The Science Behind Iwaspoisoned

IWasPoisoned.com: A Crowdsourced Food Safety Surveillance Platform (2009–Present)

IWasPoisoned.com (IWP) is an online platform launched in 2009 that allows the public to report foodborne illness incidents in real time. It serves as a form of participatory "syndromic surveillance," crowdsourcing consumer reports of suspected food poisoning to supplement traditional public health monitoring. Since its inception, IWasPoisoned has collected hundreds of thousands of consumer reports worldwide, and its data have been used by health departments and industry as an early-warning system for outbreaks.

Below is a categorized bibliography of references (academic studies, government and regulatory documents, conference proceedings, policy papers, and media/industry coverage) that analyze, cite, or discuss IWasPoisoned.com as a tool for food safety surveillance, from 2009 onward. Each entry includes context on how IWP is referenced (e.g. case study, data source, or commentary), along with citation details or excerpts. An impact analysis and citation network summary follow the bibliography.

Academic Publications

Peer-Reviewed Journal Articles

Numerous scientific studies in public health, informatics, and food safety have examined IWasPoisoned.com or included it as a data source. The platform's efficacy, data quality, and role in outbreak detection have been focal points in the literature.

Quade & Nsoesie (2017) – JMIR Public Health Surveillance. This seminal study introduced IWasPoisoned.com as "a platform for crowdsourced foodborne illness surveillance" and described the characteristics of its users and reports. Quade (IWP's founder) and Nsoesie presented descriptive statistics: over 49,000 illness reports from 89 countries by 2017, with 95% implicating restaurants. They highlighted instances where IWP reports presaged official outbreak confirmations by local health departments. The article positions IWasPoisoned as an augmentative surveillance tool to address underreporting in traditional systems.

Barreto et al. (2023) – Foodborne Pathogens and Disease. (Titled "Foodborne Illness Complaint Systems Detect, and Restaurant Inspection Programs Prevent, Restaurant-Associated Foodborne Illness Outbreaks.") A survey-based study of U.S. local health departments' practices, which reports agencies receive reports via privately managed consumer platforms like IWasPoisoned.com. It links robust consumer complaint surveillance (including third-party crowdsourced systems) with increased outbreak detection, reinforcing that integrating IWP-style data streams can improve public health response. (Context: IWP is cited as an example of an external food poisoning reporting site used by health agencies.)

Li et al. (2022) – International Journal of Infectious Diseases. (Study on "Detecting Foodborne Disease Outbreaks in Florida through Consumer Complaints"). This research evaluated multiple consumer-generated data sources for outbreak detection, including Twitter and IWasPoisoned.com reports. The authors found that mining crowdsourced complaints (tweets, IWP posts, emails) helped identify outbreaks that might not be caught through formal channels. IWasPoisoned is referenced as a key novel data stream feeding into Florida's syndromic surveillance, underscoring its practical utility for health departments. (This extends earlier work in NYC and elsewhere by applying text mining/NLP to consumer complaints.)

Altenburger & Ho (2022) – Journal of Institutional and Theoretical Economics. (Working paper circulated in 2018 as "When Algorithms Import Private Bias into Public Enforcement".) In a broader study on algorithmic bias in food safety inspections, the authors cite IWasPoisoned.com as a startup that "has attempted to crowd-source food-poisoning complaints". IWP is discussed alongside other consumer-data initiatives (e.g. Yelp reviews) as part of predictive analytics in food safety enforcement, noting the popular appeal of such tools for directing inspections. The paper raises concerns about data validity and potential biases in crowdsourced complaints, referencing IWP to illustrate new digital approaches that regulators are exploring.

Albuquerque et al. (2019) – Annual Review of Food Science and Technology. In a review of machine learning applications in food safety, the authors discuss emerging data sources for outbreak prediction. They specifically mention "participatory systems such as IWasPoisoned.com" as novel surveillance inputs, citing Quade & Nsoesie (2017) as evidence of IWP's role. This indicates IWP's recognition in the academic community as a noteworthy example of crowdsourced epidemiology integrated into food safety analytics.

Krause et al. (2021) – NPJ Digital Medicine 4, Article 8. (Study on NLP for food safety – "Using natural language processing to characterize and predict foodborne outbreaks.") This research mined Amazon product reviews to identify food safety signals and complemented the data with IWasPoisoned.com reports as a targeted source of consumer illness reports. Each IWP report is curated to mitigate inauthentic posts, making it a rich, pre-filtered dataset. The inclusion of IWP in this study's methodology underlines its value in academic efforts to harness non-traditional data (reviews, crowdsourced reports) for outbreak prediction.

Conference Proceedings and Presentations

IWasPoisoned has also featured in professional conferences and symposia, often as a case example of crowdsourced outbreak detection. These entries include abstracts and papers presented at food safety and public health meetings:

IAFP Annual Meeting 2023 – Technical Abstract, "From Anecdotal to Analytical: Correlating Self-Reported Norovirus-Like Illness with NoroSTAT Data." Quade et al. presented an analysis comparing IWasPoisoned data on norovirus-like symptoms to CDC's NoroSTAT outbreak surveillance. The study (Patrick Quade collaborating with researchers from NC State) found a strong correlation (r≈0.67, p<0.001) between IWP illness trends and official norovirus outbreak reports, with Granger tests suggesting IWP reports can lead trends. Significance: This conference paper provided evidence that curated crowdsourced data (IWP) can predict or parallel formal surveillance, bolstering IWP's credibility in outbreak detection.

ID Week 2024 – Los Angeles, CA. In 2024, a multi-institutional team presented a poster at IDWeek showing a statistically significant correlation between IWasPoisoned.com norovirus-like illness reports and official CDC NoroSTAT data (Pearson r = 0.669, p < 0.001; Granger causality p = 0.003). The findings suggest that curated crowdsourced illness reports can meaningfully align with formal surveillance data and serve as an early signal for norovirus activity.

Global Food Safety Initiative Conference (GFSI) 2019 – Nice, France. Patrick Quade presented on "Informing and empowering the consumer" in the session "The Changing Face of Retail & Food Service." The presentation addressed the shift in technology and social behavior over the previous decade and its connection to food safety. Quade shared the panel with food safety executives from Ecolab, Starbucks, Dubai Municipality, Metro France, and Wegmans Foods. GFSI brings together global food industry leaders including board members from Cargill, Coca-Cola, McDonald's, Nestle, and Amazon.

Government and Regulatory References

U.S. FDA and Federal Initiatives

Regulatory agencies have begun to acknowledge and utilize crowdsourced data from IWasPoisoned in policy initiatives and contracts:

FDA "New Era of Smarter Food Safety" (2019–2020) – In a public FDA meeting on modernizing food safety, industry speakers cited IWasPoisoned.com as an example of heightened consumer connectivity and real-time reporting impacting food safety decisions. One presenter noted that "iwaspoisoned.com is a great example" of consumers sharing unvetted information that can nonetheless prompt rapid responses by companies and regulators. The mention appears in the official meeting transcript, illustrating FDA's awareness of IWP's influence on outbreak visibility and corporate risk (e.g., reference to Chipotle's stock price drop following IWP-reported incidents).

FDA Contract with DineSafe/IWasPoisoned (2021) – The FDA's Center for Food Safety and Applied Nutrition (CFSAN) awarded a contract to DineSafe (the parent company of IWasPoisoned) to supply AI-driven, crowdsourced foodborne illness data during the COVID-19 pandemic. According to the press release, IWP's real-time consumer reports, analytics, and clustering alerts will help FDA researchers understand foodborne illness dynamics under pandemic conditions. FDA officials framed this as part of the agency's commitment to new data streams in its New Era of Smarter Food Safety blueprint. Notably, food safety scientists (e.g. Prof. Ben Chapman and Prof. Lee-Ann Jaykus) endorsed the public-private partnership, emphasizing that crowdsourced data can provide unique, timely signals to regulators that traditional surveillance might miss.

FDA 2025 - Bazaco et al. (2025) – JMIR Public Health and Surveillance 11:e58797. (Titled "Recent Use of Novel Data Streams During Foodborne Illness Cluster Investigations by the United States Food and Drug Administration: Qualitative Review"). This peer-reviewed publication by FDA scientists provides the first comprehensive documentation of how federal agencies systematically integrate "private participatory reporting platforms" (PPRPs) including IWasPoisoned into official outbreak investigations. The study presents three detailed case studies where FDA's Coordinated Outbreak Response & Evaluation (CORE) Network used novel data streams during foodborne illness investigations involving dry breakfast cereal, ketogenic shakes, and plant-based crumble products. In each case, PPRPs provided either early outbreak signals or supplementary data that informed federal response efforts. The authors note that traditional laboratory-based surveillance faces limitations including underreporting and lag times, while PPRPs can "address gaps that exist in traditional foodborne illness surveillance methods and empower consumers." The study documents how NDSs enabled quicker signal detection and provided supplementary information for vehicle confirmation, though it also identifies challenges including incomplete epidemiologic data and the need for manual data extraction. Significantly, the paper concludes that "signal detection and data collection through NDSs will likely become more prevalent" and calls for standardized approaches to maximize their potential in federal investigations. This represents formal acknowledgment by FDA scientists that platforms like IWasPoisoned have become integral components of federal foodborne illness surveillance infrastructure.

State & Local Health Agencies - agencies covering over 90% of the U.S. (by population) monitor IWP for supplementary outbreak data. Several state and local agencies (e.g., Florida Department of Health, NYC DOHMH, Las Vegas, Chicago) have piloted use of IWasPoisoned feeds. For instance, the Florida Department of Health study (above) integrated IWP alerts into their outbreak surveillance algorithms, demonstrating the government uptake at the state level.

Connecticut Department of Public Health (2019) – Connecticut Epidemiologist, Vol. 39, No. 2. In an official outbreak investigation report titled "Norovirus Outbreaks at a Food Service Establishment, Connecticut 2018–2019," the Connecticut DPH documented that during a December 2018 outbreak, complaints (n=101) were reported through three channels: the local health department, DPH, and "the website https://iwaspoisoned.com." This represents formal acknowledgment in an official state health department publication that IWasPoisoned served as a legitimate reporting mechanism alongside traditional government channels during an actual outbreak response. The report demonstrates real-world integration of crowdsourced data into state epidemiological investigations.

University of Minnesota - A 2024 study by the University of Minnesota and others found that most foodborne illness outbreaks in restaurants are detected through consumer complaint surveillance rather than lab-confirmed cases. The research showed that agencies using centralized complaint databases detected twice as many outbreaks as those without, and that higher complaint volume was positively associated with detection rates. Notably, 23% of local health agencies surveyed reported receiving reports from privately managed platforms such as IWasPoisoned.com, reinforcing the value of crowdsourced data as an early warning system for public health.

Additionally, the Council to Improve Foodborne Outbreak Response (CIFOR) included third-party complaint systems in its 2019 guidelines, paving the way for health departments to leverage sites like IWP (though not mentioning by name, the practice is reflected in surveys).

Legislative/Legal Documents

While no U.S. federal law explicitly references IWasPoisoned, it has surfaced indirectly in legal contexts. The U.S. Department of Justice's 2020 case against Chipotle (which resulted in a $25 million fine for outbreaks) acknowledged that hundreds of customers across the country got sick from 2015–2018. Media coverage of that case noted IWasPoisoned's role in bringing some outbreaks (e.g. Simi Valley and Powell, OH norovirus incidents) to light before official case counts were available. In effect, IWP has been indirectly credited in regulatory narratives as whistleblower data, though official DOJ/FDA documents focus on the company's violations rather than the reporting mechanism.

Legal and Policy Analysis

Law Review and Policy Papers

Legal scholarship on food safety and data analytics has discussed IWasPoisoned in the context of regulatory innovation, public health law, and liability:

Daniel E. Ho (2019) – Food Safety Magazine (Dec/Jan 2020) – though not a law review, this policy-oriented piece by a law professor distilled lessons for AI in food safety. It explicitly references IWasPoisoned.com as one approach among "scores of ideas and pilots" using big data for foodborne disease surveillance. Ho notes that health departments in multiple cities have experimented with Twitter, Yelp, 311 calls and IWasPoisoned for surveillance. The article's policy context emphasizes that while media headlines have touted these tools as revolutionary, one must guard against "big data hubris," ensuring that crowdsourced reports are used to complement – not fully replace – official inspections. (This serves as a policy caution, advocating for evidence-based integration of platforms like IWP into enforcement.)

World Bank/IFPRI "Voices" (2019) – Policy blog: "How crowdsourcing can improve food safety." Co-authored by World Bank's Colin Finan and IWP's Patrick Quade, this essay (published in English, Spanish, Russian, Chinese) positions IWasPoisoned as a successful public-private approach to strengthening food safety in developing economies. It provides case examples (Chipotle outbreaks identified; E. coli in Tennessee well water; norovirus in Connecticut) to illustrate impact. Policy context: the authors argue that empowering consumers to report illnesses can overcome surveillance blind spots in countries where official reporting is limited. The blog encourages policymakers to leverage citizen data and invest in digital platforms similar to IWP, noting challenges like ensuring report verification and government responsiveness.

Think Tank Reports (2018–2021): Various food safety think tanks and advocacy groups have featured IWasPoisoned. For instance, Stop Foodborne Illness (non-profit) published a 2018 interview with Patrick Quade highlighting IWP's role in industry transparency and accountability. Quade explained how aggregating consumer reports can motivate restaurants to improve. Similarly, a 2020 digital epidemiology panel cited IWasPoisoned as a practical example in food systems. These policy discussions collectively view IWP as a model of civic tech improving food safety, while also probing how regulators can formally incorporate such data without being misled by noise or malicious inputs.

Media and Industry Coverage

News Media (General)

IWasPoisoned's crowd-sourced outbreak reports have drawn significant media attention, especially when high-profile incidents were involved. News articles often discuss IWP in the context of specific outbreaks or as a new trend in food safety:

The New York Times (Feb 13, 2018) – "Too Much Power to the People? A Food Safety Site Tests the Limits." A prominent feature that branded IWasPoisoned.com as a "controversial powerhouse in the restaurant industry". The NYT story (by Kevin Roose) recounts how IWP flagged a 2017 Chipotle norovirus outbreak in Sterling, VA, sickening over 100 people and tanking the stock. It explores tension points: restaurants worrying about false reports or panic, versus the platform's success in alerting the public quickly. Cornell food safety professor Martin Wiedmann is quoted, acknowledging IWP's value but urging caution and verification. This widely read piece raised IWP's profile nationally, highlighting both its consumer protection role and the potential "limits" of crowdsourced power (e.g., need for quality control, as IWP's moderation practices were described in detail).

NPR – The Salt (Nov 30, 2017) – "'I Was Poisoned': Can Crowdsourcing Food Illnesses Help Stop Outbreaks?". An in-depth NPR article by Jill Neimark, covering Patrick Quade's backstory (a personal bout of food poisoning in 2008 spurred him to start IWP) and the platform's growth to 75,000 reports in 46 states by 2017. It presents IWasPoisoned as an innovative early-warning system, including examples of health departments acting on IWP alerts. It also notes downsides – e.g., misattribution: people might mistakenly blame the last restaurant they ate at (an issue food safety experts like attorney Bill Marler raised). NPR's balanced take concluded that crowdsourcing can indeed help if used carefully, reinforcing the site's emerging legitimacy in outbreak prevention (and even quoting skeptics to underline the need for confirmatory evidence).

Business Insider (Aug 2, 2018) – "More than 400 people say they got sick after eating at a Chipotle….". This piece by Hayley Peterson reported on a major outbreak in a Chipotle in Ohio, citing IWasPoisoned.com data as the leading indicator: at the time of reporting, 228 IWP user submissions indicated 418 people ill, far above what had been officially logged. Business Insider noted that IWP saw a surge in reports immediately after news of the closure broke, demonstrating a feedback loop between media and crowdsourced reporting. The article treated IWP as a credible source, even directly quoting founder Patrick Quade on the stats. Subsequently, many outlets (CNN, USA Today, local AP wires) picked up similar angles, solidifying IWP's reputation as the go-to real-time outbreak tracker for journalists covering restaurant incidents.

Food Safety News (2017–2023) – The food industry trade journal Food Safety News has featured IWasPoisoned in various contexts:

Television and Local News: Local TV affiliates have run consumer-focused stories on IWasPoisoned, often titled along the lines of "Website tracks food poisoning reports at restaurants". For example, WFTV (ABC Orlando) and ActionNewsJax (Florida) in 2018 demonstrated the site on air, explaining how diners can check if others got sick at a given restaurant. These pieces usually present IWP as a public service tool and mention that health departments "frequently work closely" with the site when clusters appear. Such coverage has helped drive public awareness and reporting participation, including outside the U.S., by appearing in syndicated segments and online videos.

Industry and Trade Publications

Restaurant industry media have noted IWasPoisoned as both a risk factor and an ally for businesses, reflecting a nuanced view from the industry perspective:

Nation's Restaurant News (July 21, 2017) – "Food poisoning website to restaurants: 'We're an ally'." A profile of IWP by Jonathan Maze, describing how the platform actually aims to help restaurants by sharing data with them and preventing large outbreaks. It recounts Quade's founding story and emphasizes that IWP works with chains to alert them quickly to any red flags, positioning the site as a partner in food safety rather than just a "gotcha" for bad PR. This article is notable for mentioning that IWasPoisoned had already been credited with catching a Chipotle outbreak (2017 Virginia) and includes industry voices who recognize its value. The "ally" framing suggests that forward-thinking restaurant executives were beginning to subscribe to IWP's data feeds to manage risk proactively.

Restaurant Business Online (2019) – featured IWasPoisoned in an article on tech trends in food safety, highlighting that many major fast-food chains and food service companies were quietly subscribing to IWP's alert system. It described how, for instance, Chipotle and Starbucks corporate food safety teams monitor daily reports from the site. The tone was that in the age of social media, ignoring crowdsourced complaints is perilous, and embracing them can help companies respond before issues escalate. (This industry acceptance is also evidenced by IWP's claim of 17,000 subscribers to its free alert service by 2017.)

Impact and Citation Mapping

Impact Summary

Over the past decade, IWasPoisoned.com has transitioned from a niche website to a recognized component of the food safety ecosystem. Its impacts include:

(1) Earlier Outbreak Detection – Numerous outbreaks (especially norovirus incidents in restaurants) have been identified faster or exclusively through IWP reports, as documented in case studies (Chipotle 2015–2018, etc.) and academic correlations.

(2) Public Awareness and Engagement – By giving consumers a reporting outlet, IWP has raised awareness that reporting matters; media stories and user testimonials credit the site for empowering diners and pressuring restaurants to address food safety lapses.

(3) Integration with Officials – Health departments in almost every U.S. state now monitor or receive IWP data, and federal agencies like FDA are actively experimenting with it. This public-private data sharing represents a new model for foodborne illness surveillance.

(4) Industry Changes – Major restaurant chains have invested in monitoring tools (some via DineSafe, IWP's business service) to get ahead of outbreaks. The fact that IWP has been called an "ally" by industry media and is used in corporate risk management indicates a cultural shift: what was once seen as a threat (public reporting of illness) is now also viewed as a diagnostic tool for quality control.

Forward Citations (Post-2017)

The 2017 Quade & Nsoesie paper – the first peer-reviewed study on IWasPoisoned – has been cited by a variety of subsequent research. It appears in general reviews of digital epidemiology and machine learning (e.g., Albuquerque et al. 2019 Annual Review), demonstrating that IWP is considered a key example of crowdsourced health data in food safety. It's also cited in applied studies: for instance, Li et al. (2022) on Florida outbreaks used Quade & Nsoesie to justify using IWP data, and Krause et al. (2021) acknowledged IWP's model of curated consumer reports when blending those data with e-commerce reviews. In Participatory Surveillance research (e.g., related to event-based surveillance during mass gatherings like the 2016 Rio Olympics), IWP is referenced as a parallel to apps like Flu Near You, indicating cross-domain influence. Forward citations also include policy literature: Ho (2019) in Food Safety Magazine cites Quade & Nsoesie (2017) to contrast academic enthusiasm with on-the-ground realities. In summary, forward citations show IWasPoisoned being used as a proof-of-concept for crowdsourced outbreak reporting, an data source for novel detection algorithms, and a case study in public health innovation.

Backward Citations (Pre-2009 influences and references within seminal works)

The concept of crowdsourcing food safety did not exist pre-2009 in its current form, but IWP's development drew on related threads. The 2017 JMIR article's references include: global disease burden of foodborne illness (WHO 2015) to emphasize underreporting; traditional complaint-based surveillance literature (e.g., Hedberg et al. 2008 on complaint effectiveness) to which IWP is an adjunct; and early digital detection efforts (like NYC's use of Yelp reviews and Las Vegas's Twitter project) as precedent. Notably, IWP's approach is conceptually rooted in ProMED-mail (the 1990s listserv for global disease alerts) – Ben Chapman explicitly connected the dots in 2023: ProMED was the first crowdsourced public health platform, and IWasPoisoned follows that model for food illnesses. Within law review circles, earlier scholarship on "participatory sensing" and 311 systems (e.g., civic complaint lines) laid groundwork for viewing IWP as an extension of citizen reporting mechanisms to public health. Thus, while IWasPoisoned was novel, its bibliography reveals it was inspired by a tapestry of work on outbreak surveillance, consumer empowerment, and internet-based reporting.

Areas for Continued Development

While IWasPoisoned has demonstrated significant impact, the literature identifies several areas where continued research and development could enhance the platform's effectiveness:

Data Validation and Quality Assurance: A consistent theme across academic and media coverage is the importance of maintaining data accuracy as the platform scales. IWasPoisoned currently addresses this through manual curation, with each report reviewed by expert teams. However, as volume increases, there are opportunities to develop more systematic validation approaches. Recent research comparing IWP data against lab-confirmed outbreak datasets (such as the NoroSTAT correlation study) shows promising alignment, but additional peer-reviewed evaluations of sensitivity, specificity, and timeliness could further strengthen the platform's scientific foundation.

Demographic Representation and Reporting Patterns: Current studies indicate that younger adults and women are overrepresented among platform users. While this provides valuable data from these populations, research suggests there may be underrepresentation of certain communities or types of food establishments due to varying reporting behaviors. Future work could explore strategies to broaden participation across diverse demographic groups and develop analytical approaches that account for these patterns in outbreak detection algorithms.

Public Health System Integration: Although over 500 health departments covering 90% of the U.S. population utilize IWasPoisoned data, integration approaches vary significantly across jurisdictions. Developing standardized protocols and technical solutions for seamless data integration into existing surveillance workflows could enhance the platform's utility for public health agencies while maintaining data quality and security standards.

International Implementation and Cultural Adaptation: Most published research and implementation examples focus on the U.S. context. While IWasPoisoned has developed interfaces in seven languages, limited documentation exists regarding effectiveness in different cultural and regulatory environments. Systematic research on international implementation, including adaptation strategies for varying food safety systems and reporting cultures, could inform global expansion efforts.

Policy Framework Development: The widespread informal adoption by health departments highlights the need for formal policy guidance on incorporating crowdsourced data into food safety enforcement. Current use varies by jurisdiction, and establishing evidence-based standards for data utilization, investigation triggers, and integration with traditional surveillance methods could help maximize the platform's public health impact while ensuring appropriate procedural safeguards.

In conclusion, the body of publications and documents from 2009–2025 reflects growing validation and adoption of crowdsourced food safety surveillance via IWasPoisoned.com. Academic studies, government programs, and media reports collectively recognize that harnessing citizen reports can significantly enhance outbreak detection and public awareness. At the same time, they underscore the importance of careful curation, cross-verification, and policy support to fully realize the benefits while managing the limitations. IWasPoisoned's trajectory – from a single-user idea to a global data stream referenced by scholars and FDA officials alike – illustrates a critical shift toward participatory epidemiology in food safety. The comprehensive references above provide a roadmap of that journey, documenting both the enthusiasm for this novel approach and the continued need for rigorous analysis and integration.