Making health care safer II: an updated critical analysis of the evidence for patient safety practices.

OBJECTIVES: To review important patient safety practices for evidence of effectiveness, implementation, and adoption. DATA SOURCES: Searches of multiple computerized databases, gray literature, and the judgments of a 20-member panel of patient safety stakeholders. REVIEW METHODS: The judgments of the stakeholders were used to prioritize patient safety practices for review, and to select which practices received in-depth reviews and which received brief reviews. In-depth reviews consisted of a formal literature search, usually of multiple databases, and included gray literature, where applicable. In-depth reviews assessed practices on the following domains: • How important is the problem? • What is the patient safety practice? • Why should this practice work? • What are the beneficial effects of the practice? • What are the harms of the practice? • How has the practice been implemented, and in what contexts? • Are there any data about costs? • Are there data about the effect of context on effectiveness? We assessed individual studies for risk of bias using tools appropriate to specific study designs. We assessed the strength of evidence of effectiveness using a system developed for this project. Brief reviews had focused literature searches for focused questions. All practices were then summarized on the following domains: scope of the problem, strength of evidence for effectiveness, evidence on potential for harmful unintended consequences, estimate of costs, how much is known about implementation and how difficult the practice is to implement. Stakeholder judgment was then used to identify practices that were "strongly encouraged" for adoption, and those practices that were "encouraged" for adoption. RESULTS: From an initial list of over 100 patient safety practices, the stakeholders identified 41 practices as a priority for this review: 18 in-depth reviews and 23 brief reviews. Of these, 20 practices had their strength of evidence of effectiveness rated as at least "moderate," and 25 practices had at least "moderate" evidence of how to implement them. Ten practices were classified by the stakeholders as having sufficient evidence of effectiveness and implementation and should be "strongly encouraged" for adoption, and an additional 12 practices were classified as those that should be "encouraged" for adoption. CONCLUSIONS: The evidence supporting the effectiveness of many patient safety practices has improved substantially over the past decade. Evidence about implementation and context has also improved, but continues to lag behind evidence of effectiveness. Twenty-two patient safety practices are sufficiently well understood, and health care providers can consider adopting them now.

Republished paper: Improving safety culture on adult medical units through multidisciplinary teamwork and communication interventions: the TOPS Project.

AIM: The goal of this project was to improve unit-based safety culture through implementation of a multidisciplinary (pharmacy, nursing, medicine) teamwork and communication intervention. METHOD: The Agency for Healthcare Research and Quality Hospital Survey on Patient Safety Culture was used to determine the impact of the training with a before-after design. RESULTS: Surveys were returned from 454 healthcare staff before the training and 368 staff 1 year later. Five of eleven safety culture subscales showed significant improvement. Nurses perceived a stronger safety culture than physicians or pharmacists. CONCLUSION: While it is difficult to isolate the effects of the team training intervention from other events occurring during the year between training and postevaluation, overall the intervention seems to have improved the safety culture on these medical units.

Improving safety culture on adult medical units through multidisciplinary teamwork and communication interventions: the TOPS Project.

AIM: The goal of this project was to improve unit-based safety culture through implementation of a multidisciplinary (pharmacy, nursing, medicine) teamwork and communication intervention. METHOD: The Agency for Healthcare Research and Quality Hospital Survey on Patient Safety Culture was used to determine the impact of the training with a before-after design. RESULTS: Surveys were returned from 454 healthcare staff before the training and 368 staff 1 year later. Five of eleven safety culture subscales showed significant improvement. Nurses perceived a stronger safety culture than physicians or pharmacists. CONCLUSION: While it is difficult to isolate the effects of the team training intervention from other events occurring during the year between training and postevaluation, overall the intervention seems to have improved the safety culture on these medical units.

Making health care safer: a critical analysis of patient safety practices.

OBJECTIVES: Patient safety has received increased attention in recent years, but mostly with a focus on the epidemiology of errors and adverse events, rather than on practices that reduce such events. This project aimed to collect and critically review the existing evidence on practices relevant to improving patient safety. SEARCH STRATEGY AND SELECTION CRITERIA: Patient safety practices were defined as those that reduce the risk of adverse events related to exposure to medical care across a range of diagnoses or conditions. Potential patient safety practices were identified based on preliminary surveys of the literature and expert consultation. This process resulted in the identification of 79 practices for review. The practices focused primarily on hospitalized patients, but some involved nursing home or ambulatory patients. Protocols specified the inclusion criteria for studies and the structure for evaluation of the evidence regarding each practice. Pertinent studies were identified using various bibliographic databases (e.g., MEDLINE, PsycINFO, ABI/INFORM, INSPEC), targeted searches of the Internet, and communication with relevant experts. DATA COLLECTION AND ANALYSIS: Included literature consisted of controlled observational studies, clinical trials and systematic reviews found in the peer-reviewed medical literature, relevant non-health care literature and "gray literature." For most practices, the project team required that the primary outcome consist of a clinical endpoint (i.e., some measure of morbidity or mortality) or a surrogate outcome with a clear connection to patient morbidity or mortality. This criterion was relaxed for some practices drawn from the non-health care literature. The evidence supporting each practice was summarized using a prospectively determined format. The project team then used a predefined consensus technique to rank the practices according to the strength of evidence presented in practice summaries. A separate ranking was developed for research priorities. MAIN RESULTS: Practices with the strongest supporting evidence are generally clinical interventions that decrease the risks associated with hospitalization, critical care, or surgery. Many patient safety practices drawn primarily from nonmedical fields (e.g., use of simulators, bar coding, computerized physician order entry, crew resource management) deserve additional research to elucidate their value in the health care environment. The following 11 practices were rated most highly in terms of strength of the evidence supporting more widespread implementation. Appropriate use of prophylaxis to prevent venous thromboembolism in patients at risk; Use of perioperative beta-blockers in appropriate patients to prevent perioperative morbidity and mortality; Use of maximum sterile barriers while placing central intravenous catheters to prevent infections; Appropriate use of antibiotic prophylaxis in surgical patients to prevent postoperative infections; Asking that patients recall and restate what they have been told during the informed consent process; Continuous aspiration of subglottic secretions (CASS) to prevent ventilator-associated pneumonia; Use of pressure relieving bedding materials to prevent pressure ulcers; Use of real-time ultrasound guidance during central line insertion to prevent complications; Patient self-management for warfarin (Coumadin) to achieve appropriate outpatient anticoagulation and prevent complications; Appropriate provision of nutrition, with a particular emphasis on early enteral nutrition in critically ill and surgical patients; and Use of antibiotic-impregnated central venous catheters to prevent catheter-related infections. CONCLUSIONS: An evidence-based approach can help identify practices that are likely to improve patient safety. Such practices target a diverse array of safety problems. Further research is needed to fill the substantial gaps in the evidentiary base, particularly with regard to the generalizability of patient safety practices heretofore tested only in limited settings and to promising practices drawn from industries outside of health care.

The benefits of using hospitalists.

A good hospitalist program will more than pay for itself by improving outcomes, reducing length of stay, and enhancing hospital efficiency. Moreover, the program can often attract primary care physicians who prefer to use the service of a hospitalist.

Implications of the hospitalist model for medical students' education.

At many academic health centers, medical students in internal medicine, family medicine, and pediatrics are working with a new form of attending physician, the hospitalist. Although a growing literature demonstrates the benefits of hospitalists for patients and housestaff, the influence of hospitalists on students has been underemphasized. Advantages of the hospitalist model for students can include hospitalists' expertise in general inpatient medicine, their availability to teach throughout the day, and their role-modeling of the provision of high-quality and efficient care. However, the change in the ward attending workforce from non-hospitalist generalists, subspecialists, and biomedical researchers to generalist-hospitalists potentially limits students' exposure to the broad range of career opportunities the former group represents. The authors propose a research agenda to investigate the educational impact of the hospitalist model, suggest strategies to mitigate the limitations in students' exposures to subspecialty faculty, and recommend professional development in teaching for hospitalists to ensure that student education thrives in this new environment of inpatient medicine.

Refined crystal structure of DsRed, a red fluorescent protein from coral, at 2.0-A resolution.

The crystal structure of DsRed, a red fluorescent protein from a corallimorpharian, has been determined at 2.0-A resolution by multiple-wavelength anomalous dispersion and crystallographic refinement. Crystals of the selenomethionine-substituted protein have space group P2(1) and contain a tetramer with 222 noncrystallographic symmetry in the asymmetric unit. The refined model has satisfactory stereochemistry and a final crystallographic R factor of 0.162. The protein, which forms an obligatory tetramer in solution and in the crystal, is a squat rectangular prism comprising four protomers whose fold is extremely similar to that of the Aequorea victoria green fluorescent protein despite low ( approximately 23%) amino acid sequence homology. The monomer consists of an 11-stranded beta barrel with a coaxial helix. The chromophores, formed from the primary sequence -Gln-Tyr-Gly- (residues 66-68), are arranged in a approximately 27 x 34-A rectangular array in two approximately antiparallel pairs. The geometry at the alpha carbon of Gln-66 (refined without stereochemical restraints) is consistent with an sp(2) hybridized center, in accord with the proposal that red fluorescence is because of an additional oxidation step that forms an acylimine extension to the chromophore [Gross, L. A., Baird, G. S., Hoffman, R. C., Baldridge, K. K. & Tsien, R. Y. (2000) Proc. Natl. Acad. Sci. USA 87, 11990-11995]. The carbonyl oxygen of Phe-65 is almost 90 degrees out of the plane of the chromophore, consistent with theoretical calculations suggesting that this is the minimum energy conformation of this moiety despite the conjugation of this group with the rest of the chromophore.

Response to David Meltzer's paper "Hospitalists and the doctor-patient relationship".

The hospitalist movement, in which a generalist physician assumes the physician-of-record role for hospitalized patients in place of the patients' own primary physician is transforming American hospital care. Studies, including those by David Meltzer, support the premise that hospitalist care is less expensive and possibly of higher quality than inpatient care provided by primary physicians. Meltzer's survey of randomly selected individuals indicates that, although many would prefer that their own physician orchestrate their inpatient care, relatively few would pay more than $100 for this privilege. This indicates that patient dissatisfaction with the hospitalist model is unlikely to impede its widespread implementation. Hospitalists are just the latest example of the ongoing tension between generalism and specialization in American medicine. The hospitalist--unlike most traditional medical specialists--is actually a generalist by nature whose specialty is defined by the the site of the care rather than by organ, disease, patient population, or procedure.

A pneumonia practice guideline and a hospitalist-based reorganization lead to equivalent efficiency gains.

OBJECTIVE: To compare the impact of a practice guideline for a common inpatient disorder with that of a hospitalist-based reorganization of an academic medical service. STUDY DESIGN: Retrospective cohort study. PATIENTS AND METHODS: In July 1995 we introduced a clinical practice guideline for the treatment of community-acquired pneumonia at University of California San Francisco Moffitt-Long Hospital. Simultaneously, we implemented a structural change for half of the inpatient medical service, requiring earlier and more intensive faculty intervention, primarily by hospitalists. For 1 year, we studied the effect of these interventions on hospital costs, length of stay, and resource use. RESULTS: As reported previously, the hospitalist-based intervention resulted in significant decreases in average adjusted cost ($7777 vs $7007, P = .05) and length of stay (4.9 days vs 4.3 days, P = .01) compared with both concurrent and historical controls. For patients with community-acquired pneumonia, a similar savings occurred when fiscal year 1996 was compared with fiscal year 1995 ($8164 vs $6282, P= .015; 5.0 vs 4.2 days, P= .04). However, the effect was identical for the hospitalist and nonhospitalist groups. The reduced length of stay was associated with a borderline significant reduction in readmission rates (from 4.8% to 0.7%, P = .055) and no change in mortality rates. CONCLUSIONS: In this study, a hospitalist-based reorganization improved efficiency, with its greatest impact on the care of patients with disorders not covered by a practice guideline. The introduction of a guideline for a common diagnosis improved efficiency on both hospitalist- and nonhospitalist-based services. For common diagnoses amenable to practice guidelines, successful implementation of and compliance with guidelines may be an alternative to major organizational change.

Primary care physician attitudes regarding communication with hospitalists.

Hospitalist systems create discontinuity of care. Enhanced communication between the hospitalist and primary care physician (PCP) could mitigate the harms of discontinuity. We conducted a mailed survey of 4,155 physician members of the California Academy of Family Physicians to determine their preferences for and satisfaction with communication with hospitalists. We received 1,030 completed surveys (26%). PCPs overwhelmingly stated that they "very much prefer" to communicate with hospitalists by telephone (77%), at admission (73%), and discharge (78%). Only discharge medications (94%) and discharge diagnosis (90%) were deemed "very important" by >90% of PCPs. Of the 556 respondents (54%) who had ever used a hospitalist, 56% were very or somewhat satisfied with communication with hospitalists, and 68% agreed that hospitalists are a good idea. Regarding communication at discharge, only 33% of PCPs reported that discharge summaries always or usually arrive before the patient is seen for follow-up. Only 56% of PCPs in our survey were satisfied with communication with hospitalists. Hospitalists should communicate with PCPs in a timely manner by telephone, at least at admission and discharge, and provide the specific pieces of information deemed important by the vast majority of PCPs. Hospitalists should also ensure that discharge information arrives in time to assist the PCP in reassuming care of their patients. It may be possible to tailor communication to individual PCPs. Further research could assess the impact of such communication on patient satisfaction and outcomes.

The "continuity visit" and the hospitalist model of care.

Previous analyses have focused on the importance of hospitalist-primary care physician communication to mitigate the harms of discontinuity when hospitalists care for inpatients. We believe that both patients and physicians may benefit if primary physicians visit patients (or at least speak directly to them) during hospitalizations when a hospitalist is the physician-of-record. We propose calling such encounters the "continuity visit" to emphasize that the visit is not purely "social." Moreover, we encourage research on the value of continuity visits and recommend compensation if research establishes that these visits improve the efficiency and quality of inpatient care or patient satisfaction and comfort.

Caregiver perceptions of the reasons for delayed hospital discharge.

CONTEXT: Financial pressures have increased the emphasis on expeditious hospital discharge. Identification of barriers to timely discharge may help direct efforts to decrease unnecessary hospital days. OBJECTIVE: To identify caregivers' perceptions of reasons for discharge delays at an academic medical center. DESIGN: Survey and free-form written responses using a convenience sample (overall response rate, 68%). RESPONDENTS: 104 housestaff, 34 attending physicians, and 33 nurses. RESULTS: Nurses were much more likely than housestaff or attending physicians to cite inadequate communication as a reason for discharge delays. Nurses were also more likely to attribute delays to rounds and other conferences (48% vs. 22% and 9%, respectively; P = 0.05). Physicians, however, were more likely to cite delays in testing and availability of subacute care beds. Almost all housestaff and attendings thought that discharge decisions were generally made in the morning, and over 60% felt that discharge orders were usually written before noon. In contrast, none of the nurses thought that orders were usually written before noon. CONCLUSIONS: Caregivers at the same institution perceived different barriers to discharge and believed that discharge-related activities occurred at different times. To facilitate hospital discharge, communication gaps should be addressed and traditional morning routines should be reexamined.

Physician attitudes toward and prevalence of the hospitalist model of care: results of a national survey.

PURPOSE: We sought to determine the availability and utilization of, as well as physician attitudes toward, the hospitalist model in the United States. SUBJECTS AND METHODS: Using a telephone survey, we asked physicians who were board certified in internal medicine about their inpatient practice arrangements, the availability of hospitalist services, and their attitudes toward the hospitalist model. All physicians were generalists in active clinical practice. Using multivariable methods, we determined factors associated with attitudes toward the hospitalist model. RESULTS: We were able to contact 787 of 2,829 physicians who were randomly selected from a national list of board-certified internists, of whom 400 agreed to participate. Most respondents were familiar with the term "hospitalist" and had hospitalist services available in their community, and 28% used hospitalists for their inpatients. Few (2%) reported the presence of the "mandatory" hospitalist model. Physicians reported that the model was more commonly available in Western states (84% vs. 55% to 63% in other regions, P<0.0001). Seventy-three percent thought hospitalist systems would reduce continuity of care. Only 28% thought that patients would prefer care from an inpatient specialist, but 51% thought patients might get better care, and 47% thought patients might get more cost-effective care in a hospitalist system. In multivariable models, physicians who were in solo practice, those in specialties with more inpatient practice, and those who had more patients hospitalized each month responded more negatively about the model, whereas those with hospitalists in their community were more positive. CONCLUSIONS: Although agreeing that quality of care and efficiency might be improved, physicians were concerned about patient-doctor relationships and patient satisfaction in a hospitalist model. Future studies should determine the effect of the hospitalist model on these outcomes.

Crystallographic and energetic analysis of binding of selected anions to the yellow variants of green fluorescent protein.

The fluorescence emission of yellow fluorescent proteins (YFPs) has been shown to respond rapidly and reversibly to changes in the concentration of some small anions such as halides; this allows for the use of YFPs as genetically encodable Cl(-) sensors that may be targeted to specific organelles in living cells. Fluorescence is suppressed due to protonation of the chromophore upon anion binding, with a stronger level of interaction at low pH values. At pH 6.0, the apparent dissociation constant (K(app)) for Cl(-) is 32 mM for YFP and 22 mM for YFP-H148Q, whereas at pH 7.5, K(app) is 777 mM and 154 mM, respectively. In the cytosol, YFP-H148Q appears most promising as a halide sensor due to its high degree of sensitivity towards I(-) (K(app)=23 mM at pH 7.5). To aid in the design of variants with improved levels of specificity and affinity for Cl(-), we solved apo and I(-)-bound crystal structures of YFP-H148Q to 2.1 A resolution. The halide-binding site is found near van der Waals contact with the chromophore imidazolinone oxygen atom, in a small buried cavity adjacent to Arg96, which provides electrostatic stabilization. The halide ion is hydrogen bonded to the phenol group of T203Y, consistent with a mutational analysis that indicates that T203Y is indispensible for tight binding. A series of conformational changes occurs in the amphiphilic site upon anion binding, which appear to be propagated to the beta-bulge region around residue 148 on the protein surface. Anion binding raises the chromophore pK(a) values, since delocalization of the phenolate negative charge over the chromophore skeleton is suppressed. Extraction of microscopic binding constants for the linked equilibrium between anion and proton binding indicates that anion selectivity by YFP is related to hydration forces. Specific suggestions to improve Cl(-) binding to YFP-H148Q based on size and hydration energy are proposed.

Probing the ground state structure of the green fluorescent protein chromophore using Raman spectroscopy.

We present Raman spectra, obtained using 752 nm excitation, on wild-type GFP and the S65T mutant of this intrinsically fluorescent protein together with data on a model chromophore, ethyl 4-(4-hydroxyphenyl)methylidene-2-methyl-5-oxoimidazolacetate . In the pH range 1-14, the model compound has two macroscopic pK(a)s of 1.8 and 8.2 attributed to ionization of the imidazolinone ring nitrogen and the phenolic hydroxyl group, respectively. Comparison of the model chromophore with the chromophore in wild-type GFP and the S65T mutant reveals that the cationic form, with both the imidazolinone ring nitrogen and the phenolic oxygen protonated, is not present in these particular GFP proteins. Our results do not provide any evidence for the zwitterionic form of the chromophore, with the phenolic group deprotonated and the imidazolinone ring nitrogen protonated, being present in the GFP proteins. In addition, since the position of the Raman bands is a property exclusively of the ground state structure, the data enable us to investigate how protein-chromophore interactions affect the ground state structure of the chromophore without contributions from excited state effects. It is found that the ground state structure of the anionic form of the chromophore, which is most relevant to the fluorescent properties, is strongly dependent on the chromophore environment whereas the neutral form seems to be insensitive. A linear correlation between the absorption properties and the ground state structure is demonstrated by plotting the absorption maxima versus the wavenumber of a Raman band found in the range 1610-1655 cm(-1).