Introduction
Within Australia, the coronavirus disease (COVID-19) pandemic has caused unprecedented disruption to the way most people live, with widespread physical distancing, travel restrictions and border closures, work-from-home mandates, and lockdowns enforced. Previous respiratory endemics, such as severe acute respiratory syndrome coronavirus (SARS-CoV), influenza A/H1N1, and the Middle East Respiratory Syndrome (MERS) posed significant psychological consequences onto many population groups, including symptoms of anxiety, depression, and posttraumatic stress disorder. Reference Luo, Chua and Xiong1 Given the level of upheaval associated with COVID-19, it is likely there will be both an acute and long-term impact on the mental health and well-being of Australian communities, although the scale and extent of this is only now starting to emerge. Reference Newby, O’Moore and Tang2 The impacts of COVID-19 might also be compounded by other recent disasters. For example, prior to the COVID-19 pandemic, devastating bushfires ravaged parts of the country, and entire communities were unable to recover before the nation’s focus shifted to the threat of COVID-19. The psychological health impacts of these bushfires are still emerging. Reference Jones, Mills and Gray3 Rural Australia was predominantly impacted by these events, Reference Usher, Ranmuthugala and Maple4 and, in fact, it is disproportionally impacted by extreme climate-related events in general, with bushfires, droughts, and floods common. Reference Morrissey and Reser5 This situation is unlikely to improve as climate-related disasters become more frequent and severe. Reference Morrissey and Reser5 Unfortunately, the Australian COVID-19 literature has largely focused on COVID-19 in isolation from other disasters.
International studies focused on measuring psychological distress associated with the COVID-19 pandemic using the Depression Anxiety Stress Scales-21 (DASS-21) Reference Lovibond and Lovibond6 have consistently drawn attention to the high levels of depression, anxiety, and stress-related symptoms being experienced by the general population during COVID-19. Studies from China, Mexico, and Italy have shown that between 15.7–32.4%, 18.7–28.8%, and 8.1–27.2% of people have experienced moderate to severe depression, anxiety, and stress symptoms, respectively. Reference Wang, Pan and Wan7–Reference Mazza, Ricci and Biondi10 Within Australia, 2 studies have used the DASS-21 to assess psychological distress within the general Australian population during the COVID-19 pandemic. Reference Newby, O’Moore and Tang2,Reference Rossell, Neill and Phillipou11 One study, conducted from March 27 to April 7, 2020, found that 46.3%, 40.8%, and 38.7% of respondents were experiencing moderate to extremely severe symptoms of depression, anxiety, and stress, respectively. However, it should be noted that a significant proportion of the sample had a lived experience of a mental health diagnosis (70%), which likely contributed to the high rates of psychological distress. Reference Newby, O’Moore and Tang2 Rates of psychological distress more consistent with international research were found in a separate Australian study, conducted from April 1 to 4, 2020. In this study, 21–35% of the population demonstrated moderate to extremely severe depression, anxiety, and stress, with the reported rates being greater than the population norms. Reference Rossell, Neill and Phillipou11 These studies highlight the significant psychological impacts of the COVID-19 pandemic on the general population; however, it is likely the risk of experiencing psychological distress varies, depending on a person’s characteristics and vulnerabilities. Reference Norris, Friedman and Watson12
The 2 aforementioned Australian studies also give some insights into the factors that place a person at higher risk of experiencing psychological distress. While there was variability between the studies in the specific types of factors that were found to be important, in general, the factors could be condensed into 3 overarching categories: demographic variables (eg, being female, a student, a career or stay-at-home parent), personal and financial vulnerabilities (eg, having a chronic illness or a prior history of mental health diagnosis, experiencing financial stress, loneliness), and COVID-19-specific worries and experiences (eg, having a higher mortality risk or perceiving higher illness severity). Furthermore, a number of variables stand out as being of particular relevance to a risk of increased psychological distress: having a chronic illness or a prior history of mental health diagnosis, and experiencing financial stress. Reference Newby, O’Moore and Tang2,Reference Rossell, Neill and Phillipou11 Similar categories of risk factors have also been shown in international samples. Reference Wang, Pan and Wan8–Reference Mazza, Ricci and Biondi10 Of particular relevance to the Australian context is 1 study that found that a history of stressful situations (eg, dismissal, mourning) prior to COVID-19 was associated with higher levels of depression and anxiety. Reference Mazza, Ricci and Biondi10 This finding emphasizes the need to consider the mental health impacts of COVID-19 in combination with other life stressors, such as other disasters.
Following a disaster, such as bushfires or floods, the majority of people will recover psychologically and most will not go on to develop a mental health disorder. However, many people will experience strong emotional and physical reactions. Reference Norris, Friedman and Watson12–Reference Goldmann and Galea15 At times, the adverse psychological effects can be experienced for prolonged periods after the disaster occurs, particularly when there is severe trauma exposure or additional adverse life events in the aftermath of the event. Reference Bryant, Gibbs and Gallagher16 While there is a large amount of literature on disasters and mental health, Reference Norris, Friedman and Watson12,Reference Norris, Friedman and Watson13 relatively few studies have examined psychological outcomes after multiple disasters. Within Australia, exposure to multiple (ie, 2 or more) natural disasters (eg, flood, cyclone, or earthquake) has been associated with an increased lifetime risk of panic disorder, while cumulative effects of repeated man-made disaster exposures (eg, a fire started by a cigarette or a bomb explosion) have been noted in terms of an increased lifetime risk of obsessive compulsive disorder. Reference Reifels, Mills, Dückers and O’Donnell17 Research has also identified cumulative effects of repeated disaster exposure (both man-made and natural) in relation to the risk of suicide at the Australian population level. Reference Reifels, Spittal and Dückers18 These studies are limited in that the focus on diagnosable psychiatric disorders and suicidality might not be indicative of subclinical levels of distress. Reference Norris, Tracy and Galea14,Reference Goldmann and Galea15 Taken together, these 2 studies justify a deeper analysis of the cumulative psychological impact of prior disaster experience and COVID-19. Disaster experience should not only include exposure, but also impact; given the level of psychological distress a person experiences following a disaster is impacted by how people are affected by the disaster. Reference Bryant, Waters and Gibbs19,Reference Wahlström, Michélsen, Schulman and Backheden20
This paper forms part of a larger program of research looking at the impact of COVID-19 on organizations and workforce and how organizations can be used to deliver appropriate interventions to improve employee preparedness, response, and recovery. Investigating this issue is important as workplaces are an essential part of most people’s social support systems and might offer a level of long-term protection from the psychological impacts of pandemics and disasters, Reference Goldmann and Galea15,Reference Schouten, Callahan and Bryant21,Reference Modini, Joyce and Mykletun22 especially if the capacity of organizations to provide appropriate support can be developed. Reference Hatton, Grimshaw, Vargo and Seville23,Reference Brooks, Dunn and Amlôt24 Further, novel approaches to community intervention need to be explored, Reference Norris, Friedman and Watson12 particularly in rural and regional Australia where appropriate treatment services are often limited or unavailable. 25 While these are larger issues that will be considered in other manuscripts, this paper specifically focuses on the mental health of Australian workers during the COVID-19 outbreak. It was hypothesized that Australian workers would report levels of anxiety, depression, and stress above the population norms. Reference Schouten, Callahan and Bryant21 It was also hypothesized that previous recent disaster exposure and impact from that disaster will be a risk factor for higher levels of depression, anxiety, and stress.
COVID-19 Public Health Response: Contextual Information
Prior to the survey’s distribution, people living in Australia had experienced a series of public health responses that varied depending on geographical location. Restrictions had included border closures, directives to stay at home unless for necessary shopping, health care, exercise and work and study that can’t be done remotely, and the closing of non-essential services and many schools. At the time of the survey’s distribution, these restrictions were gradually being eased. However, by June 20, restrictions were being reinstated in the state of Victoria due to an apparent “second wave” of infection. By July 7, 2020, a second period of lockdown had been introduced for metropolitan Melbourne and neighboring Mitchell Shire. Reference Duckett and Stobart26,Reference Lupton27 As of July 8, 8886 confirmed cases (56% acquired from overseas) and 106 deaths (predominantly in people ages 60 and greater) had occurred in Australia. 28
Method
Recruitment
A link to the online survey was distributed via e-mail and social media posts, initially using the authors’ personal and professional networks with subsequent snowball recruitment. The survey was available in English only. Inclusion criteria comprised residing in Australia, being age 18 years and over, and having volunteered or been employed at any time during the period from December 2019 to July 2020. Data were collected from May 29 to July 8, 2020. A follow-up survey will be conducted when restrictions have been lifted. Study data were collected and managed using REDCap electronic data capture tools hosted at The University of Melbourne.
Ethics Approval and Consent
The study was conducted in compliance with the Declaration of Helsinki, the National Statement on Ethical Conduct in Human Research (2007, updated 2018) and the University of Melbourne’s COVID-19 Central Human Ethics Committee (Approval number: 2056921.1). All participants provided electronic informed consent before participating. Participants could choose to enter a prize draw for 1 US $100 gift card and/or register their interest in a follow-up survey, by providing an email address. If participants declined to enter the prize draw or register their interest in a follow-up survey, their responses remained anonymous.
Participants
In total, 616 records were created in REDCap. Responses received after July 8, 2020, were removed from the data set. Of the remaining records, 596 participants were included in the analysis.
Measures
Demographics
Information was collected on each participant’s age, gender, postcode of residence, living arrangements (by myself, with my partner, with pre-school-aged children, with school-aged children, with friends or flatmates, with other family, other), diagnosis of COVID-19 (yes, tested and confirmed, suspected but not tested, no, don’t know), COVID-19 diagnosis in another person in the house hold (yes, tested and confirmed, suspected but not tested, no, don’t know), industry of work, 29 nature of work (paid, volunteer), employment type (full time, part time, casual, not applicable, other), work location (home, normal workplace, home and normal workplace combination, not currently working or volunteering, other), and changes to employment due to the pandemic (temporarily stood down, role terminated, hours reduced, hours not impacted, other).
Physical and Mental Health
Participants were asked whether they were living with a long-term physical health condition (yes, no, don’t know, prefer not to say), and/or a long-term mental health condition (yes, no, don’t know, prefer not to say).
The survey also included the validated DASS-21, Reference Lovibond and Lovibond6 which was used to assess symptoms of depression, anxiety, and stress. The DASS-21 is a 21-item measure consisting of 3 subscales (depression, anxiety, and stress), with 7 items per subscale. Participants were asked to read each statement (eg, I found it hard to wind down) and indicate how much the statement applied to them over the past week, on a scale from 0 (did not apply to me at all) to 3 (applied to me very much). Relevant items were combined to give a total score for each of the 3 scales, and severity ratings were given (Normal, Mild, Moderate, Severe, and Extremely Severe) based on this total. The DASS-21 possesses good internal consistency across the subscales and overall scale (α > 0.81), and convergent and discriminant validity has been established. Reference Henry and Crawford30
Disaster Exposure
Participants were asked whether they had experienced another disaster such as drought, flood, or bushfire in the past 2 years (yes, no) and whether they had experienced emotional stress, physical stress, financial stress, damage or loss of property, or other impact because of the disaster(s) they had experienced (and if so, to specify).
Recoding of Survey Questions
Each participant’s postcode of residence was recoded into the Modified Monash Model categories of 1-7 and subsequently divided into metropolitan (MMM1) and rural, regional or remote (MMM2-7). 31 Living arrangements were divided into living alone (including by myself) and living with others (including, with my partner, with pre-school-aged children, with school-aged children, with friends or flatmates, with other family, other). Responses to the pre-existing physical health and mental health condition were limited to yes or no. A new disaster exposure and impact variable, containing 3 categories, was created from the disaster questions: no additional disaster exposure (including participants who answered no to having experienced another disaster in the past 2 years), additional disaster exposure with no impact (including participants who answered yes to having experienced another disaster in the past 2 years, but did not report an impact from the disaster), and additional disaster exposure with impact (including participants who answered yes to having experienced another disaster in the past 2 years and did report an impact from the disaster).
Statistical Analysis
Categorical variables are presented as N and percent (%), while continuous variables are presented as mean (M) and standard deviation (SD). DASS-21 subscale scores were compared to the existing Australian population norms Reference Lovibond and Lovibond6 using an online t-test calculator. Between group comparisons for continuous variables were completed using analysis of variance (ANOVA). Post hoc tests were undertaken for statistically significant results using Tukey’s Honestly Significant Difference test. Separate multiple regression analyses were conducted to explore factors predictive of depression, anxiety, and stress. The independent variables of age, sex, pre-existing chronic health condition, pre-existing mental health condition, and additional disaster exposure with impact were included in the analysis. For the regressions, the variable “sex” was limited to male and female, whereas the disaster exposure with impact variable was recoded into yes or no from the free text responses.
Results
Demographics
Demographic characteristics of the participants are depicted in Table 1. Overall, participants were mostly female (80.3%) and ranged in age from 18 to 77 years (M = 43.8, SD = 12.9). The majority were from non-metropolitan areas (MMM 2-7; 78.1%), lived with other people (87.4%), had not been diagnosed with COVID-19 (98.9%), had not lived with someone diagnosed with COVID-19 (99.1%), worked in the health care and social assistance industry (45.0%), were in paid work (98.0%), worked full time (56.5%), and were either working from home (38.9%) or their normal workplace (35.5%). Over half the participants reported COVID-19 had had no impact on their work hours (67.0%). Approximately one-fifth identified as having a pre-existing chronic disease (21.3%) or mental health condition (21.2%). Over half (76.1%) had had no additional disaster exposure.
Demographic characteristics of respondents
Notes:
a Mining, manufacturing, electricity, gas, water and waste services, construction, wholesale trade, retail trade, accommodation and food services, transport, postal and warehousing, information media and telecommunications, financial and insurance services, and public administration and safety.
b Suspended activities, different workplace, changes of work location as restrictions eased.
c Increased hours, increased workload, redeployment, self-employment, leave and/or reduced pay.
Mental Health
Table 2 shows the proportion of participants who scored across the severity categories of the DASS-21 subscales, as well as the mean and standard deviation for the DASS-21 total subscale scores. Overall, 19.2% of participants were experiencing moderate to extremely severe depression, 13.4% were experiencing moderate to extremely severe anxiety, and 16.8% were experiencing moderate to extremely severe stress. When compared to the existing Australian population norms (pre-pandemic), Reference Lovibond and Lovibond6 DASS-21 scores for depression and stress were significantly higher for participants in this study (6.3 ± 7.0 vs 8.2 ± 9.0, P < 0.0001; 10.1 ± 7.9 vs 11.8 ± 9.0, P < 0.0001, respectively).
Depression, anxiety, and stress (DASS-21) severity ratings
As shown in Table 3, 1-way between-group ANOVAs revealed a significant relationship between disaster exposure type (no additional disaster exposure, additional disaster exposure without impact, additional disaster exposure with impact) and the DASS-21 stress subscale (F (2, 487) = 3.994, P = 0.019, η Reference Newby, O’Moore and Tang2 = 0.016). DASS-21 stress subscale scores were significantly higher for participants who had experienced an additional disaster exposure with impact than for those who had experienced an additional disaster without impact (6.8 ± 4.3 vs 4.4 ± 4.0).
Mean (standard deviation) depression, anxiety, and stress scores among respondents who had experienced no additional disaster exposure, additional disaster exposure without impact, and additional disaster exposure with impact
Predictors of Depression, Anxiety, and Stress Severity
Three separate multiple regressions were used to explore factors predictive of DASS-21 depression, anxiety, and stress symptom severity. Sex, age, pre-existing chronic health condition, pre-existing mental health condition, and additional disaster exposure with impact were included as independent variables in each of the analyses. The final multiple regression models are presented in Table 4.
Predictors of depression, anxiety, and stress severity (DASS-21 subscale scores)
Among participants who had experienced a disaster in the 2 years prior to the survey, higher depression scores were associated with being male (B = 2.127, P = 0.041) and having a pre-existing mental health condition (B = 5.060, P < 0.001). Higher anxiety scores were associated with having a pre-existing mental health condition (B = 0.3.013, P < 0.001). Higher stress scores were associated with pre-existing mental health condition (B = 0.3.289, P = 0.001) and having reported an impact from the disaster (B = 2.367, P = 0.003). The predictors for depression, anxiety, and stress explained 25.0% (F (5, 105) = 7.008, P < 0.001), 22.7% (F (5, 105) = 6.162, P < 0.001), and 21.4% (F (5, 109) = 5.939, P < .001) of variance, respectively.
Discussion
This study provides some insight into how the COVID-19 pandemic has impacted the mental health of people living and working in Australia, particularly those who were exposed to a disaster prior to COVID-19. Overall, 19.2%, 13.4%, and 16.8% of participants were experiencing moderate to extremely severe depression, anxiety, and stress symptoms, respectively, as measured by the DASS-21. When compared to the existing Australian population norms Reference Lovibond and Lovibond6 (pre-pandemic), the mean DASS-21 scores for depression and stress were significantly higher for participants in this study. Despite this, the levels found in this study were lower than those found in previous Australian research. Reference Newby, O’Moore and Tang2,Reference Rossell, Neill and Phillipou32 For instance, Newby et al. Reference Newby, O’Moore and Tang2 found that 46.3%, 40.8%, and 38.7% of participants reported moderate to extremely severe symptoms of depression, anxiety, and stress, respectively, whereas Rossell et al. Reference Rossell, Neill and Phillipou32 found that 21-35% of their sample demonstrated moderate to extremely severe depression, anxiety, and stress.
Estimates of the rates of psychological distress are likely to vary based on sample characteristics (eg, reported rates of pre-existing mental health conditions) and contextual factors (eg, survey timing). The proportion of participants reporting a mental health condition in the current study was 21.2%, which is representative of the Australian prevalence rates. 33 This was not the case for Newby et al.’s study, Reference Newby, O’Moore and Tang2 where the proportion was closer to 70.0%. Further, the majority of participants in this study were from non-metropolitan areas (78.1%), had not been diagnosed with COVID-19 (98.9%), were in paid work (98.0%), worked full time (56.5%), and reported COVID-19 had not impacted work hours (67.0%). The survey used in this study was also distributed at a time when COVID-19 cases in Australia were relatively under control and restrictions were being eased. Reference Duckett and Stobart26,Reference Lupton27 Given these characteristics, it is possible this sample was somewhat protected from increased psychological distress during COVID-19; they were likely experiencing fewer COVID-19 cases and less financial stress than other samples. Additionally, these results should be considered in light of the potential mental health benefits of working. Reference Modini, Joyce and Mykletun22 Although unmeasured, other protective factors might be contributing to these findings, including workplace (eg, good-quality supervision Reference Modini, Joyce and Mykletun22 ), family, or government support. Reference Lupton27,28 Overall, this highlights the need to carefully interpret research in this area.
The current study also provides insights into the variables associated with higher psychological distress during the COVID-19 pandemic. In particular, and in keeping with previous Australian research, Reference Newby, O’Moore and Tang2,Reference Rossell, Neill and Phillipou32 this study found having a pre-existing mental health condition predicted higher severity of depression, anxiety, and stress-related symptoms. This adds to the growing body of evidence highlighting the increased emotional impact of the COVID-19 pandemic on people with a lived experience of mental illness. However, it is also possible that people with a pre-existing mental health condition were experiencing higher levels of psychological distress prior to the COVID-19 pandemic, rather than as a result of the pandemic. It is difficult to differentiate this as no baseline assessment for psychological distress prior to the pandemic was conducted. This is a common limitation in all research related to disasters and mental health, and it will likely be important for future research in this area to adopt a prospective study design to examine prevalence and severity estimates prior to disasters occurring. This study also found being male predicted higher levels of depressive symptoms, but not anxiety or stress. The findings from past Australian research have varied, with 1 study finding being female was predictive of lower depression Reference Newby, O’Moore and Tang2 and another study finding lower levels of negative emotions were demonstrated by males. Reference Rossell, Neill and Phillipou32 However, international research has consistently found that being female is associated with higher levels of stress, anxiety, and depression. Reference Wang, Pan and Wan7,Reference Cortés-Álvarez, Piñeiro-Lamas and Vuelvas-Olmos9,Reference Mazza, Ricci and Biondi10 Contextual issues between countries and surveys may explain some of this variation.
This study also found that having experienced a disaster in addition to COVID-19, and having experienced an impact from that disaster, was predictive of higher stress symptoms, but not depression or anxiety. This adds to previous research into the mental health impact of COVID-19, which has shown a history of more general stressful events (eg, dismissal, mourning) is predictive of higher levels of depression and anxiety. Reference Mazza, Ricci and Biondi10 Furthermore, it extends this research, by highlighting the need to consider a person’s previous disaster experience when researching mental health outcomes during COVID-19, due to the cumulative effects of repeated disaster on mental health and psychological distress. These cumulative effects have been shown in research unrelated to COVID-19. Reference Reifels, Mills, Dückers and O’Donnell17,Reference Reifels, Spittal and Dückers18,Reference Williams, Williams and Stein34 The current study also emphasizes a need to consider how persons are impacted by a disaster, and not just whether they have been exposed to a disaster. Again, this has been suggested by extant literature unrelated to COVID-19. Reference Bryant, Waters and Gibbs19,Reference Wahlström, Michélsen, Schulman and Backheden20 That having experienced a disaster in addition to COVID-19, and having experienced an impact from that disaster, was predictive of higher stress symptoms, but not depression or anxiety, could be related to how the DASS-21 measures these symptoms. The DASS-21 conceives stress as a persistent state of over-arousal reflecting a continuing difficulty in meeting taxing life demands, depression as being related to low self-esteem and anxiety as being related to an increased anticipation of negative events. Reference Lovibond and Lovibond6 It is possible this is reflective of how people emotionally respond to multiple disasters. Alternatively, it is also possible it is too early to see increases in depression and anxiety in our participants; additional life stressors following a disaster are predictive of increased symptomatology over time Reference Goldmann and Galea15,Reference Bryant, Gibbs and Gallagher16 ; however, the long-term emotional burden of COVID-19 is still emerging. Further research with larger sample sizes would be needed to properly elucidate this.
Implications
While it is difficult to ascertain the level of distress participants in this study were experiencing prior to the pandemic, and whether these levels changed as the pandemic began and progressed, there is a need, regardless, to provide an appropriate response to lessen elevated symptoms of psychological distress. The needs of different subgroups of the population will vary widely, and responses will need to be tailored to an individual’s unique combination of risk and protective factors. Responses will likely need to vary in intensity and encompass multiple levels of intervention (eg, societal, community, family, and individual) so as to prevent subclinical symptomatology from worsening over time or developing into psychiatric disorders. Reference Norris, Friedman and Watson12,Reference Watson, Brymer and Bonanno35 Thought will need to be given to the services, organizations, or agencies best placed to deliver such responses, particularly in places where traditional mental health services are limited. In most cases, consideration will also need to be given to including local stakeholders to ensure programs and services are culturally and regionally appropriate. Reference Watson, Brymer and Bonanno35 Research regarding how to provide mental health responses following disasters is still emerging, but it is hoped the larger program of research surrounding this study will address this issue in more detail, with a focus on how organizations can be used to deliver appropriate interventions to improve employee preparedness, response, and recovery. Reference Newby, O’Moore and Tang2,Reference Rossell, Neill and Phillipou11–Reference Norris, Friedman and Watson13
Limitations and Future Research
The results of this study are based on a convenience sample of workers recruited online. Thus, the results are unlikely to generalize to the broader Australian population. Our sample was also small in comparison to other studies in this area, with the majority of participants being female (80.3%), living in a non-metropolitan area (78.1%), and having no additional disaster exposure (76.1%), which might have impacted results. Future research should focus on the mental health impact of having experienced multiple disasters and the implications of this on organizations that may have a role to play in the response, such as workplaces. It would also be helpful to assess in more detail people’s perceived experience of single and multiple disasters, and how this relates to their mental health and well-being.
Conclusions
There is a growing body of evidence that highlights the increased psychological impact of the COVID-19 pandemic on people with a lived experience of mental illness, as well as those who are experiencing, or have experienced, additional stressors (such as impact from an additional disaster). Reference Newby, O’Moore and Tang2,Reference Mazza, Ricci and Biondi10,Reference Rossell, Neill and Phillipou11 When considered alongside other COVID-19 and mental health research, our study highlights the need to provide an appropriate response to lessen a person’s severity of psychological distress; different subgroups of the population will have varied combinations of risk and protective factors, and responses will need to take this into consideration. Importantly, COVID-19 is not occurring in a vacuum, and as climate-related disasters become more intense, frequent, and longer in duration, increasing attention should be given to the cumulative mental health impact of multiple disasters.
Acknowledgments
We thank all of the participants who kindly contributed to this study.
Author Contributions
TP and KMG conceptualized the study, performed the formal analysis, investigated, and contributed to the writing of the original draft and review and editing of the manuscript.
Conflict(s) of Interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this paper.
Open access
