Respiratory illnesses account for a large proportion of the total infectious disease burden worldwide. Each year, approximately 900,000 episodes of pneumonia occur among senior citizens1 in the United States alone, and influenza and pneumonia as a group are the only infectious diseases among the 10 leading causes of death in the United States.2 Most episodes of pneumonia occur as isolated cases, but many types of pneumonia and other serious respiratory illnesses can occur as outbreaks, with cases clustering together in place and time and occurring in numbers higher than the background rate. When outbreaks occur, achieving control requires rapid coordinated measures to identify cases and the sources of infection and to ultimately deliver the appropriate interventions.
The increasing number of humans infected with H5N1 avian influenza virus has provided another reason to worry about respiratory illnesses. An article in this issue of Infectious Diseases in Clinical Practice reviews the hematologic characteristics of human cases of avian influenza, highlighting the severity of this disease.3 In today's world of rapid intercontinental travel, the threat of an influenza pandemic and other, as of yet, unknown emerging diseases makes effective detection and control of infectious illnesses especially urgent. The outbreak of severe acute respiratory syndrome (SARS) in 2003 provided a useful lesson in how quickly an outbreak of respiratory disease can spread. By the time appropriate alarm had been raised, the disease identified, and the epidemic contained, infected persons had traveled to 26 countries around the world, spawning secondary outbreaks and causing over 8000 infections and 774 deaths.4 A large-scale international response was instrumental in achieving control of this outbreak.
Clinicians are the front line in the battle against infectious diseases. Clinicians are, of course, responsible for providing health care for individual patients; in the era of avian influenza, clinicians should also think about how their patients' illnesses might have broader implications. Public health officials rely on clinicians to detect and report to them cases of infections that could have ramifications for the greater population. Essential functions of public health agencies such as disease surveillance and control depend on a good working relationship between public health officials and health care providers and for providers to have a basic sense of which diseases are, for the good of others, important to identify in their individual patients.
SURVEILLANCE-TRACKING IMPORTANT RESPIRATORY DISEASES
Disease surveillance is a core function of the modern public health system and involves collecting information about circulating and emerging illnesses to take some action to control them. Clinicians, infection control practitioners, and laboratory personnel are the primary sources of data for surveillance programs. Surveillance data can prompt the investigation of disease clusters, allow implementation and evaluation of control measures, stimulate and direct research, and inform health policy. The present threat of avian influenza is a stimulus to refine and re-emphasize disease surveillance practices, as has been done throughout history in response to new disease epidemics.
There are several historical examples of the use of public health surveillance and control measures. In the 14th century during the bubonic plague pandemic, ships were required to sit at anchor in quarantine for 40 days (quarenta giorni), and public health officials boarded ships near Venice to prevent those with plaguelike illnesses from disembarking.5 In 1878, a serious epidemic of yellow fever with approximately 120,000 cases and 20,000 deaths within 2 months struck communities in New Orleans, Memphis, and the Mississippi River Valley.6 The National Quarantine Act was subsequently passed, authorizing the US Marine Hospital Service to collect data on infectious illness morbidity from consulates abroad and to effect quarantine measures to prevent the introduction of yellow fever, as well as cholera, smallpox, and plague into the United States. Previously limited to caring for sick and injured seamen abroad, the US Marine Hospital Service thus began evolving to what is now known as the US Public Health Service. In the ensuing decades, the US Public Health Service became incorporated into federal public health agencies, and state and local health departments grew. At the same time, an increasingly sophisticated system for national disease surveillance in the United States played a crucial role in the successful control of several epidemic diseases, including malaria, poliomyelitis, and smallpox.
The National Notifiable Diseases Surveillance System in place today in the United States is a collaboration between the Council of State and Territorial Epidemiologists (CSTE) and the Centers for Disease Control and Prevention (CDC). CSTE members, who include the lead public health officials from state and territorial health departments, annually review the list of diseases that should be reported by clinicians to public health officials. The CSTE list of nationally notifiable conditions is considered a recommendation, as state public health officials and state legislatures have the final say in determining which conditions clinicians in their state should report. The list of notifiable diseases includes many specific respiratory pathogens, in addition to a broad category of "unusual illnesses," to cover outbreaks of new diseases or diseases not routinely tracked (Table 1).
The process of disease surveillance starts when clinicians, infection control practitioners, or clinical laboratory workers recognize illnesses of public health concern in their patients. In turn, those recognizing these illnesses report them to their local county health department, and local officials can take immediate action if necessary (Fig. 1). Reports are subsequently forwarded from local to state health departments, who in turn communicate with CDC. Reporting is mandated at the state level, but reporting to the CDC by state health departments is voluntary except for certain diseases for which quarantine measures may be used, including yellow fever, cholera, and plague. State health department staff will often work with local public health officials on investigations and implementation of control measures, and CDC staff can assist with state investigations if invited by state public health officials. CDC often has the lead role in investigations of disease outbreaks involving multiple states. At the state and federal levels, broader evaluations of surveillance data looking for epidemiological patterns such as clustering among demographic groups or geographic and temporal clustering can lead to the development of prevention and control policies in addition to the investigation and control of large-scale disease outbreaks.
THE ROLE OF THE CLINICIAN
Public health programs depend on the initiative of clinicians, infection control practitioners, and laboratories staff to detect and notify health department staff of concerning illnesses. First, a clinician must consider a particular illness in his or her differential diagnosis. Next, he or she must order the appropriate test, and the test must accurately detect the infection. Finally, this newly diagnosed case must be reported to the health department. The clinician bears the primary responsibility for the first steps in disease reporting: consideration of a potentially reportable illness that may present unremarkably and the decision to submit specimens for etiologic laboratory diagnosis. Well-known examples from the last decade highlight the important role clinician reporting has in the successful identification and control of infectious respiratory diseases outbreaks; SARS, hantavirus pulmonary syndrome, and inhalational anthrax were detected and ultimately controlled because of timely reporting by astute practicing clinicians who knew that the patients they were seeing had unusual illnesses.7-9
Obstacles with diagnosis and reporting contribute to the wide variation of reporting completeness for the common notifiable illnesses, ranging from 6% to 90%.10-12 Infections caused by Mycobacterium tuberculosis are an example of good practice for a reportable respiratory infection, with one study detecting only 0.5% of tuberculosis cases going unreported.13 Although atypical cases do occur, physicians are usually adept at recognizing risk factors and clinical cues for most tuberculosis infections and will appropriately pursue an etiologic diagnosis and report discovered cases, allowing control measures to be activated by the local health department.
A key to picking up respiratory infections with public health implications is looking for clues in the history and physical examination that suggest these specific infections. A thorough history may elicit clues such as high-risk exposures and patients who may fit characteristic seasonal, geographic, or demographic high-risk groups. During the SARS outbreak, health care providers became acutely sensitized to look for a history of recent travel to affected parts of Asia in patients with severe respiratory disease. Some have suggested that attention to specific clinical and epidemiological features can reliably distinguish certain pathogens, such as Hantavirus.14 However, these examples are unusual. In this month's article by Dr Wiwanitkit,3 the author describes the hematologic features of human infections with H5N1 avian influenza virus. Such findings, although helpful in expanding our familiarity with this burgeoning threat, are not pathognomonic for avian influenza infection. In general, there is poor association between clinical findings and specific respiratory disease etiology, as patients often present with only a small subset of the clinical and epidemiological cues that might aid in diagnosis. Therefore, clinicians need to have a bias toward the importance of detecting reportable diseases and look carefully for factors that might prompt inclusion of a reportable illness in their differential diagnosis if these infections are to be identified.
Several factors conspire to inhibit the second step in disease reporting: pursuit of an etiologic laboratory diagnosis. Quality microbiological samples are often difficult to obtain. Tests for some agents are not adequately sensitive or specific. There may be disincentives for etiologic laboratory testing because of cost concerns. Treatment decisions must often be made before laboratory testing would return a diagnosis. As a result, continuing empiric therapy is often preferable to pursuing a specific microbiological diagnosis. Nevertheless, multiple arguments have been proposed for the importance of establishing an etiologic diagnosis for the treatment of pneumonia.15 In addition to the benefits to public health surveillance, other benefits include optimal use of antibiotics to reduce costs and to limit both inducible resistance and the community spread of drug-resistant organisms. Balance between the use of empiric therapy and pursuing laboratory diagnostic testing is essential for improving detection of reportable diseases.
Emerging respiratory diseases such as SARS and avian influenza are high-profile respiratory threats that are capable (SARS) or may soon be capable (avian influenza) of epidemic spread, and the importance of definitive diagnosis and reporting of these infections is obvious. On the other hand, the most common causes of community-acquired pneumonia, such as Streptococcus pneumoniae, are less likely to raise alarm as a public health concern and are generally covered by empiric antimicrobial regimens.15 Nonetheless, pathogens such as pneumococcus that are common causes of sporadic pneumonia may cause disease outbreaks, especially in institutional settings such as nursing homes, homeless shelters, and army barracks,16-18 and these outbreaks require investigation and intervention. These outbreaks are typically identified and controlled because a clinician recognizes a cluster of "routine" illnesses, orders appropriate etiologic testing, and alerts local health authorities to intervene.
Identification and control of travel-related outbreaks of respiratory disease present a special challenge for public health. These epidemics often manifest as a geographically scattered cluster of sick patients, as people may be exposed at a common location but return home before their symptoms emerge.19 Outbreaks of Legionnaires disease frequently occur among travelers, as they have been linked to contaminated water sources in hotels and on cruise ships. Bioterrorism-related outbreaks might also first present among travelers, if an infectious agent is intentionally introduced in an area of high transit. Cases from travel-related outbreaks might never be linked unless clinicians independently obtain etiologic diagnosis for single cases and report them to their local health departments, allowing activation of contact tracing and case-finding activities.20 The SARS outbreak provides an atypical example of how cases in travelers helped to identify a large local disease cluster that went undetected; initial SARS cases in China were not recognized because atypical pneumonia was common for that time of year and because the cases did not seem to be connected.7 A history of recent travel in a pneumonia patient, especially if the patient is severely ill, should prompt testing for Legionnaires disease and, depending on the site of travel and activities, other specific pathogens.
DETECTING AND REPORTING AVIAN INFLUENZA
Based on computer modeling of a potential avian influenza epidemic, a cluster of 14 human cases could indicate a potential antigenic shift and serve as a trigger for more aggressive surveillance and containment efforts.21 According to the World Health Organization (WHO) Global Influenza Preparedness Plan, the current influenza situation has entered phase 3 of 6, the last phase of the "Pandemic Alert Period," characterized by rare instances of spread of a novel influenza strain between humans who are in close contact.22 The WHO recommendations for all countries during the Pandemic Alert Period emphasize an increased awareness of the possibility of spread and the importance of immediate reporting of suspicious cases to local health departments for further testing, local control, and communication between organizations.
The CDC has released recommendations to guide testing for avian influenza during this phase. Testing for avian influenza A (H5N1) is indicated for all hospitalized patients with a severe respiratory illness such as pneumonia or acute respiratory distress syndrome for which an alternate etiologic diagnosis has not been established and who have a travel history to a country with documented H5N1 avian influenza in either poultry or humans. In addition, this testing should be considered on a case-by-case basis, in consultation with state and local health departments, for hospitalized or ambulatory patients with documented fever, upper respiratory symptoms, and any history of contact with poultry or a human with of H5N1 influenza (known or suspected) in an affected country within 10 days of symptom onset.23 Sources of updated recommendations include the CDC, state or local health departments, professional organizations, and international bodies such as the WHO.
Finally, several studies have shown that although clinicians usually are aware of their obligation to report, a significant proportion is unaware of how to do so.24,25 Larger institutions may have a designated infection control practitioner who is responsible for reporting, whereas in a smaller setting, the responsibility may lie with the individual clinician or laboratory worker to contact the health department. Reporting mechanisms vary by state, but in general, all provide for electronic, paper, and telephone means of contacting the state and local health departments. A lack of familiarity with the typical route for reporting should not stop you from notifying your local health officials immediately about a concerning illness; you can always simply call the health department and ask for the person on call for communicable diseases. Most health departments will have someone on call 24 hours a day.
The scale of morbidity and mortality caused by the next significant epidemic or pandemic will be a reflection of the public health infrastructure's ability to detect and respond to vanguard cases. Practicing clinicians find themselves at the crux of this system, intermediaries at the front line between a budding epidemic and the coordinated response that can contain it. Clinicians are therefore in a unique position to drastically alter the natural history of a potential or ongoing epidemic. By arming ourselves with a renewed sense of this responsibility and striving to improve notifiable disease reporting practices, we as clinicians can redouble our impact on the health and safety of our patients and the larger community.
1. Jackson ML, Neuzil KM, Thompson WW, et al. The burden of community-acquired pneumonia in seniors: results of a population-based study. Clin Infect Dis
2. Health, United States, 2004
. Hyattsville, MD: National Center for Health Statistics; 2004.
3. Wiwanitkit V. Hematological manifestations of bird flu, H5N1, infection. Infect Dis Clin Pract
. 2006. In press.
5. Rosen G. A History of Public Health [expanded edition]
. Baltimore, MD: Johns Hopkins University Press; 1993.
6. Ellis JH. Yellow Fever & Public Health in the New South
. Lexington, KY: University Press of Kentucky; 1992.
8. Duchin JS, Koster FT, Peters CJ, et al. Hantavirus pulmonary syndrome: a clinical description of 17 patients with a newly recognized disease. The Hantavirus Study Group. N Engl J Med.
9. Bush LM, Abrams BH, Beall A, et al. Index case of fatal inhalational anthrax due to bioterrorism in the United States. N Engl J Med
10. Monson RR, Halperin W, Baker EL. Harvard School of Public Health: Public Health Surveillance
. New York: Van Nostrand Reinhold; 1992.
11. Teutsch SM, Churchill RE, eds. Principles and Practice of Public Health Surveillance
. 2nd ed. Oxford, NY: Oxford University Press. 2000:186-189.
12. Progress in improving state and local disease surveillance-United States, 2000-2005. MMWR Morb Mortal Wkly Rep
13. Curtis AB, McCray E, McKenna M, et al. Completeness and timeliness of tuberculosis case reporting. A multistate study. Am J Prev Med
14. Moolenaar RL, Dalton C, Lipman HB, et al. Clinical features that differentiate hantavirus pulmonary syndrome from three other acute respiratory illnesses. Clin Infect Dis
15. Bartlett JG, Dowell SF, Mandell LA, et al. Practice guidelines for the management of community-acquired pneumonia in adults. Infectious Diseases Society of America. Clin Infect Dis
16. Gratten M, Morey F, Dixon J, et al. An outbreak of serotype 1 Streptococcus pneumoniae
infection in central Australia. Med J Aust
17. Feikin DR, Moroney JF, Talkington DF, et al. An outbreak of acute respiratory disease caused by Mycoplasma pneumoniae
and adenovirus at a federal service training academy: new implications from an old scenario. Clin Infect Dis
18. Nuorti JP, Butler JC, Crutcher JM, et al. An outbreak of multidrug-resistant pneumococcal pneumonia and bacteremia among unvaccinated nursing home residents. N Engl J Med
19. Benin AL, Benson RF, Arnold KE, et al. An outbreak of travel-associated Legionnaires disease and Pontiac fever: the need for enhanced surveillance of travel-associated legionellosis in the United States. J Infect Dis
20. Lee JV, Joseph C. Guidelines for investigating single cases of Legionnaires' disease. Commun Dis Public Health
21. Ferguson NM, Fraser C, Donnelly CA, et al. Public health. Public health risk from the avian H5N1 influenza epidemic. Science
24. Harvey I. Infectious disease notification-a neglected legal requirement. Health Trends
25. Allen CJ, Ferson MJ. Notification of infectious diseases by general practitioners: a quantitative and qualitative study. Med J Aust