Abstract
The National Poison Data System (NPDS) comprises self-reported information from people who call US poison center hotlines. NPDS data have proven to be important in identifying emerging public health threats. We used NPDS to examine records of people who had self-reported exposure to harmful algal blooms (HABs). Participating poison centers then contacted people who had called their centers from May through October 2019 about their HAB exposure to ask about exposure route, symptoms, health care follow-up, and awareness of possible risks of exposure. Of 55 callers who agreed to participate, 47 (85%) reported exposure to HABs while swimming or bathing in HAB-contaminated water. Nine callers reported health symptoms from being near waters contaminated with HABs, suggesting potential exposure via aerosolized toxins. Symptoms varied by the reported routes of exposure; the most commonly reported symptoms were gastrointestinal and respiratory. More public and health care provider education and outreach are needed to improve the understanding of HAB-related risks, to address ways to prevent HAB-related illnesses, and to describe appropriate support when exposures occur.
Harmful algal blooms (HABs) occur in water bodies around the world and can adversely affect those who use these water bodies for recreation, agriculture, or drinking. 1 People can be exposed to HABs or HAB toxins when they swim or wade in contaminated water, eat contaminated fish or shellfish, or use contaminated drinking water. 1 Exposure can also occur through aerosolized HAB toxins when one is near a HAB-contaminated water site.2 -5 Using the National Poison Data System (NPDS), we examined information collected from callers who contacted 1 of the 5 participating US poison centers from May through October 2019 about HAB-related exposures. Participating poison centers contacted callers to ask about exposure route, symptoms, health care follow-up, and public health communications around HABs.
Methods
Five regional poison centers participated in this review: Florida/US Virgin Islands Poison Information Center–Jacksonville (Jacksonville, Florida), Rocky Mountain Poison and Drug Safety (Denver, Colorado), Central Ohio Poison Center (Columbus, Ohio), Utah Poison Control Center (Salt Lake City, Utah), and Washington Poison Center (Seattle, Washington). People who called any of the 5 poison centers to report HAB exposure and an associated illness from May through October 2019 were considered for inclusion in our analyses. NPDS coding for HABs does not distinguish between marine HABs and freshwater cyanobacterial HABs; thus, we used the term HAB to refer to both types of events. We included callers if their cases were related to human exposure to HABs and if they presented with at least 1 clinical effect (ie, signs or symptoms). Animal-related calls, informational calls, and calls not pertaining to the clinical effects of HAB exposure were excluded.
We asked the participating poison centers to survey the callers who met our inclusion criteria. We allowed only adults aged ≥18 years to answer survey questions; if a child was the exposed individual, a parent or guardian answered on the child’s behalf. The survey was composed of 14 questions on exposure to HABs and related health symptoms. Participants were asked about the source or sources of exposure (bathing/swimming in contaminated waters, being near contaminated waters, drinking water contaminated with toxins, or eating contaminated fish or shellfish), symptoms temporally related to the exposure, any medical treatment received to alleviate symptoms, information provided to them by the poison center, and any health messaging received before or during the exposure. Basic demographic information (age and sex) and route of exposure had been collected in NPDS at the time of the original call, and we merged these data with the questionnaire dataset. We instructed the poison centers to attempt at least 3 follow-up calls before being considered lost to follow-up.
We used descriptive statistics to summarize the counts and percentages of responses to the questionnaire items (SAS version 9.4; SAS Institute Inc). We compared data by the source of exposure to highlight any differences between exposure routes. We created symptom categories from the symptoms reported on the questionnaire. We used Fisher exact tests to test the equality of proportions between exposure groups, with P < .05 considered significant. The Centers for Disease Control and Prevention National Center for Environmental Health Office of Science determined that this study was not human subjects research and waived formal ethical review.
Results
The 5 poison centers contacted 99 people who had called about HAB exposure from May through October 2019. Of 99 callers, 55 (56%) agreed to answer survey questions. Of 55 respondents, 38 (69%) had originally contacted the Utah Poison Center. Of 53 respondents, 29 (55%) had reported symptoms from exposure to HABs involving children or adolescents (Table 1). Forty-two (76%) stated that exposure to HABs occurred through multiple routes, with the 2 most common being dermal contact (n = 47, 85%) and unintentional ingestion (n = 33, 60%). Ingestion was always coded with dermal contact.
Characteristics of callers who self-reported harmful algal bloom exposure to participating poison centers, May–October 2019 a
Participating poison centers were Florida/US Virgin Islands Poison Information Center–Jacksonville, Rocky Mountain Poison and Drug Safety, Central Ohio Poison Center, Utah Poison Control Center, and Washington Poison Center.
All values are number (%) unless otherwise indicated.
Individuals could report >1 source or route of exposure.
The most frequently reported source of exposure was bathing or swimming in waters contaminated with HABs (n = 47, 85%). The only other stated source of exposure was being near waters contaminated with a HAB (n = 9, 16%). One respondent reported exposure from swimming and being near contaminated waters. Most exposures were in or near freshwater sources (n = 49, 89%).
Fifty-four (98%) respondents indicated that they called a poison center because they or their children felt ill. Thirty-nine (71%) were worried about HAB exposure, and 37 (67%) wanted to know more about HAB exposure. Twenty-five (45%) stated that the water they were near or were swimming or bathing in was unusually colored.
Among the 47 people who were exposed through swimming or bathing in HAB-contaminated waters, 35 (74%) reported having >1 symptom, and 17 (36%) indicated having symptoms in >1 symptom category (Table 2). The most common symptom category was gastrointestinal (n = 27, 57%). Among 45 respondents to the question on timing between HAB exposure through swimming or bathing and symptom onset, 10 (22%) indicated symptom onset <6 hours after exposure; 21 (47%), 6-24 hours after exposure; and 14 (31%), >24 hours after exposure.
Frequency of symptoms among people who self-reported harmful algal bloom (HAB) exposure to participating poison centers, May–October 2019 a
Participating poison centers included Florida/US Virgin Islands Poison Information Center–Jacksonville, Rocky Mountain Poison and Drug Safety, Central Ohio Poison Center, Utah Poison Control Center, and Washington Poison Center. One caller reported exposure to HABs through swimming/bathing and being near to water with HABs and different symptoms based on the exposure type.
Gastrointestinal symptoms included nausea, vomiting, diarrhea, and abdominal pain; upper respiratory symptoms included respiratory irritation, asthma attack, watery eyes, throat irritation, burning sensation in eyes, and trouble breathing; dermal symptoms included rash, itchiness, blisters on skin, and burning sensation on skin; neurologic symptoms included headache; and other symptoms were those not fitting into the other categories. Headache was not included in the questionnaire symptom list but was reported often enough in the “other” category to be included as a symptom category.
P value generated from Fisher exact tests. P < .05 was considered significant.
Among 9 individuals who were exposed to HABs by being near contaminated waters, 8 had >1 symptom, and 7 had symptoms in >1 symptom category. The most common symptom category was upper respiratory (n = 8). All 9 of those who indicated being near waters contaminated with HABs had symptoms immediately to <6 hours after exposure.
Common advice given by the poison center was to keep hydrated, shower, and see a health care provider immediately or if the individual’s condition worsened. Of 55 people surveyed, 12 (22%) stated that they or their children went to a health care facility. When asked about treatments or tests received, respondents had a variety of responses, including receiving medications, computed tomography or magnetic resonance imaging scans, and blood tests.
Of 55 respondents, only 13 (24%) reported hearing or reading warnings about the dangers of HABs just before or during the exposure, whereas 5 reported hearing from friends or family, 3 from watching television, 3 from signs posted near bodies of water, 2 from newspapers or magazines, 2 from a recreational or utility website, and 8 from other sources (including social media), with some hearing from multiple sources. Of 55 respondents, 34 (62%) did not know what exposure prevention methods were in place at the exposure site before the exposure (eg, warning signs posted near bodies of water).
Discussion
NPDS has been an important source of information for emerging public health threats, including poisonings from lamp oil and detergent pods and the abuse of tianeptine, a drug not authorized for use in the United States.6 -8 We found that NPDS also provided information about another important emerging health issue: exposure to HABs.
Among 55 people who were interviewed about HAB exposure, 67% had called the Utah Poison Center. In 2019, the first harmful blooms in Utah Lake were reported in early June. 9 The Utah Department of Environmental Quality website noted cell densities (range, 20 000–10 million cells/mL) and microcystin levels (4-2000 µg/L) that indicated a moderate probability of acute health risk, prompting a warning advisory for some parts of the lake.10,11 By September 16, 2019, toxin levels ranged from <2.0 to 2.8 µg/L of microcystin and from 0.11 to 0.21 µg/L of anatoxin-a, which prompted a lake-wide warning advisory. Harmful blooms affected other lakes in central Utah, resulting in beach closures. 10 We suggest that calls about HAB exposures in Utah were prompted by Utah’s monitoring activities combined with efforts to reach out to its communities with cyanobacterial toxin–related messaging. Warning and closure signs displayed the poison center help line for those concerned about exposure. These public health efforts in Utah and the other 4 states could explain why many people self-reported HAB exposure but fewer than half (45%) indicated that they noticed water discoloration.
We anticipated that the most commonly reported health symptoms would be gastrointestinal, presumably from swallowing water, and respiratory, presumably from breathing aerosols, as noted in the literature.2 -4,12 -15 The reports of headaches (neurologic effects) suggest that there may be exposures to toxins that we had not considered. The self-reports of neurologic effects are somewhat consistent with news reports of canine poisonings that occurred after dogs ingested benthic cyanobacterial mats (anchored to rocks or sediment), although these events resulted in severe neurologic effects and fatalities. 16 The cyanobacteria composing the mats produced the potent neurotoxins (anatoxin-a and its derivatives). The presence of benthic mats may also explain why 55% of respondents did not report discolored water. Although 45% in our study stated that the water associated with their complaints was discolored, we suggest that this visual clue may not have dissuaded them from entering the water. Periodically reviewing NPDS data may provide additional guidance on what to look for in the environment and what health effects to anticipate.
Future public outreach and educational materials on HABs could be refined. For example, outreach materials could emphasize water color as a warning sign and explain that clear water is not necessarily free from risks. Current information suggests that benthic blooms with the potential to produce anatoxin-a thrive in “pristine” waters. 17 Public health messaging could encourage people who see discolored or unusually colored waters, detect smells of rotting plants, or have read reports about risks from using their streams and lakes to contact their local public health or environmental health department or poison center to inquire about posted health advisories and recent environmental monitoring.
Our study had 4 limitations. First, the survey response rate was 56%, which may have led to bias in the results; however, callers to poison centers who did not respond to the survey, as shown in the NPDS data, were similar to survey respondents in sex and age. Second, although callers were self-reporting symptoms related to presumed HAB exposure, the person may not have been exposed to HABs. Gastrointestinal complaints may have been from waterborne bacteria or parasitic infections rather than from HAB toxins. Third, some of the treatments (eg, antibiotics) suggest that the health care provider considered a diagnosis other than exposure to HABs or HAB toxins. Finally, because individuals were contacted to answer questions on exposures and symptoms a few months after their original calls to the poison center, data may have been subject to recall bias.
We suggest that people had identified a potential association between an environmental exposure (eg, recreational waters) and a subsequent health outcome, resulting in exposed individuals seeking further information from a known entity, such as a poison center. More public and healthcare provider education and outreach are needed to improve the understanding of the risks posed by HABs, prevent HAB-related illnesses, and provide support when exposures occur. We suggest that public health information obtained from NPDS can be improved by refining the product codes (exposure categories) to differentiate harmful marine algal blooms from harmful freshwater cyanobacterial blooms and to separate the different forms of cyanobacterial blooms (benthic, surface).
Footnotes
Correction (July 2023):
Article updated to add a co-author.
Disclaimer
The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention (CDC).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received the following financial support for the research, authorship, and/or publication of this article: Non-CDC authors received financial support for data collection activities through a cooperative agreement (National Poison Data System Annual Progress Report and Evaluation and Performance Measurement Plan cooperative agreement no. 1UE1EH001314).
