Association of Visitor Contact Precautions With Estimated Hospital-Onset Clostridioides difficile Infection Rates in Acute Care Hospitals | Gastroenterology | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Layout of the Agent-Based Model
Layout of the Agent-Based Model

Agent-based simulation model of Clostridiodides difficile transmission in an acute care, tertiary adult hospital. This model replicated events associated with C difficile infection in a 200-bed dynamic hospital environment representing an average-sized US adult hospital and included 4 types of individuals as agents: patients, visitors, nurses, and physicians.

Figure 2.  Logic of Visitor Agent Flow in the Agent-Based Model
Logic of Visitor Agent Flow in the Agent-Based Model

Pathways of visitors’ Clostridiodides difficile propagation based on visitor behavior, as shown in Barker et al.4

Figure 3.  Compliance Associated With at Least 1% Decrease in Hospital-Onset Clostridioides difficile Infections per 10 000 Patient-Days by Intervention
Compliance Associated With at Least 1% Decrease in Hospital-Onset Clostridioides difficile Infections per 10 000 Patient-Days by Intervention
Table 1.  Modeled Experiment Scenarios
Modeled Experiment Scenarios
Table 2.  Association of Deimplementation vs Implementation of VCPs With CDI Rates by Experiment
Association of Deimplementation vs Implementation of VCPs With CDI Rates by Experiment
1.
Lessa  FC, Mu  Y, Bamberg  WM,  et al.  Burden of Clostridium difficile infection in the United States.   N Engl J Med. 2015;372(9):825-834. doi:10.1056/NEJMoa1408913 PubMedGoogle ScholarCrossref
2.
McDonald  LC, Gerding  DN, Johnson  S,  et al.  Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA).   Clin Infect Dis. 2018;66(7):e1-e48. doi:10.1093/cid/cix1085 PubMedGoogle ScholarCrossref
3.
Munoz-Price  LS, Banach  DB, Bearman  G,  et al.  Isolation precautions for visitors.   Infect Control Hosp Epidemiol. 2015;36(7):747-758. doi:10.1017/ice.2015.67 PubMedGoogle ScholarCrossref
4.
Barker  AK, Alagoz  O, Safdar  N.  Interventions to reduce the incidence of hospital-onset Clostridium difficile infection: an agent-based modeling approach to evaluate clinical effectiveness in adult acute care hospitals.   Clin Infect Dis. 2018;66(8):1192-1203. doi:10.1093/cid/cix962 PubMedGoogle ScholarCrossref
5.
Anderson  DJ, Kirkland  KB, Kaye  KS,  et al.  Underresourced hospital infection control and prevention programs: penny wise, pound foolish?   Infect Control Hosp Epidemiol. 2007;28(7):767-773. doi:10.1086/518518 PubMedGoogle ScholarCrossref
6.
Dubberke  ER, Carling  P, Carrico  R,  et al.  Strategies to prevent Clostridium difficile infections in acute care hospitals: 2014 update.   Infect Control Hosp Epidemiol. 2014;35(S2)(suppl 2):S48-S65. doi:10.1017/S0899823X00193857 PubMedGoogle ScholarCrossref
7.
Sharma  A, Pillai  DR, Lu  M,  et al.  Impact of isolation precautions on quality of life: a meta-analysis.   J Hosp Infect. 2020;105(1):35-42. doi:10.1016/j.jhin.2020.02.004 PubMedGoogle ScholarCrossref
8.
Abad  C, Fearday  A, Safdar  N.  Adverse effects of isolation in hospitalised patients: a systematic review.   J Hosp Infect. 2010;76(2):97-102. doi:10.1016/j.jhin.2010.04.027 PubMedGoogle ScholarCrossref
9.
Day  HR, Perencevich  EN, Harris  AD,  et al.  Depression, anxiety, and moods of hospitalized patients under contact precautions.   Infect Control Hosp Epidemiol. 2013;34(3):251-258. doi:10.1086/669526 PubMedGoogle ScholarCrossref
10.
Findik  UY, Ozbaş  A, Cavdar  I, Erkan  T, Topcu  SY.  Effects of the contact isolation application on anxiety and depression levels of the patients.   Int J Nurs Pract. 2012;18(4):340-346. doi:10.1111/j.1440-172X.2012.02049.x PubMedGoogle ScholarCrossref
11.
Smith  L, Medves  J, Harrison  MB, Tranmer  J, Waytuck  B.  The impact of hospital visiting hour policies on pediatric and adult patients and their visitors.   JBI Libr Syst Rev. 2009;7(2):38-79. doi:10.11124/jbisrir-2009-181PubMedGoogle ScholarCrossref
12.
Engström  A, Söderberg  S.  Receiving power through confirmation: the meaning of close relatives for people who have been critically ill.   J Adv Nurs. 2007;59(6):569-576. doi:10.1111/j.1365-2648.2007.04336.x PubMedGoogle ScholarCrossref
13.
Barker  AK, Scaria  E, Safdar  N, Alagoz  O.  Evaluation of the cost-effectiveness of infection control strategies to reduce hospital-onset Clostridioides difficile infection.   JAMA Netw Open. 2020;3(8):e2012522-e2012522. doi:10.1001/jamanetworkopen.2020.12522 PubMedGoogle ScholarCrossref
14.
Codella  J, Safdar  N, Heffernan  R, Alagoz  O.  An agent-based simulation model for Clostridium difficile infection control.   Med Decis Making. 2015;35(2):211-229. doi:10.1177/0272989X14545788 PubMedGoogle ScholarCrossref
15.
National Healthcare Safety Network, Centers for Disease Control and Prevention. Surveillance for C. difficile, MRSA, and other drug-resistant infections. Accessed July 7, 2020. https://www.cdc.gov/nhsn/ltach/cdiff-mrsa/index.html
16.
Python Software Foundation. Python. Accessed January 20, 2021. https://www.python.org/
17.
Barratt  RL, Shaban  R, Moyle  W.  Patient experience of source isolation: lessons for clinical practice.   Contemp Nurse. 2011;39(2):180-193. doi:10.5172/conu.2011.180 PubMedGoogle ScholarCrossref
18.
Stelfox  HT, Bates  DW, Redelmeier  DA.  Safety of patients isolated for infection control.   JAMA. 2003;290(14):1899-1905. doi:10.1001/jama.290.14.1899 PubMedGoogle ScholarCrossref
19.
Day  HR, Perencevich  EN, Harris  AD,  et al.  Association between contact precautions and delirium at a tertiary care center.   Infect Control Hosp Epidemiol. 2012;33(1):34-39. doi:10.1086/663340 PubMedGoogle ScholarCrossref
20.
Gammon  J.  Analysis of the stressful effects of hospitalisation and source isolation on coping and psychological constructs.   Int J Nurs Pract. 1998;4(2):84-96. doi:10.1046/j.1440-172X.1998.00084.x PubMedGoogle ScholarCrossref
21.
Granzotto  EM, Gouveia  AM, Gasparetto  J, Dantas  LR, Tuon  FF.  Depression and anxiety in hospitalized patients on contact precautions for multidrug-resistant microorganisms.   Infect Dis Health. 2020;25(3):133-139. doi:10.1016/j.idh.2020.01.002 PubMedGoogle ScholarCrossref
22.
Haessler  S, Martin  EM, Scales  ME,  et al.  Stopping the routine use of contact precautions for management of MRSA and VRE at three academic medical centers: An interrupted time series analysis.   Am J Infect Control. 2020;48(12):1466-1473. doi:10.1016/j.ajic.2020.06.219PubMedGoogle ScholarCrossref
23.
Martin  EM, Russell  D, Rubin  Z,  et al.  Elimination of routine contact precautions for endemic methicillin-resistant staphylococcus aureus and vancomycin-resistant enterococcus: a retrospective quasi-experimental study.   Infect Control Hosp Epidemiol. 2016;37(11):1323-1330. doi:10.1017/ice.2016.156PubMedGoogle ScholarCrossref
24.
Guh  AY, Mu  Y, Winston  LG,  et al; Emerging Infections Program Clostridioides difficile Infection Working Group.  Trends in U.S. burden of Clostridioides difficile infection and outcomes.   N Engl J Med. 2020;382(14):1320-1330. doi:10.1056/NEJMoa1910215 PubMedGoogle ScholarCrossref
25.
McLaughlin  AM, Canavan  JB, Adams  EJ,  et al  A survey of MRSA awareness and knowledge among the general public and patients’ visitors.   J Infect Prev. 2008;9(5):18-23. doi:10.1177/1469044608095540 Google Scholar
26.
Philipp  R, Hughes  A, Wood  N, Burns-Cox  C, Cook  N, Fletcher  G.  Information needs of patients and visitors in a district general hospital.   J R Soc Health. 1990;110(1):10-12, 14. doi:10.1177/146642409011000104 PubMedGoogle ScholarCrossref
27.
Seibert  G, Ewers  T, Barker  AK,  et al.  What do visitors know and how do they feel about contact precautions?   Am J Infect Control. 2018;46(1):115-117. doi:10.1016/j.ajic.2017.05.011 PubMedGoogle ScholarCrossref
28.
Stone  SP, Fuller  C, Savage  J,  et al.  Evaluation of the national CleanYourHands campaign to reduce Staphylococcus aureus bacteraemia and Clostridium difficile infection in hospitals in England and Wales by improved hand hygiene: four year, prospective, ecological, interrupted time series study.   BMJ. 2012;344:e3005. doi:10.1136/bmj.e3005 PubMedGoogle Scholar
29.
Ragusa  R, Giorgianni  G, Lupo  L,  et al.  Healthcare-associated Clostridium difficile infection: role of correct hand hygiene in cross-infection control.   J Prev Med Hyg. 2018;59(2):E145-E152.PubMedGoogle Scholar
30.
Mayfield  JL, Leet  T, Miller  J, Mundy  LM.  Environmental control to reduce transmission of Clostridium difficile.   Clin Infect Dis. 2000;31(4):995-1000. doi:10.1086/318149 PubMedGoogle ScholarCrossref
31.
Jacobs A. Grave shortages of protective gear flare again as Covid cases surge. New York Times. Published July 8, 2020. Accessed July 13, 2020. https://www.nytimes.com/2020/07/08/health/coronavirus-masks-ppe-doc.html
32.
Livingston  E, Desai  A, Berkwits  M.  Sourcing personal protective equipment during the COVID-19 pandemic.   JAMA. 2020;323(19):1912-1914. doi:10.1001/jama.2020.5317 PubMedGoogle ScholarCrossref
33.
Morgan  DJ, Pineles  L, Shardell  M,  et al.  The effect of contact precautions on healthcare worker activity in acute care hospitals.   Infect Control Hosp Epidemiol. 2013;34(1):69-73. doi:10.1086/668775 PubMedGoogle ScholarCrossref
34.
Dhar  S, Marchaim  D, Tansek  R,  et al.  Contact precautions: more is not necessarily better.   Infect Control Hosp Epidemiol. 2014;35(3):213-221. doi:10.1086/675294 PubMedGoogle ScholarCrossref
35.
Barker  AK, Scaria  E, Alagoz  O, Sethi  AK, Safdar  N.  Reducing C. difficile in children: an agent-based modeling approach to evaluate intervention effectiveness.   Infect Control Hosp Epidemiol. 2020;41(5):522-530. doi:10.1017/ice.2020.14 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Infectious Diseases
    February 26, 2021

    Association of Visitor Contact Precautions With Estimated Hospital-Onset Clostridioides difficile Infection Rates in Acute Care Hospitals

    Author Affiliations
    • 1Department of Industrial and Systems Engineering, University of Wisconsin–Madison, Madison
    • 2Department of Internal Medicine, University of Michigan, Ann Arbor
    • 3Population Health Sciences, School of Medicine and Public Health, University of Wisconsin–Madison, Madison
    • 4Division of Infectious Diseases, Department of Medicine, School of Medicine and Public Health, University of Wisconsin–Madison, Madison
    • 5William S. Middleton Memorial Veterans Hospital, Madison, Wisconsin
    JAMA Netw Open. 2021;4(2):e210361. doi:10.1001/jamanetworkopen.2021.0361
    Key Points

    Question  Is there an association between visitor contact precautions and hospital-onset Clostridioides difficile infection in acute care hospitals?

    Findings  In this simulation study of a generic acute care adult hospital, in all experiments, use of visitor contact precautions was associated with a decrease of 1% or less in the rate of hospital-onset C difficile infection; small improvements in hand hygiene and environmental cleaning were associated with larger decreases in hospital-onset C difficile infection.

    Meaning  In this study, visitor contact precautions were associated with minimal reductions in hospital-onset C difficile infections, suggesting that hospitals may achieve a larger reduction through other interventions.

    Abstract

    Importance  Visitor contact precautions (VCPs) are commonly used to reduce the transmission of Clostridioides difficile at health care institutions. Implementing VCPs requires considerable personnel and personal protective equipment resources. However, it is unknown whether VCPs are associated with reduced hospital-onset C difficile infection (HO-CDI) rates.

    Objective  To estimate the association between VCPs and HO-CDI rates using simulation modeling.

    Design, Setting, and Participants  This simulation study, conducted between July 27, 2020, and August 11, 2020, used an established agent-based simulation model of C difficile transmission in a 200-bed acute care adult hospital to estimate the association between VCPs and HO-CDI while varying assumptions about factors such as patient susceptibility, behavior, and C difficile transmission. The model simulated hospital activity for 1 year among a homogeneous, simulated adult population.

    Interventions  No VCP use vs ideal use of VCPs under different hospital configurations.

    Main Outcomes and Measures  The rate of HO-CDI per 10 000 patient-days according to the Centers for Disease Control and Prevention’s definition of HO-CDI.

    Results  With use of the simulation model, the baseline rate of HO-CDI was 7.94 10 000 patient-days (95% CI, 7.91-7.98 per 10 000 patient-days) with no VCP use compared with 7.97 per 10 000 patient-days (95% CI, 7.93-8.01 per 10 000 patient-days) with ideal VCP use. Visitor contact precautions were not associated with a reduction of more than 1% in HO-CDI rates in any of the tested scenarios and hospital settings. Independently increasing the hand-hygiene compliance of the average health care worker and environmental cleaning compliance by no more than 2% each was associated with greater HO-CDI reduction compared with all other scenarios, including VCPs.

    Conclusions and Relevance  In this simulation study, the association between VCPs and HO-CDI was minimal, but improvements in health care worker hand hygiene and environmental cleaning were associated with greater reductions in estimated HO-CDI. Hospitals may achieve a higher rate of reduction for HO-CDI by focusing on making small improvements in compliance with interventions other than VCP.

    Introduction

    Clostridioides difficile infection (CDI) is a major infection associated with health care, and its containment is considered a public health priority.1 In health care settings, C difficile transmission is believed to occur primarily through the physical interactions among patients, health care workers, visitors, and the environment. Recent guidelines from leading infectious diseases societies recommend several infection control interventions to curtail C difficile transmission.2,3 Thus, many institutions have developed CDI prevention bundles, which incorporate multiple intervention strategies simultaneously. Implementing complex bundles with high fidelity requires extensive resources.4 However, infection control programs are markedly underfunded at hospitals, with only half of hospitals budgeting money specifically for an infection control department.5 In the setting of known financial constraints, it is crucial to identify and prioritize the most promising interventions to reduce hospital-onset CDI (HO-CDI).

    Prevention bundles for CDI typically include a visitor contact precautions (VCPs) component, which requires visitors to don gowns and gloves when entering the room of a patient with CDI. In a recent survey of 245 North American hospitals, 71% offered education materials to visitors about contact precautions for C difficile and 82% had specific contact-precaution policies for visitors of patients with CDI.3 The 2014 infectious diseases guidelines consider the use of contact precautions for visitors who enter the room of a patient with CDI as an unresolved issue without substantial supporting evidence; infection control guidelines recommend the use of VCPs for patients with CDI but also note a lack of supporting evidence.3,6 The 2017 infectious diseases guidelines specifically recommend contact precautions for health care workers.2

    Using contact precautions for patients with CDI when they are cared for by health care workers is essential to reduce transmission from patient to patient and from patient to health care worker. However, visitors, unlike health care workers, do not interact with multiple patients and do not usually participate in the physical care of the patient. Thus, the benefit of VCPs is unclear. Moreover, VCPs may be associated with adverse effects for patients if their visitors are required to don gowns and gloves before interacting with them. Unlike many other infection control interventions, in addition to the cost and resource implications, there may be an association between quality of life and contact precautions.7 Some studies have reported increased delirium and depression in patients for whom contact precautions are used; other studies have not found such an association.7-10 It is possible that adverse effects associated with contact precautions are increased by VCPs because isolating patients from their visiting family and friends could be distressing. Receiving visitors has been shown to be associated with greater satisfaction with health care among patients, but patients under contact precautions may receive a smaller number of visitors.11,12 Any potential benefits associated with VCPs must be considered in the context of detriments in patient outcomes. The risks and costs associated with VCPs in typical conditions may outweigh the benefits for 2 other health care–associated infections: methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus (VRE) infections. Recent guidelines recommend against VCPs for endemic MRSA and VRE infections because these infections are prevalent in the community.3

    Data on the association of VCPs with CDI transmission are essential to guide decision-making regarding this intervention. To our knowledge, no study has estimated the association of VCPs with hospital-onset CDI. The purpose of this study was to use simulation modeling to estimate the association between VCPs and HO-CDI. Estimating this association was challenging because it may differ according to hospital and community characteristics. For example, there is likely a differential association of VCPs with the rate of HO-CDI in a hospital where most patients are susceptible to CDI compared with a hospital where a small proportion of the patients are susceptible. Simulation modeling is inherently able to test large numbers of configurations, and as such, it is a useful alternative to conventional epidemiologic studies in this context.

    Methods
    Overview of the Agent-Based Simulation Model

    This simulation study was conducted between July 27, 2020, and August 11, 2020, using a previously developed agent-based simulation model of C difficile transmission in an acute care, tertiary adult hospital.4 This model replicated events associated with CDI in a 200-bed dynamic hospital environment representing an average-sized US adult hospital and included 4 types of individuals as agents: patients, visitors, nurses, and physicians (Figure 1).4 The University of Wisconsin Health Sciences Institutional Review Board approved the study and waived the need for informed consent because the data were deidentified. The study followed the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) reporting guideline.

    The model represented the CDI status of the patients using 1 of 9 possible disease states. The model updated the clinical state of each patient in the hospital every 6 hours using a discrete-time Markov chain structure (eTable 1 and the eFigure in the Supplement). Patients in the colonized, infected, recolonized, or recurrent infection states can transmit C difficile to other agents and the environment through interactions with them. Visitors, nurses, and physicians who are contaminated with C difficile from patients or the environment can in turn transmit C difficile to other susceptible patients and the environment. Similarly, any agent interacting with a contaminated environment can become exposed to C difficile. The probability of transmission for each interaction depends on the agent types involved and the duration of the interaction. The key parameters associated with C difficile transmission and other model inputs are presented in eTable 2 in the Supplement.

    The initial model was developed using NetLogo software, version 5.3.1 (Center for Connected Learning and Computer-Based Modeling, Northwestern University) and subsequently translated into the Java programming language (Oracle Corp) for faster speeds and flexibility as described in the eAppendix and eTables 3-7 in the Supplement.4 All experiments were conducted using the Java version of the model. The original NetLogo model was validated through face validation, parameter sensitivity analyses, and cross-validation, which involved comparing model outputs with those from C difficile infection–control studies in the literature. Both versions used common random numbers generated by the Colt Project’s Mersenne Twister algorithm to reduce variation and to directly compare runs under different intervention scenarios.4,13 A detailed description of the model is provided elsewhere.4,14

    Visitor Agent Logic

    On each day in the simulation, there was a probability that a patient received up to 2 visitors. Visitors did not interact with other patients or health care workers in the hospital. Visitors could become transiently exposed to and contagious with C difficile through interactions with the patient with whom they visited and with contaminated environments. The probability of visitor exposure to C difficile in a patient room increased with the visit duration and depended on the transfer efficiency between surfaces. Visitors could wash their hands after exiting from the patient room according to hand hygiene recommendations regardless of VCP use. Visitors exited the hospital through a ward common room, which they contaminated with probability dependent on their length of stay in the common room. The baseline model assumed that visitors remained in the common room for 5 minutes before exiting. Each ward common room may also have been used by patients staying within the same ward. Therefore, a patient may have become exposed to C difficile through a common area contaminated by a visitor. Environment-to-agent contamination was possible through high-touch surfaces, which represent a portion of the room. Common rooms may have been cleaned daily depending on compliance with the bleach cleaning intervention. We assumed that visitors were healthy adults who were not colonized or infected with C difficile. Therefore, visitors could not shed C difficile but could expose and contaminate the environment. Figure 2 shows the logic governing visitor actions in the simulation, and the eAppendix in the Supplement further describes visitor-agent logic.

    Simulated Interventions

    The model included the following core infection control interventions: nurse and physician hand hygiene, nurse and physician contact precautions, daily environmental cleaning, terminal environmental cleaning, visitor hand hygiene, visitor contact precautions, patient hand hygiene, and C difficile surveillance screening at admission. All interventions were implemented at 3 increasing levels of compliance: baseline, enhanced, and ideal. The baseline level corresponded to typical compliance with minimal institutional support or, for the surveillance intervention, no asymptomatic screening at admission. Enhanced and ideal implementation represented increasing levels of institutional support. Assumptions of baseline, enhanced, and ideal compliance for the core infection-control interventions were derived from the studies cited in eTable 2 in the Supplement. Parameters representing high-risk antibiotic use in the hospital remained constant for all interventions and were only changed when explicitly examined by the experiments.

    Simulation Settings and Outcomes

    We simulated C difficile transmission at a generic 200-bed hospital for 13 months, including a 1-month warmup period. Data were collected during the following 12 simulated months. The primary output was the HO-CDI rate, calculated as the number of HO-CDI diagnoses made per 10 000 patient-days. In accordance with the definition from the Centers for Disease Control and Prevention, a CDI was classified as hospital onset if it was detected more than 3 days after admission.15 All CDIs detected within 3 days of admission were classified as community onset.

    Simulation Scenarios

    We conducted 10 sets of simulation experiments to test the value of VCPs in different hospital settings under a variety of infection control implementation strategies. These experiments tested parameter assumptions associated with community distribution of CDI and susceptibility to infection, compliance of health care workers with hand hygiene protocols, C difficile transmission from visitors, number of visitors in the hospital, compliance of visitors with hand hygiene protocols, compliance with other infection control interventions, the discrete-time Markov chain representing patients’ disease-state transitions, health care worker distribution and behavior, and timing of patient arrivals (Table 1). The 10 experiment sets also included 1 experiment defined a posteriori, which considered the “worst case” hospital using parameters from analogous experiments most conducive to infection transmission.

    Baseline assumptions for the experiments are provided in eTable 2 in the Supplement. Individual experiments are described in detail in Table 1. In all the experiments, we compared 2 scenarios to estimate the maximum possible association between VCPs and HO-CDI: VCPs deimplemented (ie, no use of VCPs) vs VCPs implemented at an ideal level of compliance (93.5%). All interventions that were not assessed in the simulation scenario were implemented at their baseline level.

    We then estimated the minimum required improvement in compliance from baseline that the interventions of health care workers’ hand hygiene, daily cleaning, and terminal cleaning would need to achieve a similar or greater association with HO-CDI rate reduction compared with ideal implementation of the VCP intervention.

    Statistical Analysis

    We ran 5000 replications for each scenario to obtain stable estimates. Means and 95% CIs were calculated for all experiments. We used the SciPy Stats package (Enthought, Inc) to build 95% CIs with the Python programming language.16 We considered the association of VCPs with HO-CDI to be negligible if the difference between the control and experimental means was less than 1%.

    Results

    The overall baseline rate of HO-CDI when all model input parameters were set at baseline and VCPs were deimplemented was 7.94 per 10 000 patient-days (95% CI, 7.91-7.98 per 10 000 patient-days). This changed to 7.97 per 10 000 patient-days (95% CI, 7.93-8.01 per 10 000 patient-days) with VCPs implemented at the ideal level (Table 2). The association between VCPs and HO-CDI rates followed expected trends. For example, an increased proportion of susceptible patients (HO-CDI rate with VCPs deimplemented, 17.18 per 10 000 patient-days; 95% CI, 17.13-17.24 per 10 000 patient-days), lower hand-hygiene compliance among health care workers (HO-CDI rate with VCPs deimplemented, 14.43 per 10 000 patient-days; 95% CI, 14.38-14.48 per 10 000 patient-days), and lower environmental cleaning compliance (HO-CDI rate with VCPs deimplemented, 11.32 per 10 000 patient-days; 95% CI, 11.28-11.37 per 10 000 patient-days) were associated with higher estimated rates of HO-CDI. Across all experiments, VCPs were associated with a change of less than 1% in the rates of HO-CDI (Table 2). The experiments varying the proportion of susceptible, colonized, and infected patients at admission, which correspond to assumptions regarding the prevalence of CDI in the community, suggested that the association between VCPs and HO-CDI was negligible across all admission populations. The reduction in HO-CDI rates was less than 1% even in the worst-case scenario.

    The experiments with the largest changes in the baseline HO-CDI rate were associated with patient disease states (experiment set 1 in Table 2) and high-impact interventions, such as hand hygiene of health care workers (experiment 2a) and daily cleaning (experiment 6a). Experiments sets 4 and 5, which changed the distribution and behavior of patients in the simulation, were associated with a change of <1% in the overall baseline HO-CDI rate of 7.94 per 10 000 patient days.

    Figure 3 and eTable 8 in the Supplement summarize the results of incremental-improvement experiments that increased the mean compliance with hand hygiene among health care workers from a baseline estimate of 55% to 56% (60% to 61% for nurses and 50% to 51% for physicians) and compliance with environmental cleaning from 47% to 49% (46% to 47% for daily cleaning and 47% to 50% for terminal cleaning). For health care workers’ hand hygiene and environmental cleaning, improvements in compliance of 3% or less were associated with greater CDI reduction compared with ideal VCP implementation.

    Discussion

    To our knowledge, this is the first study to estimate the association of VCPs with HO-CDI rates. This simulation study suggests that VCPs have a minimal association with the overall HO-CDI rate and that similar benefits could be obtained by increasing compliance with core infection-control interventions. such as health care workers’ hand hygiene or environmental cleaning. The VCPs were not associated with a great reduction in HO-CDI using simulation models with any of the examined hospital configurations or patient profiles. The adaptability of this model may enable these experiments to be run for different hospital configurations and may help estimate the association between VCPs and HO-CDI in specific settings.

    Because no previous study, to our knowledge, has evaluated the association of VCPs with HO-CDI, direct comparison with previous studies was not possible. However, multiple studies17-21 have described the harms associated with contact precautions for patients. Previous studies reported increased delirium and depression associated with contact precautions.18-20 Patients under contact precautions also reported lower care satisfaction.21 It is likely that these negative outcomes are associated with the use of VCPs.17

    Other studies have questioned the effectiveness of VCPs for MRSA and VRE infections.3,22,23 The 2015 Society for Healthcare Epidemiology of America (SHEA) report3 states that VCPs are an ineffective intervention because of the high prevalence of MRSA and VRE in the community. As community spread is increasingly recognized in the context of C difficile, with some studies noting a general increase in community-onset CDI since the early 1990s, a similar phenomenon may explain the potential low impact of VCPs.24 The SHEA guidelines3 also suggest that visitor behavior, particularly that visitors normally do not interact with multiple patients, may be a reason why this intervention is associated with little reduction in rates of HO-CDI.

    High fidelity and sustained implementation of VCPs require considerable resources. The SHEA guidelines3 report the difficulty of educating visitors and enforcing compliance with contact-precaution practices as reasons against the use of contact precautions among visitors of patients with MRSA and VRE infection. These reasons likely generalize to C difficile because the recommended gowns, gloves, and visitor education for MRSA and VRE infections are similar to those for CDI.

    Our findings may have implications for health care institutions, infection prevention programs, and clinicians. The minimal changes to the baseline HO-CDI rate associated with VCPs suggest that any restrictions regarding visitor length of stay or number of visitors allowed may be relaxed with minimal change in HO-CDI rates. Removing VCPs for CDI control may also be associated with decreased health care worker burden related to enforcing and managing compliance with VCPs, allowing additional time and effort to be allocated toward more effective interventions.25-27 These interventions include hand hygiene of health care workers and environmental cleaning, which have consistently been associated with significant decreases in HO-CDI.28-30

    The personal protective equipment (PPE) resources saved by removing VCPs can also be used to improve PPE inventory for use during acute infectious outbreaks. For example, in the coronavirus disease 2019 outbreak, guaranteeing PPE to health professionals became a challenge, with calls for removing contact precautions for patients with MRSA and VRE infection during the pandemic.31,32 Minimizing PPE use among visitors to rooms of patients with CDI would provide much-needed additional resources for health care workers at seemingly minimal risks for worsening C difficile transmission.

    Limitations

    This study has limitations. First, all data analyzed in this study were generated through simulation and not through clinical experiments. Therefore, any conclusions drawn from this study should at most be used to influence the purpose or design of such clinical experiments. Second, we did not consider the effect of community-onset CDI on hospital visitors. The model assumed that visitors could not be infected or colonized, and thus, visitors were not able to expose patients to C difficile spores from the community. This study did not consider the benefit associated with VCPs among visitors. The model also simulated each ward uniformly and did not account for differential CDI risk across wards. We also did not consider how the use or deimplementation of VCPs may affect compliance with other infection control measures, such as hand hygiene or precautions for health care worker contact. It is possible that deimplementing VCPs would suggest to visitors and health care workers a diminished risk of C difficile, which in turn may be associated with decreased compliance with other infection-control practices that are highly effective against HO-CDI. A recent study of health care workers found that the use of contact precautions was associated with increased risk-mitigating behavior, such as improved hand hygiene.33 However, another study suggests that a threshold of patients designated under isolation exists at which compliance with contact precautions decreases.34 These studies suggest interactions between infection control interventions and their perceived importance; further work is required to understand whether such an association could be present after deimplementation of VCPs among patients with CDI. This study also did not examine the impact of VCPs in a pediatric setting, where visitor-patient interactions differ from the adult, acute care hospital setting. However, a previous agent-based model35 of C difficile transmission in a pediatric hospital showed little reduction in HO-CDI in association with implementation of VCPs at an ideal level under baseline assumptions. Further research is required to estimate the contribution of VCPs in a pediatric care setting under different configurations. In addition, the assumption that visitors can only contaminate the environment, not health care workers or other patients, may be a limitation. However, given that most visitors do not physically interact with more than 1 patient, it is unlikely that changing this assumption would greatly affect the current model outcomes.

    Conclusions

    In this simulation study, VCPs were associated with minimal change in HO-CDI rates. In the context of these findings and financial and supply-chain constraints, the infection control community should reevaluate the value of VCPs for patients with CDI. If this model is correct, removing VCPs, which are associated with minimal change in HO-CDI rates, from prevention bundles may allow better allocation of resources for containment.

    Back to top
    Article Information

    Accepted for Publication: January 10, 2021.

    Published: February 26, 2021. doi:10.1001/jamanetworkopen.2021.0361

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Scaria E et al. JAMA Network Open.

    Corresponding Author: Nasia Safdar, MD, PhD, Division of Infectious Diseases, Department of Medicine, School of Medicine and Public Health, University of Wisconsin–Madison, 5138 Medical Foundation Centennial Building, 1685 Highland Ave, Madison, WI, 53705 (ns2@medicine.wisc.edu).

    Author Contributions: Drs Alagoz and Safdar were co–senior authors for this article. Ms Scaria and Dr Alagoz had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Scaria, Alagoz, Safdar.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Scaria, Alagoz, Safdar.

    Critical revision of the manuscript for important intellectual content: Scaria, Barker, Alagoz.

    Statistical analysis: Scaria, Alagoz.

    Obtained funding: Safdar.

    Administrative, technical, or material support: Safdar.

    Supervision: Alagoz, Safdar.

    Conflict of Interest Disclosures: Dr Alagoz reported receiving grants from the National Institute of Allergy and Infectious Disease, National Institutes of Health (NIH) during the conduct of the study, serving as a paid consultant to Biovector Inc, and receiving personal fees from Biovector Inc and Bristol Myers Squibb. Dr Safdar reported receiving grants from NIH during the conduct of the study. No other disclosures were reported.

    Funding/Support: This work was supported by the National Institutes of Health under National Institute of Allergy and Infectious Diseases Institute (NIAID) grant 1DP2AI144244-01 (Dr Safdar).

    Role of the Funder/Sponsor: The funder had no role in the design and conduct of the research; collection, analysis, and interpretation of the data; preparation, review, and approval of the manuscript; or the decision to submit the manuscript for publication.

    Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

    References
    1.
    Lessa  FC, Mu  Y, Bamberg  WM,  et al.  Burden of Clostridium difficile infection in the United States.   N Engl J Med. 2015;372(9):825-834. doi:10.1056/NEJMoa1408913 PubMedGoogle ScholarCrossref
    2.
    McDonald  LC, Gerding  DN, Johnson  S,  et al.  Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA).   Clin Infect Dis. 2018;66(7):e1-e48. doi:10.1093/cid/cix1085 PubMedGoogle ScholarCrossref
    3.
    Munoz-Price  LS, Banach  DB, Bearman  G,  et al.  Isolation precautions for visitors.   Infect Control Hosp Epidemiol. 2015;36(7):747-758. doi:10.1017/ice.2015.67 PubMedGoogle ScholarCrossref
    4.
    Barker  AK, Alagoz  O, Safdar  N.  Interventions to reduce the incidence of hospital-onset Clostridium difficile infection: an agent-based modeling approach to evaluate clinical effectiveness in adult acute care hospitals.   Clin Infect Dis. 2018;66(8):1192-1203. doi:10.1093/cid/cix962 PubMedGoogle ScholarCrossref
    5.
    Anderson  DJ, Kirkland  KB, Kaye  KS,  et al.  Underresourced hospital infection control and prevention programs: penny wise, pound foolish?   Infect Control Hosp Epidemiol. 2007;28(7):767-773. doi:10.1086/518518 PubMedGoogle ScholarCrossref
    6.
    Dubberke  ER, Carling  P, Carrico  R,  et al.  Strategies to prevent Clostridium difficile infections in acute care hospitals: 2014 update.   Infect Control Hosp Epidemiol. 2014;35(S2)(suppl 2):S48-S65. doi:10.1017/S0899823X00193857 PubMedGoogle ScholarCrossref
    7.
    Sharma  A, Pillai  DR, Lu  M,  et al.  Impact of isolation precautions on quality of life: a meta-analysis.   J Hosp Infect. 2020;105(1):35-42. doi:10.1016/j.jhin.2020.02.004 PubMedGoogle ScholarCrossref
    8.
    Abad  C, Fearday  A, Safdar  N.  Adverse effects of isolation in hospitalised patients: a systematic review.   J Hosp Infect. 2010;76(2):97-102. doi:10.1016/j.jhin.2010.04.027 PubMedGoogle ScholarCrossref
    9.
    Day  HR, Perencevich  EN, Harris  AD,  et al.  Depression, anxiety, and moods of hospitalized patients under contact precautions.   Infect Control Hosp Epidemiol. 2013;34(3):251-258. doi:10.1086/669526 PubMedGoogle ScholarCrossref
    10.
    Findik  UY, Ozbaş  A, Cavdar  I, Erkan  T, Topcu  SY.  Effects of the contact isolation application on anxiety and depression levels of the patients.   Int J Nurs Pract. 2012;18(4):340-346. doi:10.1111/j.1440-172X.2012.02049.x PubMedGoogle ScholarCrossref
    11.
    Smith  L, Medves  J, Harrison  MB, Tranmer  J, Waytuck  B.  The impact of hospital visiting hour policies on pediatric and adult patients and their visitors.   JBI Libr Syst Rev. 2009;7(2):38-79. doi:10.11124/jbisrir-2009-181PubMedGoogle ScholarCrossref
    12.
    Engström  A, Söderberg  S.  Receiving power through confirmation: the meaning of close relatives for people who have been critically ill.   J Adv Nurs. 2007;59(6):569-576. doi:10.1111/j.1365-2648.2007.04336.x PubMedGoogle ScholarCrossref
    13.
    Barker  AK, Scaria  E, Safdar  N, Alagoz  O.  Evaluation of the cost-effectiveness of infection control strategies to reduce hospital-onset Clostridioides difficile infection.   JAMA Netw Open. 2020;3(8):e2012522-e2012522. doi:10.1001/jamanetworkopen.2020.12522 PubMedGoogle ScholarCrossref
    14.
    Codella  J, Safdar  N, Heffernan  R, Alagoz  O.  An agent-based simulation model for Clostridium difficile infection control.   Med Decis Making. 2015;35(2):211-229. doi:10.1177/0272989X14545788 PubMedGoogle ScholarCrossref
    15.
    National Healthcare Safety Network, Centers for Disease Control and Prevention. Surveillance for C. difficile, MRSA, and other drug-resistant infections. Accessed July 7, 2020. https://www.cdc.gov/nhsn/ltach/cdiff-mrsa/index.html
    16.
    Python Software Foundation. Python. Accessed January 20, 2021. https://www.python.org/
    17.
    Barratt  RL, Shaban  R, Moyle  W.  Patient experience of source isolation: lessons for clinical practice.   Contemp Nurse. 2011;39(2):180-193. doi:10.5172/conu.2011.180 PubMedGoogle ScholarCrossref
    18.
    Stelfox  HT, Bates  DW, Redelmeier  DA.  Safety of patients isolated for infection control.   JAMA. 2003;290(14):1899-1905. doi:10.1001/jama.290.14.1899 PubMedGoogle ScholarCrossref
    19.
    Day  HR, Perencevich  EN, Harris  AD,  et al.  Association between contact precautions and delirium at a tertiary care center.   Infect Control Hosp Epidemiol. 2012;33(1):34-39. doi:10.1086/663340 PubMedGoogle ScholarCrossref
    20.
    Gammon  J.  Analysis of the stressful effects of hospitalisation and source isolation on coping and psychological constructs.   Int J Nurs Pract. 1998;4(2):84-96. doi:10.1046/j.1440-172X.1998.00084.x PubMedGoogle ScholarCrossref
    21.
    Granzotto  EM, Gouveia  AM, Gasparetto  J, Dantas  LR, Tuon  FF.  Depression and anxiety in hospitalized patients on contact precautions for multidrug-resistant microorganisms.   Infect Dis Health. 2020;25(3):133-139. doi:10.1016/j.idh.2020.01.002 PubMedGoogle ScholarCrossref
    22.
    Haessler  S, Martin  EM, Scales  ME,  et al.  Stopping the routine use of contact precautions for management of MRSA and VRE at three academic medical centers: An interrupted time series analysis.   Am J Infect Control. 2020;48(12):1466-1473. doi:10.1016/j.ajic.2020.06.219PubMedGoogle ScholarCrossref
    23.
    Martin  EM, Russell  D, Rubin  Z,  et al.  Elimination of routine contact precautions for endemic methicillin-resistant staphylococcus aureus and vancomycin-resistant enterococcus: a retrospective quasi-experimental study.   Infect Control Hosp Epidemiol. 2016;37(11):1323-1330. doi:10.1017/ice.2016.156PubMedGoogle ScholarCrossref
    24.
    Guh  AY, Mu  Y, Winston  LG,  et al; Emerging Infections Program Clostridioides difficile Infection Working Group.  Trends in U.S. burden of Clostridioides difficile infection and outcomes.   N Engl J Med. 2020;382(14):1320-1330. doi:10.1056/NEJMoa1910215 PubMedGoogle ScholarCrossref
    25.
    McLaughlin  AM, Canavan  JB, Adams  EJ,  et al  A survey of MRSA awareness and knowledge among the general public and patients’ visitors.   J Infect Prev. 2008;9(5):18-23. doi:10.1177/1469044608095540 Google Scholar
    26.
    Philipp  R, Hughes  A, Wood  N, Burns-Cox  C, Cook  N, Fletcher  G.  Information needs of patients and visitors in a district general hospital.   J R Soc Health. 1990;110(1):10-12, 14. doi:10.1177/146642409011000104 PubMedGoogle ScholarCrossref
    27.
    Seibert  G, Ewers  T, Barker  AK,  et al.  What do visitors know and how do they feel about contact precautions?   Am J Infect Control. 2018;46(1):115-117. doi:10.1016/j.ajic.2017.05.011 PubMedGoogle ScholarCrossref
    28.
    Stone  SP, Fuller  C, Savage  J,  et al.  Evaluation of the national CleanYourHands campaign to reduce Staphylococcus aureus bacteraemia and Clostridium difficile infection in hospitals in England and Wales by improved hand hygiene: four year, prospective, ecological, interrupted time series study.   BMJ. 2012;344:e3005. doi:10.1136/bmj.e3005 PubMedGoogle Scholar
    29.
    Ragusa  R, Giorgianni  G, Lupo  L,  et al.  Healthcare-associated Clostridium difficile infection: role of correct hand hygiene in cross-infection control.   J Prev Med Hyg. 2018;59(2):E145-E152.PubMedGoogle Scholar
    30.
    Mayfield  JL, Leet  T, Miller  J, Mundy  LM.  Environmental control to reduce transmission of Clostridium difficile.   Clin Infect Dis. 2000;31(4):995-1000. doi:10.1086/318149 PubMedGoogle ScholarCrossref
    31.
    Jacobs A. Grave shortages of protective gear flare again as Covid cases surge. New York Times. Published July 8, 2020. Accessed July 13, 2020. https://www.nytimes.com/2020/07/08/health/coronavirus-masks-ppe-doc.html
    32.
    Livingston  E, Desai  A, Berkwits  M.  Sourcing personal protective equipment during the COVID-19 pandemic.   JAMA. 2020;323(19):1912-1914. doi:10.1001/jama.2020.5317 PubMedGoogle ScholarCrossref
    33.
    Morgan  DJ, Pineles  L, Shardell  M,  et al.  The effect of contact precautions on healthcare worker activity in acute care hospitals.   Infect Control Hosp Epidemiol. 2013;34(1):69-73. doi:10.1086/668775 PubMedGoogle ScholarCrossref
    34.
    Dhar  S, Marchaim  D, Tansek  R,  et al.  Contact precautions: more is not necessarily better.   Infect Control Hosp Epidemiol. 2014;35(3):213-221. doi:10.1086/675294 PubMedGoogle ScholarCrossref
    35.
    Barker  AK, Scaria  E, Alagoz  O, Sethi  AK, Safdar  N.  Reducing C. difficile in children: an agent-based modeling approach to evaluate intervention effectiveness.   Infect Control Hosp Epidemiol. 2020;41(5):522-530. doi:10.1017/ice.2020.14 PubMedGoogle ScholarCrossref
    ×