A Comparative Analysis of a Self-Reported Adverse Events Analysis after Receiving One of the Available SARS-CoV-2 Vaccine Schemes in Ecuador

The COVID-19 pandemic has put a lot of pressure on health systems worldwide. Mass vaccination against SARS-CoV-2 has reduced morbidity and mortality worldwide. Despite their safety profiles, vaccines, as with any other medical product, can cause adverse events. Yet, in countries with poor epidemiological surveillance and monitoring systems, reporting vaccine-related adverse events is a challenge. The objective of this study was to describe self-reported vaccine adverse events after receiving one of the available COVID-19 vaccine schemes in Ecuador. A cross-sectional analysis based on an online, self-reported, 32-item questionnaire was conducted in Ecuador from 1 April to 15 July 2021. Participants were invited by social media, radio, and TV to voluntarily participate in our study. A total of 6654 participants were included in this study. Furthermore, 38.2% of the participants reported having at least one comorbidity. Patients received AstraZeneca, Pfizer, and Sinovac vaccines, and these were distributed 38.4%, 31.1%, and 30.5%, respectively. Overall, pain or swelling at the injection site 17.2% (n = 4500) and headache 13.3% (n = 3502) were the most reported adverse events. Women addressed events supposedly attributable to vaccination or immunization [ESAVIs] (66.7%), more often than men (33.2%). After receiving the first dose of any available COVID-19 vaccine, a total of 19,501 self-reported ESAVIs were informed (87.0% were mild, 11.5% moderate, and 1.5% severe). In terms of the vaccine type and brand, the most reactogenic vaccine was AstraZeneca with 57.8%, followed by Pfizer (24.9%) and Sinovac (17.3%). After the second dose, 6776 self-reported ESAVIs were reported (87.1% mild, 10.9% moderate, and 2.1% severe). AstraZeneca vaccine users reported a higher proportion of ESAVIs (72.2%) in comparison to Pfizer/BioNTech (15.9%) and Sinovac Vaccine (11.9%). Swelling at the injection site, headache, muscle pain, and fatigue were the most common ESAVIs for the first as well as second doses. In conclusion, most ESAVIs were mild. AstraZeneca users were more likely to report adverse events. Participants without a history of COVID-19 infection, as well as those who received the first dose, were more prone to report ESAVIs.


Introduction
Vaccination is the most cost-effective public health intervention after water sanitation for the control and management of preventable, contagious, and life-threatening diseases [1]. Vaccines induce artificial immunity to several types of microorganisms, avert  The Pfizer Phase III study included over 46,000 participants globally, and the results found 95.0% efficacy in preventing infection 7 days after the second dose administration in participants, with these results being pivotal in the FDA granting an EUA for the vaccine in the US.14 On 31 March 2021, Pfizer and BioNTech announced remarkable 100% vaccine efficacy in young adolescents aged 12-15 years, after phase 1/2/3 trials [6,7].
Followed by other companies, vaccine trials were starting to show communitarian benefits, and several countries started to vaccinate their populations [8][9][10] (Table 1).
In general, regardless of the immunologic mechanism used by any of the COVID-19 vaccines, serious adverse events are very rare, and most of the symptoms associated with the vaccines are mild [11]. For instance, in Poland, one ESAVI was reported for every 1100 vaccines applied [12]. Similarly, Beaty et al. reported that serious ESAVIs were very rare in a cohort of more than 19,000 patients in the U.S. [13]. Most ESAVIs are mild and self-limited and can present as systemic or local reactions. A wide spectrum of ESAVIs has been reported, with pain, redness, or swelling at the vaccine injection site, fever, fatigue, headache, muscle pain, nausea, vomiting, itching, chills, and joint pain being the most common [14,15]. In very rare cases, vaccines can also cause serious adverse reactions such as anaphylactic shock, vaccine-induced immune thrombotic thrombocytopenia [VITT], or even death, with a disputable rate of less than 0.0025% of deaths according to the Centers for Disease Control and Prevention [CDC] [16].
The identification and reporting of events supposedly attributable to vaccination or immunization (ESAVI) is a priority [18]. Although COVID-19 vaccines have shown numerous benefits, some people are still wary of COVID-19 vaccines [19]. This study aimed to describe the self-reported vaccine adverse events after the application of the available COVID-19 vaccines in Ecuador.

Study Design and Sample Selection
We conducted a cross-sectional study by circulating a 32-item, self-reported, online questionnaire through the internet-based survey platform Survey Monkey. We gathered anonymous responses from all over the country using a non-probability sampling method from 1 April to 15 July 2021.

Settings and Population
Participants were all Ecuadorian residents who received one of the available COVID-19 vaccines. According to the current government plan, 9 million people were vaccinated by July 2021. In that sense, with a confidence level of 95% and a margin of error of 1.5%, our minimum estimated sample was 4267 responses.
Using the expected population to be vaccinated, the sample size "n" and margin of error "E" are given by the following formula [20,21]: where "N" is the population size, "r" is the fraction of responses (50%), "Z(c/100)" is the critical value for the confidence level "c", "x" is the expected population, and Sqrt is the square root.
All responses that came from respondents who voluntarily agreed to participate in the study and who completed all 32 questions were included. Moreover, only participants that reported they received the mRNA vaccine (Pfizer/BioNTech), viral vector (AstraZeneca), or inactivated vaccine (Sinovac) were included.

Survey Development and Measures
The data were collected using a 32-item online anonymous questionnaire to evaluate self-reported symptoms and adverse events after receiving one or two doses of the available COVID-19 vaccines. Participants' consent was obtained at the beginning of the questionnaire with an explanation of the objective of the study. Participants could proceed with the full questionnaire only after obtaining consent by accepting (electronically marking) a 'Terms and Conditions' and 'Participation Agreement' consent form. A novel questionnaire instrument was developed for this study using items adapted from information about COVID-19 published by the CDC and the World Health Organization [WHO] alongside items used in previous COVID-19 surveys [22,23]. The questionnaire was developed and fielded in Spanish and later translated into English for reporting purposes. The full survey instrument is available in Additional file 1.
The questionnaire consisted of four sections: Demographic characteristics, comorbidities, past medical history of COVID-19, and vaccination information. The demographic section collected information on age, sex, occupation, and place of residency. The comorbidities section includes questions about respondents' medical history not related to COVID-19 (smoking status, medications, and other diseases). The COVID-19-related medical history was composed of 6 questions that include information related to previous infections. Finally, the vaccination information section was composed of 20 questions that evaluated the type, date, place, and number of doses of the administered vaccine. Moreover, it evaluated the timing and type of vaccine-related symptoms and adverse events.
The questionnaire was sent through national broadcasting news and social media, using snowballing as a sample method. Weekly reminders were sent to participants during the period of the questionnaire.

Data Management
We reviewed data case by case to ensure the highest accuracy possible on our results. As such, we identified cases where responses did not match the questions posed. For example, when they answered that they had not received the second dose of the vaccine, but in the section of adverse events attributed to the second dose they chose a response, this type of case was automatically eliminated. Moreover, we excluded responses coming from the same IP address to avoid malicious and duplicate answers.

Statistical Analysis
Descriptive and inferential analysis was conducted using the software IBM SPSS Statistics for Windows Version 24.0. The results of each item in the questionnaire were reported as men and women in percentages and absolute frequencies with no further intersex variability analysis.
The Chi-Square test was used to test the association between nominal qualitative variables. A Chi-Square test for the independence of variables was chosen to search for a relationship between variables such as gender, age, comorbidities, history of COVID-19, type of vaccine, and ESAVIs. The method for calculating the p-value was based on the frequentist approach to the test. For Tables 3, 5 and 6, the data were stratified for each adverse effect reported, and contingency tables were constructed for each group based on the variables mentioned. Finally, p-values were obtained according to the definition of the chi-square distribution. In the case of Table 4, the same approach was used; however, the variable "dose number" (first and second doses) was used to verify independence with the following variables: Time of ESAVIs onset, ESAVIs severity (mild, moderate, and severe), and measures to mitigate self-reported ESAVIs.
An independent t-test was employed to evaluate the existence of an association between quantitative variables (age and number of ESAVIs). Figure 4 presents a contrast between age groups and the number of ESAVIs. The t-test was calculated using the null hypothesis that age level "i" has the same number of adverse effects as age level "i-1". For age ranges with p-values < 0.05, the null hypothesis was rejected (different number of adverse effects between age groups).
Statistical significance was accepted at p < 0.05. Confidence intervals at 95% from means and proportions were also computed.

Reliability and Validation
Reliability was examined using a test-retest questionnaire using the final version of the survey. Since this questionnaire was created only for this project, we tested within the cohort of experts previously selected for informal interviews.

Ethical Approval
This project is part of our COVID-19 analysis program, a nationwide study of the epidemiology of COVID-19 in Ecuador, which received an exemption letter from the UDLA's IRB named CEISH on 10 March 2020. The current analysis included anonymized, un-identifiable information and it complies with all local and international guidelines regarding the ethical use of anonymous, non-identifiable information stipulated in such documents as the Declaration of Helsinki and the Good Clinical Practice Guidelines (GCP).

Demographic Information and Past Medical History
A total of 11,927 responses were collected at the national level; however, after revalidation, 6654 responses were considered valid for analysis. From the total responses, 4018 (60.4%) were women, 2631 (39.5%) were men, and 5 (0.1%) defined themselves as others. Most respondents were aged between 21 and 50 years (73.9%) and most of the participants were employed (79.60%); of them, 21.2% were healthcare workers. A total of 4104 participants reported having no comorbidities, and out of the remaining 2547 participants, Vaccines 2022, 10, 1047 6 of 19 the most frequent comorbidity was hypothyroidism (24.5%) and chronic diseases such as obesity (19.9%) and arterial hypertension (17.0%). Most of the participants did not have COVID-19 at the time of the questionnaire. The three vaccine types available in Ecuador were similarly distributed ( Table 2). At the time of the survey, for 6654 participants, a higher vaccination rate per 100,000 Ecuadorian inhabits was found in AstraZeneca vaccine users for "younger" participants between 20 and 60 years of age, while the highest vaccination rate in participants older than 60 years was found in the Pfizer vaccine. (Figure 1).

General Characteristics
Overall, the number of adverse effects reported by participants between the first and second doses was n = 26,325, and women had a higher frequency of self-reported ESAVIs 66.7% (n = 17,507). Most symptoms were reported by participants who claimed not to have comorbidities 58.0% (n = 15,206) and in those lacking a personal history of COVID-19 infection 72.7% (n = 19,081) ( Table 3). At the time of the survey, for 6654 participants, a higher vaccination rate per 100,000 Ecuadorian inhabits was found in AstraZeneca vaccine users for "younger" participants between 20 and 60 years of age, while the highest vaccination rate in participants older than 60 years was found in the Pfizer vaccine. (Figure 1).

General Characteristics
Overall, the number of adverse effects reported by participants between the first and second doses was n = 26,325, and women had a higher frequency of self-reported ESAVIs 66.7% (n = 17,507). Most symptoms were reported by participants who claimed not to have comorbidities 58.0% (n = 15,206) and in those lacking a personal history of COVID-19 infection 72.7% (n = 19,081) ( Table 3). After the first vaccine dose, a total of 19,501 self-reported ESAVIs were reported; 87.0% were mild, 11.5% were moderate, and 1.5% were severe. For the second dose, there were 6776 self-reported ESAVIs (87.1% mild, 10.9% moderate, and 2.1% severe) ( Table 4).
The behavior of the reported effects shows a common trend when analyzed among all different age groups (nine groups) in favor of pain or swelling at the injection site in all age groups as the most reported 17.2% (n = 4500); moreover, headache is presented as the second most frequent with 13.3% (n = 3502). (Table 5). According to the ESAVIs, for participants between 10 and 40 years and over 70 years of age, the most frequent symptom was fatigue, while for the group between 40 and 70 years old, the most frequent symptom was pain or swelling at the injection site and headache (Table 5).     Table 3 shows the frequency distribution of ESAVIs; the numbers in parentheses show the row percentages for each variable (gender, comorbidities, history of COVID-19 infection, and age group). * The p-value was obtained by Chi-Square test.
Vaccines 2022, 10, 1047 10 of 19 Self-reported ESAVIs were higher among young adults (21 to 39 years), with 55.8% and adults (40 to 64 years), with 33.2%, compared to older adults (>65 years), with 7.6%; however, there are no significant differences in terms of the occurrence of adverse events after performing the corresponding statistical tests (Additional file 2). Figure 2 shows the effect of the comparison of the number of adverse events between the different age ranges. It was found that the effects due to vaccination differ only between the age ranges up to 70 years (p < 0.05). In participants older than 70 years, the number of adverse effects is very similar (p > 0.05) (Figure 2). Along the same lines, when addressing the influence of age between only two age groups, taking 50 years as the cut-off point, it was shown that the group under 50 years of age presents various symptoms more frequently (p < 0.05) ( Table 3). Table 4. Self-reported ESAVIs after receiving the first and second dose of any of the available COVID-19 vaccines in Ecuador (n = 6645).  the different age ranges. It was found that the effects due to vaccination differ only be-tween the age ranges up to 70 years (p < 0.05). In participants older than 70 years, the number of adverse effects is very similar (p > 0.05) (Figure 2). Along the same lines, when addressing the influence of age between only two age groups, taking 50 years as the cutoff point, it was shown that the group under 50 years of age presents various symptoms more frequently (p < 0.05) ( Table 3). The analysis of the development of ESAVIs for each age group shows that for participants in the 30 to 40 years age group, a higher number of ESAVIs were reported for Astra-Zeneca vaccine users compared to Sinovac vaccine (p = 0.016) as well as for participants in the 40 to 50 years age group (p = 0.028) (Figure 3). The analysis of the development of ESAVIs for each age group shows that for participants in the 30 to 40 years age group, a higher number of ESAVIs were reported for AstraZeneca vaccine users compared to Sinovac vaccine (p = 0.016) as well as for participants in the 40 to 50 years age group (p = 0.028) (Figure 3).

Self-Reported ESAVIs after COVID-19 Vaccination
Vaccines 2022, 10, x FOR PEER REVIEW 12 of 18 Figure 3. T-test between number of ESAVIs and vaccine type by age range.

Self-Reported ESAVIs after the First Vaccine Dose
The proportion of ESAVIs after the first vaccine dose was higher in users of Astra-Zeneca (57.8%), followed by Pfizer (24.9%) and Sinovac (17.3%). In the same way, the most common self-reported ESAVIs with the AstraZeneca vaccine were headache (9.1%), mus-

Self-Reported ESAVIs after the Second Vaccine Dose
Second-dose AstraZeneca vaccine users reported a higher proportion of ESAVIs (72.21%) compared to Pfizer/BioNTech (15.85%) and Sinovac (11.93%) vaccines. For the As-traZeneca vaccine, headache (8.7%), pain or swelling at the injection site (7.7%), and fatigue or tiredness (7.2%) were the more common self-reported ESAVIs. After the Pfizer/BioNTech vaccine, the most common self-reported ESAVIs were pain or swelling at the injection site (10.3%), headache (7.2%), and muscle pain (6.7%). Finally, with the Sinovac vaccine, participants self-reported pain or swelling at the injection site (9.5%), headache (7.6%), and fatigue or tiredness (6.8%) more frequently (Table 7). In general, the effect of the available vaccines in Ecuador showed that the most common adverse event was pain at the injection site, followed by headache (Figure 4). In general, the effect of the available vaccines in Ecuador showed that the most common adverse event was pain at the injection site, followed by headache (Figure 4).

Discussion
To the best of our knowledge, this is the first study in Latin America examining selfreported ESAVIs after one or two COVID-19 vaccine doses. This analysis revealed that women, young adults, participants without a COVID-19 infection history, participants af-

Discussion
To the best of our knowledge, this is the first study in Latin America examining selfreported ESAVIs after one or two COVID-19 vaccine doses. This analysis revealed that women, young adults, participants without a COVID-19 infection history, participants after the first vaccine dose, and AstraZeneca users developed ESAVIs more frequently. Most self-reported ESAVIs were mild and moderate.
Although this sample only represents a small subset of all vaccinated people in Ecuador, our results are similar to those described in other reports. Regarding sex, we found a higher proportion of female participants, who showed more pain or swelling at the injection site, muscle pain, and swollen glands (p < 0.05) compared to the male group and other groups. These results are similar to those seen in several studies [24][25][26]. Despite the differences in the distribution of the participants according to age, in the group younger than 50 years (81.2%) and the group older than 50 years (18.8%), the presentation of self-reported ESAVIs showed significant results (p < 0.05) that participants younger than 50 years had more ESAVIs, especially mild and moderate symptoms, in comparison with those older than 50 years. This is in line with data reported by Zare [22,27,28].
One interesting and controversial point is the relationship between COVID-19 history and ESAVIs. In our study, participants without a history of infection at the time of the questionnaire reported more adverse events. Our findings contrast with several studies [22,27]. Regarding these differences, an increase in reactogenicity related to an increase in antibody titers conditioned by a previous infection in vaccinated individuals has been proposed. However, to date, this causal relationship has not been fully demonstrated, assuring us that the role of the history of the previous infection does not cause an increased number of ESAVIs in patients receiving vaccines against the SARS-CoV-2 virus.
The most common reported ESAVIs were pain or swelling at the injection site, headache, and fatigue, as observed by several authors in other investigations based on their own reports, such as Menni et al. in the UK, Elgendy et al. in the Egyptian population, as well as in the clinical trial by Polack et al. [22,24,29]. When comparing the frequency of ESAVIs according to the number of doses, no important differences were found between the ESAVIs reported after the first and second doses, although the absolute number of ESAVIs reported was higher after the first dose (n = 19,501 vs. n = 6776), a finding that is similar to recent reports [29]. In addition, among the vaccines types available in Ecuador, for AstraZeneca and Sinovac, a similar trend was seen with a higher number of ESAVIs reported in the administration of the first dose, while with the Pfizer vaccine, a slight increase in ESAVIs was found in the administration of the second dose. Regarding these discrepancies, previous studies with larger populations showed that the participants studied had a greater number of adverse effects caused by the second dose of the AstraZeneca and Pfizer vaccines [23,24].
Regarding differences in the severity of symptoms, it is interesting that the most frequent moderate ESAVIs reported, in general, were fever and nausea, while within the severe ESAVIs, changes in heart rhythm (tachycardia) occurred in more than half of the reports.
The severe ESAVIs found in our study were very infrequent (overall n < 30). In this group, self-reports of Guillain-Barre Syndrome, transverse myelitis, anaphylaxis, and coagulation disorders were obtained for the three vaccines available in Ecuador (AstraZeneca, Pfizer, Sinovac), in addition, the frequency of presentation of severe ESAVIs was higher after receiving the first dose of vaccines. In the available literature of self-reporting studies, no data have been presented regarding the severe adverse effects reported in our research. Only one self-reporting study developed by Mathioudakis et al. found reports of anaphylaxis [30], the difficulty in accepting the reports of severe adverse effects as valid is likely due to the self-reported nature of the investigations, and, in the case of our study, we had no way of assuring that the diagnoses were made by physicians.
The origin and cause of these findings are difficult to explain, and in the case of the relationship between vaccines against COVID-19 and the development of Guillain-Barre syndrome, an analysis performed by Keh et al. in databases from the United Kingdom stated that there are no demographic or phenotypic differences that can ensure the development of Guillain-Barre syndrome due to COVID-19 vaccines; however, most of the population studied was of the UK white ethnicity (90.3%) with a minimal presence of Latin Americans [26]. Otherwise, in the case of anaphylaxis, an extensive meta-analysis managed to demonstrate that, although reports of anaphylaxis due to the possible cause of SARS-CoV-2 virus vaccines are higher than with other rare adverse effects, these cannot be attributed to the receipt of these vaccines, as they occur with a wide range of vaccines and are mainly related to vaccine components [25].
Regarding the ESAVIs reported by users who received the CoronaVac vaccine (Sinovac) in our study, the most frequently recorded were pain or swelling at the injection site, headache, and fatigue for the first and second doses. In agreement with these data, Zhang et al. found that the most frequent ESAVIs recorded were fatigue, muscle pain, and headache for both doses of Sinovac in a Chinese population, coinciding only with headache in our findings [31]; while in a Turkish population, the findings by Riad et al. were shown to be more similar to what was found in our study, since the most reported ESAVIs were pain at the injection site, fatigue, headache, and muscle pain [32].
Our findings show differences in the frequency of reporting of ESAVIs when comparing the type of vaccine, and the data showed that the greatest number of ESAVIs was attributed to the ChAdOx1 viral vector vaccine, the same as that presented by Menni [22,33,34]. Furthermore, among the 30 to 40 years and 40 to 50 years age groups, participants who received the AstraZeneca vaccine had a higher number of reported ESAVIs compared to users of other vaccines; these differences could be due to the fact that, in participants aged 20 to 60 years, the most used vaccine was ChAdOx1.
Addressing these differences in the number of doses, the reported distribution shows different patterns. In the case of the first dose, the ChAdOx1 vaccine was responsible for the majority of adverse events, while among participants that received the second dose of vaccination, the BioNTech vaccine was the one that was attributed to more than half of ESAVIs, and this trend is similar to that found in Egypt [29].
In the analysis of the time of ESAVIs onset, we found that the majority of ESAVIs significantly occurred within the first day after receiving the vaccine. Other authors observed the same behavior in their participants, although they did not find statistically significant differences [29,34,35]. On the other hand, the measures that the adverse events caused our participants to take were characterized by taking medication to correct the symptoms in most cases, and in a very small proportion, to visit a doctor. These behaviors have been previously observed in similar proportions in the population of Saudi Arabia [34], and almost zero participants required hospitalization due to adverse events. Likewise, this trend was similarly evidenced in studies that evaluated adverse effects due to AstraZeneca, Pfizer vaccines, and Sinopharm [34,35].

Limitations
Our study has several limitations inherent to the cross-sectional self-reported design. Since the questionnaire was distributed through social networks, the information belonging to the population that did not have the resources to access the questionnaire was relegated from the study. Similarly, most older adults handle electronic devices and the Internet with difficulty. This may cause selection bias; however, we consider that our sample calculation provided a considerable sample size to reduce this bias. Social desirability bias is another potential limitation that could have affected responses due to the self-report nature of the questionnaire. In other words, negative or positive self-reported symptoms may be underreported or overreported because respondents want to mark what they consider "socially acceptable" responses. However, the use of an anonymous online questionnaire should have somewhat mitigated the risk of this bias. Despite these limitations, we managed to obtain a large amount of information provided by many participants based on official data on the number of vaccinated in Ecuador. Likewise, our data were subjected to extensive filtering to provide valid results. For these reasons, we consider this study to be a reliable approximation of the self-reported ESAVIs of Ecuadorians that can help to inform their symptoms and somehow decrease COVID-19 hesitancy.

Conclusions
This is the first report in Ecuador and the Latin-American region on the adverse events attributed to COVID-19 vaccines. The results of our study allow us to conclude that, in Ecuador, the majority of ESAVIs are mild to moderate and occur more frequently after receiving the AstraZeneca vaccine, followed by Pfizer and finally Sinovac. We also noted that women are more likely to report ESAVIs compared to men and that young people report them more frequently than women. In general, it can be observed that the adverse events reported are mostly mild and transitory, which further supports the fact that COVID-19 vaccines are safe. Mild ESAVIs were the most self-reported events described in Ecuador. We found that first doses were associated with a higher proportion of ESAVIs compared to subjects who received a subsequent dose. We also found that participants without a past medical history of COVID-19, AstraZeneca users, and younger populations are more likely to develop ESAVIs. Funding: This work did not receive a formal grant. However, it received financial support associated with the publication fee from the University of the Americas in Quito, Ecuador.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The dataset with the total responses can be obtained from the authors upon reasonable request. The questionnaire summary and its results are included in this article.