MEASLES IN THE US: GOING, GOING, NOT GONE

Tuesday, 24th of May 2011 Print

Two items on measles in the US from the Journal of Infectious Diseases.

One elephant in the parlor is that measles in the industrialized countries, though at very low levels of incidence, imposes heavy costs on their governments, costs which will end only when the virus is cleared from the entire planet.

Ostroff’s editorial, also at
http://jid.oxfordjournals.org/content/203/11/1507.full discusses implications of the investigation by Chen et al., below and online at http://jid.oxfordjournals.org/content/203/11/1517

Good reading.

BD

 

Measles: Going, Going, But Not Gone

  1. 1.   Stephen M. Ostroff 

+ Author Affiliations

  1. 1.    Pennsylvania Department of Health, Harrisburg
  2. Correspondence: Stephen M. Ostroff, MD, Pennsylvania Department of Health, Rm 933, 625 Forster St, Harrisburg, PA 17120 (sostroff@state.pa.us).

(See the article by Chen et al, on pages 1517–25.)

For those of us engaged in disease investigation and response at the state and local level, the report by Chen and colleagues [1] in this issue of the Journal makes for sobering reading. It describes an outbreak of measles in Arizona where virus transmission predominantly occurred in the health care setting, a scenario of great concern to us all. In reading through the report, I was repeatedly reminded of the adage “What a fool does in the end, the wise do in the beginning.” One hopes that a report of this nature will spur at least some health care systems, hospitals, and physicians’ offices to act wisely before they too are confronted with a case of measles in their facilities. The Tucson outbreak also highlights many of the challenges faced by public health departments around the country with respect to a disease that, vaccine controversies notwithstanding, has been receding in memory and importance for many health care practitioners, institutions, and the public

In the United States, we entered the “postelimination” era in 2000 [2]. But in the context of measles, “elimination” does not mean that there are no cases occurring. This is because the disease continues to be still too common in other parts of the world, and international travels produce opportunities for continued introduction [3]. As a result, between 2000 and 2008, an average of 56 cases per year have been confirmed in the United States [3]. And paradoxically, the number of cases may actually be rising as segments of the population increasingly opt out of vaccination, producing uneven vaccination rates and pockets of susceptibility [4]. This raises concerns that sustained transmission can occur if measles is introduced into the wrong setting at the wrong time.

Consequently, even a single case of measles sets off alarms in every health department around the country and often prompts an extensive investigation like the one described in Tucson. Such investigations usually involve tracking large numbers of contacts; hastily arranged mass vaccinations; isolation, quarantine, and exclusion; expensive laboratory testing; and an enormous drain on resources [5, 6]. These actions are geared toward rapid containment to minimize the potential for transmission and, especially, multigeneration outbreaks. A major take-home lesson from Tucson is that some of the actions taken, and certainly many of the costs, were avoidable had common-sense measures been in place beforehand, rather than after the fact. At least one hopes that that lesson was learned and that these common-sense measures were applied after the fact.

First, as so well described by Chen and colleagues, case diagnosis and reporting were repeatedly delayed in Tucson. This happened even after the presence of measles was known, presumably the medical community had been alerted, and statewide active surveillance for measles was instituted. With the index case who was an international traveler, a full week elapsed between rash onset, establishment of a definitive diagnosis, and reporting of the case to the health department. Even while the patient was hospitalized, 3 days elapsed before the diagnosis of measles was even considered for this patient, followed by 2 more days before a lab test (which unfortunately had a negative result) was ordered. Only after a second test came back with a positive result was the case reported to health department investigators. With a highly transmissible infection such as measles, every day is crucial for successful containment. Even when the disease is only suspected, it should be immediately reported so that health authorities can get the jump on contact tracing (eg, the airline passengers on the patient's flight and care providers and patients in the emergency department), identifying susceptible individuals, and implementing prevention measures (prophylaxis and exclusion) while they are still feasible. Given the many challenges in obtaining airline manifests (especially for international flights) and tracking down widely dispersed travelers, the ≥2 weeks that elapsed before passengers were contacted are clearly too long for effective intervention.

Prompt tracking and investigation of the emergency department contacts of the index case patient would have led to the second case patient (presuming she was there as a patient) and allowed the health department to alert her medical team to her exposure and the possible diagnosis during her first hospitalization, rather than during her second. This would have avoided exposure by this patient of the health care worker who so unfortunately transmitted measles to her family member and to another patient. A similar missed opportunity occurred with an 11-month-old boy who was in the same emergency department at the same time as another case patient but was not tracked down as a contact or diagnosed until 2 weeks later, despite 3 visits in the interim to the pediatrician's office.

If for no other reason than its public health implications, measles should remain high in the differential diagnosis of any febrile rash illness, but especially when the patient is an international traveler or has an unknown (or inadequate) vaccination history. Although “only” 14 cases occurred in this outbreak, the consequences were severe and the costs exorbitant. The fact that the last 5 cases had no clear link to other confirmed cases or health care settings suggests that unidentified transmission occurred in the community and that the outbreak was larger than appreciated. In places with large transient populations due to tourism or migration, like Arizona, when even small proportions of the population are undervaccinated, the risk for subsequent transmission increases, as evidenced by a 2008 outbreak that occurred in San Diego, California [6, 7].

A second equally vexing concern relates to occupational health issues in the involved hospitals. We are currently in the midst of a major national drive for electronic health records [8]. So how in 2011 can it be acceptable to still have paper-based employee health records? Yet experience suggests that this remains the case in many hospitals, long-term care facilities, and outpatient clinics. Medical records are often poorly maintained, incomplete, or out of date. Often a substantial proportion of workers (employees or contractors) lack documentation of immunization (and not just for measles). In Tucson-area hospitals, this was the case for 30% of the employees. Insufficient records resulted in excessive testing, use of vaccine, and worker furloughs and produced the bulk of the $800,000 in costs for outbreak containment. These costs were fully avoidable.

There is virtually no reason (except for vaccine contraindications or failure) that any health care worker in the United States should lack measles immunity [9]. Health care workers are at substantially higher risk of exposure than the general population, and a health care worker with measles will inevitably result in large numbers of exposed, high-risk patients [10, 11]. And as seen in Tucson, illness from health care worker–to–patient transmission can be very serious.

Any facility with a substantial number of health care workers should maintain electronic occupational health records and should require that their workforce be uniformly immune to vaccine-preventable infections such as measles. The latter is a longstanding recommendation of both the Advisory Committee for Immunization Practices and the Healthcare Infection Control Practices Advisory Committee [12, 13] and is supported by professional societies. Yet in the Tucson metropolitan area's 7 hospitals, documentation of measles immunity was lacking in 30% of the workforce, and 9% of the workers tested in the 2 outbreak hospitals were found to be nonimmune. That represents an unacceptable and avoidable patient safety and liability risk that likely also exists elsewhere in the country. There is now a rising tide of support for mandatory vaccination of health care workers against influenza [14, 15]. This requires an annual vaccination campaign with its attendant complexities. But if we can't accomplish universal health care worker vaccination for measles, which requires only a single (or 2-dose) vaccination, how can we possibly achieve a better outcome for influenza vaccination?

Until measles elimination efforts make substantially more progress elsewhere in the world than they have to date, we will continue to deal with the potential for disease importation and subsequent transmission in the United States [16]. Today measles remains a substantial public health concern. However, health care settings should not contribute to the likelihood for transmission of this virus. Although we cannot eliminate the measles threat, through continued vigilance for the diagnosis, prompt reporting of suspected cases to health authorities, adherence to recommendations requiring documented health care worker vaccination, and use of administrative measures such as masking and prompt patient isolation, if we act wisely, it is within our ability to eliminate additional outbreaks like the one reported by Chen and colleagues from occurring.

 

Next Section

Footnotes

  • Potential conflicts of interest: none reported.
  • Received January 31, 2011.
  • Accepted January 31, 2011.

 

Previous Section

 

References

  1. 1.
    1. Chen SY,
    2. Anderson S,
    3. Kutty PK,
    4. et al

. Healthcare-associated measles outbreak in the United States after an importation: challenges and economic impact. J Infect Dis 2011;203:1517-25.

Abstract/FREE Full Text

  1. 2.
    1. Orenstein WA,
    2. Papania MJ,
    3. Wharton ME

. Measles elimination in the United States. J Infect Dis 2004;189 suppl 1:S1-3.

FREE Full Text

  1. 3.
    1. Parker Fiebelkorn A,
    2. Redd SB,
    3. Gallagher K,
    4. et al

. Measles in the United States during the postelimination era. J Infect Dis 2010;202:1520-8.

Abstract/FREE Full Text

  1. 4.

Centers for Disease Control and Prevention, Summary of notifiable diseases—United States, 2008. MMWR Morb Mortal Wkly Rep 2010;57(54):1-100.

  1. 5.
    1. Chen TH,
    2. Kutty P,
    3. Lowe LE,
    4. et al

. Measles outbreak associated with an international youth sporting event in the United States, 2007. Pediatr Infect Dis J 2010;29:794-800.

CrossRefMedlineWeb of Science

  1. 6.
    1. Sugerman DE,
    2. Barskey AE,
    3. Delea MG,
    4. et al

. Measles outbreak in a highly vaccinated population, San Diego 2008: role of the intentionally undervaccinated. Pediatrics 2010;125:747-55.

Abstract/FREE Full Text

  1. 7.

Centers for Disease Control and Prevention, Update: measles—United States—January–June 2008. MMWR Morb Mortal Wkly Rep 2008;57:893-6.

Medline

  1. 8.
    1. Blumenthal D

. Launching HITECH. N Engl J Med 2010;362:382-5.

CrossRefMedlineWeb of Science

  1. 9.
    1. Weber DJ,
    2. Rutala WA,
    3. Schaffner W

. Lessons learned: protection of healthcare workers from infectious disease risks. Crit Care Med 2010;38 suppl 8:S306-14.

CrossRefMedlineWeb of Science

10.  10.

  1. Atkinson WL,
  2. Markowitz LE,
  3. Adams NC,
  4. Seastrom GR

. Transmission of measles in medical settings—United States 1985–1989. Am J Med 1991;91 suppl 3B:S320-4.

CrossRefMedline

11.  11.

  1. Botelho-Nevers E,
  2. Chevereau L,
  3. Brouqui P

. Spotlight on measles 2010: measles in healthcare workers—vaccination should be revisited. Euro Surveill 2010;15:19687. http://www.eurosurveillance.org/ViewArticle.aspx?Articleid=19687. Accessed 29 January 2011.

Medline

12.  12.

Centers for Disease Control and Prevention. Immunization of health-care workers: recommendations of the Advisory Committee on Immunization Practices (ACIP) and the Hospital Infection Control Practices Advisory Committee (HICPAC). MMWR Recomm Rep 1997;46:1-42.

Medline

13.  13.

Advisory Committee on Immunization Practices. ACIP provisional recommendations for measles-mumps-rubella (MMR) “evidence of immunity” requirements for healthcare personnel. http://www.cdc.gov/vaccines/recs/provisional/downloads/mmr-evidence-immunity-Aug2009-508.pdf. Accessed 29 January 2011.

14.  14.

  1. Talbot TR,
  2. Babcock H,
  3. Caplan AL,
  4. et al

. Revised SHEA position paper: influenza vaccination of healthcare personnel. Infect Control Hosp Epidemiol 2010;31:987-95.

CrossRefMedlineWeb of Science

15.  15.

  1. Ottenberg AL,
  2. Wu JT,
  3. Poland GA,
  4. Jacobson RM,
  5. Koenig BA,
  6. Tilburt JC

. Vaccinating health care workers against influenza: the ethical and legal rationale for a mandate. Am J Public Health 2011;101:212-6.

Abstract/FREE Full Text

16.  16.

  1. Moss WJ

. Measles control and the prospect of eradication. Curr Top Microbiol Immunol 2009;330:173-89.

 

 

 

Best viewed at http://jid.oxfordjournals.org/content/203/11/1517

Health Care–Associated Measles Outbreak in the United States After an Importation: Challenges and Economic Impact

  1. 1.   Sanny Y. Chen1,2,
  2. 2.   Shoana Anderson2,
  3. 3.   Preeta K. Kutty3,
  4. 4.   Francelli Lugo4,
  5. 5.   Michelle McDonald4,
  6. 6.   Paul A. Rota3,
  7. 7.   Ismael R. Ortega-Sanchez3,
  8. 8.   Ken Komatsu2,
  9. 9.   Gregory L. Armstrong3,

10. Rebecca Sunenshine2,5 and

11. Jane F. Seward3

+ Author Affiliations

1.    1Epidemic Intelligence Service, Office of Workforce and Career Development, Centers for Disease Control and Prevention, Atlanta, Georgia
2.    2Bureau of Epidemiology and Disease Control Services, Arizona Department of Health Services, Phoenix
3.    3Division of Viral Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia
4.    4Division of Disease Control and Prevention, Pima County Health Department, Tucson, Arizona
5.    5Career Epidemiology Field Office, Office of Public Health Preparedness and Response, Centers for Disease Control and Prevention, Atlanta, Georgia
  1. Correspondence: Sanny Y. Chen, PhD, MHS, Centers for Disease Control and Prevention, 1600 Clifton Rd, MS E-30, Atlanta, GA 30333 (e-mail: sychen@cdc.gov).

 

Abstract

(See the editorial commentary by Ostroff, on pages 1507–9.)

Background. On 12 February 2008, an infected Swiss traveler visited hospital A in Tucson, Arizona, and initiated a predominantly health care–associated measles outbreak involving 14 cases. We investigated risk factors that might have contributed to health care–associated transmission and assessed outbreak-associated hospital costs.

Methods. Epidemiologic data were obtained by case interviews and review of medical records. Health care personnel (HCP) immunization records were reviewed to identify non–measles-immune HCP. Outbreak-associated costs were estimated from 2 hospitals.

Results. Of 14 patients with confirmed cases, 7 (50%) were aged ≥18 years, 4 (29%) were hospitalized, 7 (50%) acquired measles in health care settings, and all (100%) were unvaccinated or had unknown vaccination status. Of the 11 patients (79%) who had accessed health care services while infectious, 1 (9%) was masked and isolated promptly after rash onset. HCP measles immunity data from 2 hospitals confirmed that 1776 (25%) of 7195 HCP lacked evidence of measles immunity. Among these HCPs, 139 (9%) of 1583 tested seronegative for measles immunoglobulin G, including 1 person who acquired measles. The 2 hospitals spent US$799,136 responding to and containing 7 cases in these facilities.

Conclusions. Suspecting measles as a diagnosis, instituting immediate airborne isolation, and ensuring rapidly retrievable measles immunity records for HCPs are paramount in preventing health care–associated spread and in minimizing hospital outbreak–response costs.

Measles is a highly infectious viral disease spread by airborne transmission. During the late 1950s, an estimated 3–4 million measles cases occurred annually in the United States, with 48,000 reported hospitalizations and 450 reported deaths [1, 2]. After implementation of a 1-dose measles vaccine program in 1963, measles cases decreased [1]. In 1989, administration of a second dose of measles, mumps, and rubella (MMR) vaccine was recommended routinely for school children, health care personnel (HCP), students attending post–high school institutions, and international travelers without acceptable evidence of immunity [3]. Elimination of endemic measles transmission was declared in the United States in 2000 [1]. During 2001–2008, a median of 56 cases (range, 37–140 cases) were reported annually, with importations causing outbreaks among unvaccinated populations in certain community settings [4].

Because of measles severity, patients often seek care in health care settings, posing a high risk for transmission to other patients and HCP [5, 6]. Studies conducted during the 1980s documented that HCP have a 2–19-fold higher risk of acquiring measles than the general population [6, 7]. Health care–associated outbreaks can disrupt health care delivery and result in substantial morbidity or mortality among immunocompromised persons [5].

On 20 February 2008, a measles case was reported to the Arizona Department of Health Services (ADHS; Phoenix, AZ) and confirmed by the Arizona State Public Health Laboratory (ASPHL; Phoenix, AZ). Through 21 July, an additional 13 confirmed cases were identified in health care settings and the community. This report describes the epidemiology of the outbreak, examines outbreak-associated costs and risk factors that might have contributed to health care–associated transmission, and provides guidance to prevent outbreaks in health care settings.

Previous SectionNext Section

METHODS

Case Investigation and Outbreak Response

We used the Council of State and Territorial Epidemiologists measles clinical case definition: (1) fever (temperature, ≥101°F [≥38.3°C]), (2) a generalized maculopapular rash lasting ≥3 days, and (3) presence of cough, coryza, and/or conjunctivitis [8]. Confirmed cases were those that were laboratory confirmed or met the clinical case definition and were epidemiologically linked to another confirmed case. Suspected cases were those in which a generalized rash illness and a fever (temperature, ≥38.3°C) were present. Vaccination status was determined by written confirmation of receipt of a measles-containing vaccine. Self-reported receipt of vaccine without written documentation was classified as unknown vaccination status.

After the first measles case report, Pima County Health Department (PCHD; Tucson, AZ) and ADHS enhanced passive surveillance and established active surveillance in health and laboratory facilities at all 7 major community hospitals in Tucson. Activities included daily surveillance of emergency departments (EDs), urgent care, and inpatient and outpatient logs for febrile rash illnesses and measles tests. Commercial laboratories throughout the state were reminded to report positive measles test results. A screening tool was developed to evaluate ED patients who presented with a generalized rash and fever, to guide infection-control measures, and to prompt immediate reporting of suspected measles cases to PCHD.

Patients with suspected measles were interviewed to obtain demographic, clinical, and medical information and a list of potential contacts. Urine, serum, and nasopharyngeal specimens were collected for laboratory testing. Recommendations were made to place any hospitalized or ED patient in whom the diagnosis of measles was suspected or confirmed on airborne precautions during their period of infectiousness.

Household contacts who lacked evidence of measles immunity were offered MMR vaccine or immune globulin (IG) [9]. Community contacts were informed of their potential exposure through telephone calls, letters, television, radio, or print media and instructed to contact their physicians. A hospital contact was defined as any HCP (physicians, nurses, technicians, clerical and support staff, trainees, and volunteers) who had worked in the medical facility on the day of the exposure or as any patient or visitor who had shared the same room at the same time as (or within 4 h after) the patient with confirmed or suspected measles. Inpatient contacts without evidence of immunity were vaccinated or offered IG. Other hospital contacts who lacked evidence of immunity were directed to special immunizations clinics.

Laboratory

Laboratory testing was conducted at ASPHL and confirmed at the Centers for Disease Control and Prevention (CDC; Atlanta, GA). Cases were classified as laboratory confirmed by demonstration of measles IgM antibody in acute-phase serum samples by enzyme immunoassay (Microimmune), isolation of measles virus in cell culture (Vero/hSLAM cells), or detection of measles RNA by reverse-transcription polymerase chain reaction [9]. Nucleic acid from positive viral cultures was extracted, amplified, and sequenced to determine measles virus genotype [10, 11].

Evidence of Immunity for HCPs

We used the 1998 definition of acceptable evidence of measles immunity for HCP during a measles outbreak from the Advisory Committee on Immunization Practices and the Healthcare Infection Control Practice Advisory Committee [9], modified to exclude documentation of physician-diagnosed measles, as follows: (1) serologic evidence of immunity or (2) written confirmation of receipt of measles-containing vaccine according to birth year (defined as at least 1 dose for HCP born before 1957 and 2 doses for HCP born during or after 1957).

We reviewed HCP immunization records for evidence of measles immunity in 2 of 7 Tucson community hospitals. HCPs without evidence of measles immunity had blood samples drawn for serological testing and were offered MMR or IG on the same day. A second dose was administered ≥28 days after the first dose to those who were seronegative before the first MMR vaccine dose. All HCP without evidence of immunity were furloughed from work on days 5–21 after last exposure.

Hospital Costs

Costs were assessed in 2 hospitals. Data collected included the number of HCPs furloughed, time spent reviewing employee records for evidence of measles immunity (median, 15 min per record), and time spent conducting serologic tests and administering vaccine doses (median, 10 min per HCP). Furloughed hours were calculated by multiplying the number of HCP furloughed, the number of days in furlough, and a normal work shift (8 h). For the dollar value estimate of personnel time furloughed or spent reviewing employee records, we used the mean hourly earnings for full-time hospital health care practitioner and technical occupations in Arizona ($29.39), as reported by the US Bureau of Labor Statistics. The number of test kits or vaccine doses was assumed to be equal to the number of titers drawn and HCP vaccinated. The average unitary price for testing kits ($35) was used to calculate the dollar value of the tests performed. No other laboratory costs were included. For the dollar value of vaccine and vaccine administration, we used a $10 vaccine administration cost plus the average cost per dose ($48.31), as listed by the CDC for the private sector.

Previous SectionNext Section

RESULTS

Descriptive Epidemiology

During the period 13 February through 21 July 2008, there were 363 suspected, 8 probable (discarded after laboratory testing), and 14 confirmed measles cases identified in Arizona (Figures 1 and 2). All patients with confirmed cases were unvaccinated or had unknown vaccination status before exposure, and 2 patients (patients 3 and 8) had received their first dose of MMR vaccine on the same day as exposure. The median age of patients with confirmed cases was 20 years (range, 8 months–50 years), 7 (50%) were male, 4 (29%) were hospitalized for ≥24 h, and 2 (14%) required intensive care. No deaths occurred. Eleven patients (79%) accessed health care while infectious, and of these, 10 (91%) did not receive a prompt measles diagnosis after rash onset; only 1 (9%) was masked and isolated after presenting with rash and fever at a health care facility (Table 1).

View this table:

Table 1.

Demographic and Clinical Characteristics of Patients With Confirmed Measles, Tucson, Arizona, February 13–July 21, 2008 (n = 14)

View larger version:

Figure 1.

Distribution of confirmed measles cases, by week of rash onset, Tucson, Arizona, February 13–July 21, 2008 (n = 14).

View larger version:

Figure 2.

Confirmed measles cases and transmission patterns, by week of rash onset, Tucson, Arizona, 13 February–21 July 2008 (n = 14).

Twelve cases were laboratory confirmed, and 5 patients had measles virus sequences identical to each other and to the sequence of the genotype D5 viruses associated with a concurrent outbreak in Switzerland [12].

Exposure and Contact Investigations

Table 2 shows potential places of exposure, clinical and laboratory characteristics, and contact investigations of confirmed measles cases. The index case occurred in an unvaccinated female Swiss traveler aged 37 years. She arrived in Arizona on 2 February, traveled to Mexico on 3 February, developed fever (temperature, 39.6°C) on 8 February, and returned to Arizona on 9 February. She developed respiratory symptoms on 12 February (first ED visit) and rash on 13 February, at which time she was admitted to hospital A with a diagnosis of acute viral illness. She was not isolated until 15 February, when measles was first suspected. The results of initial measles IgM testing from specimens collected on 17 February, 4 days after rash onset, were negative. Because of the high suspicion of measles, additional IgM testing was conducted on samples drawn on 18 and 19 February (5 and 6 days after rash onset), and measles IgM was detected. Laboratory confirmation from ASPHL was received on 20 February. Because measles can be transmitted 4 days before rash onset, airline passengers from her flight from Mexico to the United States on 9 February were potentially exposed; these persons were investigated.

View this table:

Table 2.

Places of Exposure, Clinical Characteristics, Hospitalizations, and Contact Investigations of Confirmed Measles Cases, Tucson, Arizona, 13 February–21 July, 2008 (n = 14).

Patient 2 was a 50-year-old woman with unknown vaccination status, who was exposed to the index case for 1 h on 12 February in the ED waiting room of hospital A. Fever (temperature, 38.3°C) developed on 21 February, and respiratory symptoms developed on 24 February (second ED visit). She was admitted to the hospital on 24 February for asthma exacerbation and remained until 26 February, but was neither masked nor isolated throughout her stay. She presented again to the ED on 28 February experiencing cough and rash; she was immediately isolated and admitted for pneumonia and an allergic drug reaction. She received a measles diagnosis on 2 March and was discharged from the hospital on 3 March.

Patient 3 was a 41-year-old HCP with unknown vaccination status, who had first been exposed while providing care to patient 2 on 25 February, before receiving MMR vaccine on the same day. Fever (temperature, 38.3°C) developed on 5 March, and she presented to the ED experiencing fever and shortness of breath on 7 March. On 9 March, a rash developed, and she returned to the ED experiencing fever, cough, and rash. Measles was diagnosed, and she was masked and isolated immediately.

Patient 4 was an unvaccinated 11-month-old boy, who had spent 45 min in an ED room across the hall from patient 2 at hospital A on 24 February . Fever (temperature, 38.9°C) developed on 4 March, and a maculopapular rash developed on 10 March. He was also examined at a pediatrician's office on 3 separate occasions while infectious and was not isolated during his first 2 visits.

Patients 5 and 6 were siblings aged 3 and 5 years, respectively, who had not been vaccinated because of parental opposition to vaccination. Both children were exposed to patient 2 while visiting their mother at hospital A on 24 and 25 February. Their fever onsets occurred on 5 March (temperature, 39.5°C) and 6 March (38.9°C), respectively. Rash developed in both patients on 9 March, and both were examined at a pediatrician's office on 10 March; the pediatrician referred them to a commercial laboratory for blood sample collection. Neither patient was masked or isolated at the pediatrician's office or at the commercial laboratory.

Patient 7 was a 47-year-old woman with unknown vaccination status, who was exposed to patient 3 on 7 March in an ED room of hospital A. She developed fever (temperature, 38.9°C) on 19 March and presented to the ED experiencing dehydration and hematuria. She was admitted with a diagnosis of acute heat exhaustion and urinary tract infection and discharged the next day with a regimen of antibiotics. Rash developed on 21 March, and the patient was brought in by ambulance to the ED the next day for fever, cough, chills, and a rash. She was admitted to the intensive care unit for pneumonia and was isolated immediately.

Patient 8 was an unvaccinated 1-year-old girl who was exposed to patient 4 in the pediatrician's office on 10 March while waiting to receive MMR vaccine. Fever (temperature, 38.5°C) developed on 19 March, a generalized maculopapular rash developed on 20 March, and earache developed on 20 March. Medical care was not sought for her symptoms.

Patient 9 was a 41-year-old man with unknown vaccination status, who was exposed at his home to patient 3. Patients 10–14 were presumably exposed in the community because none had visited health care settings in the 3 weeks before the onset of illness and none had contact with any of the 8 patients with health care–associated infections. Patient 10 was an unvaccinated 2-year-old boy who had not received MMR vaccine, representing a missed opportunity for vaccination; he was admitted to the intensive care unit at hospital B for 6 days for febrile seizures. Patients 11 and 12 self-reported having received 1 dose of MMR vaccine, but no documentation was provided. No viruses were identified or sequenced from these patients. Four of the 5 patients with community-acquired infections had accessed health care while infectious, but they were neither masked nor isolated during their health care visits.

A total of 8231 contact investigations were conducted for all 14 patients; 4793 (58.2%) were hospital or clinic patients, 2868 (34.8%) were HCP, and 550 (7.0%) were other contacts. A total of 6470 investigations (78.6%) were attributable to exposures to the index case and 7 patients with confirmed healthcare-associated acquired measles.

HCP Measles Immunization Verification

None of the 7 community hospitals maintained electronic records of the immunity status of their HCP, requiring review of paper records. Of 14,844 HCP employed at the 7 community hospitals, 10,396 (70%) had acceptable evidence of measles immunity. Of 4448 HCP without proof of immunity, 1856 (42%) were born before 1957, and 2592 (58%) were born during or after 1957.

The proportion of HCP with measles immunity documentation at all 7 Tucson hospitals ranged from 58.8% to 85.7%. In hospitals A and B, 5419 (75%) of 7195 screened HCPs had evidence of immunity (Figure 3). Serologic testing was performed for 1583 (89%) of 1776 HCPs without documented immunity; 121 (11%) of 1077 HCPs born during or after 1957, and 18 (4%) of 506 HCPs born before 1957 were seronegative.

View larger version:

Figure 3.

Evaluation of evidence of measles immunity and serology testing of health care personnel at 2 major community hospitals, Tucson, Arizona, 2008 (n = 7195). IgG, anti-measles immunoglobulin G. aNo receipt of a measles-containing vaccine and no laboratory evidence of measles immunity. bLaboratory evidence of immunity or receipt of 1 measles-containing vaccine if born before 1957 or receipt of 2 measles-containing vaccine if born during or after 1957.

Hospital Costs

Approximately 15,120 h were lost in furloughs because of presumptive exposure, disease, or lack of evidence of immunity. Overall estimated economic impact for both hospitals was US$799,136, with HCP furloughs constituting 56% of the cost (Table 3). This represents a mean cost of response and containment of US$105,347 per case at hospital A and US$167,052 for the 1 case at hospital B.

View this table:

Table 3.

Estimated Costs Associated With Measles Outbreak for Hospitals A and B, Tucson, Arizona, 2008

Previous SectionNext Section

DISCUSSION

This measles outbreak in Arizona, with 14 confirmed cases, including 7 health care–associated infections, is the largest reported health care–associated measles outbreak in the United States since 1989 [8] and the first to be described in the post-elimination era. All patients were unvaccinated; one-half were infected in health care settings, including 6 in a single hospital. Health care–associated transmission included patient-to-HCP, patient-to-patient, patient-to-visitor, and HCP-to-patient transmission. The outbreak was both costly and disruptive to hospitals and to the state and local health departments. More than one-third of patients were hospitalized, and 2 required intensive care treatment, highlighting the potential severity of measles. Despite thousands of potential exposures, because of the high levels of population immunity (in 2008 in Arizona, coverage with 1 dose MMR vaccine among children aged 19–35 months was 92%, and reported coverage with 2 doses among children entering kindergarten was 96% in public schools and 91% in private schools.) and the highly efficacious MMR vaccine, no cases occurred among vaccinated persons.

This outbreak extends previous work documenting the high cost that hospitals can incur responding to measles in their facilities in the postelimination era [13]. During this outbreak, 2 hospitals spent almost US$800,000 responding to 7 patients with measles. Lack of readily available electronic HCP immunity status led to unnecessary serologic testing and vaccination of HCP who were immune to measles, which was funded largely by the health care facilities. Despite advances in measles control worldwide, in 2007, an estimated 20 million cases of measles occurred globally [4], and measles importations into the United States will continue. Optimal preparedness for measles exposures includes ensuring that all HCP have documented and easily retrievable measles immunity records to guide case management and outbreak response. Failure to implement these recommendations resulted in the continued exposure of nonimmune HCP to measles in the hospital, putting them and patients at further risk. A hospital patient infected by an unvaccinated HCP patient required intensive-care management for 4 days. This confirms previous findings that patients exposed in hospital settings might be at increased risk for severe outcomes of measles, given their relatively high prevalence of underlying medical conditions [6]. A measles-related death after health care–associated transmission in a hospital ED was reported elsewhere [14].

This outbreak posed considerable logistical challenges for hospital and health department staff. The outbreak response required rapid review of measles documentation of 14,844 HCP at 7 hospitals and emergency vaccination of ∼4500 HCP who lacked documentation of measles immunity. The number of seronegative HCPs identified (n = 138) at 2 hospitals would have been sufficient to sustain a sizeable health care–associated outbreak. Although under routine circumstances, birth before 1957 is considered acceptable presumptive evidence of measles immunity, during a measles outbreak, HCP born before 1957 should receive 2 doses of MMR vaccine or be excluded from work, unless they can demonstrate other evidence of immunity [9]. Because performing rapid serology testing during an outbreak is costly and disruptive, health care facilities should have serologic evidence of immunity available for all HCP to facilitate rapid vaccination response during a measles outbreak.

Health care–associated measles outbreaks after measles importations have been reported from other countries that have interrupted endemic measles transmission. Common risk factors identified include unvaccinated contacts and HCP [15], delayed measles diagnosis, and delayed implementation of infection-control procedures [16, 17]. During this outbreak, the following factors likely contributed to health care–associated transmission. First, none of the 14 patients with confirmed measles had been vaccinated before exposure. Besides 1 infant infected early in the outbreak and a 12-month-old child infected in the pediatrician's office the day that she received her routine MMR vaccine, 11 US cases were potentially preventable through adherence to US vaccine policy recommendations. Second, patients often accessed health care early during their illness before rash onset, resulting in substantial numbers of exposures to other patients and HCP. Strict adherence to infection-control guidance for persons in health care settings experiencing respiratory symptoms is the only available method to decrease risk for transmission at this stage of the illness [18]. Third, infection-control recommendations were only implemented for 1 of 11 patients who presented after rash onset. To prevent measles transmission in health care including ambulatory settings, all persons with an illness clinically compatible with measles should be immediately isolated in an examination room with a closed door or in a negative–air pressure room if available; and they should wear a size-appropriate mask, if feasible and tolerated, to prevent transmission while in common waiting areas [19]. Fourth, delays occurred in diagnosis and laboratory confirmation of measles. In the post-elimination era, when physicians are less familiar with diagnosing measles, a high index of suspicion is needed especially in persons with history of travel overseas or contact with someone with measles. Health care providers who suspect measles should obtain appropriate specimens for laboratory testing including specimens for viral isolation. Genotype data from this outbreak demonstrate the role of molecular epidemiologic surveillance in linking domestic measles outbreaks with imported measles cases. Health care providers are also reminded that the CDC recommends that negative measles IgM results for serum samples collected within the first 72 hours after rash onset should be confirmed with a second serum obtained 72 hours or longer after rash onset [20]. Laboratory testing results for the index case in this outbreak (IgM negative 4 days after rash onset) highlight that when there is a strong suspicion for measles, repeated testing is prudent even >72 h after rash onset.

The following limitations apply to our findings. Because some measles cases appeared to be community acquired with no demonstrable epidemiological links, it is likely that cases were missed in the community and that the outbreak was larger. HCP measles serologic data analysis was based on information collected at 2 hospitals, corresponding to 48% of all HCPs employed in Tucson during the outbreak, and therefore might not be representative. Birth year was not available for all HCPs, limiting our ability to determine the proportion of all HCPs born before 1957 who were measles seronegative. Our estimated outbreak costs likely underestimated the true cost for hospitals A and B. Data limitations prevented estimation of costs related to IG use, in-house or contracted laboratory work, overtime payments, HCP volunteers furloughed, and other administration or liability costs incurred by the hospitals. Our cost analysis was performed from a hospital perspective and did not include costs incurred by the state and local health departments, private and public insurance, and indirect costs borne by the patients and families. Better strategies are needed to ensure compliance with current guidelines to prevent measles transmission in medical settings [3, 5]. Hospitals that are not in compliance risk incurring substantial costs when faced with a measles exposure. Standard guidance for preventing measles transmission in health care settings include (1) increasing measles awareness among providers, especially among persons presenting with fever, rash, and travel history; (2) ensuring all HCP have evidence of measles immunity at the time of employment and have such data electronically available at the work site; (3) allowing only HCP with evidence of measles immunity to provide care to patients with measles [18]; and (4) instituting a screening plan to identify suspected measles cases for immediate isolation during a measles outbreak.

Previous SectionNext Section

Funding

Centers for Disease Control and Prevention and the State of Arizona provided funding to support this investigation.

Previous SectionNext Section

Acknowledgments

We acknowledge the staff at the Employee Health Services at Tucson Medical Center and Northwest Medical Center for their support and assistance in estimating the cost of the measles outbreak, infection-control practitioners at all 7 major community hospitals in the Tucson area and the Measles Investigation Team for their hard work in controlling the measles outbreak, and Kris Bisgard for careful review of the manuscript. Members of the Measles Investigation Team are as follows: Pima County Health Department: R. Beeson, E. Botwright, S. Daniels, G. Diaz, D. Douglas, J. Guthrie, L. Hulette, B. Johnson, I. Luna, E. MacNeil, R. Melland, K. Merritt, F. Miller, R. Norrish, D. Perkins, R. Peyton, K. Reeve, N. Siemsen, A. Taylor, P. Taylor, L. Valenzuela, P. Woodcock; Arizona Department of Health Services: A. D'Souza, L. Erhart, K. Fredrickson, P. Gast, S. Goodykoontz, D. Herrington, W. Humble, S. Imholte, K. Lewis, C. Martinez, J. Meyer, C. Ogden, M. Seiter, C. Tsang, V. Waddell, J. Weiss, C. Wiedeman, C. Yu; Centers for Disease Control and Prevention (asterisks indicate members of the Epidemic Intelligence Service) A. Allman, W. Bellini, E. Bolyard, D. Ehrhardt*, K. Gallagher, P. Gould*, P. Gresham, J. Leung, L. Lowe, C. Pannazo, J. Rosen*, J. Rota, D. Sarecha, A. Srinivasan, M. Wikswo.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. Use of trade names and commercial sources are for identification purposes only and does not imply endorsement by the Public Health Service or the US Department of Health and Human Services.

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: none reported.
  • Presented in part: 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy/Infectious Diseases Society of America (ICAAC/IDSA) 46th Annual Meeting; Washington, DC; October 2008. Abstract 4978.
  • Received September 28, 2010.
  • Accepted December 30, 2010.

 

Previous Section

 

References

  1. 1.
    1. Orenstein WA,
    2. Papania MJ,
    3. Wharton ME

. Measles elimination in the United States. J Infect Dis 2004;189:S1-S3.

FREE Full Text

  1. 2.
    1. Bloch AB,
    2. Orenstein WA,
    3. Stetler HC,
    4. et al

. Health impact of measles vaccination in the United States. Pediatrics 1985;76:524-32.

Abstract/FREE Full Text

  1. 3.

Centers for Disease Control and Prevention. Measles, mumps, and rubella—vaccine use and strategies for elimination of measles, rubella, and congenital rubella syndrome and control of mumps: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1998;47:1-57.

Medline

  1. 4.
    1. Fiebelkorn AP,
    2. Redd SB,
    3. Gallagher K,
    4. et al

. Measles in the United States during the post-elimination era. J Infect Dis 2010;202:1520-8.

Abstract/FREE Full Text

  1. 5.
    1. Biellik RJ,
    2. Clements CJ

. Strategies for minimizing nosocomial measles transmission. Bull World Health Organ 1997;75:367-75.

MedlineWeb of Science

  1. 6.
    1. Steingart KR,
    2. Thomas AR,
    3. Dykewicz CA,
    4. Redd SC

. Transmission of measles virus in healthcare settings during a communitywide outbreak. Infect Control Hosp Epidemiol 1999;20:115-9.

CrossRefMedlineWeb of Science

  1. 7.
    1. Atkinson WL,
    2. Markowitz LE,
    3. Adams NC,
    4. Seastrom GR

. Transmission of measles in medical settings—United States, 1985–1989. Am J Med 1991;91 Suppl 3B:320S-324S.

CrossRefMedline

  1. 8.

Centers for Disease Control and Prevention. Nationally notifiable infectious diseases, United States 2009. Atlanta, GA; http://www.cdc.gov/ncphi/disss/nndss/casedef/measles_2009.htm. Accessed 4 March 2009 Atlanta, GA.

  1. 9.

Centers for Disease Control and Prevention. ACIP provisional recommendations for measles-mumps-rubella (MMR) 2009 ‘evidence of immunity’ requirements for healthcare personnel. www.cdc.gov/vaccines/recs/provisional/default.htm. Accessed 9 March 2009 Centers for Disease Control and Prevention, Atlanta, GA.

10.  10.

World Health Organization. Standardization of the nomenclature for describing the genetic characteristics of wild-type measles viruses. Wkly Epidemiol Rec 1998;73:265-72.

Medline

11.  11.

World Health Organization. New genotype of measles virus and update on global distribution of measles genotypes. Wkly Epidemiol Rec 2005;40:347-51.

12.  12.

  1. Richard JL,
  2. Masserey-Spicher V

. Large measles epidemic in Switzerland from 2006 to 2009: consequences for the elimination of measles in Europe. Euro Surveill 2009;14. pii: 19443.

13.  13.

  1. Parker AA,
  2. Staggs W,
  3. Dayan GH,
  4. et al

. Implications of a 2005 measles outbreak in Indiana for sustained elimination of measles in the United States. N Engl J Med 2006;355:447-55.

CrossRefMedline

14.  14.

  1. Rivera ME,
  2. Mason WH,
  3. Ross LA,
  4. Wright HT Jr.

. Nosocomial measles infection in a pediatric hospital during a community-wide epidemic. J Pediatr 1991;119:183-6.

CrossRefMedlineWeb of Science

15.  15.

  1. Grgic-Vitek M,
  2. Frelih T,
  3. Ucakar V,
  4. et al

. Spotlight on measles 2010: a cluster of measles in a hospital setting in Slovenia, March 2010. Euro Surveill 2010;15. pii: 19573.

16.  16.

  1. Weston KM,
  2. Dwyer DE,
  3. Ratnamohan M,
  4. et al

. Nosocomial and community transmission of measles virus genotype D8 imported by a returning traveler from Nepal. Commun Dis Intell 2006;30:358-65.

Medline

17.  17.

  1. Groth C,
  2. Bottiger B,
  3. Plesner A,
  4. Christiansen A,
  5. Glismann S,
  6. Hogh B

. Nosocomial measles cluster in Denmark following an imported case, December 2008–January 2009. Euro Surveill 2009;14. pii: 19126.

18.  18.

  1. Siegel JD,
  2. Rhinehart E,
  3. Jackson M,
  4. Chiarello L

. Healthcare Infection Control Practices Advisory Committee. 2007 guideline for isolation precautions: preventing transmission of infectious agents in healthcare settings. http://www.cdc.gov/ncidod/dhqp/pdf/isolation2007.pdf. Accessed 9 March 2010.

19.  19.

Committee on Infectious Diseases, American Academy of Pediatrics. Infection prevention and control in pediatric ambulatory settings. Pediatrics 2007;120:650-665. http://www.pediatrics.org/cgi/content/full/120/3/650. Accessed 22 November 2010.

Abstract/FREE Full Text

20.  20.

  1. Helfand RF,
  2. Heath JL,
  3. Anderson LJ,
  4. Maes EF,
  5. Guris D,
  6. Bellini WJ

. Diagnosis of measles with an IgM capture EIA: the optimal timing of specimen collection after rash onset. J Infect Dis 1997;175:195-9.

Abstract/FREE Full Text

Related articles

EDITORIAL COMMENTARIES:

  • Stephen M. Ostroff

Editor's Choice: Measles: Going, Going, But Not Gone J Infect Dis. (2011) 203(11): 1507-1509 first published online April 28, 2011 doi:10.1093/infdis/jir125

Articles citing this article

Special Postings

;

Highly Accessed

Website Views

1933198