To the Editor—There is a lack of robust data on the health, infection control, and economic consequences of chickenpox (varicella) among healthcare workers. Chickenpox is potentially fatal, and adults contribute to most cases of chickenpox-related mortality.Reference Noah ¹ From 1985 to 1997 there was an average of 9.22 case fatalities per 100,000 population in England and Wales due to chickenpox.Reference Noah ¹ Many individuals in the tropics, especially those coming from rural areas, may be nonimmune to varicella. For instance, only 5 (3.3%) of 153 urban adults were seronegative for varicella zoster virus (VZV) immunoglobulin G (IgG) in India compared with 74 (30.1%) of 246 rural adults. Ninety-six percent of urban adults were immune by the age of 25, compared with 42% in the rural group.Reference Mandal, Mukherjee, Murphy, Mukherjee and Naik ² In our center, of the 956 VZV IgG tests performed for immunity, 593 (62%) were found to be reactive (immune to varicella) from May 2011 to June 2015; also, 26 samples had indeterminate VZV IgG reactivity. The live attenuated varicella vaccine (contraindicated in immunosuppressed or pregnant patients as well as those with previous anaphylaxis) is relatively safe with few adverse effects (injection site pain, redness, or mild rash in 10% of adults).Reference Salisbury, Mary Ramsay and Noakes ³ Although many developed countries offer the varicella vaccine (eg, National Health Service, United Kingdom) to nonimmune healthcare workers, in India and many other developing countries this vaccine is not offered free to staff mainly because of cost issues. A single dose of varicella vaccine costs 1,200 rupees (US $20) and for an adult 2 doses are needed, spaced 4–8 weeks apart.Reference Salisbury, Mary Ramsay and Noakes ^{3 , 4}

In this report we document the economic cost of a chickenpox outbreak in an oncology center in eastern India. We consider which is a more favorable option in terms of staff health and health economics: free varicella vaccination to nonimmune staff or allowing the virus to take its natural course and infect susceptible contacts. From November 1, 2014, through June 30, 2015, there were 32 cases of chickenpox documented among healthcare workers of Tata Medical Center. This included 13 nurses (41%), 1 radiology technologist (3%), 12 housekeeping (HK) staff (38%), 4 customer care staff (13%), and 2 doctors (6%). The median (range) age of the affected staff was 25 (20–37) years. There were 12 men (38%) and 20 women (63%). None of the female staff affected were pregnant. The median (range) duration of rash was 5 (4–8) days. Complete data about the rash were not available for 3 staff. Complete information about antiviral (acyclovir) therapy (800 mg 5 times daily for 7 days orally) was available in 21 (66%) of 32 staff, and all of them had taken the prescribed antiviral medication. Data for antiviral therapy was not available for the remaining 11 affected staff (these were housekeeping staff who were seen externally by other staff health physicians). Complications (eg, hepatitis, pneumonitis, encephalitis) were found in none of the affected staff. Previous VZV IgG serology was known in 16 of 32 staff; of these 16 staff, 15 (94%) were found to be nonreactive to VZV IgG, suggesting absence of immunity against chickenpox. None of the staff who were nonimmune had previously received VZV vaccination. The suspected index case (source patient) was known in 10 (31%) of the 32 cases. The apparent source of infection was the hospital in 15 cases (47%), staff hostel in 5 (16%), and unknown in 12 (38%). The total number of staff days lost was 555 days for the 32 staff; the median (range) was 14 (4–53) days. The median (range) duration of leave beyond the resolution of rash was 9 (0–47) days. Data about sick leave taken were not available for 3 staff. No deaths occurred. We performed additional investigations as follows: complete blood count (2 cases), renal function tests (urea, creatinine, sodium, potassium; 1 case), liver function tests (2), chest radiograph (2), bacterial culture (1), and other viral serology (0). The total cost of management of all the cases (eg, investigations, medicines [antiviral agents, antipyretic]) was 18,464 rupees (US $290). The median cost was 805 rupees (US $13). In terms of human resource days lost, the total cost was 220,667 rupees (US $3,678); the median cost was 6,750 rupees (US $112) (online Table 1).

From a health economic point of view, the issue of universal screening of all staff followed by universal vaccination of susceptible staff is complex. In our institution we have screened all medical, nursing, and technical staff free at a cost of 600 rupees (US $10) per VZV IgG screen. The hospital has followed a free screening policy but relied on a voluntary—but not free—VZV vaccination policy. A cost-effectiveness model of varicella vaccination supported the “screen, then vaccinate” strategy of employees. In the model, vaccination of all employees prevented 35 employee infections and 674 patient exposures for every 10,000 potentially susceptible employees. The cost of preventing 1 employee infection was approximately US $15,000, and the cost of preventing 1 patient exposure was approximately US $775.Reference Gayman ⁵ The hidden cost of a chickenpox outbreak among staff must also be taken into account. This includes the chance of spread of disease to vulnerable patients in a cancer hospital as well as staff absenteeism. It appears reasonable to offer free varicella vaccine for those staff who are in close contact with severely immunocompromised patients. Preventing outbreaks would require greater staff awareness and more-affordable varicella vaccines.