Descriptive statistics and the structure of the OCS were established. The unidimensional model that consists of four items obtained the required value of chi-square: χ(2)2=0.37, p= 0.831, N = 709, and the model-to-data fit indices were very satisfactory: RMSEA = 0 [0, 0.044, 90% CI], SRMR = 0.01, CFI = 1.00, and TLI = 1.00. All factor loadings were adequate because their values varied from 0.41 to 0.83. Item-scale Pearson's correlations varied from r = 0.49 to r = 0.85 (see Table 1) and indicate good properties of the scale.

Then, we tested whether the OCS had the same measurement characteristics across all groups involved in the study (gender, inhabitancy, infection experience). Multivariate normality of the OCS in all groups was tested (see Table 2). The results indicated that the assumption was violated. Because the variables were ordinal, we applied the DWLS estimator in all tested models (36).

First, measurement invariance across gender was established. The results indicated perfect model fit-to-the data for girls and boys. Moreover, in accordance with adopted criteria, equivalence on the configural, metric, and scalar level was fulfilled (see Table 3). All factor loadings were acceptable (>0.4, see Table 2). Then, we tested measurement invariance across inhabitancy. The results indicated an excellent model fit for adolescents from urban and rural areas. Additionally, the results confirmed multivariate invariance on the configural, metric, and scalar level (see Table 3). Although in urban children loading factor in OCS4 was slightly below the adopted criteria (β = 0.35), other loading factors were acceptable (>0.4, see Table 2). Finally, we tested measurement invariance across infection experience. The results indicated a very good model-fit-to-the-data for infected and non-infected children. It can be assumed that configural, metric, and scalar equivalence exists. Although we observed between configural and metric level ΔSRMR = 0.02, and according to assumptions, this change is slightly higher than criterion—should be <0.015—other criteria were fully met (see Table 3). In non-infected children we observed loading factor below 0.4 in OCS4 (β = 0.35), but others loading factors were acceptable (>0.4, see Table 2). Descriptive statistics for each group are presented in Table 2.

Then, we compared groups. The results indicated that girls significantly exceed boys in the level of COVID-19 obsession thoughts: t₍₇₀₆₎ = 2.35, p = 0.019 and difference is small (d = 0.18). There was no differences in the OCS score between adolescents from rural and urban areas: t₍₇₀₇₎ = −0, 80, p = 0.422, d = 0.06 and groups of infected and non-infected: U = 36832.5, p = 0.628, d = 0.09. The results in all groups indicated right-skewed distribution, which means that most adolescents revealed a low level of obsession with COVID-19.

Summing up, results indicated that the OCS is a unidimensional scale that met the criteria for configural, metric, and full scalar invariance across gender (girls, boys), inhabitancy (urban, rural), and infection experience (infected, non-infected). Considering the full invariance of the OCS between groups, the descriptive statistics of the scale were calculated for the whole Iranian sample. The average level of OCS was weak (M = 0.76, SD = 0.66, N = 709, range 0–3), its distribution was right-skewed (skewness = 1.05, W = 0.90, p < 0.001) and close to mesokurtic (kurtosis = 0.69). The results in all groups indicated that most adolescents revealed a low level of obsession with COVID-19. Descriptive statistics for each group are presented in Table 2.

Finally, the reliability of the OCS was established for each group: Cronbach's α [95% CI] = 0.78 [0.74, 0.81] for girls, 0.75 [0.70, 0.79] for boys, 0.78 [0.74, 0.81] for rural children, 0.74 [0.69, 0.79] for urban children, 0.83 [0.81, 0.85] for infected with COVID-19, and 0.74 [0.66, 0.81] for those who were not infected. Composite reliability for the whole scale was 0.779.

The main objective of this study was to test the factor structure and reliability of the Obsession with COVID-19 Scale in an Iranian adolescents. The results confirmed unidimensionality of the OCS and its measurement invariance across gender, inhabitancy, and infections experience. Moreover, the Persian language version of OCS had a high internal consistency in entire sample and across various groups. The findings provided more evidence for a universality of the OCS structure presented by researchers from various countries (27–30). Thus, the Persian language version of the scale may contribute to further international research on the level of COVID-19 obsession.

Although the obsessive thoughts of COVID-19 disease is raised worldwide (42), the current study revealed relatively low level of obsessive thoughts about pandemic in Iranian adolescents. However, it doesn't mean that there are no adolescents who experience such obsessive thoughts. Students differ in the level of experienced COVID-19 thoughts, therefore OCS may be used as a screening tool that enables the observation of extremely high level of obsessive thoughts. There were no differences in the level of obsessive thought related to COVID-19 due inhabitancy and infection experience. It is likely that regardless of these factors, adolescents may count on appropriate help and access to medical care. However, girls reported more obsessive thoughts than boys. This result is consistent with previous findings that girls tend to report significantly higher depression and anxiety levels (43). Small differences between girls and boys may result from specific situation which is pandemic.

Our study has clear advantages and limitations. The OCS has only four items which facilitates its use in a short time. In the future, however, it can be consider modifying the scale and removing the last item. The fourth item had relatively lower psychometric properties compared to the other items, which is probably due to its separate content (related to dreams, not to daily thoughts). One of the limitations of our analysis is also that the infected/non-infected and rural/urban groups were unequal. Although measurement invariance in these groups is confirmed, it should be remembered that this group bias may influence the mean results of the tested population (especially the inhabitancy; infection is not something permanent).

Moreover, our sampling was not random and is limited to the group that we had access. We also did not controlled the language fluency and nationality of the participants. We tested adolescents who attended Iranian school and we assumed that all of them speak fluently in Persian language. This point is more than important because our study has all the disadvantages of self-report research (44). The results could be affected by temporary mood, level of mindfulness in reading items and instructions, the belief that participation in the study is important and makes sense, willingness to present themselves in a chosen way (45). The presented results should be treated as a starting point for further validation studies. The study was conducted only among adolescents which limits the possibility of generalizing the results to the whole Iranian population. Adults may experience more COVID-19 thoughts as they try to deal with own problems as well as resolve their children issues.

Moreover, this study was dedicated to establishing factor structure and reliability but convergent and divergent validity also should be tested. The future study should focus on relations between OCS to e.g. behaviors to avoid contamination or relations with obsessive-compulsive personality traits in adolescents and adults. Further studies should also check test-retest reliability to establish whether the responses are stable over time.

Diagnosis of pandemic related obsessive thoughts in adolescents is needed as the COVID-19 pandemic is still on going. As experts point out, it can be expected that the effects of the pandemic will be observed in the coming years (46). The period of adolescence is the time when young people establish relationships. During pandemic such developmental aim is difficult to meet. Pandemic as a global and extremely dynamic event has become a source of stress for billions of people around the world demanding unusual countermeasures (7). The lockdown led to the forced isolation of entire societies. A broadly defined lifestyle required change: the way we spend our free time, work and study (6). As this is the first time that modern generations have faced this kind of threat, it is important to test impact of pandemic on the young people mental health. As the pandemic is a global phenomenon, it is a special opportunity to check the impact of state policies on the mental health of citizens in different countries.

During crises, such as a global pandemic, seeking social support is often one of the most adaptable ways to deal with stress. Still, many government agencies have published guidelines on social distancing and hygiene. In other words, many of the recommendations to prevent the spread of COVID-19 include social distancing, which might prevent effective social support, which is required for well-being (47, 48). Many other strategies for dealing with stress, such as active coping with stress to alleviate problems, have also been reported to be effective during a crisis. Others, such as coping with substance use, have been found to be universally incompatible. Other strategies, such as distraction, might be context-dependent (49, 50). Prolonged restrictions of this type may have negative effects on mental health not only immediately but also in the long term. The use of a screening tool will help to prevent negative effects of pandemic on adolescents' future life.

Summing up, the COVID-19 pandemic triggered physical, psychological, social, and economic impacts that have resulted in intense anxiety, depression, obsession, compulsion, etc. Adolescents are particularly vulnerable to emotional difficulties, therefore, it is crucial to support such group. To do so, it is essential to have reliable tools to diagnose the thoughts related to pandemic. It is possible thanks to the Persian version of the OCS which is a unidimensional and reliable scale. It can be dependably used for psychological research and individual diagnosis across various groups, including boys, girls, infected, non-infected, rural, and urban adolescents. As this study should be treated as a starting point for further studies, more data should be collected in the future to establish validity of the OCS in Iranian population.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Counseling Department Review Board in University of Isfahan, Iran. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

FA, MS, and AA contributed to the concept of the study. FA organized and conducted the research. MS performed the statistical analysis. All authors wrote the manuscript and accepted its final version.

This work was supported by funds of the vice-rector of Pedagogical University of Krakow, granted to MS on the basis of the application number BN.302.257.2021.RN.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.