Partnership Demo and Whitepaper

(Download whitepaper)

Poison control is an unsung public health triumph: it has been quietly saving lives and healthcare dollars for decades by preventing unnecessary emergency department (ED) visits, reducing health care expenses, and improving poisoning outcomes.^1–11 Even so, poison control phone call volume has been decreasing in recent years,¹² while unintentional poisoning has become the leading cause of injury death in the US.¹³ Traditionally, poison control has been accessible only by telephone, until the development of webPOISONCONTROL^®.

The following problems and needs led to the development of webPOISONCONTROL:

Limited awareness of poison control and reluctance to call: Although poison control hotlines are free, available 24/7/365, confidential, and staffed by toxicology experts, there is a lack of public awareness of poison control services and an increasing unwillingness to call.¹⁴ More and more people prefer to use the internet over the telephone to access health information.^15–21 Almost 6,000 people call US poison centers for guidance each day, yet an estimated 4,000 to 9,000 others experience a poison exposure but do not call, thereby losing out on the health and cost benefit of poison control guidance.²²

Inaccurate & unsafe information online: Inaccurate and unsafe poisoning guidance abounds on the internet.^16,23,24 webPOISONCONTROL provides a trusted online resource developed by toxicology experts, the same experts that direct US poison control centers. 

The need to reduce unnecessary healthcare costs even more than the savings already realized from traditional poison control phone services: As with telephone calls to poison control, most webPOISONCONTROL cases are managed at home, saving the cost and burden of unnecessary ED visits and freeing up specialists who answer the poison control hotline to manage the more serious and clinically complex cases, which are on the rise.¹² However, compared to a traditional phone-based service, it costs much less to manage a poison exposure case online, and webPOISONCONTROL can manage even a 10-fold volume increase at a minimal additional cost.

Poison centers struggle to secure adequate, reliable funding: Many people do not realize that poison control is not a government entity. Poison centers depend on patchwork funding from a variety of sources and are chronically under-funded.^25–27 webPOISONCONTROL cuts costs without cutting services, managing poisonings online at 1/3 the cost of the personnel-intensive phone-based method without compromising the quality of care, all while remaining free to users.

webPOISONCONTROL is the product of a collaboration of experts from poison control centers led by the National Capital Poison Center [NCPC; a 501(c)3 nonprofit organization)] and a 17-member software development team. Today, webPOISONCONTROL manages around 160,000 cases per year from all over the US, and even internationally. User feedback is overwhelmingly positive, and the project was a top 10 finalist for the 2018 Drucker Prize for Nonprofit Innovation and a semifinalist for the 2020 National Safety Council Green Cross for Safety Innovation Award. It was also featured in The Washington Post.²⁸
 

Who are we?

webPOISONCONTROL is brought to you by a group of forward-thinking poison control centers. It is the first fully automated virtual poison center. The logic, algorithms, and recommendations that power the tool are written by board-certified toxicology experts, each with decades of experience in poison control. As of December 2020, 24 accredited U.S. poison centers are participating in the webPOISONCONTROL project.

What is webPOISONCONTROL?

webPOISONCONTROL is an online tool and app that determines the appropriate triage for suspected poisonings without having to call poison control.

To get a personalized recommendation, users are prompted to enter the substance, amount, age, and time since the poison exposure. They can also scan or enter a unique product code (UPC, a.k.a. barcode) or type in a pill imprint, helping ensure substance accuracy.

Upon answering a few brief questions about the exposure, users immediately receive one of three triage recommendations:

1. it is safe to stay home because toxicity is minimal,
2. ED evaluation is required, or
3. a call to poison control is warranted because further information is needed to evaluate the case.     

If the tool determines the case can be managed at home, the user is given instructions for treatment and observation, specific symptoms likely to occur (yet not of concern), and symptoms that should trigger a call to poison control or an ED visit, if they develop. The user receives a copy of the detailed recommendation by email and automated follow-up ensues. During automated follow-up, the user chooses from a list of specific symptoms that developed, if any. The symptoms are compared to the worrisome effects of the poison, triggering a change in the triage recommendation if indicated. Follow-up reminders are sent at intervals appropriate to the kinetics of the implicated product or substance.

Real-Time Surveillance Dashboard

The webPOISONCONTROL surveillance dashboard provides a mechanism to visualize poison exposure data and trends in near real-time, focusing on a single product, group of products, or brand family, compared to a reference group of the user's choice. Data are from cases from 2017 up to midnight on the day prior to the analysis. As of December 2020, webPOISONCONTROL has managed more than 500,000 cases.

The following data points are viewable in the dashboard:

For a more detailed surveillance dashboard description with screenshots, please see Appendix B in the downloadable whitepaper (see PDF link, at the top of this page).

What is webPOISONCONTROL's source of product information?

We used multiple sources of product information to build webPOISONCONTROL's robust substance database and continue to add products and UPCs daily. We have licenses to use information from a few different proprietary databases from which we can also pull in barcodes, but we also use a lot of publicly available information, such as product labels and a drug database for pharmaceutical ingredients. We also frequently reach out to product manufacturers to ask for product ingredients directly, and usually once they understand why we need the information, how we use it, and who has access to it, they are more than happy to help us in the interest of public safety. Our preferred method is obtaining product information directly from the manufacturer because it is the most accurate. Having accurate information on product ingredients prevents us from having to assume a worst-case scenario for a potentially toxic ingredient's concentration in a product. When we are forced to assume a worst-case scenario, it leads to unnecessary excess referrals to emergency rooms and unnecessary extra calls to poison control. Neither public users of webPOISONCONTROL nor our partners or any other company has access to proprietary product information.

How does webPOISONCONTROL determine whether a poison exposure is serious?

webPOISONCONTROL is powered by more than 2,083 ingredient-based algorithms, each matched to the corresponding ingredient(s) in 110,428 products and substances. As of December 2020, 844,867 product barcodes are linked to these 110,000+ substances. More algorithms, products and barcodes are added daily. The application core, the ingredient algorithms, provide age- or weight-based thresholds for each ingredient. Algorithms also outline the justification for the threshold, list the expected minor symptoms and the symptoms which require further medical evaluation, recommend specific home treatment where appropriate, define the onset and duration of symptoms, and set a risk window beyond which significant toxicity is unlikely if clinical manifestations have not already begun. Special logic is incorporated to handle each formulation type, multi-ingredient products, multiple routes (swallowed, eye, skin, inhalation, bite/sting, injection), unknown amounts, unknown weight, and the minimum possible weight for each age. Users are encouraged (but not required) to provide an email address to receive a copy of the case and recommendations. The email address also serves as the key to case follow-up. Users are emailed at intervals appropriate to the substance kinetics and urged, for their own safety, to follow a link to a follow-up module. That module gathers information on what was actually done (stayed home, went to ER, admitted to hospital, etc.) and what specific symptoms developed, if any. Symptoms are further evaluated and compared to worrisome effects of the poison exposure, triggering a change in the triage recommendation for the case if indicated.

How was webPOISONCONTROL developed?

In the early stages of development, our toxicologists worked with software UX experts to transform the unstructured poison control phone interview into a user-friendly but structured series of questions that would determine the triage recommendation for a specific poison exposure scenario. We focused on asking the fewest questions necessary for an individual case, asking those questions in the easiest way possible, and providing enough answer choices that the user would find one that fit their situation. User feedback guided the many subsequent revisions. From the users, we learned which questions needed to be simplified, omitted, made optional, or re-arranged, what new capabilities were needed, and what products and algorithms were missing.

While software developers implemented these revisions, toxicologists expanded the scope of the effort by:

• writing algorithms to cover more product ingredients;
• developing a database of household products, plants and creatures that bite or sting, to expand beyond pharmaceuticals;
• adding eye, skin, inhalation and bite/sting exposure routes;
• expanding to double dosing of medicines, a common reason for poison control utilization;
• adding barcode scanning capabilities so more specific recommendations can be provided and faster;
• developing an interface so participating poison centers could also use the product information and algorithms to facilitate telephone-based case management;
• developing analytics to measure usage, detect hazards, and track trends.

Outcomes

The pilot phase of webPOISONCONTROL was launched on December 30, 2014. An analysis published in The American Journal of Emergency Medicine demonstrates its safety, feasibility, and public acceptance.²⁹ Maintenance and development work have continued since the conclusion of the pilot phase.

Safety

When assessed in 2016, over 80% of the cases triaged to home with follow-up had no effect, with minor effects reported in 10.4% of cases and moderate effects in 1.4%. No major effects or deaths were reported. These results indicate the appropriateness of the triage thresholds that determine when it is safe for the exposed person to stay home, with follow-up. Twenty-six cases initially triaged to home (1.2% of 2,118 home-triage cases with follow-up) had a changed triage recommendation on follow up, with a subsequent deliberately cautious recommendation to call poison control (21 cases) or go to an ED (5 cases). Thus, the automated follow-up function of webPOISONCONTROL proved to be an important safety feature. Over the next 3 years following the conclusion of the pilot, poison control experts conducted next-day audits of 100% of 272,000 webPOISONCONTROL completed, nonduplicated cases, confirming the accuracy and safety of recommendations. More recent data show that severe outcomes are rare.

Feasibility and Public Acceptance

Between February 11 and 25, 2016, 1,339 online feedback survey respondents had used webPOISONCONTROL for a poison emergency, and 97.3% of respondents found it easy or somewhat easy to use; 91.3% found the time to get a recommendation was quick or very quick; and 95.8% found it met or more than met their needs. When considering all 15,000 survey responses collected to date, the tool continues to impress: 98% say the tool is easy or somewhat easy to use; 95% say it met or more than met their needs; 97% would likely or very likely use the tool again; and 96% say the speed is quick to very quick. On average, it takes 2½ minutes to enter one's information in webPOISONCONTROL and get a recommendation. Please see examples of real user feedback in the downloadable whitepaper (see PDF link, at the top of this page).

Cost Benefits

The financial impact of correct versus incorrect triage has never been directly measured, although at least a dozen studies estimate the return on investment for poison control services at $6 to $36 per dollar spent. One of the most comprehensive analyses shows an ROI of $13.39 in decreased healthcare cost and saved productivity for every dollar spent on a traditional poison control center.⁹ Nationally, it costs about $50.04/case for a traditional poison center to handle a case at home (without healthcare referral). It costs only 1/3 as much ($17.46) for webPOISONCONTROL to manage a case completely online (without referral to a healthcare facility or to a phone-based poison center). Unlike the personnel-intensive traditional phone-based poison center, the marginal cost for webPOISONCONTROL to handle 10 times the case volume is minimal. Thus, promotion of webPOISONCONTROL and subsequent increased use could markedly decrease the cost of providing poison control guidance to less than $2 per home-managed case. 

Other Outcomes

We have also observed the following, unforeseen benefits of webPOISONCONTROL:

• Disaster response: The tool has a tremendous disaster response capability. webPOISONCONTROL can easily manage surges in demand since it does not require dedicated 24/7 staffing, as does the call-based service. Between early March and the end of April 2020, webPOISONCONTROL case volume increased by 51%, with prominent spikes in exposures to hand sanitizing products, household cleaners, and disinfectants.
• Optimizing limited resources: webPOISONCONTROL frees up limited poison control resources by handling the less complicated cases online, allowing call center staff to dedicate more time to the increasingly complex, and often more clinically serious cases. However, webPOISONCONTROL will not eliminate the need for traditional poison control centers because its scope is focused, designed to handle unintentional poison exposures in a person who is not pregnant and has no underlying serious illness. Definitive guidance is only provided for cases that can safely be managed at home, without intervention by a healthcare provider. That leaves the more difficult cases – the complex, intentional, nuanced, or serious cases - to be handled by the human experts at traditional phone-centric poison centers. It also leaves the consults poison control centers provide for healthcare professionals caring for poisoned patients, education for health professional trainees, hazard and chemical/bioterrorism surveillance, and input into state and local responses to emerging substances of abuse, foodborne outbreaks, product hazards, preparedness and planning, all of which traditional poison control centers provide.
• Consistent care: The evidence-based algorithms created for this tool are based on case data and arduous scientific literature review. webPOISONCONTROL uniformly adheres to these algorithms, a phenomenon not observed in call-based poison centers because of insufficient numbers of algorithms and individual deviation from them where they exist. Adherence to guidelines enhances patient care, minimizes conflicting recommendations and patient confusion, and makes triage decisions more of a science than an art. As more centers (currently 24) use webPOISONCONTROL algorithms to guide the telephone management of poison exposures, the algorithms help eliminate regional differences in responses and harmonize triage guidelines nationally.

Opportunities for Improvement and Future Development

webPOISONCONTROL is a game changer. For the first time, poison control services are delivered in a fully automated manner, at a lower cost than the traditional phone-based service.

Future changes and ways webPOISONCONTROL can continue to improve include:

• Increased public usage of online poison control services by those who do not want to call. We plan to continue driving up utilization through continued SEO efforts, word-of-mouth, and social and traditional media coverage. We also hope to form strategic partnerships and secure endorsements from respected health- and parenting-related organizations, to increase awareness.
• Increased scope in terms of the addition of products, barcodes, images, and triage algorithms so we can provide definitive guidance to as many users as possible.
• Continued quality assurance. As more clinical and research data become available, we update algorithms to optimize safety and efficiency. We now audit 30% of completed, nonduplicated public cases (down from 100% that were audited from 2015 to Jul 2019). Weekly quality assurance processes are performed to capture and address any anomalies. We also closely monitor and act on user feedback.
• Optimizing triage thresholds. While it is important for the tool to be conservative in its advice, a threshold that is too low will result in too many users referred unnecessarily to EDs or directed to call poison control.
• Increased use by traditional phone-based poison centers leading to harmonized triage in the US.

Funding

$2 million/year annual budget covers maintenance and enhancements, including:

• $1.2 million/year software development and maintenance
• $800,000/year quality assurance, product entry, algorithm development, toxicologists, databases

Additional enhancement funding needed to expand the app's scope and features during the next 3 years:

• $450,000 would fund development of a pilot API (application program interface) to provide a seamless exchange of case information between webPOISONCONTROL and traditional poison control centers, facilitating continuity of recommendations.
• $250,000 would expand analytic capabilities, improving reporting, trending, and quality assurance at both the national and center level.
• $1,000,000 would substantially complete algorithm development, leaving only the ongoing addition of algorithms for unique ingredients in newly marketed products and continuous review of existing algorithms.
• $700,000 would substantially complete the product database, except for new products that would need to be added as they are introduced.
• $2,000,000 would keep the app operational for a full year.

Conclusion & Partnership Opportunities

In a time when unintentional poisoning is the leading cause of injury death in the U.S. and EDs are once again in danger of being overwhelmed by COVID-19 patients, poison control plays a critical role in the nation's public health response. webPOISONCONTROL is an innovative solution that expands the public's access to safe, accurate, and quick triage for poisoning, mitigates the burden of unnecessary ED visits, and renders the entire poison control enterprise more cost-efficient. In developing webPOISONCONTROL, the National Capital Poison Center, a 501(c)3 nonprofit organization, put forth the enormous time and resources required to meet the challenges of the evolving information age and has proven the power of innovation to effect change.

Now that we have shown what bringing poison control into the information age can do, we need your help.

What do we need?

It is important to understand that while in the context of poison control's long history webPOISONCONTROL may be new, it is not a "startup." It was launched just over 5 years ago and has been sustained ever since with philanthropic funding. It has been proven safe and easy to use and is beloved by those who have used it. Now, we feel it is time to diversify our partnership base and invite those who manufacture and/or sell the products and substances for which consumers are using webPOISONCONTROL to participate.

Contact Us

We hope you will consider contacting us to discuss a partnership model that works for your organization.

Kelly Johnson-Arbor, MD, FACEP, FUHM, FACMT Medical Toxicologist and Co-Medical Director National Capital Poison Center johnson-arbor@poison.org	Toby Litovitz, MD Director of webPOISONCONTROL National Capital Poison Center toby@poison.org
Jessica Youniss COO & CFO National Capital Poison Center youniss@poison.org	Krista Osterthaler, MPH Director of National Strategic Alliances National Capital Poison Center osterthaler@poison.org

References

1. King WD, Palmisano PA. Poison control centers: can their value be measured? Southern medical journal. 1991;84(6):722-726. 
2. Miller TR, Lestina DC. Costs of poisoning in the United States and savings from poison control centers: A benefit-cost analysis. Annals of Emergency Medicine. 1997;29(2):239-245. 
3. Phillips KA, Homan RK, Hiatt PH, et al. The Costs and Outcomes of Restricting Public Access to Poison Control Centers: Results from a Natural Experiment. Medical Care. 1998;36(3):271-280.  
4. Guyer B, Mavor A, Institute of Medicine Committee on Poison Prevention and Control. Forging a poison prevention and control system: report of an Institute of Medicine committee. Ambulatory pediatrics : the official journal of the Ambulatory Pediatric Association. 5(4):197-200.
5. Zaloshnja E, Miller T, Jones P, et al. The potential impact of poison control centers on rural hospitalization rates for poisoning. Pediatrics. 2006;118(5):2094-2100. 
6. Vassilev ZP, Marcus SM. The impact of a poison control center on the length of hospital stay for patients with poisoning. Journal of Toxicology and Environmental Health - Part A: Current Issues. 2007;70(2):107-110. 
7. Bunn TL, Slavova S, Spiller HA, Colvin J, Bathke A, Nicholson VJ. The effect of poison control center consultation on accidental poisoning inpatient hospitalizations with preexisting medical conditions. Journal of Toxicology and Environmental Health - Part A: Current Issues. 2008;71(4):283-288. 
8. Zaloshnja E, Miller T, Jones P, et al. The impact of poison control centers on poisoning-related visits to EDs--United States, 2003. The American journal of emergency medicine. 2008;26(3):310-315.
9. The Lewin Group. Final Report on the Value of the Poison Center System.; 2012.
10. Wyckoff AS. AAP had hand in first poison control center. AAP News. 2013;34(10):45-45.  
11. Tak CR, Malheiro MC, Bennett HKW, Crouch BI. The value of a poison control center in preventing unnecessary ED visits and hospital charges: A multi-year analysis. American Journal of Emergency Medicine. 2017;35(3):438-443. 
12. Gummin DD, Mowry JB, Spyker DA, Brooks DE, Osterthaler KM, Banner W. 2017 Annual Report of the American Association of Poison Control Centers' National Poison Data System (NPDS): 35th Annual Report. In: Clinical Toxicology. Vol 56. Taylor and Francis Ltd; 2018:1213-1415. 
13. About NCHS - NCHS Fact Sheets - NCHS Data on Drug-poisoning Deaths. Accessed November 17, 2019. 
14. Report to Congress, Poison Help Campaign, Fiscal Year 2012.; 2012. Accessed November 17, 2019. 
15. JSI. Poison Control Innovation/Transformation Project Market Research Report.; 2014. Internal report; not available online.
16. Cline RJW. Consumer health information seeking on the Internet: the state of the art. Health Education Research. 2001;16(6):671-692. 
17. Fox S, Duggan M. Health Online 2013 | Pew Research Center. Published 2013. Accessed November 17, 2019. 
18. Census Bureau U. Computer and Internet Use in the United States: 2016. Accessed November 17, 2019. 
19. Jacobs W, Amuta AO, Jeon KC. Health information seeking in the digital age: An analysis of health information seeking behavior among US adults. Cogent Social Sciences. 2017;3(1). 
20. Amante DJ, Hogan TP, Pagoto SL, English TM, Lapane KL. Access to care and use of the internet to search for health information: Results from the US National Health Interview Survey. Journal of Medical Internet Research. 2015;17(4):e106. 
21. Prestin A, Vieux SN, Chou WYS. Is Online Health Activity Alive and Well or Flatlining? Findings from 10 Years of the Health Information National Trends Survey. Journal of Health Communication. 2015;20(7):790-798.  
22. National Capital Poison Center. "What is webPOISONCONTROL?" Infographic Supporting Information. Accessed November 20, 2019. 
23. Cole J, Watkins C, Kleine D. Health Advice from Internet Discussion Forums: How Bad Is Dangerous? Journal of medical Internet research. 2016;18(1):e4. 
24. el Sherif R, Pluye P, Thoër C, Rodriguez C. Reducing Negative Outcomes of Online Consumer Health Information: Qualitative Interpretive Study with Clinicians, Librarians, and Consumers. Journal of medical Internet research. 2018;20(5):e169.
25. Woolf AD, Karnes DK, Kirrane BM. Preserving the United States's poison control system. Clinical Toxicology. 2011;49(4):284-286. 
26. Litovitz T, Kearney TE, Holm K, Soloway RA, Weisman R, Oderda G. Poison control centers: Is there an antidote for budget cuts? American Journal of Emergency Medicine. 1994;12(5):585-599. 
27. The State of Poison Control in the US. Accessed December 8, 2019. 
28. Blakemore E. Worried you ingested something deadly? This virtual poison control website can be a lifesaving tool. - The Washington Post. The Washington Post. Accessed November 17, 2019.  
29. Litovitz T, Benson BE, Smolinske S. webPOISONCONTROL: can poison control be automated? American Journal of Emergency Medicine. 2016;34(8):1614-1619.