**2. The burden of salmonellosis in the United States**

Salmonellosis causes more disease burden than any other foodborne pathogen. An estimated 93.8 million cases (90% CI, 61.8-131.6 million) of gastroenteritis caused by *Salmonella* species occur globally each year and of these, nearly 80.3 million cases are foodborne (Majowicz et al., 2010). In the United States, an estimated 1 million incident cases of human salmonellosis occur annually (Scallan et al., 2011); however, only a small portion of these cases are recognized clinically (*see section 2.2*). In industrialized countries as few as 1% of clinical cases are actually reported (Heymann, 2008). Collectively, *Salmonella* infections in the United States account for roughly 19,336 hospitalizations, 17,000 quality adjusted life

infection was 17.6 cases per 100,000 persons in 2010. This was more than twice the U.S. Healthy People 2010 objective of 6.8 cases per 100,000 persons (Figure 2) (Matyas et al.,2010). Moreover, a recent report released by the Centers for Disease Control and Prevention (CDC) revealed that the incidence of *Salmonella* infections in 2010 was significantly higher than during 2006-2008 representing an increase of about 10% (95% Confidence Interval (CI), 4- 17%). However, other foodborne infections, such as *Campylobacter*, *Listeria*, *Shigella*, STEC O157, *Vibrio*, and *Yersinia*, have all actually decreased during this same period (CDC, 2011). The disease burden of salmonellosis has remained substantial in the United States in spite of ongoing public health and regulatory efforts to prevent and control this infectious disease.

Overall *Salmonella* incidence rate in 2010

Healthy People 2020 objective (11.4 per 100,000 persons)

> Healthy People 2010 objective (6.8 per 100,000 persons)

(17.6 per 100,000 persons)

Fig. 2. Laboratory-confirmed *Salmonella* incidence rate per 100,000 population, by age group, as compared to the overall incidence rate and the national health objectives (Healthy People)

The present chapter discusses the trends in morbidity, mortality, and years of potential life lost attributed to human salmonellosis in the United States. In addition, this chapter provides a snapshot of U.S. public health measures and control policies that are currently in

Salmonellosis causes more disease burden than any other foodborne pathogen. An estimated 93.8 million cases (90% CI, 61.8-131.6 million) of gastroenteritis caused by *Salmonella* species occur globally each year and of these, nearly 80.3 million cases are foodborne (Majowicz et al., 2010). In the United States, an estimated 1 million incident cases of human salmonellosis occur annually (Scallan et al., 2011); however, only a small portion of these cases are recognized clinically (*see section 2.2*). In industrialized countries as few as 1% of clinical cases are actually reported (Heymann, 2008). Collectively, *Salmonella* infections in the United States account for roughly 19,336 hospitalizations, 17,000 quality adjusted life

for 2010 and 2020, United States, 2010 (CDC, 2011).

place to protect the public against *Salmonella* infection.

**2. The burden of salmonellosis in the United States** 

years lost (QALYs), and \$3.3 billion in total medical expenditures and lost productivity each year (Batz et al., 2011).

#### **2.1 Clinical manifestations, serotypes, and outbreaks**

*Salmonella* gastroenteritis is usually a self-limited disease in which the symptom of fever typically resolves within 48 to 72 hours and diarrhea within three to seven days. Complications from the infection may include severe dehydration, shock, collapse, and/or septicemia. Symptoms are usually more severe among infants, young children, elderly, and those who are immune-compromised (Scallan et al., 2011).

Although there are many serotypes of *Salmonella* that are pathogenic to both humans and animals (i.e., approximately 2,500 serotypes have been identified), the vast majority of human *Salmonella* isolates are serotype *S. enterica* subsp. *enterica* (Heymann, 2008)*. S*erovars Typhi and Paratyphi of this serotype, *S. enterica* subsp. *enterica,* are the etiologic agents that cause typhoid and paratyphoid fevers. These types are also common, but are generally found in developing countries, such as those in South America, Africa, and parts of Asia (Heymann, 2008). In developed countries where there is active, coordinated foodborne disease surveillance, other serovars such as Typhimurium and Enteritidis are frequently reported.

Sixty to eighty percent of all human salmonellosis cases in the United States occur intermittently and sporadically throughout the population. Clusters of large outbreaks in restaurants, institutions for children, hospitals, and nursing homes have occurred recently and remain major public health threats. These outbreaks are usually the product of contamination from a production source, such as chicken farms, feed blending mills, and slaughterhouses. One of the more well-known *Salmonella* outbreaks in the United States occurred in 2010. This outbreak resulted from contamination in the food production chain, leading to a massive egg recall of over half a billion eggs and more than 2,000 reported cases of *Salmonella*-related illness (Hutchison, 2010). Although less common, outbreaks from food handling by an ill person or carrier have been reported in recent years (Cruickshank et al., 1987; Khuri-Bulos et al., 1994). For instance, in 2000 an ill food handler in a bakery that supplied hamburger buns to restaurants was found responsible for an outbreak among several burger restaurants across Southern California and Arizona. This outbreak was atypical in that it resulted from consumption of commercially distributed bread, which is a highly unusual vehicle for most foodborne infectious agents (Kimura et al., 2005).

Outbreaks from person-to-person transmission can also be of particular concern, especially among hospital workers who have the potential to spread the bacterium with their hands or through contaminated instruments. Outbreaks of *Salmonella* infection have occurred in places like maternity wards where staff members with contaminated hands and/or the use of contaminated medical instruments result in the transmission of *Salmonella* to babies and mothers (Rowe et al., 1969). In 2008, an outbreak strain of *Salmonella* serotype Tennessee occurred in a neonatal intensive care unit in the United States, where limited access to sinks for hand washing likely facilitated the transmission to infants (Boehmer, 2009).

#### **2.2 Salmonellosis incidence – FoodNet data**

Based on FoodNet surveillance data (*see section 4.1*) for nine selected foodborne pathogens from 10 states and three federal agencies (CDC, U.S. Food and Drug Administration, and U.S. Department of Agriculture), a total of 19,089 laboratory-confirmed cases of foodborne infections, 4,247 hospitalizations, and 68 deaths were identified for the year 2010 in the U.S.

The Burden of Salmonellosis in the United States 5

**Age-Adjusted Mortality Rate (95% CI)** 

0.021 (0.019-0.022) 0.038 (0.035-0.041)

0.023 (0.021-0.024) 0.059 (0.045-0.073) 0.057 (0.050-0.064) 0.031 (0.025-0.038) 0.025 (0.005-0.045)

**Age-Specific Mortality Rate**  0.086 (0.064-0.107) 0.006 (0.003-0.009) 0.002 (0.007-0.003) 0.003 (0.002-0.005) 0.012 (0.009-0.014) 0.012 (0.009-0.014) 0.020 (0.017-0.024) 0.030 (0.025-0.035) 0.073 (0.064-0.082) 0.160 (0.143-0.177) 0.314 (0.274-0.354)

Total 1,372 0.028 (0.027-0.030) N/A 21,417

Note: 95% CI = confidence interval; Years of Potential Life Lost were calculated by subtracting the age in years at the time of death from 75 years. \*Mortality rates are age-specific rates, not age-adjusted rates.

Asians had the highest age-adjusted rate ratio of 2.63 (95% CI, 2.45-2.82; n=76 deaths) relative to whites, the referent group (Table 1). While whites had the highest absolute number of deaths (n = 893), they had the lowest age-adjusted mortality rate (0.023 per 100,000 population; 95% CI, 0.021-0.024). Reasons for disparities in *Salmonella* mortality based on gender and race/ethnicity have been discussed in a previously published paper (Cummings PL et al., 2010). California and New York had the highest number of deaths (n=219 and n=105, respectively), but relatively low age-adjusted mortality rates (0.04 per 100,000 population, 95% CI, 0.04-0.05 and 0.03 per 100,000 population, 95% CI, 0.02-0.04, respectively). Although Hawaii and District of Columbia had smaller numbers, they had the highest age-adjusted

Table 1. Age-adjusted nontyphoidal *Salmonella*-related mortality rates per 100,000 population and mortality rate ratios by sex, race/ethnicity and age group, United States,

**Age-Adjusted Rate Ratio (95% CI)** 

Referent 1.84 (1.68-2.02)

Referent 2.63 (2.45-2.82) 2.53 (2.36-2.72) 1.39 (1.28-1.50) 1.11 (1.02-1.21)

> -- -- -- -- -- -- -- -- -- -- --

**Age-adjusted Rates of Potential Life Years Lost** 

> 7,970 13,447

9,768 1,256 7,095 3,050 173

4,575 1,315 797 1,204 3,797 3,236 3,262 2,004 1,227 0 0

**Frequency (N%)** 

583 (42.5%) 789 (57.5%)

893 (65.1%) 76 (5.5%) 279 (20.3%) 116 (8.5%) 7 (0.5%)

61 (4.4%) 18 (1.3%) 12 (0.9%) 22 (1.6%) 85 (6.2%) 91 (6.6%) 128 (9.3%) 134 (9.8%) 243 (17.7%) 342 (24.9%) 235 (17.1%)

**†** Numbers may not add up to total, due to missing data.

**Sex**  Female Male

**Race/Ethnicity**† White Asian/Pacific Islander Black Hispanic Native American

**Age group (years)**\*†

<1 1-4 5-14 15-24 25-34 35-44 45-54 55-64 65-74 75-84 >85

1990-2007.

(CDC, 2011). Of the nine pathogens monitored, including *Campylobacter*, *Listeria*, *Salmonella*, *Shigella*, STEC O157, *Vibrio*, *Yersinia*, *Cryptosporidium*, and *Cyclospora*, salmonellosis was the most common infection reported and had the highest number of associated hospitalizations and deaths. A total of 8,256 infections (17.6 illnesses per 100,000 persons); 2,290 hospitalizations; and 29 deaths were attributed to this pathogen in 2010. Ninety-two percent (7,564 out of 8,256) of these isolates were subsequently serotyped through PulseNet (*see section 4.1*), with Enteritidis (22%), Newport (14%), and Typhimurium (13%) representing the most common serotypes. The FoodNet data indicate that the rate of infection from *Salmonella* remains substantially high and has not declined for over a decade, as compared to the other eight foodborne pathogens tracked through FoodNet. These data support ongoing control efforts in the United States that target *Salmonella*, particularly in response to the costs associated with treatment of this infection – approximately \$365 million in direct medical costs each year (CDC, 2011).

#### **2.3 Salmonellosis-related mortality**

Current estimates indicate that there are about 155,000 salmonellosis-related deaths each year worldwide (Majowicz et al., 2010); between 400-600 of them are in the United States (Mead et al., 1999; CDC, 2008). While risk of death and actual deaths from salmonellosis are not typically common in the general population, the infection can be particularly virulent in vulnerable groups, especially among young children, older adults, and those who are immune-compromised (*see section 3*).

Table 1 presents the most updated analysis of multiple cause-of-death (MCD) data based on death certificates in the United States. From 1990 to 2007, there were 1,372 nontyphoidal *Salmonella*-related deaths. Among these reported deaths, *Salmonella* was listed as an underlying cause of death on 785 (57.2%) death certificates and as an associated cause of death on 587 (42.8%) death certificates. Fifty-six deaths occurred in 2007 alone, resulting in an age-adjusted mortality rate of 0.018 per 100,000 population (95% CI, 0.013-0.022). The average age-adjusted mortality rate over the entire study period, from 1990 to 2007, was 0.028 per 100,000 population (95% CI, 0.027-0.030; n=1,372). This represents a total of 21,417 years of potential life lost (Table 1).

Between 1990 and 2006 the age-adjusted mortality rate for human salmonellosis declined from 0.06 per 100,000 population (95%CI, 0.05-0.07; n=136 deaths) to 0.01 per 100,000 population (95%CI, 0.01-0.02; n=45 deaths). The variance between deaths and incidence, in terms of trends over the past decade show that deaths have decreased (Cummings PL et al., 2010), but incidence has increased (CDC, 2011). This difference could potentially be the result of better medical treatment or other contributing factors accounting for the decline in deaths. In 2007, however, a slight increase (albeit not significant) in the frequency and rate of *Salmonella*-related deaths was observed (Figure 3). The mean age of decedents with *Salmonella* infection listed on their death certificate for the period 1990-2007 was 63.1 years. Overall, males were more likely than females to have *Salmonella* listed as a cause of death (either underlying or associated) on their death certificate and have more years of potential life lost – 13,447 years for males versus 7,970 years for females (Table 1). Infants (< 1 year of age) and older adults (> 65 years of age) had the highest frequency of *Salmonella-*related deaths over the 18-year period (Table 1). The highest age-specific mortality rates during this period were observed among infants (0.086 per 100,000 population), those aged 75-84 (0.160 per 100,000 population), and those 85 years and older (0.314 per 100,000 population). Asian, black, and Hispanic race/ethnicity had higher rates of mortality from Salmonella infection as compared to whites.

(CDC, 2011). Of the nine pathogens monitored, including *Campylobacter*, *Listeria*, *Salmonella*, *Shigella*, STEC O157, *Vibrio*, *Yersinia*, *Cryptosporidium*, and *Cyclospora*, salmonellosis was the most common infection reported and had the highest number of associated hospitalizations and deaths. A total of 8,256 infections (17.6 illnesses per 100,000 persons); 2,290 hospitalizations; and 29 deaths were attributed to this pathogen in 2010. Ninety-two percent (7,564 out of 8,256) of these isolates were subsequently serotyped through PulseNet (*see section 4.1*), with Enteritidis (22%), Newport (14%), and Typhimurium (13%) representing the most common serotypes. The FoodNet data indicate that the rate of infection from *Salmonella* remains substantially high and has not declined for over a decade, as compared to the other eight foodborne pathogens tracked through FoodNet. These data support ongoing control efforts in the United States that target *Salmonella*, particularly in response to the costs associated with treatment of this infection – approximately \$365 million in direct

Current estimates indicate that there are about 155,000 salmonellosis-related deaths each year worldwide (Majowicz et al., 2010); between 400-600 of them are in the United States (Mead et al., 1999; CDC, 2008). While risk of death and actual deaths from salmonellosis are not typically common in the general population, the infection can be particularly virulent in vulnerable groups, especially among young children, older adults, and those who are

Table 1 presents the most updated analysis of multiple cause-of-death (MCD) data based on death certificates in the United States. From 1990 to 2007, there were 1,372 nontyphoidal *Salmonella*-related deaths. Among these reported deaths, *Salmonella* was listed as an underlying cause of death on 785 (57.2%) death certificates and as an associated cause of death on 587 (42.8%) death certificates. Fifty-six deaths occurred in 2007 alone, resulting in an age-adjusted mortality rate of 0.018 per 100,000 population (95% CI, 0.013-0.022). The average age-adjusted mortality rate over the entire study period, from 1990 to 2007, was 0.028 per 100,000 population (95% CI, 0.027-0.030; n=1,372). This represents a total of 21,417

Between 1990 and 2006 the age-adjusted mortality rate for human salmonellosis declined from 0.06 per 100,000 population (95%CI, 0.05-0.07; n=136 deaths) to 0.01 per 100,000 population (95%CI, 0.01-0.02; n=45 deaths). The variance between deaths and incidence, in terms of trends over the past decade show that deaths have decreased (Cummings PL et al., 2010), but incidence has increased (CDC, 2011). This difference could potentially be the result of better medical treatment or other contributing factors accounting for the decline in deaths. In 2007, however, a slight increase (albeit not significant) in the frequency and rate of *Salmonella*-related deaths was observed (Figure 3). The mean age of decedents with *Salmonella* infection listed on their death certificate for the period 1990-2007 was 63.1 years. Overall, males were more likely than females to have *Salmonella* listed as a cause of death (either underlying or associated) on their death certificate and have more years of potential life lost – 13,447 years for males versus 7,970 years for females (Table 1). Infants (< 1 year of age) and older adults (> 65 years of age) had the highest frequency of *Salmonella-*related deaths over the 18-year period (Table 1). The highest age-specific mortality rates during this period were observed among infants (0.086 per 100,000 population), those aged 75-84 (0.160 per 100,000 population), and those 85 years and older (0.314 per 100,000 population). Asian, black, and Hispanic race/ethnicity had higher rates of mortality from Salmonella infection as compared to whites.

medical costs each year (CDC, 2011).

**2.3 Salmonellosis-related mortality** 

immune-compromised (*see section 3*).

years of potential life lost (Table 1).


Note: 95% CI = confidence interval; Years of Potential Life Lost were calculated by subtracting the age in years at the time of death from 75 years. \*Mortality rates are age-specific rates, not age-adjusted rates. **†** Numbers may not add up to total, due to missing data.

Table 1. Age-adjusted nontyphoidal *Salmonella*-related mortality rates per 100,000 population and mortality rate ratios by sex, race/ethnicity and age group, United States, 1990-2007.

Asians had the highest age-adjusted rate ratio of 2.63 (95% CI, 2.45-2.82; n=76 deaths) relative to whites, the referent group (Table 1). While whites had the highest absolute number of deaths (n = 893), they had the lowest age-adjusted mortality rate (0.023 per 100,000 population; 95% CI, 0.021-0.024). Reasons for disparities in *Salmonella* mortality based on gender and race/ethnicity have been discussed in a previously published paper (Cummings PL et al., 2010). California and New York had the highest number of deaths (n=219 and n=105, respectively), but relatively low age-adjusted mortality rates (0.04 per 100,000 population, 95% CI, 0.04-0.05 and 0.03 per 100,000 population, 95% CI, 0.02-0.04, respectively). Although Hawaii and District of Columbia had smaller numbers, they had the highest age-adjusted

The Burden of Salmonellosis in the United States 7

As the present generation of baby boomers (those born between 1946 and 1964) reach age 65 and older, the trend in *Salmonella-*related deaths is expected to change, suggesting that more deaths could ensue, given that older adults frequently experience more severe infections and require hospitalization more often from this foodborne illness than younger adults (Kennedy et al., 2004). Trends showing increased chronic disease prevalence in the population for such conditions as cancer, autoimmune disorders, and other diseases requiring treatment with immune-suppressive therapies parallel the aging of the population and foreshadow the continual burden of human salmonellosis in the United States

Today's global market in meats, poultry, vegetables, fruits, farm animals, and pets, represents potential sources of *Salmonella* contamination that are complex and sometimes difficult to control. For example, in 2008 there was a multi-state outbreak of *Salmonella*  Typhimurium associated with frozen vacuum-packed rodents that are used to feed snakes (Fuller et al., 2008). This occurrence represents a rare, but wide-spread outbreak associated with commercially distributed rodents. Likewise, the illicit selling and importation of many animals from abroad have caused several unanticipated salmonellosis outbreaks, as well as agricultural problems for the region. In Los Angeles County, the illegal selling of red-eared slider turtles (< 4 inches in diameter) has become an important public health problem. Because caring for these animals is exceedingly difficult, they are often abandoned or dumped by their owners into wildlife preserves and adquaducts. A local animal control agency in Los Angeles County found that an increasing number of turtles have been dumped over the years; they impounded over 6,000 illegally sold, undersized red-eared slider turtles from 2000-2007 (unpublished data). The upward trend in the abandonment of turtles and the turtles' high fecundity rates may also increase the risk of transmission to native species (Perez-Santigosa et al., 2008). Nearly 10% of all reported cases of human salmonellosis in Los Angeles County have been attributed to direct or indirect contact with reptiles, namely the red-eared slider turtle, the most common reptile source found in more than 50% of these cases (LACDPH, 2008). Continual monitoring and targeted improvements to regulate the illegal selling of these animals remain key control measures for protecting the

**2.4.1 An aging population and increased burden of chronic disease** 

(Altekruse et al., 1997).

**2.4.2 An increasingly global market** 

public against acquiring *Salmonella* infections from reptiles.

**3. Salmonellosis in vulnerable groups with comorbid conditions** 

Clinical evidence suggests that infection with nontyphoid *Salmonella* often results in more severe manifestations of clinical disease than from any other foodborne pathogen (Helms et al., 2006). Comorbid health conditions and their related immuno-suppressive treatments may be particularly problematic, especially among vulnerable groups at high risk of progressing to severe forms of salmonellosis (Trevejo et al., 2003; Cummings PL et al., 2010). For example, those with HIV/AIDS, certain types of cancers (e.g., leukemia, bone marrow), or autoimmune disorders are at significantly greater risk for death, as compared to persons without these conditions. In the updated analysis of *Salmonella*-related mortality as described in *section 2.3,* a matched case-control study showed that HIV (matched odds ratio (MOR) =7.42; 95% CI, 5.26-10.47), leukemia (MOR=2.95; 95% CI, 1.48-5.88), connective tissue disorders (MOR=2.36, 95%CI, 1.42-3.93), lupus (MOR=3.83; 95% CI, 1.72-8.55), and

mortality rates during 1990-2007 (0.08 per 100,000 population, 95% CI, 0.04-0.12; n=18 and 0.08 per 100,000 population, 95% CI, 0.02-0.13; n=8, respectively).

Methods used in this updated analysis are similar to those previously described in Cummings PL et al., 2010. Briefly, years of potential life lost (YPLL) were calculated by subtracting the age in years at the time of death from 75 years (Gardner, 1990). Deaths were defined as any observation listed as either the underlying cause or the associated cause of death with the following International Classification of Diseases, 9th revision (ICD-9) and 10th revision (ICD-10) codes: 003.0-003.9 and A02.0-A02.9, respectively. These ICD codes included infection or foodborne intoxication due to any *Salmonella* species, other than serovars Typhi and Paratyphi, which are the microbial agents that cause typhoid and paratyphoid fevers. Since these latter conditions are rare in the United States and predominately occur in developing countries (e.g., countries in Southeast Asia, Africa, and South America), serovars Typhi and Paratyphi were excluded from the analysis.

Fig. 3. Number of nontyphoidal *Salmonella*-related deaths and age-adjusted mortality rates per 100,000 population by year, United States, 1990-2007.

#### **2.4 Changing trends in factors that may contribute to human salmonellosis**

Although mortality rates are important indicators of health status, they often do not tell the entire story. Factors such as the aging population; increased burden of chronic diseases that can suppress immunity; and an increasingly global market in meats, poultry, vegetables, fruits, farm animals, and pets (e.g., chicks and reptiles) are all emerging influences that can potentially amplify the risk and burden of human salmonellosis in the United States.

mortality rates during 1990-2007 (0.08 per 100,000 population, 95% CI, 0.04-0.12; n=18 and 0.08

Methods used in this updated analysis are similar to those previously described in Cummings PL et al., 2010. Briefly, years of potential life lost (YPLL) were calculated by subtracting the age in years at the time of death from 75 years (Gardner, 1990). Deaths were defined as any observation listed as either the underlying cause or the associated cause of death with the following International Classification of Diseases, 9th revision (ICD-9) and 10th revision (ICD-10) codes: 003.0-003.9 and A02.0-A02.9, respectively. These ICD codes included infection or foodborne intoxication due to any *Salmonella* species, other than serovars Typhi and Paratyphi, which are the microbial agents that cause typhoid and paratyphoid fevers. Since these latter conditions are rare in the United States and predominately occur in developing countries (e.g., countries in Southeast Asia, Africa, and

South America), serovars Typhi and Paratyphi were excluded from the analysis.

Fig. 3. Number of nontyphoidal *Salmonella*-related deaths and age-adjusted mortality rates

Although mortality rates are important indicators of health status, they often do not tell the entire story. Factors such as the aging population; increased burden of chronic diseases that can suppress immunity; and an increasingly global market in meats, poultry, vegetables, fruits, farm animals, and pets (e.g., chicks and reptiles) are all emerging influences that can

**2.4 Changing trends in factors that may contribute to human salmonellosis** 

potentially amplify the risk and burden of human salmonellosis in the United States.

per 100,000 population by year, United States, 1990-2007.

per 100,000 population, 95% CI, 0.02-0.13; n=8, respectively).

#### **2.4.1 An aging population and increased burden of chronic disease**

As the present generation of baby boomers (those born between 1946 and 1964) reach age 65 and older, the trend in *Salmonella-*related deaths is expected to change, suggesting that more deaths could ensue, given that older adults frequently experience more severe infections and require hospitalization more often from this foodborne illness than younger adults (Kennedy et al., 2004). Trends showing increased chronic disease prevalence in the population for such conditions as cancer, autoimmune disorders, and other diseases requiring treatment with immune-suppressive therapies parallel the aging of the population and foreshadow the continual burden of human salmonellosis in the United States (Altekruse et al., 1997).

#### **2.4.2 An increasingly global market**

Today's global market in meats, poultry, vegetables, fruits, farm animals, and pets, represents potential sources of *Salmonella* contamination that are complex and sometimes difficult to control. For example, in 2008 there was a multi-state outbreak of *Salmonella*  Typhimurium associated with frozen vacuum-packed rodents that are used to feed snakes (Fuller et al., 2008). This occurrence represents a rare, but wide-spread outbreak associated with commercially distributed rodents. Likewise, the illicit selling and importation of many animals from abroad have caused several unanticipated salmonellosis outbreaks, as well as agricultural problems for the region. In Los Angeles County, the illegal selling of red-eared slider turtles (< 4 inches in diameter) has become an important public health problem. Because caring for these animals is exceedingly difficult, they are often abandoned or dumped by their owners into wildlife preserves and adquaducts. A local animal control agency in Los Angeles County found that an increasing number of turtles have been dumped over the years; they impounded over 6,000 illegally sold, undersized red-eared slider turtles from 2000-2007 (unpublished data). The upward trend in the abandonment of turtles and the turtles' high fecundity rates may also increase the risk of transmission to native species (Perez-Santigosa et al., 2008). Nearly 10% of all reported cases of human salmonellosis in Los Angeles County have been attributed to direct or indirect contact with reptiles, namely the red-eared slider turtle, the most common reptile source found in more than 50% of these cases (LACDPH, 2008). Continual monitoring and targeted improvements to regulate the illegal selling of these animals remain key control measures for protecting the public against acquiring *Salmonella* infections from reptiles.

### **3. Salmonellosis in vulnerable groups with comorbid conditions**

Clinical evidence suggests that infection with nontyphoid *Salmonella* often results in more severe manifestations of clinical disease than from any other foodborne pathogen (Helms et al., 2006). Comorbid health conditions and their related immuno-suppressive treatments may be particularly problematic, especially among vulnerable groups at high risk of progressing to severe forms of salmonellosis (Trevejo et al., 2003; Cummings PL et al., 2010). For example, those with HIV/AIDS, certain types of cancers (e.g., leukemia, bone marrow), or autoimmune disorders are at significantly greater risk for death, as compared to persons without these conditions. In the updated analysis of *Salmonella*-related mortality as described in *section 2.3,* a matched case-control study showed that HIV (matched odds ratio (MOR) =7.42; 95% CI, 5.26-10.47), leukemia (MOR=2.95; 95% CI, 1.48-5.88), connective tissue disorders (MOR=2.36, 95%CI, 1.42-3.93), lupus (MOR=3.83; 95% CI, 1.72-8.55), and

The Burden of Salmonellosis in the United States 9

Salmonellarelated deaths (N=1,371),a N(%)b

Matched control deaths (N=5,484), N(%)b

Matched odds ratios (95%CI)

(0.76-1.18)

(2.08-2.92)

(1.14-2.00)

(5.26-10.47)

(1.96-2.87)

(3.09-6.01)

(0.63-1.52)

(1.55-2.58)

(2.25-3.32)

(3.16-32.91)

ICD-9 and ICD-10 codes (respectively)

unspecified) 480-488, J10-J18, P23 107 (7.80) 445 (8.11) 0.95

system (all types) 520-579, K00-K93 246 (17.94) 447 (8.15) 2.46


HIV 042, B20-B24 133 (9.70) 121 (2.21) 7.42

Renal Failure (all types) 580-589, N17-N19 197 (14.37) 374 (6.82) 2.38




septicemia) 038, A40.9, A41 193 (14.08) 304 (5.54) 2.73

Sickle-cell disorders 282, D57 13 (0.95) 6 (0.11) 10.2

a One case was excluded due to missing variables. b Numbers may not add up to total due to missing

In the United States, surveillance for *Salmonella* infections has been an ongoing effort since 1996. Managed by the CDC, the Foodborne Diseases Active Surveillance Network (FoodNet) collects active, population-based surveillance data on laboratory-confirmed infections for nine different pathogens that are commonly transmitted through food (as listed in *section 2.2*). These pathogens include *Campylobacter*, *Listeria*, *Salmonella*, *Shigella*, STEC O157, *Vibrio*, *Yersinia*, *Cryptosporidium*, and *Cyclospora* (the latter two are parasites). This surveillance effort includes 10 state health departments (i.e., California, Colorado, Connecticut, Georgia, Maryland, Minnesota, New Mexico, New York, Oregon, Tennessee), the U.S. Department of Agriculture's Food Safety and Inspection Service (USDA-FSIS), the Food and Drug Administration (FDA), and the CDC. The total surveillance area accounts for approximately 15% of the United States population, representing about 46 million people. The national *Salmonella* database in PulseNet, which is the national molecular subtyping network for foodborne disease surveillance, was established by the CDC to subtype

Table 2. Comorbid conditions associated with nontyphoidal *Salmonella* mortality in the

**4. Current surveillance efforts, prevention, and next steps** 

**4.1 Current surveillance efforts in the United States** 

Comorbid condition

Flu/Pneumonia (organism

Septicemia (including other

United States, 1990-2007.

data.

Diseases of the digestive

rheumatoid arthritis (MOR=2.24; 95% CI, 1.10-4.55) were more likely to be reported on death certificates with *Salmonella* infection listed as an underlying or associated cause of death than controls when matched on age, sex, and race/ethnicity (Table 2). Other conditions found to be listed with *Salmonella* infection on death certificates included: septicemia; various types of renal failure and disorders of fluid, electrolyte, and acid-base balance; and sickle-cell disorders (Table 2). The matched analysis examined comorbid conditions most often listed on death certificates of those who died from *Salmonella* infection, as either an underlying cause or associated cause of death. Table 2 lists these diseases within a broader category (e.g., all types of cancer, all types of renal failure) and their corresponding ICD codes. For example, Leukemia is one type of cancer that affects the bone marrow.


rheumatoid arthritis (MOR=2.24; 95% CI, 1.10-4.55) were more likely to be reported on death certificates with *Salmonella* infection listed as an underlying or associated cause of death than controls when matched on age, sex, and race/ethnicity (Table 2). Other conditions found to be listed with *Salmonella* infection on death certificates included: septicemia; various types of renal failure and disorders of fluid, electrolyte, and acid-base balance; and sickle-cell disorders (Table 2). The matched analysis examined comorbid conditions most often listed on death certificates of those who died from *Salmonella* infection, as either an underlying cause or associated cause of death. Table 2 lists these diseases within a broader category (e.g., all types of cancer, all types of renal failure) and their corresponding ICD codes. For example, Leukemia

> ICD-9 and ICD-10 codes (respectively)

Alcohol and drug abuse 303-305, K70, F10-F19 40 (2.92) 237 (4.32) 0.65

Cancer (all types) 140-239, C00-D48 192 (14.00) 1,373 (25.04) 0.47

200, 203-205, C85, C88,

710, 714, M05-M06, M08, M32-M35




Diabetes 250, E10-E11, E14 109 (7.95) 422 (7.70) 1.04

system 390-459, I00-I99 674 (49.16) 2,938 (53.57) 0.82

electrolyte, acid-base balance 276, E87 57 (4.16) 78(1.42) 3.03

Salmonellarelated deaths (N=1,371),a N(%)b

170-175, C40-C49 13 (0.95) 102 (1.86) 0.50

150-159, C15-C26 33 (2.41) 336 (6.13) 0.37

240-279, E00-E90 242 (17.65) 669 (12.20) 1.56

C90-C92 46 (3.36) 92 (1.68) 2.01

24 (1.75) 43 (0.78) 2.36

Matched control deaths (N=5,484), N(%)b

Matched odds ratios (95%CI)

(0.46-0.92)

(0.40-0.56)

(0.28-0.90)

(0.26-0.54)

(1.41-2.87)

(1.48-5.88)

(1.42-3.93)

(1.72-8.55)

(1.10-4.55)

(0.83-1.30)

(0.73-0.93)

(1.32-1.83)

(2.12-4.31)

is one type of cancer that affects the bone marrow.

Comorbid condition



Cancers affecting bone marrow (all types)

Connective tissue disorders

Diseases of the circulatory


Endocrine, nutritional, metabolic diseases, and immunity disorders (all

skin, breast)

(all types)

types)


a One case was excluded due to missing variables. b Numbers may not add up to total due to missing data.

Table 2. Comorbid conditions associated with nontyphoidal *Salmonella* mortality in the United States, 1990-2007.
