Jon Mark Hirshon, Angela C. Comer, Joseph H. Rosenberg, J. Kristie Johnson, Susan L. Moore, Thomas D. MacKenzie, Kendall K. Hall, Jon P. Furuno
Abstract
Although antibiograms have been observed to improve empiric antibiotic prescribing in hospitals, data on their effectiveness in nursing homes (NHs) are limited. In this study, our objective was to develop and implement antibiograms in three Maryland NHs. Data for antibiogram development were collected via chart review from NH residents during a 6-month period and included residents' characteristics, microbiological cultures, and antibiotic use. Additionally, we identified the primary emergency department (ED) to which NH residents were transferred and reviewed their acute-care charts for admission diagnoses and microbiological and antibiotic data during their first 48 hours of hospitalization. Finally, at each participating NH, an infection control or quality assurance nurse was interviewed as a key informant regarding resources for diagnosing and treating infections. Specific challenges faced in antibiogram development and implementation included identifying facility champions, availability of NH nursing staff and physicians, time required for individual chart review, and low volume of cultures available for antibiogram development. Lessons learned included:
- NHs may have a one-to-one or a one-to-many relationship with laboratories.
- Organisms in NH antibiograms mainly represented uropathogens.
- Incorporating culture data from residents transferred to EDs may not improve the quality of NH antibiograms unless there are sufficient numbers of transfers.
- Implementation of antibiograms in NHs required a significant investment of time and effort, including multiple educational inservices with staff and clinicians.
Development and implementation of nursing home antibiograms based upon chart reviews is possible but time consuming. Additionally, maintenance of the antibiograms requires consistent effort, which may be beyond the limited resources of many NHs. Identifying key, consistent project champions at NHs is important for successful development and implementation.
Introduction
Bacterial infections are a significant cause of morbidity and mortality among older adults residing in nursing homes (NHs).1–3 Available prevalence data are sparse, but it has been estimated that there are between 1.64 million and 3.83 million endemic infections per year in long-term care facilities, with an annual cost that exceeds $1 billion.1 These infections, and their associated poor health outcomes, are complicated by the increasing prevalence of antibiotic resistance among pathogens in NH residents, which in turn reduces treatment options and increases the probability of treatment failure.4 Treatment of infections in NHs is frequently initiated empirically (prior to definitive diagnosis based upon organism identification and antibiotic susceptibilities) because NHs often have limited resources for timely microbiological identification and antibiotic susceptibility testing. Although aggressive broad-spectrum empiric antibiotic therapy has been associated with improved clinical outcomes among infected patients, use of unnecessarily broad-spectrum antibiotics increases the incidence of antibiotic-associated side effects and creates selective pressure that promotes antibiotic resistance in surviving organisms. On the other hand, narrow-spectrum empiric therapy may not have sufficient coverage for the antibiotic susceptibilities of the infecting organism and, thus, may result in treatment failure and adverse outcomes, leading to increased potential morbidity and mortality.5,6
Antibiograms, which present a display of the cumulative antimicrobial susceptibility data of common bacteria isolated from a health care facility, are useful tools to guide empiric antibiotic prescribing.7 Antibiograms commonly display information structured in a summary table, with bacterial isolates shown along one axis and antimicrobial agents arranged along a second axis, as seen in Figure 1. Percentages in table cells indicate which bacterial organisms are either resistant or susceptible to each agent tested (susceptibility data). Antibiograms utilize clinical culture data from the index facility to delineate the ward- or facility-wide prevalence of antibiotic resistance. Thus, antibiograms help to aggregate and track laboratory results for strains of bacteria that may show decreased susceptibility to certain antibiotics over time. However, the development of antibiograms is labor intensive and requires laboratory and information technology resources. As a result, the use of antibiograms is primarily limited to acute care settings. However, implementation of antibiograms in NHs has the potential to improve antibiotic prescribing for suspected infections and should be considered despite the resource requirements.
Denver Health and Hospital Authority and the University of Maryland, Baltimore (UMB) partnered with the Agency for Healthcare Research and Quality (AHRQ) to (1) determine antibiotic susceptibilities for bacteria isolated in clinical cultures from NH residents in three Maryland NHs; (2) generate NH-specific antibiograms based on the collected data; and (3) develop a toolkit to aid NHs and affiliated laboratories in creating and maintaining NH-specific antibiograms. Further, we assessed whether NH-specific antibiograms could be implemented for use both within the facility and also be transmitted to local emergency departments (EDs) to impact the empiric management of presumed bacterial infections in NH residents. This paper focuses on the methodological challenges associated with meeting these objectives.
Figure 1. Example of nursing home antibiogram
a WHONET Software, World Health Organization. Available at www.who.int/drugresistance/whonetsoftware/en.
Methods
Overall Description
We collected antibiotic susceptibility data from 6 months of medical records for three NHs through retrospective chart review at the NHs and at each NH's primary receiving ED/acute care facility. Additionally, we conducted needs assessments in each NH prior to creating and implementing the antibiogram to gather information regarding infection control resources in each facility. The antibiograms were created using WHONET,a a free, Windows-based software program that was established for the purpose of analyzing microbiology data by the World Health Organization (WHO) Collaborating Centre for Surveillance of Antimicrobial Resistance, based at the Brigham and Women's Hospital in Boston, MA.8 Implementation was structured based on NH preferences. We presented the antibiograms to NH staff and physicians during several inservices at the facilities. Followup chart review was conducted in the first NH (NH-A), 6 months after the antibiogram was implemented. Key informant interviews were also conducted at NH-A and its related acute care facility to determine usability and usefulness of the antibiogram.
Facility Recruitment
The three NHs were chosen based on location, number of beds, and the facility's prior research experience with UMB (Table 1). NH-A was a rural NH with prior research experience with UMB. This site was chosen in order to work through the methodological and operational challenges of creating and implementing an antibiogram before expanding to two additional NHs. As an initial step, the study team obtained administrative approvals to proceed with the research from each facility's administrator and quality assurance nurse. Institutional review board (IRB) approval was first obtained from UMB for the overall study and then from each NH and acute care facility whose ED received the majority of medical transfers from the identified NH. For institutions without an IRB, the UMB IRB served as the IRB of record.
Table 1. Nursing home characteristics
Characteristic | NH-A | NH-B | NH-C |
---|---|---|---|
Number of Beds | 118 | 147 | 167 |
Dedicated Short Stay Beds? | Yes | Yes | No |
Nonprofit Status | For-profit | Not-for-profit | For-profit |
Location | Rural | Urban | Suburban |
Median Resident Age | 79 years | 84 years | 72 years |
% Female | 84 | 76 | 59 |
% Caucasian | 86 | 46 | 41 |
% African American | 10 | 51 | 53 |
Type of Laboratory Used | Hospital-based | Hospital-based | Private and hospital-based |
% Transfers to One Hospital | 79 | 98 | 48 |
Charting | Paper-based | Paper-based | Paper and electronic |
Medical Record Review
Data were collected via chart review using a standardized form for patients who resided in the NH during a specified 6-month period. Patients were included in the review if they resided at the NH the day chart review began or if they were at the NH within the previous 6 months and provided a clinical culture or were transferred to an acute care hospital for an acute medical reason, such as an infection. The parameters were guided by the information needed to create and evaluate the antibiograms, namely, culture results and transfers to the hospital that may have resulted in cultures.
All the charts were reviewed and data abstracted manually by one of the authors (ACC or JHR) at all three facilities. If a question occurred while abstracting the data, the two reviewers would confer. One NH (NH-C) had portions of its records available electronically, but all information extracted was handwritten to a data collection form. As part of the research, the chart reviewers recorded basic demographic information; type of infections occurring during the 6-month review period; infection treatment; signs and symptoms of infection; antibiotics prescribed; culture information including dates, antibiotic susceptibilities, and any hospitalizations in the 6-month period; and information about indwelling devices.
The acute care charts of hospitalized residents identified during the NH chart review were reviewed at the affiliated acute care facilities. We reviewed their charts for admission diagnoses and microbiological and antibiotic data during the first 48 hours of hospitalization.
Needs Assessments
We performed a qualitative needs assessment at each NH to examine the NH's organizational structure, particularly as it related to decisions about empiric antibiotic prescribing and to self-perceived concerns about this structure related to diagnosing and treating infections. This assessment was necessary in order to focus the antibiogram implementation phase such that antibiogram usage would meet the needs of staff at the NH and be sustainable. The quality assurance nurse at each NH, who also served as the infection control nurse, was interviewed for this purpose.
Through the needs assessment, we ascertained the handling of microbiological cultures in the NH, including the collection, charting, and result retrieval processes, as well as the estimated frequency with which cultures were ordered. The infection control nurse also provided information about how a resident was transferred to the ED, who made the decision to transfer, and what information accompanied the resident. Finally, the assessment gathered information about the antibiotic prescribing practices within the NH. The infection control nurse provided details about the types of antibiotics available on site, the typical symptoms an empiric antibiotic might be prescribed to treat, and the process used by the nurses to communicate culture results to the prescribing physician responsible for initiating an antibiotic.
Antibiogram Development
Data collected from both the NH and acute care hospital charts were entered into a Microsoft Access 2007 database (Microsoft Corporation, Redmond WA) designed for the project. The study microbiologist then used WHONET to display the data and to compute percentage susceptibilities to each antibiotic for each organism. These percentages were then manually entered into an antibiogram template matrix. Consistent with Clinical Laboratory Standards Institute (CLSI) performance standards, only the results for the first organism-specific positive culture per resident in the time period were included.9
Antibiotic susceptibilities were assessed for bacteria found in the NH and the associated acute care hospital independently as well as together. Since only acute care culture data obtained within the first 48 hours of admission were included, and thus likely originated from the transferring facility, we expected to see similar antibiotic susceptibility data in the hospital and NH charts.
Implementation
For each NH, we expected that three antibiograms (NH only, hospital only, NH–hospital combined) would be created based on the chart review data collected at the NH and the hospital (Figure 1). For two NHs, the hospital culture results were not combined with the NH culture results in an antibiogram because there were not enough hospital cultures to substantially augment the NH culture data. For NH-B, which transferred 98 percent of its residents to the same hospital and had a higher number of patients transferred to the single facility, the hospital cultures added significant information related to pathogen antibiotic resistance to the antibiogram. The combined antibiogram was implemented in this facility.
Implementation in NH-A was a multi-step process. The antibiogram was first circulated to the NH nursing team for feedback on the format of the antibiogram and how best to organize the information. An index card format was agreed upon, with gram-positive results on one side and gram-negative results on the other. We laminated the antibiogram on a 3×5-in. index card so that the antibiogram could be easily carried by physicians and nurses, placed in residents' charts, or posted around the facility. After discussions with the clinical and administrative staff, the NH staff decided to photocopy the antibiogram onto the back of the transfer forms to ensure that the antibiogram traveled with the residents to the ED. The final antibiogram was formally presented via an inservice with NH staff and separately to physicians, administrators, and nurse managers.
In NH-B and NH-C, the same index card format was used for the antibiograms. Multiple inservices were held with NH staff (nurses and administrators) and physicians in each facility to present the antibiograms and to discuss their implementation and utility. The inservices each lasted about 1 hour, during which the antibiogram was presented, and information about antibiogram use and infection control measures was discussed. Results of the initial chart reviews specific to each NH were highlighted during the inservices. These inservices were arranged through the NH administrators or the infection control nurses. Available staff were expected to participate but were not required to do so.
Evaluation
The study period only allowed for complete evaluation of the impact of antibiogram use on antibiotic prescribing and bacterial susceptibility trends at NH-A. Following implementation of the antibiogram, the same data collected at the beginning of the study were collected a second time for the evaluation, including basic demographic information; type of infections occurring during the 6-month review period; infection treatment; signs and symptoms of infection; antibiotics prescribed; culture information including dates, antibiotic susceptibility data, and any hospitalizations in the 6-month period; and information about indwelling devices. Antibiograms were recreated based on the followup data to observe if there were any changes in susceptibilities. Changes in culturing and antibiotic prescribing patterns were also examined.
Research study staff made numerous attempts to interview physicians at NH-A and the associated ED about their use and opinions of the antibiogram. These efforts included multiple telephone calls and emails to the quality assurance nurse at NH-A and to the clinical director at the associated RD. No NH physicians were available to participate, and only one ED physician, who had not seen the antibiogram despite multiple shifts in the ED, participated.
Toolkit Development
A toolkit was compiled for NHs to create antibiograms for their facilities. The toolkit consisted of several components: background information, instructions on obtaining culture data and updating the antibiogram, instructions to supplement the WHONET tutorials, templates for data entry and antibiogram structure, and instructions on entering data from WHONET into the antibiogram. The toolkit was revised with input from the infection control nurse at NH-A, who planned to update the antibiogram within the year. The toolkit was developed to be publishable online, with the elements downloadable using Microsoft Word and Excel (Microsoft 2003).
Results
All three NHs involved in this study were located in Maryland (Table 1). We consider this group of NHs to be diverse but not necessarily representative of all NHs in Maryland or in other States. The NHs differed in type of location—rural, urban, and suburban—and their populations differed in distribution by sex and race. NH-A and NH-B used the laboratory of the hospital to which they transferred the majority of their residents as the central laboratory that processed their cultures. NH-C used a private central laboratory during the week and a hospital-based laboratory on the weekend. All three facilities employed a full-time nurse who served as both the quality assurance nurse and the infection control nurse.
A total of 623 charts from the three NHs and 216 charts from the three associated acute care hospitals were reviewed. This proved to be the most time-consuming portion of the study. Archived records had to be retrieved for residents who had died or were no longer at the NH. Identifying signs and symptoms related to antibiotic prescribing proved to be the most difficult element of the review in both the NH charts and the hospital charts.
The needs assessments revealed several similarities in culturing practices across all three facilities: urine cultures were by far the most common type of culture ordered, and wound cultures were rarely ordered. Laboratory results were received by fax in all three facilities, and the results were manually entered in the residents' charts.
We were able to create antibiograms for all three NHs. The amount of culture results available for the antibiograms based on our 6-month chart review period was much less than recommended by the CLSI guidelines. However, the facilities were able to see susceptibility data for the few organisms that were included on their specific antibiogram. The antibiograms were successfully implemented in all three facilities using multiple inservices.
Discussion and Project Challenges
There were a number of methodological and operational barriers to overcome in the development of NH-specific antibiograms. These can be broken down into domains corresponding to project development and implementation steps. These steps include facility recruitment, data collection, antibiogram development, antibiogram implementation, and project evaluation (Table 2). Each step provided unique challenges—some foreseeable and some not.
The recruitment of facilities required multiple telephone and email contacts to obtain both administrative and IRB approval. The responsiveness of administrators and contact personnel varied, and it was not uncommon to have to work with several individuals initially before a primary institutional contact was identified. Because this was a research project, IRB approval was required from the university, the NHs, and the acute care facilities. Not all facilities had the same ability to conduct research; for example, one NH required assistance in renewing its Federalwide Assurance for the Protection of Human Subjects (FWA). The overall approval process required between 2 and 5 months for each NH.
Data collection also presented challenges. As was expected, medical chart review was time consuming and required approximately 45 minutes on average for each chart. The amount of time varied, depending on the format (paper or electronic) and the availability of the medical charts. While this study utilized experienced chart reviewers and dedicated research resources, the time required for medical chart review highlights feasibility concerns with regard to dependence on medical chart review by NH personnel for antibiogram creation as part of routine clinical practice.
The needs assessment at the three facilities uncovered some interesting results. First, the NHs had variable relationships both with acute care hospitals and with testing laboratories. NH-B was closely affiliated with a hospital and its laboratory, so their records were consistent. Another facility, NH-C, transferred patients to different hospitals depending on the potential diagnosis (one hospital for medical patients and a different hospital for patients deemed to have a psychiatric problem). In addition, this NH used one laboratory during the week and another on the weekend.
The actual development of the antibiogram required the ability to use WHONET software. While this program is publicly available and relatively easy to use, there is still a learning curve required to become proficient with its use. Additionally, time is required to input the data, which may be challenging for quality assurance nurses who already have multiple roles in the NH. The low number of culture results from the NHs also made it difficult to create antibiograms with truly reliable facility-specific antibiotic susceptibilities. CLSI guidelines recommend including a pathogen on an antibiogram if 30 or more isolates are available. The available NH culture results would not have produced an antibiogram for any facility if this recommendation had been followed, due to the small number of cultures obtained within the 6-month study period. Our microbiologist suggested including an organism if four or more isolates were available. This may have impacted the reliability of the antibiograms. However, even with this smaller number of cultures, facilities would have been able to see susceptibility trends over time. The acute care facility cultures only augmented the NH antibiogram for one NH (NH-B). Looking at data from a longer period of time, perhaps 1 or 2 years as compared to 6 months, might provide more comprehensive culture information. Further, the NHs in this study mainly ordered urine cultures, resulting in antibiograms primarily for uropathogens. Antibiotic susceptibility patterns for other pathogens will be relatively unknown without a change in culturing practices in NHs.
Table 2. Challenges in antibiogram development and implementation
Steps | Challenge/Obstacle | How Resolved and Effort Needed |
---|---|---|
Facility Recruitment | ||
Permission to participate | Obtaining administrative approval | Multiple phone and email contacts were required to schedule a meeting to present the project to the appropriate individuals. The nursing home (NH) administrator had to give the final approval, but his/her availability was very limited. For NH-A and NH-C, only one meeting was required; however, for NH-B, multiple phone contacts and two meetings were required, with a span of 2 months between the meetings. |
Institutional Review Board (IRB) approval | Conducting research in NHs without IRBs or Federalwide Assurances (FWA) and multiple IRBs for acute care facilities | Approval was obtained from four IRBs. Time for each approval ranged from 1 week to 3 months. Renewing the FWA for NH-C took 6 weeks. |
Medical Record Review | ||
Manual review of NH medical charts | NHs did not have patient infection and culture data available in a format to be shared electronically—mainly paper records. | Baseline chart review took 4 months at NH-A (150 hours, over 1 hour per chart) but decreased to 6 weeks at NH-C (120 hours, about 45 minutes per chart). The followup chart review at NH-A required 6 weeks (85 hours—about 40 minutes per chart). |
Manual review of acute care (AC) hospital medical charts | Acute care hospital records were electronic, but the infection and culture variables still had to be abstracted. | Baseline chart review took 3 weeks at AC 1 (40 hours—about 1 hour per chart), 5 weeks at AC 2 (80 hours—about 45 minutes per chart), and 2 weeks at AC 3 (20 hours—about 45 minutes per chart). For the followup chart review at AC 1, 1 week (20 hours—about 35 minutes per chart) was required. |
Needs Assessments | ||
Conducting interview | Scheduling an interview with the quality assurance nurse depended on his/her availability. This person was very busy and not always on site at the NH. | Scheduling this interview required multiple phone and email contacts with the nurse. Only one interview lasting about 1 hour was necessary for the assessment. |
NH and laboratory relationships | Two NHs had one hospital laboratory that processed all cultures for the facility. NH-C had two laboratories that processed its cultures. NHs were not always clear whether their affiliated laboratory could produce an antibiogram for their facility. | For a NH to maintain and update its facility's antibiogram, it may consider having its affiliated laboratory create the antibiogram. A poor relationship between the two entities makes this difficult. Dealing with two laboratories increases complications. |
Antibiogram Development | ||
Data entry for analysis | Manually abstracted data from medical records had to be entered into a database for analysis. | The study team spent about 45 hours inputting and resolving the handwritten chart review forms. |
WHONET software | WHONET software required a learning curve to import data appropriately and to create the antibiogram results as needed. | Study team members learned to create antibiograms in WHONET for the NHs to better inform the toolkit that was developed. This involved following tutorials on the WHONET site as well as using test data to create sample antibiograms. |
Clinical Laboratory Standards Institute (CLSI) guidelines for antibiograms | CLSI guidelines suggest a minimum of 30 isolates per organism to be reported in an antibiogram. Nursing homes do not culture to this extent. | Following the recommendation of the team microbiologist, organisms with more than four isolates were included, which can affect the reliability of the results. |
Prevalence of urine cultures | Urine cultures were collected far more often than other types of cultures in NHs. | The antibiogram's effectiveness may be strongest with uropathogens. |
Contribution of AC cultures to NH antibiograms | For NHs with few monthly transfers to the ED, the culture results from the acute care hospital did not augment the NH antibiograms. | Antibiograms for NH-A and NH-C included only NH cultures and resulted in low numbers of isolates. NH-B had many more isolates because ED cultures were included. |
Implementation | ||
Presenting antibiogram results to NHs | After the antibiogram was created, it took multiple inservices at each nursing home to present the antibiogram to all the interested parties. | At least two inservices were necessary to present the antibiogram at each NH. |
Transferring antibiogram to the ED | Each nursing home had to determine a method of transporting the antibiogram with patients to the ED. | NH-A was able to photocopy the antibiogram to the back of its transfer sheet, but this was not possible for the other NHs. The antibiogram had to be sent as a separate document. |
Evaluation | ||
Interviewing NH physicians | No NH physician was available to provide feedback about his/her antibiogram use. | 10–15 attempted contacts per NH physician were not enough to schedule an interview for this project. Anecdotal input from NH staff suggested that the physicians did use the antibiogram. Evaluation of the medical charts 6 months after the antibiogram was implemented also showed some change in prescribing patterns. |
Interviewing ED physicians | One ED physician was available for an interview, but he had not seen an antibiogram from the NH. | Study staff approached ED staff prior to the antibiogram intervention at the NH to alert them that the antibiogram would be sent with the patients. Only ED nurses were available at the time. The ED director also sent a notice to all the practitioners. Since only one ED physician was available to be interviewed, it is not clear if others saw the NH antibiogram. However, evaluation of the ED charts did not show any change in prescribing patterns. |
The implementation and evaluation of the antibiograms required multiple inservices and discussions at each NH. The most successful implementation occurred at NH-A, where we were invited to present to the doctors and senior staff at a facility event. As an innovative strategy to improve communication between NHs and their affiliated EDs, the antibiogram was replicated on the back of the transfer sheet that was to accompany any transferred patient. Our ability to evaluate the effectiveness of the antibiogram to improve antibiotic prescribing was limited, due in part to lack of time because of project completion and to difficulty in contacting responsive physicians. The one full-time emergency physician we were able to contact at the acute care facility affiliated with NH-A had not seen the antibiogram, despite working multiple shifts during the 6 months of antibiogram implementation.
This study highlights the challenges in adoption and dissemination of NH antibiograms, as well as the day-to-day challenges associated with communication in clinical settings. Specific challenges included identifying facility champions, availability of NH nursing staff and physicians, time required for individual chart review, and low volume of cultures available for antibiogram development. Lessons learned include the following:
- NHs may have a one-to-one or a one-to-many relationship with laboratories.
- Organisms in NH antibiograms mainly represented uropathogens.
- Incorporating culture data from residents transferred to EDs may not improve the quality of NH antibiograms unless there are sufficient numbers of transfers.
- Implementation of antibiograms in NHs required a significant investment of time and effort, including multiple educational inservices with staff and clinicians.
Conclusion
Through this project, significant barriers to antibiogram development and implementation were identified, including complex nursing home–clinical laboratory relationships and difficulties in ensuring effective utilization of antibiograms. As a research project, the development of nursing home antibiograms based on chart reviews is possible, but it is time consuming and resource intensive. Alternatives to chart review, such as having laboratories affiliated with the NHs create antibiograms as part of their clinical contract, should be considered. Additionally, maintenance of the antibiograms will require consistent effort, which may be beyond the limited resources of many NHs. However, the project also had the unexpected benefit of having NH administrative and clinical staff focus on the issue of empiric antibiotic prescribing practices, which led to increased awareness of the issue of antibiotic resistance. Identifying key, consistent partners will be important for successful development and implementation of risk reduction strategies for healthcare-associated infections.
Acknowledgments
This project was funded under contract number HHSA290200600020I, ACTION Task Order 9, "Using Nursing Home Antibiograms to Improve Antibiotic Prescribing and Delivery," from the Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health and Human Services. The findings and conclusions in this document are those of the authors, who are responsible for its content, and do not necessarily represent the views of AHRQ. No statement in this report should be construed as an official position of AHRQ or of the U.S. Department of Health and Human Services.
We acknowledge and thank Dr. Nimalie Stone from the Centers for Disease Control and Prevention, Atlanta, GA, for technical support with this project.
Authors' Affiliations
University of Maryland School of Medicine, Baltimore, MD (JMH, ACC, JHR, JKJ). Denver Health and Hospital Authority, Denver, CO (SLM, TDM). Agency for Healthcare Research and Quality, U.S. Department of Health and Human Services, Rockville, MD (KKH). College of Pharmacy, Oregon State University/Oregon Health & Science University, Portland, OR (JPF).
Address correspondence to: Jon Mark Hirshon, M.D., M.P.H., Ph.D., University of Maryland School of Medicine, 110 South Paca Street, Room 4S-127, Baltimore, MD 21201; Email: jhirs001@umaryland.edu.
References
1. Strausbaugh LJ, Joseph CL. The burden of infection in long-term care. Infect Control Hosp Epidemiol 2000 Oct;21(10):674-9. PMID: 11083186.
2. Moro ML, Jans B, Cookson B, et al. The burden of healthcare-associated infections in European long-term care facilities. Infect Control Hosp Epidemiol 2010 Nov;31 Suppl 1:S59-62. PMID: 20929373.
3. Heudorf U, Boehlcke K, Schade M. Healthcare-associated infections in long-term care facilities (HALT) in Frankfurt am Main, Germany, January to March 2011. Euro Surveill 2012 Aug 30;17(35):pii=20256. Available at: www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20256.
4. van Buul LW, van der Steen JT, Veenhuizen RB, et al. Antibiotic use and resistance in long term care facilities. J Am Med Dir Assoc 2012 Jul;13(6):568.e1-13. PMID: 22575722.
5. Cosgrove SE, Carmeli Y. The impact of antimicrobial resistance on health and economic outcomes. Clin Infect Dis 2003 Jun 1;36(11):1433-7. PMID: 12766839.
6. National Nosocomial Infections Surveillance System. National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2004, issued October 2004. Am J Infect Control 2004 Dec;32(8):470-85. PMID: 15573054.
7. WHO Collaborating Centre for Surveillance of Antimicrobial Resistance. WHONET. Available at: www.whonet.org/DNN. Accessed April 1, 2013.
8. Pakyz AL. The utility of hospital antibiograms as tools for guiding empiric therapy and tracking resistance. Insights from the Society of Infectious Diseases Pharmacists. Pharmacotherapy 2007 Sep;27(9):1306-12. PMID: 17723084.
9. Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing; Twenty-First Informational Supplement. CLSI M100-S21. Wayne, PA; 2011.