Attitudes towards COVID‐19 vaccination: A comparison between persons with different chronicity of pre‐pandemic depressive, anxiety or obsessive–compulsive disorders

Persons with mental disorders face an increased risk of COVID19related hospitalization and mortality.1– 3 This gives reasons to prioritize them for vaccination,4 but their willingness to receive a COVID19 vaccine remains unclear. We investigated the association of attitudes towards COVID19 vaccination with mental disorders and symptom severity in persons with and without depressive, anxiety or obsessive– compulsive disorders (OCD). Between 1 April 2020 and 11 June 2021, fourteen waves of online COVID19 questionnaires were distributed among participants (n = 2748) of three psychiatric case– control cohorts that started in early 2000: Netherlands Study of Depression and Anxiety, Netherlands Study of Depression in Older Persons, and Netherlands Obsessive Compulsive Disorder Association Study. Detailed information has been reported elsewhere.5 The question ‘What is your current situation or view regarding vaccination against the COVID19 virus?’ was not included in the online COVID19 questionnaire until 23 February 2021 (the twelfth wave), when vaccines started to become available in the Netherlands. The current study consisted of 936 individuals (response rate 34.1%) with (n = 709) and without (n = 227) mental disorders who responded to the question of attitudes towards COVID19 vaccination at least once at the twelfth, thirteenth and fourteenth waves (mean age 58.7 [SD 12.4] years, 63.1% female). We used participants' first response of attitudes towards COVID19 vaccination because their responses appeared to be consistent across the three waves of online questionnaires. They were categorized as willing to get vaccinated (including vaccinated persons) or unsure/ unwilling to get vaccinated. We repeatedly collected information on whether participants themselves or household members had been diagnosed with COVID19, and whether close contacts died from COVID19 throughout the fourteen waves of online questionnaires. Participants who reported the experience at any wave were categorized as yes, otherwise no. We assessed symptom severity using four rating scales which have been validated in psychiatric cohorts: 16item Quick Inventory of Depressive Symptoms, 21item Beck Anxiety Inventory, 11item Penn State Worry Questionnaire and 6item de Jong Gierveld Loneliness scale,5 at the same time as we collected information on attitudes towards COVID19 vaccination. Mental disorders were diagnosed using DSMIVbased clinical interview at prepandemic waves, including major depressive disorder, dysthymia, panic disorder, generalized anxiety disorder, agoraphobia, social anxiety disorder and OCD. We counted the number of waves a participant attended and the number of waves at which a participant had a current (6month recency) disorder between 2006 and 2016. We divided the number of waves with a disorder by the number of waves attended yielding a ratio that represented the chronicity of disorders. Chronicity was categorized into no lifetime disorder (healthy controls), remitted disorder (participants who at baseline were diagnosed with a disorder that was persistently remitted during the remaining waves), lowtomedium chronicity (1%– 50% of previous waves with disorders) and high chronicity (51%– 100% of previous waves with disorders), conforming to prior analyses.5 Considering the relatively large proportion of nonresponders in our study, we applied propensity score weighting6 to address the potential problem of selective attrition bias. The propensity scores were estimated based on age, gender, education, source cohort and the chronicity of mental disorders. We performed logistic regression models to estimate the association of mental disorders, symptom severity, demographics and COVID19related factors with attitudes towards vaccination, weighted for the propensity scores.