Problem Statement
- We confront a threatening array of pathogens
- Healthcare facilities amplify dissemination
- We have limited resources for containment
- We lack local guidance for containment
Mission
- Facility Level Approaches To Infectious Control Engineering (FLATICE).
- To provide easy to use, broadly applicable analytic tools for optimizing pathogen containment policies in outpatient healthcare facilities, such as clinics or emergency departments.
We confront ominous viral, mycobacterial, and bacterial pathogens such as influenza,
extremely drug resistant Mycobacterium tuberculosis, Severe Acute Respiratory Syndrome (SARS) coronavirus, and methicillin resistant Staphylococcus aureus. Healthcare facilities can promote pathogen dissemination. Facility or network level interventions that limit the dissemination of pathogens will increase patient and provider safety, decrease morbidity, and reduce healthcare and societal costs. Computational approaches can elucidate the dynamics of pathogen dissemination and inform containment policies.
Influenza, whether seasonal or pandemic, is a particular concern. Given anticipated delays and limited supplies of vaccines, antiviral drugs, and infection control materiel, alternative strategies for attenuating pathogen dissemination in healthcare facilities are urgently needed. However, little guidance regarding optimal containment strategies is available for individual facilities. No facility level models directly address optimal resource utilization or caregiver assignment policies, key issues in developing viable containment strategies. We address these deficiencies along two axes.
Similarly, multidrug resistant Mycobacterium tuberculosis and the more recently
reported “extremely drug resistant” M. tuberculosis are growing threats with significant
public health implications. An effective vaccine is not available, and drug-based
treatments are rapidly becoming limited. Simple measures to attenuate transmission
might prove extremely helpful in containing these pathogens.
We have developed stochastic agent based models emulating a single healthcare provider, a single outpatient clinic, a network of multiple outpatient
clinics, and inpatient facilities of varying size, contact intensity, and spatial
complexity. Such “digital laboratories” allow the user to specify the length and number of queues, the certainty with which incident patients can be classified as infectious or noninfectious, pathogen transmissibility (allowing for asymmetrical transmission probabilities and multiple transmission routes), and various containment strategies, including changes in hand hygiene, barrier precautions of various types (mask, mask and gloves, etc), environmental decontamination, and changes in the scheduling and assignment policies of caregivers and patients. The models can be readily configured to reflect specific clinics, facilities, or networks of interest.
These analytic tools can be used to predict the relative benefits of different infection control strategies at the level of individual caregivers, facilities, and networks, delineate the sensitivity of these predictions to uncertainty in input parameters, and define anticipated variability in outcomes. Outputs include quantitative comparisons and rank ordering of the effectiveness and robustness of different infection control strategies, the risk of contamination faced by caregivers and uninfected patients, the cost per contamination event prevented, and the total cost to the system of each approach.
We are also developing software for characterizing and monitoring transmission dynamics. These tools should prove helpful in estimating the transmissibility of novel, poorly understood pathogens. Such tools could also be used to identify system-level failures in policy compliance (such as a decline in hand hygiene compliance or violation of scheduling practices), to monitor performance as a component of an incentive program, or both.
Although we have elected to focus on influenza, these tools are readily adaptable for analyses relating to other viral, mycobacterial, and bacterial pathogens.
Our initial work was funded by NIH grant NIAID 7R21AI055818-02
Current ASP.NET 2.0 and Windows 2.0 development services are being provided by
Holley Associates