Economic Burden of Foodborne Illnesses Acquired in the United States

Abstract

Foodborne disease burden estimates inform public health priorities and can help the public understand disease impact. This article provides new estimates of the cost of U.S. foodborne illness. Our research updated disease modeling underlying these cost estimates with a focus on enhancing chronic sequelae modeling and enhancing uncertainty modeling. Our cost estimates were based on U.S. Centers for Disease Control and Prevention estimates of the numbers of foodborne illnesses, hospitalizations, and deaths caused by 31 known foodborne pathogens and unspecified foodborne agents. We augmented these estimates of illnesses, hospitalizations, and deaths with more detailed modeling of health outcomes, including chronic sequelae. For health outcomes, we relied on U.S. data and research where possible, supplemented by the use of non-U.S. research where necessary and scientifically appropriate. Cost estimates were developed from large insurance or hospital charge databases, public data sources, and existing literature and were adjusted to 2023 dollars. We estimated the cost of foodborne illness in the United States circa 2023 to be $75 billion. Deaths accounted for 56% and chronic outcomes for 31% of the mean cost. The costliest pathogen was nontyphoidal Salmonella at $17.1 billion followed by Campylobacter at $11.3 billion. Toxoplasma ($5.7 billion) and Listeria ($4 billion) followed due primarily to deaths and chronic outcomes from pregnancy-associated cases. Per-case cost ranged from $196 for Bacillus cereus to $4.6 million for Vibrio vulnificus. Unspecified agents accounted for 38% of the total cost of foodborne illness, but these illnesses were generally mild (per-case cost $781). These cost estimates can help inform food safety priorities. Our pathogen-specific per-case cost estimates can also help inform benefit–cost analysis required for new federal food safety regulations.

Introduction

This article provides new estimates of the cost of U.S. foodborne illness. Disease burden estimates inform public health priorities and can help the public understand disease impact. Like other summary measures of health, cost-of-illness estimates provide an aggregate measure of the impact of a disease across outcomes and can be used to compare the impact of diseases with diverse health outcomes. In addition, cost-of-foodborne illness estimates inform food safety program priorities and budgets, and the cost–benefit analyses needed to issue new regulations. Cost-of-illness estimates also provide the public with information on disease impact in familiar terms.

The most recent prior cost of foodborne illness estimates published by the U.S. Department of Agriculture (USDA), Economic Research Service (ERS) included 15 pathogens with an estimated annual cost of $17.6 billion ($2018) (Hoffmann and Ahn, 2021). Our study estimated the cost of foodborne disease caused by 31 major known pathogens and unspecified agents whose incidence has been estimated by the Centers for Disease Control and Prevention (CDC) (Scallan et al. 2011a; Scallan et al., 2011b) (forthwith Scallan et al., 2011, unless otherwise specified). This expanded coverage from 9 million cases of foodborne illness to 48 million (Hoffmann et al., 2012). Our research also updated disease modeling underlying these cost estimates with a focus on enhancing chronic sequelae modeling and uncertainty modeling.

Economics views cost-of-illness as an approximation of individuals’ willingness-to pay (WTP) to avoid an illness, the theoretically correct measure of the value people place on efforts to prevent illness (Johansson, 1991). Formally, WTP is the sum of the cost of medical treatment, the value of time lost to normal activities (“lost productivity”), plus WTP to reduce risk of pain and suffering from the illness and WTP to reduce risk of death, offset by individuals’ personal expenditures on avoiding illness (Harrington and Portney, 1987). Like prior ERS estimates, we approximated an individuals’ WTP to avoid foodborne illness as the sum of the cost of medical treatment, lost productivity, and WTP to reduce risk of deaths. We omit pain and suffering and individual expenditures on avoidance due to lack of research and/or data.

Materials and Methods

Cost-of-illness research requires both augmenting estimates of illnesses, hospitalizations, and deaths with more detailed modeling of health outcomes and estimation of the cost of those outcomes. We sought to keep the structure of our disease modeling consistent with prior ERS modeling (Hoffmann et al., 2012; Batz et al., 2014). Our estimates use CDC estimates of foodborne illnesses, hospitalizations, and deaths from major pathogens and unspecified agents (Scallan et al., 2011). For modeling of other health outcomes, we relied on U.S. data and research where possible, supplemented by the use of non-U.S. research where necessary and scientifically appropriate (Supplementary Appendix SA2). Cost estimates are reported in $2023. For outpatient and hospital costs estimated from primary data on charges, cost estimates are the average of costs for 2016–2019 updated for inflation to $2023. As discussed below, this time period choice was driven by changes in ICD coding and the COVID-19 pandemic. Costs for chronic sequelae and WTP to reduce risk of premature death rely on estimates from prior primary research updated for inflation to $2023.

All estimates were based on deidentified public data or previously published research. No interaction or intervention with human subjects occurred, and no personally identifiable information was used, collected, or transmitted. This analysis did not meet the definition of human subjects research (as defined in the U.S. Code of Federal Regulations, Title 45 Part 46) and was not subject to review by an institutional review board.

Disease-outcome modeling

For most pathogens and unspecified acute gastrointestinal illness (AGI), our disease outcome trees included the following acute health states: “no physician visit, recovered”; “physician visit, recovered, no hospitalization” (including emergency department visits not resulting in hospitalization); “hospitalized, recovered”; and “hospitalized, died” (Supplementary Appendix SA1, Fig. 1). As did prior ERS estimates, we assumed that hospitalizations were preceded by a physician visit and that deaths were preceded by a hospitalization. Prior ERS estimates included separate disease outcome tree branches for hospitalizations with and without complications. We have only one branch for total hospitalizations in our disease outcome trees but do provide “side estimates” for major complications. Neither our estimates nor prior ERS estimates incorporated chronic sequelae into the disease outcome trees due to lack of information on the health outcome that preceded the sequelae (Supplementary Appendix SA1, Supplementary Fig. SA1.1, and Supplementary Table SA 1.1). Different disease outcome model structures were used for pregnancy-associated Listeria monocytogenes and Toxoplasma gondii and Mycobacterium bovis due to their distinct disease progressions (Supplementary Appendix SA2).

Acute health outcomes

Scallan et al. (2011) modeled hospitalizations and deaths independently, while we assumed that all deaths were preceded by hospitalizations. As a result, we simulated the number of hospitalized cases who died (with uncertainty bounds) using a nested transition probability (i.e., the proportion of hospitalized cases who died from Scallan et al., 2011) applied to the number of hospitalized cases from Scallan et al. (2011). We derived the number of cases seeking any medical care (outpatient or inpatient) from data provided in the appendices in Scallan et al. (2011b) (Supplementary Appendix SA3). Throughout this article, we assume that illnesses caused by unspecified agents are milder than those with pathogen-specific diagnoses as doctors do not generally order the laboratory tests needed to establish the pathogen-specific diagnosis when an illness is mild. We assumed that cases caused by unspecified agents sought care at the same rate as norovirus. Cases recovering without seeking medical care were calculated as the number of illnesses less those cases seeking medical care. The number of cases only visiting a physician (no hospitalization) was calculated as the number seeking care less the number hospitalized. Ahn et al. (2022) provide our estimates of the pathogen-specific rates and costs of cases hospitalized with sepsis. There we estimate that foodborne illnesses result in 43,000 sepsis cases annually at a cost of $1 billion $2023 (Ahn et al., 2022). Decisions to include other complications and estimation of their rates were based on an expert workshop, literature review, and individual expert reviews of our final models (Supplementary Appendix SA1, Supplementary Table SA1.1) (Hoffmann and Scallan Walter, 2020).

Duration of acute illnesses

Sources used to estimate the duration of illness varied by pathogen and severity of illness (Supplementary Appendix SA2). For cases not seeking medical care, we used FoodNet Population Survey data on duration of cases with diarrhea (for bacterial pathogens) or vomiting (for viral pathogens and unspecified agents), CDC Foodborne Disease Outbreak Surveillance (FDOSS) data (for bacterial toxins: B. cereus, Clostridium perfringens, Staphylococcus aureus), and estimates from prior cost-of-illness studies (for parasites) (Batz et al., 2014). For cases receiving outpatient care only, we used duration of illness in confirmed outbreaks with no reported hospitalizations (for pathogens with >10 outbreaks), prior cost-of-illness studies (for pathogens with ≤10 outbreaks), and analogized to norovirus for unspecified agents(CDC FDOSS, 2024, Minor et al., 2015, Friedman et al., 2004; White et al., 2019).

We estimated durations of hospitalizations from the National Inpatient Sample (NIS) (Healthcare Cost and Utilization Project [HCUP], 2016–2019). We used pathogen-, pathogen/complication-specific, and “any AGI” ICD codes in any diagnostic position to identify hospitalizations and their complications (Dhaliwal et al., 2021, Supplementary Appendix SA1 Tables SA1.2, 1.3). Duration of post-hospitalization recovery was modeled as 2 times the length of hospitalization for most pathogens and unspecified AGI and 3 times the length of hospitalization for pathogens causing more severe disease (Brucella spp., Clostridium botulinum, Hepatitis A, Vibrio cholerae, L. monocytogenes, T. gondii) (Supplementary Appendix SA2). These multipliers were informed by quantitative indicators of disease severity in the NIS data and judgments from an expert panel advising previous ERS post-hospitalization recovery duration estimates (Ahn et al., 2022; NIS 2014–2019; Batz et al., 2014) (Supplementary Appendix SA2). We calculated workdays lost as a proportion of the duration of illness for each health state, with an additional 1.6 workdays lost for hospitalized cases to account for prehospitalization illness.

Chronic health outcomes

A major strength of this study is its thorough review and analysis of primary research on chronic sequelae incidence and costs relevant to U.S. conditions. Bradford Hill criteria guided our decisions on which sequelae to include (Bradford Hill, 1965). We focused on U.S. research, but also looked to research outside the United States where that evidence was relevant to the U.S. Decisions on which sequelae to include were based on prior literature and expert advice. In addition to an expert workshop advising on the strength of evidence relevant to inclusion decisions, we examined which chronic sequelae were included in prior cost-of-illness studies in the United States and internationally, commissioned several systematic reviews, and conducted conventional reviews of prior literature (Hoffmann and Scallan, 2020; Pogreba-Brown et al., 2022; Schaefer et al., 2022; Supplementary Appendix SA3). Finally, we sought subject matter experts’ review of our inclusion decisions, modeling, and evidentiary support (Supplementary Appendix SA3). Estimates of the duration of chronic sequelae were based on prior literature (Supplementary Appendix SA2).

We included chronic sequelae for Campylobacter spp., nontyphoidal Salmonella, Shigella, Yersinia, STEC O157 and non-O157, pregnancy-related L. monocytogenes, and both pregnancy-related and acquired toxoplasmosis (Table 3, Supplementary Appendix SA1 Table SA1.1). There were areas where our results diverged from other research efforts in the United States and elsewhere. (Supplementary Appendix SA1 Table SA 1.1, Supplementary Appendix SA3).

Table 1.

Estimated Annual Number of Episodes of Domestically Acquired Foodborne Illnesses, United States, by Disease Severity

		Health State
Pathogen	Total Acute Illnesses (Mean)^a	No Physician’s Visit	Physician’s Visit	Hospitalized, Recovered	Hospitalized, Died	Chronic Sequelae^b
Bacteria
Bacillus cereus	63,400	50,023	13,357	20	0	—
Brucella spp.	839	614	170	54	1	—
Campylobacter spp.	845,024	664,189	172,372	8387	76	85,652
Clostridium botulinum	55	0	13	33	9	—
Clostridium perfringens	965,958	762,141	203,379	412	26	—
STEC O157	63,153	44,144	16,871	2118	20	9
STEC non-O157	112,752	87,834	24,647	271	0	1
ETEC	17,894	14,118	3764	12	0	—
Diarrheagenic E. coli other than STEC and ETEC	11,982	9454	2520	8	0	—
Listeria monocytogenes (nonpregnancy associated)^c ^, ^d	1352	0	0	1107	245	—
L. monocytogenes (pregnancy associated)^c	231	—	—	110	57	22
Mycobacterium bovis	60	—	29	28	3	—
Salmonella spp., nontyphoidal	1,027,561	810,746	197,479	18,958	378	103,784
Salmonella Typhi	1821	1184	440	197	0	—
Shigella spp.	131,254	104,741	25,057	1446	10	13,257
Staphylococcus aureus	241,148	190,266	49,818	1058	6	—
Streptococcus spp., Group A	11,217	8850	2366	1	0	—
Vibrio cholerae, toxigenic	84	54	28	2	0	—
V. vulnificus	96	0	3	57	36	—
V. parahaemolyticus	34,664	27,350	7214	96	4	—
Vibrio spp., other	17,564	13,858	3623	75	8	—
Yersinia enterocolitica	97,656	79,687	17,436	504	29	2539
Parasites
Cryptosporidium spp.	57,616	47,245	10,161	206	4	—
Cyclospora cayetanensis	11,407	8704	2692	11	0	—
Giardia intestinalis	76,840	63,009	13,606	223	2	—
Toxoplasma gondii (nonpregnancy associated)^d	86,686	68,405	13,853	4101	327	1719
T. gondii (pregnancy associated)^d	67	—	—	45	22	45
Trichinella spp.	156	101	49	6	0	—
Viruses						—
Astrovirus	15,433	14,044	1302	87	0	—
Hepatitis A virus	1566	1192	275	92	7	—
Norovirus	5,461,731	4,970,175	476,893	14,514	149	—
Rotavirus	15,433	12,346	2739	348	0	—
Sapovirus	15,433	14,044	1302	87	0	—
Subtotal	9,388,133	8,068,518	1,263,458	54,564	1419	206,723
Unspecified agents	38,392,704	34,937,361	3,383,465	70,192	1686
TOTAL	47,717,208	43,005,879	4,646,923	124,756	3105	206,723

Total acute illnesses, hospitalizations, and deaths are from Scallan et al. (2011). Additional health states (no physician’s visit, physician’s visit) were derived from multipliers provided in Scallan et al. (2011) appendices. (Supplementary Appendix SA2).

Estimates are the number of chronic sequelae resulting from acute infections. One patient may have multiple chronic sequelae (Supplementary Appendices SA2 and SA3).

Estimates of the incidence of listeriosis in Scallan et al. 2011 represent the combined incidence of nonpregnancy and pregnancy associated illness. This article separately estimated the costs for nonpregnancy associated listeriosis and pregnancy associated listeriosis. We were unable to account for hospitalizations for infected infants.

For toxoplasmosis and listeriosis, “Hospitalized, died” includes fetal loss, stillbirths, and neonatal deaths.

Totals are modeled to reflect uncertainty simultaneously across outcomes.

ETEC, enterotoxigenic Escherichia coli; STEC, Shiga toxin–producing Escherichia coli.

Economic costs

Cost estimates were developed from large insurance or hospital charge databases, public data sources, and existing literature and were adjusted to 2023 dollars.

Medical treatment costs

For outpatient costs, we relied on MarketScan Commercial Claims and Encounters data (Whitham et al., 2022), which includes >90 million people in the United States covered by employer-sponsored health insurance. We used the estimated cost of treating AGI for all pathogens. Since diagnostic tests are generally only ordered for more severe or persistent outpatient cases, using pathogen-specific estimates would overestimate the cost of outpatient treatment. Outpatient cost estimates included treatment costs of outpatient visits, clinic visits, and emergency department visits not resulting in hospitalization, as well as associated procedures, laboratory tests, and prescription medication costs.

We estimated hospitalization costs from 2016 to 2019 NIS hospitalization discharge data. This period is after the U.S. hospitals shifted from using ICD-9 to ICD-10 in coding charges (October 15, 2015) and before the COVID pandemic (2020). Charges were adjusted to cost using NIS charge-to-cost ratios and updated for inflation to 2023 dollars. Emergency department visits were included in the hospitalization cost if they led to admission. For norovirus, sapovirus, C. perfringens, and unspecified agents, we used the cost of hospitalizations with a diagnosis of unspecified AGI, as most cases are undiagnosed (Scallan et al., 2011b; Lopman et al., 2011; Gastañaduy et al., 2013). Based on expert panel advice for prior ERS estimates, for Vibrio parahaemolyticus and other Vibrio spp. we used the cost of Salmonella hospitalizations, and for V. vulnificus we used the costs of listeriosis (Hoffmann et al. 2012). For all other pathogens, we used pathogen-specific hospitalization costs. Since diagnosis is more common for hospitalized cases due to their severity, we used hospitalizations with pathogen-specific diagnoses to estimate hospitalization costs (Dhaliwal et al., 2021).

Value of time

We valued only that time that would otherwise be spent in the labor market, a conservative measure of impacted time that was used in prior ERS estimates (Hoffmann et al., 2012). We calculated lost productivity as the product of workdays lost, labor force participation rate, and the daily before tax total compensation for labor. We based our estimates on U.S. Bureau of Labor Statistics and U.S. Census estimates of mean hourly wages plus benefits (total compensation), average weekly hours worked, an assumption of a 5-day work week, and average U.S. employment rate (Supplementary Appendix SA2). We calculated average daily hours worked from average weekly hours worked divided by 5, adjusted for the average U.S. employment rate (Supplementary Appendix SA2).

Costs of chronic sequelae

We based our estimates of the cost of chronic sequelae on prior U.S. and non-U.S. literature (Appendices 2 and 3). These estimates generally include the cost of treatment and the value of lost time.

Value of preventing premature mortality

Deaths were valued using 2010 U.S. Environmental Protection Agency (EPA) estimates of the value of a statistical life (VSL) updated for inflation and income growth to $2023 (EPA 2011) (Supplementary Appendix SA2). The VSL is calculated from estimates of individual WTP to reduce their own risk of premature death. These values are scaled up to a population level so that in aggregate they provide an estimate of the U.S. population’s WTP to prevent 1 premature death. The EPA VSL estimate is based on a meta-analysis of primary WTP studies (EPA 2011). Because of the lack of literature on WTP to avoid miscarriages and stillbirths and only having separate estimates of miscarriages/stillbirths for Listeria, we use the VSL to value miscarriages, stillbirths, and neonatal deaths. It seems likely that at least miscarriages are valued differently from neonatal deaths. We explore a lower bound on the cost of pregnancy-related listeriosis by valuing miscarriages and stillbirths at $0 (Supplementary Appendix SA1 Table SA 1.2).

Uncertainty analysis

In each disease-outcome model, uncertainty was captured through independent random draws from individual parameters, including case numbers, transition probabilities, and costs (Supplementary Appendix SA2). After assigning candidate distributions to parameters in each disease-outcome model, we generated lower and upper bound values by anchoring to the deterministic lower bound estimates from data sources described previously. We ran Monte Carlo simulations to randomly draw values from probability distributions across multiple parameters 1000 times (inner loop) and recording the posterior distribution (mean and 95% credibility interval [CrI]) of 100 outcomes (outer loop) in the disease-outcome trees. We used Microsoft Excel and Microsoft Visual Basic for Applications programming because of the wide accessibility of Microsoft products and the transparency of coding language and formulas. on the uncertainty analysis, please see Supplementary Appendix SA3.

Results

Of 48 million people with domestically acquired foodborne illnesses each year in the United States, we estimated that 43 million did not seek medical care (Table 1). Of those that did, 4.6 million sought outpatient care only, 125,000 were hospitalized and subsequently recovered, and 3100 were hospitalized and died. These illnesses resulted in 206,723 cases of sequelae. Of those with foodborne illness caused by unspecified agents, 35 million (91%) did not seek medical care, 3.4 million (9%) sought outpatient care only, 70,000 (0.2%) were hospitalized and recovered, and 1700 (0.004%) died. Among those with illnesses caused by 31 major known pathogens, 8 million (86%) did not seek medical care, 1.3 million (13%) sought outpatient care only, 55,000 (0.6%) were hospitalized and recovered, and 1400 (0.02%) died (Table 1).

We estimated that foodborne illnesses in the United States imposed a mean cost of $75 billion annually (CrI 28–105) (Table 2). This included $51.3 billion from acute illnesses and $23.4 billion from sequelae. For acute illnesses, $5.9 billion (12% of acute illness cost) was from nonhospitalized cases, $2.7 billion (5%) from hospitalizations, and $42.6 billion (83%) from deaths. The illnesses from the 31 major known pathogens cost an estimated $44.7 billion (60% of total cost); those caused by unspecified agents cost an estimated $30 billion.

Table 2.

Annual Costs of Domestically Acquired Foodborne Illnesses (Millions $2023), United States, by Disease Severity^d

	Health State (Acute Illnesses)
Pathogen	No Physician’s Visit,Mean (90% CrI)	Physician’s Visit,Mean (90% CrI)	Hospitalized, Recovered,Mean (90% CrI)	Hospitalized, Died,Mean (90% CrI)	Sequelae,Mean (90% CrI)	Total (Acute + Sequelae)Mean (90% CrI)
Bacteria
B. cereus	1.7 (0.3–4.4)	10.4 (0.7–33.1)	0.3 (0–1.5)	0 (0–0)	NA	12.4 (1–39)
Brucella spp.	0.1 (0–0.1)	0.2 (0–0.5)	1.8 (1–2.8)	13.7 (8.2–20.3)	NA	15.8 (9.3–23.8)
Campylobacter spp.	72.4 (18.5–165.5)	174.5 (26.3–478.2)	144.6 (68.3–250.2)	994.5 (504–1616.9)	9941.3 (3665.1–19,067.7)	11,327.3 (4,282.2–21,578.5)
C. botulinum	0 (0–0)	0 (0–0.1)	2.4 (1.1–4)	113.2 (56.1–175.7)	NA	115.6 (57.1–179.8)
C. perfringens, foodborne	26.7 (5.3–62.5)	158.8 (9.7–516.2)	8.5 (0.8–23.3)	409.8 (41.7–1101)	NA	603.8 (57.5–1703)
STEC O157	4.8 (1.4–10.4)	16.2 (2.1–45.6)	51.7 (12.7–113.7)	313.1 (81.9–667.1)	71.3 (71.3–71.3)	457.1 (169.4–908.1)
STEC non-O157	9.6 (1.3–26.2)	23.6 (2.3–71.2)	5.7 (0–28.8)	0 (0–0)	7.9 (7.9–7.9)	46.8 (11.5–134.2)
ETEC	1.5 (0–6.4)	3.8 (0–16.8)	0.2 (0–1)	0 (0–0)	NA	5.5 (0–24.3)
Diarrheagenic E. coli other than STEC and ETEC	1 (0–4.3)	2.5 (0–11.1)	0.2 (0–0.8)	0 (0–0)	NA	3.7 (0–16.1)
L. monocytogenes (nonpregnancy associated)^a	0 (0–0)	0 (0–0)	50.9 (17.4–99.9)	3,099.1 (1083.1–5941)	NA	3150.1 (1100.4–6040.9)
L. monocytogenes (pregnancy associated)^a			4.6 (1.6–8.8)	124.9 (44.3–237.1)	60 (19.4–109.9)	814.2 (284.4–1548.8)
M. bovis^b	0 (0–0)	0.3 (0.1–0.4)	1.7 (1–2.5)	41.9 (28.4–57.3)	NA	43.7 (29.4–60.1)
Salmonella spp., nontyphoidal	88.1 (29.9–176.7)	191 (37.4–485.5)	358.3 (148.3–658.5)	4668.1 (2053.6–8139.7)	11,821.6 (6588.4–18,637.7)	17,127 (8857.7–28,098.1)
Salmonella Typhi	0.1 (0–0.4)	0.6 (0–2.5)	1.8 (0–8.3)	0 (0–0)	NA	2.5 (0–11.3)
Shigella spp.	11.4 (1.6–30.8)	24.6 (1.5–77.8)	23.6 (4.4–56.5)	109.6 (20.7–254.4)	1514.3 (269.6–3655.7)	1,683.5 (297.7–4075.1)
S. aureus	4 (1.1–8.6)	37.7 (2.8–117)	29.2 (4.5–72.5)	106.9 (17.3–260.4)	NA	177.8 (25.8–458.4)
Streptococcus spp., Group A, foodborne	0.4 (0–1.6)	2.2 (0–9.7)	0 (0–0.1)	0 (0–0)	NA	2.6 (0–11.4)
V. cholerae, toxigenic	0 (0–0)	0 (0–0.1)	0 (0–0.2)	0 (0–0)	NA	0.1 (0–0.3)
V. vulnificus	0 (0–0)	0 (0–0.1)	2.1 (1.2–3)	436.3 (275.7–614.6)	NA	438.4 (277–617.7)
V. parahaemolyticus	3 (0.9–6.2)	6.9 (1–18.6)	1.8 (0.8–3.2)	65.1 (32.7–106.7)	NA	76.8 (35.4–134.7)
Vibrio spp., other	1.5 (0.5–3)	3.5 (0.5–9.2)	1.4 (0.8–2.3)	96.1 (59.4–140.4)	NA	102.5 (61.2–154.9)
Y. enterocolitica	8.6 (1.8–21.1)	16.3 (1.8–47.5)	15.7 (0–65.7)	353.6 (0.3–1471.8)	2.5 (0.6–5.9)	396.7 (4.6–1612)
Parasites
Cryptosporidium spp.	6.6 (1.1–16.7)	11.2 (1.1–33.5)	5.3 (1.4–11.3)	45.4 (12.7–94.2)	NA	68.5 (16.3–155.7)
C. cayetanensis	1.2 (0–4.4)	3.8 (0–14.9)	0.2 (0–1.1)	0 (0–0)	NA	5.3 (0–20.4)
Giardia intestinalis	6.8 (2.4–13.5)	18.5 (4.7–42.8)	6.2 (3.7–9.5)	27.1 (17–39)	NA	58.6 (27.8–104.8)
T. gondii (nonpregnancy associated)	9.6 (4.5–16.5)	25.5 (5.8–60)	221.8 (128.5–339.1)	4,339 (2583.6–6460.4)	755.7 (566.4–968.8)	5351.5 (3288.8–7844.8)
T. gondii (pregnancy associated)^c				298.3 (24–829.7)	69.5 (57.7–81.1)	367.8 (81.6–910.8)
Trichinella spp.	0 (0–0)	0.1 (0–0.5)	0.1 (0–0.7)	0 (0–0)	NA	0.3 (0–1.2)
Viruses
Astrovirus	0.6 (0.2–1.2)	1.1 (0.1–3.5)	3 (0.9–6.3)	0 (0–0)	NA	4.8 (1.2–11)
Hepatitis A virus	0.6 (0.2–1.1)	0.4 (0–1.1)	1.9 (0.8–3.6)	94.8 (40.6–167.1)	NA	97.7 (41.7–172.9)
Norovirus	261.5 (146.4–407.6)	393.5 (39.5–1164.1)	304.9 (159–500.1)	2008.4 (1111.5–3118.1)	NA	2968.3 (1456.4–5189.9)
Rotavirus	0.6 (0.2–1.1)	2.3 (0.2–6.9)	7.2 (2.5–14.2)	0 (0–0)	NA	10 (2.9–22.2)
Sapovirus	0.6 (0.2–1.2)	1.1 (0.1–3.4)	1.8 (0.6–3.6)	0 (0–0)	NA	3.5 (0.9–8.3)
Subtotal	524.4 (367.2–705)	1127.2 (576.4–1892.7)	1266 (946.4–1634.1)	18,350.8 (14,015.1–23,513.4)	24,244 (11,246–42,606)	44,743.3 (30,869.6–61,554.3)
Unspecified agents	1588.8 (796.7–2616.7)	2703.4 (254.9–8113.7)	1478.1 (194.5–3806.7)	24,202.4 (3358.9–60,761.2)	NA	29,972.8 (4605–75,298.2)
TOTAL	2113.2 (1163.9–3321.6)	3830.7 (831.3–10,006.3)	2744.1 (1140.9–5440.8)	42,553.2 (17,374–84,274.6)	24,244 (11,246–42,606)	74,716.1 (35,474.6–136,852.5)

For pregnancy-associated L. monocytogenes, “hospitalized, recovered” costs are for infected mothers only; “hospitalized, died” includes fetal loss, stillbirths, and neonatal deaths; and “sequelae” are for infected infants.

All M. bovis cases are treated as acute even though treatment generally takes at least a year. Costs of multidrug resistant (MDR) and non-MDR cases were estimated separately, but costs were combined in this table to “hospitalized, recover” (Supplementary Appendix SA2).

For pregnancy-associated T. gondii, “hospitalized, died” includes fetal loss, stillbirths, and neonatal deaths.

Totals are modeled to reflect uncertainty simultaneously across outcomes.

CrI, 95% credibility interval; ETEC, enterotoxigenic Escherichia coli; STEC, Shiga toxin–producing Escherichia coli.

The mean total costs for known pathogens ranged from $0.1 million for V. cholerae to $17.1 billion for nontyphoidal Salmonella (Table 2). The 5 pathogens with the highest mean total cost were as follows: nontyphoidal Salmonella, Campylobacter, Toxoplasma, L. monocytogenes, and norovirus (Fig. 1).

Mean per-case cost ranged from $196 for B. cereus to $4.6 million for V. vulnificus (Table 3). Across known pathogens, the distribution of per-case cost was highly skewed with a mean of $326,515 and a median of $1373. Four pathogens (V. vulnificus, L. monocytogenes [total], C. botulinum, M. bovis) had mean per-case costs exceeding $500,000. Toxoplasma and hepatitis A had mean per-case costs of approximately $65,000, Salmonella $17,000, and Campylobacter, Brucella, and Shigella approximately $13,000–$19,000. Eight pathogens had mean per-case costs ranging from $1000 to $7000. The remainder of specified pathogens had per-case costs less than $1000. Unspecified agents had a mean per-case cost of $781.

Chronic sequelae imposed a mean total cost of $23.4 billion (Table 4). By pathogen, chronic sequelae costs ranged from $2.5 million (CrI: 0.6–6 million) for yersiniosis to $11.8 billion (CrI: 6.6–18.6 billion) for salmonellosis (Table 4). By sequelae, irritable bowel syndrome imposed the highest total cost among the sequelae included in this study ($23 billion). End-stage renal disease from STEC 0157 and non-O157 imposed the highest per-case cost ($7.9 million), followed by congenital outcomes ($1.9 million). Reactive arthritis had the lowest per-case cost ($1226) followed by IBS ($153,099) (Table 3).

FIG. 1.

Mean cases by pathogen versus mean total cost by pathogen.

Table 3.

Foodborne Illnesses: Pathogen Ranking by Per-Case Cost ($2023), Total Cost ($2023), and Case Incidence

Pathogen	Per-Case CostMean	Total CostMean (Rank)	CasesMean (Rank)
T. gondii (total)	5,551,287	5,719,300,000 (3)	86,753 (9)
V. vulnificus	4,566,667	438,400,000 (9)	96 (28)
L. monocytogenes (total)	2,504,296	3,964,300,000 (4)	1583 (24)
C. botulinum	2,101,818	115,600,000 (12)	55 (31)
M. bovis	728,333	43,700,000 (19)	60 (30)
Hepatitis A virus	62,388	97,700,000 (14)	1566 (25)
Brucella spp.	18,832	15,800,000 (20)	839 (26)
Salmonella spp., nontyphoidal	16,668	17,127,000,000 (1)	102,756 (2)
Campylobacter spp.	13,405	11,327,300,000 (2)	845,024 (4)
Shigella spp.	12,826	1,683,500,000 (6)	131,254 (6)
STEC O157	7238	457,100,000 (8)	63,153 (12)
Vibrio spp., other	5836	102,500,000 (13)	17,564 (16)
Y. enterocolitica	4062	396,700,000 (10)	97,656 (8)
V. parahaemolyticus	2216	76,800,000 (15)	34,664 (14)
Trichinella spp.	1923	300,000 (30)	156 (27)
Salmonella Typhi	1373	2,500,000 (29)	1821 (23)
V. cholerae, toxigenic	1190	100,000 (31)	84 (29)
Cryptosporidium spp.	1189	68,500,000 (16)	57,616 (13)
Giardia intestinalis	763	58,600,000 (17)	76,840 (10)
S. aureus	737	177,800,000 (11)	241,148 (5)
Rotavirus	648	10,000,000 (22)	15,433 (17)
C. perfringens	625	603,800,000 (7)	965,958 (3)
Norovirus	543	2,968,300,000 (5)	5,461,731 (1)
C. cayetanensis	465	5,300,000 (24)	11,407 (21)
STEC non-O157	415	46,800,000 (18)	112,752 (7)
Astrovirus	311	4,800,000 (25)	15,433 (18)
Diarrheagenic E. coli other than STEC and ETEC	309	3,700,000 (26)	11,982 (20)
ETEC	307	5,500,000 (23)	17,894 (15)
Streptococcus spp., Group A	232	2,600,000 (28)	11,217 (22)
Sapovirus	227	3,500,000 (27)	15,433 (19)

ETEC, enterotoxigenic Escherichia coli; STEC, Shiga toxin–producing Escherichia coli.

Table 4.

Chronic Sequelae to Foodborne Disease, Incidence and Cost (Millions $2023), United States

Pathogen	CasesMean (90% CrI)	Cost (Millions $2023)Mean (90% CrI)
Bacteria
Campylobacter spp.	85,652	9941 (3665–19,068)
Guillain–Barre syndrome	304	205 (72–409)
Irritable bowel syndrome	63,377	9710 (3587–18,594)
Reactive arthritis	21,971	27 (7–65)
STEC O157
End-stage renal disease	9	71
STEC non-O157
End-stage renal disease	1	8
L. monocytogenes (pregnancy associated)
Neurological disorder in infants	22	60 (19–110)
Salmonella spp., nontyphoidal	103,784	11,822 (6588–18,638)
Irritable bowel syndrome	77,067	11,789 (6578–18,568)
Reactive arthritis	26,717	33 (11–70)
Shigella spp.	13,257	1514 (270–3656)
Irritable bowel syndrome	9844	1510 (269–3644)
Reactive arthritis	3414	4 (1–12)
Y. enterocolitica
Reactive arthritis	2539	3 (1–6)
Parasites
T. gondii (nonpregnancy associated) (severe visual impairment)	1719	756 (566–969)
T. gondii (pregnancy associated [infants])	43	69 (58–81)
Visual impairment only	22	11 (9–13)
Visual and cognitive impairment	20	55 (45–64)
Visual, cognitive, and hearing impairment	1	4
TOTAL	206,723	24,244 (11,246–42,606)

Estimates for chronic sequelae from literature, see Supplementary Appendix SA2.

Totals are modeled to reflect uncertainty simultaneously across outcomes.

CrI, 95% credibility interval; ETEC, enterotoxigenic Escherichia coli; STEC, Shiga toxin–producing Escherichia coli.

Discussion

CDC estimates that there are 48 million foodborne illnesses annually in the United States (Scallan et al. 2011; CDC 2023). This study found that these illnesses cost the United States $75 billion annually. Like other high-income countries, the United States invests substantial effort in food safety. But further progress in the United States is still possible and desirable. Our pathogen cost estimates can help inform food safety priorities. Our pathogen-specific per-case cost estimates will help inform benefit–cost analysis required for new federal food safety regulations.

While less than 1% of foodborne illnesses require hospitalization, serious illness and death do occur and are the major drivers of cost. Prevention of death often accounts for over 80% of economic benefit from regulations designed to protect health (e.g., United States, Environmental Protection Agency). In this study, 0.01% of cases resulted in deaths but accounted for 56% of total cost ($42.6 billion). We estimate that the 47.4 million cases of foodborne illnesses occurring annually in the United States result in 206,723 cases of sequelae (0.004 sequelae cases per case of foodborne illness). Yet, chronic sequelae accounted for an additional 31% of total cost ($23.4 billion). Only 0.3% of cases were hospitalized and recovered; they account for 4% of the total cost. Among cases of known pathogen origin, chronic sequelae accounted for 52% of total costs, deaths 41%, and hospitalized and recovered cases 3%.

A comparison to prior ERS estimates demonstrates the effect of updating chronic sequelae estimates. ERS 2014 estimates of the annual costs of 15 leading foodborne pathogens found that deaths accounted for 83% of total cost and chronic sequelae 15% (Hoffmann et al., 2015). In our study, deaths accounted for only 39% of the total cost for these same 15 pathogens, whereas chronic sequelae accounted for 54%. Since both sets of estimates build on the same CDC disease incidence estimates and are based on similar methods and the same VSL values, the difference is driven primarily by improved estimates of the frequency and cost of sequelae.

Distributional impact is also a concern to decision makers. Severity of disease is relevant to these concerns. Per-case cost estimates provide an aggregate indication of severity. V. vulnificus, C. botulinum, and M. bovis rank in the top 5 for per-case cost but do not rise to the top 5 on any of the total impact measures (Table 5).

Table 5.

Top 5 Pathogens Ranked by Outcome Measure

Rank	Total Cost	Per-Case Cost	Illnesses	Hospitalizations	Deaths	DALYs
1.	Salmonella spp. (nontyphoidal)	V. vulnificus	Norovirus	Salmonella spp. (nontyphoidal)	Salmonella spp. (nontyphoidal)	Salmonella spp. (nontyphoidal)
2.	Campylobacter spp.	L. monocytogenes	Salmonella spp. (nontyphoidal)	Norovirus	T. gondii	T. gondii
3.	T. gondii (acquired)	C. botulinum	C. perfringens	Campylobacter spp.	L. monocytogenes	Campylobacter spp.
4.	L. monocytogenes (acquired)	M. bovis	Campylobacter spp.	T. gondii	Norovirus	Norovirus
5.	Norovirus	T. gondii	S. aureus	Shiga toxin E. coli O157 (STEC)	Campylobacter spp.	L. monocytogenes

ETEC, enterotoxigenic Escherichia coli; DALY, disability-adjusted life year; STEC, Shiga toxin–producing Escherichia coli.

Our results support prioritization among known pathogens and provide insight into the importance of non-pathogen-specific prevention efforts. Due to the large number of cases, unspecified agents accounted for 40% of the total cost of foodborne illness in the United States. But 18 known pathogens (25% of all cases with known pathogen cause) had greater per-case costs than unspecified agents ($781) (Table 5). Rankings can help public health and food safety policymakers identify where to focus prevention efforts (Table 5, Fig. 1). Rankings on cost-of-illness are similar to those based on physical outcomes or disability-adjusted life year estimates of known pathogens but in a metric that can be compared to program costs (Table 5).

In our review of the sequelae literature, we found areas where additional epidemiological research would improve the estimation of the burden or cost of foodborne illness. We found emerging literature on a number of sequelae (norovirus and chronic diarrhea, foodborne V. vulnificus and amputations, toxoplasmosis and schizophrenia, sequelae to meningitis, and generalized hypertension linked to chronic kidney disease from toxogenic E. coli) but judged the evidence not yet strong enough to support their inclusion (Supplementary Appendix SA3). U.S. literature on congenital toxoplasmosis is limited due to a lack of surveillance data. Still, due to differences in prevalent serotypes and differences in disease surveillance and treatment, reliance on the larger European literature on congenital toxoplasmosis would likely lead to an underestimation of the cost of congenital toxoplasmosis in the United States (Supplementary Appendix SA3). While the impact of antibiotic resistance was reflected in our total costs, we were unable to find primary epidemiological research that would have allowed us to estimate it separately for all pathogens. Finally, epidemiological studies often report on clinical outcomes that are difficult to translate into functional life outcomes that can be valued. For example, studies of congenital toxoplasmosis report intracranial calcifications, hydrocephalus, and central nervous system abnormalities rather than mental impairment (by degree of severity). These are all areas where continued epidemiological research is needed to support cost estimation.

This study has several limitations. By relying on CDC incidence estimates (Scallan et al., 2011), we do not reflect recent disease trends, most notably the drop in number of cases of hepatitis A and toxoplasmosis in recent years (Ly and Klevens 2015; Hou et al., 2018; Jones and Dubey 2012). Another limitation imposed by deferring to CDC disease estimates is that the disease estimates are circa 2006, whereas our cost estimates are circa 2023. Like prior cost-of-illness studies, we did not capture the value of time the ill would have spent on home production, leisure, or children’s activities. Similarly, we did not include WTP for pain and suffering due to lack of data and methodological concerns. There is a lack of research on WTP to avoid miscarriages and stillbirths. We see our estimates valuing them using the VSL as an upper bound (Supplementary Appendix SA1). Research that uses expert medical review of hospitalization records to assess the accuracy of the ICD coding would be useful in assessing the accuracy of hospitalization identification used in this study. These are all areas where further research would strengthen cost-of-illness studies, including the ERS cost of foodborne illness estimates.

Conclusions

This study estimated the cost of foodborne illness in the United States circa 2023 at $75 billion. Deaths accounted for 56% of the mean cost and chronic outcomes for 31%. These new estimates include 9.4 million cases of foodborne illness due to 31 foodborne pathogens and unspecified foodborne agents (Scallan et al., 2011). The costliest pathogen was Salmonella at $17.1 billion followed by Campylobacter at $11.3 billion. Toxoplasma ($5.7 billion) and Listeria ($4.0 billion) followed due primarily to deaths and chronic outcomes from pregnancy-associated cases. Per-case cost ranged from $196 for B. cereus to $4.6 million for V. vulnificus. Forty percent of the total cost-of-illness was due to foodborne illness of unspecified pathogen cause, but these cases were generally mild (per-case cost $781). With the growing interest in global burden of foodborne illness and in the cost of foodborne illnesses, it should be noted that countries vary substantially in accessibility and use of medical care services and the cost of medical care services. The United States has high medical treatment costs compared with other advanced countries (Papanicolas et al., 2018).

Footnotes

Acknowledgments

The authors are grateful to the Agency for Healthcare Research and Quality (AHRQ) for allowing them to use hospital discharge and cost data in the National Inpatient Sample (NIS) database developed by the Healthcare Cost and Utilization Project (HCUP) (). The authors thank reviewers of their methods: Michael Batz (FDA) methods, Maria Negron Sureda (CDC) brucellosis, Jennifer Unukwube Okaro (CDC) Streptococcus, Sarah Collier (CDC) methods, Despina Contopoulos-Ioannidis (Stanford U.) congenital toxoplasmosis, Megan Hofmeister (CDC) hepatitis A, Arie Havelaar (U. of Florida) methods, Anne Straily (CDC) toxoplasmosis and trichinosis, Anita Kambhampati (CDC) viral disease, Lauren Ahart (CDC) trichinosis, Jitender Dubey (USDA ARS) toxoplasmosis, Patricia Griffin and team (CDC) enterics, and Mark Riddle (U. of Nevada) chronic bowel sequelae. In particular, the authors thank Suzanne Marks (CDC) for her extensive advice on the Mycobacterium bovis modeling and Rima McLeod (U. of Chicago) for her advice on toxoplasmosis modeling.

Ethics

All estimates were based on deidentified public data or previously published research. No interaction or intervention with human subjects occurred and no personally identifiable information was used, collected, or transmitted. This analysis did not meet the definition of human subjects research (as defined in the US Code of Federal Regulations, Title 45 Part 46) and was not subject to review by an institutional review board.

Disclaimer

The findings and conclusions in this article are those of the authors and should not be construed to represent any official USDA or U.S. Government determination or policy.

Authors’ Contributions

S.H.: Conceptualization, methodology, validation, resources, writing—original draft, review and editing, visualization, supervision, project administration, and funding acquisition. A.E.W.: Methodology, software, validation, formal analysis, data curation, and writing—review and editing. R.B.M.: Methodology, software, validation, formal analysis, data curation, and writing—review and editing. J.-W.A.: Software, validation, formal analysis, data curation, and writing—review and editing. L.B.G.-S.: Software, validation, formal analysis, data curation, and writing—review and editing. E.J.S.W.: Conceptualization, methodology, validation, resources, writing—original draft, review and editing, visualization, supervision, project administration, and funding acquisition.

Disclosure Statement

No competing financial interests exist.

Funding Information

This research was supported by the U.S. Department of Agriculture, Economic Research Service.

Supplementary Material

Supplementary Appendix SA1

Supplementary Appendix SA2

Supplementary Appendix SA3

References

Ahn

, Scallan Walter

, White

, et al. Identifying sepsis from foodborne hospitalization: Incidence and hospitalization cost by pathogen. Clin Infect Dis, 2022; 75(5):857–866; doi: 10.1093/cid/ciab1045

Batz

, Hoffmann

, Morris

JG.

Jr ., Disease-outcome trees, EQ-5D scores, and estimated annual losses of quality-adjusted life years (QALYs) for 14 foodborne pathogens in the United States. Foodborne Pathog Dis, 2014; 11(5):395–402; doi: 10.1089/fpd.2013.1658

Bradford Hill

. The environment and disease: Association or causation? Proc Royal Soc Med, 1965; 58(5):295–300.

Dhaliwal

, Hoffmann

, White

, et al. Cost of hospitalizations for leading foodborne pathogens in the United States: Identification by ICD coding and variation by pathogen. Foodborne Pathog Dis, 2021; 18(11):812–821.

Friedman

, Hoekstra

, Samuel

, et al. Emerging Infections Program FoodNet Working Group. Risk factors for sporadic Campylobacter infection in the United States: A case-control study in FoodNet sites. Clin Infect Dis, 2004; 38(Suppl 3):S285–296; doi: 10.1086/381598

Gastañaduy

, Hall

, Curns

, et al. Burden of norovirus gastroenteritis in the ambulatory setting—United States, 2001-2009. J Infect Dis, 2013; 207(7):1058–1065; doi: 10.1093/infdis/jis942

Gastañaduy

, Harrington

, Portney

. Valuing the benefits of health and safety regulation. Journal of Urban Economics, 1987; 22(1):101–112; doi: 10.1016/0094-1190(87)90052-0

Healthcare Cost and Utilization Project (HCUP).. 2016-2019. Healthcare Cost and Utilization Project (HCUP) National Inpatient Sample (NIS). Agency for Healthcare Research and Quality: Rockville, MD. www.hcup-us.ahrq.gov/nisoverview.jsp

Hoffmann

, Ahn

J-A

. Updating economic burden of foodborne disease estimates for inflation and income growth. USDA ERS, Econ Research Report 2021; 297. 2021. Available from: https://www.ers.usda.gov/publications/pub-details/?pubid=102639

10.

Hoffmann

, Batz

, Morris

JG.

Jr ., Annual cost of illness and quality-adjusted life year losses in the United States due to 14 foodborne pathogens. J Food Prot, 2012; 75(7):1292–1302; doi: 10.4315/0362-028X.JFP-11-417

11.

Hoffmann

, Maculloch

, Batz

. Economic burden of major foodborne illnesses acquired in the United States. USDA ERS, Econ Research Rpt, 2015; 1476:120935; doi: 10.22004/ag.econ.205081

12.

Hoffmann

, Scallan Walter

. Acute complications and sequelae from foodborne infections: Informing priorities for cost of foodborne illness estimates. Foodborne Pathog Dis, 2020; 17(3):172–177; doi: 10.1089/fpd.2019.2664

13.

Hou

, Patel

, Farooq

, et al. Decline in ocular toxoplasmosis over 40 years at a tertiary referral practice in the United States. Ocular Immun Inflamm, 2018; 26:577–583.

14.

Johansson

P-O

. An Introduction to Modern Welfare Economics. Cambridge University Press: Cambridge UK; 1991.

15.

Jones

, Dubey

. Foodborne toxoplasmosis. Clin Infect Dis, 2012; 55(6):845–851.

16.

Lopman

, Hall

, Curns

, et al. Increasing rates of gastroenteritis hospital discharges in US adults and the contribution of norovirus, 1996-2007. Clin Infect Dis, 2011; 52(4):466–474; doi: 10.1093/cid/ciq163

17.

, Klevens

. Trends in disease and complications of hepatitis A virus infection in the United States, 1999–2011: A new concern for adults. J Infect Dis, 2015; 212(2):176–182.

18.

Minor

, Lasher

, Klontz

, et al. The per case and total annual costs of foodborne illness in the United States. Risk Anal, 2015; 35(6):1125–1139; doi: 10.1111/risa.12316

19.

Papanicolas

, Woskie

, Jha

. Health care spending in the United States and other high-income countries. JAMA, 2018; 319(10):1024–1039.

20.

Pogreba-Brown

, Austhof

, Armstrong

, et al. Chronic gastrointestinal and joint-related sequelae associated with common foodborne illnesses: A scoping review. Foodborne Pathog Dis, 2020; 17(2):67–86; doi: 10.1089/fpd.2019.2692

21.

Pogreba-Brown

, Boyd

, Schaefer

, et al. Complications associated with foodborne listeriosis: A scoping review. Foodborne Pathog Dis, 2022; 19(11):725–743; doi: 10.1089/fpd.2022.0012

22.

Scallan

, Griffin

, Angulo

, et al. Foodborne illness acquired in the United States unspecified agents. Emerg Infect Dis, 2011a;17(1):16–22; doi: 10.3201/eid1701.P21101

23.

Scallan

, Hoekstra

, Angulo

, et al. Foodborne illness acquired in the United States–major pathogens. Emerg Infect Dis, 2011b;17(1):7–15; doi: 10.3201/eid1701.P11101

24.

Scallan

, Hoekstra

, Mahon

, et al. An assessment of the human health impact of seven leading foodborne pathogens in the United States using disability adjusted life years. Epidemiol Infect, 2015; 143(13):2795–2804; doi: 10.1017/S0950268814003185

25.

Scallan Walter

, Crim

, Bruce

, et al. Incidence of Campylobacter-associated Guillain-Barré syndrome estimated from health insurance data. Foodborne Pathog Dis, 2020; 17(1):23–28; doi: 10.1089/fpd.2019.2652

26.

Schaefer

, Austhof

, Boyd

, et al. Septicemia due to Listeria Monocytogenes infection: A systematic review and meta-analysis. Foodborne Pathog Dis, 2022; 19(2):104–114.

27.

U.S. Centers for Disease Control and Prevention (CDC). (2023). Estimated Flu disease burden, by season—United States, 2010-2012 thorough 2021-2022. Available from: cdc.gov/flu/about/burden/past-seasons.html

28.

White

, Ciampa

, Chen

, et al. Characteristics of Campylobacter, Salmonella infections and acute gastroenteritis in older adults in Australia, Canada, and the United States. Clin Infect Dis, 2019; 69(9):1545–1552.

29.

Whitham

, Gilliland

, Collier

, et al. Direct outpatient healthcare costs of acute gastrointestinal illness among commercially insured persons – common foodborne pathogens, 2012–2015. Foodborne Path Dis, 2022; 19:558–568.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.90 MB

2.40 MB

0.69 MB