HIGHER RATES OF HOSPITAL TREATMENT FOR PARASUICIDE ARE TEMPORALLY ASSOCIATED WITH COVID-19 LOCKDOWNS IN NEW ZEALAND CHILDREN

During the New Zealand COVID-19 lockdowns, paediatricians reported an increase in parasuicides in children under their care. Such an adverse trend in mental health of children has been noted overseas, with reports of increased childhood eating disorders in Victoria (Australia) that filled hospital wards and emergency departments during lockdowns.¹ The most restrictive lockdowns in New Zealand started on 25 March for 6 weeks, then they recommenced in Auckland during 12–31 August 2020. This resulted in children being confined to their homes with consequent increases in screen time.

To investigate trends in mental health, we requested the last 5 years of monthly counts of hospital diagnoses for children aged 10–14 years from the Ministry of Health, with the International Classification of Disease discharge codes for parasuicide (version 9 codes E950–E958) for all of New Zealand. The data included counts from mid-2015 to the end of 2020.

Trends in the rate of hospital admissions for parasuicides were investigated using seasonal trend decomposition, a descriptive technique, which decomposes a time series into long-term trend and seasonal components. R (version 4.0.4 (R core team, R Foundation for Statistical Computing, Vienna, Austria)) software was used.

The results are portrayed in Figure 1 and show a clear upward trend in the latter half of 2020 from a stable baseline. The raw data are found in the uppermost plot with a sharp increase from base rates observed in August 2020 from a baseline of about 40 children per month to a peak of 90 cases. Rates have remained high, but have subsequently declined, but not back to baseline. The raw counts are decomposed into a long-term ‘trend’ (second plot), which reinforces the impression gained from the raw data. The ‘seasonal’ component shows peaks of parasuicide occurring in October of each year, with a gradually increasing magnitude of seasonal fluctuation. The ‘irregular’ plot shows the difference between the observed and the sum of seasonal and trend components, and represents a measure of model fit. The fit is worst for the peak observed in September 2020.

Anecdotal clinical experience from paediatricians during the 2020–2021 COVID-19 period suggests not only increases in parasuicides, but also in children with somatic symptoms, which are likely related to anxiety. This has led to an increase in violent and aggressive behaviour on wards and consequent stress for health-care professionals involved in their care.

In a meta-analysis of studies, an adverse association between lockdowns and youth mental health was observed, manifesting as depression and anxiety.² In a survey of Chinese primary school students in Hubei province during the lockdown, almost a quarter of respondents reported depressive and anxiety-related symptoms.

Several studies do not support the use of lockdowns to contain cases and fatalities related to COVID-19 overseas.^{3, 4} Here, we have illustrated the clear detrimental effect of COVID-19 lockdown policies on child mental health, which is consistent with clinical and overseas experience. We suggest that this evidence is considered when contemplating the use of lockdowns in New Zealand and overseas.