The Practice of Epidemiology in Community Hospitals

August J. Valenti

The methods of epidemiology and infection control are broadly applicable regardless of the size, location, or affiliation of a healthcare facility. The day-to-day practice of infection prevention in most community hospitals is normally more a matter of developing and implementing policies, educating staff, applying appropriate isolation precautions, conducting surveillance, and responding to regulatory and accrediting agencies’ requirements than engaging in the kind of science common to larger academic centers. The success of infection preventionists (IPs) in any hospital depends not only on the training and dedication of those managing infection control and the resources allocated to their efforts, but also on the strength of their own interpersonnel skills, their antibiotic stewardship programs, their clinical microbiology laboratories, the availability of healthcare quality expertise, and their access to the decision-making apparatus of their institutions.

The patient safety movement appropriately regards healthcare-associated infections (HAIs) as preventable adverse events. The “culture” of patient safety compels healthcare facilities, regardless of size, to improve their collection, analysis, and feedback of HAI data. Regulatory agencies, such as the Department of Health and Human Service’s Centers for Medicare and Medicaid Services (CMS) and accrediting bodies (particularly the Joint Commission [TJC] (1), have responded to pressures to reduce errors and adverse outcomes in hospitals by giving greater attention to process and quality indicators as well as outcome measures related to HAIs. Increasingly, CMS reimbursement is tied to certain HAIs, with hospitals being denied payment for hospital-acquired catheter-associated urinary tract infections (CA-UTIs), central line-associated bloodstream infections (CLA-BSIs), and certain surgical site infections (SSIs). Such measures are expanding.

The appetite for healthcare quality data should motivate hospital leadership to pay greater attention to their HAI rates and to seek appropriate benchmarks for comparisons. Along with these changes, there has been a growing tendency toward transparency in disclosing hospital-associated adverse events. Naturally, this requires data gathering and graphic display in ways that are accessible and understandable to consumers as well as providers. The electronic medical health record and data-mining programs often come with the promise of reducing the work of surveillance and improving compliance, but tailoring these systems to the particular needs of an institution (or healthcare system) takes time, money, and expertise. Unfortunately, there are as many ways to report and display data as there are groups within and outside of hospitals who wish to see it. These demands impact all hospitals, but the challenges for rural and smaller community hospitals may not be the same as for larger institutions.

Until recently, surprisingly little had been published on infection control practices in the community despite the fact that community hospitals provide more of the healthcare in the United States than larger, tertiary care centers (2). Evidence of increasing interest in the state of infection control in community hospitals is reflected by a growing literature examining how these infection control programs are configured and how they are responding to the challenges of resource availability, drug-resistant organisms, SSIs, Clostridium difficile infection, and antibiotic stewardship (3).

The most current available data reveals that nongovernment, not-for-profit community hospitals accounted for 58% of the 4,973 acute care hospitals in the United States (4). As hospitals’ bed numbers decrease, the percentage of those hospitals classified as rural markedly increases. Overall, 40% of the 4,973 hospitals were classified as rural and roughly 60% as urban. Most of the rural hospitals had ≤99 beds. The types of services offered by rural hospitals vary significantly by bed size, with more services offered by hospitals with the greater number of beds (5).

There is a paucity of data from these hospitals in the area of studies of infectious diseases, HAI prevention and control, as well as patient safety (7,8,9,10). Fortunately, the National Nosocomial Infections Surveillance (NNIS) system of the CDC always included a number of such community hospitals, but before 2006, >2,300 hospitals with <100 beds could not participate in NNIS. The NNIS was merged into the National Healthcare Safety Network (NHSN) in January 2006. This Web-based system is making access to data entry and analysis for all hospitals more accessible and useful.

A collaborative set of evidence-based recommendations published in February 1998 (11) remains useful for targeting some key discussion points. In March 2006, the authors of the previous edition of this chapter conducted a survey at the Society for Healthcare Epidemiology of America (SHEA) in Chicago inviting colleagues in hospital epidemiology to share their top issues and key frustrations with us. A similar survey was conducted at the May 2006 statewide Wisconsin Association of Professionals in Infection Control and Epidemiology (APIC) meeting. For over 5 years, the Maine Infection Prevention Collaborative (a consortium consisting of the Maine Hospital Association, the Maine CDC, the Northeast Health Care Quality Foundation, hospital epidemiologists, IPs, and administrators
representing every hospital in the State) also conducted surveys and gap analyses (see an example of a gap analysis survey in Appendix 1) for all participating hospitals in Maine that echo the concerns of IPs and hospital epidemiologists expressed in other “convenience sample” surveys.

To be sure, more extensive data and gap analyses in infection prevention programs are needed, but the consistency of concerns expressed in these surveys by IPs and hospital epidemiologists should encourage TJC in its efforts to foster a culture of safety, which must include a team of experts in HAI prevention and control, an absolute requirement and responsibility of the hospital administration and leadership (Table 12.1). Lack of time, an increasing scope of work, a frequent lack of administration-provided resources, and sense of value highlight the results. A few quotes from the Wisconsin data are relevant. A common thread in the era of pandemic influenza and bioterrorism preparedness concerns the lack of time (hours, staff, or both): “I have the desire to be proactive rather than always reactive, but there is not enough time.…” The frustration with lack of administration support is articulated best from two quite different hospitals. From a 600-bed hospital: “The Infection Control Department is perceived as a thorn in their (administration’s) side instead of an important component of healthcare,” and from a 100-bed facility: “Infection control has always been seen as a (cost center) rather than an income-producing department.”

Our informal surveys underscore some of the findings published in a formal survey conducted among Volunteer Hospitals of America (VHA) hospitals in various regions of the country by Christenson et al. (12). These authors conducted a demographic survey of 31 hospitals ranging in size from <50 beds to >500 beds to assess the staffing, structure, and functions of infection control departments in participating hospitals. Participants were asked to conduct observational studies of compliance with process measures, such as hand hygiene practices, and ventilator-associated pneumonia (VAP), CLA-BSI, and CA-UTI rates. One-third of the participants reported levels of infection control staffing below the level of one IP per 100 occupied beds. Only one hospital reported data entry and analytic support within the infection control department. The results of the process observations showed variability in compliance with evidence-based strategies for preventing HAIs. Sustained compliance with best practices frequently requires repeated educational interventions, and one can expect varying degrees of availability of trained IPs performing these interventions to impact results. As Christenson et al. note, the size of the study imposes limitations on the conclusions one can draw, but it is hoped that such studies will encourage broader examinations of infection control practices and contribute to evidence-based practices aimed at better control of HAIs globally. Clearly, more study is needed to examine whether there is a correlation between staffing and compliance with evidence-based practices and, ultimately, outcomes.

TABLE 12.1 The Top Issues and Frustrations of Infection Prevention and Control Team Members in 2006

Top Issues		Top Frustrations
1.	Collecting, interpreting, and using HAI infection surveillance data	1.	Need to find enough time to do multiple tasks and the ability to usefully and appropriately prioritize those tasks. Dealing with the ever-increasing scope of work
2.	Getting HCWs, especially nurses and doctors, to consistently use evidence-based guidelines, especially for isolation and hand hygiene	2.	Lack of administrative understanding, support, and buy-in of the central role of HAI prevention and control as the paradigm for patient safety
3.	Having the problem of MDROs and the need for appropriate antibiotic use and surgical prophylaxis	3.	The ability and willingness of the HCWs, including both doctors and nurses, to use policies correctly and understand the benefits of those policies for their patients
4.	Trying to use and keep up with new evidence and guidelines to create useful and up-to-date policies, procedures, and care measures that the HCWs will understand and use
5.	Addressing the increasing need for both old and new disease reporting, planning for new crises, and potential outbreaks and still doing the preceding 1-4
Source: 94 responses from hospital epidemiologists and infection control professionals at the March 2006 SHEA meeting and the May 2006 Wisconsin APIC Chapters Meeting.

Even with data, it remains to be seen whether hospital infection prevention programs will receive adequate resources. In October 2006, Wright et al. surveyed the SHEA membership in an attempt to develop data on the responsibilities, resources, and compensation for members and their departments. Their conclusions suggest that resources for healthcare epidemiology and infection control were below those recommended in older as well as newer studies, surveys, and expert panels published in the literature. They conclude that “The quickly evolving external mandates that have arisen since our survey have placed additional strain and expectations on our already resource-scarce departments.” (13). Furthermore, the effects that the changing financial landscape of healthcare as well as the increasing interest of government (local and federal) in exercising control over HAIs will have on infection prevention programs are uncertain.

The sundry internal and external forces facing infection control (IC) programs in community hospitals prompted Anderson and Sexton to comment forcefully on the challenges facing the programs in community hospitals and their
survival (14). On the basis of research within their network of 39 hospitals affiliated with the Duke Infection Control Outreach Network (DICON), they cite the following as three of the most important factors facing IC programs in community hospitals:

Deficits in administrative leadership
- IC a low priority
- Frequent leadership changes that “stall” the progress of IC programs when hospital infrastructure changes
Lack of specialization
- Reliance on volunteer services of busy physicians
- Inadequate expertise
  - Even among ID physicians who may not be trained in hospital epidemiology
- Lack of reimbursement for hospital epidemiologists
Limited numbers of personnel despite the rise in the type and severity of problems facing IPs
- Do not meet minimum standards of staffing
- Having duties outside of IC more common among ICPs in community hospitals as compared with those in tertiary hospitals
- Lack of support staff
- Lack of resources for continuing education

Good surveillance and infection control activities reduce multidrug-resistant organism (MDRO) rates in acute care hospitals, yet two surveys of Canadian hospitals (where MDROs are less a problem than in the United States, but are, nonetheless, on the rise) revealed that effective surveillance and control activities are not in place in many Canadian hospitals (15,16,17). In 2000, a survey of 72% of Canadian hospitals with >80 acute care beds revealed that there was <1 IP per 250 beds in 42% of hospitals, and only 60% of infection control programs had physicians or doctoral professionals with infection control training. SSI rates were provided to surgeons in only 37% of hospitals. A follow-up study indicated that surveillance and control activities in acute care hospitals in Canada are being performed in roughly two-thirds of the 120 hospitals responding to a survey.

It should be readily apparent from the preceding discussion that the reasons that infection control programs in many community hospitals are struggling to implement best practices are complex and certainly worthy of further study. We propose a number of recommendations that we hope will be useful to small hospitals facing these challenges.

Most of the top issues in Table 12.1 are discussed in detail in other chapters of this book. Scheckler et al. have articulated the principal goals for infection control and epidemiology (11). More recently, Cook, Marchaim, and Kaye published widely applicable and sound guidance for developing a successful infection prevention program, which hospitals of any size or scope will find helpful (18). They emphasize the importance of a written mission statement, establishing and documenting a vision (future goals) and core values (the day-to-day operation of the program). The following goals, adapted from Sheckler et al., are helpful when developing these three essential components of a program:

Protect patients.
Protect healthcare workers (HCWs), visitors, and others in the healthcare environment.
Accomplish these two goals in a cost-effective manner, whenever possible.

They also note both the value and necessity of measuring the effectiveness of the procedures, policies, and programs put in place to accomplish the three goals. Most hospital epidemiologists agree that, as with medications, conducting prospective, controlled trials of a current procedure compared with a newer procedure is optimal. The outcome of most interest, of course, is the HAI rate targeted by the procedure. Rarely can studies like the medication studies using a double-blind controlled method be done. And in small hospitals, small denominators make this especially difficult. Complicating HAI studies further is that outbreak situations and endemic situations may not show the same results in separate studies because the underlying issues may be different. At times, these realities lead to a conflict in recommendations (19,20,21).

We recommend a careful review of this article by our epidemiologist and IP colleagues in community hospitals (11). Our review of the article indicates that the discussion and the 23 recommendations are still useful. Since the publication of this paper in 1998, additional studies have supported and increased the evidence supporting these recommendations.

KEY ISSUES IN COMMUNITY HOSPITALS

ADEQUATE HAI PREVENTION AND CONTROL TEAM

Who, How Much Time, and How Many. The 1999 Institute of Medicine report To Err Is Human (22) brought the concept of patient safety, or lack of it, to the forefront of public discourse. Lost in this report, however, was the long-standing paradigm of HAI prevention; it was barely mentioned. A subsequent paper showed the value of infection control in the safety push (23). As our colleagues indicate, lack of time, expanding responsibilities, and support are their principal challenges. As Cook et al. point out, the IP, physician epidemiologist (if available), a data analyst, and an administrative assistant are the core elements of the infection prevention team (18).

IPs are usually nurses (but not exclusively) with special training and certification, preferably by APIC. They are primarily responsible for the day-to-day functioning of the infection control program in most hospitals. They are especially important where there is no trained physician epidemiologist. They should have managerial and information technology skills as well as a good working knowledge of microbiology, infectious diseases epidemiology, and quality improvement. In small hospitals, they may play a role in employee health, quality, and safety.

According to one study by Stone et al. (24), IPs spend only 13% to 15% of their time in preventative roles, such as teaching, isolation, and preparing policies. The majority of the IPs’ time (44.5%) is spent on surveillance activities. Because of the surveillance, data entry, and analysis responsibilities of IPs, it is now more common for infection control activities to be decentralized, with surveyors on nursing units or in clinical departments acting either independently or as liaisons of the infection control program.

Another important study on staffing using the Delphi method reported similar results with respect to the predominance of
surveillance activities (39% of IPs’ time) (25). The Delphi project’s expert panel recommended that there be 0.8 to 1.0 IPs per 100 occupied beds. It was also pointed out that more than just bed numbers should influence decisions about staffing.

Smaller hospitals may not have a physician epidemiologist (usually a specialist in infectious diseases with epidemiological training) available on site. As the Delphi project demonstrates, the complexity of modern healthcare has made simple calculations of the full-time equivalent (FTE) size of the HAI prevention and control team difficult. Scheckler et al. did not use a number. The literature has no recommendations correlating bed size or discharges with hours needed for a physician hospital epidemiologist.

Our recommendations, which follow, are based on our own extensive experience, our review of the literature. The reality of increased reporting requirements and responsibilities such as reviewing renovation and new building projects or planning for natural or man-caused disasters were considered in our recommendations.

Recommendations

For every 100 staffed beds in an acute care hospital, there should be at least one FTE trained IP and at least 4 hours of a paid physician hospital epidemiologist.
The smallest hospital should have the services of an IP for at least 8 hours a week (20% time), which should include some time—at least 3 days a week—at the hospital. Likewise, a physician hospital epidemiologist should be accessible to even the smallest institution by e-mail and telephone, and should be paid for the time spent consulting.
Adequate information technology (IT) support, Internet access, and time for continuing education are essential for the IPs no matter what the size of the hospital.
With acute care hospitals of 200 staffed beds, one FTE of staff support—secretarial and/or medical record specialist—must be available. This person would function as an administrative assistant in facilitating the role of the IPs and physician epidemiologist.

These recommendations are bare-bone essentials. Ample evidence indicates that a proactive HAI prevention team can substantially help reduce adverse events in the patients the hospital serves. The critical issue is making a convincing business case for a program’s needs. SHEA has published valuable guidelines (25) for developing a cogent business argument for IC-related proposals. Cook et al. have summarized these and other useful suggestions from the literature for making a strong business case (18). These exercises are of particular value to small hospitals with limited resources. They emphasize the importance of soliciting feedback and garnering support from administrators, key stakeholders, and institutional business leaders prior to framing a proposal.

Administrative Understanding and Support. TJC attempted in its 2006 requirements for HAI prevention and control and patient safety to require that the administrative authorities of the hospital—chief executive officers (CEO) and board—support the staff and resources to do the job. The fact that these requirements appear to have made little impact in our surveys and our experience with many hospitals suggest that they are either too new to be effective or that they are low on TJC’s surveyors’ priority list. There may also be reluctance on the part of IPs and physician epidemiologists to risk getting marked down on a survey by reporting lack of administrative support, which might rebound to the detriment their programs. Finally, the lack of business expertise on the part of infection prevention personnel may hinder their efforts to effectively communicate their needs to business-oriented administrators.

Recommendations

Every acute care hospital must have in place a budget plan for HAI prevention and control. This plan should include the IP and hospital epidemiologist positions and time required per our four recommendations noted in the preceding section.
TJC must articulate and enforce its new requirement concerning hospital administrative responsibility in the area of HAI prevention and control as an integral part of patient safety.
Patient safety activities and quality-improvement initiatives in a hospital must complement the necessary roles of IPs and hospital epidemiologists. The decades of experience in HAI prevention and control and the science base developed should be used by those responsible for overseeing quality and safety in our hospitals. Similarly, IPs and hospital epidemiologists also should recognize the value of well-trained quality professionals in improving safety and outcomes.

CHANGING HUMAN BEHAVIOR—AN ESSENTIAL COMPONENT OF INFECTION PREVENTION

There is not enough data to say with certainty that smaller hospitals have an easier time getting employees and medical staff to accept appropriate vaccinations, postexposure follow-up recommendations, proper use of personal protective equipment (PPE), and the rationale for policies and procedures related to infection prevention and control. However, there is universal agreement that hand hygiene remains a problem in healthcare facilities of any size.

Hand hygiene is essential before and after contact with a patient or his or her immediate environment whether using gloves or not. One hospital used the “Clean In, Clean Out” sign in every patient room. However, multiple studies since the CDC’s Healthcare Infection Control Practices Advisory Committee (HICPAC) Hand Hygiene Guidelines was released (27) have shown only modest compliance with this intervention, which has proven effective since the time of Semmelweis. HCWs seem better at hand hygiene after rather than before seeing a patient. Nurses tend to be much better than doctors. And, most surprisingly, nonsurgical primary care physicians and intensivists are better than their surgical colleagues (28).

Another long-standing concern of IPs and hospital epidemiologists is the attitude of some nurses and physicians toward isolation precautions. The apparent lack of understanding regarding the rationale for careful hand hygiene and the use of PPE by those involved in direct patient care continues to frustrate the infection control community. The severe acute respiratory syndrome (SARS) outbreak, the appearance of a new SARS-like corona virus infection from the Middle East (MERS-CoV), the emergence of highly resistant bacteria, and novel-strain influenza viruses should be adequate incentive enough for all HCWs to consistently practice the appropriate use of PPE and hand hygiene. Moreover, the increased scrutiny of the
public on how HCWs manage these threats should heighten a sense of professional responsibility.

Recommendations

One effective way to measure the success of hand hygiene implementation is by direct observation of the hospital medical and nursing staff’s use of the CDC HICPAC Hand Hygiene Guidelines with direct feedback to the units and types of staff on their percentage use of appropriate hand hygiene.
Select an alcohol-based hand rub (foam or gel) that is most acceptable to staff in trials.
Authorize all HCWs to ask for a “time-out” to review missed opportunities for the proper use of PPE in patient isolation situations.
Enable the admitting and/or supervising nurse on a unit to put a new patient in isolation if the patient’s admitting diagnosis or condition warrants.
Whenever possible, use a “forcing function” or engineering control rather than human behavioral change to implement a new policy.
Establish a culture of accountability for clinicians on all units/wards. It is the responsibility of clinicians and chiefs/directors of wards/units, rather than only infection control, to ensure compliance with infection control recommendations.

CONTROL OF RESISTANT ORGANISMS

No matter what the setting or size of a hospital, the problem of MDRO infection is a major preoccupation for ICPs and has led to calls for action from many quarters. Data comparing percentages of healthcare-associated Staphylococcus aureus infections with methicillin-resistant Staphylococcus aureus (MRSA) in NNIS hospitals with <200 beds to those with >200 beds between 1992 and 2002 demonstrates that smaller hospitals, formerly behind larger hospitals in their percentage of MRSA-HAIs, have caught up with larger hospitals (29). There is a need for more study of MDRO rates in the community healthcare setting; however, most studies of the epidemiology and control of MDROs have come from large, academic centers. Diekema et al. conducted a survey of >400 US hospitals and found that antimicrobial resistance rates were strongly associated with the size, geographic location, and academic affiliation of hospitals (30).

Hospitals, especially those with intensive care units (ICUs), and long-term care facilities are important epicenters and repositories of MDROs. Increasing rates of community-onset MRSA (CO-MRSA) also are having an impact on hospitals as CO-MRSA strains are recognized as causes of HAI infections as well (31). Unfortunately, many centers, discouraged by the increase in MRSA and the cost of control, relaxed previously recommended infection control practices in the recent past. In contrast, some northern European countries adopted national, comprehensive programs resulting in impressive control of resistant organisms.

Cost and safety issues surround these infections—increased morbidity and mortality, more expensive and limited treatment options, longer hospital stays, patient dissatisfaction, the cost and inconvenience of precautions, litigation, and adverse publicity for healthcare facilities (especially where drug resistance is publicly reported and/or considered a measure of quality)—and are an increasing reality in today’s consumer-driven patient safety movement.

With the identification of vancomycin-resistant S. aureus (VRSA), CO-MRSA, more virulent strains of C. difficile, extended-spectrum beta-lactamase (ESBL)-resistant organisms, carbapenem-resistant Enterobacteriaceae (CRE), fungal resistance, multidrug-resistant Mycobacterium tuberculosis, and other emerging organisms, it is imperative that community hospitals have a complete understanding of the resistance patterns in their institutions, their affiliates, and their region. Similarly, they must keep up with developing recommendations for controlling these pathogens, even if such pathogens have not yet affected them.

While antibiotic resistance is a daunting problem, a substantial body of literature demonstrates that these organisms can be successfully controlled with multidisciplinary efforts that include active surveillance testing, contact or barrier precautions, careful environmental cleaning, effective antimicrobial stewardship, and strict compliance with evidence-based hand hygiene practices. The degree to which control measures should be adopted universally, particularly in regions with low prevalence rates of resistance, is one of the issues surrounding the debate among experts over developing guidelines for managing resistance. This can confuse and deter smaller hospitals that are attempting to determine and implement best practices for the identification and control of these organisms.

RECOMMENDATIONS

In 2003, SHEA published an extensive review of the literature and guidelines for the control of MRSA and vancomycin-resistant enterococcus (VRE) (15). It is widely known that the publication of this guideline during the simultaneous development of draft guidelines by HICPAC sparked a vigorous debate among experts as to how to interpret and apply the best science to control these organisms. Most hospitals must decide what they will take from each of the guidelines to develop strategies for preventing the dissemination of resistance. Both guidelines should be studied carefully. The CDC HICPAC MDRO Guideline recommends that if a hospital’s rate of infection with these pathogens is not decreasing, it should implement more aggressive measures, including active surveillance testing. We present an approach to managing MRSA and VRE adopted by the MaineHealth^® Infection Control Consortium in Appendix 2.

The most salient difference between the two guidelines is the screen and isolate or “search and destroy” practice (20). The SHEA guidelines recommend aggressive use of active surveillance testing to identify patients colonized with these organisms (because routine clinical cultures do not detect the colonized reservoir) and placing those patients on contact precautions. The HICPAC guidelines favor a graduated, two-tiered approach with increased intensity of control activities in settings where baseline measures fail to decrease transmission rates. A major argument used by those who find the SHEA document too rigorous is that many studies used multiple interventions and are unable to weigh the strength of any single intervention or combination of interventions (an argument never made with any of the other CDC HICPAC guidelines, although this is also true of them). However, Muto et al. vigorously defend the SHEA approach of active surveillance testing and contact isolation (also called active detection and isolation or ADI), citing the success of northern European countries and
Western Australia in controlling these infections. They contrast this with the failure of standard precautions and failure to control these organisms in other areas of Europe and Australia where a less aggressive approach is used (32,33). In European countries, the delivery of medical care may differ from that in US healthcare facilities, the intended target for these guidelines. In countries and institutions where an aggressive approach, such as that recommended in the SHEA guidelines, has been adopted, however, impressive control of MRSA and VRE have been achieved.

Translating the science coming out of larger academic centers into practice at the level of smaller, nonteaching hospitals is complicated by a lack of understanding of which interventions are most effective and an imprecise understanding of the epidemiology of these organisms outside of large hospitals. Some experts argue that it is better to direct efforts and resources at reducing SSIs and device-related HAIs as a means to control resistance rather than to implement the type of “search and destroy” strategies that have been successful in Europe. Indeed, the former approach is gaining momentum as a result of national quality efforts toward implementing proven methods of reducing such infections by encouraging the utilization of “bundles” of evidence-based best practices for prevention (20). However, it is far from clear that this will be an adequate national strategy for decreasing the rates of MRSA and VRE in the United States and in community hospitals.

West et al. have published a paper on the effect of targeted surveillance cultures of high-risk populations for control of MRSA in a community hospital system (34

Only gold members can continue reading. Log In or Register to continue