Discussion

The data presented in Table 1, alongside Figs. 1 and 2, provide a comprehensive overview of referral patterns to our PLP service from 1 January 2018 to 31 December 2022. Table 1 highlights a substantial increase in the total number of referrals over the 5-year period with an escalation post the emergence of COVID-19, underscoring the profound impact of the pandemic on children's mental health. This increase is mapped on a month-by-month basis in Fig. 1, which shows three sizable spikes around May 2020, April 2021 and December 2021. These align with three periods of intensified public health measures and societal disruptions, including lockdowns and transitions in educational settings in Ireland. Furthermore, Fig. 2 expands on this analysis by categorising referrals by diagnostic criteria, revealing notable increases in self-harm, suicidal ideation and eating disorders during the pandemic. The authors speculate that this rise may be indicative of the heightened stress and isolation experienced by children during lockdown periods.

Using time series analysis, we observed that the COVID-19 pandemic had a significant effect on the overall number, gender distribution and clinical presentations of referrals. Overall numbers were increased as was the number of females and males referred to the PLP post pandemic onset. There was also an effect found with an increase in the number presenting with eating disorders, low mood, self-harm and oppositional disorder. This aligns with evidence from other Irish studies, showing increased mental health attendances^{Reference McDonnell, Barrett, McNicholas, Barrett, Conlon and Cummins11} and admissions^{Reference McDonnell, Conlon, McNicholas, Barrett, Barrett and Cummins17} despite reductions in general paediatric attendances for all other reasons.

Hernandez-Calle et al^{Reference Hernandez-Calle, Andreo-Jover, Curto-Ramos, Garcia Martinez, Valor and Juarez18} conducted a very similar study, also using interrupted time series analysis, examining rates of referrals to a Madrid paediatric emergency department pre- and post-COVID-19. They reported stable pre-COVID-19 referrals (October 2018–February 2020), followed by a significantly sustained increase, post-COVID-19, up to April 2021. They reported a specific increase in referrals for suicide-related phenomena, anxiety and depression. Unlike many other studies, they did not find any increase in referrals for eating disorders. Our study supports and extends these findings, with a larger cohort and a longer duration of study. Consistent with Hernandez-Calle et al, we also found an increase in post-COVID-19 referrals, which continued until December 2022, but did not find any effect on referrals for anxiety or depression. Consistent with many national and international studies,^{Reference McDonnell, Conlon, McNicholas, Barrett, Barrett and Cummins17,Reference Hansen, Stephan and Menkes19,Reference Otto, Jary, Sturza, Miller, Prohaska and Bravender20} and at odds with the Spanish cohort, we found an increase in referrals of eating disorders. To what extent these differences reflect difference in access to other services during COVID-19 has not been examined. These increases were over and above what would have been expected by a continuation of the pre-COVID-19 upward trend (Fig. 3).

In addition, a number of systematic reviews examining international data indicated increased rates of general depression symptoms^{Reference Ludwig-Walz, Dannheim, Pfadenhauer, Fegert and Bujard21} and self-harm during COVID-19. However, published data suggest a stronger increase for females than males,^{Reference Madigan, Korczak, Vaillancourt, Racine, Hopkins and Pador22} in contrast to our finding, which indicates an increase for both genders. However, this increase in both genders may be because of the overall increased number of referrals experienced during COVID-19. A UK study reported self-rated psychological problems, depression and functional impairment, and psychiatric referrals appeared higher among youths after school re-opening post lockdowns.^{Reference Sayal, Partlett, Bhardwaj, Dubicka, Marshall and Gledhill12} The ROI and New Zealand have reported increases in eating disorders, with descriptions of a ‘tsunami-like' increase of presentations with eating disorders,^{Reference McDonnell, Conlon, McNicholas, Barrett, Barrett and Cummins17,Reference Hansen, Stephan and Menkes19,Reference Leask23} which are consistent with our results. Also, we find the COVID-19 pandemic did not result in any significant effect on the number of referrals with suicidal ideation alone, or the number of youth presenting with other diagnostic disorders, such as anxiety, ADHD, ASD and major depression. Similar findings of reduced referrals for behavioural difficulties and ADHD have been reported when examining general practitioner (GP) referrals to, and attendance at, CAMHSs.^{Reference FitzPatrick, Antczak, Lynch, Lynch and McNicholas24} Rates of ASD admissions to paediatric hospitals have also been reported to be reduced.^{Reference McDonnell, Conlon, McNicholas, Barrett, Barrett and Cummins17} This finding can be attributed to the decreased visibility of symptoms to external observers, as social restrictions could have led to fewer opportunities for teachers and coaches at school who typically observe and refer youth for these issues to interact with them, potentially underreporting or delaying recognition of symptoms. In addition, the limited social interactions during lockdowns might have made it less likely for exacerbations in symptoms that typically prompt hospital admissions to occur.

The finding of a lack of specific impact of the COVID-19 phases of restrictions on self-harm or psychiatric disorder type, such as depression or eating disorders, does not imply that restrictions did not have an effect on the genesis of illness and help seeking, but rather it was not observable at the time the restrictions were in place. It also recognised that disaster-induced psychological impacts are often delayed. Whilst onset of various disorders may commence during restrictions and be triggered by them, presentation to services may be delayed. This may be linked to ‘stay at home’ requests or the gradual emergence of symptoms to reach levels of impairments sufficient for help seeking. For example, presentation of eating disorders may be delayed until significant weight has been lost, leading to medical compromise. Equally, the emergence of low mood, reduced social interest or poor attention may be perceived by parents as understandable, expected or justified during periods of restrictions, assumed to recover post restrictions, and hence help seeking delayed until restrictions are removed but obvious impairment persists. Anxiety disorders triggered by restrictions may only reach a threshold for presentation following return to school and/or social re-engagement. This might explain why over the period of COVID-19, presentations of eating, depressive and anxiety disorders were elevated, but not specifically linked to restriction phases. There is already published evidence that for youth with premorbid social anxiety, school closures were linked with improvement in youth mental health.^{Reference Widnall, Winstone, Plackett, Adams, Haworth and Mars25}

There is a variety of reasons for the increased number of referral and clinical presentations owing to COVID-19 and the associated restrictions. Childhood and adolescence represent sensitive developmental periods, and prolonged deprivations of personal, educational and societal liberties are now recognised to have had significant adverse effects.^{Reference Loades, Chatburn, Higson-Sweeney, Reynolds, Shafran and Brigden26,Reference Viner, Russell, Saulle, Croker, Stansfield and Packer27} This in part was reflected by findings from studies showing higher rates of referrals as restrictions were lifted and access re-opened,^{Reference Viner, Russell, Saulle, Croker, Stansfield and Packer27} and higher referral rates following school re-opening and re-establishing academic and peer contact after a period of absence.^{Reference Sayal, Partlett, Bhardwaj, Dubicka, Marshall and Gledhill12}

The significant psychological impact of COVID-19 and associated restrictions have been well documented.^{Reference Chadi, Ryan and Geoffroy14} Many countries examined the lived experiences of youth during COVID-19 and the impact of COVID-19 restrictions. Large cohorts (>20 000) of youth reported decreasing levels of mental health well-being as the pandemic progressed, particularly in anxiety and depression. Youth reported increased worry, anxiety, depression and concerns related to missing friends and school.^{Reference Ougrin, Wong, Vaezinejad, Plener, Mehdi and Romaniuk28,Reference Panda, Gupta, Chowdhury, Kumar, Meena and Madaan29} Others reported increased difficulties linked to school return and difficulty accessing specialist services.^{Reference Ougrin, Wong, Vaezinejad, Plener, Mehdi and Romaniuk28,Reference Young30} Qualitative studies highlight feelings of social isolation, depression and anxiety, and increases in maladaptive behaviour during restrictions, and families of children with ASDs reported increased mental health difficulties linked to changes in routine.^{Reference O'Sullivan, Clark, McGrane, Rock, Burke and Boyle31}

Beyond the direct effects of the pandemic, increased family interactions during prolonged periods at home may have also contributed to the rise in mental health referrals. As families spent more time together owing to public health restrictions, parents and caregivers became more aware of changes in their children's behaviours and emotional states, potentially leading to greater recognition and reporting of mental health concerns. Parenting stressors relating to working from home, addressing educational needs of children and lack of suitable working space, may all have led to heightened expressed emotions. In addition, the intensified family dynamics and reduced access to external support systems during lockdowns could have exacerbated existing issues, further driving the demand for paediatric mental health services. Moreover, evidence from this period reveals that parental anxiety and depression became increasingly prevalent during lockdowns,^{Reference Adegboye, Williams, Collishaw, Shelton, Langley and Hobson32} with the findings of a direct correlation between financial strains and parental mental health, which indirectly, yet significantly, influenced the mental well-being of their children.^{Reference Adegboye, Williams, Collishaw, Shelton, Langley and Hobson32}

Notwithstanding adverse mental health impacts for many, there have been a number of children whose mental health improved during the pandemic. Surprisingly, both UK and US studies have reported better outcomes in those with pre-existing mental health difficulties.^{Reference Soneson, Puntis, Chapman, Mansfield, Jones and Fazel33,Reference Penner, Hernandez Ortiz and Sharp34} It is possible that for some working from home facilitated lower parental stress and increased quality time with parents; school closures and online learning may have reduced certain school-related stressors and social anxiety. What is less certain is whether any initial benefits persisted as the restrictions waxed and waned, schools re-opened and closed, financial aspects of reduced or altered working hours changed and whether resilience tanks ran dry. Even in studies reporting an average reduction in mental health outcomes in their cohorts, there are subgroups who did well, ranging from 8% to 30%, depending on whether self- or parent-rated.^{Reference Soneson, Puntis, Chapman, Mansfield, Jones and Fazel33}

Examining the impact of public health measures in addition to the impact of the pandemic is important to ensure that future public health measures, when enforced, are done so in an informed manner, weighing the balances for and against associated risks. For example, Ludwig-Walz et al^{Reference Ludwig-Walz, Dannheim, Pfadenhauer, Fegert and Bujard21} systematically examined levels of depression among children and adolescents across Europe. They found a post-pandemic increase in depression symptoms in general but noted that this increase was more pronounced when public health measures were more stringent or associated with school closures. This suggests that social distancing policies enforced as part of the pandemic containment may have had specific negative impacts on youth. They also reported that public health measures had a higher impact among males than females, possibly linked to lower levels of depression in males pre-pandemic. This might have relevance to our study, which found a pandemic-related increase in overall numbers specific to both female and male gender, but we did not find any gender-specific effect when examining COVID-19 restrictions. Neither the pandemic onset nor restrictions had any specific impact on MDD. Taking pre-pandemic trends and trajectories into account in study analysis is also crucial.