The study aimed to report the anxiety attributed to COVID-19, self-rated disease knowledge, and intention to vaccinate against the disease among general public. Moreover, the interplay among these three outcomes was also investigated.

This study was designed as a cross-sectional survey and was conducted for 2 months, i.e., September–October 2021 in three cities of the Dammam Region of Saudi Arabia.

The target segment was the adult population. Adult male and female participants who aged 18 and above and eligible for a vaccine, were invited to participate in this study. Incomplete responses were excluded from study.

The data was collected using a convenient sampling method. Participants who could be conveniently approached were contacted. The mode of survey was physical as well as online. Both forms were available as per the convenience of participants. The online survey was conducted using an electronic tablet, while the hardcopies were handed to participants and later collected. The venues selected were mostly open public spaces such as public parks and open areas of shopping malls. The data was collected once per participant and there was no follow-up.

The sample size was calculated using an online sample size calculator. Dammam Region consists of several cities. Our study included cities namely Dammam, Dhahran, and Khobar. According to available estimates at the time of this writing, the population of Dammam city was 7,68,602 while Dhahran and Khobar had a population of 99,540 and 1,65,799 respectively (18). Since there was no description of numbers related to the vaccine eligible population, the sum of all three figures, i.e., population of 1,033,941 individuals, was considered the total population of these cities and thus, our target population.

The sample size was calculated using a margin of error of 5% and a confidence interval of 97.5%. From the initial calculation, the number of samples obtained was 503 (19). Later, the sample size was adjusted for non-response/missing error rate which was considered at 20%. Finally, the required sample size turned to 629.

The research questionnaire consisted of a demographic section, a section dedicated to vaccination intention and, a section containing Coronavirus Anxiety Scale (CAS) (20). The demographic section contained questions related to the age, sex, nationality, education level, marital status, occupation, residence, and income. The second section contained items related to the participants' medical history, any exposure to COVID-19, self-rated knowledge about COVID-19, level of compliance to recommendations aimed at preventing COVID-19 spread, and vaccination intention. The third section was CAS itself, a validated scale to measure anxiety related to the disease. The scale contained 4 items related to the COVID-19 anxiety on a person's daily life in last 14 days. The scale was Likert-format, and each item had 5 possible options. Each options awarded a score. A cumulative score of ≥ 9 indicated COVID-19 anxiety (20).

A formal permission was obtained from the developer of the scale through email. The questionnaire was formulated in native Arabic and English languages. The CAS scale was already available in both English and Arabic languages at the time of study. The survey was piloted in few participants before actual study.

Data were checked for incomplete and missing responses. At the beginning 558 responses were received and 71 samples of incomplete responses were excluded from the data set. Of 23 partially incomplete responses, 16 were excluded due to untreatable nature, whereas 7 (seven) responses were treated using the “last observation carried forward” statistical method. Finally, a total of 542 complete responses were analyzed. The potential sources of bias considered during this study were selection bias due to the convenient sampling, and information bias attributed to the self-reporting format of this study. The study outcomes were the intention to vaccinate and COVID-19 anxiety. A secondary outcome considered was COVID-19 self-reported knowledge.

The data were coded and entered for analysis in IBM SPSS version 23. The demographic data was reported using sample counts (N) and percentages (%) for descriptive data. Simple and multiple logistic regression methods were used to report the determinants of the outcomes. Only the significant variables were included in the multiple regression model. The significance of those variables was determined by simple regression analysis and necessary model fitness parameters were checked by required statistics that are mentioned in footnotes of Tables 4–6.

The participants were briefed about the study and their consent was sought. The participation was voluntary. An electronic informed consent was designed and was shown at the front page of the online survey, and it was also available in hardcopy. The participants in the online survey could only access the survey if they consent to participate while those who preferred the hardcopy were asked to provide their consent before accessing the survey. The nature of consent was implied, i.e., participants were not required to provide their personal identifiable details. The study was approved by the Institutional Review Board of Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia (IRB -2021-05-297).

Most respondents were female (N = 352, 64.9%) and aged between 18 and 29 years (N = 306, 56.5%). Most of them were Saudi nationals (N = 513, 94.6%), students (N = 219, 40.4%), resided in urban area (N = 457, 84.3%) and had postgraduate qualification (N = 230, 42.4%). Slightly less than half (N = 237, 43.7%) had a monthly family income above Saudi Arabian Riyal (SAR) 10,000, i.e., > United States Dollar (USD) 2666, (Table 1).

Most participants did not have any chronic illnesses (N = 449, 82.8%) and mental illnesses (N = 492, 90.8%). More than half of participants rated their knowledge of COVID-19 as good (N = 304, 56.1%). The majority did not suffer from COVID-19 (N = 458, 84.5%) however, had COVID-19 patients in their family (N = 469, 86.5%). Most of the participants never missed a doctor recommended vaccine (N = 508, 93.7%) and intended to vaccinate against COVID-19 (N = 311, 57.4%) (Table 2).

Slightly less than half of the participants mentioned that they follow the recommendations from authorities regarding COVID-19 prevention at most times (Figure 1).

The majority reported an acceptable level of knowledge of COVID-19 viral infection (N = 432, 79.7%). More than half of the participants had positive intent regarding COVID-19 vaccination (N = 311, 57.4%). Most of them had no anxiety attributed to COVID-19 (N = 499, 92.1%). The reliability of the Coronavirus Anxiety Scale (CAS) was 0.845, i.e., Cronbach's alpha value (Table 3).

The model for COVID-19 anxiety highlighted those individuals in age groups 18–29 and 30–45 years were more likely to have COVID-19 anxiety, i.e., (AOR 5.33) and (AOR 6.48), respectively, compared to individuals aged 46 years and above, when adjusted for other demographics (p < 0.05). Besides, individuals who mentioned having any long-term mental health condition were roughly three times more likely to have COVID-19 anxiety (AOR 2.66) compared to those without any mental illness when other demographics are considered (p < 0.05). Some of the variables were significant determinants of COVID-19 anxiety alone and became non-significant when adjusted for all different demographics (Table 4).

The model for a positive intention toward COVID-19 vaccination reported that compared to the females, the male respondents had a higher likelihood of positive intention (AOR 2.27) after adjusting for all demographic characteristics (p < 0.001). Besides, those who were single (AOR 1.77) had a higher likelihood of positive intention (p < 0.01). Further, compared to those individuals having a monthly income above SAR 10,000, those with an income of SAR 5,000 (AOR 0.59) and SAR 7,500–10,000 (AOR 0.57), had negative intention when adjusted for other demographics (p < 0.05). Individuals who rated their knowledge of COVID-19 as good were more likely to vaccinate (AOR 1.78), while those who never contracted the disease showed positive intention (1.74) toward vaccination when adjusted for all other demographics (p < 0.05). Some of the variables were significant determinants of positive intention alone and become non-significant when adjusted for all other demographics (Table 5).

The model for self-rated good knowledge of COVID-19 disease revealed that compared to individuals who lived in rural localities, individuals living in urban areas were more likely to rate their knowledge as good (AOR 1.71) when other demographics were considered (p < 0.05). Besides, individuals who did not have any long-term mental health condition were three times more likely to rate their knowledge of COVID-19 as good (AOR 2.74) when adjusting other demographics (p < 0.01). Moreover, individuals who intended to receive COVID-19 vaccine were more likely to rate their knowledge of the disease as good (AOR 1.75) when other variables were considered (p < 0.05). Some of the variables were significant determinants of positive intention alone and become non-significant when adjusted for all other demographics (Table 6).

Our findings highlight that the self-rated knowledge of COVID-19 among most participants was at an acceptable level. This confirmed the findings of previous studies which reported that public in Saudi Arabia is quite aware of the infection (26–28). However, one of the study reported that it may be higher among individuals with higher education and income (28). In our study, most participants had higher education and income. To this end, this study adds to the existing body of literature by carrying out the analysis to report the determinants of self-rated good knowledge of COVID-19. The residence may impact an individual's life. People living in urban areas may have more opportunities for obtaining education and finding better work as most educational institutes and work related opportunities are usually concentrated in urban areas. Our findings reveal that individuals who resided in urban localities were more likely to rate their knowledge as good compared to those living in rural areas. Besides, individuals with a higher degree such as bachelor or a postgraduate qualification were more likely to rate their knowledge as good compared to others. However, the variable of higher education was only a significant determinant when other demographics were not considered. Further, our data could not establish the aspect of income as a significant determinant of self-rated good knowledge of the infection. In another study, it was reported that older participants were more knowledgeable (26). However, in our study we found that participants aged 30 years or younger were more likely to rate their knowledge as good. Though, the variable of age was only significant when other variables were not considered.

Our findings highlight that individuals who did not suffer from any long-term mental health condition were more likely to rate their knowledge of COVID-19 as good. This aspect of viewing mental health issues alongside knowledge of disease was not reported from this population. This finding also raises the question of evaluating the impact of mental health on disease knowledge. In this context, an interesting finding was reported by Al Dhaheri et al. (29) that the disease had some level of psychological impact. It was also observed in our study that the individuals who intended to receive a vaccine for the disease were roughly two times more likely to rate their knowledge as good.

In regards to the intention to receive a vaccine for COVID-19, more than half of the target segment, i.e., 57.4% had a positive intention to receive a vaccine. This was similar to the results reported by previous studies (30, 31). In addition, the data highlights that the male respondents and those who never contracted the disease before, had a higher likelihood of positive intention as compared to female respondents and individuals with a history of COVID-19 infection, respectively. Being male and having no history of COVID-19 infection were previously reported as predictors of positive intention in this population (31). In addition, our data highlighted that individuals who rated their knowledge of COVID-19 as good were more likely to vaccinate. A possible explanation for this occurrence could be that better understanding about the disease may prompt an individual to seek a positive change, i.e., treatment. Hence, patients who rated their knowledge as good understood the disease and its treatment. Therefore, they indicated their acceptance for vaccination.

It was also observed that most respondents had no anxiety owing to the viral infection. Individuals with long-term mental illnesses appeared prone to having COVID-19 anxiety. This was in line with results of a previous study where it was reported that the disease had a psychological impact (29). Therefore, it is likely that respondents in our study who suffered from mental illness were more anxious. Our study reported that younger individuals were more likely to have anxiety as compared to the older ones. To this end, a previous study reported a high level of anxiety in young adults between 18 and 29 years during COVID-19 resultant lockdown in Saudi Arabia (32). Thus, our findings highlight the presence of continued anxiety in this age group. This age group usually contains individuals who are either involved in their studies or are working. COVID-19 has caused disruptions in education and work sectors such as switching to online education (33). It has also resulted in switching to work from home (34). This coupled with new outbreaks and emergence of new variants of the virus may continuously bother such individuals. It is extremely important to address the mental issues pertaining to individuals in this age group.

The study has some limitations, due to the convenient nature of data collection, there may be a selection bias as the researchers gather data from individuals who could be easily approached. Although the study gathered data from a large number of participants with reasonable demographic diversity, the findings should be interpreted with caution. This also affects the generalizability of the findings. However, since it was carried out in several sites at different times of day. This would have offered a considerable level of random data collection within the convenient realm. Therefore, it is expected that the findings may be limited to local settings and may not be generalized.

Secondly, there may an element of information bias attributed to the self-reporting format of this study particularly with answers related to the COVID-19 disease anxiety, disease knowledge, compliance to social distancing protocols provided by the health authority. We believe that there may be a slight overestimation by respondents as they may have tended to select a socially acceptable response. Although, they were informed that the data was collected without any personal identifier, and they cannot be identified once the response has been submitted. Besides, it was explicitly requested to provide an honest response for this exact purpose. Moreover, some of the findings of this study were similar to the ones reported previously. Lastly, study gathered data from a small number of non-Saudi respondents and therefore, all comparisons made based on nationality must be interpreted with caution.