Type of information source and healthcare avoidance: insights from two population-based studies during the COVID-19...

The COVID-19 pandemic generated major disruptions in primary and specialist care, and there were global trends of reduced healthcare-seeking behaviour. Previous studies showed that use of information sources is linked to psychological well-being and compliance with public health measures. It is unclear if the type of information source is also associated with healthcare avoidance.

Between April and October 2020, we sent a questionnaire to participants of two population-based cohort studies including middle-aged and older adults from Rotterdam, the Netherlands: the Rotterdam Study (N = 8,732, response rate 71.5%) and the Generation R Study (N = 4,904, response rate 22.7%). We defined healthcare avoidance as not seeking care despite experiencing symptoms that participants would otherwise interpret as requiring medical attention. We pooled the data from both studies and used logistic regression analyses to examine the adjusted association between type and number of information sources and healthcare avoidance. Additionally, we investigated if symptoms of depression and anxiety modified and/or mediated this association.

Among 6,702 respondents (mean age 67.0 years, 61.8% women), 96.0% used traditional media, 15.5% used social media, and 1,197 (17.9%) avoided healthcare. Compared to non-avoiders, avoiders more often reported poor/fair self-appreciated health (28.9% versus 9.7%), and symptoms of depression (31.9% versus 11.6%) or anxiety (29.2% versus 11.1%). Odds of healthcare avoidance were lower for those who consulted traditional media (adjusted OR (aOR): 0.72, 95% CI: 0.53-1.00) and higher among individuals using social media (aOR: 1.21, 95% CI: 1.02-1.44), compared to those who did not use these sources. Additional adjustment for self-appreciated health and anxiety/depressive symptoms attenuated the latter association. Stratified analyses showed a stronger association between social media use and healthcare avoidance among individuals with anxiety symptoms (aOR: 1.34, 95% CI: 0.94-1.89), yet not statistically significant. Additionally, in a fully adjusted model, part of the association between social media and healthcare avoidance was mediated by anxiety symptoms, although the effect estimate of the mediated effect was small (p-value 0.14, natural indirect effect OR: 1.03, 95% CI: 1.01-1.05).

Social media use was associated with higher odds of healthcare avoidance as compared to not using social media, whereas use of traditional media was related to lower avoidance. We found suggestive evidence for a role of anxiety symptoms in these associations, which should be validated by future studies. These findings emphasise the importance of disseminating accessible, objective information across a variety of media platforms, where needed tailored to vulnerable populations, to support informed decision-making during and beyond health crises.