COVID-19 incidence, hospitalizations and mortality trends in Croatia and school closures

Public Health. 2021 Sep;198:164-170. doi: 10.1016/j.puhe.2021.07.030. Epub 2021 Aug 3.


Objectives: COVID-19 pandemic affected majority of students worldwide because school closures were one of the first and frequently taken measures in tackling epidemic. This study analyzed trends in COVID-19 morbidity and mortality from the beginning of pandemic in Croatia, in relation to schools opening and closing.

Study design: Retrospective data review.

Methods: Data on COVID-19 positive patients in Croatia from week 9 of 2020 to week 10 of 2021 in Croatia were analyzed using joinpoint regression. Analysis also included hospitalizations and mortality trends for age groups 26 to 65 and 66+ from week 30 of 2020.

Results: Schools opened in fall after the summer holidays in week 37. Joinpoint regression analysis revealed a statistically significant increase in cumulative incidence rates of COVID-19 in all age groups until 50th week, except in the 19-25 age group which saw an increase until 49th week. During the period of increase, there were periods of moderate increases and rapid increases in incidence that were present between 39/41 week and 43/44 week in all age groups except in those 0-6 years [from 40th till 43rd week in age groups 7-14 and 15-18, average percentage change (APC) = 87.41, P = 0.035, and APC = 83.47, P = 0.013; from 39th till 43rd in 19-25, APC = 91.90, P = 0.002; from 40th till 44th in 26-65, APC = 74.79, P < 0.001; from 41st till 44th in 66+, APC = 81.95, P = 0.004]. Steeper increase in hospitalizations was seen in 40th week for age groups 26 to 65 (40th to 45th week APC = 34.67, P < 0.001) and 66+ (40th to 45th week APC = 38.76, P < 0.001). Steeper increase in mortality started in 41st week for both age groups 26 to 65 and 66+ (41st to 46th week APC = 59.59, P < 0.001 and 41st to 45th week APC = 70.28, P < 0.001). Schools were closed for winter holidays in week 51. A steep decrease occurred in week 50 for cases and in week 51 for mortality and hospitalizations. There was no significant increase in hospitalizations and mortality after schools were re-opened in week 03 of 2021 (primary schools) and week 07 (secondary schools).

Conclusion: COVID-19 morbidity and mortality trends in Croatia observed in fall 2020 in Croatia perhaps cannot completely exclude potential association of school opening in all age groups. However, in winter 2021 effect was completely lacking and numbers were independent of schools' dynamics. The observed inconsistent pattern indicates that there were no association of school openings and COVID-19 morbidity and mortality trends in Croatia and that other factors were leading to increasing and decreasing numbers. This emphasizes the need to consider the introduction of other effective and less harmful measures by stakeholders, or at least to use school closures as a last resort.

Keywords: COVID-19; Joinpoint regression; Return to school.

MeSH terms

  • COVID-19*
  • Child
  • Child, Preschool
  • Croatia / epidemiology
  • Hospitalization
  • Humans
  • Incidence
  • Infant
  • Infant, Newborn
  • Pandemics*
  • Retrospective Studies
  • SARS-CoV-2
  • Schools