Table 3.
Topic number | Salient terms (top 10 in italics) | Label based on salient terms | Label based on sample tweets |
T1 | “people,” “children,” “kids,” “covid,” “long,” “get,” “many,” “risk,” “vaccinated,” “deaths,” “know,” “cases,” “vaccine,” “schools,” “death,” “even,” “also,” “still,” “die,” “amp,” “term,” “vaccines,” “young,” “school,” “delta,” “health,” “protect,” “masks,” “spread,” and “immunity” | Long COVID in people, including children, in the context of vaccination | More objective perspectives: research, monitoring, experience, and news items highlighting significant impact of symptoms of long COVID in adults and children, including potential benefits of vaccination in reducing long COVID risk |
T2 | “day,” “feel,” “like,” “back,” “days,” “year,” “got,” “time,” “pain,” “last,” “week,” “feeling,” “months,” “since,” “one,” “better,” “friend,” “ago,” “still,” “felt,” “work,” “went,” “hope,” “years,” “body,” “go,” “march,” “going,” “first,” and “today” | Duration and suffering associated with long COVID | More subjective perspectives: suffering and frustration associated with long COVID |
T3 | “symptoms,” “19,” “covid,” “long,” “study,” “post,” “patients,” “term,” “infection,” “haulers,” “new,” “covid19,” “months,” “syndrome,” “coronavirus,” “fatigue,” “acute,” “effects,” “common,” “studies,” “viral,” “via,” “persistent,” “reported,” “sars,” “found,” “weeks,” “experiencing,” “brain,” and “lingering” | Persistent symptoms of long COVID | Persistent symptoms of long COVID, with greater focus on formal and informal advocacy and awareness-raising around long COVID |
T4 | “longcovid,” “mecfs,” “research,” “please,” “cfs,” “support,” “amp,” “pwme,” “help,” “thank,” “patients,” “community,” “need,” “share,” “us,” “longcovidkids,” “funding,” “sign,” “patient,” “awareness,” “join,” “millionsmissing,” “myalgicencephalomyelitis,” “treatments,” “illness,” “treatment,” “guidelines,” “nicecomms,” “needs,” and “thanks” | Need for research on long COVID treatment | Calls to address the plight of people with long COVID, either by researchers or by governments, or frustration at having been neglected, and frustration at having ongoing symptoms. |
T5 | “fewer,” “ontario,” “vast,” “matters,” “measure,” “amongst,” “likelihood,” “possibility,” “polio,” “contract,” “ha,” “tracking,” “victims,” “de,” “clinically,” “eg,” “vs,” “admitted,” “telegraph,” “exposure,” “estimate,” “fda,” “surge,” “impacted,” “reducing,” “70,” “american,” “ages,” “yo,” and “largely” | Measuring long COVID symptoms | The large number of people affected by, or projected to be affected, by long COVID symptoms, and associated societal impacts. |
aDerived from topic modeling of Twitter content from 2021, with labels for each topic based on human interpretation of keywords and associated sample tweets. This table shows the topics identified, the salient terms associated with each topic, and the final labels given to each topic based on the interpretation of sample tweets.