key: cord-0874749-jnt2i30i
authors: Nikolova, Irina; Caniëls, Marjolein C. J.; Curseu, Petru L.
title: COVID‐19 Rumination Scale (C‐19RS): Initial psychometric evidence in a sample of Dutch employees
date: 2021-04-04
journal: Int J Health Plann Manage
DOI: 10.1002/hpm.3165
sha: cd48d7b44096a3c502bd18cfbef29e3bf9dd9f99
doc_id: 874749
cord_uid: jnt2i30i

OBJECTIVES: Starting with the spring of 2020, COVID‐19 pandemic has impacted nearly every aspect of our lives. Due to its threatening nature, along with the rapid rise in contamination and mortality figures, the spread of the virus has caused a considerable rise in individuals' anxieties. To enable the assessment of the COVID‐19‐triggered individual rumination, we developed and tested a COVID‐19 Rumination Scale (C‐19RS). DESIGN AND METHODS: Demographics (i.e., gender, age and education) and several items assessing the proximity of one's exposure to the virus (i.e., whether one's family and close friends are affected) were evaluated as antecedents of C‐19RS that provided evidence for the criterion validity of the scale. A sample of 523 Dutch employees working in different companies and sectors completed the online survey in March 2020. RESULTS: Results showed that women, older individuals and workers with lower educational level ruminated considerably more about COVID‐19. In keeping with prior theoretical and empirical work on stress and coping, we established that COVID‐19 ruminative thoughts can unlock withdrawal coping reactions (i.e., self‐handicapping) and drain individual's energy (i.e., causing emotional exhaustion), whereby providing evidence for the predictive validity of the new instrument. In addition, we examined how the COVID‐19 rumination evolved during the nearly 3‐week period of the data collection, a time‐frame that coincided with the introduction of the national restrictive measures in the Netherlands. Results showed a drop in the level of rumination, which might be indicative of potential habituation with the stressor. CONCLUSIONS: The results supported the sound psychometric qualities of the scale.

(TMSC), 9 we explain why a new and threatening situation, such as the COVID-19 pandemic, may cause a rise in individuals worries about dealing with the threat at hand.

The items included in our new scale tap into the affective and detachment aspects of ruminative thoughts that occur when a stressful event (e.g., a virus pandemic) is perceived as overwhelming and overpowering. Existing scales about fear or anxiety of a health-threatening disease assess concerns about developing a disease or carrying on a hereditary disease to one's offspring. 10 Our scale differs from these types of scales, in that the impact of a pandemic is much broader than affecting only individual's own health, or the health of their offspring. Furthermore, because the individual's perceived control over a pandemic is likely to be nihil, a pandemic may trigger ruminative thoughts. Given that pandemics are generally rare, rumination triggered by the life-threatening, wide spreading COVID-19 is likely to have widespread repercussions. To enable research of the far-reaching consequences of the coronavirus, a new diagnostic tool is needed.

Providing scholars with an instrument that can help tap into individuals' worries about the corona spread has theoretical and practical implications. Gaining insights into the segments of the population and the extent to which corona-related worries preoccupy our daily life, is vital for further theory development regarding the individual's response to, and coping with, high-impact worldwide health threatening situations, such as presented by the corona pandemic. Also, by aiding future scholarly work in this area, we strive (indirectly/by extension) to help health professionals, policymakers and organisations to accurately predict individuals' reactions to highly impactful, widespread public-health stressors. Studying the effect of the different mental well-being-supportive initiatives that were undertaken over the past few weeks (e.g., online buddy support, virtual psychological support) is key. Extending knowledge on the effectiveness of this type of measures may be imperative for policymakers and organisations who rely on empirical evidence when making decisions about financing and promoting such programs and initiatives.

Studies from the occupational health and work psychology domain have consistently showed that situations marked by uncertainty (e.g., organisational changes, or life events) can unlock anxieties and negative emotions among the affected individuals and can cause ill-health. [11] [12] [13] [14] To date, much of the existing work on individuals' stress and coping with uncertainty builds on the theoretical premises of the TMSC. 9 TMSC posits that individuals continuously monitor their environment for changes, and that they evaluate the potential of every new situation to cause loss of precious resources or to challenge their coping capacities (labelled 'primary appraisal'). In response to the situational stressor and based on the individual's primary appraisal, in a next step, one evaluates the needed resources for coping next to the resources one already possesses for dealing successfully with the situation at hand (labelled 'secondary appraisal'). When a stressor is appraised as harmful and exceeding the individual's resources to deal with it (i.e., a threat appraisal), it will trigger a stress response and strain symptoms. Sustained over time, high levels of strain can in turn impact the individual's appraisal, causing more negative evaluations of the source of strain (i.e., the strain-inducing situation) to arise. 15, 16 When confronted with an uncertain situation that exceeds the individual's control, one is likely to experience increased anxieties and negative emotions. 9 In keeping with TMSC, 9 we posit that events such as the wide spread of the COVID-19 virus, which are highly impactful and jeopardising public health, will be appraised as a threat. Job insecurity research has provided considerable evidence that macrolevel factors (e.g., national unemployment rate and the country's economic climate) [17] [18] [19] are viewed as a threat, because they are beyond the control of the individual. Stressors, in general, and the COVID-19 situation in particular, can be highly alarming because individuals perceive little or no control at all over it. 14 The combination of, on the one hand, the high importance and impact of the stressful event (i.e., one's own and his or her family members' lives are at stake), and on the other individuals' inability to control it, implies that psychological detachment from this stress stimulus (i.e., being able to 'switch off' worrying and disturbing thoughts about the virus) may be impaired. As a result, individuals are likely to remain preoccupied with persistent worries about theirs and their loved one's health, which in time can exhaust them.

In prior research, 2 rumination has often been featured as a coping mechanism that is characterised with negative moods and self-focused attention. 20 To date the ample research on rumination streaming from various disciplines (e.g., clinical and neuropsychology, psychiatry and work and organisational psychology) has provided abundant, yet somewhat contradicting evidence on the outcomes of rumination. Some of these inconsistencies can be attributed to the different definitions and operationalisations of rumination used by scholars. Generally, studies on rumination carried out by clinical psychology and psychiatry researchers, tend to work with clinically inspired (i.e., rumination as a symptom of psychiatric disorder) definitions and conceptualisations of rumination. 20, 21 This view spilt over into occupational health research on rumination, where scholars primarily discuss rumination as a 'nonadaptive' cognitive process and an inability to solve problems. 2, 22 Researchers sharing this negatively tinted view 23, 24 presumed that rumination is a form of maladaptive coping and is psychologically debilitating because the individual fails to escape from the vicious cycle of overpowering NIKOLOVA ET AL. thoughts. However, other scholars 25 suggested that there is a positive side to rumination as well. For instance, Segerstrom et al. 26 proposed that ruminative thoughts can be positive and that a distinction should be made based on the purpose or focus of the thought (i.e., problem-solving vs. searching for meaning). Similarly, Pravettoni et al. 22 argued that besides the negative (i.e., 'repetitive' ruminative thoughts), rumination encompasses also positive cognitive aspects (i.e., 'creative' rumination). Furthermore, Cropley and Zijlstra 27 differentiated between two variants of rumination: a negatively emotionally loaded type, called 'affective rumination', and a more positive one, called 'problem-solving pondering'. The main difference being the type of emotional arousal. Finally, Querstret and Cropley 25 distinguished between three kinds of rumination-affective rumination, problem-solving pondering and detachment-where the problem-solving pondering reflects the creative thought process that is positive by nature and triggers positive outcomes.

In keeping with views from clinical psychology research, we argue that ruminative thoughts occur when a stressful event (e.g., coronavirus pandemic) is perceived as unsolvable and overpowering. 22, [28] [29] [30] '. However, we did not include any problem-solving aspects, because the trigger of the anxiety (i.e., the virus) is beyond an individual's ability to provide an effective solution to (e.g., fully preventing contamination, illness or death caused by the virus). It is unlikely that individuals can engage in any form of active problem-solving, as it is inherent to the nature of rumination that individuals find themselves trapped into a repetitive thought process fueled by negative emotions (e.g., fears) and cognitions (e.g., self-doubt). [28] [29] [30] Our scale incorporates affective and cognitive items alike. For instance, an item that taps into the emotional aspect of COVID-19 rumination is 'I am worried about the coronavirus (COVID-19)'; an example of an item that captures ruminative cognitions is 'I notice that I think about the coronavirus (COVID-19) several times a day'. Importantly, rumination drains the individual's energy because the arousal remains high during a substantial period of time (i.e., the period when one is trapped in a vicious affect-driven cognitive cycle 27 ). This is the case because the repetitive thoughts are often driven by a discrepancy between the current state (e.g., living in fear of potential contamination) and the ideal state (e.g., the contamination threat does not exist) which the individual cannot overcome. 31 

Of the invited 1289, precisely 523 Dutch employees completed the online survey (response rate of 40.6%). The questionnaire was administered by a Dutch International Organisation for Standardisation certified research company on 23 March 2020. Only 7 days prior to the data collection, the restrictive measures for limiting the COVID-19 virus spread in the Netherlands were announced (i.e., all individuals, except those representing essential occupations such as medical and public care staff, were asked to remain home; schools were closed). Respondents could take part in the survey until 9 April 2020. During the entire period the data collection was carried on, the restrictive COVID-19 measures remained in place. From the onset of the outbreak in the Netherlands until the closing date of the survey, 20,549 individuals tested positively, 7735 needed hospitalisation and 2248 died from the coronavirus. The procedures used to collect data did not incur significant load on participants, and were not assumed to create distress or harm to participants as no new invasive procedures were used.

To ensure a sample that reflects the demographic distribution of the Dutch working population, the marketing company in charge of the data collection used a stratification procedure based on demographics of the general working population in the Netherlands as reflected in the annual reports of the Central Office for Statistics of the Netherlands (e.g., employees age, gender and education). Yet, some requirements posed by the researchers (i. e., include only participants who have colleagues and direct supervisor) might have caused for some deviation of what could constitute a nearly representative sample. These sampling restrictions were necessary because of other planned studies, for which data should include only the responses of employees who have colleagues and a direct supervisor. The youngest participant in the survey was 18 and the oldest was 67 years old (M = 45.29 years; SD = 11.31 years). The sample distribution with regard to employees' educational level was as follows: lower educational training (17.4%), high school (38.0%) and higher educational training (42.6%). Of the 523 surveyed employees, 43.6% were female. Precisely 88.5% were employed with a permanent contract.

The coronavirus pandemic and the rising need among researchers to measure individuals' persistent corona-related worries in a valid and reliable manner inspired us to develop the COVID-19 Rumination Scale (C-19RS). Because circulating lengthy surveys can fatigue participants, and can cost time and money to organisations, we chose to include a limited number of items. The six items incorporated in the scale were adapted from an earlier scale developed by one of the authors for measuring rumination. Three experts in the field of psychometry and wellbeing research were consulted on the content and wording of the items. In terms of their content, the items were formulated based on a thorough review of existing scales for measuring rumination. Before the data collection was set out, the authors of the study once again reflected on the theoretical relevance and wording of each item. All items tap into affective rumination or impaired detachment facets of rumination. We incorporated both cognitive and affective-worded items to emphasise the thought aspects and the anxiety aspects of rumination, respectively.

We used the following sentence to instruct the respondents when answering the COVID-19 rumination questions:

'To what extent do you agree with the following statements?'. Responses could be given on a 5-point Likert scale ranging from 1 = 'totally disagree' to 5 = 'totally agree'.

In addition to demographic characteristics, two constructs (i.e., emotional exhaustion and self-handicapping) were used to test the predictive validity of the C-19RS. Participants were able to select a response on a 5-point Likert scale, for emotional exhaustion ranging from 1 = 'never' to 5 = 'always', and for self-handicapping from 1 = 'totally disagree' to 5 = 'totally agree'. We evaluated emotional exhaustion using the 4-item Burnout Assessment Tool (BAT) developed by Schaufeli et al. 32 A sample item for BAT is: 'At work, I feel mentally exhausted' (α = 0.86). Selfhandicapping was assessed with a 10-item instrument developed by Strube. 33 A sample item is 'I tend to put things off until the last moment' (α = 0.73).

To investigate the factor structure of the scale we conducted a principal component analysis (PCA) without rotation. The results show that all items load on a single factor that has an eigenvalue of 3.62 and covers NIKOLOVA ET AL.

-5 more than 60% of variance in the scores. Table 1 presents the results of the PCA, item descriptives and the internal reliability scores.

Cronbach's α for the scale is 0.85 supporting a good internal consistency of the items. As this scale is newly developed, we have used the procedure and macro presented by Hayes and Coutts 34 to compute the omega for this scale based on the results of the factor analysis. The scale's omega of 0.87 confirms a good internal consistency of the items. For subsequent analyses concerning the testing of the scale validity we use the average item score as well as the Bartlett dominant factor that is an accurate indicator of the true underlying dominant factor score of the scale. 35 Both the average item score and the dominant factor scores are presented in Table 2 , showing the descriptive statistics and intercorrelations among the study variables.

We have used several demographic variables as antecedents of corona-related worries and we have also adopted and adapted a procedure similar to the criterion group validation to further explore the scale's psychometric properties. The criterion group validation procedure contrasts the scores obtained using C-19RS for groups that are known to potentially differ with regard to the criterion that is being evaluated. 36 In the context of the COVID-19 pandemic, we expect that respondents who have family members or friends infected with corona would be naturally more worried concerning COVID-19 as compared to respondents whose family members and immediate social network are not impacted. We have used the following items to capture the extent to which respondents had Table 3 presents the results of the regression analysis. Table 3 reports that women tend to worry more about COVID-19 than men (β = 0.14, p = 0.002), a pattern that is aligned with meta-analytic evidence 37, 38 showing that women are more prone to rumination 

Rumination as a cognitive tendency of excessively focussing on a stressor in an attempt to spur the cognitiveemotional processing, precedes more serious emotional disturbance and ultimately emotional disorders, such as anxiety and depression. Literature to date explored a variety of consequences of general rumination in relation to various stressors and the results converge towards the fact that rumination is an antecedent of self-handicapping strategies 39 and exhaustion. 40 ***p < 0.001, **p < 0.01, *p < 0.05, †p < 0.10.

To test the predictive validity of the C-19RS, we used self-handicapping and exhaustion as dependent variables. Given that our data is cross-sectional and we cannot draw causal claims based on this data structure, we selected gender as an instrumental variable to clarify the mediation path between COVID-19 rumination, on the one hand, and self-handicapping and exhaustion, on the other hand. We expect that rumination mediates the association between gender and self-handicapping on the one hand and between gender and exhaustion on the other hand. To test these mediation claims, we have used a bootstrapping procedure and the macros described in Hayes. 42 The results show that the indirect association between gender and self- 

Our main aim was to provide initial support for the reliability and validity of a novel scale that evaluates rumination in relation to COVID-19. We have shown that the C-19RS scale has a good internal consistency and a unitary factor structure, with all six items loading into a dominant factor score. In terms of antecedents for the COVID-19 rumination, gender is a significant predictor. In line with meta-analytic evidence for gender differences in rumination, 37, 38 our results show that women report a higher tendency of ruminating about COVID-19 than men.

Moreover, our results show that the tendency to ruminate about COVID-19 dilutes over time, a pattern associated with the stress reaction phases. Our survey started when the first protective public measures were taken in the Netherlands, therefore we could capture ruminations at the very beginning of this period as well as later on when the public concern was somehow attenuated. The scores of the C-19RS reflect a likely habituation with the stressor. Finally, in terms of predictors, our criterion group approach showed that individuals who report having close relatives seriously ill due to COVID-19 tend to report higher rumination in relation to the virus than respondents who did not report having such cases in their proximal social network. This result supports the sound psychometric properties of the scale, by showing that it discriminates between the groups that differ in their exposure to and proximity to the virus.

Our findings support the good predictive validity of C-19RS as it significantly predicted both emotional exhaustion and self-handicapping. These results corroborate prior findings and align with the key assumptions of TMSC 9 about individuals' appraisal and coping with stressful situations. Rumination related to COVID-19, which is a primary appraisal of the COVID-19 pandemic, triggered a withdrawal response (i.e., self-handicapping) presumably because individuals tend to mentally distance themselves from situations over which they have little or no control (e.g., a coronavirus outbreak). 9 Self-handicapping may serve to help individuals justify or rationalise the situation of reduced personal ownership or control, whereby they can, to some extent, psychologically distance NIKOLOVA ET AL.

-9

themselves from the stressor. In these circumstances, self-handicapping represents a more proximal outcome of COVID-19 worries compared to the more distal emotional exhaustion. Also supporting the scale's predictive validity, we found that COVID-19 worries triggered emotional exhaustion; this result feeds back into the theoretical notion that a hindrance stressor, because it can spur feelings of helplessness (i.e., anticipated loss of resources that spans beyond one's control), can tax individual's energy level. 14-43

Despite its theoretical and practical value, the current study has some limitations.

First, the nonprobabilistic sampling procedure used for the data collection prevents us from unconditionally extrapolating our findings to the general population. Only working individuals were included in our study, which excluded people younger than 18 and older than 67. Even though care was taken to preselect participants based on demographics, this preselection was aimed at representing as accurately as possible the working population in the Netherlands and not the general public as a whole; yet, the COVID-19 pandemic impacts everyone, including the nonworking individuals. Future studies may wish to test our instrument on samples including a broader range of age groups, which would allow generalisation of the results for nonworking individuals as well.

Second, an experimental study design could allow a more thorough exploration of the discriminant properties of our instrument. Testing the scale among individuals who are at immediate risk of being affected by the pandemic compared to individuals who are not (e.g., countries where the virus is spreading compared to countries that are not impacted yet) could allow a better understanding of the pre and during-outbreak effects. Given the nonexperimental design of the current study, future contributions may tackle this shortcoming by surveying individuals before, during and after a (corona)virus outbreak.

Perceived risk of COVID-19 pandemic: the role of public worry and trust

Rumination reconsidered: a psychometric analysis

Job strain and rumination about work issues during leisure time: a diary study

Work stressors and impaired sleep: rumination as a mediator

The impact of rumination on sleep quality following a stressful life event. Personality Individ Differ

Effects of ruminative and distracting responses to depressed mood on retrieval of autobiographical memories

Cognitive processes in social anxiety: the effects of self-focus, rumination and anticipatory processing

Effects of work stress on work-related rumination, restful sleep, and nocturnal heart rate variability experienced on workdays and weekends

The Cancer Worry Scale: detecting fear of recurrence in breast cancer survivors

Uncertainty during organizational change: managing perceptions through communication

Uncertainty during organizational change: types, consequences, and management strategies

The reciprocal relationship between job insecurity and depressive symptoms: a latent transition analysis

Threat of losing valued job features: the role of perceived control in mediating the effect of qualitative job insecurity on job strain and psychological withdrawal

Differential challenge stressor-hindrance stressor relationships with job attitudes, turnover intentions, turnover, and withdrawal behavior: a meta-analysis

Extending the challenge-hindrance model of occupational stress: the role of appraisal

Is job insecurity on the rise? Evidence from Canadian perception data

Understanding unemployed people's job search behaviour, unemployment experience and well-being: a comparison of expectancy-value theory and self-determination theory

Towards an Explanation of the Job Insecurity-Outcome Relationship: The Role of Perceived Control

Self-perpetuating properties of dysphoric rumination

The perseverative cognition hypothesis: a review of worry, prolonged stressrelated physiological activation, and health

The differential role of mental rumination among industrial and knowledge workers

The role of rumination in depressive disorders and mixed anxiety/depressive symptoms

Ruminative coping with depressed mood following loss

Exploring the relationship between work-related rumination, sleep quality, and work-related fatigue

A multidimensional structure for repetitive thought: what's on your mind, and how, and how much?

Work and Rumination. Handbook of Stress in the Occupations

Perceived self-efficacy and cognitive rumination

Perfectionism cognitions, rumination, and psychological distress

Self-focused attention in clinical disorders: review and a conceptual model

Attention and Self-Regulation: A Control Theory Approach to Human Behavior

An analysis of the self-handicapping scale

Use omega rather than Cronbach's alpha for estimating reliability

Understanding and using factor scores: considerations for the applied researcher

Validity of psychological assessment: validation of inferences from persons' responses and performances as scientific inquiry into score meaning

Gender differences in rumination: a meta-analysis

Specificity of rumination in anxiety and depression: a multimodal meta-analysis

Consequences of self-handicapping: effects on coping, academic performance, and adjustment

Co-rumination partially mediates the relationship between social support and emotional exhaustion among graduate students

Passion for work and emotional exhaustion: the mediating role of rumination and recovery

PROCESS: A Versatile Computational Tool for Observed Variable Mediation, Moderation, and Conditional Process Modeling

The effects of job insecurity on psychological health and withdrawal: a longitudinal study

COVID-19 Rumination Scale (C-19RS): Initial psychometric evidence in a sample of Dutch employees

The authors thank the Open Universiteit for awarding the Wetenschapsprijs to one of the authors, which made this research possible.

The authors declare that there are no conflict of interests.

The study complies with the Declaration of Helsinki, and the data collection method has been GDPR approved by

the Open Universiteit, The Netherlands.

https://orcid.org/0000-0003-0908-0724