CDC Logo Tuberculosis Information CD-ROM   Image of people
jump over main navigation bar to content area
TB Guidelines
Surveillance Reports
Slide Sets
TB-Related MMWRs and Reports
Education/Training Materials
Ordering Information


U.S. Department of Health and Human Services


TB Notes 1, 2005

No. 1, 2005

Highlights from State and Local Programs

New and Old Ideas Blend Well at Harbor Light Shelter

In spring 2003, the St. Louis City Health Department TB control program noted an increase in TB cases among the city’s homeless, with a possible link to a shelter called Harbor Light. Owned and managed by the Salvation Army, Harbor Light is the city’s largest homeless shelter. The state and city TB staff had been noting this trend through program assessments and case reviews. The city’s TB nurse case manager requested a thorough investigation by the state health department, which was completed on April 29, 2003. The goal of the investigation was to confirm or rule out Harbor Light as a significant site of transmission and to develop new strategies for curbing this outbreak.

The investigation revealed that since February 2001, 16 homeless persons with active TB had been diagnosed and epidemiologically linked to stays at Harbor Light. Fourteen had AFB-positive smears at time of diagnosis; two were HIV infected. Two died, one shortly after diagnosis and one while hospitalized at the Missouri Rehabilitation Center (Missouri’s inpatient TB facility). Isolates were available for 13, allowing for confirmation of epidemiological links (identified by chart review and shelter log records) through genotyping.

From analysis, two distinct outbreaks emerged. In the first, four patients had epidemiological links to each other and to the shelter and also had the same genotype pattern. These cases were diagnosed between June 2001 and January 2002. In the second, nine others were linked, having another distinct and matching genotype pattern, and became the focus of the investigation. The first two cases in this group were diagnosed in July 2001. One of these case patients was found critically ill at the shelter and died shortly after hospitalization. Both were deemed to have been highly infectious while staying at the shelter between April and July 2001. Eight cases were linked back to this exposure. Four out of five shelter workers also converted their skin tests at this time. Two cases did not have culture isolates available, but were also epidemiologically linked to this second group. A second critical exposure period existed between January and March 2003, when the second patient who died had stayed in the shelter, and two additional infectious cases were diagnosed.

Eleven of these surviving patients were completely treated. One reactivated and was treated twice; two were lost to follow-up. Compared with the other TB patients in Missouri, patients in this outbreak were four times more likely to be HIV infected, eight times more likely to abuse alcohol, nine times more likely to use noninjected drugs, and 25 times more likely to inject drugs.

The outbreak persisted and reached 19 cases by August 2003, despite the city health department staff’s excellent efforts in contact follow-up, targeted testing, and symptom reviews. With the prospect of another cold Missouri winter and the likelihood that homeless persons would crowd this large shelter for another season, several unconventional recommendations were considered. Many of these included primary prevention activities, such as environmental controls at the shelter, to reduce the incidence of TB transmission at this high-risk congregate setting.  The following discussion will explain what was done and why.

Managing TB in Homeless Shelters in the St. Louis Area

During May through June 2003, 250 clients, staff, and frequent visitors (e.g., ministers) were given a TB skin test at the Harbor Light shelter in response to two cases and a death that were reported during the previous 2 months. A few positive skin tests were reported and evaluated, but no active TB cases were identified.

In August 2003, the City of St. Louis Health Department TB control program began to use a homeless shelter client tracking system called ROSIE (Regional Online Service Information Exchange). This tracking system keeps a current record of all movement of homeless clients throughout the city’s shelter network. The value of this system is that it allows the TB Control Program to identify contacts who may have moved from one city shelter to another and be able to find, evaluate, and provide treatment as necessary. Access to this system resulted in identifying six homeless contacts from previous cases.

In September 2003, a TB outbreak containment meeting was held at the Harbor Light Shelter and was attended by representatives of the City of St. Louis Health Department TB Control Program; CDC’s DTBE, including Dr. Paul Jensen, who has extensive engineering experience working in shelters and prisons in countries such as Russia and South Africa; CDC’s National Institute for Occupational Safety and Health (NIOSH); the Missouri State Department of Health and Senior Services (DHSS); the Salvation Army shelter management staff; and the Salvation Army’s new facilities management company, Lenzy Hayes. As a result of that meeting, a strategy was developed to conduct a comprehensive evaluation of the entire heating, air conditioning, and ventilation (HVAC) system in the shelter. Their recommendations included cleaning and retrofitting all 23 air handlers, upgrading the filters inside the air handlers to higher-rated, less porous filters, and installing TB-killing ultraviolet germicidal irradiation (UVGI) lights as funds were available.

In October 2003, funds were secured to conduct mobile, on-site radiographs at the Harbor Light Shelter. This provided a baseline evaluation of 250 clients, staff, and frequent visitors prior to the implementation of the suggested engineering improvements. There was a potential concern regarding how to proceed if large numbers of clients were found with abnormal radiograph results. To address this concern, in October the Florida Department of Health Laboratory in Jacksonville agreed to a memorandum of understanding (MOU) with the Missouri Department of Health and Senior Services and the City of St. Louis Health Department TB Control Program to provide the rapid, 1-day nucleic acid amplification (NAA) Mycobacteria Tuberculin Direct (MTD) Test for any homeless clients. Eight clients were found to have abnormal radiographs and referred to the TB control program for further evaluation. All eight were determined to be negative for TB disease. Abnormal results varied from old, healed gunshot wounds to a referral for possible carcinoma.

On November 26, 2003, the state health department issued a Public Health Advisory. The Advisory alerted the St. Louis–area medical community to the outbreak, and recommended that providers 1)  “think TB” for homeless shelter users and workers with respiratory illness, 2) screen for TB disease risk factors and symptoms in area emergency departments, 3) send all sputum specimens to the state TB lab, and 4) report all TB suspects within 24 hours. The state health department issued an epi-X report shortly following the advisory, in an effort to alert neighboring states. (Epi-X, or the Epidemic Information Exchange, is a secure, moderated means of communication between public health officials for reporting and discussing outbreaks and other acute health events.)

During November-December 2003, the Salvation Army paid to have the 23 air handlers, including two roof units, completely cleaned, scrubbed, and retrofitted. This investment by the Salvation Army division headquarters resulted in thorough air quality improvement (versus a band-aid measure). The visiting nurses who see clients a few hours a day reported seeing a significant reduction in complaints for respiratory ailments (such as coughs, colds, and allergies). During this 2-month period of time, the air filters were replaced with higher-rated air filters a couple of times to capture residual particles from the cleaning effort.

In April 2004, funds in the amount of $25,000 were secured through the State of Missouri Department of Health and Senior Services to purchase TB-killing ultraviolet lights. Lumalier Incorporated, located in Memphis, Tennessee, was awarded the contract. Mr. Charley Dunn, Chairman, has worked on a number of shelter and jail projects and as a consultant with CDC on projects throughout the United States, Eastern Europe, Africa, and South America. A number of UVGI lights were purchased based upon the evaluation of Dr. Paul Jensen (CDC/DTBE), Drs. Chris Coffey and Steve Martin (CDC/NIOSH), and Mr. Charley Dunn. This type of UVGI lighting was designed to provide the best type of application based on the size of the room, the amount of airflow, and the client capacity.

In July 2004, a nine-member engineering team from CDC/NIOSH conducted a week-long tracer gas study in the Harbor Light Shelter Annex unit (the area where all TB cases were identified) in order to determine the air flow patterns and the volume of air circulated and to measure for static air locations. This is important in that some homeless clients are permitted to remain in the shelter sleeping quarters during the day because of illness or injury.

An article clipping from the St. Louis Post-Dispatch, July 15, 2004. A member of NIOSH team takes an air sample in Harbor Light shelter

A member of the NIOSH team takes an air sample in Harbor Light shelter (from the St. Louis Post-Dispatch, July 15, 2004).

In July 2004, a second phase of UVGI light installation ($10,000) was completed. Mr. Charley Dunn of Lumalier Incorporated designed a new type of air circulation/UVGI light combination called a “Silent Air Mover” (referred to as SAM). This new light helps to continually circulate the air throughout the room (remains on 24 hours a day) and into the attached UVGI TB-killing chamber.

SAM units in bunk area

Photo shows SAM units in bunk area.

In October 2004, another team from NIOSH returned to conduct a tracer gas study in the Harbor Light Shelter main building. This is where clients enrolled in long-term drug or work programs reside and is also where the staff offices as well as the dining and meeting rooms are located.

Other intervention measures

The installation of large, unusual lighting devices and of large tanks releasing tracer gas, along with the other uncommon activities being conducted in the shelter, were deemed potentially intimidating to the residents, and we felt that the perceptions of the residents and staff needed to be considered. Ongoing in-service meetings were scheduled with clients to reassure them that the “strangers” in their facility were there to improve the air quality and make it safer. Educational TB literature, handouts, and posters were provided and time was allotted for questions and answers. The fact that staff and residents noted the improved air quality in their shelter helped foster a positive reception.

Lessons learned

Several key lessons were learned during the last year and a half of this project.

  • Persevere; it is important to keep sight of the goals of the project strategy despite obstacles. Throughout this project there were a number of delays. Some were expected, such as scheduling members of the team so they could be at the shelter at the same time. Other obstacles were more difficult to overcome. Obtaining funding for the UVGI lights and getting approval through the various levels of government and shelter management was time consuming, but worth the effort.
  • Provide education and outreach. The shelter working conditions during the project were made easy in part by the ongoing information sessions held with clients and staff. Once they realized that different people would be coming into their facility to make improvements, everyone was helpful.
  • “Toot your own horn.” Members of the press, the television and radio stations, and the St. Louis Science Center were invited to two different media events held at the shelter. The TB Program, CDC, the State of Missouri Health Department, and the Salvation Army all received positive press (St. Louis Post-Dispatch) and TV and radio network coverage (Fox, CBS, NBC, and Media Network Radio). We coordinated the events through the St. Louis City Health Department and Salvation Army public relations offices. The media will come if there is news.
  • Promote communication and teamwork. Keeping all parties informed of even small accomplishments or obstacles helped to forge a team atmosphere among the 11 different agencies and vendors. By doing this, everyone came on-site more prepared and were able to order, manufacture, or bring with them needed supplies or equipment.


There have been no TB cases reported from the shelters since August 2003. As we proceeded with installation of the engineering improvements throughout the winter and spring, we remained focused on identifying any other measures that we could implement. Although some of the homeless clients only stay at the shelter for a few days, on any given day a client with active TB could enter the shelter and put the resident population at risk. Engineering improvements are one part of the overall TB prevention strategy; TB education is another important ongoing prevention element. Staff and visiting nurses are provided with signs-and-symptoms checklists, and posters are placed in sleeping quarters and shower hallways.

The feasibility of implementing some or all of the TB strategy of this project within other homeless shelter networks in other parts of the nation is greatly dependent upon the size of the shelters, commitment of the participating shelter management, and availability and negotiation posture of the parties involved. Above all, remaining persistent and forging sincere partnerships were the keys to the success of this project.

—Submitted by Ted Misselbeck
PHA, City of St. Louis Health Department
and Lynelle Phillips, RN, MPH
PHA, Missouri Department of Health and Senior Services


Released October 2008
Centers for Disease Control and Prevention
National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention
Division of Tuberculosis Elimination -

Please send comments/suggestions/requests to:, or to
CDC/Division of Tuberculosis Elimination
Communications, Education, and Behavioral Studies Branch
1600 Clifton Rd., NE - Mailstop E-10, Atlanta, GA 30333