Horizontal Versus Vertical: Two Approaches to HAI Prevention

By Kelly M. Pyrek

Two approaches to infection prevention that are being used in hospitals today bear continued scrutiny as multidrug-resistant organisms proliferate, emphasize experts writing in a recent commentary in the journal Infection Control and Hospital Epidemiology. Edward Septimus, MD, of the Texas A&M Health Science Center College of Medicine in Houston, Texas, and of Hospital Corporation of America in Nashville, Tenn. and his co-authors urge clinicians to carefully consider the clinical advantages and cost-related disadvantages to each strategy.

A horizontal approach to infection prevention and control measures refers to broad-based approaches attempting reduction of all infections due to all pathogens, while a vertical approach refers to a narrow-based program focusing on a single pathogen. While vocal advocates of each approach exist, some experts also concede that vertical approaches are generally indicated when standard measures have failed. Generally, the goal of a vertical approach is to reduce infection or colonization due to specific pathogens while a horizontal approach aims to eliminate all infections and is population-based. While the vertical approach is selective or universal when it comes to application, while the horizontal approach is generally universal. In terms of resource utilization, the vertical approach is typically high while the horizontal approach is usually lower; in addition, a vertical program promotes exceptionalism (some organisms are more important that others), while a horizontal approach is more utilitarian. A vertical approach favors the hospital while the horizontal approach favors the patient, and both approaches differ in terms of time needed to see results -- vertical is a short-term approach while horizontal is designed both for the present and for the long-term. Finally, the approaches diverge in terms of interventions; vertical programs encompass active surveillance and vaccination of healthcare workers, while horizontal embraces hand hygiene, CHG bathing, care bundles and activities that reduce presenteeism among healthcare workers. (Edmond, 2011)

As Septimus, et al. (2014) write, "Over the last decade, the general approaches to healthcare-associated infection (HAI) prevention have taken two conceptually different paths: (1) vertical approaches that aim to reduce colonization, infection, and transmission of specific pathogens, largely through use of active surveillance testing (AST) to identify carriers, followed by implementation of measures aimed at preventing transmission from carriers to other patients, and (2) horizontal approaches that aim to reduce the risk of infections due to a broad array of pathogens through implementation of standardized practices that do not depend on patient-specific conditions. Examples of horizontal infection prevention strategies include minimizing the unnecessary use of invasive medical devices, enhancing hand hygiene, improving environmental cleaning, and promoting antimicrobial stewardship. Although vertical and horizontal approaches are not mutually exclusive and are often intermixed, some experts believe that the horizontal approach under usual endemic situations may offer the best overall value given the diversity of microorganisms that can cause HAIs and the constrained resources available for infection prevention efforts. When informed by local knowledge of microbial epidemiology and ecology and supported by a strong quality improvement program, this strategy allows healthcare facilities to focus on approaches that target all rather than selected organisms in the absence of an organism-specific epidemic."

The Vertical Approach
Septimus, et al. (2014) explain that vertical approaches are often based on the results of active surveillance testing (AST), the rationale being that "multidrug-resistant organisms (MDROs) such as methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant enterococci (VRE), multidrug-resistant (MDR) gram-negative organisms, and Clostridium difficile share several epidemiological features: colonization can precede infection, transmission can occur by direct patient contact or indirect contact with contaminated equipment or environmental surfaces, the number of asymptomatic 'source' patients greatly exceeds the number of infected patients, and asymptomatic carriers can serve as the reservoir for spread to other patients. AST is used to identify patients who are carriers of these target pathogens so that these patients can be isolated from non-carriers and, in some situations, can undergo decolonization in order to eradicate pathogen carriage."

Studies indicate that the vertical approach has been used primarily for the prevention of MRSA transmission and infection. As Septimus, et al. (2014) observe, "More than 100 observational studies have evaluated the use of MRSA AST to target MRSA carriers for contact precautions, with or without supplemental decolonization. The effectiveness of AST in preventing MRSA transmission and infection continues to be controversial, and studies on this topic have yielded varying conclusions." For example, one study (Jain, et al) described a nationwide intervention in Veterans Affairs acute-care hospitals that included MRSA AST and contact precautions for MRSA carriers, improved compliance with hand hygiene, and an institutional culture change that was temporally associated with a large decline in infections caused by MRSA as well as other pathogens. However, in  another study (Huskins, et al.) -- a multicenter cluster-randomized, controlled trial in intensive care units (ICUs) -- demonstrated that an intervention involving MRSA AST plus universal gloving until a patient’s colonization status was known to be negative did not impact rates of MRSA colonization or infection.

The Horizontal Approach
The horizontal approach encompasses a number of interventions: Hand hygiene; decolonization of all patients in high-risk settings using topical chlorhexidine gluconate (CHG); selective digestive tract decontamination (SDD); antibiotic stewardship, and thorough environmental cleaning.

A number of experts say that the multi-modal horizontal approach works best for most hospitals' HAI prevention efforts. Wenzel and Edmond (2010) observe that "A very important point is that no hospital with a vertical (MRSA) approach has shown a major reduction in the rate of all infections or of all bloodstream infections. A recent analysis suggested why the MRSA (vertical) program is a flawed approach: the favorable outcomes of a horizontal program dwarf the vertical program in terms of reduced mortality, years of life lost, and costs. (Wenzel, et al. 2008)  In most hospitals, MRSA comprise 14 percent of all nosocomial infections. There are also new concerns related to the biology of S. aureus and nosocomial community-acquired MRSA (CA-MRSA): throat carriage only of S. aureus has been shown in 25 percent of carriers, and MRSA throat-carriage only in 13 percent to 15 percent of carriers. (Widmer, et al. 2007; Marshall et al. 2007)  Moreover, in a prospective cohort of carriers or infected patients with CA-MRSA, the organism was found in the nares in only 41 percent. (Zafar, et al. 2007) Thus, nasal screening will fail frequently with this emerging nosocomial pathogen. A fair question is, 'Is there any value to MRSA screening?' A recent meta-analysis suggests that for preoperative carriers of non-MRSA S. aureus, screening followed by decolonization would prevent S. aureus surgical site infections. (van Rijen, et al. 2008) Thus, currently it is reasonable to screen patients undergoing high-risk procedures such as orthopedic implants and cardiac surgery."

They add that, "The future of infection control will rely on a platform of strong horizontal programs committed to repeatedly reducing infections by 50 percent every three to four years. The critical infection control question that must be addressed before an institution invests new resources for a vertical program is this: What is its incremental value above that of the existing horizontal program?"

Septimus says he believes that the horizontal approach to infection prevention is still the best tactic, especially in light of the evolution of bacterial and viral strains and an age of inappropriate antibiotic prescribing. "It benefits many pathogens and sites, and the one exception is dealing with outbreaks." Septimus acknowledges that the type of approach can be decided on the local level, given individual healthcare institutions' experiences with pathogens, but that one strategy observed by all healthcare facilities regardless of local circumstances could reduce variation in practice. "If a facility has sporadic HAIs and are not experiencing high endemic or outbreaks, then a horizontal approach provides greater value," Septimus says. "However, if rates are high with a specific pathogen, then a vertical approach short-term may be preferable."

A Mixed Approach
There are some unique situations where using the best of both approaches may work -- especially when it comes to asymptomatic colonization in Clostridium difficile transmission leading to hospital-associated CDI. Curry, et al. found that among 114 active surveillance (AS) cultures performed on patients more than eight days before their first positive toxin assay (when they were symptomatic), 10 cultures from seven patients were positive for isolates highly related to the isolate recovered later from the patient's first toxin-positive stool. The researchers reported that at least 29 percent of hospital-acquired CDI results from transmission from asymptomatic carriers.

McDonald (2013) observes that "Although intriguing, this finding by Curry, et al. … is [not] sufficient to provoke revision in current surveillance definitions. The total number of patients in whom the prior cultures were performed was not reported; however, assuming a similar ratio of cultures to patients (i.e., 10:7) for all 114, <10 percent of evaluable CDI patients had incubation periods over seven days-possibly representing the 'trailing off' of a skewed normal distribution of incubation periods with a median of <3 days and a somewhat larger mean." McDonald (2013) adds that "The main conclusion of the report by Curry, et al. is sensitive to several important factors. First, the study setting was a tertiary hospital with a seasoned infection control program focused on containing transmission from CDI patients. An average rate of 5.6 HA-CDI cases per 10 000 patient-days (2006-2012) supports the assertion of an effective program at interrupting transmission from CDI cases, especially given the hospital's tertiary care status with intensive antibiotic exposures. Hospitals with less-effective infection control would have a higher proportion of all cases resulting from transmission from CDI patients."

McDonald (2013) says there are several hurdles to overcome before AS can be used in CDI prevention: "First is the need to rapidly detect colonization. Although Curry, et al. previously demonstrated accurate detection of colonization through use of a broth amplification culture followed by commercial NAAT, the preamplification incubation could last as long as 72 hours, calling into question the utility of the approach. Although existing commercial diagnostics could be refined or new ones developed to improve the rapid, sensitive detection of nearly all colonized patients, currently available NAATs may serve a useful purpose even without broth preamplification. For example, currently available NAATs alone may reliably detect a subset of colonized patients with higher organism loads who are more contagious as reflected by skin and environmental contamination. Once colonized patients are identified, special precautions will be needed to reduce transmission from these patients to other susceptible patients. In the case of other MDROs, this has most commonly entailed isolating colonized patients using contact precautions. However, criticism has been leveled at AS partly for this reason. It may be more cost-effective to apply broad, horizontal approaches to prevent all MDROs and healthcare-associated infections, such as improved hand hygiene for all patients, over a more burdensome vertical approach such as AS-directed isolation of all specific MDRO-colonized patients."

McDonald (2013) emphasizes that due to C. difficile being a spore-forming cause of healthcare-associated infection, "AS-directed special precautions may have particular relevance for C. difficile compared to other MDROs. For example, enhanced cleaning with a C. difficile sporicidal disinfectant and the use of gloves for patient care are more efficacious than standard methods at reducing environmental and hand contamination, respectively. Thus one approach would be to perform enhanced environmental cleaning, possibly along with the use of gloves for all patient care, for patients colonized with C. difficile." McDonald (2013) continues, "Even if its spore-forming status strengthens the case for using a vertical approach of AS-focused precautions to reduce C. difficile transmission, there are several special horizontal approaches to consider. One possible approach on particular ward locations with high rates of colonization and infection is universal gloving along with enhanced cleaning and environmental disinfection. Perhaps in the future, improved hand hygiene and environmental disinfection products can be developed that better remove or deactivate spores and yet can be used more broadly. Regardless, another important horizontal approach, the importance of which cannot be overstated, is antibiotic stewardship. Because reducing unnecessary antibiotic exposures decreases the risk of colonization and infection by a number of MDROs, especially CDI, a highly effective stewardship intervention could easily overshadow the impact of further reducing C. difficile transmission through AS-focused measures. A third possible horizontal approach with rapidly broadening evidence base in the prevention of CDI is the use of probiotics in patients receiving antibiotics. However, thus far there is not a similar level of evidence for probiotics preventing colonization or infection from MDROs other than C. difficile."

Financial Considerations
While clinical imperatives may exist for each strategy, fiscal ramifications cannot be overlooked. As Septimus, et al. (2014) explain, "In addition to comparing the strength of evidence supporting each approach, it is also important to take into account financial costs and potential consequences associated with various infection prevention strategies, including the impact on hospital personnel effort and on aspects of patient care; for example, placing patients on isolation precautions may lead to fewer healthcare provider visits. These comparisons are difficult to make because of conflicting study results, at least partly reflecting the heterogeneity of study designs and settings (i.e., where the prevalence of the target pathogen ranges from rare to endemic to epidemic) and the paucity of high-quality cost-effectiveness analyses that are needed to estimate the economic impact of specific HAI prevention interventions."

Septimus and his colleagues state that it is "increasingly important to understand the relative benefits and costs of pathogen-specific screening and intervention strategies compared with reliable application of more generic methods to mitigate transmission and infection." Septimus acknowledges that healthcare worker practices drive this momentum and that, "Most HAIs that are preventable usually results from a breakdown on basic practices. Using a pathogen-specific approach for routine practice defers vital resources from hard wiring basic practices that reduce infections for all pathogens such as CHG bathing in the ICU."
Wide variation in infection prevention and epidemiology practice could be contributing to mixed results in pathogen control and prevention, as well as patient outcomes. As Septimus says, "Human factors and variability of practices does contribute to increased risk. Understanding the local culture and supporting teamwork reduces risk and decreases turnover rates.  It is very important to include bedside healthcare personnel in discussions so teams can understand local culture and work flow. A redesign of the process should be local, creating local ownership for best practice which results in better compliance and importantly sustainability. There is a big difference between ownership and buy-in."

In their commentary, Septimus and his colleagues admit that "Our current ability to adequately compare the cost effectiveness of horizontal and vertical HAI prevention strategies or combinations of these strategies across healthcare settings is severely limited by the absence of robust data."
In terms of boosting the data, Septimus says, "We are seeing better funding for pragmatic trial which are more generalizable to the average facility.  Project are now adding a cost analysis to more of their studies to make sure the interventions provide not just better care, but better value."

Septimus, et al. (2014) note that "Reliable implementation is critical for either vertical or horizontal strategies. As new data emerge, prevention measures known to be effective should be integrated into care, applied reliably, and sustained. Quality improvement programs can play an important role in facilitating change and ensuring that implementation and intra-institutional spread respect local contexts. Unintended consequences should be anticipated and monitored." Septimus says that "Translational/behavioral science is limited, but what has been reported is that facilities with better teamwork environment have better outcomes and lower turnover rates. Understanding the adaptive local culture should be part of designing and implementing a safety initiative."

Septimus and his colleagues add in their commentary that Given the evolving epidemiology of MDROs and the complexity of managing the multiplicity of epidemiologically important pathogens across heterogeneous healthcare settings, however, we recommend (1) using robust quality improvement methods to ensure reliable performance of basic infection prevention practices known to mitigate transmission of MDROs and the infections they cause; (2) ensuring adherence to evidence-based universally applied HAI prevention strategies including hand hygiene, antimicrobial stewardship, and adequate environmental cleaning; (3) applying other evidence-based, horizontal strategies such as universal decolonization in settings where benefits are likely to outweigh risks and costs; and (4) using AST and other vertical approaches selectively when epidemiologically important pathogens are newly emerging and rare to a given institution or region or to control outbreaks of specific pathogens."

References

Edmond MB. Developing your approach to infection prevention. HAIControversies.blogspot.com Jan. 4, 2011.

McDonald LC. Looking to the Future: Vertical vs Horizontal Prevention of Clostridium difficile Infections. Clin Infect Dis. 2013.

Marshall C, Spelman D. Re: is throat screening necessary to detect methicillin-resistant Staphylococcus aureus colonization in patients upon admission to an intensive care unit? J Clin Microbiol 2007;45:3855.

Septimus E, MD, Weinstein RA, Perl TM, Goldmann DA and Yokoe DS. Commentary: Approaches for Preventing Healthcare-Associated Infections: Go Long or Go Wide? Infection Control and Hospital Epidemiol. Vol. 35, No. 7, July 2014.

van Rijen MM, Bonten M, Wenzel RP, Kluytmans JA. Intranasal mupirocin for reduction of Staphylococcus aureus infections in surgical patients with nasal carriage: a systematic review. J Antimicrob Chemother 2008;61:254-61.

Wenzel RP and Edmond MB. Infection control: the case for horizontal rather than vertical interventional programs. International Journal of Infectious Diseases 14S4 (2010) S3–S5.

Wenzel RP, Bearman G, Edmond MB. Screening for MRSA: A flawed hospital infection control intervention. Infect Control Hosp Epidemiol. 2008;29:1012-8.

Widmer A, Mertz D, Frei R. Necessity of screening both the nose and the throat to detect methicillin-resistant Staphylococcus aureus colonization in patients upon admission to an intensive care unit. J Clin Microbiol 2008; 45:835.

Zafar U, Johnson LB, Hanna M, Riederer K, Sharma M, Fakih MG, et al. Prevalence of nasal colonization among patients with community-acquired methicillin-resistant Staphylococcus aureus infection and their household contacts. Infection Control Hosp Epidemiol 2007;28:966-9.