Update: Influenza Activity—United States, 1997-98 Season | Geriatrics | JAMA | JAMA Network
[Skip to Navigation]
Citations 0
News From the Centers for Disease Control and Prevention
April 15, 1998

Update: Influenza Activity—United States, 1997-98 Season

JAMA. 1998;279(15):1155-1158. doi:10.1001/jama.279.15.1155-JWR0415-2-1

MMWR. 1998;47:196-200

In collaboration with the World Health Organization (WHO), its collaborating laboratories, and state and local health departments, CDC conducts surveillance to monitor influenza activity and to detect antigenic changes in the circulating strains of influenza viruses. This report summarizes influenza surveillance in the United States from September 28, 1997, through March 7, 1998, and presents reports of outbreaks in long-term care facilities (LTCFs) in three states and at a military base. The findings indicate that this season has been dominated by influenza A(H3N2) viruses and characterized by a sustained elevation in pneumonia and influenza (P&I)-related deaths.

Influenza activity in the United States began during October, increased sharply during December and January, peaked during late January through early February, then declined. From September 28, 1997, through March 7, 1998, WHO collaborating laboratories tested 64,421 clinical specimens for respiratory viruses, and 10,264 (16%) were positive for influenza. Of these, 10,247 (99.8%) were influenza A, and 17 (0.2%) were influenza B. Of 2453 influenza A isolates that were subtyped, 2447 (99.8%) were A(H3N2), and six (0.2%) were A(H1N1). Of the H3N2 influenza A viruses, 188 were antigenically characterized by CDC; 44 (23%) were similar to A/Nanchang/933/95(H3N2), the A/Wuhan/359/95(H3N2)-like component in the 1997-98 influenza vaccine, and 144 (77%) were similar to A/Sydney/05/97(H3N2), a related but antigenically distinguishable variant of the A(H3N2) component of the 1997-98 influenza vaccine. All eight antigenically characterized influenza B and five of six antigenically characterized influenza A(H1N1) viruses were similar to the 1997-98 influenza vaccine components.

State and territorial epidemiologists first reported widespread influenza activity* from Pennsylvania for the week ending December 20. Influenza activity peaked in the United States during the week ending February 7, when 46 states and New York City reported regional or widespread activity. During the week ending March 7, the number of states reporting regional or widespread influenza activity declined to 27.

The percentage of patient visits to sentinel physicians for influenza-like illness (ILI) first exceeded baseline levels (0-3%) during the week ending January 3, peaked at 5% from January 18 through February 7, and returned to baseline levels during the week ending February 21. The percentage of deaths attributed to P&I as reported by the vital statistics offices of 122 cities first exceeded the epidemic threshold† during the week ending January 10 and has remained elevated for 9 consecutive weeks.

As of March 7, a total of 359 outbreaks of ILI in LTCFs have been reported to CDC from the state health departments in Connecticut, New York, and Virginia. Three outbreaks in LTCFs and one on a military base are described in this report. In these investigations, disease and influenza vaccination status of residents of LTCFs and vaccination status of military squadron members were ascertained by medical record review. Among staff of LTCFs and among military squadron members, disease status was ascertained by self-administered questionnaires. ILI was defined as either (1) a positive culture or rapid-antigen test for influenza in a person with respiratory symptoms or (2) cough and either perceived or measured fever (≥100 F [≥37.8 C]) or chills. For the LTCF in Connecticut, measured fever was defined as a temperature ≥100.5 F (≥38.1 C). An influenza-related death was defined as a death that occurred within 2 weeks of onset of ILI, with no intervening asymptomatic period and no alternative explanation.1 Vaccine effectiveness (VE) was calculated as: VE=[ARU−ARV/ARU] × 100; ARU is the attack rate in unvaccinated persons, and ARV is the attack rate in vaccinated persons.2


All Connecticut LTCFs are required to report outbreaks of respiratory disease to the Connecticut Department of Public Health (CDPH). When reports are received, LTCFs are encouraged to test for influenza. Rapid-antigen testing and/or culture are made available at no cost by the state laboratory during the influenza season. LTCFs are encouraged to implement influenza outbreak control measures as recommended by the Advisory Committee on Immunization Practices (ACIP).3

From December 1, 1997, through February 28, 1998, a total of 118 (44%) of 271 LTCFs reported respiratory outbreaks to CDPH; 21 were confirmed as influenza A outbreaks. On December 12, 1997, a LTCF in New Haven County reported an outbreak of influenza A. Because this was the first confirmed influenza outbreak in the state for the 1997-98 season, an epidemiologic investigation was conducted.

The LTCF has 172 staff and 131 residents distributed in four units. Of nasopharyngeal swab specimens obtained from 42 residents with ILI, 20 (48%) were positive for influenza A by rapid-antigen testing. Influenza A (H3N2) was identified by culture in nine specimens at the state laboratory, and three isolates were further characterized at CDC by hemagglutination-inhibition testing as A/Sydney/05/97(H3N2)-like. Medical records of all residents were reviewed. From December 6, 1997, through January 3, 1998, a total of 57 (49%) of 116 vaccinated residents and seven (47%) of 15 unvaccinated residents developed ILI (VE=-5% [95% confidence interval (CI)=-87%-41%]). Five (4%) vaccinated residents and one (7%) unvaccinated resident died from influenza-related complications (VE=35% [95% CI=-416.8%-91.9%]). Beginning December 17, amantadine treatment was provided to two persons with ILI, and starting December 19, amantadine prophylaxis was provided to 21 residents who were asymptomatic.

New York

Each year, the New York State Department of Health sends a memorandum to LTCFs and other institutions recommending vaccination of residents, use of rapid-antigen testing during outbreaks of ILI, and rapid implementation of ACIP-recommended outbreak-control measures if influenza is confirmed.3

From October 30, 1997, through February 17, 1998, a total of 213 (33%) of 650 LTCFs in New York state reported laboratory-confirmed influenza A by rapid-antigen test or culture, representing a 245% increase over the 87 laboratory-confirmed influenza A outbreaks reported during the 1996-1997 influenza season. Of 47 facilities from which complete data were available, all reported prophylactic use of amantadine/rimantadine, and the median ILI attack rate was 12% (range: 2%-49%).

On January 7, 1998, a LTCF in Westchester County reported a severe outbreak of ILI. The facility has 180 day-shift staff and 270 residents in six units. On December 24, 1997, respiratory specimens were analyzed by a rapid immunofluorescent antibody test and were negative for influenza. However, on January 7, 1998, two specimens cultured at the state laboratory were positive for influenza A(H3N2). One isolate was further characterized at CDC as A/Sydney/05/97(H3N2)-like. Although rimantadine prophylaxis was administered to eligible residents on January 7, 1998, the outbreak had already peaked. From December 16, 1997, through January 7, 1998, a total of 59 (22%) of 264 vaccinated residents and one (17%) of six unvaccinated residents developed ILI (VE=-34% [95% CI=-714%-78%]). Four (2%) vaccinated residents and one (17%) unvaccinated resident died of influenza-related complications (VE=91% [95% CI=30.3%-98.8%]). Among the staff, 172 (96%) of 180 day-shift staff persons completed a self-administered questionnaire; 18 (30%) of 60 vaccinated and 36 (32%) of 111 unvaccinated persons developed ILI (VE=7.5% [95% CI=-48.1%-42.2%]).


During the 1997-98 influenza season, the Virginia Department of Health (VDH) conducted active surveillance for outbreaks of ILI in LTCFs and recommended that LTCFs confirm influenza using rapid-antigen tests provided by the state laboratory and implement ACIP-recommended outbreak-control measures.3

From January 26 through February 27, 1998, the VDH received reports of respiratory disease outbreaks from 28 (10%) of 290 licensed LTCFs. On January 26, a LTCF in Henrico County reported an outbreak of ILI. On January 31, influenza A was cultured at the state laboratory from five (71%) of seven nasopharyngeal swab specimens obtained from ill residents. Four isolates were further characterized at CDC as A/Sydney/05/97(H3N2)-like. The facility had 202 staff members and 190 residents in five units.

During January 7-31, a total of 42 (28%) of 150 vaccinated residents and 15 (38%) of 40 unvaccinated residents developed ILI (VE=25% [95% CI=-20.1%-53.6%]). Nine (6%) vaccinated residents and two (5%) unvaccinated residents died from influenza-related complications (VE=-20% [95% CI=-434%-73%]). When all deaths associated with respiratory complications during the outbreak period were included, including those not meeting the ILI case definition, 10 (7%) deaths occurred among the vaccinated and four (10%) among the unvaccinated (VE=33% [95% CI=101.5%-77.9%]). Among the staff, 16 (16%) of 101 vaccinated persons and 18 (18%) of 101 unvaccinated persons developed ILI (VE=11% [95% CI=-64.3%-51.9%]). Outbreak control measures, including antiviral prophylaxis, were fully implemented by January 31.

Military Base

On January 15, 1998, an outbreak of ILI was reported among members of a Navy squadron in Hawaii. Influenza type A was isolated at the base laboratory from four nasopharyngeal swab specimens collected from squadron members. One isolate was further characterized at CDC as A/Sydney/05/97(H3N2). Of 362 squadron members, 254 (70%) completed the questionnaire.

During January 1-30, 1998, a total of 40 (20%) of 197 vaccinated squadron members and 13 (24%) of 54 unvaccinated squadron members had ILI (VE=16% [95% CI=-46.0%-51.3%]). Median duration of illness was 6 days (range: 2-14 days) among vaccinated members and 5 days (range: 3-21 days) among the unvaccinated. Twenty-four (63%) of 38 vaccinated persons who had ILI and seven (54%) of 13 unvaccinated persons who had ILI and who responded to the questionnaire reported being sent home by the squadron's doctor and staying in bed because of symptoms (relative risk=1.17; 95% CI=0.7-2.1). Amantadine was not provided for prophylaxis, but was used to treat 12 cases.

Reported by:

ML Cartter, MD, Coordinator, Epidemiology Program, NL Barrett, MS, Connecticut Dept of Public Health; DR Mayo, ScD, SH Egbertson, Connecticut State Laboratory, Hartford, Connecticut. D Ackman, MD, S Kondracki, G Brady, H Leib, ME Hennessy, R Gallo, L Grady, PhD, P Smith, MD, State Epidemiologist, New York State Dept of Health. S Jenkins, VMD, Acting State Epidemiologist, D Woolard, PhD, M Linn, MURP, E Barrett, DMD, J Rullan, MD, Office of Epidemiology, Virginia Dept of Health; J Pearson, DrPH, B Meisel, Virginia Div of Consolidated Laboratory Svcs; C Thorpe, MD, P Young, Henrico Health District; BG Regirer, LLM, S Jones, MD, P Gershonoff, V Altman, Henrico County long-term care facility, Richmond, Virginia. N Anderson, Univ of Michigan, Ann Arbor. HJ Beecham III, MD, AJ Yund, MD, Navy Environmental and Preventive Medicine Unit No. 6; MB Weigner, MD, J Herbst, BS Wiseman, Navy Medical Clinic, Pearl Harbor, Hawaii. LC Canas, Project Gargle, Brooks Air Force Base, San Antonio, Texas. Participating state and territorial epidemiologists and state public health laboratory directors. World Health Organization collaborating laboratories. State Br, Epidemiology Program Office; Influenza Br, Div of Viral and Rickettsial Diseases, National Center for Infectious Diseases, CDC.

CDC Editorial Note:

Both the 1996-97 and the 1997-98 seasons have been dominated by influenza A(H3N2) viruses and characterized by sustained elevations in P&I-related excess deaths. The predominant A(H3N2) strains identified in the United States during the 1997-98 season have been A/Sydney/05/97(H3N2)-like, which are variants of the strain contained in the 1997-98 vaccine. Although influenza outbreaks among all age groups have been reported to CDC, most have been reported in elderly nursing home residents.

The outbreak investigations reported here all were associated with A/Sydney/05/97(H3N2)-like viruses and suggest that protection provided by the current vaccine against illness caused by this variant strain may have been low. This is consistent with previous reports with variant strains.4-6 In the outbreaks in Connecticut and New York, influenza vaccination appeared to reduce death rates, even when it failed to prevent ILI. Although the reduced risk was statistically significant in only one of the outbreaks, this also is consistent with previous studies3,5-8 and underscores the importance of vaccinating persons at high risk for influenza-related complications and death even in years when the match between vaccine and circulating strain is not optimal.3 The timely implementation of outbreak control measures within institutions, including vaccination of residents, reduced contact between ill and non-ill persons, and antiviral prophylaxis of all non-ill persons and antiviral treatment of ill persons when the outbreak is caused by influenza type A, may reduce morbidity and mortality.3

Throughout the influenza season, surveillance data collected by CDC are updated weekly and are available through the CDC voice information system, telephone (888) 232-3228, or the fax information system, telephone (888) 232-3299, by requesting document number 361100, or through CDC's Influenza Branch, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases World-Wide Web site http://www.cdc.gov/ncidod/diseases/flu/weekly.htm. Information about local influenza activity is available from many county and state health departments.

References 8 available.
*Levels of activity are (1) no activity; (2)sporadic—sporadically occurring influenza-like illness (ILI) or culture-confirmed influenza with no outbreaks detected; (3)regional—outbreaks of ILI or culture-confirmed influenza in counties with a combined population of <50% of the state's total population; and (4) widespread—outbreaks of ILI or culture-confirmed influenza in counties with a combined population of ≥50% of the state's total population.
† The epidemic threshold is 1.645 standard deviations above the seasonal baseline. The expected seasonal baseline is projected using a robust regression procedure in which a periodic regression model is applied to observed percentages of deaths from P&I since 1983.