ARTICLE

A Case-Control Study of the Epidemiology of

Sporadic Salmonella Infection in Infants
Timothy F. Jones, MDa, L. Amanda Ingram, MPHa, Kathleen E. Fullerton, MPHb, Ruthanne Marcus, MPHc, Bridget J. Anderson, PhDd,
Patrick V. McCarthy, PhDe, Duc Vugia, MD, MPHf, Beletshachew Shiferaw, MDg, Nicole Haubert, BSh, Stephanie Wedel, MPHi,
Frederick J. Angulo, DVM, PhDb

aTennessee Department of Health, Nashville, Tennessee; bCenters for Disease Control and Prevention, Atlanta, Georgia; cConnecticut Emerging Infections Program,

New Haven, Connecticut; dNew York State Department of Health, Albany, New York; eCenter for Food Safety and Applied Nutrition, Food and Drug Administration,
Washington, DC; fCalifornia Department of Health Services, Berkeley, California; gOregon Department of Human Services, Portland, Oregon; hColorado Department of
Public Health and Environment, Denver, Colorado; iMinnesota Department of Health, Minneapolis, Minnesota

The authors have indicated they have no financial relationships relevant to this article to disclose.

ABSTRACT
OBJECTIVE. Rates of Salmonella infection are highest in infants, but little is known
about potential sources of infection in this high-risk population. We performed a
www.pediatrics.org/cgi/doi/10.1542/
case-control study to identify dietary and environmental risk factors for sporadic peds.2006-1218
salmonellosis among infants. doi:10.1542/peds.2006-1218
PATIENTS AND METHODS. In 2002–2004, the Foodborne Diseases Active Surveillance Key Words
Salmonella, infant, epidemiology
Network conducted a population-based, case-control study of sporadic salmonel-
Abbreviations
losis among infants ⬍1 year of age in 8 states. Cases were identified via active FoodNet—Foodborne Diseases Active
laboratory-based surveillance. Healthy controls were frequency matched by age Surveillance Network
and identified through birth registries or published birth announcements. We OR— odds ratio
CI— confidence interval
assessed diet and environmental exposures in the 5 days before illness onset or
Accepted for publication Jun 8, 2006
interview. Data were analyzed by using logistic regression adjusting for age. Address correspondence to Timothy F. Jones,
MD, Communicable and Environmental
RESULTS. The study enrolled 442 subjects and 928 controls. Compared with healthy Disease Services, Tennessee Department of
controls, infants with Salmonella infection were less likely to have been breastfed Health, 4th Floor, Cordell Hull Building, 425
5th Ave N, Nashville, TN 37247. E-mail: tim.f.
and more likely to have had exposure to reptiles, to have ridden in a shopping cart jones@state.tn.us
next to meat or poultry, or to have consumed concentrated liquid infant formula PEDIATRICS (ISSN Numbers: Print, 0031-4005;
during the 5-day exposure period. Travel outside the United States was associated Online, 1098-4275); published in the public
domain by the American Academy of
with illness in infants 3 to 6 and ⬎6 months of age. Attending day care with a child Pediatrics
with diarrhea was associated with salmonellosis in infants ⬎6 months of age.
CONCLUSIONS. We identified a number of modifiable protective and risk factors for
salmonellosis in infants. Attention should be directed at developing effective
preventive measures for this high-risk population.

2380 JONES et al
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019
S ALMONELLA IS ESTIMATED to cause 1.4 million ill-
nesses and 400 deaths each year in the United
States.1 The incidence of laboratory-confirmed nonty-
phoidal Salmonella infection is much greater in infants
(children ⬍1 year of age) than it is among other age
groups (Fig 1).2 Furthermore, infants can suffer substan-
tial morbidity from invasive Salmonella,3 and infants are
at higher risk of complications from diarrheal disease.4
Much of what is understood about the epidemiology of
this pathogen is derived from investigation of foodborne
outbreaks, which typically have included few infants.
Although there have been dramatic point-source out-
breaks of salmonellosis in infants because of foods, fo-
mites, and nursery exposures,3,5–8 the large majority of
reported cases are not associated with recognized clus-
ters.
Sources of Salmonella in infants may be substantially
different from those in older populations because of
markedly different behaviors and more limited dietary
and environmental exposures. However, aside from oc-
casional case reports,9–13 few epidemiologic studies have
examined potential sources of sporadic Salmonella infec-
tion in infants. Of the limited studies that were reported,
many were performed in Guam, which has had a pro-
tracted epidemic of infant salmonellosis.14–17 Some stud-
ies have demonstrated that infants infected with Salmo-
nella are less likely to be breastfed,16,18 more likely to be
fed high-iron formula,16 and more likely to have a
household member with diarrhea19 than healthy infants.
We performed a large case-control study of sporadic
infant salmonellosis to more completely examine factors
associated with illness in this high-risk population.
METHODS
A case-control study was performed in the Foodborne
Diseases Active Surveillance Network (FoodNet) surveil-
lance area during a 24-month period in 2002–2004, with
variable start dates among sites. FoodNet is a collabora-
tive project involving the Centers for Disease Control
and Prevention’s Emerging Infections Program, the US
FIGURE 1
Incidence (cases per 100 000 population) of non–Typhi
Salmonella infection in FoodNet, 1996 –2004, according to
age group. a Infant ages are shown in months, older per-
sons’ ages are shown in years.
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019
Food and Drug Administration, the US Department of
Agriculture Food Safety Inspection Service, and selected
state health departments. During the study period, pop-
ulation-based active surveillance at all clinical laborato-
ries for laboratory-confirmed Salmonella infections was
performed in 8 sites (Connecticut, Minnesota, Oregon,
and selected counties in California, Colorado, Georgia,
New York, and Tennessee) covering a population of
⬃35.2 million persons (12.1% of the US population),
including ⬃450 000 infants.20 This study was approved
by the institutional review boards of the Centers for
Disease Control and Prevention and all participating
sites.
All infants (children ⬍1 year of age) residing within
the catchment area during the study period were eligible
for the study. A case was defined as a Salmonella infec-
tion in an infant from whom any serotype of Salmonella
other than Typhi was isolated from a clinical specimen,
with onset of disease or collection of the positive speci-
men occurring before their first birthday. Subjects were
excluded if the family could not be contacted after at
least 15 attempts (including evenings and weekends) or
within 45 days of the specimen collection date, if the
caretaker refused or did not speak English or Spanish (or
Mandarin or Cantonese in California), or if the infant
was asymptomatic or a chronic carrier (defined as per-
sistent positive stool cultures ⬎30 days after an acute
illness). Subjects were not eligible if the illness was part
of an outbreak investigated by public health officials and
in which the vehicle of transmission was identified.
Controls were frequency matched by 1-month age
categories on the basis of the incidence of laboratory-
confirmed infant Salmonella and Campylobacter infection
in each site in the preceding 3 years. Controls were
selected randomly from all births in the catchment area
reported in state birth registries. For part of the study
period, the New York site used published birth an-
nouncements to identify controls. Controls were ex-
cluded if they could not be contacted after at least 15
attempts (including evenings and weekends), if their
PEDIATRICS Volume 118, Number 6, December 2006 2381
caretaker did not speak a language included in the study,
or if they reported culture-confirmed Salmonella infec-
tion within the previous 30 days. A study of infant
Campylobacter infection was performed concurrently in
participating sites with identical questionnaires, and the
same controls were used for both studies.21 Some sites
offered $10 retail store gift certificates or checks to sub-
jects and controls after completing the survey.
After obtaining informed consent, an extensive ques-
tionnaire was administered by telephone to the parents
or guardians of the subjects and controls. Questionnaires
were administered in English, Spanish, Mandarin, or
Cantonese (the latter 2 in California only) to collect
information on demographic characteristics, general
health status and medications, symptoms of illness, food
exposures, and potential social and environmental risk
factors. Extensive information was collected on expo-
sures including water source, formula types, various
foods and manner of preparation, previous antibiotic
use, and animal exposures. Subject caregivers were in-
terviewed regarding exposures in the 5 days preceding
onset of illness. Control caregivers were interviewed
regarding the 5 days before interview. Because of the
time lag in the availability of birth registries, information
for controls with a target age of 0 to 2 months was
gathered for a 5-day period preceding a specified date
within the first 2 months of age.
FIGURE 2
Status of enrolled and nonenrolled case patients in the
FoodNet infant Salmonella case-control study.
2382 JONES et al
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019
Data were entered into an electronic database at each
participating site. Analyses were performed by using Epi
Info 3.3.2 (Centers for Disease Control and Prevention,
Atlanta, GA) and SAS 9.1 (SAS Institute, Inc, Cary, NC).
Unconditional logistic regression models were adjusted
for age of the case patient and education of the care-
giver. Variables included in the model were those hy-
pothesized a priori as potential risk factors and those that
were significant on univariate analysis.
RESULTS
During the study period, 7194 cases of non-Typhi Sal-
monella infection were ascertained in the surveillance
population, of which 677 (9.4%) occurred in infants. Of
the infant cases, 15 (2.2%) were not eligible for inclu-
sion because they were associated with recognized out-
breaks or were identified in chronic carriers (Fig 2). Of
the 10 outbreak cases, 5 were associated with infant
chick exposure, 1 with reptiles, 1 with honeydew melons,
and 3 with day care centers. No subjects were excluded
because of being asymptomatic. Of the eligible subjects,
442 (67%) were enrolled. Among the 217 nonenrolled
subjects, the most common reasons included being un-
able to contact the family (61%), refusal (19%), and not
speaking a study language (10%). Nonenrolled subjects
did not differ significantly from enrolled subjects with
respect to age, gender, month of illness, type of specimen
from which Salmonella was isolated, or outcome (data TABLE 2 Characterization of Illness of Infants With Salmonella
not shown). A total of 928 controls were enrolled in the Infection Enrolled in the Case-Control Study
study (Table 1). Of those enrolled, 28 (6%) subject and N ⫽ 442
14 (2%) control interviews were performed in Spanish,
No. %
with 6 (1%) subject and 2 (0.2%) control interviews
Diarrhea 432 98
completed in Cantonese. Fever 323 73
Diarrhea and fever were commonly identified symp- Vomiting 148 33
toms among infants with salmonellosis (Table 2). The Mean (median) duration of diarrhea, d 10 7
median duration of diarrhea was 7 days (range: 1–31 Salmonella isolated from normally sterile site 33 7
days). Half of the case patients were treated with anti- (eg, blood cerebrospinal fluid)
Treated with antibiotics 231 52
biotics, 30% received intravenous hydration, and one Received intravenous hydration 132 30
quarter were hospitalized for a median of 3 days (range: Admitted to the hospital 112 25
1–38 days). Fatal infections were reported in 2 (0.5%) Mean duration of hospital stay, median, d 5 3
case patients. Among case patients, 405 (92%) had Sal- Mean (median) workdays missed by family 4 3
monella isolated from stool specimens, 32 (7%) from member due to infant’s illness
Household member reported diarrhea in 5 d 56 13
blood, 1 (0.2%) from urine, 1 (0.2%) from cerebrospinal prior to infant’s illness
fluid, and 3 from an unknown site. Infant illness was Household member reported diarrhea within 91 21
associated with a substantial burden on families, result- 10 d after infant’s illness
ing in a median of 3 days (range: 0 –21 days) of parental Received antibiotics during 4 wk before this 33 7
or guardian work absence; 21% of households reported illness
Received antacids during 4 wk before this illness 21 5
another person with diarrhea within 10 days after the
Missing data are excluded.
infants’ illness.
In multivariable regression analysis, case infants were
significantly more likely than controls to have traveled
outside the United States (odds ratio [OR]: 8.9; 95% meat or poultry (OR: 3.2; 95% CI: 2.1–5.1), or to have
confidence interval [CI]: 2.6 –31.0), to have attended had exposure to reptiles (OR: 5.2; 95% CI: 3.4 –7.9). The
day care with a child with diarrhea (OR: 4.4; 95% CI: reptiles to which case infants were most commonly ex-
1.8 –10.7), to have ridden in a shopping cart next to posed included iguanas (25), lizards (25), snakes (20),

TABLE 1 Demographic Characteristics of Enrolled Cases and Controls

Cases (N ⫽ 442) Controls (N ⫽ 928)
No. % No. %
Age, mean (median), mo 6 6 6 6
Male 229 52 452 49
Race known 404 891
White 255 63 732 82
Black 108 27 96 11
Asian 26 6 43 5
Other 15 4 20 2
Hispanic 92 23 97 11
Born outside US 4 1 0 0
Live in household with another child ⬍5 y old 283 64 594 64
Caregiver educational level
Less than high school 216 49 229 25
More than high school 225 51 696 75
Residence
Urban/suburban 355 80 726 78
Town/rural 75 17 180 19
Farm 9 2 19 2
Medical insurance
Medicaid 170 38 147 16
Other insurance 269 61 755 81
No insurance 4 1 9 1
Annual household income known 381 847
Less than $30 000 210 55 201 24
More than $30 000 171 45 646 76
Discharged from hospital ⬎5 d after birth 38 9 93 10
Underlying immunosuppressive condition 19 4 24 3
Missing data are excluded.

PEDIATRICS Volume 118, Number 6, December 2006 2383

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019
and turtles or tortoises (14). Infection was also associated
with consumption of meat and concentrated liquid in-
fant formula during the 5-day exposure period (Table 3).
Salmonellosis was not associated with consumption of
ready-to-drink liquid formula or powdered infant for-
mula (data not shown). Breastfeeding was protective;
case patients were less likely to have been breastfed
during the 5-day exposure period than controls (OR: 0.5;
95% CI: 0.3– 0.6). Infant mobility (crawling or walking)
and preparation of raw meats within the home in the
previous 5 days were also associated with a lower risk of
illness.
Logistic regression models controlled for infant age by
month and parental education status. Additional analy-
ses controlling for insurance status and family income
did not significantly change the results. Exclusion of data
from the New York site (the only site which temporarily
used birth announcements to identify controls) did not
affect any results significantly.
To better understand potential risk factors and their
associated population-attributable fraction, the multiva-
riable regression analysis was repeated with cases and
controls categorized by age into 3 groups: ⬍3, 3 to 6, and
⬎6 months of age (Table 3). Previous antacid and anti-
biotic use were included in the age-specific models but
not in the overall model because of complex interactions
of this variable with age. Illness was associated with
reptile exposure and riding in a shopping cart next to
meat or poultry in all age groups. Recent antacid use was
a significant risk factor in infants ⬍3 months of age.
Illness was associated with travel outside the United
States in infants ⱖ3 months of age. Attending day care
with a child with diarrhea was associated with illness
among infants ⬎6 months of age, and recent antibiotic
use was associated with decreased risk in infants in this
group. Illness was associated with consumption of con-
centrated infant formula in infants 3 to 6 months of age
and was inversely associated with breastfeeding among
those ⱕ6 months of age. No additional risk factors were
identified among the subsets of enrollees speaking Span-
ish, Mandarin, or Cantonese when analyzed separately.
Among case patients, the most common Salmonella
serotypes included Salmonella typhimurium (20%), Salmo-
nella newport (9%), Salmonella heidelberg (6%), Salmonella
enteritidis (5%), and Salmonella montevideo (3%). These
proportions are similar to the distribution of serotypes
within FoodNet among case patients ⱖ1 year of age
during the same period, except that in the older popu-
lation, S enteritidis accounts for 15% of cases. The results
of multivariable regression analyses restricted to cases
with the most common specific serotypes are shown in
Table 4. Analyses of less common serotypes could not be
performed because of low numbers.
DISCUSSION
Laboratory-confirmed salmonellosis in infants is associ-
ated with a number of potentially modifiable risk factors.
Not unexpectedly, potential sources in this population
are remarkably different from those identified among
older persons, primarily because of their substantially
different dietary and environmental exposures. Consis-
tent with other studies of Salmonella and many other
enteric pathogens in the general population, illness
among infants was significantly associated with interna-
tional travel, but this exposure accounts for only a small
proportion (3%) of cases in this population.
Breastfeeding was protective against salmonellosis
among infants, particularly infants who were 6 months
of age and younger, in whom other dietary exposures
are limited and breastfeeding is more common. Other
studies have also demonstrated this protective ef-
fect,16,18,22 and this finding is consistent with the large

TABLE 3 Risk Factors for Salmonella Infection Within the 5 Days Before Illness Onset by Age Group-on Multivariable Regression Analysis
Age Group All Infants
⬍3 mo ⬎6 mo 3–6 mo
OR (95% CI) PAF, % OR (95% CI) PAF, % OR (95% CI) PAF, % OR (95% CI) PAF, %
Infant could crawl or walk — — 0.4 (0.3–0.8) — 0.8 (0.4–1.5) — 0.5 (0.4–0.8) —
Attended day care with another child with diarrhea — — 3.4 (0.8–14.9) 3.1 6.0 (1.9–19.0) 5.6 4.4 (1.8–10.7) 3.3
Travel outside US 2.4 (0.1–56.1) 0.6 12.1 (1.2–118.3) 2.7 18.7 (1.9–182.1) 3.7 8.9 (2.6–30.9) 2.6
Reptile exposure 6.1 (2.0–18.4) 18.5 5.4 (2.9–10.3) 21.2 3.6 (1.8–7.2) 12.2 5.2 (3.4–7.9) 17.4
Farm animal contact — — 1.4 (0.1–17.7) 0.2 2.5 (0.4–17.1) 0.7 2.1 (0.5–9.3) 0.4
Breastfed 0.2 (0.1–0.5) — 0.6 (0.3–0.9) — 0.6 (0.4–1.1) — 0.5 (0.3–0.6) —
Consumed concentrated infant formula 2.2 (1.0–5.0) 15.1 2.2 (1.2–3.9) 14.3 1.8 (1.0–3.2) 11.1 2.0 (1.4–2.8) 13
Consumed partially cooked egg — — — — 1.8 (0.5–5.9) 2.5 2.7 (0.9–7.6) 2
Consumed any meat — — 1.4 (0.4–5.2) 0.9 1.6 (1.0–2.5) 17.8 1.7 (1.1–2.7) 8.3
Any meat or poultry prepared in home 0.5 (0.2–1.0) — 0.6 (0.3–1.1) — 0.5 (0.3–0.8) — 0.5 (0.4–0.7) —
Rode in shopping cart next to meat or poultry 6.0 (1.4–25.6) 8.4 3.4 (1.4–8.0) 9.3 2.8 (1.5–5.1) 14.7 3.2 (2.1–5.1) 11.3
Took antibiotics in previous 4 wk — — 1.3 (0.6–3.0) 2.1 0.4 (0.2–0.8) — — —
Took antacids in previous 4 wk 5.9 (1.0–34.4) 6.1 2.5 (1.0–6.7) 4.3 0.4 (0.1–2.8) — — —
PAF indicates population-attributable fraction. Missing values affected the numbers of observations fit in each model. Number of infants in each age group: ⬍3 months, 86 cases, 176 controls; 3– 6
months, 146 cases, 310 controls; ⬎6 months, 151 cases, 3 controls; all infants, 385 cases, 837 controls.

2384 JONES et al
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019
 TABLE 4 Exposures Associated With Infection With the Most tion of 11%. To our knowledge, no previous studies
Common Salmonella Serotypes Among Infants: have examined this risk factor specifically. A recent
Multivariable Regression Analysis study from the United Kingdom demonstrated substan-
Salmonella Serotype OR 95% CI tial levels of contamination with pathogens on the out-
Typhimurium (n ⫽ 84) side of packaged meats and poultry.26 Our findings sug-
Exposure gest that attention should be directed at better
Reptiles 3.1 1.4–6.5 understanding and preventing this potential route of
Concentrated formula 2.6 1.5–4.6 infection. This commonly reported exposure might be
Consumption of meat 2.9 1.4–5.9
addressed with nonleaking packaging and separation of
Breastfeeding 0.4 0.2–0.7
Infant mobility 0.3 0.2–0.7 infants from high-risk foods in the shopping cart. Mo-
Meat preparation in home 0.5 0.3–1.0 bility might also be expected to lead to the potential for
Newport (n ⫽ 38) more exposure to environmental contamination. Al-
Exposure though we found that illness among 3 to 6 month olds
Travel outside US 24.0 2.6–221.2
was inversely associated with increased mobility of the
Breastfeeding 0.2 0.1–0.5
Meat preparation in home 4.0 0.2–0.7 infant (crawling or walking), few patients in this age
Enteriditis (n ⫽ 22) group are mobile. Illness was associated with environ-
Exposure mental exposure in studies in Guam, where Salmonella
Riding in shopping cart next to meat 3.4 1.0–11.3 contamination in the environment may be more com-
Heidelberg (n ⫽ 19)
mon.17 Environmental contamination may be less likely
Exposure
Contact with farm animals 20.8 1.5–288.8 in the United States in the absence of exposures to
Concentrated formula 4.8 1.5–14.9 reptiles, farm animals, or other identifiable animal sources.
Montevideo (n ⫽ 11) Although previous studies of infant salmonellosis have
Exposure included culturing of specimens from the household
Consumption of partially cooked eggs 161.9 6.6-undefined
environment,27 which can provide additional informa-
Consumption of meat 21.1 2.6–169.1
tion about potential mechanisms of transmission, this
was not performed in this study.
The inverse association of illness with recent prepa-
body of data suggesting that breastfeeding decreases the ration of meat or poultry in the home (independent of
risk of a large array of diseases in infants.21,23 In our consumption of meat or poultry by the infant) was also
study, salmonellosis was associated with consumption of unexpected. An inverse association between illness and
concentrated liquid infant formula in the 5-day expo- diet diversity was noted in a study of sporadic S enteriditis
sure period, even when controlling for breastfeeding. disease in the general population,28 and it is possible that
Among infants ⬍3 months of age, this was the second our finding is somehow correlated with dietary diversity
most important risk factor for disease (population attrib- in the family in a manner we could not identify further.
utable fraction: 15.1). In contrast, salmonellosis was not Ongoing work in FoodNet is directed at better under-
associated with consumption of ready-to-drink liquid standing the potential implications of dietary diversity
infant formula or powdered infant formula. Powdered on foodborne illness in all populations.29
infant formula is not a sterile product and has been Salmonella infection in infants is clearly a substantial
associated with outbreaks of Salmonella and other patho- burden to the patients and their affected families. Rou-
gens.7,24,25 Concentrated liquid formula is sterile, but tine treatment of Salmonella infection with antimicrobial
added water and manipulation of the food during prep- agents in low-risk populations is not recommended, al-
aration are potential sources of contamination. How- though infants may be at increased risk for severe dis-
ever, illness was not associated independently with con- ease compared with older persons.3,4 It is interesting,
sumption of water from various sources or with therefore, that the proportion of infants in this study
powdered formula, to which water is also added. Details receiving antibiotics was less than the proportion of case
of formula preparation were not examined in this study, patients ⱖ1 year of age who took antibiotics for S newport
although it is possible that differences in storage and infections (67%) or S enteritidis infections (70%) in re-
handling of opened cans of concentrated formula may be cent studies in FoodNet sites.30 Attention should be de-
associated with illness. Additional studies may further voted to better understanding determinants of antimi-
elucidate the specific risks associated with concentrated crobial use and encouraging adherence to national
formula. Of note, salmonellosis was not associated with guidelines for diagnosis and treatment of bacterial gas-
consumption of iron-containing infant formula, in con- troenteritis.31 Subsequent diarrhea was reported in other
trast to the results of a previous study.16 household members in more than one fifth of our cases.
Illness in all infant age groups was associated with Although these illnesses were not confirmed to be sal-
infants riding in a shopping cart with meat or poultry monellosis, this rate is substantially higher than the rate
placed next to them, with a population-attributable frac- of diarrhea in the general population,32 suggesting that

PEDIATRICS Volume 118, Number 6, December 2006 2385

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019
intrafamilial spread after infant illness may be com-
mon.33 Likewise, 13% of infant case patients with Sal-
monella had illness onset within 5 days after diarrhea in
another household member, suggesting that person-to-
person spread may be an important source of infant
infection. Unfortunately, data on diarrhea in household
members was not collected for controls, so we were
unable to quantify the magnitude of this risk. Further-
more, household members with diarrhea before or after
the infant case were not cultured as part of this study,
thus limiting conclusions that can be made about house-
hold transmission. Likewise, we did not have culture
confirmation of illness among other day care attendees
to assess person-to-person transmission in that setting.
That Salmonella infection in children is associated with
reptile exposure was well documented in several previ-
ous studies,34,35 although none have studied infants spe-
cifically. In this study, reptile exposure had the highest
population-attributable fraction of any of the risk factors
identified: 17% for all infants. Although it is possible
that some older infants could have had direct contact
with reptiles within their homes, it is more likely that
most infants were infected indirectly either through con-
tact with the reptile-contaminated environment or via
contaminated hands of caretakers. The Centers for Dis-
ease Control and Prevention has recommended that
households with children ⬍5 years of age should not
own reptiles.34
In the general population, a wide variety of potential
sources of Salmonella infection have been identified, with
substantial variation among serotypes. Most of these
potential sources involve consumption of specific food
items. Because we suspected that foodborne transmis-
sion is less commonly a factor in infants than it is in older
populations, we analyzed data for all serotypes com-
bined. As expected, the primary risk factors identified
were associated with environmental exposures and not
with specific food products. Analysis of potential sources
associated with specific Salmonella serotypes was limited
somewhat by low numbers of cases, particularly for less
common serotypes. Because the serotype-specific anal-
yses involved multiple comparisons without adjustment
for multiple testing, the associated uncertainties may
be underrepresented. Despite this, the findings are con-
sistent with our initial assumptions. For example, Salmo-
nella infection in infants was not associated with consump-
tion of eggs and poultry, which are established sources
of S enteriditis in older populations; however, S enteriditis
infection in infants was associated with exposure to
meat and poultry in shopping carts.
Subjects and controls differed with regard to race,
caregiver educational status, medical insurance cover-
age, and household income, each of which may correlate
with socioeconomic status. We included caregiver edu-
cational status in the final model as a proxy for socio-
economic status. Inclusion of additional factors such as
2386 JONES et al
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019
income and insurance status led to loss of data (because
of missing information) and did not change results sig-
nificantly. Additional understanding of the risk factors
associated with sociodemographic status will be an im-
portant part of further controlling this disease among
infants.
To our knowledge, this is the largest reported popu-
lation-based case-control study of infant salmonellosis.
In contrast to older populations with diverse dietary
exposures, potential sources among infants reflect their
more limited diets and the relatively greater contribution
of environmental exposures to disease transmission.
Nevertheless, contamination of the environment can oc-
cur, in part, through sources of Salmonella (such as poul-
try and reptiles) that are common among older popula-
tions. Many of the risk factors identified in this study are
potentially modifiable through targeted preventive edu-
cation and behavioral change among the caretakers of
infants. Attention should be directed at developing prac-
tical and effective measures to prevent Salmonella infec-
tion in this high-risk population.
REFERENCES
1. Voetsch AC, Van Gilder T, Angulo FJ, et al. FoodNet estimate of
the burden of illness caused by nontyphoidal Salmonella infec-
tions in the United States. Clin Infect Dis. 2004;38(suppl 3):
S127–S134
2. Centers for Disease Control and Prevention. FoodNet Annual
Report, 2003. Atlanta, GA: Centers for Disease Control and
Prevention; 2005
3. Cleary TG. Salmonella. In: Feigin RD, Cherry JD, Demmler GJ,
Kaplan SL, eds. Textbook of Pediatric Infectious Diseases. 5th ed.
Philadelphia, PA: WB Saunders Co; 2004:1473–1487
4. Cohen MB. Etiology and mechanisms of acute infectious diar-
rhea in infants in the United States. J Pediatr. 1991;118(4 pt
2):S34 –S39
5. Rushdy AA, Stuart JM, Ward LR, et al. National outbreak of
Salmonella senftenberg associated with infant food. Epidemiol
Infect. 1998;120:125–128
6. International Investigation Collaborating Units. Preliminary re-
port of an international outbreak of Salmonella anatum infec-
tion linked to an infant formula milk. Euro Surveill. 1997;2:
22–24
7. Usera MA, Echeita A, Aladuena A, et al. Interregional food-
borne salmonellosis outbreak due to powdered infant formula
contaminated with lactose-fermenting Salmonella virchow. Eur J
Epidemiol. 1996;12:377–381
8. Rowe B, Begg NT, Hutchinson DN, et al. Salmonella ealing
infections associated with consumption of infant dried milk.
Lancet. 1987;2(8564):900 –903
9. Sato Y, Mori T, Koyama T, Nagase H. Salmonella virchow
infection in an infant transmitted by household dogs. J Vet Med
Sci. 2000;62:767–769
10. Cyriac J, Wozniak ER. Infantile Salmonella meningitis associ-
ated with gecko-keeping. Commun Dis Public Health. 2000;3:
66 – 67
11. Qutaishat SS, Stemper ME, Spencer SK, et al. Transmission of
Salmonella enterica serotype typhimurium DT104 to infants
through mother’s breast milk. Pediatrics. 2003;111:1442–1446
12. Cartolano GL, Moulies ME, Sequier JC, Boisivon A. A parent as
a vector of Salmonella brandenburg nosocomial infection in a
 neonatal intensive care unit. Clin Microbiol Infect. 2003;9:
560 –562
13. Buck JJ, Nicholls SW. Salmonella arizona enterocolitis acquired
by an infant from a pet snake [letter]. J Pediatr Gastroenterol
Nutr. 1997;25:248 –249
14. Haddock RL. The origins of infant salmonellosis. Am J Public
Health. 1993;83:772
15. Haddock RL, Nocon FA. Infant salmonellosis and vacuum
cleaners. J Trop Pediatr. 1994;40:53–54
16. Haddock RL, Cousens MA, Guzman CC. Infant diet and sal-
monellosis. Am J Public Health. 1991;81:997–1000
17. Haddock RL, Malilay J. A search for infant salmonellosis risk
factors on Guam. Southeast J Trop Med Public Health. 1986;17:
38 – 42
18. France GL, Marmer DJ, Steele RW. Breast feeding and Salmo-
nella infection. Am J Dis Child. 1980;134:147–152
19. Delarocque-Astagneau E, Desenclos JC, Bouvet P, Grimont
PAD. Risk factors for the occurrence of sporadic Salmonella
enterica serotype enteriditis infections in children in France: a
national case-control study. Epidemiol Infect. 1998;121:561–567
20. Allos BM, Moore MR, Griffin PM, Tauxe RV. Surveillance for
sporadic foodborne disease in the 21st century: the FoodNet
perspective. Clin Infect Dis. 2004;38:S115–S120
21. Fullerton K, Ingram LA, Jones TF, et al. Sporadic Campylobacter
infection in infants, a FoodNet case-control study. Pediatr Infect
Dis J. In press
22. Rowe SY, Rocourt JR, Shiferaw B, et al. Breast feeding de-
creases risk of sporadic salmonellosis among infants in FoodNet
sites. Clin Infect Dis. 2004;38(suppl 3):S262–S270
23. Golding J, Emmett PM, Rogers IS. Gastroenteritis, diarrhoea
and breast feeding. Early Hum Dev. 1997;49(suppl):S83–S103
24. Centers for Disease Control and Prevention. Enterobacter saka-
zakii infections associated with the use of powdered infant
formula: Tennessee, 2001. MMWR Morb Mortal Wkly Rep. 2002;
51:297–300
25. Thurm V, Gericke B. Identification of infant food as a vehicle in
a nosocomial outbreak of Citrobacter freundii: epidemiological
Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019
subtyping by alloenzyme, whole-cell protein and antibiotic
resistance. J Appl Bacteriol. 1994;76:553–558
26. Burgess F, Little CL, Allen G, Williamson K, Mitchell RT.
Prevalence of Campylobacter, Salmonella, and Escherichia coli on
the external packaging of raw meat. J Food Prot. 2005;68:
469 – 475
27. Schutze GE, Sikes JD, Stefanova R, Cave MD. The home en-
vironment and salmonellosis in children. Pediatrics. 1999;103(1).
Available at: www.pediatrics.org/cgi/content/full/103/1/e1
28. Kimura AC, Reddy V, Marcus R, et al. Chicken consumption is
a newly identified risk factor for sporadic Salmonella enterica
serotype Enteritidis infections in the United States: a case-
control study in FoodNet sites. Clin Infect Dis. 2004;38(suppl
3):S244 –S252
29. Hardnett FP, Hoekstra RM, Kennedy M, Charles L, Angulo FJ.
Epidemiologic issues in study design and data analysis related to
FoodNet activities. Clin Infect Dis. 2004;38(suppl 3):S121–S126
30. Devasia RA, Varma JK, Whichard J, et al. Antimicrobial use
and outcomes in patients with multidrug-resistant and pansus-
ceptible Salmonella Newport infections. Microbial Drug Resist.
2005;11:356 –362
31. Guerrant RL, Van Gilder TJ, Steiner TS, et al. Practice guide-
lines for the management of infectious diarrhea. Clin Infect Dis.
2001;32:331–350
32. Jones TF, McMillian MB, Scallan E, et al. A population-based
estimate of the substantial burden of diarrheal disease in the
United States: FoodNet, 1996 –2003. Epidemiol Infect. In press
33. Wilson R, Feldman RA, Davis J, LaVenture M. Salmonellosis in
infants: the importance of intrafamilial transmission. Pediatrics.
1982;69:436 – 438
34. Centers for Disease Control and Prevention. Reptile-associated
salmonellosis, selected states, 1998 –2002. MMWR Morb Mortal
Wkly Rep. 2003;52:1206 –1209
35. Mermin J, Hutwagner L, Vugia D, et al. Reptiles, amphibians,
and human Salmonella infection: a population-based case-
control study. Clin Infect Dis. 2004;38(suppl 3):S253–S261
PEDIATRICS Volume 118, Number 6, December 2006 2387
A Case-Control Study of the Epidemiology of Sporadic Salmonella Infection in
Infants
Timothy F. Jones, L. Amanda Ingram, Kathleen E. Fullerton, Ruthanne Marcus,
Bridget J. Anderson, Patrick V. McCarthy, Duc Vugia, Beletshachew Shiferaw,
Nicole Haubert, Stephanie Wedel and Frederick J. Angulo
Pediatrics 2006;118;2380
DOI: 10.1542/peds.2006-1218

Updated Information & including high resolution figures, can be found at:
Services http://pediatrics.aappublications.org/content/118/6/2380
References This article cites 29 articles, 2 of which you can access for free at:
http://pediatrics.aappublications.org/content/118/6/2380#BIBL
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Infectious Disease
http://www.aappublications.org/cgi/collection/infectious_diseases_su
b
Permissions & Licensing Information about reproducing this article in parts (figures, tables) or
in its entirety can be found online at:
http://www.aappublications.org/site/misc/Permissions.xhtml
Reprints Information about ordering reprints can be found online:
http://www.aappublications.org/site/misc/reprints.xhtml

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019

A Case-Control Study of the Epidemiology of Sporadic Salmonella Infection in
Infants
Timothy F. Jones, L. Amanda Ingram, Kathleen E. Fullerton, Ruthanne Marcus,
Bridget J. Anderson, Patrick V. McCarthy, Duc Vugia, Beletshachew Shiferaw,
Nicole Haubert, Stephanie Wedel and Frederick J. Angulo
Pediatrics 2006;118;2380
DOI: 10.1542/peds.2006-1218

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pediatrics.aappublications.org/content/118/6/2380

Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it
has been published continuously since 1948. Pediatrics is owned, published, and trademarked by
the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,
60007. Copyright © 2006 by the American Academy of Pediatrics. All rights reserved. Print ISSN:
1073-0397.

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 18, 2019