Charting the Course for the Future of Science in Healthcare Epidemiology: Results of a Survey of the Membership of the Society of Healthcare Epidemiology of America
To describe the results of a survey of members of the Society for Healthcare Epidemiology of America (SHEA) that (1) measured members’ perceptions of gaps in the healthcare epidemiology knowledge base and members’ priorities for SHEA research goals, (2) assessed whether members would be willing to participate in consortia to address identified gaps in knowledge, and (3) evaluated the need for training for the next generation of investigators in the field of healthcare epidemiology.
Electronic and paper survey of members of SHEA, a professional society formed to advance the science of healthcare epidemiology through research and education.
All society members were invited to participate.
Of 1,289 SHEA members, 593 (46.0%) responded. Respondents identified the following issues as important for the Research Committee of SHEA: setting the scientific agenda for healthcare epidemiology, developing collaborative infrastructure to conduct research, and developing funding mechanisms for research. Respondents ranked multidrug‐resistant gram‐negative organisms, antimicrobial stewardship, methicillin‐resistant Staphylococcus aureus, adherence to effective hand hygiene guidelines, and Clostridium difficile infections as the most important scientific issues facing the field. Respondents ranked inadequate project funding, lack of protected time for research, and inability to obtain a grant, contract, and/or outside funding as the most significant barriers to conducting research. More than 92% of respondents support creating a SHEA research consortium; more than 40% would participate even if no additional funding were available; nearly 90% identified developing research training as a key function for SHEA.
These data provide a road map for the SHEA Research Committee for the next decade.
Healthcare‐associated infections (HAIs) have become a source of increasing scrutiny—both from the general public as well as from policy makers—over the past decade. Data amassed by the US Public Health Service’s Centers for Disease Control and Prevention document the magnitude of the problem of HAIs for our society. In both the United States and Western Europe, a high percentage of hospitalizations (perhaps as many as 10% in certain clinical settings) are complicated by HAI.1 In the United States alone, as many as 1.7 million people per year develop an HAI, and as many as 100,000 individuals per year die of either the HAI or complications associated with this type of infection.2 Because of the extraordinary human and economic costs associated with nosocomial infections, investigators have focused on documenting the effectiveness of the systematic implementation of interventions suspected to reduce or eliminate HAIs, in some instances with great success.3‐7 These studies are often limited by the fact that many practices proposed for widespread use to prevent or control HAIs are not based on solid scientific data.8 The gap between science and practice has recently been summarized in a white paper published by the Society for Healthcare Epidemiology of America (SHEA).8 SHEA members are optimally positioned to determine the research agenda for HAIs in the United States.
We recently surveyed the membership of SHEA to gain insight into the their views about the HAI problem and their assessment of the most important obstacles to designing effective interventions. The major goals for the survey were (1) to identify gaps in our knowledge base and determine the membership’s priorities for research goals, (2) to assess whether SHEA members would be willing to participate in research consortia to address identified gaps in scientific knowledge (either without or with additional funding), and (3) to evaluate the need for additional training for the next generation of investigators in the field of healthcare epidemiology and the membership’s view of SHEA’s role in such training. This paper describes the results of the survey of the SHEA membership.
The survey was designed by members of the SHEA Research Committee and was initially distributed electronically on September 30, 2008. Four reminders were distributed to the membership via e‐mail on October 9, 2008, November 6, 2008, November 12, 2008, and February 3, 2009. In addition, a printed copy of the survey was mailed to all SHEA members on November 25, 2008, at the time dues renewal notices were distributed. Responses were included in the survey results if they were received before February 15, 2009.
Data were analyzed by use of SAS, version 9.1 (SAS Institute), specifically by use of the χ2 test (sometimes with the Kruskal‐Wallis modification), the Fisher exact test, and the Z test (for proportions). In some instances, data were stratified by academic affiliation and by extent of experience in the field of healthcare epidemiology, to make further comparisons. A P value of less than .05 was considered to be statistically significant.
Response Rate and Demographics
At the time the survey was distributed in September 2008, SHEA had 1,289 members; by November 2009, the membership had increased to 1,697 members. This analysis is based on the 1,289 individuals who were members at the time the survey was initially distributed. Of the 1,289 members, 593 (46.0%) responded to the survey. Although SHEA does not maintain an exhaustive demographic database, we were able to make some comparisons between the overall membership and the respondents to the survey. With respect to demographics, the respondents, although generally comparable to the overall SHEA membership, were more likely to be women (
|Characteristic||No. (%) of respondentsa|
|Age of respondents (
|20–29 years||5 (0.8)|
|30–39 years||93 (15.7)|
|40–49 years||143 (24.2)|
|50–59 years||237 (40.0)|
|60–69 years||102 (17.2)|
|⩾70 years||17 (2.9)|
|Duration of experience in the field of healthcare epidemiology (
|<1 year||11 (1.9)|
|1–3 years||51 (8.7)|
|4–8 years||84 (14.3)|
|9–12 years||75 (12.7)|
|13–20 years||147 (25.0)|
|21–30 years||142 (24.1)|
|>30 years||79 (13.4)|
|Current jobb (
|Infection preventionist||124 (20.9)|
|Infection control research nurse||4 (0.7)|
|Hospital epidemiologist||228 (38.4)|
|Infectious diseases physician||318 (53.6)|
|Infectious diseases fellow||9 (1.5)|
|Healthcare administrator||38 (6.4)|
|Training and/or degreeb (
|Registered nurse||36 (6.1)|
|Medical technologist||11 (1.9)|
|Bachelor of science in nursing||33 (5.6)|
|Bachelor of science||28 (4.7)|
|Master of public health||93 (15.7)|
|Master of science||84 (14.2)|
|Doctor of philosophy||56 (9.4)|
|Doctor of medicine||423 (71.3)|
|Primary hospital affiliation (
|University academic medical center||150 (28.9)|
|Academic medical center||83 (16.0)|
|Behavioral health hospital||1 (0.2)|
|Community hospital, with no academic affiliation||100 (19.3)|
|Community hospital, academically affiliated||30 (5.8)|
|Community hospital, teaching||117 (22.5)|
|Federal, military||3 (0.6)|
|Federal, nonmilitary||26 (5.0)|
|Long‐term care facility||6 (1.1)|
|Outpatient facility||2 (0.4)|
|Rehabilitation facility||1 (0.2)|
|Primary hospital size (
|0–49 beds||5 (1.0)|
|50–99 beds||42 (8.3)|
|100–199 beds||228 (44.8)|
|200–499 beds||15 (3.0)|
|500–749 beds||102 (20.0)|
|750–999 beds||78 (15.3)|
|>1,000 beds||39 (7.7)|
|Sourcea||Proportion (%) of respondents|
|National Institutes of Health||109/552 (19.7)|
|Centers for Disease Control and Prevention||123/550 (22.4)|
|Agency for Healthcare Research in Quality||71/542 (13.1)|
|US Department of Veterans Affairs||46/534 (8.6)|
|National Patient Safety Foundation||22/543 (4.1)|
|Other public or private foundation||106/544 (19.5)|
|Pharmaceutical and/or device manufacturer||157/558 (28.1)|
Issues of Importance for Consideration by the SHEA Research Committee
With respect to respondents’ views of the role that should be played by the SHEA Research Committee (Table 3), respondents were asked to rate options using a 5‐point Likert scale; the options were “important,” “somewhat important,” “neither important nor unimportant,” “somewhat unimportant,” and “unimportant.” More than 92% of the respondents categorized SHEA’s involvement in setting the scientific research agenda for the field of healthcare epidemiology as either important or somewhat important. Respondents from academic hospitals were significantly more likely than those from institutions without academic affiliation to rank setting the research agenda as important (
|Issue||Percentage of respondents
|Ranked as important||Ranked as important or somewhat important|
|Setting the scientific agenda for healthcare epidemiology||63.1||92.4|
|Developing collaborative infrastructure to conduct research||51.8||91.8|
|Developing funding mechanisms for research||56.9||89.9|
|Creating a SHEA research consortium||46.0||85.2|
|Developing infrastructure for research training||44.4||88.6|
|Establishing national policies for healthcare epidemiology||43.4||81.1|
|Creating training programs for researchers||37.2||83.4|
|Developing or establishing certification process for healthcare epidemiology||23.7||62.2|
Nearly 92% of respondents characterized as important or somewhat important the concept of creating a collaborative infrastructure among SHEA members and partners to conduct research in healthcare epidemiology. Interestingly, experience in the field of healthcare epidemiology was inversely associated with identifying the development of the collaborative infrastructure as “important,” which suggests that the more experienced investigators felt that the development of a collaborative infrastructure was less important (
Respondents also endorsed the concept of establishing a consortium of members’ healthcare institutions to address important research questions in healthcare epidemiology. Responses from individuals in academic and nonacademic institutions were not different, nor were responses from individuals with or without existing research funding. Responses from individuals who have worked in the field of healthcare epidemiology for different periods of time also did not vary significantly.
Nearly 90% of respondents believe that the SHEA Research Committee should try to develop mechanisms to fund important research in healthcare epidemiology. No differences were identified among respondents who have varying levels of experience, between academic and nonacademic respondents, or between funded and nonfunded investigators with respect to this issue.
One important aspect of the establishment of a research consortium that can efficiently and effectively address the numerous outstanding scientific issues facing the field of healthcare epidemiology is funding. We asked whether SHEA members would participate in the initial projects associated with such a consortium, even if no additional funding were made available to support the studies. Whereas 342 (59.8%) of 572 respondents were either uncertain about their institutions’ willingness to participate or thought that lack of funding would preclude participation, 230 (40.2%) noted that they would be willing to participate.
The survey also asked with whom the membership thought SHEA should partner to establish a research consortium and to conduct healthcare epidemiology research. A total of 507 respondents identified organizations as possible partners. These organizations included the Infectious Diseases Society of America (459 respondents [90.5%]), the Association for Professionals in Infection Control and Epidemiology (405 respondents [79.9%]), the Centers for Disease Control and Prevention (346 respondents [68.2%]), the Pediatric Infectious Diseases Society (209 respondents [41.2%]), the Institute for Healthcare Improvement (209 respondents [41.2%]), and the National Institutes of Health (165 respondents [32.5%]).
Scientific Issues of Importance to the SHEA Membership
With respect to the most important scientific questions facing the field of healthcare epidemiology (Table 4), the 5 issues that were most frequently ranked as number 1 were (in descending order) multidrug‐resistant gram‐negative organisms, antimicrobial stewardship, methicillin‐resistant Staphylococcus aureus, adherence to effective hand hygiene guidelines, and Clostridium difficile infection. If one includes the issues that respondents ranked first, second, and third, the order of importance of the issues on the list remains unchanged. Although differences among stratified responses did not vary substantially, individuals who had less than 4 years of experience tended to rank the problem of multidrug‐resistant gram‐negative organisms much higher than did the more experienced individuals (
|Considering the first choice||Considering the first 3 choicesa|
|Multidrug‐resistant gram‐negative organisms||1||1|
|Methicillin‐resistant Staphylococcus aureus||3||3|
|Effective hand hygiene||4||4|
|Clostridium difficile infection||5||5|
|Intravascular catheter–associated infection||6||6|
|Surgical site infections||7||7|
|Preventing transmission of highly contagious infectious diseases||8||8|
|Implantable device–associated infection||9||9|
|Catheter‐associated urinary tract infection||11||11|
|Vancomycin‐resistant enterococcal infection||12||12|
|Improving healthcare worker safety||13||13|
|Preventing noninfectious healthcare‐associated medical complications||14||14|
Barriers to the Conduct of Clinical Research in Healthcare Epidemiology
The Research Committee also was interested in learning what SHEA members identified as the most significant barriers to the conduct of effective clinical research in healthcare epidemiology. Issues identified by respondents are summarized in Table 5.
|Considering the first choice||Considering the first 3 choicesa|
|Inadequate funding for specific projects||1||1|
|Lack of protected time for research||2||2|
|Inability to obtain grant, contract, and/or outside funding||3||4|
|Inadequate number of personnel||4||3|
|Human subjects protection approval process||5||5|
|Research is not a part of institutional mission||6||8|
|Inadequate clinical research training||7||6|
|Inadequate biostatistical support||8||7|
|Inadequate mentoring during training||9||9|
|Uninterested in clinical research||10||12|
|Regulations and regulatory oversight||12||11|
|Inadequately trained personnel||13||11|
Certification in Healthcare Epidemiology
SHEA members have expressed interest in the development of a program for subspecialty certification in healthcare epidemiology. This concept was the least enthusiastically endorsed of all the issues identified in the survey, and no differences were found among respondents when stratified according to whether they worked at an academic or a nonacademic institution, whether they were funded or not funded, or the extent of their experience in the field. Individuals who have been SHEA members for less than 4 years were significantly more likely than other SHEA members to advocate for the development of a certification process (
This paper describes the findings of a survey of the membership of the SHEA. Our findings identify both significant challenges and significant opportunities. One of the most troubling findings is that the community of professionals in the field of healthcare epidemiology in the United States is aging. More than 60% of the respondents were over 50 years of age, and 20% were older than 60 years. As a society, SHEA needs to plan for the future, including developing and implementing plans to recruit young, scientifically oriented members, developing and implementing succession plans for individual programs in healthcare epidemiology, as well as developing and implementing succession plans for the leadership of the Society. SHEA’s sister society, the Association for Professionals in Infection Control and Epidemiology, has identified a similar concern. Interestingly, respondents identified inadequate training and mentoring as contributing to the difficulty in conducting effective clinical research in healthcare epidemiology. Nearly 90% of respondents identified “developing infrastructure for research training” as a key function for the Research Committee and for the Society. In addition, inadequate training and/or mentoring were among the most commonly cited barriers (seventh and ninth, respectively) to the conduct of effective clinical research. If, as we suspect, these responses are reflective of the entire membership, the Society must focus on research training over the next decade. Such training will clear the path for young investigators to become clinical researchers in the field of healthcare epidemiology and will likely serve as a magnet to attract young scientists into the Society.
A striking finding from the survey was the virtually uniform agreement among respondents about the importance of SHEA setting the research agenda for healthcare epidemiology for the foreseeable future. More than 90% of respondents identified setting the research agenda as an important function for the Society and its Research Committee. In the United States, more than half of the states have passed legislation mandating that institutions report HAI rates. Both state legislatures and many consumer advocacy groups have argued that mandatory screening and reporting will reduce HAIs, despite the fact that no data support these beliefs. These developments underscore the need for additional research into the epidemiology and pathogenesis of HAI, as well as research into the efficacy of infection prevention strategies.8
More recently, the US Department of Health and Human Services published a National Action Plan to prevent HAIs.9 This plan focuses on reporting HAIs and assuring the implementation of existing guidelines. As noted above and elsewhere,8 the science base for many existing infection prevention guidelines is far from robust. Many, if not most, existing infection control guidelines are based on “expert opinion,” experience, empiricism, and/or common sense.8 Whereas such approaches are seemingly rational, the evidence base is far from definitive. Certainly these kinds of data are not adequate to support mandatory guideline implementation. SHEA has expressed the concern that the National Action Plan9 pays insufficient attention to the need for additional basic and clinical research in the field of healthcare epidemiology, noting that mandating the implementation of infection control interventions that are based on inadequate scientific evidence may limit their effectiveness or may have unanticipated consequences.8
The SHEA membership has identified several critical issues that need further scientific explication. The emergence of infections caused by multidrug‐resistant gram‐negative organisms, such as Acinetobacter baumannii, has become the scourge of intensive care units in tertiary referral centers. Even in 2010, we have an extremely rudimentary understanding of both the pathogenesis of infection caused by A. baumannii and the nosocomial epidemiology of this perplexing pathogen. We have a great deal to learn about what types of strategies might be effective in encouraging either antimicrobial stewardship or appropriate hand hygiene practices. For the familiar pathogens (ie, methicillin‐resistant S. aureus, C. difficile, and vancomycin‐resistant enterococci), our understanding of their epidemiology and pathogenesis and of effective preventive practices is far from complete. For successful prevention strategies to be developed, we must conduct the necessary scientific studies to delineate these factors.
The fact that many of these important questions have been identified for years but remain unanswered underscores the complexity of the topics. Individual institutions cannot accrue enough patients to address these questions in high‐quality studies. For this reason, the SHEA Research Committee has advocated for the formation of a SHEA research consortium. The consortium approach offers the opportunity for a sophisticated collaborative study design, substantially larger patient populations, intellectual partnerships with other professional societies, and quick turnaround times for individual studies. The Centers for Disease Control and Prevention is planning to use its National Health Safety Network creatively to address important questions in healthcare epidemiology. Successful implementation of the SHEA consortium, with subsequent publication of study findings, would likely also place the consortium in a favorable position for grant or contract funding for subsequent studies.
A major concern of survey respondents relates to availability of funding for healthcare epidemiology studies. With public and legislative attention currently focused on the problem of HAIs, opportunities for funding may materialize if the Society can develop an effective consortium‐based approach.
Respondents’ comments about barriers to effective clinical research in healthcare epidemiology provide insight into the current state of healthcare epidemiology. The top 4 barriers identified relate directly to insufficient resources for basic science: (1) inadequate funding for specific projects, (2) lack of protected time for research, (3) inability to obtain grant, contract, and/or outside funding, and (4) an inadequate number of personnel in the field of healthcare epidemiology. Although by no means guaranteed, successful implementation of the consortium model that produces tangible beneficial outcomes may well offer an opportunity for individual hospital epidemiologists to approach their healthcare administrators, as well as funding agencies, for additional resources. One encouraging finding for the Society is the fact that more than 70% of respondents reported that they would participate in such a consortium, and more than 40% reported that they would participate in such a consortium even if additional resources were not made available for participation.
A survey of this type has limitations. Our response rate was slightly less than 50% (despite the fact that, for completing the survey, members received a $25 reduction in their annual dues), so those who answered the survey may not be representative of the entire Society. Individuals who responded may feel more strongly about certain issues mentioned in the survey than those who did not respond. The remarkable uniformity of the responses across all age groups and levels of experience nonetheless suggests a high likelihood that the responses are representative of the Society’s membership. Not all respondents answered each question. Several questions consisted of menus of selections generated by members of the Research Committee. In all instances, respondents were given the opportunity to enter additional choices. In no instance was any “other” issue identified by more than 6 respondents; nonetheless, our selection of the menus of items may have introduced bias.
Perhaps the single most striking finding from the survey was the extent to which the respondents were of “one mind” about several of the important issues facing the Society in the year 2010 and beyond. Approximately 90% of respondents agreed about each of the issues listed in Table 6. Such unanimity of opinion is distinctly unusual, in our experience. The intense focus on the problem of HAIs has underscored what SHEA and its membership can and should do.
|1. SHEA should identify the research agenda for the discipline of healthcare epidemiology.|
|2. SHEA and its members should develop succession plans and plans to recruit young investigators into the Society.|
|3. The Society should establish a collaborative infrastructure to support clinical research in healthcare epidemiology and should create a research consortium to facilitate the conduct of appropriately designed, controlled, and powered studies to address important unanswered scientific questions in the field of healthcare epidemiology.|
|4. The Society should develop robust strategies and infrastructures to support training in healthcare epidemiology research.|
|5. The Society should work with partners to develop funding strategies to support these studies.|
In our view, the results from this survey provide straightforward recommendations for the Society (Table 6) as well as a road map for the science efforts of the Society for the next decade. In an effort to begin influencing the research agenda in healthcare epidemiology, the Research Committee has already published a white paper describing the Society’s priorities.8 We are working collaboratively to establish the infrastructure for a SHEA research consortium. A subcommittee of the Research Committee has been formed to identify the first study question(s) to be addressed. Clearly, the attention focused on the problem of HAIs represents a call to action for the community of professionals in the field of healthcare epidemiology.
Members of the Research Committee of SHEA
David K. Henderson, MD, Chair; Marc Bonten, MD; Neil Fishman, MD; Steve Gordon, MD, MPH; Stephan Harbarth, MD; Anthony D. Harris, MD, MPH; Ebbing Lautenbach, MD, MPH, MSCE; Tara N. Palmore, MD; Eli N. Perencevich, MD, MS; Trish M. Perl, MD, MSc; Richard Platt, MD; Sanjay Saint, MD, MPH; Matthew Samore, MD; Robert Sherertz, MD; David Weber, MD, MPH; Robert A. Weinstein, MD; Valerie Deloney, MBA.
Potential conflicts of interest. All authors report no conflict of interest relevant to this article.
- 1.Humphreys H, Newcombe RG, Enstone J, et al; Hospital Infection Society Prevalence Survey Steering Group. Four country healthcare associated infection prevalence survey 2006: risk factor analysis. J Hosp Infect 2008;69(3):249–257.
- 2.Klevens RM, Edwards JR, Richards CL Jr, et al. Estimating health care‐associated infections and deaths in U.S. hospitals, 2002. Public Health Rep 2007;122(2):160–166.
- 3.Berenholtz SM, Pronovost PJ, Lipsett PA, et al. Eliminating catheter‐related bloodstream infections in the intensive care unit. Crit Care Med 2004;32(10):2014–2020.
- 4.Eggimann P, Harbarth S, Constantin MN, Touveneau S, Chevrolet JC, Pittet D. Impact of a prevention strategy targeted at vascular‐access care on incidence of infections acquired in intensive care. Lancet 2000;355(9218):1864–1868.
- 5.Pronovost P, Needham D, Berenholtz S, et al. An intervention to decrease catheter‐related bloodstream infections in the ICU. N Engl J Med 2006;355(26):2725–2732.
- 6.Hanna H, Benjamin R, Chatzinikolaou I, et al. Long‐term silicone central venous catheters impregnated with minocycline and rifampin decrease rates of catheter‐related bloodstream infection in cancer patients: a prospective randomized clinical trial. J Clin Oncol 2004;22(15):3163–3171.
- 7.Raad II, Hanna HA. Intravascular catheter‐related infections: new horizons and recent advances. Arch Intern Med 2002;162(8):871–878.
- 8.Research Committee of the Society of Healthcare Epidemiology of America. Enhancing patient safety by reducing healthcare‐associated infections: the role of discovery and dissemination. Infect Control Hosp Epidemiol 2010;31(2):118–123.
- 9.U.S. Department of Health and Human Services. HHS action plan to prevent healthcare‐associated infections. HHS.gov Web site: http://www.hhs.gov/ophs/initiatives/hai/actionplan/index.html. Updated June 2009. Accessed November 30, 2009.