Peter M. Houck

Antimicrobial Prophylaxis for Surgery: An Advisory Statement from the National Surgical Infection Prevention Project

In January 2003, leadership of the Medicare National Surgical Infection Prevention Project hosted the Surgical Infection Prevention Guideline Writers Workgroup (SIPGWW) meeting. The objectives were to review areas of agreement among the most-recently published guidelines for surgical antimicrobial prophylaxis, to address inconsistencies, and to discuss issues not currently addressed. The participants included authors from most of the groups that have published North American guidelines for antimicrobial prophylaxis, as well as authors from several specialty colleges. Nominal group process was used to draft a consensus paper that was widely circulated for comment. The consensus positions of SIPGWW include that infusion of the first antimicrobial dose should begin within 60 min before surgical incision and that prophylactic antimicrobials should be discontinued within 24 h after the end of surgery. This advisory statement provides an overview of other issues related to antimicrobial prophylaxis, including specific suggestions regarding antimicrobial selection.In January 2003, leadership of the Medicare National Surgical Infection Prevention Project hosted the Surgical Infection Prevention Guideline Writers Workgroup (SIPGWW) meeting. The objectives were to review areas of agreement among the most-recently published guidelines for surgical antimicrobial prophylaxis, to address inconsistencies, and to discuss issues not currently addressed. The participants included authors from most of the groups that have published North American guidelines for antimicrobial prophylaxis, as well as authors from several specialty colleges. Nominal group process was used to draft a consensus paper that was widely circulated for comment. The consensus positions of SIPGWW include that infusion of the first antimicrobial dose should begin within 60 min before surgical incision and that prophylactic antimicrobials should be discontinued within 24 h after the end of surgery. This advisory statement provides an overview of other issues related to antimicrobial prophylaxis, including specific suggestions regarding antimicrobial selection.

Treatment of Acute Myocardial Infarction and 30-Day Mortality among Women and Men

BACKGROUND Previous studies have suggested that women with acute myocardial infarction receive less aggressive therapy than men. We used data from the Cooperative Cardiovascular Project to determine whether women and men who were ideal candidates for therapy after acute myocardial infarction were treated differently. METHODS Information was abstracted from the charts of 138,956 Medicare beneficiaries (49 percent of them women) who had an acute myocardial infarction in 1994 or 1995. Multivariate analysis was used to assess differences between women and men in the medications administered, the procedures used, the assignment of do-not-resuscitate status, and 30-day mortality. RESULTS Among ideal candidates for therapy, women in all age groups were less likely to undergo diagnostic catheterization than men. The difference was especially pronounced among older women; for a woman 85 years of age or older, the adjusted relative risk was 0.75 (95 percent confidence interval, 0.68 to 0.83). Women were somewhat less likely than men to receive thrombolytic therapy within 60 minutes (adjusted relative risk, 0.93; 95 percent confidence interval, 0.90 to 0.96) or to receive aspirin within 24 hours after arrival at the hospital (adjusted relative risk, 0.96; 95 percent confidence interval, 0.95 to 0.97), but they were equally likely to receive beta-blockers (adjusted relative risk, 0.99; 95 percent confidence interval, 0.95 to 1.03) and somewhat more likely to receive angiotensin-converting-enzyme inhibitors (adjusted relative risk, 1.05; 95 percent confidence interval, 1.02 to 1.08). Women were more likely than men to have a do-not-resuscitate order in their records (adjusted relative risk, 1.26; 95 percent confidence interval, 1.22 to 1.29). After adjustment, women and men had similar 30-day mortality rates (hazard ratio, 1.02; 95 percent confidence interval, 0.99 to 1.04). CONCLUSIONS As compared with men, women receive somewhat less aggressive treatment during the early management of acute myocardial infarction. However, many of these differences are small, and there is no apparent effect on early mortality.

Facemasks and Hand Hygiene to Prevent Influenza Transmission in Households: A Cluster Randomized Trial

Context Hand hygiene and use of facemasks are key elements of influenza pandemic preparedness plans, but their effects on preventing transmission of infection have not been demonstrated. Contribution In this cluster randomized trial, hand washing and facemasks seemed to prevent influenza transmission when healthy family members started using these measures within 36 hours of symptom onset in an infected family member. Caution Adherence to the interventions was low. Implication Hand hygiene and facemasks seem to reduce influenza virus transmission when implemented early after symptom onset. The Editors Interpandemic human influenza virus infects millions of people every year. Some infections are mild, but othersespecially in young or elderly personscan result in more severe illness requiring hospitalization. Influenza is associated with hundreds of thousands of deaths worldwide annually (1, 2). The 2009 swine-origin influenza A (H1N1) pandemic highlighted the importance of identifying public health measures to mitigate influenza virus transmission. Many countries would use nonpharmaceutical interventions, including facemasks, improved hand hygiene, cough etiquette, isolation of sick and quarantine of exposed individuals, social distancing measures, and travel restrictions, as their primary means to mitigate an influenza pandemic, particularly at its beginning (310). However, data are scarce on the effectiveness of simple personal protective measures, such as facemasks and hand hygiene, against pandemic or interpandemic influenza and on the modes of influenza virus transmission among people (5, 11). After a pilot study in 2007 (12), we conducted a prospective cluster randomized trial to test whether improved hand hygiene or surgical facemasks reduce the transmission of interpandemic influenza in households. We used a cluster design with randomization to interventions at the household level to avoid difficulties in blinding and potential contamination of interventions. Methods Design From 45 outpatient clinics in the private and public sectors across Hong Kong, we enrolled persons who reported at least 2 symptoms of acute respiratory illness (temperature37.8C, cough, headache, sore throat, or myalgia); had symptom onset within 48 hours; and lived in a household with at least 2 other people, none of whom had reported acute respiratory illness in the preceding 14 days. After participants gave informed consent, they provided nasal and throat swab specimens, which were combined and tested with the QuickVue Influenza A+B rapid diagnostic test (Quidel, San Diego, California). Participants with a positive rapid test result and their household contacts were randomly assigned to 1 of 3 study groups: control (lifestyle measures), control plus enhanced hand hygiene only, and control plus facemasks and enhanced hand hygiene. Table 1 provides detailed descriptions of the interventions. Data on clinical signs and symptoms were collected for all participants. An additional nasal and throat swab specimen was collected for laboratory confirmation of influenza virus infection by reverse-transcription polymerase chain reaction (RT-PCR). Table 1. Study Interventions Randomization lists were prepared by a biostatistician. The households of eligible study index patients were allocated to 3 groups in a 1:1:1 ratio under a block randomization structure with randomly permuted block sizes of 18, 24, and 30 by using a random-number generator (R software, R Development Core Team, Vienna, Austria). Interventions were assigned to households by the study manager on the basis of the randomization sequence. The allocation to specific intervention groups was concealed to recruiting physicians and clinics throughout the study. Participants and people who administered the interventions were not blinded to the interventions, but participants were not informed of the specific nature of the interventions applied to other participating households. After randomization, a home visit was scheduled within 2 days (ideally within 12 hours) to implement the intervention and to collect informed consent, baseline demographic data, and nasal and throat swab specimens from all household members 2 years of age or older. During the home visit, index patients and household contacts were instructed in the proper use of a tympanic thermometer. During the 6 days after the initial home visit, all household contacts were asked to keep daily symptom diaries. Further home visits were scheduled around 3 and 6 days after the baseline household visit to monitor adherence to interventions and to collect further nasal and throat swab specimens from all household members regardless of illness. During the final home visit, study nurses collected and reviewed symptom diaries, and they evaluated adherence to interventions by interview and by counting the number of surgical masks remaining and weighing the amount of soap and alcohol left in bottles and dispensers. Households were reimbursed for their participation with a supermarket coupon worth approximately U.S.

Viral Shedding and Clinical Illness in Naturally Acquired Influenza Virus Infections

25. All participants 18 years or older gave written informed consent. Proxy written consent from parents or legal guardians was obtained for persons 17 years or younger, with additional written assent from those 8 to 17 years of age. The study protocol was approved by the institutional review board of The University of Hong Kong and the Hospital Authority Hong Kong West Cluster. Outcome Measures The primary outcome measure was the secondary attack ratio at the individual level: the proportion of household contacts infected with influenza virus. We evaluated the secondary attack ratio by using a laboratory definition (a household contact with a nasal and throat swab specimen positive for influenza by RT-PCR) as the primary analysis and 2 clinical definitions of influenza based on self-reported data from the symptom diaries as secondary analyses (12). The first definition of clinical influenza was at least 2 of the following signs and symptoms: temperature 37.8C or greater, cough, headache, sore throat, and myalgia (13); the second was temperature 37.8C or greater plus cough or sore throat (14). An additional secondary outcome measure was the secondary attack ratio at the household (cluster) level: the proportion of households with 1 or more secondary case. Laboratory Methods Specimens collected from index patients at recruitment were stored in a refrigerator at 2 to 8C. Specimens collected during home visits were stored in an ice chest with at least 2 ice packs immediately after collection. Before the end of the day of a home visit, study nurses obtained samples to the nearest collection point for storage in a refrigerator at 2 to 8C. Samples stored at 2 to 8C in ice chests were delivered to the central testing laboratory at Queen Mary Hospital by courier. Samples were eluted and cryopreserved at 70C immediately after receipt. All specimens were tested by RT-PCR for influenza A and B viruses using standard methods (1517). The Appendix provides additional details of the laboratory procedures that we used. Statistical Analysis On the basis of data collected in our pilot study (12) and other studies with similar design (18, 19), we assumed that 10% to 15% of household contacts in the control group would develop RT-PCRconfirmed influenza, with an average household size of 3.8 and an intracluster correlation coefficient of 0.29. Specifying 80% power and a significance level of 5%, we aimed to follow 300 households in each intervention group to allow us to detect differences in secondary attack ratios of 35% to 45%, depending on the actual secondary attack ratios in the control group (15% or 10%, respectively). Recruiting 100 or 200 households to each group would allow 80% power to detect 55% to 70% and 45% to 55% differences in secondary attack ratios, assuming a secondary attack ratio of 10% to 15% in the control group. To evaluate and compare secondary attack ratios by intervention group, we estimated 95% CIs by using a cluster bootstrap technique with 1000 resamples (20) and chi-square tests and multivariable logistic regression models adjusting for potential within-household correlation (21, 22). We estimated the intracluster correlation coefficient from the mean squared errors in the secondary attack ratio between and within households (21). For the multivariable logistic regression models, we used forced-entry methods to include plausible confounders, including the intervention allocated, the age and sex of the household contacts and their corresponding index patients, vaccination status of the household contacts, and antiviral use in corresponding index patients, whereas missing data on the exact age of 14 household contacts were imputed by comparison with their relationship with the index patient or occupation. Participants were analyzed in the group to which they were randomly assigned, regardless of adherence to the intervention or use of hand washing or facemasks in groups not assigned that intervention. Our protocol specified that households with more than 1 member with RT-PCRconfirmed influenza virus infection at baseline (coindex patients) or index patients in whom influenza virus infection could not be confirmed by RT-PCR would be excluded from analyses. We excluded from analyses participants who dropped out before receiving the intervention and the few participants who dropped out after the intervention but before data on the primary outcome measure were collected (23). In sensitivity analyses, we analyzed all households in which the intervention was applied, using multiple imputation for unobserved outcomes (24) and including an additional explanatory variable for households with more than 1 index patient. Statistical analyses were conducted in R, version 2.7.1 (R Development Core Team). Role of the Funding Source The study was funded by the Centers for Disease Control and Prevention; the Research Fund for the

Preliminary Findings of a Randomized Trial of Non-Pharmaceutical Interventions to Prevent Influenza Transmission in Households

Abstract Background. Volunteer challenge studies have provided detailed data on viral shedding from the respiratory tract before and through the course of experimental influenza virus infection. There are no comparable quantitative data to our knowledge on naturally acquired infections. Methods. In a community-based study in Hong Kong in 2008, we followed up initially healthy individuals to quantify trends in viral shedding on the basis of cultures and reverse-transcription polymerase chain reaction (RT-PCR) through the course of illness associated with seasonal influenza A and B virus infection. Results. Trends in symptom scores more closely matched changes in molecular viral loads measured with RT-PCR for influenza A than for influenza B. For influenza A virus infections, the replicating viral loads determined with cultures decreased to undetectable levels earlier after illness onset than did molecular viral loads. Most viral shedding occurred during the first 2–3 days after illness onset, and we estimated that 1%–8% of infectiousness occurs prior to illness onset. Only 14% of infections with detectable shedding at RT-PCR were asymptomatic, and viral shedding was low in these cases. Conclusions. Our results suggest that “silent spreaders” (ie, individuals who are infectious while asymptomatic or presymptomatic) may be less important in the spread of influenza epidemics than previously thought.

Effects of Oseltamivir Treatment on Duration of Clinical Illness and Viral Shedding and Household Transmission of Influenza Virus

Background There are sparse data on whether non-pharmaceutical interventions can reduce the spread of influenza. We implemented a study of the feasibility and efficacy of face masks and hand hygiene to reduce influenza transmission among Hong Kong household members. Methodology/Principal Findings We conducted a cluster randomized controlled trial of households (composed of at least 3 members) where an index subject presented with influenza-like-illness of <48 hours duration. After influenza was confirmed in an index case by the QuickVue Influenza A+B rapid test, the household of the index subject was randomized to 1) control or 2) surgical face masks or 3) hand hygiene. Households were visited within 36 hours, and 3, 6 and 9 days later. Nose and throat swabs were collected from index subjects and all household contacts at each home visit and tested by viral culture. The primary outcome measure was laboratory culture confirmed influenza in a household contact; the secondary outcome was clinically diagnosed influenza (by self-reported symptoms). We randomized 198 households and completed follow up home visits in 128; the index cases in 122 of those households had laboratory-confirmed influenza. There were 21 household contacts with laboratory confirmed influenza corresponding to a secondary attack ratio of 6%. Clinical secondary attack ratios varied from 5% to 18% depending on case definitions. The laboratory-based or clinical secondary attack ratios did not significantly differ across the intervention arms. Adherence to interventions was variable. Conclusions/Significance The secondary attack ratios were lower than anticipated, and lower than reported in other countries, perhaps due to differing patterns of susceptibility, lack of significant antigenic drift in circulating influenza virus strains recently, and/or issues related to the symptomatic recruitment design. Lessons learnt from this pilot have informed changes for the main study in 2008. Trial Registration ClinicalTrials.gov NCT00425893 HKClinicalTrials.com HKCTR-365

Volume, quality of care, and outcome in pneumonia

BACKGROUND Large clinical trials have demonstrated the therapeutic efficacy of oseltamivir against influenza. We assessed the indirect effectiveness of oseltamivir in reducing secondary household transmission in an incident cohort of influenza index patients and their household members. METHODS We recruited index outpatients whose rapid test results were positive for influenza from February through September 2007 and January through September 2008. Household contacts were followed up for 7-10 days during 3-4 home visits to monitor symptoms. Nose and throat swabs were collected and tested for influenza by reverse-transcription polymerase chain reaction or viral culture. RESULTS We followed up 384 index patients and their household contacts. Index patients who took oseltamivir within 24 h of symptom onset halved the time to symptom alleviation (adjusted acceleration factor, 0.56; 95% confidence interval [CI], 0.42-0.76). Oseltamivir treatment was not associated with statistically significant reduction in the duration of viral shedding. Household contacts of index patients who had taken oseltamivir within 24 h of onset had a nonstatistically significant lower risk of developing laboratory-confirmed infection (adjusted odds ratio, 0.54; 95% CI, 0.11-2.57) and a marginally statistically significant lower risk of clinical illness (adjusted odds ratio, 0.52; 95% CI, 0.25-1.08) compared with contacts of index patients who did not take oseltamivir. CONCLUSIONS Oseltamivir treatment is effective in reducing the duration of symptoms, but evidence of household reduction in transmission of influenza virus was inconclusive.

Factors affecting QuickVue Influenza A + B rapid test performance in the community setting

Context Better outcomes have been associated with care by high-volume providers for many surgical procedures and medical conditions. It is not clear whether the same is true for other common conditions, such as pneumonia. Contribution Using data from the Medicare National Pneumonia Quality Improvement Project, these investigators found that after adjusting for severity of illness and other factors, there was no association between high physician or hospital volume and 30-day mortality rates. Adherence to guidelines was better and length of stay was shorter in low-volume hospitals. Implications High volume does not necessarily lead to better process of care or better outcomes in all clinical conditions. The Editors Approximately 1.2 million patients with pneumonia are hospitalized each year in the United States. Inpatient mortality rates average 5.8%, and these admissions result in more than

Performance measures for pneumonia: are they valuable, and are process measures adequate?

20 billion in direct health care costs (1). Opportunities to improve the care of patients with pneumonia have been well documented at both the state and national levels (2). For example, the timeliness of antibiotic administration and the selection of antibiotics are suboptimal, despite the dissemination of national guidelines (3-5). Moreover, essential hospital-based prevention strategies, such as the administration of influenza and pneumococcal vaccines, are markedly underused (6). In this context, improving the quality of care for patients with pneumonia remains a strategic priority in the United States. Continuous methods of quality improvement adapted from the U.S. manufacturing industry provide the basis for most ongoing efforts to improve care of patients with pneumonia. Common strategies used by hospitals involve the establishment of clinical practice guidelines, the development of standard order sets and reminder systems, and the use of measurement of and feedback on performance (7-11). During the past 2 decades, an increasing body of research has shown a positive correlation between hospital and physician volume and patient outcomes across a wide range of clinical conditions (12). This has not only been recognized within the medical community but has aroused the interest of both purchasers and payers of health care (13). Although surgical procedures have received more attention than medical treatment, there is evidence to suggest that patients with acute myocardial infarction, breast cancer, and HIV infection all benefit when their physicians treat more patients with these conditions (14-18). Although it may be impractical to regionalize the care of a common condition like pneumonia to high-volume hospitals, this is certainly feasible at the physician level. The growth of the hospitalist model concentrates the care of general medical patients into the hands of a small number of high-volume providers (19). In light of the potential for these organizational factors to impact patient care, we sought to determine whether physician or hospital pneumonia volume is associated with quality of care or outcomes. Methods Design, Setting, and Participants Details of the implementation of the Centers for Medicare & Medicaid Services (CMS) national quality improvement projects have been previously published (2). As part of its Medicare National Pneumonia Quality Improvement Project, CMS used Medicare fee-for-service hospital claims data to identify patients who were discharged from the hospital with a principal diagnosis of pneumonia (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 480 to 483.8, 485 to 486, or 487.0) during one of two 6-month periods: 1 July through 31 December 1998 and 1 September 1998 through 31 March 1999. As many as 850 patients were randomly selected from each state and from the District of Columbia, and all eligible patients were included if there were fewer than the targeted number of patients available. We excluded patients from the analysis if they were younger than 65 years of age, if pneumonia was not documented as a working diagnosis, or if the patient lacked radiographic findings consistent with pneumonia. Patients were also excluded if they died on the day of admission, received comfort measures only, left the hospital against medical advice, or were transferred to another acute care facility. We further limited our analysis to patients treated in acute care hospitals and those treated by an attending physician who was a general internist, family physician, or general practitioner. Informed consent and institutional review board approval were not required because CMS has statutory access to the medical records of Medicare beneficiaries. Data Collection Hospitals sent photocopies of the selected medical records to 1 of 2 CMS-contracted clinical data abstraction centers that used computerized data collection tools with explicit predefined criteria for manual chart review and data entry. Continuous monitoring of the quality of data collection through interrater reliability testing was done throughout the process for each clinical topic. Quality and Outcome Measures Quality-of-care measures were developed by the CMS in collaboration with national experts representing the American Thoracic Society, the Infectious Diseases Society of America, and the Centers for Disease Control and Prevention (11). These measures were extensively field-tested by state Quality Improvement Organizations and have been used widely for several years. Specific measures included the proportion of patients whose antibiotics were administered within 4 hours of hospital arrival, the proportion whose antimicrobial regimen was consistent with current guideline recommendations, the proportion in whom blood cultures were drawn before antibiotics were administered, and the proportion who were screened for or were given the influenza vaccine between October and December and who were screened for or were given pneumococcal vaccine year-round. Length of stay and in-hospital mortality rates were abstracted directly from the charts, and 30-day mortality rates were obtained from Medicare enrollment files. Physician and Hospital Volume Unique provider identification numbers associated with study patients were matched with all Medicare Part A claims that met study inclusion criteria during the same calendar year to calculate each attending physicians total annual Medicare inpatient pneumonia volume. The annual hospital pneumonia volume was calculated similarly. Statistical Analysis Summary statistics for the overall sample were constructed by using simple frequencies and proportions for categorical data and by using means, standard deviations, and medians for continuous variables. Physicians and hospitals were grouped into balanced quartiles of inpatient pneumonia volume. Within each quartile, we calculated the proportion of patients for whom each quality measure was done. Patients with inadequate documentation of dates or times for hospital arrival or process-of-care performance were excluded from analyses. Patients who did not receive antibiotics during the hospitalization, those who received antibiotics 36 hours or more after arrival, or those who had blood cultures drawn 24 hours or more before hospital arrival or after hospital discharge were also excluded. Chi-square tests were used to compare performance on quality-of-care indicators across quartiles of pneumonia volume at both the physician and hospital level. To account for the effects of clustering of patients within physicians and physicians within hospitals, the association among physician volume, hospital volume, and mortality rates was examined by using generalized estimating equations (SAS software, version 9.1, PROC GENMOD, SAS Institute, Inc., Cary, North Carolina). Because the optimal method of accounting for such effects remains uncertain, our analyses were repeated using 3-level multivariable hierarchical linear modeling for binary outcomes (HLM, version 6.0, Scientific Software International, Lincolnwood, Illinois) (20). These regression models included 3 sets of variables: level 1 or patient characteristics, including all of the components of the Pneumonia Severity Index (21); level 2 or physician characteristics, which was the physicians annual pneumonia volume; and level 3 or hospital characteristics, including teaching status, bed size, whether the hospital was located in an urban or rural setting, and the hospitals annual pneumonia volume. These 2 approaches produced nearly identical results, and we report the output of the generalized estimating equations only. Observations with missing hospital characteristics were omitted from multivariable analyses. There was little correlation between physician volume and hospital volume (r= 0.08), and the inclusion of an interaction term for physician and hospital volume of patients with pneumonia in the regression models yielded only marginal changes in the results. All analyses were done by using SAS software, version 9.1, and HLM, version 6.0. Role of the Funding Source The analyses on which this publication is based were performed under contract number 500-99-P619, titled Utilization and Quality Control Peer Review Organization for the State of Oklahoma, sponsored by CMS, Department of Health and Human Services. This article is a direct result of the Health Care Quality Improvement Program initiated by CMS, which has encouraged identification of quality improvement projects derived from analyses of patterns of care, and therefore required no special funding on the part of this contractor. The funding source had no role in the design, analysis, or interpretation of the study or in the decision to submit the manuscript for publication. Results Patients A total of 35347 patients with a principal diagnosis of pneumonia were identified on the basis of ICD-9-CM codes, and their charts were abstracted for the project. Chart review failed to confirm pneumonia as a working diagno

Detection of antimicrobial resistance by small rural hospital microbiology laboratories: comparison of survey responses with current NCCLS laboratory standards.

Rapid diagnosis of influenza can facilitate timely clinical management. We evaluated the performance of the QuickVue Influenza A + B test (Quidel, San Diego, CA) in a community setting and investigated the factors affecting test sensitivity. We recruited 1008 subjects from 30 outpatient clinics in Hong Kong between February and September 2007. Each subject provided 2 pooled pairs of nose and throat swabs; 1 pair was tested by the QuickVue rapid test on site, and the other pair was sent to a laboratory for reference tests. Among 998 enrolled subjects with valid results, the rapid test had overall sensitivity of 0.68 and specificity of 0.96 compared with viral culture. Sensitivity for both influenza A and B was significantly higher for specimens with viral loads greater than 5 log(10) copies/mL. The QuickVue Influenza A + B test has similar sensitivity in point-of-care community settings to more controlled conditions.