Infection Control: A Decade of Lessons Learned
ACVIM 2008
Paul S. Morley, DVM, PhD, DACVIM
Fort Collins, CO, USA


Ten years ago, Colorado State University created a rather unique faculty position to meet a perceived need at the Veterinary Teaching Hospital: a faculty position whose clinical appointment was dedicated to oversight of infection control in the hospital, a director of 'biosecurity.' The stimulus for creating this position was primarily the result of an outbreak of nosocomial Salmonella infections that occurred in the large animal hospital in 1996 which resulted in widespread contamination of the facility.1 Unfortunately, despite best efforts at mitigation after recognition of the problem, new infections continued to occur until all large animal patients were discharged and the entire facility was thoroughly decontaminated. The consequences of this outbreak and the 3-month closure required to decontaminate the facility were significant, both in terms of financial losses and also for actions deemed necessary to meet our operational goals. The primary goal of our hospital is to deliver the very best veterinary care possible anywhere in the world (as I am sure other referral hospitals can also claim). Our faculty's experience with this outbreak created a belief that we need to be systematic and purposeful in our approach to infection control in order to reliably achieve this goal. Infection control is probably not the most critical activity defining excellent medical care, but excellence in patient care cannot be achieved without adequate infection control. The purpose of this talk is to share some key lessons that we have learned while developing our biosecurity program during the past decade.

"I Thought It Would Be About Science"

I had the privilege of joining the faculty at CSU as Director of Biosecurity for the Veterinary Teaching Hospital in 1998. Given my background in epidemiology and internal medicine, I could foresee that my activities would emphasize protecting the wellbeing of the larger hospital population while optimizing care for individual animals. As an internist, I could foresee that we would work with the best available scientific data and make logical extrapolations where data did not exist. As a researcher, I could foresee the need to identify important gaps in our knowledge regarding infection control and target research efforts to fill these holes. Having previously taught public health, I could foresee the need to address personnel concerns regarding zoonotic disease risks. The unifying concept is that before I took the job, I thought the biggest challenges for developing and refining infection control for a major referral hospital would mostly revolve around scientific issues. However, the reality that I quickly learned is that the biggest challenges in this job, the most persistent difficulties, the things that take most of my time, the things that slow progress the most, the things that frustrate me most, and my biggest mistakes all relate to human relations and communication issues. While I thought it would mostly be about science, it really is mostly about human relations and achieving compliance and buy-in. The truth is that one misstep with hospital personnel can cause as many problems as any infectious agent.

"What Data?"

In the best circumstances, data would be available to guide all decisions that we need to make regarding infection control (Does this horse really need to go to isolation? What disinfectant should we use? Do footbaths reduce patient risk? When is active surveillance most useful? etc). Unfortunately, there is a dearth of evidence regarding many of the most important issues related to infection control. For example, we have almost no information about the efficacy of many of our most commonly used preventive strategies. We don't know how best to determine if we are being successful (e.g., If we use very rigorous and expensive control strategies and no outbreaks occur, how can we determine if these strategies are responsible for the apparent success?). Most hospitals have no idea of how many nosocomial infections occur, let alone how many should be expected and how many are preventable. Which brings us to another important concept:

"Not All Nosocomial Infections are Preventable"

There is no question that nosocomial infections are a known and unavoidable risk associated with administering medical care and hospitalization of patients. As such, there has to be some minimum rate of occurrence that can be expected to occur under the best circumstances. The key, therefore, is to attempt to identify and minimize the occurrence of preventable infections that occur beyond those that cannot be avoided. Even if it is difficult to label individual incidents as being preventable, it is possible to create standards for expected rates of infections and then determine whether facilities are experiencing greater than expected rates of infection. For over 2 decades, human health care facilities in the U.S. have participated in national surveillance efforts that help to establish expected rates of infection as well as provide a check of whether they are experiencing problems. Unfortunately, there are no established norms for determining expected rates of nosocomial infections in veterinary care facilities. This is one of the biggest deficiencies we have regarding infection control in veterinary medicine and one which we must remedy as soon as possible if we are to make systematic progress in infection control across the entire profession.

"We Need To Talk"

Because many veterinarians succumb to the belief that every occurrence of nosocomial infection is an indication that mistakes were made (i.e., that every infection is preventable). As such, there is a tendency across the entire profession to conceal and not talk openly about these instances or even about major outbreaks. Unfortunately, this lack of discussion only reinforces the impression that all infections are preventable and therefore someone is liable for losses in every instance. While every adverse event is regrettable,2 clearly not all nosocomial infections or adverse outcome is preventable. As such, I strongly believe that we need to embrace open disclosure about the risks related to nosocomial infections. If veterinarians, clients, and students are to gain a better understanding about what constitutes normal risk and what is preventable then we need to do a much better job of investigating and communicating about expected and extraordinary risks related to nosocomial infections.

Nosocomial infections are clearly not a rare problem in veterinary medicine. In a recent survey3 of veterinary teaching hospitals at AVMA accredited institutions, 82% (31/38) reported identifying outbreaks of nosocomial infections during the past 5 years, and 57% (17/31) of these reported identifying problems with >1 infectious agent. Salmonella was the most common agent identified (20/31) followed by methicillin-resistant Staphylococcus aureus (MRSA, 13/31). More than half (58%, 22/38) of the hospitals had restricted admissions during this period to reduce patient risk or mitigate contamination, 68% (15/22) of which were equine facilities. A third of all hospitals (32%, 12/38) closed part or all of their facilities in order to control the problem.

By investigating, reporting, and discussing the occurrence of nosocomial infections, I believe we promote a better understanding of the true risks among the entire profession, which in turn protects us all from inappropriate judgments from clients, colleagues, or the legal system.

"Not Everybody Believes"

We try to base as many decisions as possible on data and objective evidence, but there is still a large portion of our actions that must be guided by belief and 'best' judgment. While failure to take any precautions related to infection control is obviously malpractice, it is often not clear about how much is 'enough.' Based upon our primary practice goal (be the very best), we have adopted a very risk-averse approach to infection control. From experience in talking with other veterinarians, however, it is clear that not everyone believes this is necessary. Worse, a few people have even said that we sometimes impair patient care with our rigorous procedures and precautions. We disagree, but how can we convince others that being more risk-averse is an appropriate stance for our circumstances? We must continue to generate data that guides toward the best practice standards possible. Veterinary medicine is probably at least a decade behind human medicine in terms of knowing what works, what doesn't, what is necessary, and what isn't. In our survey of veterinary teaching hospitals,3 only about a third of institutions (37%, 14/38) reported that they included any disclosure of nosocomial infection risks in informed consent statements that their clients sign despite the common occurrence of outbreaks of nosocomial disease that was reported. Additionally, 64% (32/50) of designated biosecurity experts from these institutions reported that the true risk of nosocomial infection (as opposed to increased concern or awareness) was greater or much greater than it was 10 years ago. Clearly, different types of operations will have different hazards related to nosocomial infection, and the rigor and practices that are employed must be tailored to each individual operation.4,5 As such, what is appropriate for one practice with higher inherent risks because of the level of care and the types of patients could be unnecessary for another type of practice. However, there is still a distinct impression that many of those who vocally dismiss infection control procedures as being unnecessary often do so without having a clear understanding of the risks in their own or other practices. Obtaining more and better data is the only solution to clarifying these questions.


1.  Tillotson K, et al. J Am Vet Med Assoc 1997;211:1554-1557.

2.  Harvard Hospitals, Massachusetts Coalition for the Prevention of Medical Errors, 2006 [available].

3.  Benedict KM, et al. J Am Vet Med Assoc 2008 [in press].

4.  Morley. Vet Clin N Am Food Anim Pract 2002;18:133-155.

5.  Morley and Weese. Smith BP, ed. Large Animal Internal Medicine, 4th ed [In Press].

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Paul Morley, DVM, PhD, DACVIM
Ft. Collins, CO