Examining the research taxonomy of artificial intelligence, deep learning & machine learning in the financial sphere-a bibliometric analysis.

This paper surveys the extant literature on machine learning, artificial intelligence, and deep learning mechanisms within the financial sphere using bibliometric methods. We considered the conceptual and social structure of publications in ML, AI, and DL in finance to better understand the research's status, development, and growth. The study finds an upsurge in publication trends within this research arena, with a bit of concentration around the financial domain. The institutional contributions from USA and China constitute much of the literature on applying ML and AI in finance. Our analysis identifies emerging research themes, with the most futuristic being ESG scoring using ML and AI. However, we find there is a lack of empirical academic research with a critical appraisal of these algorithmic-based advanced automated financial technologies. There are severe pitfalls in the prediction process using ML and AI due to algorithmic biases, mostly in the areas of insurance, credit scoring and mortgages. Thus, this study indicates the next evolution of ML and DL archetypes in the economic sphere and the need for a strategic turnaround in academics regarding these forces of disruption and innovation that are shaping the future of finance.

Safely Managed On-Site Sanitation: A National Assessment of Sanitation Services and Potential Fecal Exposure in Indonesia.

Sustainable Development Goal target 6.2 calls for universal access to adequate and equitable sanitation, setting a more ambitious standard for 'safely managed sanitation services'. On-site sanitation systems (e.g., septic tanks) are widely used in low- and middle-income countries (LMICs). However, the lack of indicators for assessing fecal exposure risks presents a barrier to monitoring safely managed services. Furthermore, geographic diversity and frequency of disasters require a more nuanced approach to risk-informed decision-making. Taking Indonesia as an example, the purpose of this paper is to provide insights into current status and practices for on-site sanitation services in the contexts of LMICs. Using a dataset from a national socio-economic survey (n = 295,155) coupled with village census (n = 83,931), we assessed (1) household sanitation practices across Indonesia stratified by city-level population density and meteorological factors, (2) factors associated with septic tank emptying practice, and (3) inequalities in potential fecal exposure as measured by population density and WASH access by wealth quintile. We found a high reliance on on-site sanitation facilities (80.0%), almost half of which are assumed to be 'uncontained' septic tanks and one in ten facilities discharging untreated waste directly into the environment. The most densely populated areas had the highest rates of septic tank emptying, though emptying rates were just 17.0%, while in the lowest population density group, emptying was rarely reported. Multivariate regression analysis demonstrated an association between flooding and drought occurrence and septic tank emptying practice. Higher groundwater usage for drinking among poorer households suggests unsafe sanitation may disproportionally affect the poor. Our study underscores the urgent need to strengthen the monitoring of on-site sanitation in LMICs by developing contextualized standards. Furthermore, the inequalities in potential fecal exposure require greater attention and tailored support mechanisms to ensure the poorest gain access to safely managed sanitation services.

The cost-effectiveness of preimplantation genetic testing for aneuploidy in the United States: an analysis of cost and birth outcomes from 158,665 in vitro fertilization cycles.

BACKGROUND: A controversial and unresolved question in reproductive medicine is the utility of preimplantation genetic testing for aneuploidy as an adjunct to in vitro fertilization. Infertility is prevalent, but its treatment is notoriously expensive and typically not covered by insurance. Therefore, cost-effectiveness is critical to consider in this context. OBJECTIVE: This study aimed to analyze the cost-effectiveness of preimplantation genetic testing for aneuploidy for the treatment of infertility in the United States. STUDY DESIGN: As reported to the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System, a national data registry, in vitro fertilization cycles occurring between 2014 and 2016 in the United States were analyzed. A probabilistic decision tree was developed using empirical outputs to simulate the events and outcomes associated with in vitro fertilization with and without preimplantation genetic testing for aneuploidy. The treatment strategies were (1) in vitro fertilization with intended preimplantation genetic testing for aneuploidy and (2) in vitro fertilization with transfers of untested embryos. Patients progressed through the treatment model until they achieved a live birth or 12 months after ovarian stimulation. Clinical costs related to both treatment strategies were extracted from the literature and considered from both the patient and payer perspectives. Outcome metrics included incremental cost (measured in 2018 US dollars), live birth outcomes, incremental cost-effectiveness ratio, and incremental cost per live birth between treatment strategies. RESULTS: The study population included 114,157 first fresh in vitro fertilization stimulations and 44,508 linked frozen embryo transfer cycles. Of the fresh stimulations, 16.2% intended preimplantation genetic testing for aneuploidy and 83.8% did not. In patients younger than 35 years old, preimplantation genetic testing for aneuploidy was associated with worse clinical outcomes and higher costs. At age 35 years and older, preimplantation genetic testing for aneuploidy led to more cumulative births but was associated with higher costs from both perspectives. From a patient perspective, the incremental cost per live birth favored the no preimplantation genetic testing for aneuploidy strategy from the <35 years age group to the 38 years age group and beginning at age 39 years favored preimplantation genetic testing for aneuploidy. From a payer perspective, the incremental cost per live birth favored preimplantation genetic testing for aneuploidy regardless of patient age. CONCLUSION: The cost-effectiveness of preimplantation genetic testing for aneuploidy is dependent on patient age and perspective. From an economic perspective, routine preimplantation genetic testing for aneuploidy should not be universally adopted; however, it may be cost-effective in certain scenarios.

Population-based trends in hospitalizations due to injection drug use-related serious bacterial infections, Oregon, 2008 to 2018.

BACKGROUND: Injection drug use has far-reaching social, economic, and health consequences. Serious bacterial infections, including skin/soft tissue infections, osteomyelitis, bacteremia, and endocarditis, are particularly morbid and mortal consequences of injection drug use. METHODS: We conducted a population-based retrospective cohort analysis of hospitalizations among patients with a diagnosis code for substance use and a serious bacterial infection during the same hospital admission using Oregon Hospital Discharge Data. We examined trends in hospitalizations and costs of hospitalizations attributable to injection drug use-related serious bacterial infections from January 1, 2008 through December 31, 2018. RESULTS: From 2008 to 2018, Oregon hospital discharge data included 4,084,743 hospitalizations among 2,090,359 patients. During the study period, hospitalizations for injection drug use-related serious bacterial infection increased from 980 to 6,265 per year, or from 0.26% to 1.68% of all hospitalizations (P<0.001). The number of unique patients with an injection drug use-related serious bacterial infection increased from 839 to 5,055, or from 2.52% to 8.46% of all patients (P<0.001). While hospitalizations for all injection drug use-related serious bacterial infections increased over the study period, bacteremia/sepsis hospitalizations rose most rapidly with an 18-fold increase. Opioid use diagnoses accounted for the largest percentage of hospitalizations for injection drug use-related serious bacterial infections, but hospitalizations for amphetamine-type stimulant-related serious bacterial infections rose most rapidly with a 15-fold increase. People living with HIV and HCV experienced increases in hospitalizations for injection drug use-related serious bacterial infection during the study period. Overall, the total cost of hospitalizations for injection drug use-related serious bacterial infections increased from $16,305,129 in 2008 to $150,879,237 in 2018 (P<0.001). CONCLUSIONS: In Oregon, hospitalizations for injection drug use-related serious bacterial infections increased dramatically and exacted a substantial cost on the health care system from 2008 to 2018. This increase in hospitalizations represents an opportunity to initiate substance use disorder treatment and harm reduction services to improve outcomes for people who inject drugs.

Achieving the Sustainable Development Goals for water and sanitation in Indonesia - Results from a five-year (2013-2017) large-scale effectiveness evaluation.

BACKGROUND: Access to safe sanitation and the elimination of open defecation are pre-conditions for improved child health and nutrition and wider achievement of the Sustainable Development Goals (SDGs). While Indonesia has a solid policy framework, the country ranks third globally in terms of numbers of people practicing open defecation. OBJECTIVES: Our aim was to assess the effectiveness of a five-year strategy to reduce open defecation through accelerating implementation of the national sanitation program across districts receiving variable levels of external support. METHODS: Among three provinces with poor sanitation program performance, districts were selected to receive one of three levels of external support. High intensity districts (n = 6) benefitted from enabling environment strengthening support including political and social mobilization, direct capacity development, and efforts to strengthen planning, budgeting, monitoring and supervision; learning districts (n = 16) benefitted from cross-district learning opportunities and political mobilization through provincial government advocacy efforts; and comparison districts (n = 58) were monitored under routine program conditions. Outcomes included open defecation free (ODF) status and new toilet facility construction and were assessed through village level monitoring systems across all districts. Negative binomial regression and multivariate analysis were used to assess associations between levels of intervention intensity and outcomes. FINDINGS: Among districts receiving high-intensity external support improvements in political commitment, planning, coordination, financing, monitoring and supervision were observed. Relative to comparison districts, high intensity districts were more likely to be ODF (aRR 4.65, CI 2.12-10.20) with greater increase in household toilet coverage (aRR 11.15 CI 1.04-119.82). Weaker non-significant associations with ODF were observed among learning districts relative to comparison districts. INTERPRETATION: Efforts to strengthen provincial and district government capacity to implement sanitation programming in Indonesia can yield substantial improvements in outcomes in a relatively short period of time.

Spatial and temporal clustering of patients hospitalized with laboratory-confirmed influenza in the United States.

BACKGROUND: Timing of influenza spread across the United States is dependent on factors including local and national travel patterns and climate. Local epidemic intensity may be influenced by social, economic and demographic patterns. Data are needed to better explain how local socioeconomic factors influence both the timing and intensity of influenza seasons to result in national patterns. METHODS: To determine the spatial and temporal impacts of socioeconomics on influenza hospitalization burden and timing, we used population-based laboratory-confirmed influenza hospitalization surveillance data from the CDC-sponsored Influenza Hospitalization Surveillance Network (FluSurv-NET) at up to 14 sites from the 2009/2010 through 2013/2014 seasons (n = 35,493 hospitalizations). We used a spatial scan statistic and spatiotemporal wavelet analysis, to compare temporal patterns of influenza spread between counties and across the country. RESULTS: There were 56 spatial clusters identified in the unadjusted scan statistic analysis using data from the 2010/2011 through the 2013/2014 seasons, with relative risks (RRs) ranging from 0.09 to 4.20. After adjustment for socioeconomic factors, there were five clusters identified with RRs ranging from 0.21 to 1.20. In the wavelet analysis, most sites were in phase synchrony with one another for most years, except for the H1N1 pandemic year (2009-2010), wherein most sites had differential epidemic timing from the referent site in Georgia. CONCLUSIONS: Socioeconomic factors strongly impact local influenza hospitalization burden. Influenza phase synchrony varies by year and by socioeconomics, but is less influenced by socioeconomics than is disease burden.

Short-Term Incidence of Sequelae of HCV Infection Among Medicaid Beneficiaries in Oregon.

OBJECTIVES: Given the known high morbidity and mortality of hepatitis C virus (HCV) infection in Oregon, we sought to develop a practical method of estimating the severe sequelae of HCV infection among Medicaid beneficiaries in Oregon. METHODS: We assembled a retrospective cohort that identified all Oregon Medicaid beneficiaries with HCV infection enrolled for at least 1 year during 2009-2013. We linked this cohort to 3 data sets to identify HCV-related deaths, cases of hepatocellular carcinoma (HCC), and first hospitalizations for advanced liver disease (ALD). We calculated incidence density rates and used multivariable Cox regression modeling to calculate adjusted hazard ratios (aHRs) to evaluate the association between demographic characteristics (birth year, sex, race, ethnicity) and these 3 outcomes. RESULTS: Of 11 790 Oregon Medicaid beneficiaries with HCV infection, 474 (4.0%) had an HCV-related death, 156 (1.3%) had HCC, and 596 (5.1%) had a first hospitalization for ALD. Adjusted hazard ratios for deaths were 2.2 (95% confidence interval [CI], 1.6-2.8) among persons born in 1945 through 1965 (vs persons born after 1965), 2.1 (95% CI, 1.7-2.5) among males (vs females), and 1.9 (95% CI, 1.2-2.9) among Asian/Pacific Islanders and 2.2 (95% CI, 1.5-3.2) among American Indian/Alaska Natives (vs white persons). The same risk groups had significant aHRs for first hospitalizations for ALD. Persons born before 1945 (aHR = 17.0; 95% CI, 5.2-55.8) and in 1945 through 1965 (aHR = 12.8; 95% CI, 4.1-40.3) vs born after 1965, males (aHR = 3.3; 95% CI, 2.3-4.8) vs females, and Asian/Pacific Islanders (aHR = 3.9; 95% CI, 2.3-6.7) vs white persons had higher risks for HCC. CONCLUSIONS: Continued assessments using the methods piloted in this study will allow Oregon to monitor trends in severe sequelae of HCV infection over time.

Current Epidemiology and Trends in Invasive Haemophilus influenzae Disease-United States, 2009-2015.

Background: Following Haemophilus influenzae serotype b (Hib) conjugate vaccine introduction in the 1980s, Hib disease in young children dramatically decreased, and epidemiology of invasive H. influenzae changed. Methods: Active surveillance for invasive H. influenzae disease was conducted through Active Bacterial Core surveillance sites. Incidence rates were directly standardized to the age and race distribution of the US population. Results: During 2009-2015, the estimated mean annual incidence of invasive H. influenzae disease was 1.70 cases per 100000 population. Incidence was highest among adults aged ≥65 years (6.30) and children aged <1 year (8.45); many cases in infants aged <1 year occurred during the first month of life in preterm or low-birth-weight infants. Among children aged <5 years (incidence: 2.84), incidence was substantially higher in American Indian and Alaska Natives AI/AN (15.19) than in all other races (2.62). Overall, 14.5% of cases were fatal; case fatality was highest among adults aged ≥65 years (20%). Nontypeable H. influenzae had the highest incidence (1.22) and case fatality (16%), as compared with Hib (0.03; 4%) and non-b encapsulated serotypes (0.45; 11%). Compared with 2002-2008, the estimated incidence of invasive H. influenzae disease increased by 16%, driven by increases in disease caused by serotype a and nontypeable strains. Conclusions: Invasive H. influenzae disease has increased, particularly due to nontypeable strains and serotype a. A considerable burden of invasive H. influenzae disease affects the oldest and youngest age groups, particularly AI/AN children. These data can inform prevention strategies, including vaccine development.

Social determinants of influenza hospitalization in the United States.

BACKGROUND: Influenza hospitalizations result in substantial morbidity and mortality each year. Little is known about the association between influenza hospitalization and census tract-based socioeconomic determinants beyond the effect of individual factors. OBJECTIVE: To evaluate whether census tract-based determinants such as poverty and household crowding would contribute significantly to the risk of influenza hospitalization above and beyond individual-level determinants. METHODS: We analyzed 33 515 laboratory-confirmed influenza-associated hospitalizations that occurred during the 2009-2010 through 2013-2014 influenza seasons using a population-based surveillance system at 14 sites across the United States. RESULTS: Using a multilevel regression model, we found that individual factors were associated with influenza hospitalization with the highest adjusted odds ratio (AOR) of 9.20 (95% CI 8.72-9.70) for those ≥65 vs 5-17 years old. African Americans had an AOR of 1.67 (95% CI 1.60-1.73) compared to Whites, and Hispanics had an AOR of 1.21 (95% CI 1.16-1.26) compared to non-Hispanics. Among census tract-based determinants, those living in a tract with ≥20% vs <5% of persons living below poverty had an AOR of 1.31 (95% CI 1.16-1.47), those living in a tract with ≥5% vs <5% of persons living in crowded conditions had an AOR of 1.17 (95% CI 1.11-1.23), and those living in a tract with ≥40% vs <5% female heads of household had an AOR of 1.32 (95% CI 1.25-1.40). CONCLUSION: Census tract-based determinants account for 11% of the variability in influenza hospitalization.

Generalisability of vaccine effectiveness estimates: an analysis of cases included in a postlicensure evaluation of 13-valent pneumococcal conjugate vaccine in the USA.

OBJECTIVES: External validity, or generalisability, is the measure of how well results from a study pertain to individuals in the target population. We assessed generalisability, with respect to socioeconomic status, of estimates from a matched case-control study of 13-valent pneumococcal conjugate vaccine effectiveness for the prevention of invasive pneumococcal disease in children in the USA. DESIGN: Matched case-control study. SETTING: Thirteen active surveillance sites for invasive pneumococcal disease in the USA. PARTICIPANTS: Cases were identified from active surveillance and controls were age and zip code matched. OUTCOME MEASURES: Socioeconomic status was assessed at the individual level via parent interview (for enrolled individuals only) and birth certificate data (for both enrolled and unenrolled individuals) and at the neighbourhood level by geocoding to the census tract (for both enrolled and unenrolled individuals). Prediction models were used to determine if socioeconomic status was associated with enrolment. RESULTS: We enrolled 54.6% of 1211 eligible cases and found a trend toward enrolled cases being more affluent than unenrolled cases. Enrolled cases were slightly more likely to have private insurance at birth (p=0.08) and have mothers with at least some college education (p<0.01). Enrolled cases also tended to come from more affluent census tracts. Despite these differences, our best predictive model for enrolment yielded a concordance statistic of only 0.703, indicating mediocre predictive value. Variables retained in the final model were assessed for effect measure modification, and none were found to be significant modifiers of vaccine effectiveness. CONCLUSIONS: We conclude that although enrolled cases are somewhat more affluent than unenrolled cases, our estimates are externally valid with respect to socioeconomic status. Our analysis provides evidence that this study design can yield valid estimates and the assessing generalisability of observational data is feasible, even when unenrolled individuals cannot be contacted.

An assessment of the screening method to evaluate vaccine effectiveness: the case of 7-valent pneumococcal conjugate vaccine in the United States.

The screening method, which employs readily available data, is an inexpensive and quick means of estimating vaccine effectiveness (VE). We compared estimates of effectiveness of heptavalent pneumococcal conjugate vaccine (PCV7) against invasive pneumococcal disease (IPD) using the screening and case-control methods. Cases were children aged 19-35 months with pneumococcus isolated from normally sterile sites residing in Active Bacterial Core surveillance areas in the United States. Case-control VE was estimated for 2001-2004 by comparing the odds of vaccination among cases and community controls. Screening-method VE for 2001-2009 was estimated by comparing the proportion of cases vaccinated to National Immunization Survey-derived coverage among the general population. To evaluate the plausibility of screening-method VE findings, we estimated attack rates among vaccinated and unvaccinated persons. We identified 1,154 children with IPD. Annual population PCV7 coverage with ≥1 dose increased from 38% to 97%. Case-control VE for ≥1 dose was estimated as 75% against all-serotype IPD (annual range: 35-83%) and 91% for PCV7-type IPD (annual range: 65-100%). By the screening method, the overall VE was 86% for ≥1 dose (annual range: -240-70%) against all-serotype IPD and 94% (annual range: 62-97%) against PCV7-type IPD. As cases of PCV7-type IPD declined during 2001-2005, estimated attack rates for all-serotype IPD among vaccinated and unvaccinated individuals became less consistent than what would be expected with the estimated effectiveness of PCV7. The screening method yields estimates of VE that are highly dependent on the time period during which it is used and the choice of outcome. The method should be used cautiously to evaluate VE of PCVs.

Intrathecal baclofen in multiple sclerosis: too little, too late?

The majority of patients with multiple sclerosis (MS) have symptoms of spasticity that increasingly impair function as the disease progresses. With appropriate treatment, however, quality of life can be improved. Oral antispasticity medications are useful in managing mild spasticity but are frequently ineffective in controlling moderate to severe spasticity, because patients often cannot tolerate the adverse effects of increasing doses. Intrathecal baclofen (ITB) therapy can be an effective alternative to oral medications in patients who have a suboptimal response to oral medications or who cannot tolerate dose escalation or multidrug oral regimens. ITB therapy may be underutilized in the MS population because clinicians (a) are more focused on disease-modifying therapies rather than symptom control, (b) underestimate the impact of spasticity on quality of life, and (c) have concerns about the cost and safety of ITB therapy. Delivery of ITB therapy requires expertly trained staff and proper facilities for pump management. This article summarizes the findings and recommendations of an expert panel on the use of ITB therapy in the MS population and the role of the physician and comprehensive care team in patient selection, screening, and management.

Socioeconomic and racial/ethnic disparities in the incidence of bacteremic pneumonia among US adults.

OBJECTIVES: We examined associations between the socioeconomic characteristics of census tracts and racial/ethnic disparities in the incidence of bacteremic community-acquired pneumonia among US adults. METHODS: We analyzed data on 4870 adults aged 18 years or older with community-acquired bacteremic pneumonia identified through active, population-based surveillance in 9 states and geocoded to census tract of residence. We used data from the 2000 US Census to calculate incidence by age, race/ethnicity, and census tract characteristics and Poisson regression to estimate rate ratios (RRs) and 95% confidence intervals (CIs). RESULTS: During 2003 to 2004, the average annual incidence of bacteremic pneumonia was 24.2 episodes per 100 000 Black adults versus 10.1 per 100 000 White adults (RR = 2.40; 95% CI = 2.24, 2.57). Incidence among Black residents of census tracts with 20% or more of persons in poverty (most impoverished) was 4.4 times the incidence among White residents of census tracts with less than 5% of persons in poverty (least impoverished). Racial disparities in incidence were reduced but remained significant in models that controlled for age, census tract poverty level, and state. CONCLUSIONS: Adults living in impoverished census tracts are at increased risk of bacteremic pneumonia and should be targeted for prevention efforts.

The epidemiology and clinical characteristics of patients with newly diagnosed alcohol-related liver disease: results from population-based surveillance.

GOALS: We describe the epidemiology of outpatients newly diagnosed with chronic alcoholic liver disease and describe predictors of cirrhosis and referral for specialty care. BACKGROUND: Alcohol is a major cause of liver disease in the United States. Most previous work has described hospitalized patients. STUDY: Participants were identified through prospective population-based surveillance in gastroenterology practices Multnomah County, Oregon and New Haven County, Connecticut; and primary care and gastroenterology practices from Kaiser Permanente Northern California in Alameda County during 1999 to 2001. Patients were interviewed, a blood specimen obtained, and their medical record reviewed. RESULTS: We identified 82 patients from gastroenterology practices with newly diagnosed alcoholic liver disease. Their median age was 50.0 years. 72.0% were male and 79.3% were White. The median age at initiation of alcohol use was 17.0 years. 43.9% of patients had evidence of cirrhosis at the time of diagnosis. Only 40.2% reported alcohol as the cause of their liver disease. Patients with cirrhosis were more likely to be older, have a higher median number of years of heavy alcohol consumption, and to have been hospitalized for a liver-related complication than noncirrhotic patients. An additional 83 primary care patients were more likely to be older, to be drinking alcohol at study interview, and to not have cirrhosis than patients referred for gastroenterology care. CONCLUSIONS: Patients with alcoholic liver disease may present at a late stage and may not identify alcohol as a cause for their liver disease. Improved patient screening and education may limit morbidity and mortality.

Antimicrobial prescribing for upper respiratory infections and its effect on return visits.

BACKGROUND AND OBJECTIVES: Antibiotic resistance is a growing problem that complicates the treatment of various illnesses. This study analyzes Medicaid encounter data to (1) determine antibiotic prescribing rates for common respiratory tract infections in Oregon and (2) assess the effect of receiving an antibiotic at an index visit on whether there was a return visit within 30 days. METHODS: Subjects included in this study were Medicaid patients in Oregon between 2001--2003 who were enrolled in Medicaid for a full year and were diagnosed with an upper respiratory tract infection, including bronchitis, sinusitis, acute otitis media (AOM), pharyngitis, and upper respiratory infections (URIs). Claims data were analyzed to determine receipt of an antibiotic within 3 days of the initial visit and if there was a return visit within 30 days. RESULTS: During 2001--2003, the proportion of patients receiving antibiotics for bronchitis and sinusitis decreased, from 70% to 61%, and from 78% to 74%, respectively, while antibiotic prescribing for AOM, URI, and pharyngitis changed little. After controlling for age, gender, race/ethnicity, Medicaid plan type, and location, we determined that patients who had received antibiotics during the index visit for AOM, URI, and pharyngitis were more likely to return with a respiratory tract infection during the subsequent 30 days than patients who did not receive antibiotics. CONCLUSIONS: Antibiotic prescribing among Medicaid patients in Oregon has decreased. Receiving an antibiotic does not decrease the rate of subsequent return visits.

Effectiveness of pharmacy policies designed to limit inappropriate vancomycin use: a population-based assessment.

OBJECTIVE: In Oregon in 1994, a population-based study of 66 nonpsychiatric hospitals indicated that 40% of vancomycin orders were inappropriate according to Centers for Disease Control and Prevention guidelines. We repeated the study to determine whether vancomycin use had been affected by pharmacy policies implemented following the 1994 study. METHODS: We surveyed pharmacists in nonpsychiatric hospitals in Oregon regarding vancomycin use policies in their hospitals. Using pharmacy records, we identified and abstracted the charts of all patients in Oregon hospitals receiving vancomycin during a 3-week period to determine appropriate use of vancomycin. RESULTS: Thirteen (20%) of 64 hospitals had implemented a vancomycin restriction policy since 1994; none ofthe remaining hospitals in the study had a policy. In 1999, hospitals with vancomycin restriction policies had substantially decreased rates of inappropriate vancomycin use compared with hospitals without such policies (1.0 vs 1.8 orders per 1,000 patient-days; P = .01). Compared with 1994 baseline rates of inappropriate use, hospitals that adopted policies experienced a decrease (from 1.5 orders per 1,000 patient-days in 1994 to 1.0 in 1999; P= .13), whereas hospitals without policies experienced a statistically significant increase (from 0.9 orders per 1,000 patient-days in 1994 to 1.8 in 1999; P= .001). Restriction policies were most effective at reducing rates of inappropriate use for treatment of confirmed gram-positive infections and prophylaxis. CONCLUSION: Vancomycin restriction policies were associated with a decrease in inappropriate therapeutic and prophylactic vancomycin use, but had no effect on inappropriate empiric use. Hospitals considering limits regarding inappropriate use should consider implementation of institution-based vancomycin restriction policies as part of an overall strategy.