The COVID-19 pandemic is said to be the impetus of a sequelae of mental health issues [1
]. While any flu-like symptom increases psychological distress, the impact of the COVID-19 pandemic on mental health was anticipated to be more serious. Among the psychiatric disorders, anxiety and depression are said to be the leading manifestation [2
], with a high probability of increased new-onset health-related anxiety [3
]. Health anxiety is defined as worries and anxiety due to perceived threat to one’s health. This is conceptualized as a dimensional construct on a continuum that ranged from the absence of health awareness to disordered health anxiety such as illness anxiety disorder [4
]. There were also reports on the emergence of “psychogenic COVID-19” described as a conversion reaction that presented with upper respiratory tract symptoms and psychogenic fever [7
], which may be within the continuum of health anxiety.
The pandemic adds an emotional burden to the infected person through its neuropsychiatric sequelae associated with the viral infection and its medical management [8
]. Some people who underwent quarantine experienced many psychological effects from trauma, social isolation, and stigma [3
]. Anxiety is also reflected in the society as seen from reported public behaviour, such as panic buying and hoarding of essential and non-essential items [9
]. The sudden change in lifestyle, including restricted traveling, working or studying at home, social distancing, and self-isolation, may also contribute to the rise of mental health issues [1
It is hypothesized that the pandemic will have a considerable psychological impact on HCW in particular as they may have perceived themselves to be at highest risk of exposure to infection. This fear of being infected with COVID-19 possibly due to concerns about their own vulnerability in contracting the disease, worries about inadequate medical supply or protective gears, and/or long working hours. This is supported by an early study of COVID-19 cases in China, where a high rate of 3.8% of exposed HCW (1716 of 44,672) were infected [8
]. World Health organization reported that 14%–35% of COVID-19 cases reported to WHO were among HCW [9
]. In a previous MERS-CoV outbreak, a study reported a significant number of hospital workers in Saudi Arabia had anxiety of acquiring the infection [10
]. Similarly, a recent multi-center survey in China involving 1563 medical staff found the prevalence of depression, anxiety, insomnia, and stress-related symptoms to be 50.7%, 44.7%, 36.1%, and 73.4%, respectively during the COVID-19 pandemic [11
]. Attention to the psychological health of frontline healthcare workers is therefore crucial, as their clinical services are much needed for the care of sick patients.
In Malaysia, the COVID-19 pandemic led to a nationwide Movement Control Order (MCO) implemented by the government beginning 18 March 2020. It was enforced in phases until it entered a recovery phase starting on 10 June onwards [12
]. The worldwide shortage of personal protection equipment (PPE) during this early phase affected Malaysia’s HCW as well, which was an important source of distress. Having accurate knowledge and a positive attitude with the correct practice of managing the risk of infection may be fundamental to good mental health of the HCW. In turn, HCW with good emotional wellbeing are reliable advocators and change agents in their families and communities. Nevertheless, the extent of knowledge, attitude, and practice regarding COVID-19 of the HCW in relation to their tendency to health anxiety has not been explored. This study specifically investigates the correlations between health anxiety traits and knowledge, attitude, and practice of Malaysian healthcare workers (HCW) during the initial phases of the first Movement Control Order (MCO) in Malaysia, i.e., between 18 March and 28 April 2020. This is important to understand further measures that could be implemented to ensure the concordance of physical and mental health wellbeing of the HCW during a pandemic.
Data was collected from HCW in Malaysian healthcare setting who completed the survey. Of 776 samples, 51.5% were doctors, 25.1% were nurses, 1.4% were pharmacists, while the rest were mainly supporting staffs and administrators. About three quarters of the respondents were female (77.4%) while a majority were aged between 31–40 years old (54.4%), Malay (81.1%), married (70.0%), had a degree as their minimum education level (62.6%), and were permanent staffs (96.5%). The majority of HCW reported receiving information on how to protect from COVID-19 (98.4%), how COVID-19 was transmitted (97.7%), information on COVID-19 symptoms (98.6%), what to do if you have symptoms (95.6%) and risk and complication of COVID-19 (93.9%). A high number of respondents received COVID-19 information from the Facebook application (95.1%), other healthcare workers (90.3%), and other social media applications (86.3%). The majority of respondents did not trust Facebook (72.2%) and other social media applications (79.3%) as trusted channels to receive information, and the most trusted source of information was other healthcare workers (83.2%).
Data from thirteen respondents were removed due to more than 30% missing data and sensitivity analysis revealed non-significant difference (p = 0.671). Sixty-one respondents reported known to have anxiety disorder, and hence were excluded from the analysis. Total data taken for analysis was 709.
illustrates the association between source of information and self-risk perception. From the table, 61.7% respondents received information from the radio have significantly high self-perceived risk. A higher proportion of respondents (58.7%) who had high self-perceived risk did not obtain information from NGO and this association is statistically significant.
demonstrates the association between source of information and knowledge, attitude, and practice scores. In terms of knowledge component, those who received information from Facebook, social media, family and friends had significantly higher knowledge score than those who did not. There was no significant difference in knowledge scores between those who sought information from radio, television, healthcare worker, NGO, community, and religious leaders. As for attitude component, those who received information from television had significantly higher attitude score among those who did not. There was no significant difference in attitude scores between those who sought information from other sources of information. As shown in Table 2
, those who received information from radio, television, family members, friends, NGO, community leaders and religious leaders had significantly higher practice score compared to those who did not receive from these sources respectively. Those who received information from another healthcare worker had significantly lower practice scores compared to those who did not.
presents the mean score of HAI by sociodemographic variables and self-perceived COVID-19 risk. Overall, there is a significant difference between the groups perceived at risk with no perceived risk group (p
< 0.001, 95% CI means difference: 1.55, 10.23). A difference is also seen between those with high self-perceived COVID-19 risk who had significantly higher mean HAI scores and those with moderate self-perceived COVID-19 risk (p
= 0.002, 95% CI mean difference: 1.10, 7.58).
As shown in Table 4
, simple linear regression revealed the significant predictors of HAI were self-perceived COVID-19 risk and attitude scores based on KAP. In multiple linear regression, high self-perceived risk remained significant. Those who perceived themselves to have high self-perceived COVID-19 risk was associated with an increase in 1.3 HAI score compared to those with low self-perceived COVID-19 risk. Higher scores in attitude component in KAP reflects a greater cautious attitude towards COVID-19. Each additional attitude score is significantly associated with an increase of 0.7 HAI score. The predictors measured in this study were able to explain 6% of the HAI score (adjusted R2
Past epidemics such as SARS in 2003 [22
], H1N1 in 2009/2010 [23
], and Ebola in 2014/2016 [24
] have revealed that health-related anxiety and safety behavior are pervasive. To the best of the authors’ knowledge, there is no local research done specifically on health anxiety during the pandemic, although local studies reported anxiety among healthcare workers during the COVID-19 outbreak [25
]. Healthcare workers’ self-perception of risk and their attitude towards the infectious disease contribute to health anxiety, as demonstrated by this study. We found that self-perceived high risk of getting infected with COVID-19 and a cautious attitude towards the infection significantly correlated with health anxiety traits.
Overestimation of the threat posed by a viral pandemic has been linked with increased anxiety [29
]. This is also established in this study as the healthcare workers’ self-perceived risk independently and positively correlated with health anxiety. Self-perception of having high risk might be counterproductive and made worse by increasing work demand, moreover with a sudden surge of cases during a pandemic. Particularly for HCW, they presumably have a greater risk of direct or indirect exposure with the virus than the general public. This impression could lead to perceiving the situation as more real and might amplify their risk perceptions, thus increasing their health anxiety traits. This is supported by [31
] as they studied across ten countries and found that people who have had personal and direct experience with the infection would have significantly higher risk perception. It is also important to consider the background of the study period, which was during the early phase of MCO. During this period, the cases were peaking and therefore this would also contribute to the participants’ health anxiety. This study was conducted from 18 March 2020, which coincided with the beginning of the virus outbreak in Malaysia where an accumulation of 790 cases with 2 death were reported. COVID-19 cases started to double from 238 cases on 14 March 2020 to 428 cases on 15 March 2020 and continued to increase exponentially since then. The highest daily COVID-19 positive cases were 190 cases on 15 March 2020 [32
]. Consideration of the timeframe of the study is essential to better understand the development of anxiety during the pandemic. It helps to identify its potential resilience and preventive factors in accordance to the development of the pandemic [33
]. A study during different stages of the SARS epidemic showed that risk-specific worries significantly associated with compliance to protective behaviours at different stages of the epidemic, but cognitive risk appraisal may inform individual protective behavior later in the epidemic trajectory [34
In this era, one important factor that highly influences people’s knowledge, attitude, and risk appraisal of pandemics is the internet. As reported in this study, social media was among the most sought-after sources of information, as 94.3% and 85.3% of the subjects obtained information from Facebook and other social media apps, respectively, besides relying on other HCW (89.4%). The social media has a role in precipitating and perpetuating anxiety as observed by way of excessive COVID-19-related Internet use as a form of safety-seeking behavior [35
]. This occurs by many ways, whether in the form of misinterpretation, distorted information, dissemination or even fabrication of information on the internet, which could further enhance the distressing safety-seeking behavior [36
]. Conversely, accurate perceptions of personal and societal risk factors importantly contribute to the success of measures and policies placed to flatten the epidemic curve during a pandemic [37
]. As outlined in the protection motivation theory [38
], threat appraisal and risk perception are determinants of the public’s willingness to cooperate and adopt health-protective behaviors during pandemics, including sanitization practices, physical distancing, and wearing face masks [40
Interestingly, although respondents reported receiving the majority of information from social media, the majority also reported not trusting social media as a source of COVID-19 information. A possible explanation is that the majority of respondents are highly educated with a health educational background; they are most likely aware of the trustworthiness of social media as a source of medical information for COVID-19. However, because social media is currently so integrated in our lifestyle, it allows faster and easier spread of unconfirmed or false information. Therefore, it becomes challenging to sift through the abundance of information. In addition, due to the unprecedented nature of the COVID-19 pandemic, a lot of information is still unknown. Hence, it is difficult to judge which information is true or false. The social media ‘infodemic’ has been linked to trigger panic and anxiety [43
]. Nonetheless, it was information sourced from the radio that was found to be significantly related to high self-perceived risk. A recent European survey reported that broadcast media, which includes radio and television, continues to be the most credible media source in contrast to social media [45
]. The same survey has also listed the latter as the least trustworthy, as concurred by this study finding.
It is also noteworthy to observe the different source of information influenced the knowledge, attitude and practice scores of the HCW. Multiple sources of information seemed to influence mostly the practice, followed by knowledge, while attitude was only significantly singly linked to television source (Table 2
While there is individual variation in people’s stress responses to outbreaks, collectively [46
], people’s behavior can influence and modify its contagion [47
]. Therefore, the adoption of certain attitude, specifically one that is parallel to the government policy may influence health anxiety as a manifestation of stress response. In this study, the attitude scale examined the respondents’ viewpoint or outlook on the pandemic. It constitutes of their own personal opinion on the dangerousness of COVID-19, its social implications including stigma against certain groups including HCW as a high-risk group, protective measures employed by themselves, their workplace, and the government in general, perceived risk to themselves and family members, and whether they think they were in line or against the government measures (Refer Supplementary Table
This study demonstrated that those with higher cautious attitude score had higher traits of health anxiety and attitude was not greatly influenced by the source of information except for broadcast media ie. television. Again, adoption of the “right” attitude is influenced by multiple factors, including the availability, dissemination, accuracy, and interpretation of information about the pandemic, risk perception, and individual and societal characteristics. A few of the questions in the Attitude scale such as ‘Do you think COVID-19 is dangerous?’, ‘Do you think handling infected COVID-19 cases will threaten the safety of healthcare workers?’ are ‘anxiety related’, but crucial in assessing one’s attitude on the pandemic. Harper et al. reported that fear at an early stage of pandemic has a functional role, which increases compliance to public health recommendations in order to prevent further transmissions of the disease. Health anxiety has been conceptualized as existing on a spectrum [3
]. Hence, the extent of anxiety may determine the performance whether one can perform optimally, underperform, or experience burn-out [50
The strengths of this study include that it examined health anxiety traits in a sample population where those with pre-existing anxiety disorders were excluded. Therefore, the findings were more specific on the threat of covid-19 to healthcare workers’ psychological health. It also sampled on a heterogeneous study population whereby the findings could be implied for a wider generalizability. There are several limitations of the study. Firstly, it is not representative of all Malaysian HCW. There was overrepresentation of Malay ethnicity; not the typical expected proportion of ethnicity group as among the Malaysian citizens, Malays constituted 63%; followed by other ethnic groups such as Chinese (24%) and Indians (7%) [51
]. Female respondents were also overrepresented in this study. That is, despite the fact the Malaysian population was constituted 16.8 million as compared to 15.8 million females [51
]. Secondly, the content validity, internal consistency, and/or test-retest reliability of the “knowledge, attitude and practices (KAP) on COVID-19” measure were not assessed, thus could be a potential source of misclassification bias. Thirdly, many confounding factors including presence of other health problems and their own traumatic experience dealing with pandemic, were not incorporated in this study. The researchers also need to ensure that the survey length was appropriate for the participants to complete to ensure genuine responses and prevent fatigue effects of answering a lengthy survey. Fourthly, we could not determine the response rate of the survey as it was sent through the representatives of the professional bodies who then distributed the form via online whereby the numbers of subjects who received the invitation to participate were not reported. Finally, the online platform of the survey also meant that it is biased towards those who were more internet-proficient.