Evidence Table

Quality Methods

SourceQuality or Safety Issue Related to Clinical PracticeDesign TypeStudy Outcome Measure(s)Study Setting & Study PopulationStudy InterventionKey Finding(s)
Failure Modes and Effects Analysis (FMEA)
Adachi 200578Medication safetyQuality improvementMedication errors, targeting wrong dose errors (Level 4)422-bed hospital in CaliforniaFMEA used to develop strategies
– Standard order sets were revised, items from the formulary were removed, and the use of unapproved abbreviations was eliminated.
– Used IV pumps with enhanced safety features.
1 year after medication strategies were implemented, medication errors associated with IV infusion were reduced slightly (from 59 to 46), and error related to IV pumps decreased from 41% of dosing errors to 22%. Errors related to wrong drug concentration were completely eliminated.
Apkon 2004115Medication safetyQuality improvementInfusion drug errors (Level 4)11-bed pediatric intensive care unit (ICU) in a children’s hospitalNoneStandardization of the infusion delivery process, with the combined effect of prolonging infusion hang times from 24 to 72 hours, shifting preparation to the pharmacy, and purchasing premanufactured solutions resulted in 1,500 fewer infusions prepared by nurses per year; process changes preferred by nurses and patients.
Burgmeier 200294Blood transfusionQuality improvementErrors associated with blood products administered to patients (Level 4)1 hospital in OhioFollowing the FMEA, implemented the following changes: a standardized form listing choices for blood products and documenting medical necessity, form is faxed to the blood bank; used a blood-barrier system; required staff training; and changes in policies and procedures.Following the new process changes for blood transfusions, no outcome errors were reported within the first 3 months.
New process continued to be assessed, finding more failures to be addressed, and data are aggregated and reported monthly.
Flowcharting before beginning the FMEA process itself was important.
FMEA process was time consuming, tedious, and difficult.
Day 2006124Dialysis treatmentQuality improvementRisks for error in the process of administering dialysis (Level 4)1 hospital in UtahNoneRisk factors included inconsistent nephrology consult/dialysis communication process; dialysis technicians performing beyond their scope of work; scheduling treatments for chronic dialysis patients without a formal consult/order; nurses inconsistently involved in dialysis process; nurses not reviewing dialysis orders or treatment plan before treatment; and lack of a formal handoff report before treatment.
Esmail 2005107Medication safetyQuality improvementSystematic analysis for improvement in the ordering and administration of potassium chloride and potassium phosphate using HFMEA (Level 4)4 adult ICUs in 3 hospitals in CanadaImplemented standardized protocol for potassium chloride and potassium phosphate.Using the HFMEA, recommendations were made for the hospital and ICUs, including who, where, and how the drugs should be mixed, and identifying and developing standard labels for look-alike and sound- alike products.
HFMEA helped prioritize the critical steps of a complex medication process (from ordering to administration), making it more objective.
While the process took time to conduct, it was instrumental in discovering that the vials of intravenous potassium needed to be stored and packaged differently.
Gering 2005123Patient transferPretest and post-test, quality improvementAdverse events (Level 3)2 VA medical centersA series of strategies to merge patients into one facilityNurses were critical in the actual move of patients from one hospital to the next.
After integration, there were no disruptions in patient care, operating room (OR) cancellations decreased, there were no MRSA infections, and clinic wait times decreased.
Kim 2006116Medication safety, CPOEPretest, post- test studyMedication order errors (Level 3)Pediatric oncology patients in 1 academic medical center in MarylandImplementation of a CPOE systemAfter CPOE implementation, there was a decrease in improper dosing, incorrect dosing calculations, missing cumulative dose calculations, and incomplete nursing checklists.
There was no difference in the likelihood of improper dosing on treatment plans, and a higher likelihood of not matching medication orders to treatment plans.
Papastrat 2003111Medication safetyChanging practice projectError detection associated with medication administration (Level 4)First-semester baccalaureate nursing students at 1 university in PennsylvaniaNew teaching methodProblem-based learning enabled students to use findings from topic-specific research to develop solutions for clinical problems.
Students applied knowledge to clinical settings.
Semple 200492Patient monitoringQuality improvementResponse time (Level 4)1 unit with telemetry in a hospital in ConnecticutProcedure changes to enable nurse to respond to telemetry alarmsProblem areas were identified as the nurses’ inability to see critical alarm screen color change, hear critical alarms, and to know when their patient’s alarm is sounding.
A series of changes were implemented to enable nurse response.
Response to telemetry alarms decreased from 12 minutes to 1.57 minutes.
Singh 2004107Error risk detectionPretest, post- test studyPerceived type/cause of error (Level 3)1 academic rural primary care practice with 32 staff membersImplementation of electronic medical recordPerceived risk of errors decreased in nurse- physician and physician-chart interactions, but hazards increased in physician-patient interaction in the assessment stage as well as nurse-chart interactions.
Singh 200799Error risk detectionQuality improvementPerceived type/cause of error (Level 4)2 primary care practices serving rural populations in New YorkNoneNurses perceived being in a hurry, fatigued, stressed, or ill as well as not using available resources for help as the most prevalent type and cause of errors.
Hazard scores at site 2 were consistently higher, indicating that staff perceived greater frequency and/or severity of the errors in their practice.
Smith 200587Medication safetyQuality improvementMedication errors and adverse drug events (ADEs) (Level 4)1 hospital in IllinoisPharmacist staffing on patient care units to review orders and stock medications reduced errors by 45%; adult IV medications were standardized, and nonstandard doses were prepared by the pharmacy.There was a significant (a 66% drop in the FMEA score) reduction in ADEs.
van Tilburg 2006108Medication safety, CPOEQuality improvementErrors associated with chemotherapy (Level 4)Pediatric oncology patients in a hospital in the NetherlandsNoneBecause changes in ordered prescriptions could be made without being noticed by the nurse, a standardized procedure for changes in chemotherapy treatment schedules was made.
Because of administration errors, the procedure was changed so that only pediatric oncologists were allowed to administer vincristine via peripheral IV access.
Weir 2005101Medication safetyQuality improvementADEs associated with patient- controlled analgesia (PCA) (Level 4)1 hospital and clinics in CaliforniaNoneAreas needing change included using a standard IV PCA dosage or concentration protocol; adding the patient’s age to CPOE medication order screen; handwritten orders; PCA pumps programmed incorrectly; and monitoring patients using PCAs.
71% of ADEs were associated with PCA programming error, followed by human factors (15%), equipment problems (9%), and ordering errors (5%).
Plan-Do-Study-Act (PDSA)
Baird 2001103Medication safetyQuality improvementPatient outcomes and reduced costs in the ICU (Level 4)Physicians, nurses, and clinical pharmacists in a 115 adult ICU beds in 1 large medical center in TexasUsing a new heparin administration protocol in ICUInitial findings with 10 patients found that 90% of patients received optimal bolus doses (compared to 8.6% of the historical patients) and all received optimal infusion doses (compared to 3.4% of historical patients).
Patients received better heparin therapy because they received the right loading dose, reached a therapeutic level of the drug more quickly, and maintained the therapeutic level.
Nursing efficiency improved with fewer dose changes and laboratory tests. Medication and laboratory test costs decreased as did the patient’s length of stay.
Bolch 200597Care transitionsQuality improvementPatients having a documented discharge plan, patients screened for risk, patients receiving followup care within 10 days of discharge (Level 4)Patients ages ≥ 65, admitted to a hospital in South AustraliaModified the nursing assessment/risk assessment toolImprovements in the initiation and followup of discharge planning resulted in more documented discharge plans, increased risk assessment, increased referrals to community services, and improved communication between hospital staff and community providers.
Buhr 200680Pain managementQuality improvementImproved assessment and management of chronic pain (Level 4)Patients and nurses (licensed practical nurses (LPNs), certified nursing assistants (CNAs), and registered nurses (RNs)) in 1 nursing home in North CarolinaIncreased knowledge of chronic pain assessment and management through education.
Implemented updated policies and procedures, and used new tools for pain assessment and management.
Revised standing orders for pain management.
Pain assessment and management understanding improved in staff, especially in the CNAs.
Patient and family satisfaction increased, and feeling that pain was adequately addressed increased.
Docimo 200089Throughput in emergency department (ED)Quality improvementTime in ED for minor illnesses and injuries (Level 4)1 ED in 1 hospital in MarylandImproved both the processes and relationships of hospital staff using PDSA cyclesNonacute patients were fast-tracked to an average time of 1 hour, 47 minutes by not waiting behind higher-acuity patients for registration.
Physician assistants, nurses, and technicians reported improved working conditions and team spirit.
Dodds 2006119Practice variationQuality improvementLength of stay, reduced variation in process of care (Level 4)Patients with chronic obstructive pulmonary disease (COPD)Redesigned service delivery by using a continuous quality improvement methodology and PDSA cyclesDecrease in average length of stay.
Increase in the numbers of patients admitted directly to the emergency medical unit and transferred to the respirator department.
Improved the management of patient information and communication with patients.
Dunbar 2006100Pain management Practice variationQuality improvementFrequency of painful procedures, managing pain associated with painful procedures (Level 4)11 neonatal ICUsImplemented evidence-based practices for pain management and sedation in neonates using PDSA cyclesThe combination of using collaborative quality improvement techniques and local quality improvement efforts resulted in better patient outcomes.
Eisenberg 2002109IV incidentsQuality improvementIV care patient outcomes (Level 4)4 community hospitalsEducation of all staff nurses on IV site care and assessment, as well as assessment of central line, total parenteral nutrition (TPN). Revised 35 IV policies into 5, revised documentation flow sheets, and provided a resource manual.Reductions in complications and costs. Improved patient satisfaction.
No formal complaints about IV care.
Erdek 200493Pain managementProspective studyPain management and assessment (Level 4)2 surgical ICUs in 1 hospital in MarylandImplemented 4 PDSA cycles, including educating staff on pain management, modifying pain scales at patients’ bedsides, residents documenting pain scores for past 24 hours, and creating expectation that pain > 3 is a defect.Pain assessment improved from 42% to 71%, and pain management improved from 59% to 97%.
Documentation of pain assessment improved among nurses.
Farbstein 2001106Medication safetyQuality improvementTypes of medication administration errors (Level 4)6 improvement projects in hospitals in MassachusettsImplementation of best practices, using PDSA to assess impactThe results presented from the 6 improvement projects included faster therapeutic anticoagulation for patients receiving heparin; fewer look-alike/sound- alike errors; fewer PCA administration adverse events; safer administration of coumadin; improved patient information on their medication; and improved processing of the morning dispensing of medications in the pharmacy.
The investigators described success factors of medication safety projects as using data to measure outcomes; using forcing functions built into the process; pacing changes sequentially, not all at one time; low cost of changes; using a consultant to mentor team leaders; and using reported errors to assess implementation impact.
Horbar 200398Neonatal intensive careQuality improvementImproved quality and safety of neonatal intensive care (Level 4)34 centersImplemented, applying 4 key habits for improvement using rapid-cycle PDSADeveloped 51 potentially better practices that were implemented by multidisciplinary neonatal ICU teams in identifying, testing, and implementing change in practice.
Horner 2005117Pain managementPretest and post-test studyImproved pain assessment and management of residents (Level 3)9 nursing homes in North CarolinaChart audit and data feedback on quality indicators, provider education, and technical support for systems change using PDSAThe number of residents receiving pain assessments increased from 8% to 29%. Residents receiving nonpharmacological pain treatments increased from 31% to 42%.
Residents with daily moderate or excruciating pain had increased probability of pain medication use.
Leape 200686Medication reconciliation, communicating critical test resultsQuality improvementImplementation of safe practices (Level 4)58 hospitals (88%) in MassachusettsInstitute for Healthcare Improvement model for improvement to care practicesParticipating hospitals did so because of the following factors: the intrinsic appeal of the practice, access to experts, and the availability of implementation strategies.
Project success was associated with active engagement of senior management, physician engagement, increased use of PDSA cycles, participation in collaborative meetings.
Pronovost 200083Access to careQuality improvementNumber of ambulance bypass hours (Level 4)1 hospital in MarylandPDSA to act on identified root causes, targeting bed sharing for patients needing ICU care that were managed in the EDSignificant reduction in hours with an estimated $6 million in additional hospital revenue.
Success was achieved by teams integrating tools that improved processes and collaborative relationships.
Salvador 2003114Medication safetyQuality improvementSafety of hospital-based antenatal home care for high-risk women (Level 4)Physicians, nurses, and clinical pharmacists in 115 adult ICU beds in 1 large medical center in TexasUsing a new heparin administration protocol in ICUNew heparin protocol resulted in better patient care, improved nursing efficiency and work satisfaction, and reduced costs by $885 on average.
There were no differences in maternal or newborn health outcomes.
van Tiel 2006118Health care associated infectionsQuality improvementCompliance with infection control measures (Level 4)1 ICU and OR in a 715-bed university hospital in the NetherlandsInstruction and training of nursing and medical staff on PDSA cyclesNot wearing a face mask during procedures decreased to 0%; not wearing jewelry decreased to 33%.
Improved compliance with wound care, including hand washing before and after wound care and the use of disposable surgical wound sets.
Warburton 2004120Adverse outcomes in EDsQuality improvementDetect patients at risk for adverse outcomes, provide a plan of care, and target care services (Level 4)1 small hospital in CanadaImplementation of the Elder Alert program using PDSA cyclesProcess evaluation audits and regular meetings of providers and academic collaborators were essential improvement tools.
Screening criteria had to be adapted to the patient population.
Wojciechowski 2006122PDSAQuality improvementIncreasing access to patient education resources (Level 4)1 rehabilitation facility in a city in the MidwestImplementation of a new patient education system for medication and disease information using PDSA cyclesDesigning a new Web-based patient education system benefits from a process promoting change incrementally and collaboration.
Root-Cause Analysis (RCA)
Gowdy 200390Patient fallsQuality improvmentIncidence of inpatient falls (Level 4)1 hospital in North CarolinaImplemented an action plan to prevent patient fallsRCA identified risks for falls associated with confusion, gait disturbance, and self- toileting.
Inpatient fall rate decreased from 6.1 to 2.6 falls per 1,000 patient days (a 43% decrease during the study period).
Luther 2002104Adverse eventsQuality improvementIncidence of ADEs, ventilator- acquired pneumonia, central-venous- catheter-related bloodstream infections (Level 4)2 hospitals in TexasIncreased staffing levels and improved education. Conducted RCA to identify issues needing to be addressed by leadership and staff.Adverse events targeted by nurses using protocols decreased ADEs by 45%, ventilator-acquired pneumonia from 47.8/1,000 ventilator days to 10.9/1,000, and decreased central-venous-catheter- related bloodstream infections from the 90th to the 50th percentile of the National Nosocomial Infection Surveillance System.
Implementation of protocols decreased length of stay from 8.1 to 4.5 days.
Middleton 2007105Root causes of errorsCross- sectional studyAdoption of recommendations detected from RCA (Level 4)12 physicians (86% response rate) and 17 nurses (100% response rate) in Sydney, AustraliaNoneNurses were more likely than physicians to view RCA recommendations as “relevant to the causal statement,” “understandable,” “achievable,” and “measurable.”
Physicians and nurses involved in the RCA were significantly more likely to believe that the RCA recommendations would “eliminate” or “control” future risks.
Some recommendations rated as “relevant to the causal statement” by nurses were significantly less likely to also be rated as “achievable.”
Mills 200582Patient fallsQuality improvementIncidence of falls and major injuries due to falls (Level 4)100 VA acute and long-term care facilitiesAggregate RCA was used to support implementation of fall prevention strategies.61.4% of strategies were fully implemented, and 20.9% were partially implemented. 34% of the facilities reported a reduction in the number of falls, and 38.9% reported a reduction in major injures related to falls.
The impact of the interventions could have been hampered by making specific clinical changes without changing policies and providing staff education.
Mutter 200395Medication safetyQuality improvementFrequency of medication administration errors (Level 4)1 451-bed acute care hospital in New JerseyAfter assessing causes of errors, established a nonpunitive environment to encourage error reporting and interviewed providers who reported errors.Improvement requires constant and continual assessment of errors.
Rapid-cycle improvement was used to decrease medication administration errors and to inform changes.
Plews-Ogan 2004102Voluntary reporting of near miss/adverse eventsCross- sectional studyError reporting (Level 4)1 ambulatory site of a large teaching hospitalSystem analysis and redesign using RCA.Two-thirds of the 70 recommended recommendations were level 1, 23% level 2 (i.e., involving more complex interventions usually requiring significant groundwork), and 10% level 3 (i.e., involving other services).
Using RCA increased error reporting as system issues were addressed, not through individual blame.
RCA identified the underlying causes of reported errors, and improvements were made on an ongoing basis.
Rex 200096Medication safetyQuality improvementRates of ADEs (Level 4)1 hospital in TexasImplemented policy changes to use forcing or constraining functions and better personnel supportRCA identified environmental factors (e.g., patient acuity, change of shit) and staffing issues (e.g., new staff).
ADEs decreased by 45%. Implementing blame-free RCA enabled identification and prioritization of performance improvement initiatives and focus on systems issues.
Willeumier 200485Medication safety Health care associated infectionsQuality improvementRates of medication error reporting and ventilator- associated pneumonia (VAP) rates (Level 4)8-hospital system in northern IllinoisImproved medication availability, standardized nursing reassessment of medications, reinforced the 5 rights of medication administration, provided medication information, revised medication policies, and standardized nursing documentation of medication administration.
Redesigned oral hygiene processes, used head positioning, and used collection and culture techniques for better diagnosis.
Identified strategies based on proactive risk assessment (a composite of RCA and FMEA).
Medication error reporting increased and VAP rates decreased.
Greatest challenges were implementing and sustaining a culture of safety, the complexity of the health care system, underreporting of patient safety events, and medical staff’s acceptance of the disclosure policy.
Improvement is dependent upon the involvement of leadership, communication with staff, and the use of the appropriate technology.
Six Sigma
Germaine 200791Surgical site infections OR patient throughputQuality improvementOR turnover (Level 4)1 hospital in MichiganImplemented OR turnaround protocolTurnover decreased from 34 minutes to an average of 18 minutes, allowing volume to increase by 5%.
Surgical site infections decreased from 2.14% to 1.07%.
Guinane 200481Groin injury in cardiac catheterization patientsQuality improvementGroin injury rates (Level 4)A team of physicians, nurses, and administrators involved in the care of cardiac catheterization patients in 1 hospitalImplemented groin management process to decrease injury rates, reduce the cost of care, and improve customer satisfactionGroin injuries decreased from 4% to less than 1% (e.g., 41,666 defects to 8,849.5 defects) – sigma value improved from 3.23 to 3.87.
Length of stay that exceeded the specified upper limit decreased from 16% of the time to only 3% of the time.
Operating costs that exceeded the specified upper limit decreased from 18% to 3% of the time.
Johnson 2005125Chest pain managementQuality improvementTime for diagnosis and evidence-based treatment of patients with chest pain1 hospital in New YorkImplemented an algorithm, preprinted orders, and use of cardiac nurse practitioners from presentation in ED through dischargeIncreases in diagnosis of cardiac disease, cardiac catheterization, and stenting/bypass surgery, especially in women, Latinos, and patients > 60 years old.
Pexton 2004113Surgical site infectionsQuality improvementRate of colon and vascular surgical site infections (Level 4)1 medical center in West VirginiaA preoperative order set with a checklist including recommended antibiotics and weight-based dosages, education of team members, physician report cards, and anesthetists and nurses prompting surgeons to use antibiotics.Surgical site infection rates decreased by 91% (2.86 sigma), with an estimated potential annual savings of more than $1 million.
Aldarrab 2006126Emergency care of patients with ST-elevation myocardial infarction (MI)Quality improvementPatients with appropriate reperfusion and adjunctive pharmacological treatment (Level 4)3-site tertiary/quaternary facility in CanadaImplementation of evidence- based guidelines for ST-elevation in MI patientsAn RCA was used to understand current processes and to assess what could be standardized.
Targets were achieved in terms of using the appropriate reperfusion strategy, meeting the median time of < 30 minutes for thrombolytic therapy and 90 minutes for percutaneous coronary intervention, appropriate thrombolytic and adjunctive treatment use.
It was noted that without continued reinforcement of the new protocol, the process would regress to prior levels of performance.
Furman 200739Error reportingQuality improvementNear-miss error reports (Level 4)1 medical center in VirginiaImplemented an error reporting system, including a 24-hour hotlineNurses reported 44% of the near misses, physicians 8%, managers 20%, nonclinical staff 23%.
Over a period of 3 years, the number of error reports increased because there was a transparent discussion and feedback process.
Jimmerson 200588Medication safety Access to medical equipmentQuality improvementEfficiency of testing patient’s glucose level at the bedside (Level 4)1 medical-surgical ICU in a hospital in UtahInstalled glucometers in each room in the ICUReduced time to do glucose check from 17 to 4 minutes.
Improved ability to consistently implement the protocol. No unlabeled specimens at risk of erroneous identification.
Fewer RN interruptions and frustration.
Nowinski 2006121Medication administrationQuality improvementMedication administration errors (Level 4)1 hospital in PennsylvaniaRevised and streamlined medication administration process based on finding from an RCARapid, substantial, and continuing improvements in patient care were achieved.
Nursing staff reported higher levels of satisfaction.
Printezis 200759Using TPS in health careLiterature reviewReviewed five quality improvement projects to reduce medical errors in hospitals (Level 4)Improvement projects in hospitalsNoneSimple pathways of root causes lead to better operational performance.
Organizing principles of TPS improve reliability and effectiveness of health care delivery systems.
Problem-solver on projects should not be a consultant, but someone who is a stakeholder.
Many problems are associated with relationships with other departments. TPS makes work-around and rework difficult to continue.
TPS helps staff learn and identify waste in daily activities.
Front-line staff need to be enthusiastic about making improvements. Clear, concise, and objective communication is key.
Thompson 200384Medication administrationQuality improvementMissing medications Complexity of the medication administration process (Level 4)Pharmacy and nursing units at 1 hospital in PennsylvaniaImplemented: specificRapid, substantial, and continuing improvements in medication administration processes were achieved.
Nursing staff reported higher levels of satisfaction, associated with workflow improvements.

From: Chapter 44, Tools and Strategies for Quality Improvement and Patient Safety

Cover of Patient Safety and Quality
Patient Safety and Quality: An Evidence-Based Handbook for Nurses.
Hughes RG, editor.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.