Introduction
Depression and anxiety are the leading causes of morbidity among youth worldwide and collectively contribute to more than ten million disability-adjusted life years annually [
1]. This results, in part, because these two conditions are responsible for half of all suicides among youth. Youth with depression or anxiety are also more likely to suffer from additional comorbid behavioral problems such as attention deficit hyper-activity disorder and conduct disorder, have decreased school performance, and are more likely to engage in risky behaviors such as risky sexual behaviors and drug and alcohol abuse [
2,
3]. For these reasons, they are also responsible for a disproportionate share of annual medical expenditures [
4‐
6]. Longer term, youth with these conditions can expect 30% lower lifetime earnings due, in part, due to lower academic achievement, and are less likely to have stable family relationships [
7,
8].
Even before the COVID-19 pandemic, rates of depression and anxiety among youth were increasing in high-income countries. Several countries, including the United States and Finland, had shown a doubling in prevalence in the decade leading up to the pandemic [
9,
10]. Increased social media use, combined with prolonged school closures, social isolation, and general household stress associated with the pandemic has further exacerbated the prevalence of depression and anxiety among youth [
11]. A recent systematic review found that the prevalence of these conditions had more than doubled to 25% and 21% respectively in the first year of the pandemic [
11].
Policymakers in Singapore, the country of focus of this effort, rely on timely information of the prevalence and economic burden of select conditions in efforts to prioritize prevention and treatment initiatives. Although estimates from other countries are available, they may not generalize to Singapore due to myriad factors, including cultural difference and differences in health seeking behavior. Prior to the COVID19 pandemic, although no economic burden data exist, Magiati et al. reported that 9.3% and 16.9% of 8–12 year-old Singaporean children had clinically elevated symptoms of anxiety and depression, respectively, in 2015 based on a representative household survey of Singaporean children [
12]. At the height of COVID-19 pandemic from April to June of 2020, home-based learning was instituted across all educational institutions [
13]. Although schools were reopened in June 2020, school activities remained restricted throughout 2020 and 2021 [
14]. Given school closures and other COVID-era restrictions in the city-state, [
15] and increases in social media use and other risk factors, rates of depression and anxiety have likely increased. Currently, no updated information is available for depression and anxiety prevalence among youth and the corresponding economic burden.
As was done by Magiati et al., the gold-standard approach for generating this information is through in-person school- or household-based surveys [
12]. However, these are time-consuming, costly, and difficult to implement. An alternative, convenient and low cost approach, which we take in this study, is to rely on parent reporting of youth prevalence and economic burden via a web panel. We use an existing web panel to estimate the prevalence of depression and anxiety symptoms among Singaporean youth based on parent report and to determine the extent to which these symptoms increase school absences, reduce school performance, hamper daily activities, and increase healthcare utilization/costs. We hypothesize that these symptoms will lead to increases in healthcare utilization and costs and reductions in school performance and performance in regular activities (outside of school). Although we focus on Singapore, this approach can be easily replicated in countries worldwide.
Discussion
This is the first study in Singapore, and among the first internationally, to estimate the prevalence and economic burden of depression and anxiety among youth using a low cost web panel approach with proxy reports from parents. The finding that 16.2% of parents reported symptoms consistent with depression or anxiety among their children and the corresponding healthcare burden and reductions in school and daily activities performance is cause for concern.
To contextualize our prevalence estimates, we compared them to the 2015 estimates of Singaporean youth ages 8 to 12 reported by Magiati et al. based on a household survey and more recent estimates from other countries. Magiati et al. reported estimates of depression and anxiety of 16.9% and 9.3%, respectively [
12]. Using the web panel, parent report, and the PHQ-4 screening criteria, we found that 16.2% of youth ages 4 to 21 had symptoms consistent with these conditions. Given the divergent methodologies a direct statistical comparison is not advisable nor is it possible to tease out the influence of COVID-19 or other factors. However, for several reasons we suspect our prevalence estimate, and therefore burden estimates, are conservative. Not only are our estimates similar to results from 2015, they are lower than those of a recent systematic review of studies post onset of COVID-19. The review found estimates of these conditions ranging from 20.5 to 25.2% among youth aged 18 years or younger [
11]. The most likely reason that our estimates are lower is underreporting in our sample due to reliance on proxy responses. Proxies may not be aware of the symptoms in some cases, may have difficulty recalling in others, or may not wish to disclose [
36]. Lack of awareness, recall bias and systematic underreporting could be exacerbated in Singapore due to the high degree of stigma associated with these conditions [
37]. To this point, our results were similar to those reported by Peng et al., where 16.3% (95% CI: 16.0 − 16.7%) and 10.3% (95% CI: 10.0 − 10.6%) of the high school students surveyed had symptoms of depression and anxiety respectively [
21]. The study was conducted in Guangdong China in April 2020, after students had been attending compulsory home-based distance learning and lockdowns, similarly experienced by youths in Singapore, and where stigma from mental illness is also likely to be high.
Despite the high level of interaction with the health system, 84.8% of children with symptoms consistent with depression or anxiety did not have a formal diagnosis from a healthcare provider. This may result from multiple factors. First may be that those who screen positive do not actually have symptoms that warrant a clinical diagnosis. It is also possible that many of these youth were seen by a healthcare provider who recognized the conditions but did not codify a diagnosis to avoid stigma or because they felt the conditions may be temporary. It is also possible that some parents were told of a mental health diagnosis but did not recall that information or did not wish to disclose it. An alternative, and more concerning, possibility is that many youths with mental health conditions in Singapore remain untreated or undertreated. More research is needed to confirm this but, if confirmed, it represents a significant opportunity to improve mental healthcare as those with a formal diagnosis are most likely to receive effective treatment.
The children whose parents reported evidence of depression and anxiety symptoms incurred significant medical expenses. This may be because the most severe cases are more likely to be reported. Results suggest that of the over $10,000 in annual expenses per child associated with these conditions, 47% are potentially avoidable. This includes ED visits, hospitalizations and expensive differential diagnostic tests. The high utilization of expensive physiological diagnostic tests among youth may be indicative of a cautionary practice for pediatric populations to eliminate all differential diagnoses prior to the diagnosis or treatment of mental illnesses [
38]. Further research is necessary to determine whether routine screening to identify and treat youth with mental health conditions, as suggested by the US Preventive Services Task Force in draft recommendations, would be cost-effective and potentially even cost saving in Singapore [
39].
Addressing mental health among youth requires a whole-of-government approach. Coordinated efforts across ministries is required to offer appropriate severity-based services. In addition to implementing screening and personalized treatment programs, the government could also ramp up programs aimed to reduce stigma of mental health conditions among youth. Examples of such services include the Response, Early Intervention and Assessment in Community Mental Health (REACH) program, and the Community Health Assessment Team (CHAT), which provide community-centric personalized services to children and youth and promote peer support through a multi-sectoral approach [
40‐
42]. They provide low-barrier access to mental health services in a stigma-free community setting. Other programs, such as “Beyond the Label” and “Nurture SG” also aim to increase awareness, reduce stigma, and promote access to care. Quantifying the cost-effectiveness of key components of such strategies should be an area of future research.
Strengths and limitations
This study has both strengths and limitations. The primary strength is the ability to generate estimates of prevalence and burden of depression and anxiety using a low cost and expedient approach that takes advantage of an existing web panel. For the chosen panel, participants are recruited country wide to take surveys on a regular basis in exchange for modest rewards. Most households remain on the panel for 2–3 years. The panel exceeds 500,000 individuals and is broadly representative of the socioeconomic, gender, and ethnicity distributions in Singapore. However, a primary limitation is that it is unclear whether their children are representative of the general population of children in Singapore overall or among the subset with mental health conditions. It is possible that parents are more likely to report the more severe cases. If so our burden estimates may be inflated. In addition, whereas PHQ-4, as the combination of the PHQ-2 and GAD-2, is a validated instrument, it is a screening tool, not a diagnostic, tool. It allows for rapid identification of individuals who have symptoms consistent with these conditions but is not confirmatory. Our burden estimates are also based on self-report and could suffer from recall or other biases. Finally, although our prevalence estimates are based on all household members, our utilization and burden estimates are limited to the oldest child with symptoms of depression or anxiety in the household whose responding parent did not also report these symptoms. If these youth are not representative of all youth with these conditions, then our estimates would be biased. Future studies should aim to improve on our results using more robust school-based samples, face-to-face survey administration directly with youth, clinical diagnostic tools, and linkages to actual healthcare utilization and claims data.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.