Performance of a rapid dermatology referral system during the anthrax outbreak

BRIEF

REPORTS

Performance of a rapid dermatology referral system during the anthrax outbreak John T. Redd, MD, MPH, FACP,a Chris Van Beneden, MD, MPH,b Nicholas A. Soter, MD,c Eric Hatzimemos, JD,* and David E. Cohen, MD, MPHc Atlanta, Georgia, and New York, New York The bioterrorism-related anthrax outbreak generated unanticipated demand for dermatologic services. In this study we sought to perform rapid, efficient, cost-effective evaluation of patients suspected of having cutaneous anthrax. During the outbreak, we developed an anthrax evaluation system featuring clinical field examination by nondermatologist physicians, followed by rapid referral of selected high-risk patients to a centralized dermatology center. We excluded anthrax in 29 previously screened high-risk patients. All were examined within 24 hours, costing $272.07 per patient. Diagnoses were established quickly (median, same day; range, 0-15 days). Among 2259 at-risk postal workers, 144 (6.4%) self-identified new (#14 days) skin lesions and were examined in the field; 8 (5.6%) were referred to our system. Our system was not the only local dermatologic resource available during the outbreak. A system featuring initial nondermatologist examination with minimal laboratory evaluation, followed by rapid centralized referral of high-risk patients, functioned efficiently in this outbreak. ( J Am Acad Dermatol 2005;52:1077-81.)

D

uring the bioterrorism-related anthrax outbreak of the fall of 2001, 8 of 22 anthrax patients were identified in New York City (NYC). Seven of the NYC patients had cutaneous anthrax.1,2 Because cutaneous anthrax typically does not result in death if treated properly,3,4 public health plans in place before the 2001 anthrax outbreak had focused almost exclusively on inhalational anthrax.5 Although the NYC anthrax outbreak was largely cutaneous, it caused a high level of public anxiety and a massive public health response.1,2,6-9 Unfortunately, the nonspecific symptoms of early cutaneous anthrax infection made it difficult for physicians to

From the Epidemic Intelligence Service, Division of Applied Public Health Training, Epidemiology Program Office, US Centers for Disease Control and Prevention (CDC),a Atlanta; the Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, CDCb; and The Ronald O. Perelman Department of Dermatology, New York University School of Medicine.c *Mr Hatzimemos was Assistant Council to the Mayor, Office of the Mayor, City of New York, at the time of the anthrax outbreak. Reprint requests: John T. Redd, MD, MPH, FACP, Prevention Branch, Division of Viral Hepatitis, National Center for Infectious Diseases, Centers for Disease Control and Prevention; Chief, Viral Hepatitis and Chronic Liver Disease Section, Infectious Diseases Branch, Indian Health Service Division of Epidemiology, 5300 Homestead Rd NE, Albuquerque, NM 87110. E-mail: [email protected]. 0190-9622/$30.00 doi:10.1016/j.jaad.2005.02.047

confidently reassure patients that they did not have anthrax.6 At the same time, however, rapid identification of anthrax cases was essential for public health response. Consequently, public health responders from the NYC Department of Health, Communicable Diseases Division (NYCDOH) and the Centers for Disease Control and Prevention (CDC) identified an urgent need for fast, convenient, and centralized dermatological evaluation of suspected cutaneous anthrax infections. The development of a cutaneous anthrax rapid referral system in the midst of an outbreak has not been previously reported in the medical literature. In this article, we recount the development of the Cutaneous Anthrax Rapid Referral System (CARRS) during the anthrax outbreak and present clinical and epidemiologic findings of the patients evaluated by this new system.

METHODS Development of CARRS Initial steps in the development of CARRS included the identification of key personnel, securing of clinic space, and distribution of information to clinicians. Key personnel included one contact person from the combined NYCDOH/CDC field investigation team, a clinical dermatologist from the medical center, the dermatology resident on call, the clinic’s medical director and administrator, and the chairman of Dermatology. A single person remained identified in each of the defined roles 24 hours a day. 1077

1078 Brief reports

J AM ACAD DERMATOL JUNE 2005

Referral procedures Our goal when developing CARRS was to design, implement, and then coordinate a multistage screening and referral process for patients with potential cutaneous anthrax infections. Patients for whom the diagnosis of anthrax was still a possibility after being screened in the field were referred to the Charles C. Harris Skin and Cancer Unit, Tisch Hospital, the University Hospital of New York University Medical Center (NYU). Patients, who were seen without regard to ability to pay, were referred from 3 sources: 1. Physician calls to the NYCDOH/CDC anthrax hotline. Calls to the anthrax hotline were initially screened by an NYCDOH nurse, who sent calls concerning potential anthrax cases to an NYCDOH clinician. The clinician had the option to refer the call to an NYCDOH or CDC epidemiologist for further investigation. The epidemiologist, in conjunction with the treating physician, had the option of using CARRS for specialty dermatological evaluation. 2. The National Pharmaceutical Stockpile Antibiotic Distribution Project. Because of the known risk encountered during mail handling, antibiotics were distributed to NYC postal workers who may have been exposed to anthrax spores from anthrax-containing letters that passed through postal sorting machines at the Morgan Processing and Distribution Center in Manhattan, NY. An antibiotic distribution project was established by the US Postal Service, CDC, and the Commissioned Corps Readiness Force of the US Public Health Service. Through this project, 2259 postal workers were offered antibiotic prophylaxis and given their first 10 days of antibiotics if prophylaxis was accepted.7 Physicians at the Distribution Project examined all new (onset #14 days) skin lesions and referred the patients to CARRS if they thought specialty consultation with dermatology was needed to definitively exclude anthrax. 3. In-house referrals on NYU patients. CARRS was a resource for evaluation of both inpatients and outpatients at NYU hospital with skin lesions suspected of being anthrax. Patient finding A patient was considered to have been evaluated by CARRS if his or her medical record indicated a visit to the Charles C. Harris Skin and Cancer Unit at NYU, or an evaluation by CARRS in another location at NYU, for the specific purpose of ruling out cutaneous infection with Bacillus anthracis between Oct 22 (when CARRS began operation) and Dec 31, 2001

(after the anthrax outbreak ended). CARRS maintained ongoing records of patients referred to the clinic during the outbreak. After the outbreak these records were supplemented by (1) a computer search of NYU medical records databases for patients with diagnoses containing anthrax-related International Classification of Diseases, Ninth Revision codes, (2) computer searching of the clinical laboratory system database for text fields containing ‘‘anthrax’’; and (3) canvassing of dermatology residents at NYU to determine whether they recalled any other consultations that were not ascertained by the other methods. Data collection Information was abstracted into a Microsoft Excel database by using a standardized data collection form. NYCDOH outbreak records were cross-referenced with NYU charts. We also abstracted data from a Microsoft Access database maintained by CDC personnel during the National Pharmaceutical Stockpile Antibiotic Distribution Project. Data collected included patient demographics; date, time, and location of clinical evaluation by CARRS; source, date, and time of referral; dermatological history; employment information (including employment with the US Postal Service); risks for exposure to anthrax (eg, receipt of suspect mail); duration of complaints; vital signs; lesion characteristics (eg, pain, erythema, presence of vesicles); laboratory testing results; final diagnoses including date and time of determination; and treatment. Cost estimation Cost estimates for personnel were calculated by estimating the percentage of an employee’s full-time workweek that he or she spent on CARRS, averaged over the 72 days the system was in operation. Costs of diagnostic procedures were calculated using the fee schedule for the Charles C. Harris Skin and Cancer Unit. Cost estimates for communications were based on current rates for Nextel service and hardware. Human subjects review The CDC Human Subjects Board and the Institutional Review Board of NYU Hospital reviewed and approved our plan for retrospective review of patient data collected during the outbreak. Our investigation was conducted after the outbreak ended.

RESULTS Twenty-nine patients were evaluated between Oct 22 and Dec 31, 2001 (Fig 1). All patients were

Brief reports 1079

J AM ACAD DERMATOL VOLUME 52, NUMBER 6

Fig 1. Number of patients evaluated for suspected cutaneous anthrax by CARRS, by date, between October, 21, 2001 and December, 31, 2001.

evaluated within 24 hours of referral. Patient visits were clustered around two important events: the start of the National Pharmaceutical Stockpile Antibiotic Distribution Project on October 24 and the announcement on Oct 29 by the NYCDOH that inhalational anthrax had been confirmed in NYC,8 which was followed 2 days later by news of the patient’s death.9 Sources of patient referrals Six of the 29 patients (21%) were referred from NYCDOH anthrax hotline calls, 7 (24%) from the National Pharmaceutical Stockpile Antibiotic Distribution Project, and two (7%) by private physicians with no call to NYCDOH documented. For 14 patients (48%), no source of referral was documented. Of the 2259 postal employees offered prophylaxis during the National Pharmaceutical Stockpile Antibiotic Distribution Project,7 667 (30%) self-identified as ill (defined as presence of a skin lesion with onset in the previous 14 days or current flu-like symptoms). Among employees who self-identified as ill, 144 (22%) reported new-onset skin lesions and were therefore examined by Public Health Service physicians who were part of the distribution project. Eight of the 144 employees (6%) had new skin lesions believed to be worrisome enough for cutaneous anthrax to warrant referral to NYU using CARRS; 7 kept their appointments. Patient characteristics The median age of the 29 patients seen at CARRS was 49 years (range, 7-74 years); 12 (41%) were female. Twenty-three (79%) were evaluated in the

Table I. Clinical findings in 29 patients referred to CARRS for evaluation of suspected cutaneous anthrax, NYC, Oct 21, 2001-Dec 31, 2001 No. of patients (%)

Clinical finding Febrile ([378C) Lesion characteristics Erythema Pruritus Erythema and/or pruritus Vesicle(s) Pain Tenderness Eschar Edema Ulcer(s) Satellite lesion(s)

1 (3) 21 14 26 4 4 2 2 2 1 0

(72) (48) (90) (14) (14) (7) (7) (7) (3) (0)

Harris Skin and Cancer Unit at NYU and 6 (21%) were evaluated in the Emergency Department. The median interval between onset of the skin lesion and seeking medical evaluation was 4.5 days (range, 1-21 days; one outlier with an interval of 150 days excluded). Nine of 23 patients (39%) with available employment data were postal workers. Clinical findings and diagnostic tests performed History and physical findings are presented in Table I. For this analysis, we assumed that variables missing from the chart had been assessed and were not present. Minimal testing was performed on the 29 patients, although some patients had multiple

1080 Brief reports

J AM ACAD DERMATOL JUNE 2005

Table II. Estimated personnel costs and material/laboratory costs of CARRS, Oct 21, 2001 through Dec 31, 2001: Personnel and material/laboratory costs for evaluation of 29 patients Type of employee

% One full-time equivalent spent on CARRS

Annual salary

Cost

10 20* 10

$85,000 $45,000 $40,000

$1676 $1774 $ 789 $4239

Cost per procedure/expense

Total cost

Attending physician Resident physician Clinic administrator Summary of personnel costs Procedure/expense

No.

Punch biopsy procedure fee Dermatopathology, routine (hematoxylin-eosin stain) Dermatopathology, Gram stain Bacterial culture procedure fee Bacterial culture laboratory fee Mobile phone purchasey Mobile phone servicey Summary of material and laboratory costs Total costs of CARRS (personnel, material/laboratory)

9 9

$40 $95

$ 360 $ 855

9 9 9 5 5

$50 $30 $49 $75 $60/phone/mo

$ 450 $ 270 $ 441 $ 375 $ 900 $3651 $7890

*Based on a 24-hour workday. y Five Nextel phone radios were provided to CARRS by the City of New York. Estimated expenses calculated by using Nextel Free Incoming 400 plan with Nextel i35s phone radios.

tests: 9 (31%) had skin biopsy, 9 (31%) had bacterial culture of a skin lesion, 1 (3%) had viral culture of a skin lesion, and 1 (3%) had a Tzanck test for herpes simplex virus. Final diagnoses For 15 patients (52%) a final diagnosis was determined within 1 day of CARRS evaluation (median, same day; range, 0-15 days). None of the 29 patients were diagnosed with cutaneous anthrax infection. The most common diagnosis (8/29, 28%) was a bacterial skin infection (ie, cellulitis, folliculitis, furuncle, impetigo). Four (14%) were diagnosed with arthropod bite; 3 (10%) with herpes simplex virus infection; 3 (10%) with eczema; 2 (7%) with pyoderma; and 1 patient (3%) with each of the following: contact dermatitis, dermatofibroma, lipoma, prurigo nodularis, pruritus, psoriasiform spongiotic dermatitis, seborrheic dermatitis, solar keratosis, and tinea pedis. Costs A tally of estimated costs is presented in Table II. The system spent an estimated $7890 to evaluate the 29 patients referred to CARRS. Costs were $3.49 per patient evaluated at the National Pharmaceutical Stockpile Antibiotic Distribution Project (n = 2259) and $272.07 per patient seen at NYU (n = 29).

DISCUSSION This is the first report of the development and performance of a dermatologic referral system in the

midst of an outbreak of bioterrorism-related infection. CARRS, established during the anthrax outbreak in NYC, evaluated patients quickly, established alternative diagnoses efficiently, and did so using only minimal laboratory testing. Although these patients were heavily selected for elevated risk of cutaneous anthrax, no patient required use of blood culture, anthrax serology, anthrax polymerase chain reaction, or nasal swab culture. Important features of the system included rapid referral to dermatologists using same-day appointments without regard to insurance or ability to pay; screening of lesions by nondermatologist physicians before referral; and referral to a facility with sophisticated in-house laboratory facilities. Data from the National Pharmaceutical Stockpile Antibiotic Distribution Project provided valuable insights. First, 6% of postal workers identified new (#14 days) cutaneous lesions. Although this rate may not apply to other populations and settings, it does provide a reference point for public health officials making bioterrorism preparedness plans. Second, by using a strategy of nondermatologist physician screening before referral to specialists, only 8 of 2259 at-risk postal workers (0.4%) were referred to Dermatology. This result suggests that our strategy of initial nonspecialist evaluation, with subsequent rapid referral of high-risk cases to a centralized facility, was an effective one. The cost of ruling out anthrax in these patients was minimal. Our cost estimates, however, are

J AM ACAD DERMATOL

Brief reports 1081

VOLUME 52, NUMBER 6

subject to some important caveats. First, many of the routine costs of patient evaluation at CARRS, such as gauze pads and suture material, are part of the overall clinic budget and were not counted separately. Second, we have evaluated only the NYUspecific costs of CARRS and have not included person-time from the NYCDOH, CDC, or the Office of the Mayor of the City of New York. Third, if the nation suffers another bioterrorism attack, cost is unlikely to be a major consideration in the affected area. However, as US medical institutions saw in 2001, worried low-risk citizens seek evaluation for possible bioterrorism-related infection even at locations far from the actual site of an attack. Cost estimates, therefore, are relevant to bioterrorism response plans in diverse localities. Our study has several important limitations. Our system was one of many possible sources of dermatological evaluation for patients in NYC suspected of having cutaneous anthrax. In addition, we cannot assess the ability of this system to perform in the midst of a larger outbreak. However, we hope that our experience will help others who are making plans for outbreaks of bioterrorism-related agents, particularly those agents that are likely to result in skin manifestations. The anthrax outbreak of 2001, in which 7 of 8 NYC cases were cutaneous, showed that dermatological resources are a crucial part of bioterrorism response. Our system’s success may have implications for planning for the response to other bioterrorism agents, such as smallpox. Dermatologists, who have not historically been an integral part of bioterrorism planning, should now be considered a crucial resource for preparedness. We gratefully acknowledge the assistance of the officers of the Epidemic Intelligence Service, CDC; the

CDC staff members deployed to the anthrax outbreaks; the resident physicians of The Ronald O. Perelman Department of Dermatology, NYU School of Medicine, New York; the Commissioned Corps Readiness Force of the US Public Health Service; the US Postal Service; the City of New York; Marcelle Layton, MD; Irwin M. Freedberg, MD; David A. Ashford, DVM, DSc, MPH; Marc A. Fischer, MD; Michael S. Phillips, MD; Theresa Munoz-DiMarco; Thomas To¨ro¨k, MD; Wallace Wilhoite, CISSP; C. Mack Sewell, DrPH, MS; Ronald E. Voorhees, MD, MPH, and James E. Cheek, MD, MPH.

REFERENCES 1. Jernigan DB, Raghunathan PL, Bell BP, Brechner R, Bresnitz EA, Butler JC, et al. Investigation of bioterrorism-related anthrax, United States, 2001: epidemiologic findings. Emerg Infect Dis 2002;8:1019-28. 2. New York City Department of Health, Office of Public Affairs. Press release, 10/12/01: Health department announces an anthrax case in New York City. 3. Inglesby TV, Henderson DA, Bartlett JG, Ascher MS, Eitzen E, Friedlander AM, et al. Anthrax as a biological weapon: medical and public health management. JAMA 1999;281:1735-45. 4. Pile JC, Malone JD, Eitzen EM, Friedlander AM. Anthrax as a potential biological warfare agent. Arch Intern Med 1998;158:429-34. 5. Centers for Disease Control and Prevention. Biological and chemical terrorism: strategic plan for preparedness and response. Recommendations of the CDC Strategic Planning Workgroup. MMWR 2000;49:1-14. 6. Genzlinger N. Jersey: Anxiety spreads, not anthrax. New York Times, Nov 11, 2001. Late Edition-Final, Section 14NJ, p 1, col 3. 7. Shepard CW, Soriano-Gabarro M, Zell ER, Hayslett J, Lukacs S, Goldstein S, et al. Antimicrobial postexposure prophylaxis for anthrax: adverse events and adherence. Emerg Infect Dis 2002;8:1124-32. 8. New York City Department of Health, Office of Public Affairs. Press release, 10/29/01: Health department announces suspected inhalation anthrax case in NYC. 9. New York City Department of Health, Office of Public Affairs. Press release, 10/31/01: Update on Anthrax Situation in New York City.