Author Affiliations: Faculty of Medicine, University of Toronto, Toronto, Ontario (Dr Naylor); Kings Fund, London, England (Dr Chantler); and Department of Public Health, University of Oxford, Oxford, England (Dr Griffiths).
The recurrence of severe acute respiratory syndrome (SARS) in China
during 2004 has highlighted the continuing threat to human health from infectious
disease outbreaks. A zoonosis caused by a novel coronavirus,1,2 SARS
first emerged among humans in the southern Chinese province of Guangdong during
November 2002. By March 2003, SARS had spread to neighboring Hong Kong and
from there to Toronto, Ontario, and many other areas in a matter of days.
The World Health Organization (WHO) has reported that by July 2003 when
the epidemic had waned, in Hong Kong there were 1755 probable cases of SARS
with 300 deaths (17%) and in Canada there were 251 probable cases with 43
deaths (17%).3 Most Canadian cases and all
deaths were in the Toronto area. Both areas had serious difficulties managing
the outbreak, and several inquiries into public health and epidemic management
have since been performed. We led the panels that first reported on SARS and
public health in each jurisdiction. Both panels worked through the summer
of 2003 and issued their reports within a week of one another in early October
2003.4,5 Herein, we compare our
findings, highlight common conclusions, and suggest some general lessons that
may be applicable to other areas.
In Toronto and throughout Canada, health care is provided through universal
health insurance plans administered by each province. Hospitals rely on public
grants and physicians work primarily as private state-dependent contractors.6 About 45% of Ontario's 12 million individuals live
in the greater Toronto area. Coordination of the response to SARS in the Toronto
area was limited by 4 factors. First, Ontario is the only Canadian province
that has no regional health authorities; thus, 28 general hospitals in Toronto
function semiautonomously. Second, although they are paid by public insurance,
physicians work as private fee-for-service practitioners. Third, public health
units are funded partly by municipalities, and therefore also operate semiautonomously.
Four different public health units in the Toronto area worked to contain the
SARS outbreak, but did not share even the same information system. Finally,
the federal role in public health and clinical care is limited constitutionally,
creating potential challenges for national and international coordination
during a health emergency.
Hong Kong is a Special Administrative Region of the People's Republic
of China and its 6.8 million residents live on a 1100-km square area at the
southeastern tip of the country. Population density is among the highest in
the world at 6300 individuals per square kilometer. Hong Kong's economy is
closely linked to the neighboring Pearl River Delta, and large numbers of
individuals move daily between Hong Kong and the mainland. Hong Kong is therefore
an international transit zone for infectious diseases that continue to emerge
in south China. Overall responsibility for health in Hong Kong rests with
the Health, Welfare, and Food Bureau. Both the Department of Health, which
oversees public health among other services, and the Hospital Authority, which
runs the public hospitals and related outpatient services for Hong Kong, report
to the Secretary of Health, Welfare, and Food. However, while the public health
system is available for all citizens, a fee-for-service private sector also
exists. The private sector is substantial, providing primary, specialized,
and institutional care services to a large number of individuals.
The Canadian system operates on 3 levels (federal, provincial, and municipal)
while Hong Kong operates using 1. However, common features include the prominent
role of autonomous private health care institutions and clinicians, uneven
connections between public health and private sector health care, and financial
pressures on publicly funded clinical systems.
In Hong Kong, the outbreak spread first in the community, but eventually
22% of all persons affected were health care workers. In Toronto, where community
spread was more limited, the proportion of SARS cases among health care workers
was even higher at 43%.3 Straightforward protection
measures against droplet and contact transmission proved reasonably effective,7 but the spread of the infection to health care workers
added to the stress both areas faced in combating the outbreak. Along with
other jurisdictions, the 2 areas fought an outbreak in which, initially, no
satisfactory laboratory tests were available to confirm the diagnosis in suspected
cases. Even after reverse-transcriptase polymerase chain reaction methods
were instituted, rapid confirmation was not possible during the presymptomatic
or early stage of infection. Furthermore, both areas also dealt with travel
advisories from other jurisdictions that contributed to economic dislocation
and public upset.
In both Toronto and Hong Kong, containment of SARS relied heavily on
application of public health and clinical infection-control measures rooted
in 19th-century science. Both jurisdictions struggled to overcome several
key problems with capacity, preparedness, coordination, and communication.
We believe similar inadequacies exist in other countries.8
Alert systems proved inadequate. Authorities from the WHO in Beijing
were informed in December 2002 about an unusual outbreak in Guangdong, but
the WHO global alert about SARS was not issued until March 2003; and China
itself did not issue appropriate alerts. Thus, the Hong Kong government was
unaware of an outbreak of a novel form of atypical pneumonia in southern China
during the months of November and December 2002. Fortunately, background surveillance
was in place because of anxiety about the avian influenza virus, and this
was enhanced in February as accounts of an unusual epidemic in China surfaced.
Canadian alert systems were slow and fragmentary, although the province of
British Columbia successfully forewarned regional clinicians, hospitals, and
public health units through its well-organized Centre for Disease Control.
Absent outbreak plans, coordination across institutions and health service
sectors was poor and strong linkages between public health and the health
care system for epidemic response were absent. Moreover, when SARS appeared,
alerts and directives initially targeted acute care institutions, while long-term
care institutions and general physicians received little guidance and support
until well into the outbreak in Toronto and Hong Kong.
As a corollary, overall lines of authority were blurred. In the Toronto
outbreak, confusion was initially compounded by a management structure involving
distinct leadership for health care, public health, and general emergency
response. In Hong Kong, it was not clear who was in charge of the epidemic
Both areas were hampered by underinvestment in public health infrastructure,
diminution of public health leadership, and weak links between health care
and public health. Health professionals and administrators in the Toronto
area criticized poor coordination across public health units. Also, there
was general criticism of the weak analytical capacity of the Ontario Public
Health Branch and its performance in leading the outbreak response. In Hong
Kong, health professionals criticized the Department of Health for a deficiency
shared by the public health system and the private health sector: a failure
to establish a common population-based approach to outbreak management and
weak links between public health and private sector health services in institutional
and community settings.
As was true for other Canadian provinces, Ontario lacked interjurisdictional
agreements for sharing public health and clinical personnel. Toronto therefore
remained short of specialized staff for the first few weeks of the outbreak.
Hong Kong accommodated the increase in demand at first through diversion of
patients into a cluster of public hospitals and service networks, but that
system was finally overwhelmed by the massive increase in health care demand
caused by SARS. All Toronto-area hospitals were initially subject to provincial
directives that shut down elective activity, leading to a massive backlog
of deferred services. During a second wave of SARS, a network of 4 Toronto
hospitals focused on SARS care, allowing full activity to be maintained at
the other 24 hospitals. Moreover, regularized processes were not in place
in either region for sharing and compensating staff appropriately during an
Institutional infection control was undercut by shortages in personnel
and deficits in knowledge of frontline caregivers. Neither jurisdiction had
enough infection control practitioners and infectious disease specialists.
For example, 42% of Canadian hospitals fail to meet the current US standard
of 1 infection control practitioner (generally nurses or laboratory technologists)
per 250 active care beds; 80% cannot attain the new Canadian standard of 1
infection control practitioner per 175 active care beds.6,9 In
both Hong Kong and Toronto, failure of institutional syndromic surveillance
allowed SARS cases to go undetected and contributed to secondary spread of
the outbreak after an initial wave of SARS cases. Both reviews highlighted
concerns about a widespread lack of knowledge of basic principles of infection
control on the part of health care workers, presumably because few had ever
faced a serious outbreak of infectious disease. This situation, along with
deficiencies in provision of personal protective equipment in multiple sectors
of both health care systems, contributed to complaints and grievances about
occupational health and safety in health care settings in Hong Kong and Toronto.
Physical institutional deficiencies were also evident. Hong Kong had
poorly designed ventilation systems, lack of basic hand-washing and sanitary
facilities, and a shortage of single rooms with independent bathrooms. In
the Toronto area, only 3.8% of acute hospital beds were in single, negative-pressure
rooms. A number of hospitals in both jurisdictions lacked infection-control
areas in their emergency departments.
Both reports noted issues with laboratory coordination and research
responses to the outbreak. In Toronto, SARS testing was performed increasingly
by local hospital laboratories after the regional public health reference
laboratories were overwhelmed by the volume of testing. None of the laboratories
was adequately linked to each other, to clinicians, or to clinical and epidemiological
data in general. Hong Kong was better able to establish laboratory capacity
and link data sources through its E-SARS network. Researchers in both Hong
Kong and Toronto were active in delineating the cause of SARS, characterizing
the agent, developing diagnostic tests, and generating initial clinical descriptions.
However, data integration greatly enhanced research in Hong Kong, underpinning
an outpouring of epidemiological studies during and after the outbreak. Notwithstanding
its smaller outbreak and larger research infrastructure, Toronto generated
much less clinical and epidemiological research on SARS.
Risk communication to health care workers and with the general public
was uneven, compounded by the lack of certainty about diagnosis, treatment,
and epidemiology throughout the outbreaks. Confusion in Hong Kong was augmented
by open public conflict between senior health care professionals and lack
of identification of a single source for information. This was remedied later
in the epidemic when daily press briefings were instituted with the director
of health. In Toronto, similar leadership tensions were kept from public view,
but regular press briefings involved 3 and sometimes 4 spokespersons, and
the overall communications strategy was strongly criticized by expert commentators.10
Various jurisdictions issued advice against travel to both Toronto and
Hong Kong during the outbreak. The WHO also began issuing travel advisories
for the first time. The WHO's advice to postpone all but essential travel
to Hong Kong and China's Guangdong Province was issued on April 2, 2003. Three
weeks later, the WHO added Toronto, Beijing, and China's Shanxi Province to
areas that travelers should avoid. The epidemiological foundation for these
travel advisories was weak and international inconsistencies were apparent,
but the adverse economic effects were profound.
Health-related border controls were another source of controversy throughout
the outbreak and afterward.6 Toronto airports
used information cards with screening questions and secondary assessments
and interviews as needed. A large-scale Canadian pilot project involving thermal
scanners resulted in millions of screening transactions and thousands of referrals
of passengers for further assessment by nurses or quarantine officers, but
not a single case of SARS was identified. In Hong Kong, border controls using
temperature checking and screening forms remain in place, especially now that
SARS has recurred in China. There is no international consensus on appropriate
travel screening and border controls.
The recommendations from both reports respond to parallel problems in
organization, capacity, and structure of health care and public health systems
(Box 1). Our comparison of
the 2 reports also suggests 5 overarching lessons
Public Health Capacity
Review and revise legislation and regulations regarding public health
and health emergencies; invest in training public health and infectious disease
specialists as well as infection-control training for all staff; develop improved
analytical and leadership capacity for public health; train more epidemiologists,
public health nurses, and microbiologists; do more cross-training to link
health care and public health clinicians; and develop flexible and progressive
career paths for public health professionals.
Create new Canadian public health agency with administrative headquarters
in nation's capital and major collaborating centers across Canada; invest
in provincial public health capacity; generate business process agreements
to link provincial and federal public health activities. Create a center for
health protection in Hong Kong and review central government departmental
structures for public health and health care.
Enhance communications with mainland China (especially Guangdong Province)
through network for communicable disease control and strengthen 2-way information
technology for disease surveillance and outbreak management across all sectors
of health care in Hong Kong.
Adopt an Internet-enabled public health information system across all
jurisdictions throughout Canada and link public health information system
with electronic patient records and other health care data sources.
Infection Control Policies, Plans, and Procedures
Develop integrated protocols for outbreak management at a population
level in Toronto and Hong Kong, linking institutional and noninstitutional
clinicians; test protocols through simulated outbreaks and emergencies; audit
preparedness on a regular basis; train health care staff in infection control
and outbreak response as applicable.
Develop efficient information systems for contact tracing, concatenating
laboratory with clinical and epidemiological data, and tracking patients through
the health care system in Toronto and in Hong Kong.
Create multidisciplinary health care emergency teams and specific epidemic
response teams; develop standardized and efficient business processes for
rapid epidemic and outbreak investigation; ensure availability of personal
Agree on mutual aid arrangements across jurisdictions throughout Canada
and remove interprovincial barriers to certification and licensure. Ensure
policies for outbreak response are graded to minimize impact on routine patient
care in Toronto and Hong Kong.
Health Care Facilities
Ensure adequate numbers of isolation and negative-pressure rooms across
differing levels of acuity, including emergency departments. Improve hospital
ward design including ventilation systems, hand-washing stations, and other
Laboratory Facilities and Research Capacity
Develop mechanisms for laboratory data sharing; define the roles of
laboratory facilities in health care and public health under outbreak scenarios;
ensure quality control for laboratories testing for infectious pathogens.
Strengthen capacity for outbreak investigation and related infectious
disease research; support research to establish best practices for infection
control and outbreak response; collect and share evidence for optimal public
health protection; set up protocols in advance for rapid activation of research
response to other potential disease outbreaks.
Enhance communication within health care and public health to ensure
rapid dissemination of alerts, updates, and directives. Train institutional,
public health, and health care leaders in risk and crisis communication.
Occupational Health Issues
Determine special compensation for staff who take on extraordinary responsibilities
during future epidemics and disasters; ensure availability of personal protective
equipment at all relevant sites; and improve internal communication regarding
infection-control policies. Strengthen occupational health systems in institutions
and other sectors of the health system.
Work with the World Health Organization and other international partners
on the rational provision of travel advice and revision of international health
regulations and strengthen international links in infectious disease surveillance.
Review access control measures (eg, airport and border screening) in Hong
Kong and across Canada, including Toronto.
Infectious Diseases Remain a Serious Global Threat
Despite many serious new epidemics starting with human immunodeficiency
virus and AIDS, the continued challenges of antibiotic resistance, and the
reality of global spread of novel pathogens in hours or days, the global health
community continues to underinvest in infectious disease control measures.
Governments Underinvest in Public Health
Because of its preventive and population focus, the public health system
is consistently overshadowed by other areas of health care. However, investments
in public health are vitally important for health protection (including safe
food and water supplies and effective sanitation), disease and injury prevention
(including outbreak containment), and health promotion.
An Outbreak Requires Attention in Advance to Clarity,
Collaboration, Communication, Coordination, and Capacity
Clarity is needed to determine the lines of authority and responsibilities
during an outbreak. Collaboration—whether within jurisdictions or across
them—presupposes shared values and goals, trust, goodwill, and agreed
rules of engagement. Communication within health care and the public health
system, with the media and public, and between nations is a cornerstone of
crisis management. Coordination of activities is essential and depends on
all the foregoing elements. No outbreak can be contained efficiently and effectively
without prior investments in capacity at all levels of health care and the
public health system.
Integration Within Health Care and Between Health
Care and the Public Health System Facilitates Crisis Management
Integration and alignment of incentives leads to enhanced quality and
efficiency of usual and customary health care services. However, severe acute
respiratory syndrome (SARS) has highlighted that integration is even more
valuable in the face of a health crisis. Given its nosocomial transmission,
the SARS outbreak particularly underscores the importance of seamless integration
between health care and the public health system.
A Global Public Health System Is Essential
A global, interdependent health system is essential for detection and
containment of infectious threats, whether naturally emerging or products
of bioterrorism. Every nation should maintain integrated surveillance systems
for infectious diseases and bioterrorist threats, and share information on
potential threats fully and freely with all other nations. As a corollary,
humanitarian measures taken by wealthier nations to support health protection
and disease control in less wealthy nations may pay dividends in global safety
and security, while promoting a fairer and stronger global economy.
Not surprisingly, there were differences in emphasis between the 2 reports.
The Hong Kong report points to a potential role for traditional Chinese medicine.
Canada's National Advisory Committee on SARS and Public Health addressed federal
and provincial issues not applicable to Hong Kong. However, the Hong Kong
report highlights the need for a clearer relationship with mainland China,
and this may be seen as an analog for the federal and provincial challenges
identified in Canada. Even on matters as specific as legislation, there are
more similarities than differences. For example, building on the debates and
experiences of quarantining residents from Amoy Gardens, the Hong Kong report
highlights out-of-date legislation and recommends a review and revision of
public health and health-related emergency regulation. The Canadian quarantine
experience was not as controversial, but a similar review of legislation and
regulations was recommended for health emergencies across the provincial and
On a positive note, the involved governments have been responding. The
Canadian federal government is moving forward to create a national agency
for public health, spun out from the current bureaucratic structures. The
2004 federal budget in Canada committed additional funds for the new agency,
for a network of collaborating centers to protect and promote health across
the country, for enhancing local public health capacity, and for provincial
immunization programs. Meanwhile, the Ontario government is actively framing
its responses to a recently released provincial report on SARS, whose recommendations
strongly reinforce the federal report in regard to provincial investment and
restructuring.11 Nonetheless, progress toward
frontline preparedness for another major outbreak has been slow and problems
with inadequate capacity for public health persist in most provinces.
In Hong Kong, the administration is progressing steadily toward the
creation of the Centre for Health Protection as recommended by the international
panel. This includes physical infrastructure, recruitment of personnel, funding,
and active links with Guangdong. Work is also under way on developing information
systems and surveillance. Officials from Hong Kong have acknowledged the importance
of integrated training for specialists in infectious diseases, microbiology,
and public health. Infection-control measures in Hong Kong nursing homes have
been augmented, and closer links are being forged slowly between the private
sector, particularly family physicians, and the public sector. Linkages with
the Hospital Authority are still suboptimal, and the suggested review and
reorganization of the central administration of health has not occurred.
The SARS story as we saw it in Hong Kong and Toronto had both tragic
and heroic elements. The outbreak took a major toll, with deaths, illness,
upheaval, and hardship on multiple levels in both these major metropolitan
areas. Thousands of public health and health care workers rose to the occasion
and ultimately contained the SARS outbreak in the 2 areas, notwithstanding
systems and resources that were suboptimal.
It remains a mystery why other major areas were spared. However, in
this era of rapid global travel, any city and area could face an outbreak
at any time. In comparing the findings of 2 inquiries after SARS in Toronto
and Hong Kong, we have been struck that these 2 areas, among others, missed
opportunities to learn from each other during the outbreak, notwithstanding
electronic communication and the laudable attempts at brokerage and facilitation
by the WHO.
The challenge now is to ensure better preparation on a global scale
for the next epidemic or threat, whether from SARS, a bioterrorism attack,
or a novel infectious agent. International communication and collaboration,
including sharing these lessons learned, remains an essential part of that
process of preparation.
Naylor CD, Chantler C, Griffiths S. Learning From SARS in Hong Kong and Toronto. JAMA. 2004;291(20):2483-2487. doi:10.1001/jama.291.20.2483