Burden of Foodborne Illness: Methods and Data Sources

In 2011, Scallan et al. estimated the overall burden of foodborne illness caused by known and unspecified agents.

This analysis included 31 pathogens known to cause foodborne illness and unspecified agents that cause acute gastroenteritis illnesses (AGI). Scallan et al. also estimated the number of hospitalizations and deaths caused by these illnesses.

Download and print factsheet [6 pages]

Estimating U.S. foodborne illnesses for 31 known foodborne pathogens

For each pathogen, we gathered data from surveillance systems and corrected for underreporting and under-diagnosis. We then multiplied the adjusted number by the proportion of illnesses that was acquired in the United States (that is, not during international travel) and the proportion transmitted by food to yield an estimated number of illnesses that are domestically acquired and foodborne. Then, we added the estimates for each of the pathogens to arrive at a total, and used an uncertainty model to generate a point estimate and 90% credible interval (upper and lower limits) (Figure 1).

Figure 1. Estimating illnesses for pathogens known to cause foodborne illness, 2011*
Figure 1 is a diagram, depicting an equation used to estimate illnesses for pathogens known to cause foodborne illness for the year 2011. The equation is as follows: For each pathogen multiply the number of illnesses reported to surveillance system multiplied by the multiplier to correct for under-reporting multiplied by the pathogen-specific multiplier to correct for under-diagnosis multiplied by the estimated proportion domestically acquired multipiled by the estimated proportion transmitted through food. That results in the estimated number of domestically acquired foodborne illnesses due to each known pathogen, which is then added to the individual pathogen estimates summed. The final result is the estimated # of domestically acquired foodborne illnesses due to known pathogens.

*Probability distributions were used to model uncertainty in each of the data inputs. Point estimates were bounded by a 90% credible interval.
**For six of the 31 pathogens, no routine surveillance data were available so alternative approaches were used to estimate illnesses.

Estimating U.S. foodborne illnesses for unspecified agents

Unspecified agents fall into four general categories:

  • Agents with insufficient data to estimate agent-specific burden
  • Known agents not yet recognized as causing foodborne illness
  • Microbes, chemicals, or other substances known to be in food whose pathogenicity is unproven
  • Agents not yet described

To estimate foodborne illnesses from unspecified agents, we used symptom-based data from surveys to estimate the total number of AGI and then subtracted the number of illnesses accounted for by known gastroenteritis pathogens. We then multiplied this number by the proportion of domestically acquired illnesses and of illnesses attributable to food, just as we did for the known agents. Finally, again as with the known-pathogens estimate, we used an uncertainty model to generate a point estimate and 90% credible interval (upper and lower limits) (Figure 2).

Foodborne illnesses due to chemicals that cause acute gastroenteritis are included in the estimate of illnesses due to unspecified agents. However, chemicals or unspecified agents that do not cause acute gastroenteritis are not included in the estimates.

Figure 2. Estimating foodborne illnesses due to unspecified agents, 2011*
Figure 2 is a diagram, depicting an equation used to estimate foodborne illness due to unspecified agents for the year 2011. The equation is as follows: Estimated number of acute gastro-enteritis illnesses (AGI) from population surveys multiplied by 2006 US population estimates minus Estimated number of illnesses due to 24 known foodborne pathogens that cause AGI. The result of that is then multiplied by Estimated proportion domestically acquired multiplied by Estimated proportion transmitted through food which will result in Estimated number of domestically acquired foodborne illnesses due to unspecified agents.

*Probability distributions were used to model uncertainty in each of the data inputs. Point estimates were bounded by a 90% credible interval.
**Estimated proportions were based on the 24 known pathogens that cause acute gastroenteritis illnesses.

 Top of Page

Estimating hospitalizations and deaths from U.S. foodborne illnesses due to known pathogens

For each known pathogen with surveillance data available, we multiplied the estimated number of reported illnesses (after correcting for underreporting) by the pathogen-specific hospitalization and death rate from surveillance data, surveys, or outbreak data. Because some people with illnesses that were not laboratory-confirmed would also have been hospitalized and died, we doubled the estimates to correct for under-diagnosis. We multiplied the adjusted hospitalization and death estimates by the proportion of illnesses that were acquired within the United States (vs. international travel-related) and the proportion transmitted by food. Finally, we used an uncertainty model to generate a point estimate and 90% credible intervals for both hospitalizations and deaths (Figures 3 and 4).

Figure 3. Estimating hospitalizations from foodborne illnesses due to known pathogens, 2011*
Figure 3 is a diagram, depicting an equation used to estimate hospitalizations from foodborne illnesses due to known pathogens for the year 2011. The equation is as follows: Estimated number of illnesses reported to surveillance system (after correcting for under-reporting), multiplied by the proportion of persons hospitalized, multiplied by 2 to correct for under-diagnosis, multiplied by estimated proportion domestically acquired, multiplied by estimated proportion transmitted through food. This results in the estimated number of hospitalizations from domestically acquired foodborne illnesses due to known pathogens.
*Probability distributions were used to model uncertainty in each of the data inputs. Point estimates were bounded by a 90% credible interval.
Figure 4. Estimating deaths from foodborne illnesses due to known pathogens, 2011*
Figure 3 is a diagram, depicting an equation used to estimate deaths from foodborne illnesses due to known pathogens for the year 2011. The equation is as follows: Estimated number of illnesses reported to surveillance system (after correcting for under-reporting), multiplied by the proportion of persons who died, multiplied by 2 to correct for under-diagnosis, multiplied by estimated proportion domestically acquired, multiplied by estimated proportion transmitted through food. This results in the estimated number of deaths from domestically acquired foodborne illnesses due to known pathogens.

*Probability distributions were used to model uncertainty in each of the data inputs. Point estimates were bounded by a 90% credible interval.

 Top of Page

Estimating hospitalizations and deaths from U.S. foodborne illnesses due to unspecified agents

To estimate hospitalizations, we applied the average hospitalization rate for all AGI, determined from survey data for 2000–2006, to 2006 U.S. population estimates and subtracted the estimated number of hospitalizations caused by the 24 known pathogens that cause AGI. For deaths, we determined the death rate for acute gastroenteritis illnesses from U.S. death certificates (specifically, multiple-cause-of-death data) for 2000–2006 and applied that rate to the 2006 U.S. population estimate. We then subtracted the estimated number of deaths from the 24 known gastroenteritis pathogens. For both hospitalizations and deaths, we multiplied the overall number by the proportion of hospitalizations and deaths from the 24 known gastroenteritis pathogens that was domestically acquired and foodborne. Finally, we used an uncertainty model to generate a point estimate and 90% credible intervals (Figures 5 and 6).

Figure 5. Estimating hospitalizations from foodborne illnesses due to unspecified agents, 2011*
Figure 5 is a diagram, depicting an equation used to estimate hospitalizations from foodborne illnesses due to unspecified agents for the year 2011. The equation is as follows: Hospitalization rate for acute gastroenteritis illness from population surveys, multiplied by 2006 US population estimates, minus number of hospitalizations due to 24 known foodborne pathogens that cause acute gastroenteritis illness. The result of that is then multiplied by the estimated proportion domestically acquired, multiplied by estimated proportion transmitted through food. This results in the estimated number of hospitalizations from domestically acquired foodborne illnesses due to unspecified agents.

*Probability distributions were used to model uncertainty in each of the data inputs. Point estimates were bounded by a 90% credible interval.

Figure 6. Estimating deaths from foodborne illnesses due to unspecified agents, 2011*
Figure 6 is a diagram, depicting an equation used to estimate deaths from foodborne illnesses due to unspecified agents for the year 2011. The equation is as follows: Death rate for acute gastroenteritis illness from US multiple cause-of-death data, multiplied by 2006 US population estimates, minus number of deaths due to 24 known foodborne pathogens that cause acute gastroenteritis illness. The result of that is then multiplied by the estimated proportion domestically acquired, multiplied by estimated proportion transmitted through food. This results in the estimated number of deaths from domestically acquired foodborne illnesses due to unspecified agents.

*Probability distributions were used to model uncertainty in each of the data inputs. Point estimates were bounded by a 90% credible interval.

 Top of Page

Data Sources for the 2011 Estimates

We used five general types of data sources to create the 2011 estimates:

Active surveillance (public health officials actively gather data from state and local health departments, laboratories, hospitals, etc.)

  • Foodborne Diseases Active Surveillance Network (FoodNet)

Passive surveillance (public health officials rely on state and local health departments, laboratories, hospitals, etc. to report data to surveillance systems)

  • National Notifiable Diseases Surveillance System (NNDSS)
  • National Tuburculosis Surveillance System (NTSS)
  • Public Health Laboratory Information System
  • Cholera and other Vibrio Illness Surveillance System (COVIS)

Outbreak surveillance

  • Foodborne Disease Outbreak Surveillance System

Surveys

  • FoodNet Population Survey
  • FoodNet Laboratory Survey
  • National Ambulatory Medical Care Survey (NAMCS)
  • National Hospital Ambulatory Medical Care Survey (NHAMCS)
  • National Hospital Discharge Survey (NHDS)
  • National Health and Nutrition Survey (NHANES)
  • Nationwide Inpatient Sample (NIS)

Vital (government) statistics

  • Multiple-cause-of-death data (from U.S. death certificates)
  • U.S. Census

 Top of Page

Descriptions of Data Sources

Table 1a. Data sources used to estimate illnesses, hospitalizations, and deaths due to known foodborne pathogens in the United States, 2011 (from Scallan et al.)
Table 6a. Data sources used to estimate illnesses, hospitalizations, and deaths due to known foodborne pathogens in the United States, 2011 (from Scallan et al.)
Data Source Data Pathogen(s) Geographic Coverage Time Frame Adjustments
Cholera and  Other Vibrio  Illness  Surveillance  System (COVIS)* Number of case-patient  reports,  proportion  hospitalized, proportion  who died Vibrio cholerae, toxigenic; Vibrio vulnificus; Vibrio parahaemolyticus; other Vibrio spp. United States 2000–2007 Underreporting; Under-diagnosis
Foodborne  Diseases Active  Surveillance  Network  (FoodNet) Number of laboratory-confirmed illnesses,  proportion hospitalized,  proportion who died Campylobacter spp.; Cryptosporidium  spp.; Cyclospora cayetanensis;  Shiga toxin–producing Escherichia coli  O157;  Shiga toxin-producing E. coli non-O157; Listeria  monocytogenes; non-typhoidal Salmonella; Salmonella  serotype Typhi; Shigella spp.; Yersinia enterocolitica FoodNet  sites 2005–2008 Geographical coverage§; Under-diagnosis
Foodborne  Disease Outbreak  Surveillance  System (FDOSS) Number of foodborne  outbreak-associated  illnesses Bacillus cereus; Clostridium  perfringens; Enterotoxigenic Escherichia coli;  Staphylococcus aureus; Streptococcus spp.Group A United States 2000–2007;  (Streptococcus  spp., Group A  1996–2007) Underreporting; Under-diagnosis
Proportion  hospitalized  and  proportion  who  died  in  foodborne  outbreaks Bacillus cereus; Clostridium perfringens; Enterotoxigenic Escherichia coli;  Staphylococcus aureus; Streptococcus spp., Group A;  Clostridium botulinum; Trichinella spp. United States 2000–2007;  (Streptococcus  spp., Group A 1981–2007) Under-diagnosis
Nationwide Inpatient Sample  (NIS) Hospitalization rate Giardia intestinalis (ICD-9-CM code 007.1), Toxoplasma gondii (ICD-9-CM codes  130.0-9) Sample of  discharge  records from  U.S. hospitals 2000–2006 Weighted to give national estimates according to  Healthcare Cost and Utilization Project (HCUP)  criteria; Under-diagnosis (Giardia intestinalis and  Toxoplasma gondii )
Death rate Giardia intestinalis  (ICD-9-CM code 007.1),  Toxoplasma gondii  (ICD-9-CM  codes  130.0-9) Sample of  discharge records from  U.S. hospitals 2000–2006 Under-diagnosis
National  Health  and  Nutrition  Examination  Survey  (NHANES) Seroprevalence Toxoplasma gondii  United States 1999–2004 Rate of infection over  time and percentage  symptomatic
National  Notifiable  Diseases  Surveillance  System (NNDSS) Number of case-patient  reports Brucella spp.; Clostridium botulinum (foodborne);  Trichinella spp.; Hepatitis A; Giardia intestinalis United States 2000–2007 (2002–2007  for Giardia  intestinalis)# Underreporting; Under-diagnosis
Hospitalization  rate Hepatitis A United States 2000–2007 Under-diagnosis
National  Tuberculosis  Surveillance  System (NTSS) Number  of  tuberculosis  case-patient  reports,  proportion  who  died Mycobacterium  bovis United States 2004–2007 Percentage  of  tuberculosis  cases  attributable  to  M.bovis;  under-diagnosis
U.S.  Census Population estimates Astrovirus, rotavirus, sapovirus United States 2006 75% of children experience an episode of clinical  illness by 5 years  of  age
Various acute  gastroenteritis  data sources (see  Table 6b) Acute gastroenteritis  illnesses,  hospitalizations,  and  deaths Norovirus See Table 6b See Table 6b Fraction of acute gastroenteritis attributable to  norovirus

* Passive surveillance from COVIS was used in preference to active surveillance from FoodNet for Vibrio spp. because most illnesses are reported by Gulf States (Florida, Alabama, Louisiana, Texas) that are not included in the FoodNet surveillance area.

Beginning in 2004, there were 10 FoodNet sites. In 2008, the population of these sites was 46 million persons, 15% of the U.S. population.

§ Incidence of laboratory-confirmed illnesses in FoodNet from 2005 to 2008 was applied to the 2006 U.S. Census population estimates.

Data from FDOSS on Streptococcus spp., Group A were included from 1996 to 2007 for illnesses and from 1981 to 2007 for hospitalizations and deaths because of a paucity of data (Appendix 1 and 3).

# Giardia intestinalis became nationally notifiable in 2002.

Table 1b. Data sources used to estimate illnesses, hospitalizations, and deaths due to acute gatroenteritis in the United States, 2011 (from Scallan et al.)
Table 6b. Data sources used to estimate illnesses, hospitalizations, and deaths due to acute gatroenteritis in the United States, 2011 (from Scallan et al.)
Data source Data Definition Geographic coverage Time frame
FoodNet Population Survey Rate of illness Average annual rate of acute gastroenteritis was derived  by multiplying the average monthly prevalence by 12, where an episode of acute gastroenteritis  was defined as diarrhea  (≥3 loose stools in 24 hours) or vomiting in the past month with both lasting >1 day or resulting in restricted daily activities. Persons with a chronic condition in which diarrhea or vomiting was a major symptom and persons with concurrent symptoms of cough or sore throat were excluded. FoodNet sites 2000–2001,  2002–2003,  2006–2007
Death certificates  —  Multiple-cause-of-death  data from the National  Vital Statistics System Death rate Norovirus;  Acute gastroenteritis deaths were identified from the underlying or contributing cause of death classified by ICD-10 diagnostic codes A00.9–A08.5 (infectious gastroenteritis of known cause) A09 (diarrhea and gastroenteritis of presumed infectious origin); and K52.9  (noninfectious gastroenteritis and colitis, unspecified); excluding  A04.7 (enterocolitis due to Clostridium difficile) and A05.1  (botulism) United States 2000–2006
National Ambulatory  Medical Care Survey  (NAMCS); National  Hospital Ambulatory  Medical Care Survey  (NHAMCS) Hospitalization  rate Norovirus;  Acute gastroenteritis hospitalizations were identified from patient visits to clinical settings, including physician offices, hospital emergency and  outpatient departments with a diagnosis of infectious enteritis ICD-9-CM diagnostic codes 001–008  (infectious gastroenteritis of known cause); 009  (infectious gastroenteritis); 558.9 (other and unspecified noninfectious gastroenteritis and colitis), or 787.9  (other symptoms involving digestive system:  diarrhea); [excluding 008.45 (Clostridium difficile colitis) and 005.1 (botulism)] or reason for visit classification codes for diarrhea (1595) or  gastrointestinal infection (1540) resulting in hospitalization. Nationally  representative  sample of U.S  clinical settings 2000–2006
Nationwide Inpatient  Sample (NIS) Hospitalization  rate Norovirus; Acute gastroenteritis hospitalizations were identified from discharges with one of the first three listed diagnoses classified by ICD-9-CM  diagnostic codes 001–008 (infectious gastroenteritis of known cause); 009 (infectious gastroenteritis); 558.9 (other and unspecified noninfectious  gastroenteritis and colitis), or 787.9 (other symptoms involving digestive system: diarrhea); excluding 008.45 (Clostridium difficile colitis) and 005.1  (botulism). Sample of  discharge  records from U.S. hospitals 2000–2006
National Hospital  Discharge Survey (NHDS) Hospitalization  rate Norovirus; Acute gastroenteritis hospitalizations were identified from discharges with one of the first three listed diagnoses classified by ICD-9-M  diagnostic codes 001–008 (infectious gastroenteritis of known cause); 009 (infectious gastroenteritis); 558.9 (other and unspecified noninfectious  gastroenteritis and colitis), or 787.9 (other symptoms involving digestive system: diarrhea); excluding 008.45 (Clostridium difficile colitis) and 005.1  (botulism). Nationally  representative  sample of  discharge  records from U.S. hospitals 2000–2006

There were 8 FoodNet sites participating in the 2000–2001 survey, 9 in 2002–2003, and 10 in 2006–2007. In 2008, the population of these sites was 46 million persons, 15% of the U.S. population.

Codes for other and unspecified noninfectious gastroenteritis and colitis were included because infectious illnesses of unknown etiology are sometimes coded as noninfectious.

 Top of Page

View Page In: PDF [559K]