Side effects of Sputnik V, Oxford–AstraZeneca, Sinopharm, and Covaxin and their associations with other variables among healthcare workers of a tertiary hospital in Iran

Abstract

Background

Misinformation about the prevalence of COVID-19 vaccines' side effects (SEs) has led to concerns about and mistrust of vaccine safety. Thus, this study aimed to evaluate the prevalence of COVID-19 vaccines' SEs.

Methods

In a cross-sectional survey-based study on the healthcare workers (HCWs) of a tertiary hospital in Iran, the SEs of Sputnik V, Oxford–AstraZeneca, Sinopharm, and Covaxin were evaluated through a face-to-face interview by a researcher-made questionnaire.

Results

A total of 368 HCWs received at least one dose of a COVID-19 vaccine. The prevalence of people with at least one SE was higher among those who received the Oxford–AstraZeneca (95.8 %) and Sputnik V (92.1 %) vaccines than those who received Covaxin (70.5 %) or Sinopharm (66.7 %). Following the first and second doses, injection site pain (50.3 % and 58.2 %), body/muscle pain (53.5 % and 39.4 %), fever (54.5 % and 32.9 %), headache (41.3 % and 36.5 %), and fatigue (44.4 % and 32.4 %) were the most common SEs. Overall, SEs were often initiated within 12 h and subsided within 72 h of vaccination. The prevalence of SEs after the first dose of Sputnik V was higher among those aged ≤ 31 years (93.3 %) than those aged > 31 years (80.5 %). In the Sputnik V group, the number of SEs experienced after the first dose was higher in women with underlying diseases than those without such diseases. Furthermore, the body mass index of participants with SEs was lower than that of participants without SEs.

Conclusion

Compared to Sinopharm or Covaxin, the Sputnik V and Oxford–AstraZeneca vaccines were associated with a higher prevalence of SEs, a greater number of SEs per individual, and more severe SEs.

1. Introduction

Accompanying preventive measures with vaccination is essential to overcome the coronavirus disease 2019 (COVID-19) pandemic [1], [2]. The vaccination program against COVID-19 began in December 2020 [3]. Because healthcare workers (HCWs) are the most at risk for COVID-19 and are also potentially affected by the disease [4], [5], their vaccination has been prioritized to benefit the healthcare system, their families, caregivers, and patients.

Since the beginning of the COVID-19 vaccination program, public concerns about vaccines' effectiveness and side effects (SEs) have been raised [6], which has affected vaccines' acceptance rates [7], [8]. Several other determinants of COVID-19 vaccine acceptance have also been addressed [7], [9]. Although some studies [9], [10] suggest a high acceptance rate, emerging new virus variants and the need to administer booster doses required governments to adopt strategies to promote vaccine acceptance and minimize vaccine hesitancy [6], [9]. A crucial part of these strategies is providing the public with transparent information about the effectiveness and safety of vaccines [11]. In this regard, studies on COVID-19 vaccines' SEs should be designed and implemented appropriately to gather evidence about vaccine safety.

The World Health Organization (WHO) defines an adverse event following immunization (AEFI) as “any untoward medical occurrence which follows immunization and which does not necessarily have a causal relationship with the usage of the vaccine [12].” Data about inactivated vaccines (Sinopharm and Covaxin) and viral vector vaccines (Oxford–AstraZeneca and Sputnik V) obtained from clinical trials reveals that most reported cases of AEFI were mild to moderate and generally disappeared within a few days after vaccination [13], [14], [15], [16], [17], [18]. Additionally, the most comprehensive safety monitoring study on the Pfizer-BioNTech and Moderna vaccines in the U.S. indicated that, in 90.8 % of the cases, the SEs were nonserious and involved local and systemic symptoms. The most frequent symptoms included headache (22.4 %), fatigue (16.5 %), and dizziness (5.16 %) [19]. However, reports that rely on data extracted from clinical trials may provide information about prevalence that differs from the information gained from real-world studies. Combining the results of clinical trials and real-world studies can provide a more comprehensive evaluation of vaccine SEs than considering either type of study alone [20].

The Government of Iran officially began the COVID-19 vaccine rollout in February 2021 with Sputnik V. During the study time frame, Sputnik V (Gam-COVID-Vac, Russia, rAd26 and rAd5 viral vectors), Sinopharm (BBIBP-CorV, China, inactivated whole virus), Oxford–AstraZeneca (AZD1222 or ChAdOx1 nCoV-19, South Korea, viral vector), and Covaxin (BBV152, Bharat Biotech, India, inactivated whole virus) were the most administered vaccines. HCWs were the first group to receive the vaccine, which could have significantly impacted the community's acceptance of or doubts about vaccination [7], [21]. Therefore, this study investigated the SEs of four available COVID-19 vaccines (Sputnik V, Sinopharm, Oxford–AstraZeneca, and Covaxin) among HCWs of Imam Hassan Hospital, a tertiary hospital serving as a COVID-19 treatment center.

2. Materials and methods

2.1. Study population

This study was a descriptive-analytical cross-sectional survey-based study conducted on HCWs of Imam Hassan Hospital; this hospital is affiliated with the North Khorasan University of Medical Sciences, Bojnurd, Iran, which is the main COVID-19 treatment center in the university. This study was performed on a group of employed personnel, including nurses, administrative personnel, service personnel, physicians, and students, who were vaccinated as part of the HCW vaccination program in the spring of 2021. This vaccination program was initially performed with a smaller number of members of the target population starting at the end of February 2021; at this time, reported SEs were sparse and indistinctive.

At the time of designing the present study, the reported SEs of vaccines were mainly related to Pfizer-BioNTech, which is unavailable in Iran. Given that the Imam Hassan HCWs were vaccinated using four available vaccines (Sputnik V, Sinopharm, Oxford–AstraZeneca, and Covaxin), the researchers decided to evaluate their SEs. The sample size was determined by the census method. Therefore, all HCWs who met the inclusion criteria were considered when evaluating the SEs of the investigated vaccines.

2.2. Inclusion and exclusion criteria

All HCWs, who received the COVID-19 vaccine in the spring of 2021, were included. The exclusion criterion was not being vaccinated. The most important reasons for not vaccinating could be COVID-19 during the last three months, a personal decision not to vaccinate, medical restrictions or contraindications (specifically a history of anaphylaxis to previous vaccines or multiple drugs), and pregnancy [22]. Other diseases unrelated to COVID-19 were not considered in the exclusion criteria (e.g., a history of cancer or autoimmune diseases).

2.3. Instrument

2.3.1. Instrument design

At the time of the study, there was a lack of information about the vaccines' SEs. Therefore, a researcher-made questionnaire was prepared by reviewing available articles and oral interviews with HCWs. In drafting the questionnaire, attention was paid to the form used by Riad et al. [3], several reports published by the WHO, CDC [23], [24], and other approved sources of vaccine SEs reports. Then, over several sessions, the researchers discussed different parts of this questionnaire to achieve a comprehensive data collection form. As a result, this form consisted of several parts, which are listed below:

• Demographic information, including name, gender, age, weight, height, occupation, workplace, years of employment, and type of employment.

• History of COVID-19 infection, its frequency, its date(s), and how it was diagnosed.

• Current history of underlying diseases and medication use.

• Information about the COVID-19 vaccine received, including the type of vaccine, injection dates, and history of COVID-19 after vaccination. Additionally, each participant was asked about their history of influenza vaccine administration.

• Information about the general SEs of vaccines using organ system-based symptomatology on different organ systems, including the skeletal, muscular, circulatory, digestive, nervous, and integumentary systems. For each SE, participants were requested to specify the onset time and duration of symptoms and whether they were related to the first dose, second dose, or both. Moreover, this section included items that could help estimate the severity of the SEs of each dose by leaving work or hospitalization due to SEs and the use of medications to relieve the symptoms. Moreover, if the participant took medication to relieve the symptoms of SEs, the researchers also evaluated the type of medication used, how long it was used, and how long it took symptoms to subside.

2.3.2. Instrument validity and reliability

2.3.2.1. Validity

The face validity of the questionnaire was assessed in writing using face-to-face interviews with the target group. Then, according to the recommendations, the items were edited several times to simplify the understanding of the queries. It continued until no new change was proposed.

The questionnaire was given to 10 experts and faculty members with sufficient knowledge and experience in this field (nursing, epidemiology, and health) to check the validity of its content. After carefully reading the questionnaire, experts were asked to present their corrections regarding grammar, word choice, and the placement of items. Items were subsequently edited according to the experts' recommendations. In addition to the qualitative validity of the content, the experts also commented on the necessity of the questionnaire items based on a triple Likert scale with the options of “essential,” “useful but not essential,” and “not necessary” [25]. Then, the experts' responses were quantified, and the relaxed content validity ratios were determined [26].

The experts were asked to rate the relevance of each item based on a four-point Likert scale to calculate the content validity index (CVI). The four choices on this scale were as follows: 1 = not relevant, 2 = somewhat relevant, 3 = quite relevant, and 4 = highly relevant [27]. Then, the CVI was calculated using the CVI formula. The CVI score of each item was calculated by dividing the number of experts who gave the item a score of 3 or 4 by the total number of experts. A CVI score of 0.78 or higher was considered acceptable [27]. Finally, the scale-level CVI was calculated as 0.90 by the averaging method [27].

2.3.2.2. Reliability

The reliability of the questionnaire was evaluated using the test–retest method by ten newly vaccinated personnel, who were interviewed twice at an interval of two weeks. The results revealed strong reliability of the instrument [28] with a mean (standard deviation, range) Kappa coefficient of 0.85 (0.12, 0.66–1.00).

2.4. Study procedure

Initially, the infection control unit of the hospital, which is responsible for coordinating the vaccination program, provided a list of the personnel working at Imam Hassan Hospital. This list included each person's name, workplace, the type of vaccine received, and vaccination date. Then, the researchers went to the participants' workplaces and completed the prepared questionnaire through face-to-face interviews.

2.5. Statistical analysis

Statistical analysis was performed using SPSS software version 20.0 (IBM SPSS Statistics for Windows, Armonk, NY: IBM Corp). Descriptive data were shown as numbers and percentages and central and dispersion indices, including median (interquartile range or IQR). Additionally, the Kruskal-Wallis H and Chi-square tests were employed if there were statistically significant differences between groups on continuous and discrete variables, respectively. Correspondingly, the Mann-Whitney U and Chi-square tests (or Fisher's exact test, if needed) were implemented in two-by-two comparisons. Significance level was considered in all tests with bilateral p-value less than or equal to 0.05.

2.6. Ethics approval and consent to participate

The North Khorasan University of Medical Sciences ethics committee approved the study protocol under the reference IR.NKUMS.REC.1400.062. Before participants joined the study, the primary objectives were explained to the eligible population, and written informed consent was obtained from all participants. The participants were told that their participation was entirely voluntary and that they could leave the study at any time. In addition, participants were assured that since the collected information was personal, it would be kept confidential. The eligible population did not receive any financial or non-financial incentives to participate in the study, as such incentives could have affected their motivation. All stages of the research and data collection were based on the principles of the Helsinki Declaration [29].

3. Results

3.1. Population characteristics and demographic data

During the data collection process, we realized that 558 of 675 people working in Imam Hassan hospital had already received at least one dose of a COVID-19 vaccine. Of these 558 individuals, 434 had received a vaccine within the target period (spring 2021) and, therefore, were eligible to participate in the study (Fig. 1 ). The recipients' claims about the vaccine type were confirmed using documents collected by the hospital's infection control unit. At the face-to-face interview stage, 17 people did not agree to participate, and another 42 were unavailable. Consequently, 375 personnel were interviewed. Ultimately, data from 368 completed forms were analyzed.

Fig. 1.

Flow diagram of the study's participant recruitment.

This study included 239 women (64.9 %) and 129 men (35.1 %) who were 21–67 years old, with a median (IQR) age of 31 (27–38). In total, 174 (47.3 %), 83 (22.5 %), 65 (17.7 %), and 46 (12.5 %) participants received at least one dose of the Sputnik V, Sinopharm, Oxford–AstraZeneca, and Covaxin vaccines, respectively. No significant differences emerged between vaccine groups' body mass index (BMI), frequency of contracting COVID-19 before vaccination, underlying diseases, or medication intake. Meanwhile, the groups differed significantly in terms of age, years of employment, sex, occupation, history of COVID-19 before vaccination, and history of receiving an influenza vaccine. These results are summarized in Table 1 .

Table 1.

Demographics and clinical characteristics of the study population.

Characteristics	Vaccine type				P-value
	Sputnik V (n = 174)	Sinopharm (n = 83)	Oxford–AstraZeneca (n = 65)	Covaxin (n = 46)
Age, median (IQR*), year	31 (28–38)	30 (27–37.5)	30 (26–37)	35 (31–41)	0.002
Body Mass Index, median (IQR), kg/m2	24.3 (22.3–26.4)	24.6 (21.4–26.5)	24.2 (21.5–26.8)	25.5 (22.7–27.1)	0.400
Years of employment, median (IQR), year	5 (3–12)	4 (3–11)	5.5 (2–12.25)	9 (4–15)	0.014
Sex, N (%)					0.011
Female	122 (70.1)	57 (68.7)	31 (47.7)	29 (63.0)
Male	52 (29.9)	26 (31.3)	34 (52.3)	17 (37.0)
Occupation, N (%)					<0.001
Nurse	99 (56.9)	43 (51.8)	11 (16.9)	19 (41.3)
Official	16 (9.2)	17 (20.5)	24 (36.9)	13 (28.3)
Patient care technician	30 (17.2)	3 (3.6)	8 (12.3)	4 (8.7)
Lab technician	16 (9.2)	5 (6.0)	1 (1.5)	0
Student	0	1 (1.2)	12 (18.5)	0
Operation room staff	2 (1.1)	7 (8.4)	2 (3.1)	0
Ward secretary	2 (1.1)	5 (6.0)	2 (3.1)	1 (2.2)
Patient advocate	5 (2.9)	1 (1.2)	0	3 (6.5)
Cooker	0	0	4 (6.2)	3 (6.5)
Pharmaceutical technician	2 (1.1)	0	0	2 (4.3)
Pharmacist	0	1 (1.2)	0	1 (2.2)
Physician	2 (1.1)	0	0	0
Physiotherapist	0	0	1 (0.3)	0
History of COVID-19 before vaccination, N (%)					0.002
Yes	110 (63.2)	49 (59.0)	25 (38.5)	32 (69.6)
No	64 (36.8)	34 (41.0)	40 (61.5)	14 (30.4)
Frequency of contracting COVID-19 before vaccination, N (%)					0.884
1	69 (62.7)	33 (67.3)	14 (58.3)	17 (53.1)
2	35 (31.8)	15 (30.6)	9 (37.5)	12 (37.5)
3	5 (4.5)	1 (2.0)	1 (4.2)	3 (9.4)
4	1 (0.9)	0	0	0
Underlying diseases, N (%)					0.514
Yes	26 (14.9)	13 (15.7)	12 (18.5)	11 (23.9)
No	148 (85.1)	70 (84.3)	53 (81.5)	35 (76.1)
Medication intake, N (%)					0.210
Yes	21 (12.1)	14 (16.9)	12 (18.5)	11 (23.9)
No	153 (87.9)	69 (83.1)	53 (81.5)	35 (76.1)
History of receiving an influenza vaccine, N (%)					0.001
Yes	118 (67.8)	44 (53.0)	27 (41.5)	32 (69.6)
No	56 (32.2)	39 (47.0)	38 (58.5)	14 (30.4)

^*IQR, interquartile range.

3.2. Side effects after the first and second doses of vaccines

3.2.1. First dose

Of 368 people who received the first dose of a vaccine, 288 (78.3 %) experienced at least one SE. The prevalence of SEs was significantly different between the vaccine groups (p < 0.001). Two-by-two comparisons determined the prevalence of SEs as follows: Oxford–AstraZeneca = Sputnik V > Covaxin = Sinopharm. Regarding the number of SEs per participant, the highest and lowest values belonged to Sputnik V (median [IQR] = 4 [2–7]) and Sinopharm (median [IQR] = 1.5 [1–2]), respectively. The two-by-two comparisons also revealed that the number of SEs experienced after the first dose did not significantly differ between the Sputnik V and Oxford–AstraZeneca groups (p = 0.944), though significant differences were revealed by other comparisons (Sputnik V = Oxford–AstraZeneca > Covaxin > Sinopharm). Details of the two-by-two comparisons of the frequency and number of SEs experienced after the first and second doses are shown in Supplemental Table S1.

As shown in Fig. 2 , the most frequent SE following the first dose of Sputnik V, Sinopharm, Oxford–AstraZeneca, and Covaxin was body/muscle pain (n = 96, 63.2 %), injection site pain (n = 21, 43.8 %), fever (n = 50, 82.0 %), and injection site pain (n = 17, 63.0 %), respectively. However, considering the whole population, the most common SE was fever (n = 157, 54.5 %). In addition, significant between-group differences were observed in the frequencies of fever, body/muscle pain, fatigue, chills, feeling unwell, joint pain, dizziness, nausea, sweating, depression, and petechiae. A detailed profile of the two-by-two comparisons of reported SEs following the first and second vaccine doses is provided in Supplemental Table S2. Further, the frequencies (percentages) of the most and least common SEs are illustrated in Supplemental Table S3 .

Fig. 2.

Frequency distribution of the most common side effects in the first and the second doses of vaccines.

3.2.2. Second dose

Of the 308 participants who received the second dose, 170 (55.2 %) had at least one SE after receiving this dose. The prevalence of SEs differed significantly between the vaccine groups (p = 0.015). The two-by-two comparisons determined that this difference was between Sputnik V and Sinopharm (p = 0.001), with Sputnik V being associated with a higher prevalence of SEs than the other vaccines. The Covaxin and Sinopharm vaccine groups showed the highest and lowest numbers of SEs per participant (median [IQR] = 3 [1.25–7.5] vs 1 [1–3], respectively). As shown by the two-by-two comparisons (Supplemental Table S1), the Sputnik V and Sinopharm groups presented a significant difference in the number of SEs experienced after the second dose (p = 0.003), as did the Sinopharm and Covaxin groups (p = 0.010). However, other comparisons did not reveal significant differences (Covaxin = Sputnik V = Oxford–AstraZeneca > Sinopharm).

As shown in Fig. 2, injection site pain was the most frequent SE following the second dose of Sputnik V (n = 58, 56.3 %), Sinopharm (n = 16, 53.3 %), Oxford–AstraZeneca (n = 9, 69.2 %), and Covaxin (n = 16, 66.7 %), and also in the whole population (n = 99, 58.2 %). Furthermore, the frequency of body/muscle pain, fever, fatigue, chills, local swelling, itching, petechiae, and other symptoms differed significantly between the vaccine groups (Supplemental Table S2 ). Moreover, the frequencies (percentages) of the most and least common SEs are illustrated in Supplemental Table S3 .

3.2.3. First dose vs The second dose

As shown in Table 2 , significantly fewer people experienced SEs after the second dose than after the first dose of Oxford–AstraZeneca (p = 0.008), Sputnik V (p = 0.030), Sinopharm (p = 0.018), and Covaxin (p = 0.014). In the whole population, the number of SEs associated with the first dose was higher than those associated with the second dose (p < 0.001). However, when the vaccine groups were investigated separately, this trend was found only for Sputnik V (p < 0.001) and Oxford–AstraZeneca (p < 0.001), whereas no significant differences were found in terms of the number of SEs between the first and second doses of Sinopharm (p = 0.139) and Covaxin (p = 0.988).

Table 2.

Characteristics of side effects, their outcomes, and medication use to relieve them.

Characteristics		Vaccine types				p-value
	Total	Sputnik V	Sinopharm	Oxford–AstraZeneca	Covaxin
Frequency of participants with side effects, N (%)
First dose	288 (78.3)	152 (87.4)	48 (57.8)	61 (93.8)	27 (58.7)	<0.001
Second dose	170 (55.2)	103 (62.4)	30 (40.0)	13 (54.2)	24 (54.5)	0.015
P-value between the first and second doses	<0.001	0.030	0.018	0.008	0.014
One of both doses*	256 (83.1)	152 (92.1)	50 (66.7)	23 (95.8)	31 (70.5)	<0.001
Both doses*	146 (47.4)	94 (57.0)	22 (29.3)	12 (50.0)	18 (40.9)	0.001
Number of side effects, median (IQR)
First dose	3 (2–6)	4 (2–7)	1.5 (1–2)	4 (3–6)	2 (1–6)	<0.001
Second dose	3 (1–5)	3 (1–5)	1 (1–3)	2 (1–4)	3 (1.25–7.5)	0.016
p-value between doses	<0.001	<0.001	0.139	<0.001	0.988
Distribution of symptoms onset, N (%)
First dose, hours						<0.001
≤ 6	451 (35.7)	266 (34.6)	46 (52.3)	96 (31.9)	43 (40.6)
6–12	620 (49.1)	388 (50.5)	24 (27.3)	176 (58.5)	32 (30.2)
12–24	161 (12.7)	99 (12.9)	15 (17.0)	24 (8.0)	23 (21.7)
> 24	31 (2.5)	15 (2.0)	3 (3.4)	5 (1.6)	8 (7.5)
Second dose, hours						<0.001
≤ 6	245 (39.3)	144 (36.5)	21 (30.0)	28 (71.8)	52 (43.3)
6–12	247 (39.6)	176 (44.7)	32 (45.7)	7 (17.9)	32 (26.7)
12–24	100 (16.1)	61 (15.5)	13 (18.6)	1 (2.6)	25 (20.8)
> 24	31 (5.0)	13 (3.3)	4 (5.7)	3 (7.7)	11 (9.2)
Distribution of symptoms duration, N (%)						<0.001
First dose
≤ 24 h	635 (50.3)	382 (49.8)	48 (54.6)	174 (57.8)	31 (29.2)
24–72 h	420 (33.2)	255 (33.2)	24 (27.3)	100 (33.2)	41 (38.7)
72 h-7 days	168 (13.3)	117 (15.2)	15 (17.0)	18 (6.0)	18 (17.0)
> 7 days	40 (3.2)	14 (1.8)	1 (1.1)	9 (3.0)	16 (15.1)
Second dose						<0.001
≤ 24 h	278 (44.6)	188 (47.7)	34 (48.6)	26 (66.7)	30 (25.0)
24–72 h	199 (31.9)	122 (31.0)	14 (20.0)	11 (28.2)	52 (43.3)
72 h-7 days	118 (18.9)	75 (19.0)	20 (28.6)	2 (5.1)	21 (17.5)
> 7 days	28 (4.5)	9 (2.3)	2 (2.8)	0	17 (14.2)
Medication use to relieve symptoms**, N (%)
First dose	199 (69.1)	115 (75.7)	21 (43.8)	48 (78.7)	15 (55.6)	<0.001
Second dose	92 (54.1)	62 (60.2)	14 (46.7)	5 (38.5)	11 (45.8)	0.246
Days of medication use, median (IQR)
First dose	1 (1–2)	1 (1–2)	1 (1–2)	1 (1–2)	1 (1–1.25)	0.766
Second dose	1 (1–2)	1 (1–2)	1 (1–1.25)	1 (1–2.5)	1.5 (1–3.5)	0.410
Improvement of symptoms***, N (%)
First dose						0.497
< 4 h	125 (62.8)	74 (64.3)	12 (57.1)	32 (66.7)	7 (46.7)
≥ 4 h	74 (37.2)	41 (35.6)	9 (42.9)	16 (33.3)	8 (53.3)
Second dose						0.411
< 4 h	55 (59.8)	39 (62.9)	9 (64.3)	3 (60.0)	4 (36.4)
≥ 4 h	37 (40.2)	23 (37.1)	5 (35.7)	2 (40.0)	7 (63.6)
Recipients' point of view regarding severity, duration, and number of side effects****, N (%)
The severity of side effects						0.148
The first dose is more than the second	94 (64.4)	63 (67.0)	11 (50.0)	10 (83.3)	10 (55.6)
The second dose is more than the first	41 (28.1)	26 (27.7)	8 (36.4)	0	7 (38.9)
Similar	11 (7.5)	5 (5.3)	3 (13.6)	2 (16.7)	1 (5.6)
Duration of side effects						0.021
The first dose is more than the second	68 (46.6)	46 (48.9)	4 (18.2)	10 (83.3)	8 (44.4)
The second dose is more than the first	26 (17.8)	15 (16.0)	7 (31.8)	0	4 (22.2)
Similar	52 (35.6)	33 (35.1)	11 (50.0)	2 (16.7)	6 (33.3)
Number of side effects						0.169
The first dose is more than the second	72 (49.3)	47 (50.0)	6 (27.3)	9 (75.0)	10 (55.6)
The second dose is more than the first	28 (19.2)	18 (19.1)	6 (27.3)	0	4 (22.2)
Similar	46 (31.5)	29 (30.9)	10 (45.5)	3 (25.0)	4 (22.2)
Hospitalization following vaccine side effects**, N (%)
First dose	6 (2.0)	5 (3.3)	0	1 (1.6)	0	0.726
Leaving work following vaccine administration§, N (%)
First dose	82 (22.3)	52 (29.9)	4 (4.8)	19 (29.2)	7 (15.2)	<0.001
Second dose	48 (15.6)	37 (22.4)	4 (5.3)	0	7 (15.9)	0.001
COVID-19 after vaccination§§	32 (8.7)	16 (9.2)	5 (6.0)	6 (9.2)	5 (10.9)	0.778

^*Among those participants who received two doses.

^**Among those participants with side effects.

^***Among those participants had side effects and received medications to relieve symptoms.

^****Among those participants who had side effects with both doses.

^§Among those participants who received the vaccine dose.

^§§Among the study population.

3.3. The start time and duration of side effects

3.3.1. First dose

As shown in Table 2 and Supplemental Table S4, the distribution of the onset time and duration of the first dose's SEs differed significantly between the vaccine groups. When these factors were accompanied by the central and diversity indices—namely, median (IQR)—as illustrated in Supplemental Table S5 , more detailed profiles of these variables were provided. The results of two-by-two comparisons of the medians (IQR) determined that the SE onset times, from fastest to slowest, were as follows: Sinopharm > Sputnik V = Oxford–AstraZeneca = Covaxin. Moreover, the SE duration, from longest to shortest, was as follows: Covaxin > Sputnik V ≥ Sinopharm = Oxford–AstraZeneca.

Among the entire population, eye pain, local warming, and injection site pain had the fastest onset times, while skin rash, itching, and aphthous stomatitis had the slowest onset times. Also, in the vaccine groups, the fastest and slowest onset times were associated with local warming and depression (Sputnik V), local stiffness and rhinorrhea (Sinopharm), local warming and eye redness (Oxford–AstraZeneca), and eye pain and rhinorrhea (Covaxin). According to the two-by-two comparisons, chills were initiated significantly faster in the Sputnik V group than in the Covaxin group, rhinorrhea was initiated significantly faster in the Sputnik V group than in the Sinopharm group, and injection site pain was initiated significantly faster in the Sinopharm group than in the Oxford–AstraZeneca group. No significant differences were observed in other comparisons.

An examination of SEs duration revealed that skin rash and fever had the longest and the shortest duration, respectively. Furthermore, considering the vaccine groups separately, the longest and shortest durations were associated with depression and fever (Sputnik V), other symptoms and rhinorrhea (Sinopharm), local redness/swelling and xerostomia (Oxford–AstraZeneca), and joint pain and rhinorrhea/local swelling (Covaxin). The two-by-two comparisons determined significant differences between the vaccine groups regarding the duration of joint pain, injection site pain, headache, fatigue, rhinorrhea, chills, and fever.

3.3.2. Second dose

As shown in Table 2 and Supplemental Table S4, the distribution of the onset time and duration of the second dose's SEs differed significantly between the vaccine groups. When accompanied by the central and diversity indices—namely, median (IQR)—a more detailed profile of onset time and duration was provided as illustrated in Supplemental Table S6. According to the two-by-two comparisons of medians (IQR), the onset times of SEs, from fastest to slowest, were as follows: Oxford–AstraZeneca > Sputnik V = Sinopharm = Covaxin. Moreover, the SE duration, from longest to shortest, was as follows: Covaxin = Sinopharm = Sputnik V > Oxford–AstraZeneca.

Among the second-dose recipients, local swelling, hypotension, itching, and injection site pain were the SEs with the fastest onset times; meanwhile, petechiae, aphthous stomatitis, and depression had the slowest onset times. When considering the vaccine groups separately, the fastest and slowest onset times were associated with hypotension and palpitation (Sputnik V), injection site pain and depression (Sinopharm), injection site pain and dizziness (Oxford–AstraZeneca), and local redness and petechiae (Covaxin). A significant difference was observed in the onset time of fever between all vaccines. Moreover, fatigue was initiated significantly faster in the Oxford–AstraZeneca group than in the Sinopharm group.

An examination of SE duration indicated that petechiae and fever had the longest and shortest durations, respectively. When considering the vaccine groups individually, the longest and shortest durations were associated with palpitation and fever (Sputnik V), depression and sweating (Sinopharm), local stiffness at the injection site/body or muscle pain/fatigue and chills (Oxford–AstraZeneca), and petechiae and sweating (Covaxin). The two-by-two comparisons indicated significant differences between the vaccine groups in terms of the duration of injection site pain, headache, chills, and fever.

3.4. Medication use to relieve symptoms

Among the first- and second-dose recipients who had SEs, 199 (69.1 %) and 92 (54.1 %) took at least one drug to relieve the symptoms, respectively (Table 2). The frequency of medication use to relieve symptoms differed significantly between the vaccine groups after the first dose (p < 0.001) but not after the second dose (p = 0.246). According to the two-by-two comparisons (Supplemental Table S4), in terms of medication use after the first dose, the Oxford–AstraZeneca and Sputnik V groups were significantly different from the Sinopharm and Covaxin groups. However, no significant differences were observed between the Sinopharm and Covaxin or Sputnik V and Oxford–AstraZeneca groups regarding medication use (Oxford–AstraZeneca = Sputnik V > Covaxin = Sinopharm).

In the whole population, the number of days during which medication was used did not differ significantly between the vaccine groups for the first dose (median [IQR] = 1 [1–2] day; p = 0.766) or the second dose (median [IQR] = 1 [1–2] day; p = 0.410). Acetaminophen was the most widely used medication by participants to relieve SE symptoms. Detailed information regarding the frequency of medication types is provided in Supplemental Table S7.

Among 199 participants who developed SEs after the first dose and took medication, the symptoms of 125 participants (62.8 %) were alleviated in less than 4 h; for the other 74 participants (37.2 %), it took more than or equal to 4 h for symptoms to subside (p = 0.497). Among the 92 participants who experienced SEs after the second dose and took medication, 55 (59.8 %) experienced reduced symptoms in less than 4 h, and 37 (40.2 %) experienced reduced symptoms in more than or equal to 4 h after taking medicine (p = 0.411).

3.5. Recipients' point of view regarding severity, duration, and number of side effects

Vaccine recipients' attitudes toward the SEs associated with the first and second doses of the vaccines show that the first dose of Oxford–AstraZeneca had the most severe and longest-lasting symptoms when compared with the second dose of the same vaccine, as well as both doses of the other vaccines (Supplemental Table S4). Regarding the number of SEs, the only significant difference is that the first dose of Oxford–AstraZeneca led to significantly more SEs than the first dose of Sinopharm (p = 0.017). As shown in Table 2, participants who experienced SEs after both doses (n = 146) expressed that the SEs caused by the first dose were more severe (p = 0.148), longer-lasting (p = 0.021), and more numerous (p = 0.169) than those caused by the second dose.

3.6. Vaccination outcomes

Of the 288 participants who experienced SEs after the first dose, 6 (2 %) were hospitalized as a result (five from the Sputnik V group and one from the Oxford–AstraZeneca group). None of the participants were hospitalized due to the second dose. An investigation of the causes of hospitalization indicated an appropriate indication for hospitalization in only one case in which the participant's blood pressure dropped, accompanied by fever, chills, and body pain after the first dose of the Sputnik V vaccine. In the other cases, the reason for hospitalization of the HCWs was the hospital availability rather than an urgent need for healthcare services.

Out of the 368 participants who received the first dose, 82 (22.3 %) had a degree of severity in symptoms that led to leaving the workplace. In the second dose, this was 48 (15.6 %). Therefore, there was a significant difference between vaccines in both doses (p ≤ 0.001). In the two-by-two comparisons of the first dose (Supplemental Table S4), it was determined that the frequency of leaving the workplace is as follows: Sputnik V = Oxford–AstraZeneca ≥ Covaxin = Sinopharm. In the second dose, it was Sputnik V = Covaxin ≥ Sinopharm = Oxford–AstraZeneca.

3.7. The relationship between the occurrence of side effects and demographic information

As shown in Supplemental Table S8, the median (IQR) age of participants who received both doses and reported SEs caused by at least one dose (31 [27.0–37.0] years old) was significantly lower than those without SEs (33 [29.0–40.0] years old); p = 0.032. A more detailed analysis revealed that this difference was significant only in the Sputnik V group (31 [27.7–36.0] vs 39 [30.5–42.5] years old); p = 0.012. Furthermore, among the first-dose recipients of the Sputnik V vaccine, the prevalence of SEs was higher in those aged ≤ 31 years (n = 84, 93.3 %) than those aged > 31 years (n = 66, 80.5 %); p = 0.012. In addition, comparisons between each reported SE and age showed that, among participants who had SEs, only fever and chills were significantly more prevalent among the first-dose recipients of Sputnik V aged ≤ 31 years than those aged > 31 years (Supplemental Table S9).

Similarly, in the Sputnik V group, the median (IQR) of BMI was significantly lower among those with SEs (24.2 [21.9–26.0] kg.m⁻²) than those without SEs (27 [23.5–30.1] kg.m⁻²); p = 0.015. A similar difference was found for the median (IQR) of years of employment (5 [3.0–11.0] vs 12 [6.0–17.5] years); p = 0.008. No significant relationships were observed for other vaccines.

The underlying disease, medication intake, gender, and history of COVID-19 were not associated with vaccine SEs in the whole population and when considering the vaccine groups separately. As shown in Supplemental Table S10, in the Sputnik V vaccine, the number of SEs per participant following the first dose was higher in women with underlying diseases than those without such diseases (median [IQR] = 7 [4.5–11.2] vs 4 [2.0–6.0]; p = 0.001). This relationship was neither observed in the male population nor in the second dose.

4. Discussion

In this survey-based study on 368 HCWs from a tertiary hospital in Iran, we evaluated the SEs of four available vaccines (Sputnik V, Sinopharm, Oxford–AstraZeneca, and Covaxin) through face-to-face interviews. Most of the analyses were complemented with two-by-two comparisons between the vaccine groups. Of the total population, 83.1 % reported at least one SE. More participants experienced SEs after the first dose (n = 288, 78.3 %) than after the second dose (n = 170, 55.2 %). This lower reactogenicity after the second dose is consistent with previous studies [[20], [30]]. Fever (54.5 %), body/muscle pain (53.5 %), and injection site pain (50.3 %) were the most common symptoms following the first dose. Meanwhile, injection site pain (58.2 %), body/muscle pain (39.4 %), and headache (36.5 %) were the most common symptoms following the second dose. Following the first dose, the fastest start time of SEs was associated with Sinopharm, and the longest SE duration was associated with Covaxin. Following the second dose, AstraZeneca had the fastest onset time and the shortest duration of SEs.

Moreover, our findings indicate more significant between-group differences for systemic SEs than local SEs. Two-by-two comparisons showed that Sputnik V and Oxford–AstraZeneca led to more common systemic SEs than Sinopharm or Covaxin (Supplemental Table S2 ). Furthermore, oral and skin reactogenicity events were rare among our paticipants, and Covaxin recipients did not report any oral SEs. Although oral and skin SEs were observed as resulting from mRNA-based COVID-19 vaccines, a previous study [31] showed that Oxford–AstraZeneca, a viral vector vaccine, has a higher prevalence of oral and skin SEs than Pfizer-BioNTech and Moderna, which are mRNA-based vaccines.

4.1. Sputnik V

Our results show that 87.4 % and 62.4 % of people receiving Sputnik V developed SEs after the first and second doses, respectively. Meanwhile, in another study on HCWs in Iran, these percentages were 67.8 % and 32.2 % [32]. Both studies showed a reduction in the frequency of SEs after the second dose. However, in one study with a mean age of vaccine recipients of 66 ± 14 years, these percentages were 53.3 % and 66.8 % [33].

In our study, 92.1 % of Sputnik V recipients had at least one SE, which is higher than the 82.7 % reported in a previous real-world study [30] and the 76.0 % indicated in a study with a predominantly geriatric population [33]. Additionally, in two studies that only reported SEs following the first dose of Sputnik V, the prevalences were 71.3 % [34] and 81.9 % [35]. The Sputnik V trial did not clearly mention what percentage of participants suffered adverse events, as it cited “7966 adverse events in 12,296 participants who received both doses [18].” The high prevalence reported in our study is probably due to the relatively young age of our participants and different study design (i.e., face-to-face interviews).

COVID-19 SEs' associations with sex and age have been addressed in previous research [30]. According to the literature, females show more SEs than men after the first dose, whereas men report more SEs following the second dose [30], [32]. Furthermore, younger participants who received the Sputnik V vaccine in a study from Iran had a greater chance of showing SEs than older individuals [30]. Interestingly, our study revealed a direct relation between a subgroup of females and SEs (i.e., females with underlying diseases reported more SEs than those without such diseases). Consistent with our finding, previous research shows that the presence of chronic diseases is correlated with the development of COVID-19 post-vaccination SEs [36]. In another stydy, suffering from chronic diseases was significantly associated with the frequency and severity of post-vaccination SEs [37]. However, no information explaining this correlation is available. Moreover, in contrast with this correlation, a study on Sputnik V found that people with underlying diseases had a lower prevalence of SEs than those without underlying conditions [35]. The researchers explained that using analgesic and anti-inflammatory drugs or having a higher threshold for pain or discomfort could explain this outcome.

Additionally, we found that participants with SEs had a lower BMI than those without SEs. A cross-sectional study in Spain [38] on participants who received Pfizer-BioNTech, Moderna, Oxford–AstraZeneca, or Janssen showed that a higher risk of some SEs (fevers of at least 38 °C, vomiting, diarrhea, and chills) among underweight people and those of a normal weight in comparison with overweight (including obese) individuals. Another study in Iran showed that the prevalence of SEs was higher following the first dose of Oxford–AstraZeneca and Covaxin (but not Sputnik V) in persons with a BMI above 25 compared to those with a BMI below 25 [35]. The present study indicated that participants with fewer years of employment had a higher chance of showing SEs than those with more years of employment. However, this is probably because participants with few years of work experience tend to be young.

In our study, the most prevalent symptoms reported after the first and second doses of the Sputnik V vaccine were body/muscle pain, fever, injection site pain, fatigue, and headache. This outcome is similar to that reported by another study from Iran, with some changes in the order [32]. Injection site pain had a similar prevalence between the two studies, but our study showed that this symptom was more frequent after the second dose (56.3 %) than the first dose (54.6 %). However, the reduction of joint and body pain after the second dose was similar between the two studies. Although the frequencies of depression, swelling, and redness at the injection site were lower in our participants than in those in the mentioned study, these SEs were uncommon among the Sputnik V recipients in both studies.

A comparision of the frequency of swelling and redness at the injection site showed similar frequencies in the first dose between studies, but the decrease in the frequency of the second dose compared to the first dose in our study was not observed in the mentioned study [32] or another study from Iran [30]. Furthermore, our findings showed no significant difference between Sputnik V and Oxford–AstraZeneca regarding the prevalence of SEs. However, this finding is inconsistent with previous studies, where Sputnik V was associated with a higher frequency of symptoms than Oxford–AstraZeneca [30] and led to fewer SEs than Oxford–AstraZeneca [35].

4.2. Sinopharm

The present study showed that Sinopharm was associated with the lowest prevalence of SEs after both the first and second doses (57.8 % and 40 %, respectively). This result is similar to the findings in Jordan [39] from an investigation of the Oxford–AstraZeneca, Pfizer-BioNTech, and Sinopharm vaccines (46.1 % and 52.8 %, respectively). In this regard, a recent meta-analysis of clinical trials revealed that inactivated vaccines, such as Sinopharm, have the lowest risk for occurrence of adverse events [40]. Additionally, the prevalences of SEs in the United Arab Emirates population following the first and second doses were 75.6 % and 86 %, respectively [41]. Inconsistent with these two studies [39], [41], our study indicates a lower frequency of SEs in the second dose of Sinopharm than in the first dose. Although the overall prevalences of SEs in a real-world study [30] and a phase 1/2 clinical trial [13] were 37.4 % and 39.0 %, respectively, our study revealed an overall prevalence of SEs that was close to that of the Iraqi population (66.7 % and 61.1 %, respectively), with a similar sample size [42].

On the other hand, we found that the injection site pain is the most common SE of Sinopharm, in line with previous findings [39], [41], [42]. Similar to other studies [30], [42], our participants' other most frequent SEs were headache, fatigue, body/muscle pain, and fever, with a different order between the two doses. In reporting the number of SEs per participant, similar studies included participants with no SEs in their calculations [30], [39] and, therefore, confounded the accurate number of occurred SEs, making any comparison between the current study and similar studies inaccessible an unimpeachable.

4.3. Oxford–AstraZeneca

Since there is an eight- to 12-week interval between administering the Oxford–AstraZeneca (AZD 1222) doses [43], most researchers have evaluated the SEs of only the first dose [35], [44], [45], [46], [47], [48], [49] rather than the SEs of both doses [39], [50]. Our study reported that Oxford–AstraZeneca was associated with the highest prevalence of SEs following the first dose (93.8 %). In a survey of Jordanian HCWs evaluating the SEs of the Oxford–AstraZeneca, Pfizer-BioNTech, and Sinopharm vaccines, Oxford–AstraZeneca led to the highest prevalence of SEs following the first and second doses (97.8 % and 98.3 %, respectively) [39]. Moreover, a large-scale study from the UK [49] showed that systemic SEs of the first dose of Oxford–AstraZeneca (33.7 %) were significantly higher than those of the first (13.5 %) or second (22.0 %) doses of Pfizer-BioNTech. However, another study from Iran [30] reported that Sputnik V led to more SEs than Oxford–AstraZeneca and Sinopharm, indicating some diversity in the prevalence of SEs in previous studies. HCWs in Ethiopia [50] reported SE prevalences of 91.3 % and 67.0 % following the first and second doses of Oxford–AstraZeneca, respectively, which supports our findings. Furthermore, some studies reported a prevalence of 57.0 % [51], 66 % [30], 69.7 % [52], 93.5 % [44], and 98.1 % [46] following the first dose administration.

Differences in study design, sample size, heterogenicity of the population, and ethnicity are among the essential contributors to the diversity in the findings noted above [47]. The female sex, a young age, and comorbidities were associated with the prevalence of SEs in Oxford–AstraZeneca recipients [47], [49]. Although we did not find any relation between these attributes and SEs, the young age of our population seems to be the most critical contributor to the high prevalence of SEs in our study. In line with this, a clinical trial phase 2/3 Oxford–AstraZeneca [16] showed that participants aged 18–55 years had the highest prevalence of systemic SEs (86 %).

In our study, fever, body/muscle pain, chills, fatigue, headache, and injection site pain were the most common SEs of Oxford–AstraZeneca, with the order of these symptoms varing between doses. Among them, headache, fatigue, and chills were also observed as the most frequent systemic SEs in a large-scale study [49] conducted in the UK. Injection site pain was reported as the most common SE, though only the first dose was considered. Furthermore, injection site pain, fatigue, headache, feverishness, and myalgia were the most common SEs in the Oxford–AstraZeneca clinical trial phase 2/3 [16].

Our findings are also consistent with similar studies [39], [44], [50], [51], though the order varies, as injection site pain was more common following the second dose than the first one. This reactogenicity can be attributed to various factors [53] related to the administrators, vaccines, and recipients that can not necessarily explain this difference between doses. Educating vaccine recipients, providing an appropriate environment for vaccination, and using suitable injection methods can reduce injection site pain [53]. In contrast to previously reported Oxford–AstraZeneca-related cases [54], we did not find laboratory confirmation of coagulation disorders or deaths caused by this vaccine.

4.4. Covaxin

We reported that 70.5 % of people who received at least one dose of Covaxin developed at least one SE (58.7 % and 54.5 % after the first and second doses, respectively). In comparison, a study in India indicated SEs in only 29.8 % of vaccine recipients (38.1 % and 26.4 % after the first and second doses, respectively) [55]. Furthermore, interim results of a Covaxin trial [15] indicated that the SE prevalence was 33 %. Injection site pain, headache, body/muscle pain, and fatigue were the most frequently reported SEs in our study.

These SEs are similar to those reported in other studies on Covaxin [15], [35], [55], albeit with different orders and prevalences. Furthermore, the rates of SEs in two of these previous studies were lower than the rates we reported [15], [55] but higher than indicated in the study from Iran [35]. These differences between studies may be due to age, comorbidity, and data collection method differences [35], [55]. Compared with another study from India [55], our study had a younger population, more comorbidities (23.9 % vs 5.8 %), and a more detailed data-gathering method (face-to-face interview vs a telephone interview), which may be responsible for the higher prevalences.

Finally, injection site pain was more common after Covaxin than the other vaccines considered in our study. This outcome is consistent with the findings of another study comparing Covaxin with Covishield [56].

4.5. Other characteristics of side effects

Our findings revealed that most SEs after the first and second doses occurred within 12 h of the injection (84.8 % and 78.9 %), and most disappeared within 72 h after vaccination (83.5 % and 76.5 %), respectively. Other studies reported similar distributions [32], [33], [42], [47], [50]. Health authorities recommend taking over-the-counter medicines, such as acetaminophen or ibuprofen, to relieve post-COVID-19 vaccine symptoms [57]. More importantly, the short-term use of non-prescription doses of analgesics/antipyretics may increase vaccine uptake by reducing SEs without affecting the vaccine's immunogenicity or efficacy [57]. In line with this, most of our participants (69.1 %) took analgesics, especially acetaminophen, after experiencing SEs from the first dose. Analgesic use was significantly more common following the first dose of Oxford–AstraZeneca and Sputnik V than the first dose of Sinopharm or Covaxin.

Five recipients of Sputnik V and one recipient of Oxford–AstraZeneca were hospitalized following the first dose. We evaluated their conditions and medical records and found that the hospitalization of only one participant, who belonged to the Sputnik V group, was warranted. Hospitalization following Sputnik V administration was previously reported [34]. Altogether, given the high incidence of leaving work following vaccine administration among first-dose recipients of the Sputnik V and Oxford–AstraZeneca vaccines, these two vaccines appear to have more severe SEs than Sinopharm and Covaxin following the first dose. Furthermore, participants' attitudes toward the first dose of Oxford–AstraZeneca were negative in terms of the severity and duration of SEs. However, after considering the other data regarding the severity and duration of SEs, we could not conclude that Oxford–AstraZeneca had more severe or longer-lasting SEs than Sputnik V.

4.6. Strengths and limitations

The main strength of our study is the inclusion of all available vaccines at the time of the study; thus, this study adequately reflected a comprehensive safety profile of non-mRNA-based COVID-19 vaccines. Moreover, most analyses were complemented by two-by-two comparisons to compare different aspects of safety. Furthermore, since data were collected through face-to-face interviews, the participants provided plenty of details about the safety profiles of the vaccines and their characteristics. As a result, the problem of missing data was not encountered, which is a major problem in other methods, such as self-repors.

Nevertheless, the present study included some limitations that should be acknowledged. First, this study was conducted in the spring of 2021, during one of the peaks of the COVID-19 epidemic in Iran. Therefore, due to the heavy workload at the time, it was impossible to include all HCWs involved in treating patients, especially physicians. Second, electronic forms and efficient software that can transfer the collected information to analysis software online were not available. Thus, the processes of defining study variables, entering information, and checking their accuracy were time-consuming, thereby delaying the presentation of the results. Third, due to resource limitations and the study's cross-sectional nature, it was impossible to follow up with some participants, especially the Oxford–AstraZeneca recipients, to collect data related to the second dose. This limitation might have affected the findings on the prevalence of SEs and other variables related to the second dose. Designing studies [58] according to the template provided by the WHO [59] could mitigate the loss of follow-ups and provide the capability to follow up on long-term SEs. Finally, we conducted this study on HCWs, making it difficult to generalize our findings.

5. Conclusion

Our study revealed that all four vaccines had an acceptable safety profile and were well tolerated among HCWs, though Sputnik V and Oxford–AstraZeneca had a greater prevalence of SEs, number of SEs per individual, and severe SEs than Sinopharm and Covaxin. Overall, the prevalence and severity of SEs were higher following the first dose than the second dose. SEs often started within 12 h and subsided within 72 h of injection. If needed, vaccine recipients should take analgesics/antipyretics to overcome the symptoms. Additional studies are required to evaluate the SEs following the third and fourth doses of available vaccines in different populations. Moreover, studies that begin data collection immediately after a vaccine injection would provide more precise and accurate results. In the current situation, in which multiple SARS-CoV-2 mutations have emerged, safety studies on vaccines with high immunogenic profiles are particularly important.

CRediT authorship contribution statement

Sahar Oghazian: Conceptualization, Methodology, Validation, Investigation, Data curation, Writing - original draft. Taraneh Tavanaei Tamanaei: Conceptualization, Methodology, Resources, Writing - original draft. Ramin Haghighi: Conceptualization, Methodology, Investigation, Resources. Mojdeh Faregh: Investigation, Resources. Mohammad Bagher Oghazian: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Visualization, Supervision, Project administration, Funding acquisition, Writing - original draft, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.