• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Forum on Microbial Threats; Knobler S, Mahmoud A, Lemon S, et al., editors. Learning from SARS: Preparing for the Next Disease Outbreak: Workshop Summary. Washington (DC): National Academies Press (US); 2004.

Cover of Learning from SARS

Learning from SARS: Preparing for the Next Disease Outbreak: Workshop Summary.

Show details

MODELING A RESPONSE STRATEGY

, M.D., , M.P.A., and , M.D., M.P.H.

Department of Public Health, Weill Medical College of Cornell University

Containment of the 2002–2003 severe acute respiratory syndrome (SARS) epidemic posed unprecedented challenges to health care delivery and public health systems worldwide. In addition to the human costs of infection in medical workers, efforts to contain the spread of the virus led to widespread disruptions in the provision of routine medical care. Response strategies for potential recurrences of SARS will need to address treatment of infected individuals, quarantine of potential victims, and health system action plans that lead to containment of the outbreak without undue impact on the delivery of care for the wider populace. Since the outbreak of SARS, several computational models have been developed to investigate the transmission dynamics of the SARS coronavirus. Although these studies have identified certain parameters (e.g., maximum allowable delay in quarantining new cases) that may lead to more efficient management of new outbreaks, further research is needed to better define the practical steps required for such optimized response strategies. This chapter summarizes the current state of theoretical modeling for SARS and proposes a research agenda to improve forecasting of resource requirements at the hospital, health system, and regional levels for containment of future outbreaks.

Eight models of SARS transmission and control were published in the English and Chinese scientific literature in 2003 (Chen, 2003; Chowell et al., 2003; Lin et al., 2003; Lipsitch et al., 2003; Lloyd-Smith et al., 2003; Riley et al., 2003; Shi, 2003; Wang and Zhao, 2003). Seven of these utilize the standard SEIR (susceptible, exposed, infectious, recovered) dynamic mathematical model of disease transmission or variations on that model accounting for the use of quarantine (Table 5-1) (Chen, 2003; Chowell et al., 2003; Lipsitch et al., 2003; Lloyd-Smith et al., 2003; Riley et al., 2003; Shi, 2003; Wang and Zhao, 2003). SEIR models can provide estimates of critical parameters for a disease outbreak, such as the basic reproductive number R0 (that is, the number of new cases for every existing case) or maximal lag time for isolation of new cases (Dye and Gay, 2003).

TABLE 5-1. SARS Dynamic Transmission Models.

TABLE 5-1

SARS Dynamic Transmission Models.

Five of these studies consider outbreak response variables that reflect both public health activity (e.g., time to isolation of each new case) and hospital-based measures (e.g., efficacy of isolation and reduction in transmission rate of virus) (Chowell et al., 2003; Lipsitch et al., 2003; Lloyd-Smith et al., 2003; Riley et al., 2003; Shi, 2003). For example, Chowell and colleagues predicted that containment of the Canadian outbreak would require a time-to-isolation of 3 to 6 days and a 50 to 90 percent reduction in person-to-person transmission from identified cases (Chowell et al., 2003). In a similar fashion, most of these papers provide model-derived threshold values, but do not focus on the practical steps needed to attain them. Only one paper, by Lloyd-Smith and colleagues (2003), went into sufficient detail about methods of disease containment to provide practical guidance for public health and hospital managers in attaining these goals. For example, these authors found that efforts to interrupt health care worker-to-patient transmission would yield greater improvements in epidemic containment than reductions in population-based transmission. This finding provided the basis for a practical recommendation to initiate hospital-wide campaigns to increase contact precautions and strict case management of infected individuals. Additionally, this report alone—among the eight model-based papers—acknowledges that containment efforts would be carried out in an environment of limited hospital resources, where scarcity of items such as gowns, gloves, and masks would require prioritization of population-wide and hospital-based strategies.

These eight reports provide the beginnings of an evidence base on which to design effective response strategies for future SARS outbreaks. In parallel with these efforts, a number of researchers have developed prediction models for medical outcomes of SARS patients (Table 5-2) (Booth et al., 2003; Chan et al., 2003; Donnelly et al., 2003; Han et al., 2003; He et al., 2003). The current challenge is to use these findings from both theoretical modeling and patient care to assist health planners in practical ways. For example, hospital administrators may benefit from guidance on determining when in the course of an epidemic it is better to cease all admissions, isolate a specific ward, or simply isolate a number of patients in individual rooms. More complex response models may begin to weigh the relative benefits of drastic steps such as shuttering entire hospitals in order to contain the spread of SARS in light of the potential harms that may accrue to affected communities through the loss of routine medical care capacity. Such cost-benefit studies will highlight the difficult choices faced by health planners and hospital administrators in the real-world setting of financial and resource constraints. Finally, with the prospect of a SARS vaccine on the horizon, new models will be needed to quantify optimal pre- and post-detection vaccination rates for disease containment given the significant resource requirements of any mass vaccination campaign. Recent efforts to model mass antibiotic prophylaxis strategies for bioterrorism response may provide insight into the methods and data requirements for this type of logistical modeling as well as techniques (e.g., Internet-based platforms) for wide dissemination of modeling tools (Hupert and Cuomo, 2003; Hupert et al., 2002).

TABLE 5-2. Prediction Models for Medical Outcomes of SARS Patients.

TABLE 5-2

Prediction Models for Medical Outcomes of SARS Patients.

Publication of data on resources consumed in isolating and treating SARS patients as well as quarantine of potentially infected individuals will assist modelers in developing realistic forecasting models capable of leading public health and hospital planners through “what if” scenarios that may require difficult tradeoffs of personnel, materials, and patient care arrangements. The more accurate the data underlying these models, the better they can serve planners and their communities. The goal of such efforts should be to give every decision maker the ability to understand, in relevant terms and for their particular institution or community, not just the knowledge that containment of SARS would require isolation of new cases within a certain number of days, but also an estimate of how to go about achieving that containment goal (i.e., how many staff, rooms, media campaigns, and other factors). Planning models that focus on critical resources in this manner can provide guidance for live exercises and may influence future investments in both infrastructure (e.g., installation of negative pressure isolation rooms) and disposable medical equipment (e.g., gowns and masks).

Copyright © 2004, National Academy of Sciences.
Bookshelf ID: NBK92482
PubReader format: click here to try

Views

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...