Design and sample
Between 27 th January and 28 th February, 2021, we invited all full- and part-time baccalaureate nursing students > 18 years of age from five Norwegian universities at ten different campuses (N = 6088) to take part in a web-based cross-sectional survey. The participating universities were Oslo Metropolitan University, Western Norway University of Applied Sciences, the University of Agder, the Norwegian University of Science and Technology and the University of Stavanger.
Measures
The survey included questions related to students’ demographics, personal health and study situation during the pandemic, specifically developed for the present research by an expert group consisting of clinicians, nursing students, university staff and researchers. Additional measures included four validated instruments for assessing fear of COVID-19, overall QoL, general health and psychological distress.
Characteristics of the respondents included age (< 25, 25-29, ≥ 30 years), household status, study site and year of study.
COVID-19 specific questions related to personal health were developed for the present study and included the number of times the student was tested for COVID-19 (never, 1, 2 , 3 or ≥ 4 times); quarantine history (never, previous, present); feelings of loneliness due to COVID-19 (rated from 1 [strongly disagree] to 5 [strongly agree]); perceived risk for complications of COVID-19 (no, uncertain, yes); history of suspected, possible or confirmed COVID-19 infection; intention to take the vaccine (already taken, yes, undecided, no); and trust in authorities’ and universities’ handling of the pandemic (rated from 1 [strongly disagree] to 5 [strongly agree]).
COVID-19 specific questions related to education addressed students’ perceived impact of the different aspects of their education, especially the impacts related to the conduct of clinical training and placements.
The Fear of COVID-19 Scale (FCV-19S) [26], which had been adapted and assessed for use with Norwegian samples, was used [27]. Seven items (e.g. ‘I am most afraid of the coronavirus’) are rated on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree), with a total score ranging from 7 to 35. Higher scores represent greater fear of COVID-19. In the present study, the average item score was used; it was calculated by dividing the total score by the number of items.
The Hopkins Symptom Checklist (SCL-5) [29] is available as a Norwegian translation [30]. It consists of five items measuring psychological distress (anxiety and depression) that are rated on a five-point scale from 1 (not at all) to 5 (extremely). The average item score was calculated by dividing the total score by the number of items answered [15]. Higher scores represent greater psychological distress.
General health was assessed using one item derived from the 36-Item Short-Form Health Survey (SF-36) [31], ‘In general, would you say your health is: excellent, very good, good, fair or poor?’ Responses were rated on a five-point scale ranging from 1 (excellent) to 5 (poor)[32]. Consistent with the SF-36 scoring algorithm, the scale was reversed scored [33]. Thus, higher scores reflect better general health, as perceived by respondents. The item was found to be as valid and reliable as multi-item scales [32].
Overall quality of life was rated on an adapted version of the Cantril Ladder, on a scale from 0 (not at all satisfied) to 10 (highly satisfied). A score of 6 or more indicates ‘high life satisfaction’ [34]. The question, ‘All in all, how satisfied are you with your life at this time?’, has been widely used in various populations and in different settings; it is considered a valid and reliable measure of overall QoL [35].
Participants’ results on the FCV-19S, SCL-5 and measures of general health and overall QoL were compared to reference data. At present, the FCV-19S data from the general population are unavailable. Thus, to compare the students’ score on the FCV-19S we used scores from an urban adult Norwegian population [27]. For the SCL-5, the nursing students’ scores were compared to those of Norwegian first-year medical students [15]. For general health and overall Qol, students’ scores were compared to the Norwegian reference values for the SF-36 [31] and the Norwegian Survey on Living Conditions [36], respectively.
The questionnaire was piloted with 9 nursing students, and after minor adjustments, a digital pilot study was conducted with 90 physiotherapy students. No adjustments were made after the digital pilot. A brief description of the study and an invitation to the web-based survey was e-mailed to 6088 baccalaureate nursing students’ registered university e-mail addresses and made available on the respective learning portals of their teaching institution. At two universities, additional announcements were made at the students’ common Facebook site. All students received at least two reminders by e-mail.
The front page of the survey contained a detailed description of the study and information about voluntary participation. By completing and submitting the survey, the students consented to participate. All responses were stored automatically in ‘SurveyXact’ (https://www.surveyxact.com). The respondents’ IP addresses were not registered and their answers could not be linked to their identities in any way; thus, their participation was anonymous. The survey was evaluated by the Data Protection Officer at the responsible institution, i.e. Western Norway University of Applied Sciences, with additional approval of each university.
Statistical analyses
Categorical variables are expressed as percentages and continuous variables as means and standard deviations (SD). The FCV-19S scores were stratified by sample characteristics, using separate one-way analysis of variance (ANOVA) tests. Differences between the sample and reference data were investigated using a one-sample t-test. Reference data, except for FCV-19S scores, were adjusted to reflect the age and gender distributions of the respondents, assuming the proportion of males was similar to that of the general nursing student population (approximately 10%). Cohen’s d was used to calculate the effect sizes of the comparisons of means. Unadjusted and fully adjusted hierarchical regression analyses, with the universities as clusters, were conducted to investigate the associations between the FCV-19S score as the independent variable, and the SCL-5 general health or overall QoL score as the dependent variable in separate models. In the regression analysis we standardised the FCV-19S and the three dependent variables where the mean = 0 and standard deviation = 1 (dependent variables were transformed to z-scores, unstandardised regression coefficients). From the fully adjusted models, the associations between other items from the questionnaire and the SCL-5, general health and overall QoL as dependent variables were assessed and reported separately if they had meaningful effect sizes, as assessed by Cohen’s d. A meaningful Cohen’s d was judged to be a difference ≥ 0.2 SD of the dependent variable per 2 SD changes in the FCV-19S or between respondents representing the lower or higher end of the discrete variables with 2-5 categories [37; 38].
Overall, the effect sizes were interpreted as follows: trivial (< 0.2 ), small (0.2 to < 0.5), moderate (0.5 to < 0.8) and large (≥ 0.8 )[39]. We reported two-tailed P-values and 95% confidence intervals (CI) as continuous indicators of the robustness of the estimates. Survey data were downloaded to Microsoft® Excel®, manually coded, and then transferred to IBM SPSS (Statistics for Windows, Version 27.0. Armonk, NY: IBM Corp) for the statistical analyses.