya A, Zimmerman, Bean M. 10 Top Patient Safety Issues for 2018. https://www.beckershospitalreview.com/10-top-patient-safety-issues-for-2018.html. Accessed January 23, 2020.
Centers for Medicare & Medicaid Services. Partnership for Patients. https://innovation.cms.gov/initiatives/partnership-for-patients/. Accessed January 23, 2020.
Whitlock EP, Lin JS, Chou R, et al. Using existing systematic reviews in complex systematic reviews. Ann Intern Med. 2008;148(10):776-82. doi: 10.7326/0003-4819-148-10-200805200-00010.
Shea B, Reeves B, Wells G, et al. AMSTAR 2: A crticial appraisal tool for systematic review that include randomised or non-randomised studies of healthcare interventions or both. BMJ. 2017;358:j4008. doi: 10.1136/bmj.j4008.
Blackman C, Chartrand J, Dingwall O, et al. Effectiveness of person-and family-centered care transition interventions: a systematic review protocol. Bio Med Central. 2017 6:158. doi: 10.1186/s13643-017-0554-z
Viswanathan M, Ansari MT, Berkman ND, et al. Assessing the Risk of Bias of Individual Studies in Systematic Reviews of Health Care Interventions. Methods Guide for Effectiveness And Comparative Effectiveness Reviews. AHRQ Publication No. 12-EHC047-EF. Rockville, MD: Agency for Healthcare Research and Quality. March 2012. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/methods-guidance-bias-individual-studies_methods.pdf
Berkman ND, Lohr KN, Ansari M, et al. Grading the Strength of a Body of Evidence When Assessing Health Care Interventions for the Effective Health Care Program of the Agency for Healthcare Research and Quality: An Update. Methods Guide for Effectiveness and Comparative Effectiveness Reviews. (Prepared by the RTI-UNC Evidence-based Practice Center under Contract No. 290-2007-10056-I). AHRQ Publication No. 13(14)-EHC130-EF. Rockville, MD: Agency for Healthcare Research and Quality. November 2013.
Owens DK, Lohr KN, Atkins D, et al. AHRQ series paper 5: Grading the strength of a body of evidence when comparing medical interventions--Agency for Healthcare Research and Quality and the Effective Health-Care Program. J Clin Epidemiol. 2010;63(5):513-23. doi: 10.1016/j.jclinepi.2009.03.009.
Atkins D, Eccles M, Flottorp S, et al. Systems for grading the quality of evidence and the strength of recommendations: Critical appraisal of existing approaches the grade working group. BMC Health Serv Res. 2004;4(1):38. doi: 10.1186/1472-6963-4-38.

SHARE: