We built a SEIR (susceptible, exposed, infected, recovered) model of smallpox transmission for New York, New York, USA, and Sydney, New South Wales, Australia, that accounted for age-specific population immunosuppression and residual vaccine immunity and conducted sensitivity analyses to estimate the effect these parameters might have on smallpox reemergence. At least 19% of New York’s and 17% of Sydney’s population are immunosuppressed. The highest smallpox infection rates were in persons 0–19 years of age, but the highest death rates were in those

Smallpox virus was eradicated in 1980 but remains a category A bioterrorism agent (

Many researchers who have developed smallpox models have been optimistic about residual vaccine-induced immunity and assumed a case-fatality ratio (CFR) of 30%, whereas estimates of outbreaks in nonimmune populations suggest a CFR of 50%–70% (

The immunologic status of the population has also changed dramatically in the decades since smallpox eradication. A larger proportion of the population today is unvaccinated, and residual immunity in persons who were vaccinated before 1980 is waning (

Persons born after 1980 have no immunity to smallpox because they have never been exposed to wild-type virus or been vaccinated. For vaccinated cohorts, immunity wanes over time, and the highest protection is present during the first 5 years after vaccination, possibly waning to zero within 5–10 years (

We used Sydney’s population in 2015 (

Characteristics of population used to model smallpox transmission, by age group, New York, NY, USA, and Sydney, New South Wales, Australia. Characteristics (e.g., size, age, immunosuppression rates) of populations from 2015 were used. A) Total population; B) immunosuppressed population.

We considered common types of immunosuppression estimated in an influenza study (

We sourced data for each city, and when only countrywide data were available, we attributed rates from the countrywide data set to the respective fraction of the population in the city. When age-specific immunosuppression prevalence data were not available, we used yearly age-specific incidence data instead to calculate prevalence age distribution (

We estimated the populations living with cancer (

In the United States, including New York, widespread smallpox vaccination occurred until 1970 (

For New York, we assumed 80% of the healthy population 40–69 years of age (born before 1975) were previously vaccinated. For Sydney, we estimated the proportion of persons vaccinated by estimating those born before 1980 in the following groups: healthcare workers in Sydney in 2015 (

In our model, we used the heterogeneous age-specific contact rates from the European mixing patterns study (

We categorized smallpox disease into 4 different types defined by infectivity (R_{0}) and CFRs: hemorrhagic, flat, ordinary, and vaccine-modified. Age-specific and other model parameters (

We assumed infected persons had different probabilities of developing each disease type, depending on their age and immunologic status. The incidence of each disease type within each age group for healthy unvaccinated persons was drawn from historical outbreaks (

We constructed a modified SEIR (susceptible, exposed, infected, recovered) model for smallpox transmission (_{0} for hemorrhagic, flat, ordinary, and vaccine-modified smallpox. Finally, we multiplied the force of infection by a parameter (α_{1,} α_{2,} α_{3,} α_{4};

The model ran for 100 simulated days. We assumed an attack in a crowded public space, such as an airport, and started the epidemic with 51 infected in New York and 29 in Sydney to reflect the same attack rate for each population. We assumed a dynamic population updated each day using the birth (

We conducted a sensitivity analysis on the assumption of waning immunity, reducing immunity by 0.7% per year (approximately half the value used in the base case scenario [i.e., the first scenario discussed]). We present results for 3 different assumptions about residual vaccine immunity: no residual immunity, base case immunity (1.41% waning immunity per year), and high residual immunity (0.7% waning immunity per year). We also conducted a sensitivity analysis to test the model outputs without considering population immunosuppression, which has been the approach in most past models (

We examined the population age distributions of New York and Sydney. Sydney has a higher percentage of persons <20 and >55 years of age than New York (

We analyzed age-specific infection (

Smallpox infection and death rates of population for base case scenario and for scenario including immunosuppression in model, by age group, New York, NY, USA, and Sydney, New South Wales, Australia. Characteristics (e.g., size, age, immunosuppression rates) of populations from 2015 were used. A) Infection rate 50 and 60 days after start of smallpox outbreak; B) cumulative deaths in population 50 and 60 days after start of smallpox outbreak.

Cumulative deaths per 1,000 population increase with age starting with persons

Looking at total rates over time, New York (

Smallpox infection and death rates over time considering different immunologic factors included in model, New York, NY, USA, and Sydney, Australia. Characteristics (e.g., size, age, immunosuppression rates) of populations from 2015 were used. A) Rates for New York, considering different levels (none, base case, and high) of residual vaccine immunity with the inclusion of immunosuppressed population. B) Rates for Sydney, considering different levels (none, base case, and high) of residual vaccine immunity with the inclusion of immunosuppressed population. C) Rates for New York, including and excluding immunosuppression with base case level of residual vaccine immunity. D) Rates for Sydney, including and excluding immunosuppression with base case level of residual vaccine immunity.

Infection and death rate estimates for New York, where vaccine coverage is more than double that of Sydney, are more sensitive to assumptions of residual immunity. New York (

Smallpox infection and death rates with different levels of residual vaccine immunity including and excluding immunosuppression in model of smallpox transmission, by age group, New York, NY, USA, and Sydney, Australia. Characteristics (e.g., size, age, immunosuppression rates) of populations from 2015 were used. A) New York 50 days after start of smallpox outbreak with no (top), base case (middle), and high (bottom) residual vaccine immunity. B) Sydney 50 days after start of smallpox outbreak with no (top), base case (middle), and high (bottom) residual vaccine immunity.

Infection and death rates increase when including (vs. excluding) immunosuppression parameters in the model; greater differences are seen between New York’s infection rates (

With each passing year, population immunosuppression is a more influential determinant than residual vaccine immunity of the severity of a smallpox epidemic. Although the spread of disease is highest in younger age groups, driven mostly by their higher contact rates, higher death rates were seen in older populations, reflecting the prevalence of immunosuppression.

The differences between New York, which has high vaccination coverage (an estimated ≈22% of the population), and Sydney, which has low (≈10%) vaccination coverage, demonstrate that residual immunity assumptions are not as influential in Sydney as in New York. However, the consideration of population immunosuppression, from medical conditions to iatrogenic factors, strongly affects disease transmission and deaths in both cities. This large population subset must be considered when modeling the impact of any infectious disease outbreak. We estimated conservatively that almost 1 in 5 persons in New York and 1 in 6 persons in Sydney (and higher for the 60–64-year age group) are living with some degree of immunosuppression. Although New York has higher rates of immunosuppression for the 25–84-year age groups, Sydney has higher rates than New York for the youngest (0–19 years) and the oldest (

Residual immunity affects age-specific infection and death rates, with both cities showing the highest infection rates for unvaccinated young persons 5–19 years of age. However, death rates rise after 40 years of age, despite higher vaccination coverage in this age group. For Sydney, even an assumption of higher immunity does not affect the infection or death rates greatly because of the low vaccine coverage before 1980. However, residual immunity becomes more influential if we use more optimistic assumptions of waning immunity. Note that persons who have been vaccinated would mount a more robust and rapid response to revaccination in the event of an outbreak and might be better protected after postexposure vaccination. Obtaining a vaccination history and checking for a consistent scar are necessary parts of outbreak management.

Although immunosuppression is a major determinant of the size and distribution of a smallpox outbreak, this fact should not be a major determinant of vaccination policy. Immunosuppression should continue to be an absolute contraindication for vaccination of persons who are not true contacts. Ensuring that persons with immunosuppression (including healthcare workers) avoid contact with persons with smallpox (if possible) should be a priority. Smallpox would always be more pathogenic than vaccinia virus, so any patient with a bona fide exposure to smallpox should be vaccinated with a fully potent vaccinia strain, such as ACAM2000 (

Our study is subject to some limitations. We used an underestimate of immunosuppression; other conditions causing immunosuppression, such as diabetes, were not considered. We also used conservative estimates for the increased risk for infection in immunosuppressed persons and grouped persons with severe and moderate immunosuppression into single categories because of the absence of more specific data to categorize them further by degree of immunosuppression. The contact matrix we used was estimated in a study conducted in the United Kingdom in 2006, which might not necessarily reflect New York or Sydney social contact patterns (

The speed and vigor with which smallpox control efforts are implemented should be major aspects of control efforts and need to be tested in a model that accounts adequately for immunosuppression. Ensuring adequate hospital care and isolation facilities will also help in epidemic control. During the Ebola epidemic in West Africa, lack of beds resulted in widespread community transmission, and modeling showed that 70% of patients needed to be in treatment facilities to control the epidemic (

Given waning smallpox vaccine immunity (nearly 4 decades since eradication and a dwindling vaccinated population), the influence of population immunosuppression is greater than that of residual vaccine immunity, yet has not been adequately considered in smallpox epidemic modeling. Advances in medicine and new endemic diseases, such as HIV, have resulted in almost 1 in 5 persons living with immunosuppression in large metropolitan cities. Immunosuppression must be considered in preparedness planning and poses a challenge for vaccination strategies during potential smallpox outbreaks.

Description of the SEIR (susceptible, exposed, infected, recovered) model of smallpox transmission and model parameters.

Dr. MacIntyre is Professor of Infectious Diseases Epidemiology at the University of New South Wales, Sydney, and leads a research program on biosecurity, bioterrorism, and emerging infectious diseases, which are her primary research interests.