Rapid and unequal decline in adolescent mental health and well-being 2012–2019: Findings from New Zealand cross-sectional surveys

Abstract

Objective:

To investigate prevalence and trends in key mental health and well-being indicators among New Zealand secondary school students.

Methods:

Representative cross-sectional youth health surveys with 2–4% of the New Zealand secondary school population were conducted in 2001, 2007, 2012 and 2019 (total n = 34,548).

Results:

In 2019, 69.1% reported good well-being (95% confidence interval = [67.6, 70.6]; World Health Organization 5-item), 22.8% reported clinically significant depression symptoms (95% confidence interval = [21.4, 24.1]; Reynolds Adolescent Depression Scale–Short Form) and 41.8% reported possible anxiety symptoms (95% confidence interval = [40.5, 43.2]; Generalized Anxiety Disorder 2, adapted). Past-year prevalence of periods of low mood (38.3%, 95% confidence interval = [36.6, 40.1]), deliberate self-harm (24.1%, 95% confidence interval = [22.8, 25.4]), suicide thoughts (20.8%, 95% confidence interval = [19.2, 22.4]) and suicide attempts (6.3%, 95% confidence interval = [5.5, 7.0]) were observed. After relative stability from 2001 to 2012, there were large declines in mental health to 2019. The proportion reporting good well-being decreased (odds ratio 0.71, 95% confidence interval = [0.65, 0.78], p < 0.001), depression symptoms increased (odds ratio 1.96, 95% confidence interval = [1.75, 2.20], p < 0.001) and past-year suicide thoughts and suicide attempts increased (odds ratio 1.41, 95% confidence interval = [1.25, 1.59], p < 0.001; odds ratio 1.60, 95% confidence interval = [1.32, 1.92], p < 0.001). Past-year deliberate self-harm was largely stable. Declines in mental health were unevenly spread and were generally greater among those with higher need in 2012 (females, Māori and Pacific students and those from higher deprivation neighbourhoods), increasing inequity, and among Asian students.

Conclusion:

Adolescent mental health needs are high in New Zealand and have increased sharply from 2012 among all demographic groups, especially females, Māori, Pacific and Asian students and those from high-deprivation neighbourhoods. Ethnic and socioeconomic disparities have widened.

Keywords

Adolescent mental health depression suicide attempts Māori

Introduction

International evidence suggests large and rapid increases in the prevalence of common mental health concerns (e.g. anxiety, depression) among adolescents since around 2012. While often presented as a global phenomenon, most of these findings are from the United Kingdom, the United States and Canada (Collishaw, 2015; Cook et al., 2021; Keyes et al., 2019; Twenge et al., 2019; Wiens et al., 2020). Important, although perhaps less marked, increases have been reported in other Western and Northern European nations (e.g. Haugan et al., 2021), while data from low- and middle-income countries are less available (Erskine et al., 2017). Many studies report the increases in mental distress as concentrated among females, but few provide trends by ethnicity or socioeconomic status.

Aotearoa New Zealand is a multi-ethnic society, with 27% of the population born overseas (Statistics New Zealand, 2020). Indigenous Māori people (comprising 17% of the total population and 25% of the youth population) were colonised largely by people from the United Kingdom. Historically, Aotearoa New Zealand has not had a good track record for adolescent mental health, with the adolescent suicide rate consistently among the highest in developed nations (e.g. Gromada et al., 2020). The burden of mental ill health is unevenly distributed, with long-standing inequities by sex, ethnicity and deprivation. Most notably, Indigenous rangatahi Māori have persistent health inequity compared with their Pākehā (New Zealand European) peers (Clark et al., 2022). In Aotearoa, Te Tiriti o Waitangi (a treaty) was signed between Māori and the Crown to confer (among other things) the right of Māori to equitable outcomes (Reid et al., 2019). There are ongoing and clear breaches of Te Tiriti o Waitangi regarding mental health, with Māori populations disproportionately affected (Clark et al., 2022; Reid et al., 2019; Waitangi Tribunal, 2019). Higher mental health needs are also consistently reported by females (although, notably, more males die by suicide), Pacific young people and those living in higher deprivation neighbourhoods (Fleming et al., 2020c), among other groups.

The Youth2000 survey series comprises large-scale surveys that assess a broad range of health and well-being factors among New Zealand secondary school students, including mental health indicators such as well-being, depression symptoms, self-harm and suicidality. We previously compared mental health results from 2001, 2007 and 2012 (Fleming et al., 2014) and reported little change, with small increases in self-reported depression symptoms, emotional symptoms, peer problems and past-year periods of low mood and self-harm between 2007 and 2012. The prevalence of past-year suicide thoughts and suicide attempts remained largely stable.

The Youth19 Rangatahi Smart Survey (Youth19; n = 7374) is the latest in the Youth2000 series, offering robust, recent, comparable data. Here, we describe the mental health and well-being of New Zealand secondary school students in 2019 and identify important changes between 2012 and 2019 among key population groups.

Methods

Survey procedure

Comprehensive, cross-sectional youth health and well-being surveys were completed by representative samples of New Zealand secondary school students in 2001, 2007, 2012 and 2019.

Detailed methods are available elsewhere (Adolescent Health Research Group, 2003, 2008; Clark et al., 2013; Fleming et al., 2020b; Rivera-Rodriguez et al., 2021). In brief, the surveys used a two-stage cluster design. In 2001, 2007 and 2012, schools throughout New Zealand with over 50 students in years 9–13 (secondary school) were randomly selected and invited to participate. Kura Kaupapa Māori (Māori-language immersion schools) were included from 2007, but sampling methods differed between waves, so Kura Kaupapa Māori are excluded from this analysis to ensure comparability.

Youth19 used a regional sample. Schools with over 50 students in years 9–13 from the Northland, Auckland and Waikato regions of New Zealand were randomly selected. This area includes 47% of New Zealand’s secondary school population and is highly diverse in ethnicity, socioeconomic deprivation and population density (e.g. urban, small town, rural setting). Numbers and response rates of participating schools and students across survey waves are shown in Table 1.

Table 1.

School and student response rates by survey year, 2001–2019.

	Schools invited	Schools participated (response rate)	Students invited	Students participated (response rate)	Surveyed students as % of New Zealand secondary student roll
2001	133	114 (86%)	12,934	9567 (74%)	3.9
2007	115	96 (84%)	12,355	9107 (74%)	3.3
2012	125	91 (73%)	12,503	8500 (68%)	3.0
2019	80	45 (56%)	12,359	7374 (60%)	2.6

In all waves, in participating schools, parents and caregivers were given information about the survey and could opt for their child to be excluded. Non-excluded students were randomly selected from school rolls and gave their own consent. Surveys were delivered on laptop computers (2001) or handheld Internet tablets (from 2007) in te reo Māori (New Zealand’s Indigenous language) or English, with optional voiceover through headphones. The surveys used a branching design, so students only received questions that were relevant to them. From 2007, during the survey, students were invited to enter their main residential address into a standalone computer to identify their census meshblock number (neighbourhood area unit), which was later linked with their survey files.

Ethics approval for each survey was granted by The University of Auckland Human Subjects Ethics Committee (Youth19: #022244).

Measures

Single-item measures are shown in Table 2. Other measures are described below.

Table 2.

Youth2000 survey measures.

Measure	Question	Answer options	Reported
Demographics
Age	How old are you?	Under 12, 12, 13, 14, 15, 16, 17, 18, 19, Over 19	Younger (15 or under)/Older (16 or over)
Ethnicity^a	Which ethnic group do you belong to? (You may choose as many as you need)	168 ethnic groups or ‘I don’t know’	Māori, Pacific, Asian, Other, Pākehā or other European
Neighbourhood deprivation^b	NA	NA	Low deprivation (deciles 1–3) Medium deprivation (deciles 4–7) High deprivation (deciles 8–10)
Locale^c	NA	NA	Urban Small town Rural
Mental health and well-being
Period of low mood	During the past 12 months, was there ever a time when you felt sad, blue or depressed for 2 weeks or more in a row?	Yes/No	Yes/No
Deliberate self-harm	During the last 12 months, have you deliberately hurt yourself or done anything you knew might harm you (but not kill you)?	No, never Yes – once or twice Yes – three or more times	Yes (including either ‘yes’ option)/No
Suicide thoughts	During the last 12 months, have you seriously thought about killing yourself (attempting suicide)?	Yes/No	Yes/No
Suicide attempts	During the last 12 months, have you tried to kill yourself (attempted suicide)?	Yes/No	Yes/No

Answer options according to the Statistics New Zealand level 4 classification. Students who reported multiple ethnicities were assigned to one of five groups according to the Ministry of Health (2017) ethnicity prioritisation method. Groups in order of priority are Māori, Pacific Peoples, Asian, Other (comprising Middle Eastern, Latin American, African and ‘residual’ ethnicities), and European (comprising Pākehā/New Zealand European and other European).

Based on neighbourhood meshblock and the New Zealand Deprivation Index 2018 (NZDep2018), which assesses nine characteristics (e.g. income, Internet access) of households in a meshblock (Atkinson et al., 2019).

Based on neighbourhood meshblock and Statistics New Zealand (2017; the Statistical Standard for Geographic Areas 2018). Urban areas contain 10,000 or more people, small towns 1000–9999 people, and rural areas fewer than 1000 people.

Sex

In 2001, 2007 and 2012, students were asked What sex are you? (Male/female). The 2019 survey included more expansive response options around sex and gender. To enable trend comparison, we used sex at birth (male/female) where this could be identified, allowing the inclusion of transgender and non-binary participants. Results for the 63 students who did not report a male or female sex at birth in 2019 will be reported elsewhere, where contexts can be fully explored.

Good well-being was measured in 2007 using the WHO-5 Wellbeing Index (World Health Organization, 1998). This widely used five-item measure assesses positive mood, vitality and general interests over the last 2 weeks. Scores range from 0 to 25, with scores ⩾ 13 indicating good well-being. The WHO-5 has good reliability and validity (Topp et al., 2015), with a Cronbach’s alpha in this sample of 0.889.

Depression symptoms were measured using the Reynolds Adolescent Depression Scale in 2001 and the short form of the same scale (RADS-SF) from 2007 (Reynolds, 2002). Prior analyses of Youth2000 data indicate that the RADS-SF provides an adequate measure of depression symptoms, with a cut off of ⩾28 providing the closest agreement with the full scale (Milfont et al., 2008). Possible scores range from 10 to 40. The RADS-SF has good reliability and validity (Milfont et al., 2008), with a Cronbach’s alpha in this sample of 0.893.

Possible anxiety symptoms were measured in 2019 only, using the Generalized Anxiety Disorder 2-item screening tool (GAD-2; Kroenke et al., 2009), adapted. Questions were consistent with the original tool: In the last two weeks, I have been bothered by feeling nervous, anxious or on edge and In the last two weeks, I have been bothered by not being able to stop or control worrying. Answer options were adapted from the original four-point scale to a six-point scale, with a scoring representative of GAD-2: All of the time (+3), Most of the time (+3), More than half of the time (+2), Less than half of the time (+1), Some of the time (+1), At no time (0). Students who scored a total of ⩾3 across both items were classified as having possible anxiety symptoms.

Statistical analysis

A data access proposal including the research question, selected variables, analysis plan and consideration for data sovereignty principles was approved as per Youth2000 policy (www.youth19.ac.nz/the-surveys). All analyses were completed using R 4.1.1 (R Core Team, 2021) and the survey package (v 1.1.1; Ellis and Schneider, 2022). Data were cleaned and weighted by the inverse probability of selection, and the variance of estimates was adjusted for a two-stage clustered and stratified survey design and calibrated for student age, sex and ethnicity to enable estimates representative of the national secondary school population (Rivera-Rodriguez et al., 2021). Missing data for individual items were not imputed.

Main analysis

We present percentages with 95% confidence intervals (CIs) for the prevalence of each outcome (overall and by sex, age, ethnicity and neighbourhood deprivation) in 2019 and other years where available. Mental health trends over earlier survey waves have been explored previously (Fleming et al., 2014), thus we explored changes between 2012 and 2019 using logistic regression. Adjusted odds ratios (ORs) are presented, showing the odds of each outcome in 2019 relative to 2012, adjusted for sex, age, ethnicity, neighbourhood deprivation and locale. A conservative p-value of <0.01 indicates statistically significant differences between waves.

Secondary analysis

To explore whether observed changes in national estimates were due to differences in the sampling frame between 2012 (national sample) and 2019 (regional sample), we ran a secondary analysis comparing mental health outcomes over time in the Auckland, Tai Tokerau and Waikato regions only (i.e. with no changes in the included region over the survey waves).

Results

In total, 34,548 secondary school students have participated in a Youth2000 survey, representing 2.6–3.9% of the New Zealand secondary school student population (see Table 1). Participants’ demographic characteristics are shown in Table 3.

Table 3.

Participant characteristics by survey year, 2001–2019.

	2001		2007		2012		2019
	N	%	N	%	N	%	N	%
Total schools	114	–	96	–	91	–	45	–
Total students	9546	100.0	9098	100.0	8487	100.0	7311	100.0
Sex^a
Female	5144	53.9	4187	46.0	4618	54.4	3990	54.6
Male	4402	46.1	4911	54.0	3869	45.6	3321	45.4
Age (years)^b
13 or under	2050	21.5	1859	20.4	1838	21.7	1323	18.1
14	2285	23.9	2100	23.1	1895	22.3	1635	22.4
15	2178	22.8	1973	21.7	1755	20.7	1620	22.2
16	1725	18.1	1743	19.2	1578	18.6	1403	19.2
17 or over	1308	13.7	1423	15.6	1421	16.7	1330	18.2
Ethnicity^c
Māori	2335	24.5	1702	18.7	1697	20.0	1188	16.2
Pacific	765	8.0	924	10.2	1200	14.1	915	12.5
Asian	679	7.1	1126	12.4	1049	12.4	1768	24.2
Other^d	365	3.8	549	6.0	523	6.2	387	5.3
Pākehā and other European	5402	56.6	4797	52.7	4018	47.3	3053	41.8
Neighbourhood deprivation^e,f
Low deprivation	–	–	3216	35.3	2715	32.0	2098	28.7
Medium deprivation	–	–	3396	37.3	2995	35.3	2772	37.9
High deprivation	–	–	2246	24.7	2670	31.5	1829	25.0
Locale^f,g
Urban	–	–	6688	73.5	6591	77.7	5085	69.6
Small towns	–	–	726	8.0	517	6.1	520	7.1
Rural	–	–	1390	15.3	1216	14.3	1095	15.0

2019 was the first Youth2000 survey in which students could report a sex other than male or female. Sixty-three students did not report a male or female sex in 2019. Their results will be reported in other outputs where contexts can be fully explored (see text for further detail).

Very few participants were younger than 13 or older than 17. These students are included in 13 or under or 17 or over to ensure adequate group sizes.

Ethnicity categorised using the NZ census ethnicity prioritisation method (Ministry of Health, 2017).

‘Other’ ethnicity includes Middle Eastern, Latin American, African, other ethnicities and ethnicity unknown.

Neighbourhood deprivation is categorised using the NZ Deprivation Index 2018 (NZDep2018; Atkinson et al., 2019). Low deprivation = deciles 1–3, medium = deciles 4–7, high = deciles 8–10.

Meshblock information was not collected in 2001; thus, NZDep and locale are not available.

Urban (10,000 people or more), small towns (between 1000 and 9999 people) and rural (fewer than 1000 people), derived from the statistical standard for geographic areas 2018 (Statistics New Zealand, 2017).

Mental health indicators

The prevalence of mental health and well-being indicators for males and females in each survey wave is shown in Figure 1. The prevalence of variables in each survey wave and changes between 2012 and 2019 by sex, age, ethnicity and neighbourhood deprivation are shown in Tables 4 –6. Key results are described below.

Figure 1.

Prevalence of mental health indicators 2001, 2007, 2012 and 2019, by sex.

Table 4.

Mental health indicators by sex^a and age (2001–2019), with change between 2012 and 2019.

	2001		2007		2012		2019^b		Change between 2012 and 2019
	n/N	% [95% CI]	n/N	% [95% CI]	n/N	% [95% CI]	n/N	% [95% CI]	OR^c [95% CI]	p value
Good overall well-being^d
Total	NA		6804 (8679)	78.2 [77.4, 79.0]	6306 (8274)	76.0 [75.0, 77.0]	4835 (7004)	69.1 [67.6, 70.6]	0.71 [0.65, 0.78] ↓	<0.001
Female			2887 (4023)	72.2 [70.9, 73.4]	3233 (4533)	70.0 [68.3, 71.6]	2331 (3845)	60.1 [58.4, 61.9]	0.66 [0.59, 0.74] ↓	<0.001
Male			3917 (4656)	84.3 [83.1, 85.4]	3073 (3741)	82.0 [81.1, 83.0]	2504 (3159)	78.1 [76.2, 80.0]	0.79 [0.70, 0.90] ↓	<0.001
⩽15			4496 (5643)	79.2 [78.3, 80.2]	4153 (5342)	77.3 [76.2, 78.5]	3128 (4342)	72.2 [70.5, 73.8]	0.77 [0.69, 0.85] ↓	<0.001
⩾16			2308 (3036)	76.5 [74.8, 78.3]	2153 (2932)	74.1 [72.4, 75.8]	1707 (2662)	65.1 [62.7, 67.4]	0.65 [0.57, 0.75] ↓	<0.001
Significant depression symptoms^e
Total	1154 (9179)	11.8 [11.1, 12.5]	981 (8550)	11.4 [10.8, 12.0]	1078 (8170)	13.0 [12.2, 13.9]	1727 (7014)	22.8 [21.4, 24.1]	1.96 [1.75, 2.20] ↑	<0.001
Female	782 (4983)	14.7 [13.5, 15.8]	643 (3972)	15.6 [14.6, 16.6]	763 (4499)	17.4 [16.1, 18.6]	1218 (3844)	29.0 [27.2, 30.9]	1.94 [1.71, 2.19] ↑	<0.001
Male	372 (4196)	8.8 [7.9, 9.7]	338 (4578)	7.2 [6.6, 7.9]	315 (3671)	8.7 [7.6, 9.8]	509 (3170)	16.4 [14.9, 17.9]	2.04 [1.72, 2.43] ↑	<0.001
⩽15	805 (6240)	12.2 [11.3, 13.0]	657 (5560)	12.1 [11.3, 12.9]	667 (5275)	12.5 [11.4, 13.6]	1053 (4368)	22.5 [21.0, 24.1]	2.03 [1.77, 2.32] ↑	<0.001
⩾16	349 (2939)	11.1 [10.0, 12.3]	324 (2990)	10.3 [9.2, 11.3]	411 (2895)	13.8 [12.6, 15.0]	674 (2646)	23.0 [21.3, 24.8]	1.88 [1.61, 2.20] ↑	<0.001
Period of low mood (past 12 months)^f
Total	NA		2368 (8764)	27.3 [26.4, 28.2]	2599 (8368)	30.8 [29.7, 31.9]	2753 (7121)	38.3 [36.6, 40.1]	1.40 [1.27, 1.53] ↑	<0.001
Female			1430 (4061)	34.7 [33.3, 36.0]	1718 (4575)	38.4 [37.0, 39.9]	1780 (3913)	44.1 [42.0, 46.2]	1.26 [1.13, 1.40] ↑	<0.001
Male			938 (4703)	19.9 [18.6, 21.1]	881 (3793)	23.3 [21.7, 24.9]	973 (3208)	32.5 [30.0, 35.1]	1.59 [1.36, 1.84] ↑	<0.001
⩽15			1509 (5714)	26.9 [25.8, 28.0]	1554 (5416)	28.4 [27.2, 29.5]	1577 (4437)	35.2 [33.3, 37.0]	1.37 [1.24, 1.51] ↑	<0.001
⩾16			859 (3050)	27.9 [26.3, 29.5]	1045 (2952)	34.4 [32.8, 36.1]	1176 (2684)	42.5 [40.0, 45.1]	1.43 [1.25, 1.64] ↑	<0.001
Self-harm (past 12 months)^g
Total	NA		1785 (8752)	20.8 [20.0, 21.7]	2009 (8367)	23.3 [22.3, 24.3]	1711 (7058)	24.1 [22.8, 25.4]	1.03 [0.94, 1.13] −	0.55
Female			1060 (4054)	25.9 [24.5, 27.3]	1330 (4572)	29.3 [27.7, 30.9]	1175 (3876)	30.2 [28.9, 31.5]	1.04 [0.95, 1.14] −	0.43
Male			725 (4698)	15.7 [14.7, 16.8]	679 (3795)	17.4 [16.1, 18.8]	536 (3182)	17.8 [16.0, 19.7]	1.03 [0.88, 1.19] −	0.75
⩽15			1225 (5704)	22.4 [21.3, 23.5]	1314 (5409)	23.8 [22.3, 25.2]	1086 (4383)	24.9 [23.2, 26.6]	1.04 [0.91, 1.17] −	0.58
⩾16			560 (3048)	18.2 [17.0, 19.5]	695 (2958)	22.7 [21.3, 24.0]	625 (2675)	23.0 [21.5, 24.6]	1.01 [0.90, 1.14] −	0.81
Thoughts of suicide (past 12 months)^h
Total	NA		1231 (8719)	14.5 [13.7, 15.2]	1307 (8314)	15.3 [14.5, 16.2]	1501 (7038)	20.8 [19.2, 22.4]	1.41 [1.25, 1.59] ↑	<0.001
Female			795 (4051)	19.4 [18.2, 20.6]	937 (4550)	20.7 [19.2, 22.3]	1002 (3867)	24.6 [22.8, 26.4]	1.23 [1.08, 1.39] ↑	<0.001
Male			436 (4668)	9.5 [8.6, 10.4]	370 (3764)	10.0 [9.0, 11.0]	499 (3171)	16.9 [14.8, 19.1]	1.77 [1.46, 2.16] ↑	<0.001
⩽15			807 (5684)	14.8 [13.9, 15.7]	822 (5373)	14.5 [13.4, 15.7]	889 (4376)	20.0 [18.1, 21.9]	1.42 [1.22, 1.66] ↑	<0.001
⩾16			424 (3035)	13.9 [12.7, 15.1]	485 (2941)	16.5 [15.3, 17.7]	612 (2662)	21.8 [20.0, 23.6]	1.40 [1.21, 1.61] ↑	<0.001
Suicide attempts (past 12 months)ⁱ
Total	NA		413 (8715)	4.8 [4.3, 5.2]	377 (8313)	3.9 [3.4, 4.3]	439 (7048)	6.3 [5.5, 7.0]	1.60 [1.32, 1.92] ↑	<0.001
Female			275 (4049)	6.6 [5.9, 7.3]	284 (4549)	5.5 [4.8, 6.3]	310 (3868)	7.4 [6.2, 8.6]	1.33 (1.12, 1.60] ↑	<0.001
Male			138 (4666)	2.9 [2.4, 3.5]	93 (3764)	2.2 [1.6, 2.8]	129 (3180)	5.1 [3.7, 6.5]	2.27 [1.53, 3.35] ↑	<0.001
⩽15			301 (5681)	5.6 [5.0, 6.2]	254 (5372)	4.0 [3.4, 4.5]	291 (4383)	6.9 [5.8, 8.0]	1.73 [1.39, 2.15] ↑	<0.001
⩾16			112 (3034)	3.4 [2.9, 4.0]	123 (2941)	3.7 [3.0, 4.4]	148 (2665)	5.5 [4.6, 6.4]	1.39 [1.05, 1.82] ↑	0.02

CI: confidence interval; OR: odds ratio.

2019 was the first Youth2000 survey in which students could report a sex other than male or female. Sixty-three students did not report a male or female sex in 2019. Their health and well-being will be reported in other outputs where contexts can be fully explored (see text for further detail).

2019 data excludes participants attending Kura Kaupapa Māori to allow comparisons to previous survey waves.

Adjusted for sex, age, ethnicity, deprivation and locale.

Good overall well-being defined as a score of 13 or more on the World Health Organization 5-item (WHO-5).

Depression symptoms measured using the Reynolds Adolescent Depression Scale (RADS) in 2001 or the Reynolds Adolescent Depression Scale – Short Form (RADS-SF) from 2007 (cut off: 28+).

Period of low mood based on the question: During the past 12 months, was there ever a time when you felt sad, blue or depressed for 2 weeks or more in a row? (Yes/No).

Self-harm based on the question: During the last 12 months, have you deliberately hurt yourself or done anything you knew might harm you (but not kill you)? (No, never/Yes – once or twice/Yes – three or more times). Either affirmative response was classified as Yes.

Thoughts of suicide based on the question: During the last 12 months, have you seriously thought about killing yourself (attempting suicide)? (Yes/No). In 2001, the question did not include the word ‘seriously’; hence, 2001 data are not included.

Suicide attempts based on the question: During the last 12 months, have you tried to kill yourself (attempted suicide)? (Yes/No). In 2001, only students who reported suicide thoughts in the past 12 months received this question. From 2007, all students received the question. Hence, 2001 data are not included.

Table 5.

Mental health indicators by ethnicity^a (2001–2019), with change between 2012 and 2019.

	2001		2007		2012		2019^b		Change between 2012 and 2019
	n (N)	% [95% CI]	n (N)	% [95% CI]	n (N)	% [95% CI]	n (N)	% [95% CI]	OR^c (95% CI)	p value
Good overall well-being^d
Total		NA	6804 (8679)	78.2 [77.4, 79.0]	6306 (8274)	76.0 [75.0, 77.0]	4835 (7004)	69.1 [67.6, 70.6]	0.71 [0.65, 0.78] ↓	<0.001
Māori			1259 (1614)	77.6 [75.7, 79.4]	1233 (1640)	75.3 [73.2, 77.4]	733 (1103)	67.1 [64.3, 70.0]	0.66 [0.55, 0.79] ↓	<0.001
Pacific			701 (850)	82.2 [79.5, 84.9]	908 (1140)	79.0 [75.8, 82.3]	622 (837)	72.7 [67.3, 78.0]	0.73 [0.57, 0.95] ↓	0.02
Asian			816 (1086)	75.0 [72.3, 77.6]	764 (1021)	76.1 [72.8, 79.3]	1173 (1723)	67.7 [64.5, 70.9]	0.68 [0.55, 0.84] ↓	<0.001
Other			377 (515)	72.5 [68.8, 76.1]	348 (501)	68.8 [64.4, 73.3]	232 (353)	69.9 [64.7, 75.2]	0.98 [0.71, 1.36] −	0.93
Pākehā and other European			3651 (4614)	78.8 [77.7, 80.0]	3053 (3972)	76.4 [75.1, 77.7]	2075 (2988)	69.4 [67.4, 71.4]	0.70 [0.62, 0.79] ↓	<0.001
Significant depression symptoms^e
Total	1154 (9179)	11.8 [11.1, 12.5]	981 (8550)	11.4 [10.8, 12.0]	1078 (8170)	13.0 [12.2, 13.9]	1727 (7014)	22.8 [21.4, 24.1]	1.96 [1.75, 2.20] ↑	<0.001
Māori	345 (2229)	14.9 [13.3, 16.5]	180 (1584)	11.3 [10.0, 12.7]	227 (1608)	13.8 [11.4, 16.3]	336 (1107)	27.9 [25.4, 30.4]	2.56 [2.01, 3.26] ↑	<0.001
Pacific	129 (710)	18.1 [14.9, 21.2]	90 (815)	11.4 [9.5, 13.3]	150 (1113)	13.7 [11.1, 16.2]	217 (841)	24.6 [20.7, 28.5]	1.92 [1.46, 2.53] ↑	<0.001
Asian	91 (646)	13.9 [11.6, 16.2]	158 (1079)	14.6 [12.7, 16.6]	131 (1017)	12.5 [10.7, 14.2]	439 (1723)	24.6 [22.2, 27.0]	2.24 [1.83, 2.74] ↑	<0.001
Other	41 (295)	12.7 [9.1, 16.3]	92 (509)	19.0 [15.8, 22.2]	84 (498)	16.5 [13.2, 19.9]	98 (355)	27.0 [22.4, 31.7]	1.93 [1.34, 2.77] ↑	<0.001
Pākehā and other European	548 (5299)	10.1 [9.2, 11.0]	461 (4563)	10.4 [9.6, 11.2]	486 (3934)	12.4 [11.4, 13.5]	637 (2988)	19.6 [18.1, 21.2]	1.75 [1.51, 2.04] ↑	<0.001
Period of low mood (past 12 months)^f
Total	NA		2368 (8764)	27.3 [26.4, 28.2]	2599 (8368)	30.8 [29.7, 31.9]	2753 (7121)	38.3 [36.6, 40.1]	1.40 [1.27, 1.53] ↑	<0.001
Māori			491 (1640)	30.3 [28.0, 32.7]	560 (1665)	33.5 [31.2, 35.8]	487 (1124)	42.2 [38.2, 46.2]	1.51 [1.27, 1.81] ↑	<0.001
Pacific			239 (857)	28.3 [25.7, 30.9]	398 (1169)	34.4 [31.2, 37.6]	336 (859)	41.1 [36.7, 45.4]	1.31 [1.09, 1.58] ↑	0.01
Asian			313 (1098)	28.8 [26.0, 31.7]	302 (1036)	28.7 [25.4, 32.0]	679 (1748)	38.3 [35.6, 40.9]	1.52 [1.25, 1.86] ↑	<0.001
Other			182 (515)	35.9 [32.4, 39.4]	186 (505)	36.1 [32.0, 40.1]	152 (369)	38.3 [32.6, 44.0]	1.11 [0.82, 1.51] −	0.50
Pākehā and other European			1143 (4654)	25.3 [24.0, 26.6]	1153 (3993)	29.2 [27.8, 30.7]	1099 (3021)	36.4 [34.8, 38.0]	1.41 [1.26, 1.57] ↑	<0.001
Self-harm (past 12 months)^g
Total	NA		1785 (8752)	20.8 [20.0, 21.7]	2009 (8367)	23.3 [22.3, 24.3]	1711 (7058)	24.1 [22.8, 25.4]	1.03 [0.94, 1.13] −	0.55
Māori			414 (1639)	25.3 [23.3, 27.2]	478 (1664)	27.3 [24.5, 30.1]	357 (1111)	32.0 [29.2, 34.8]	1.26 [1.04, 1.52] ↑	0.02
Pacific			175 (854)	21.4 [18.9, 24.0]	305 (1166)	26.1 [23.4, 28.7]	196 (850)	24.2 [20.5, 28.0]	0.87 [0.71, 1.06] −	0.17
Asian			172 (1094)	15.7 [13.4, 18.1]	211 (1036)	19.1 [16.9, 21.3]	363 (1733)	18.5 [16.6, 20.4]	0.95 [0.79, 1.15] −	0.63
Other			120 (520)	24.2 [20.7, 27.7]	140 (509)	26.9 [23.2, 30.6]	81 (361)	22.1 [19.2, 25.0]	0.70 [0.50, 0.98] ↓	0.04
Pākehā and other European			904 (4645)	19.9 [18.8, 21.0]	875 (3992)	21.9 [20.6, 23.2]	714 (3003)	22.5 [20.6, 24.4]	1.04 [0.90, 1.19] −	0.59
Thoughts of suicide (past 12 months)^h
Total	NA		1231 (8719)	14.5 [13.7, 15.2]	1307 (8314)	15.3 [14.5, 16.2]	1501 (7038)	20.8 [19.2, 22.4]	1.41 [1.25, 1.59] ↑	<0.001
Māori			285 (1628)	17.9 [15.7, 20.1]	307 (1656)	17.8 [15.7, 19.8]	284 (1109)	24.8 [21.8, 27.7]	1.49 [1.21, 1.83] ↑	<0.001
Pacific			134 (844)	16.7 [13.7, 19.8]	215 (1151)	18.0 [15.2, 20.8]	209 (848)	25.9 [22.0, 29.8]	1.50 [1.18, 1.89] ↑	<0.001
Asian			156 (1095)	14.3 [12.4, 16.2]	136 (1030)	12.4 [10.3, 14.5]	360 (1727)	19.7 [16.8, 22.7]	1.68 [1.31, 2.16] ↑	<0.001
Other			74 (518)	15.1 [12.3, 17.9]	88 (500)	17.3 [13.2, 21.3]	89 (361)	25.1 [20.8, 29.3]	##	##
Pākehā and other European			582 (4634)	13.1 [12.2, 14.0]	561 (3977)	14.4 [13.4, 15.5]	559 (2993)	18.1 [16.4, 19.9]	1.31 [1.13, 1.52] ↑	<0.001
Suicide attempts (past 12 months)ⁱ
Total	NA		413 (8715)	4.8 [4.3, 5.2]	377 (8313)	3.9 [3.4, 4.3]	439 (7048)	6.3 [5.5, 7.0]	1.60 [1.32, 1.92] ↑	<0.001
Māori			114 (1628)	7.2 [6.1, 8.4]	108 (1656)	6.2 [5.0, 7.4]	138 (1105)	12.8 [10.5, 15.1]	2.17 [1.64, 2.88] ↑	<0.001
Pacific			69 (842)	8.5 [6.9, 10.2]	95 (1151)	7.4 [5.8, 8.9]	99 (855)	11.7 [9.9, 13.5]	1.68 [1.30, 2.16] ↑	<0.001
Asian			35 (1092)	3.2 [2.3, 4.1]	38 (1030)	3.2 [2.1, 4.4]	79 (1729)	4.4 [2.6, 6.2]	1.27 [0.72, 2.26] −	0.41
Other			23 (518)	4.5 [3.1, 5.9]	29 (500)	5.5 [2.9, 8.0]	24 (361)	6.5 [5.0, 7.9]	1.20 [0.68, 2.13] −	0.53
Pākehā and other European			172 (4635)	3.8 [3.3, 4.3]	107 (3976)	2.4 [2.0, 2.9]	99 (2998)	3.2 [2.3, 4.2]	1.31 [0.90, 1.92] −	0.16

CI: confidence interval; OR: odds ratio.

Ethnicity categorised using the NZ census ethnicity prioritisation method (Ministry of Health, 2017).

2019 data excludes participants attending Kura Kaupapa Māori and those who did not report male or female sex to allow comparisons to previous survey waves.

Adjusted for sex, age, deprivation and locale.

Good overall well-being defined as a score of 13 or more on the World Health Organization 5-item (WHO-5).

Depression symptoms measured using the Reynolds Adolescent Depression Scale (RADS) in 2001 or the Reynolds Adolescent Depression Scale – Short Form (RADS-SF) from 2007 (cut off: 28 +).

Period of low mood based on the question: During the past 12 months, was there ever a time when you felt sad, blue or depressed for 2 weeks or more in a row? (Yes/No).

insufficient data to estimate.

Table 6.

Mental health indicators by neighbourhood deprivation^a (2001–2019), with change between 2012 and 2019.

	2001^b		2007		2012		2019^c		Change between 2012 and 2019
	n (N)	% [95% CI]	n (N)	% [95% CI]	n (N)	% [95% CI]	n (N)	% [95% CI]	OR^d [95% CI]	p value
Good overall well-being^e
Total	NA		6804 (8679)	78.2 [77.4, 79.0]	6306 (8274)	76.0 [75.0, 77.0]	4835 (7004)	69.1 [67.6, 70.6]	0.71 [0.65, 0.78] ↓	<0.001
Low deprivation			2509 (3153)	79.8 [78.6, 81.0]	2064 (2683)	77.2 [75.6, 78.8]	1458 (2048)	70.7 [68.3, 73.0]	0.72 [0.62, 0.83] ↓	<0.001
Medium deprivation			2558 (3286)	77.8 [76.4, 79.2]	2226 (2938)	75.9 [74.5, 77.2]	1845 (2683)	69.8 [67.6, 72.0]	0.73 [0.64, 0.83] ↓	<0.001
High deprivation			1664 (2146)	76.5 [74.8, 78.2]	1948 (2563)	74.7 [72.8, 76.6]	1155 (1699)	67.2 [63.9, 70.4]	0.66 [0.56, 0.79] ↓	<0.001
Significant depression symptoms^f
Total	1154 (9179)	11.8 [11.1, 12.5]	981 (8550)	11.4 [10.8, 12.0]	1078 (8170)	13.0 [12.2, 13.9]	1727 (7014)	22.8 [21.4, 24.1]	1.96 [1.75, 2.20] ↑	<0.001
Low deprivation	NA		315 (3126)	10.0 [8.8, 11.3]	326 (2669)	11.9 [10.4, 13.3]	407 (2045)	17.0 [15.8, 18.1]	1.49 [1.27, 1.74] ↑	<0.001
Medium deprivation			377 (3257)	11.5 [10.4, 12.5]	394 (2907)	13.2 [11.9, 14.6]	668 (2687)	23.1 [20.9, 25.2]	1.97 [1.66, 2.34] ↑	<0.001
High deprivation			276 (2083)	13.5 [11.9, 15.0]	342 (2506)	14.2 [12.5, 15.9]	504 (1709)	29.3 [25.6, 33.1]	2.80 [2.27, 3.45] ↑	<0.001
Period of low mood (past 12 months)^g
Total	NA		2368 (8764)	27.3 [26.4, 28.2]	2599 (8368)	30.8 [29.7, 31.9]	2753 (7121)	38.3 [36.6, 40.1]	1.40 [1.27, 1.53] ↑	<0.001
Low deprivation			822 (3176)	26.3 [24.5, 28.1]	809 (2699)	29.7 [28.0, 31.5]	703 (2074)	32.7 [31.2, 34.3]	1.12 [0.99, 1.27] −	0.06
Medium deprivation			907 (3325)	27.4 [26.0, 28.9]	903 (2969)	29.9 [27.7, 32.2]	1087 (2726)	39.2 [36.6, 41.8]	1.52 [1.31, 1.76] ↑	<0.001
High deprivation			616 (2171)	28.7 [27.1, 30.2]	849 (2604)	33.1 [31.2, 35.1]	730 (1736)	44.3 [40.7, 47.9]	1.67 [1.43, 1.95] ↑	<0.001
Self-harm (past 12 months)^h
Total	NA		1785 (8752)	20.8 [20.0, 21.7]	2009 (8367)	23.3 [22.3, 24.3]	1711 (7058)	24.1 [22.8, 25.4]	1.03 [0.94, 1.13] −	0.55
Low deprivation			584 (3178)	18.8 [17.5, 20.2]	584 (2697)	20.8 [19.2, 22.4]	457 (2061)	22.2 [20.7, 23.8]	1.06 [0.93, 1.21] −	0.35
Medium deprivation			703 (3313)	21.5 [20.2, 22.9]	711 (2970)	23.1 [21.5, 24.8]	646 (2710)	22.3 [20.7, 23.9]	0.96 [0.84, 1.09] −	0.50
High deprivation			483 (2173)	23.0 [21.4, 24.6]	685 (2605)	26.7 [24.6, 28.8]	458 (1712)	28.3 [24.6, 31.9]	1.12 [0.92, 1.38] −	0.26
Thoughts of suicide (past 12 months)ⁱ
Total	NA		1231 (8719)	14.5 [13.7, 15.2]	1307 (8314)	15.3 [14.5, 16.2]	1501 (7038)	20.8 [19.2, 22.4]	1.41 [1.25, 1.59] ↑	<0.001
Low deprivation			403 (3170)	12.9 [12.0, 13.9]	371 (2690)	13.5 [12.5, 14.5]	351 (2053)	16.1 [14.4, 17.8]	1.22 [1.04, 1.42] ↑	0.01
Medium deprivation			460 (3304)	14.1 [13.0, 15.3]	479 (2952)	15.8 [14.0, 17.7]	576 (2699)	20.8 [19.2, 22.4]	1.39 [1.18, 1.64] ↑	<0.001
High deprivation			351 (2159)	17.1 [15.3, 18.9]	445 (2577)	17.2 [15.5, 18.9]	436 (1707)	25.5 [21.7, 29.4]	1.74 [1.42, 2.14] ↑	<0.001
Suicide attempts (past 12 months)^j
Total	NA		413 (8715)	4.8 [4.3, 5.2]	377 (8313)	3.9 [3.4, 4.3]	439 (7048)	6.3 [5.5, 7.0]	1.60 [1.32, 1.92] ↑	<0.001
Low deprivation			111 (3169)	3.4 [2.8, 4.0]	87 (2689)	2.7 [2.1, 3.2]	62 (2053)	2.7 [2.1, 3.3]	1.00 [0.72, 1.39] −	1.00
Medium deprivation			158 (3304)	4.9 [4.2, 5.7]	112 (2952)	3.4 [2.6, 4.1]	147 (2697)	6.0 [5.0, 6.9]	1.76 [1.32, 2.35] ↑	<0.001
High deprivation			137 (2156)	6.5 [5.5, 7.5]	175 (2577)	6.3 [5.1, 7.4]	187 (1717)	10.7 [8.5, 13.0]	1.87 [1.44, 2.45] ↑	<0.001

CI: confidence interval; OR: odds ratio.

Neighbourhood deprivation categorised using the NZ Deprivation Index 2018 (NZDep2018; Atkinson et al., 2019). Low deprivation = deciles 1–3, medium = deciles 4–7, high = deciles 8–10.

Meshblock information was not collected in 2001, thus NZDep is not available.

2019 data excludes participants attending Kura Kaupapa Māori and those who did not report a male or female sex to allow comparisons to previous survey waves.

Adjusted for sex, age, ethnicity and locale.

Good overall well-being defined as a score of 13 or more on the World Health Organization 5-item (WHO-5).

Depression symptoms measured using the Reynolds Adolescent Depression Scale (RADS) in 2001 or the Reynolds Adolescent Depression Scale – Short Form (RADS-SF) from 2007 (cut off: 28+).

Period of low mood based on the question: During the past 12 months, was there ever a time when you felt sad, blue or depressed for 2 weeks or more in a row? (Yes/No).

Well-being

In 2019, 69.1% of students scored above the cut off for good well-being, which was more prevalent among males (78%, 95% CI = [76.2, 80.0]) than females (60%, 95% CI = [58.4, 61.9]) and among younger students (72%, 95% CI = [70.5, 73.8]) than older students (65%, 95% CI = [62.7, 67.4]) (Table 4). Differences in well-being among ethnic groups and neighbourhood deprivation bands were small, with overlapping 95% CIs (Tables 5 and 6).

The overall prevalence of good well-being declined substantially from 76% in 2012 to 69% in 2019 (OR 0.7), with declines similar across demographic groups, including sex (ORs for most groups, 0.7–0.8).

Depression symptoms

In 2019, 23% of students reported significant symptoms of depression, with marked demographic differences. Prevalence was higher among females (29%) than males (16%); higher among Māori (28%) and Other ethnic groups (27%) than European (20%; Pacific and Asian both 25%); and higher in high-deprivation neighbourhoods (29%; medium-deprivation 23%; low-deprivation 17%).

The prevalence of significant depression symptoms increased overall from 13% in 2012 to 23% in 2019 (OR = 2.0), with similar relative increases among males (9–16%, OR = 2.0), females (17–29%, OR = 1.9), older students (14–23%, OR = 1.9) and younger students (13–23%, OR = 2.0). The increase was statistically significant for all ethnic groups and deprivation bands, and particularly marked among Māori students (14–28%, OR = 2.6) and Asian students (13–25%, OR = 2.2) and students in high-deprivation neighbourhoods (14–29%, OR = 2.8).

Period of low mood

In 2019, 38% of students reported periods of low mood of 2 weeks or more, within the past year. Demographic differences were marked, with prevalence higher among females (44%) than males (33%) and higher among older (43%) than younger students (35%); higher among Māori (42%) and Pacific students (41%) than European (36%); and higher in both high-deprivation (44%) and medium-deprivation (39%) neighbourhoods than low-deprivation neighbourhoods (33%).

Period of low mood was more prevalent in 2019 than that in 2012 for almost all groups, with marked increases for males (23–33%, OR = 1.6; females 38–44% OR = 1.3), increases for all ethnic groups except Other, particularly among Asian students (29–38%, OR = 1.5), and increases in all deprivation bands, particularly high-deprivation (33–44%, OR = 1.7).

Self-harm

In 2019, 24% of students reported deliberate self-harm. Prevalence was particularly high among females (30%; males 18%), Māori (32%; Asian 19%, Other 22%, European 23%, Pacific 24%) and those in high-deprivation neighbourhoods (28%; medium- and low-deprivation both 22%). The prevalence of self-harm increased among Māori students between 2012 and 2019 (27–32%, OR = 1.3). Changes for other groups were not statistically significant.

Suicide thoughts

In 2019, 21% of students reported serious thoughts of suicide in the past year, with prevalence higher among females (25%) than males (17%); higher among Māori (25%), Other (25%) and Pacific students (26%) than European students (18%); and higher among those in high- (26%) and medium-deprivation neighbourhoods (21%) than those in low-deprivation neighbourhoods (16%).

Suicide thoughts were more prevalent for most demographic groups in 2019 than in 2012, with the odds of suicide thoughts increasing, particularly in males (10–17%, OR = 1.8; females 21–25%, OR = 1.2), increasing in both age groups (younger 15–20%, older 17–22%, both OR = 1.4) and across ethnic groups. There were large increases for Asian (12–20%, OR = 1.7), Pacific (18–26%, OR = 1.5) and Māori students (18–25%, OR = 1.5; European 14–18%, OR = 1.3) and those in high-deprivation neighbourhoods (17–26%, OR = 1.7; medium-deprivation 16–21%, OR = 1.4; low-deprivation 14–16%, OR = 1.2).

Suicide attempts

In 2019, 6.3% of students reported past-year suicide attempts, with prevalence higher among female students (7.4%; males 5.1%), Māori (12.8%) and Pacific students (11.7%; Other 6.5%, Asian 4.4%, European 3.2%) and those in high-deprivation neighbourhoods (10.7%; medium-deprivation 6%; low-deprivation 2.7%).

Suicide attempts were more prevalent in 2019 than 2012, with particularly marked increases among males (2.2–5.1%, OR = 2.3; females 5.5–7.4%, OR = 1.3), younger students (4.0–6.9%, OR = 1.7; older students 3.7–5.5%, OR = 1.4) and Māori (6.2–12.8%, OR = 2.2) and Pacific students (7.4–11.7%, OR = 1.7; changes in other ethnic groups were not statistically significant) and among those in high- (6.3–10.7%, OR = 1.9) and medium-deprivation (3.4–6.0%, OR = 1.8) neighbourhoods (no change observed in the low-deprivation group).

Percentage changes in mental health and well-being indicators between 2012 and 2019 are presented in Figures 2 –4. The ORs described above and shown in Tables 4 –6 highlight changes in the odds of indicators for each group.

Figure 2.

Prevalence change on mental health indicators 2012–2019, by age and sex.

Figure 3.

Prevalence change on mental health indicators 2012–2019, by ethnicity.

Figure 4.

Prevalence change on mental health indicators 2012–2019, by neighbourhood deprivation.

Possible anxiety symptoms were measured in 2019 only (Table 7). Overall, 42% of students scored at or above the cut off on this measure. Prevalence was particularly high among females (49%; males 35%), Māori (48%) and Pacific students (50%; European 38%), and those in high-deprivation neighbourhoods (50%; medium-deprivation 40%, low-deprivation 37%).

Table 7.

Possible anxiety symptoms^a 2019, by selected demographic factors.

	Possible anxiety symptoms
	n (N)	% [95% CI]
Total	3001 (6987)	41.8 [40.5, 43.2]
Sex^b
Female	1936 (3846)	48.8 [46.8, 50.7]
Male	1065 (3141)	34.8 [32.7, 36.9]
Age
⩽15	1786 (4321)	40.8 [39.1, 42.5]
⩾16	1215 (2666)	43.1 [40.9, 45.3]
Ethnicity^c
Māori	503 (1091)	47.8 [43.7, 51.8]
Pacific	412 (842)	49.8 [47.3, 52.2]
Asian	721 (1717)	40.9 [37.6, 44.1]
Other^d	163 (356)	42.4 [38.8, 46.1]
European	1202 (2981)	38.4 [36.6, 40.1]
Neighbourhood deprivation^e
High deprivation	776 (2042)	36.5 [34.4, 38.6]
Medium deprivation	1128 (2683)	40.7 [38.5, 42.8]
Low deprivation	823 (1687)	49.9 [45.6, 54.2]
Locale^f
Rural	413 (1048)	37.3 [33.8, 40.8]
Small town	211 (488)	44.5 [36.9, 52]
Urban	2104 (4877)	42.4 [40.9, 43.9]

CI: confidence interval.

Possible anxiety symptoms measured using GAD-2, adapted (see text for further details).

Ethnicity categorised using the NZ census ethnicity prioritisation method (Ministry of Health, 2017).

‘Other’ ethnicity includes Middle Eastern, Latin American, African and other ethnicities and ethnicity unknown.

Neighbourhood deprivation is categorised using the NZ Deprivation Index 2018 (NZDep2018; Atkinson et al., 2019). Low deprivation = deciles 1–3, medium = deciles 4–7, high = deciles 8–10.

Urban (population of 10,000 or more), small towns (population between 1000 and 9999 people), rural (population fewer than 1000), derived from the statistical standard for geographic areas 2018 (Statistics New Zealand, 2017).

Secondary analysis

As shown in Table 8, the secondary analysis (reporting findings from the Auckland, Tai Tokerau and Waikato regions only) also identified large declines in mental health over the 2012 to 2019 period. Adjusted odds ratios did not differ substantially from those produced in the original analysis. For example, in the regional analysis, the OR describing changes in well-being between 2012 and 2019 for the total population was 0.70 (compared to 0.71 for the primary analysis), the OR for depression symptoms was 1.97 (1.96 for the primary analysis) and the OR for suicide attempts was 1.67 (1.60 for the primary analysis). These findings suggest that the increase in distress observed in the primary analysis is not an artefact of methodological differences between 2012 and 2019.

Table 8.

Mental health indicators for total Auckland, Tai Tokerau and Waikato sample (2001–2019), with change between 2012 and 2019.

	2001		2007		2012		2019^a		Change between 2012 and 2019
	n/N	% [95% CI]	n/N	% [95% CI]	n/N	% [95% CI]	n/N	% [95% CI]	OR^b [95% CI]	p value
Good overall well-being^c	NA		3451 (4432)	77.7 [76.6, 78.9]	3446 (4489)	76.5 [74.9, 78.1]	4835 (7004)	69.1 [67.6, 70.6]	0.70 [0.63, 0.79] ↓	<0.001
Significant depression symptoms^d	468 (3388)	13 [11.6, 14.3]	513 (4375)	11.6 [10.6, 12.7]	577 (4433)	12.9 [11.4, 14.5]	1727 (7014)	22.8 [21.4, 24.1]	1.97 [1.69, 2.30] ↑	<0.001
Period of low mood (past 12 months)^e	NA		1254 (4482)	28.1 [26.3, 30]	1397 (4542)	30.3 [28.4, 32.1]	2753 (7121)	38.3 [36.6, 40.1]	1.44 [1.29, 1.61] ↑	<0.001
Self-harm (past 12 months)^f	NA		899 (4478)	20.7 [19, 22.3]	1080 (4533)	22.9 [21.2, 24.6]	1711 (7058)	24.1 [22.8, 25.4]	1.04 [0.92, 1.17] ↑	0.536
Thoughts of suicide (past 12 months)^g	NA		618 (4469)	14.2 [13, 15.4]	688 (4506)	14.7 [13.5, 15.8]	1501 (7038)	20.8 [19.2, 22.4]	1.49 [1.31, 1.69]↑	<0.001
Suicide attempts (past 12 months)^h	NA		215 (4469)	5 [4.3, 5.6]	209 (4505)	3.8 [3.2, 4.4]	439 (7048)	6.3 [5.5, 7.1]	1.67 [1.34, 2.07] ↑	<0.001

CI: confidence interval; OR: odds ratio.

2019 data excludes participants attending Kura Kaupapa Māori and those who did not report a male or female sex to allow comparisons to previous survey waves.

Adjusted for sex, age, ethnicity, locale.

Good overall well-being defined as a score of 13 or more on the World Health Organization 5-item (WHO-5).

Depression symptoms measured using the Reynolds Adolescent Depression Scale (RADS) in 2001 or the Reynolds Adolescent Depression Scale – Short Form (RADS-SF) from 2007 (cut off: 28+).

Period of low mood based on the question: During the past 12 months, was there ever a time when you felt sad, blue or depressed for 2 weeks or more in a row? (Yes/No).

Discussion

Summary of main findings

In this large-scale, representative survey series, we identified marked declines in mental health on key indicators for New Zealand secondary school students from 2012 to 2019, after relative stability from 2001 to 2012. Declines were dramatic, with the overall prevalence of depressive symptoms rising from 13% to 23% in this 7-year period. Declines were unevenly spread and were generally greater among Māori and Pacific students and those in lower socioeconomic neighbourhoods, resulting in increased inequity. The included indicators are self-reported and relate to well-being, anxiety, depression, periods of low mood, self-harm, suicide thoughts and suicide attempts. These findings from diverse population samples including detailed sub-group analyses form an important addition to the global and local adolescent mental health discourse.

Rangatahi Māori reported particularly high prevalence of all mental health indicators in both 2012 and 2019, with inequities widening over this period. Rangatahi Māori were the only ethnic group to report an increase in self-harm, and inequities for depression symptoms and suicide attempts were stark. Given Te Tiriti o Waitangi obligations and Aotearoa cross-government goals of improving equity for Māori, these findings highlight significant failures of the current system (Clark et al., 2022). Pacific students also reported high prevalence of depressive symptoms, suicide thoughts and suicide attempts in both 2012 and 2019, with large increases over the period. Almost all of the observed increase in prevalence of suicide attempts in the total population could be accounted for by increases among Māori and Pacific students. Asian students generally reported relatively low mental health need in 2012 but reported some of the largest increases in 2019, in both relative and absolute terms, including the largest increase in suicide thoughts of any ethnic group (from 12% to 20%) and large increases in depression symptoms.

Socioeconomic deprivation and mental health status had a clear inverse relationship, with those from higher deprivation neighbourhoods reporting lower mental health status on all measures in both 2012 and 2019. Declines in mental health status across this period were largest among those in high-deprivation neighbourhoods and smallest among those in low-deprivation neighbourhoods, in both relative and absolute terms, resulting in a substantial widening of pre-existing inequities.

Females reported higher prevalence of mental health problems than males on all included indicators (depression symptoms, period of low mood, self-harm, suicide thoughts, and suicide attempts) in both 2012 and 2019, in keeping with long-recognised differences (Altemus et al., 2014). Declines in mental health status from 2012 to 2019 were generally larger among males in relative terms but larger among females in absolute terms. For example, prevalence of depression symptoms increased for males from 8.7% to 16.4% (relative increase OR = 2.04, absolute increase 7.7%), and for females from 17.4% to 29.0% (relative increase OR = 1.94, absolute increase 11.6%). Although the gap in need between the sexes could be seen to be narrowing, females still reported very much higher need in 2019.

Younger students (<16) reported larger increases in the prevalence of depression symptoms, suicide thoughts and suicide attempts than older students (16+), in both absolute and relative terms, suggesting a downward shift in the age of onset of depression and suicidality, both traditionally thought to develop later in adolescence (Solmi et al., 2021).

While total population increases in distress reported here largely resemble patterns observed in other high-income English-speaking nations over a similar timeframe, there are important differences. We see less distinction in patterns of increases between males and females than in some international studies, where increases were reported as highly concentrated among females (Keyes et al., 2019; Twenge et al., 2019; Wiens et al., 2020). We also see stronger increases among Indigenous and ethnic minority youth and those from lower socioeconomic groups, while others who provide demographic analyses report larger increases among more privileged White Americans (Twenge et al., 2019), or mainly consistent across ethnicities and socioeconomic American groups (e.g. Keyes et al., 2019).

Causality for population-level changes in mental health is difficult to determine, and full consideration of this issue is outside our scope. Artefactual explanations may account for some increased reporting, but corresponding increases in related behavioural phenomena offer converging evidence in support of actual increases in the prevalence of distress (e.g. Cook et al., 2021; Cybulski et al., 2021).

Seemingly paradoxically, many established drivers of distress appear stable or moving in a desirable direction (Fleming et al., 2022b; Twenge et al., 2019). In Aotearoa, most students continued to report strong family and school connections, the prevalence of bullying appeared largely stable, and substance use and risk behaviours trended downwards, although there was some evidence of increasing financial stress (Fleming et al., 2022b).

Key among proposed drivers of increased distress is the rapid increase in adolescents’ use of new technologies (Cook et al., 2021; Twenge et al., 2019). Although the association between social media use and distress among individuals is not strong (Cook et al., 2021; Odgers and Jensen, 2020), the impact of new technologies may have important population effects. Other proposed drivers include increases in perfectionism (Curran and Hill, 2019), school pressure (Cosma et al., 2020) and future concerns such as climate anxiety (Hickman et al., 2021), although these do not provide a strong rationale for the rapid pace of change. There have been other changes in important contextual factors, such as increases in parental monitoring, changes in parenting and, in some areas, increased political polarisation. New Zealand adolescents have identified climate change as a major problem facing young people (Fleming et al., 2020a).

The increases in inequities in distress we observed (i.e. greater increases among Māori students, Pacific students and those in high-deprivation areas) are not explained by these global hypotheses. A possible contributor to increased inequity in mental health need in Aotearoa is widening income inequality. Aotearoa saw the most rapid increases in income inequality among OECD nations between the mid-1980s and the mid-2000s, reaching roughly the seventh highest level of income inequality in the OECD (Barber, 2011). Such increases have since slowed, but highincome inequality remains embedded. Continuing research is required, which should include ethnic- and socioeconomic-specific data and explore causality. Increasing inequity in distress may also be linked to compounding vulnerabilities, by which members of groups with initial higher socioeconomic need and/or affected by discrimination, racism and limited opportunities are disproportionately affected by the drivers of increasing mental health need (Simon-Kumar et al., 2022). None of these hypotheses provide an explanation on their own. Observed changes may result from complex interplays of factors at individual, family, school and society levels (Collishaw and Sellers, 2020; Fleming et al., 2022b).

Despite incomplete understandings of causal processes, these findings and international research highlight the need for an urgent and immediate response. Improving the population mental health requires efforts at multiple levels (Fusar-Poli et al., 2021). Along with continued attention to known determinants of distress (e.g. racism, discrimination, social exclusion, poverty, inequality, and exposure to abuse and neglect) youth and mental health professionals and advocates must be alert and responsive to evidence around emerging determinants. Universal public health approaches hold promise, in combination with prevention, early intervention, and more and diversified treatment options (Fusar-Poli et al., 2021). Services should be resourced to meet changing need profiles, e.g., increased supports in primary and intermediate schools, given evidence for earlier onset of mental health problems.

Population-specific responses to address profound inequities are required. In Aotearoa, reducing mental health inequity among rangatahi Māori is a priority (Government Inquiry into Mental Health and Addiction 2018; Waitangi Tribunal, 2019). Policy directives must be actively anti-racist and anti-discrimination, and address entrenched disadvantage caused by mechanisms of colonisation compounding over generations. In line with the principle of tino rangatiratanga (self-determination), Māori-led solutions and those incorporating mātauranga Māori (Māori knowledge) are needed.

Further population-specific initiatives are required to address high need among other groups reported here (Pacific students, those from high-deprivation neighbourhoods, females and, increasingly, Asian students) and those outside the scope of this paper, including adolescents who are: takatāpui, rainbow and/or gender diverse (Fenaughty et al., 2021a, 2021b); disabled (Clark et al., 2021); and care-experienced (Fleming et al., 2022a). Need is especially high among adolescents with particular intersections of identity (Roy et al., 2021), including Māori, Pacific and Asian females (Fleming et al., 2020c).

Strengths and limitations

Strengths of the Youth2000 series include large, representative samples comprising high proportions of the New Zealand adolescent population, robust sampling techniques, and a broad range of indicators, most used consistently over time.

There are also limitations. Only students attending school on the day of the survey participated. Those not enrolled in mainstream schools or absent are generally likely to be less healthy than those present, thus the true levels of some difficulties are likely underrepresented. Similarly, response rates have decreased over time, possibly resulting in increased non-response bias. Reliable and valid measures were used to assess distress, and questions were extensively piloted; however, formal diagnoses of disorders based on the clinical interview were not possible. Youth19 used a regional rather than a national sample; however, the findings of our secondary analysis of trends in the regions only provided the same, or even marginally stronger results (albeit with wider confidence intervals), suggesting that findings are unlikely to be an artefact of differences in sampling frame. Youth19 predates the COVID-19 global pandemic, thus did not capture the effects of the pandemic on student well-being. We provide a snapshot of mental health and well-being immediately preceding the pandemic, capturing earlier changes and providing a baseline for measuring future change.

Future research

We report findings for male and female sex, major ethnic groups and neighbourhood deprivation bands. High mental health needs were also reported by other groups. Further work considering these findings in context is required, including for young people who are outside the male/female binary.

We observed large changes in reported needs based on binarised variables and top-down segmentation using demographic factors. While this is useful to demonstrate population-level changes, future research should further unpack mental health need, e.g., by considering changes in scale distribution or changes on individual scale items over time. It would also be useful to investigate specific profiles of need not captured by a top-down approach. Bottom-up, data-driven techniques such as cluster analysis could be useful here. Work to understand the drivers and mechanisms of increases in need was outside the scope of this study, but is important. To enable meaningful trend comparisons in future, continued collection of comparable data is vital.

Conclusion

We found high prevalence of mental distress among New Zealand secondary school students in 2019 and sharp increases between 2012 and 2019. Needs and changes were unevenly distributed, with a general pattern of larger increases among groups with pre-existing higher prevalence (e.g. females, rangatahi Māori, those in higher deprivation neighbourhoods), resulting in widening inequities. Increases on some indicators were notably high among Asian students. Urgent work to address these needs is required, with equity as a priority. It is important to understand where the changes are, for whom, to attempt to address causal processes and inform our response. Further work is needed to understand the drivers. These findings provide a baseline of distress prior to the COVID-19 global pandemic. Continued collection of directly comparable data is required to ensure that changes (related to the pandemic and otherwise) are captured.

Footnotes

Acknowledgements

Thank you first and foremost to the students, staff and schools who have taken part in Youth2000 surveys over almost 20 years. Thank you to the members of the research teams who implemented the surveys and to our steering and advisory groups. Thank you to Stefanie Lash for editing support.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: Youth12 was funded by the Ministries of Youth Development, Social Development, Health, Education and Justice; the Department of Labour; The Families Commission; and the Health Promotion Agency. The Youth19 Rangatahi Smart Survey was funded by the Health Research Council of New Zealand (HRC project grants 17/315 and 18/473). The lead author was funded by a Victoria Doctoral Scholarship, Te Herenga Waka – Victoria University of Wellington.

ORCID iD

Kylie Sutcliffe

References

Adolescent Health Research Group (2003) A health profile of New Zealand youth who attend secondary school. New Zealand Medical Journal 116: 1171U380.

Adolescent Health Research Group (2008) Youth’07: The health and wellbeing of secondary school students in New Zealand. Report, University of Auckland, Auckland, New Zealand.

Altemus

Sarvaiya

Neill Epperson

(2014) Sex differences in anxiety and depression clinical perspectives. Frontiers in Neuroendocrinology 35: 320–330.

Atkinson

Salmond

Crampton

(2019) NZDEP 2018 Index of deprivation users’ manual. Report, University of Otago, Wellington, New Zealand.

Barber

(2011) How to get closer together: Impacts of income inequality and policy responses. Policy Quarterly 7: 62–68.

Clark

Ball

Fenaughty

, et al. (2022) Indigenous adolescent health in Aotearoa New Zealand: Trends, policy and advancing equity for rangatahi Māori, 2001–2019. The Lancet Regional Health, Western Pacific 28: 100554.

Clark

Fleming

Bullen

, et al. (2013) Health and well-being of secondary school students in New Zealand: Trends between 2001, 2007 and 2012. Journal of Paediatrics and Child Health 49: 925–934.

Clark

Kuresa

Drayton

, et al. (2021) A Youth19 brief: Young people with disabilities. Research brief for The Youth19 Research Group, Victoria University of Wellington and University of Auckland, Auckland, New Zealand. Available at: www.youth19.ac.nz (accessed 1 June 2022).

Collishaw

(2015) Annual research review: Secular trends in child and adolescent mental health. Journal of Child Psychology and Psychiatry, and Allied Disciplines 56: 370–393.

10.

Collishaw

Sellers

(2020) Trends in child and adolescent mental health prevalence, outcomes and inequalities. In: Taylor

Verhulst

Wong

, et al. (eds) Mental Health and Illness of Children and Adolescents: Mental Health and Illness Worldwide. Singapore: Springer, pp. 63–73.

11.

Cook

Hamilton

Montazer

, et al. (2021) Increases in serious psychological distress among Ontario students between 2013 and 2017: Assessing the impact of time spent on social media. Canadian Journal of Psychiatry / Revue Canadienne de Psychiatrie 66: 747–756.

12.

Cosma

Stevens

Martin

, et al. (2020) Cross-national time trends in adolescent mental well-being from 2002 to 2018 and the explanatory role of schoolwork pressure. The Journal of Adolescent Health: Official Publication of the Society for Adolescent Medicine 66: S50–S58.

13.

Curran

Hill

(2019) Perfectionism is increasing over time: A meta-analysis of birth cohort differences from 1989 to 2016. Psychological Bulletin 145: 410–429.

14.

Cybulski

Ashcroft

Carr

, et al. (2021) Temporal trends in annual incidence rates for psychiatric disorders and self-harm among children and adolescents in the UK, 2003–2018. BMC Psychiatry 21: 229.

15.

Ellis

Schneider

(2022) R survey package (v 1.1.1) (computer software). R Foundation. https://CRAN.R-project.org/package=srvyr (accessed 12 March 2022).

16.

Erskine

Baxter

Patton

, et al. (2017) The global coverage of prevalence data for mental disorders in children and adolescents. Epidemiology and Psychiatric Sciences 26: 395–402.

17.

Fenaughty

Clark

Choo

, et al. (2021a) Te āniwaniwa takatāpui whānui: Te aronga taera mō ngā rangatahi Sexual attraction and young people’s wellbeing in Youth19. Report for The Youth19 Research Group, Victoria University of Wellington and University of Auckland, Auckland, New Zealand. Available at: www.youth19.ac.nz (accessed 1 June 2022).

18.

Fenaughty

Sutcliffe

Fleming

, et al. (2021b) A Youth19 brief: Transgender and diverse gender students. Research brief for The Youth19 Research Group, Victoria University of Wellington and University of Auckland, Auckland, New Zealand. Available at: www.youth19.ac.nz (accessed 1 June 2022).

19.

Fleming

Archer

Sutcliffe

, et al. (2022a) Young people who have been involved with Oranga Tamariki: Mental and physical health and healthcare access. Research brief for The Youth19 Research Group, Victoria University of Wellington and University of Auckland, Auckland, New Zealand. Available at: www.youth19.ac.nz (accessed 1 June 2022).

20.

Fleming

Ball

Bavin

, et al. (2022b) Mixed progress in adolescent health and wellbeing in Aotearoa New Zealand 2001–2019: A population overview from the Youth2000 survey series. Journal of the Royal Society of New Zealand 52: 426–449.

21.

Fleming

Ball

Kang

, et al. (2020a) Youth19: Youth voice brief. Research brief for the Youth19 Research Group, Victoria University of Wellington and University of Auckland, Auckland, New Zealand. Available at: www.youth19.ac.nz (accessed 1 June 2022).

22.

Fleming

Clark

Denny

, et al. (2014) Stability and change in the mental health of New Zealand secondary school students 2007-2012: Results from the national adolescent health surveys. Australian and New Zealand Journal of Psychiatry 48: 472–480.

23.

Fleming

Peiris-John

Crengle

, et al. (2020b) Youth19 Rangatahi smart survey, initial findings: Introduction and methods. Report for the Youth19 Research Group, Victoria University of Wellington and University of Auckland, Auckland, New Zealand. Available at: www.youth19.ac.nz (accessed 1 June 2022).

24.

Fleming

Tiatia-Seath

Peiris-John

, et al. (2020c) Youth19 Rangatahi smart survey, initial findings: Hauora Hinengaro / emotional and mental health. Report for the Youth19 Research Group, Victoria University of Wellington and University of Auckland, Auckland, New Zealand. Available at: www.youth19.ac.nz (accessed 1 June 2022).

25.

Fusar-Poli

Correll

Arango

, et al. (2021) Preventive psychiatry: A blueprint for improving the mental health of young people. World Psychiatry: Official Journal of the World Psychiatric Association 20: 200–221.

26.

Government Inquiry into Mental Health and Addiction (2018) He Ara Oranga: Report of the government inquiry into mental health and addiction. Wellington. Available at: www.mentalhealth.inquiry.govt.nz/inquiry-report/ (accessed 1 November 2021).

27.

Gromada

Rees

Chzhen

(2020) Worlds of Influence: Understanding what shapes child well-being in rich countries. Report for UNICEF. Available at: www.unicef-irc.org/publications/pdf/Report-Card-16-Worlds-of-Influence-child-wellbeing.pdf (accessed 12 December 2021).

28.

Haugan

Frostad

Mjaavatn

(2021) Girls suffer: The prevalence and predicting factors of emotional problems among adolescents during upper secondary school in Norway. Social Psychology of Education 24: 609–634.

29.

Hickman

Marks

Pihkala

, et al. (2021) Climate anxiety in children and young people and their beliefs about government responses to climate change: A global survey. The Lancet: Planetary Health 5: e863–e873.

30.

Keyes

Gary

O’Malley

, et al. (2019) Recent increases in depressive symptoms among US adolescents: Trends from 1991 to 2018. Social Psychiatry and Psychiatric Epidemiology 54: 987–996.

31.

Kroenke

Spitzer

Williams

, et al. (2009) An Ultra-Brief Screening Scale for anxiety and depression: The PHQ-4. Psychosomatics 50: 613–621.

32.

Milfont

Merry

Robinson

, et al. (2008) Evaluating the short form of the Reynolds Adolescent Depression Scale in New Zealand adolescents. The Australian and New Zealand Journal of Psychiatry 42: 950–954.

33.

Ministry of Health (2017) HISO 10001:2017 Ethnicity data protocols. Available at: www.health.govt.nz/publication/hiso-100012017-ethnicity-data-protocols (accessed 26 June 2020).

34.

Odgers

Jensen

(2020) Annual research review: Adolescent mental health in the digital age – Facts, fears, and future directions. Journal of Child Psychology and Psychiatry, and Allied Disciplines 61: 336–348.

35.

R Core Team (2021) R: A Language and Environment for Statistical Computing (software). Vienna: R Foundation for Statistical Computing. Available at: www.R-project.org/ (accessed 12 March 2022).

36.

Reid

Cormack

Paine

S-J

(2019) Colonial histories, racism and health: The experience of Māori and Indigenous peoples. Public Health 172: 119–124.

37.

Reynolds

(2002) Reynolds Adolescent Depression Scale, 2nd Edition. Professional Manual. Lutz, FL: Psychological Assessment Resources.

38.

Rivera-Rodriguez

Clark

Fleming

, et al. (2021) National estimates from the Youth ‘19 Rangatahi smart survey: A survey calibration approach. PLoS ONE 16: e0251177.

39.

Roy

Greaves

Peiris -John

, et al. (2021) Negotiating multiple identities: Intersecting identities among Māori, Pacific, Rainbow and Disabled young people. Report for the Youth19 Research Group, Victoria University of Wellington and University of Auckland, Auckland, New Zealand. Available at: www.youth19.ac.nz (accessed 1 June 2022).

40.

Simon-Kumar

Lewycka

Clark

, et al. (2022) Flexible resources and experiences of racism among a multi-ethnic adolescent population in Aotearoa, New Zealand: An intersectional analysis of health and socioeconomic inequities using survey data. The Lancet 400: 1130–1143.

41.

Solmi

Radua

Olivola

, et al. (2021) Age at onset of mental disorders worldwide: Large-scale meta-analysis of 192 epidemiological studies. Molecular Psychiatry 27: 281–295.

42.

Statistics New Zealand (2017) Statistical standard for geographic areas 2018. Available at: www.stats.govt.nz (accessed 26 June 2020).

43.

Statistics New Zealand (2020) 2018 census totals by topic. Available at: www.stats.govt.nz/information-releases/2018-census-totals-by-topic-national-highlights-updated/ (accessed 25 July 2022).

44.

Topp

Østergaard

Søndergaard

, et al. (2015) The WHO-5 Well-Being Index: A systematic review of the literature. Psychotherapy and Psychosomatics 84: 167–176.

45.

Twenge

Cooper

Joiner

, et al. (2019) Age, period, and cohort trends in mood disorder indicators and suicide-related outcomes in a nationally representative dataset, 2005–2017. Journal of Abnormal Psychology 128: 185–199.

46.

Waitangi Tribunal (2019) Hauora: Report on stage one of the health services and outcomes Kaupapa inquiry. Wai 2575 Waitangi tribunal report. Lower Hutt, New Zealand: Legislation Direct. Available at: https://forms.justice.govt.nz/search/Documents/WT/wt_DOC_152801817/Hauora%20W.pdf (accessed 1 June 2022).

47.

Wiens

Bhattarai

Pedram

, et al. (2020) A growing need for youth mental health services in Canada: Examining trends in youth mental health from 2011 to 2018. Epidemiology and Psychiatric Sciences 29: E115.

48.

World Health Organization (1998) Wellbeing measures in primary health care/The Depcare Project. Copenhagen. Available at: www.euro.who.int/__data/assets/pdf_file/0016/130750/E60246.pdf (accessed 26 June 2020).