Endocrine Abstracts

Introduction: In the Covid-19 era mass media present people worldwide, including in Greece, to transiently flock to green spaces and parks. A study from the Western United States in 2020, indicated that indeed, park visitation may have increased by 20% compared to baseline (the immediate pre-Covid-19 period). The Google mobility index (GMI) is based on location data/visits - from smartphone users - to various areas; among areas of mobility the GMI can focus specifically on visits to parks. Studies have shown that internet searches honed on diets may have fluctuated during the Covid-19 era, while obesity may have increased.

Aim: To assess whether the GMI for parks, as a surrogate of physical activity, is related to Google Trends (GT) searches for diet and obesity.

Methods: We extracted the GMI data for Greece, regarding people’s mobility in parks, from March 2020 to February 2022; the data were analyzed by two-way analysis of variance (ANOVA) vis-à-vis season and non-lockdown (NLD)/lockdown (LD) time. For the same time period we also extracted the GT data for internet searches in Greece regarding ‘diet‘ and ‘obesity‘ (in English and Greek). We assessed the autocorrelation of the parameters, to search for periodicity/seasonality, and we performed cross-correlation analysis of the GMI vis-à-vis GT searches. Separate cross-correlation analyses were done for NLD and LD time periods also.

Results: Park visitations’ GMI vs baseline had a markedly dimorphic aspect during NLD/lD time periods; in winter, NLD: +16%, LD: -11%, in spring, NLD: +38%, LD: +9%, in summer, NLD only: +150%, in autumn, NLD: +76%, LD: -5% (p were 0.067 for comparisons by season and <0.001 for comparisons by NLD/lD status). The autocorrelations for periodicity/seasonality of the parameters did not reach statistical significance. Mostly negative cross-correlation coefficients for GMI against GT, lagging mainly from to +2 to +11 weeks (with GT trailing GMI) were noted, ranging from - 0.22 to - 0.56 (P<0.05). Analyses done separately for NLD and LD periods yielded analogous results.

Discussion: Surprisingly, during LD periods, when stay-at-home mandates were in effect, whereas walking in parks was allowed for exercise, park visitations mostly declined. Moreover, periods of increased mobility in parks during the Covid-19 era were coupled, with a time delay, to a drop in internet searches for ‘diet‘ and ‘obesity‘. Whether this represents a counterintuitive perception in the community, in Greece, that physical activity in parks renders dieting superfluous remains to be evaluated.